The European Nitrogen Assessment: Sources, Effects and Policy Perspectives

This page intentionally left blank

The European Nitrogen Assessment Sources, Effects and Policy Perspectives A century ago, when the world depended on fossil nitrogen and manure recycling, there was insufficient reactive nitrogen to feed the growing human population. With the invention of the Haber–Bosch process, humans found a way to make cheap reactive nitrogen from the almost inexhaustable supply of atmospheric di-nitrogen. What humans did not anticipate was that the massive increase in reactive nitrogen supply, exacerbated by fossil fuel burning, would lead to a web of new environmental problems cutting across all global-change challenges. The European Nitrogen Assessment presents the first full, continental-scale assessment of reactive nitrogen in the environment and sets the problem in context by providing a multidisciplinary introduction to the key processes in the nitrogen cycle. Issues of up-scaling from field, farm and city to national and continental scales are addressed in detail with emphasis on opportunities for better management at local to global levels. A comprehensive series of maps showing nitrogen pools and fluxes across Europe also highlight the location of the major threats and allow a comparison of national budgets for the first time. Five key societal threats posed by reactive nitrogen are assessed, providing a framework for a set of policies that can be used for joined-up management of the nitrogen cycle in Europe. This includes the first cost–benefit analysis for different reactive nitrogen forms and consideration of future scenarios. Incorporating a handy technical synopsis and summary for policy makers, this land-mark volume is an essential reference for academic researchers across a wide range of disciplines, as€well as for stakeholders and policy makers in Europe and beyond. It is also a valuable tool in helping communicate the key environmental issues and future challenges to the wider public. Mark Sutton is an environmental physicist investigating human alteration of the nitrogen cycle, with specific attention to ammonia. He is coordinator of the major integrated project ‘NitroEurope’, a 5-year effort, bringing together 64 research institutes to ask how nitrogen is affecting the European greenhouse gas balance. Dr Sutton is vice-chair of the ‘Nitrogen in Europe’ (NinE) programme of the European Science Foundation, the Director of the European Centre of the International Nitrogen Initiative (INI) and co-chair of the Task Force on Reactive Nitrogen of the UN-ECE Convention on Long-range Transboundary Air Pollution. Clare Howard is currently engaged in a postdoctoral fellowship in knowledge transfer, with an emphasis on research networks which focus on nitrogen. Dr Howard is project coordinator for the European Nitrogen Assessment and for the Task Force on Reactive Nitrogen, which sits beneath the Working Group on Strategies and Review of the Convention on Long Range Transboundary Air Pollution. Her research interests involve the modelling of biogeochemical cycles of nitrogen and carbon and assessing uncertainty in model systems. Jan Willem Erisman heads the Biomass, Coal and Environmental Research Unit of the Energy Research Centre of the Netherlands (ECN) and is a professor in Integrated Nitrogen studies at Vrije Universiteit, Amsterdam. His research focuses on atmosphere–biosphere exchange of gases and aerosols related to acidification and eutrophication and climate change. He was instrumental in establishing the International Nitrogen Initiative, the Nanjing Declaration on Nitrogen Management, the EU 6th Framework research program NitroEurope and for chairing the European Science Foundation project NinE and the EU COST Action 729.

Gilles Billen is research director of the Centre National de la Recherche Scientifique (CNRS) at the University Pierre and Marie Curie (Paris) where his research covers many aspects of biogeochemistry, with an emphasis on the nitrogen, phosphorus and silica cycles. His main expertise is on the assessment and modelling of the ecological functioning of hydrosystems, including marine, estuarine and freshwater environments. From 1997 to 2007, he was the Director of the PIREN-Seine programme, a large interdisciplinary research programme on the Seine river watershed. Albert Bleeker works as a senior scientist at the Energy Research Centre of the Netherlands, in the department of Air Quality and Climate Change. He has almost 20 years of experience in the field of nitrogen, where his main expertise is on the atmospheric emission, transport and deposition of nitrogen at various spatial scales, as well as studies on the effect of nitrogen in the natural environment. Currently, he is the Nitrogen in Europe (NinE) Programme Co-ordinator and a member of the COST 729 Management Committee. Peringe Grennfelt has a background in atmospheric chemistry. His research includes regional air pollution problems in Europe, in particular acidification, nitrogen deposition and tropospheric ozone. He has coordinated several national and international research programmes including the EU project Network for the support of European Policies on Air Pollution (NEPAP). He is presently leading the Mistra Climate Policy Research Programme (Clipore) and the Swedish Clean Air Research Programme (SCARP). Hans van Grinsven works at the Netherlands Environmental Assessment Agency where he conducts research and coordinates projects related to agriculture and environment, focusing on nitrogen and phosphorus, and sustainable food production. Dr van Grinsven was responsible for evaluations of national implementation of the EU Nitrates Directive and was also closely involved in the evaluations of the implementation of the EU Water framework Directive and EU NEC directive. Bruna Grizzetti is a researcher in the field of large scale modelling of nutrient and water transfer. She works on modelling nutrient pressures on water at European scale in support to the implementation of environmental European policies, such as the Water Framework Directive, Nitrates Directive and the Marine Strategy. Since 2007, Dr Grizzetti has been a member of the Coordination Team of the European Nitrogen Assessment process, supported through the European Science Foundation.

TFRN

The European Nitrogen Assessment has been prepared through coordinated action led by the Nitrogen in Europe (NinE) Research Networking Programme of the European Science Foundation, the NitroEurope Integrated Project supported by European Commission’s 6th Framework Programme and the COST Action 729. The Assessment is a contribution to the work of the Task Force on Reactive Nitrogen (TFRN), led by the UK and the Netherlands, in support of the long-term goals of the UN-ECE Convention on Long-range Transboundary Air Pollution (CLRTAP). In parallel, the Assessment represents a European contribution to the work of the International Nitrogen Initiative (INI), a joint project of the International Geosphere Biosphere Programme (IGBP) and the Scientific Committee on Problems of the Environment (SCOPE), providing evidence to underpin many United Nations and other multi-lateral agreements. The actual assessment work has been carried out by 200 experts from 21 countries and 89 organizations which kindly provided support for this work. The ENA has been conducted as a scientifically independent process. The views and conclusions expressed are those of the authors, and do not necessarily reflect policies of the contributing organizations.

Acknowledgements

The European Nitrogen Assessment was prepared by the list of contributors given on page ix, with the support of the NinE Programme of the European Science Foundation, the€ NitroEurope IP (funded by the European Commission 6th Framework Programme), the COST Action 729, the Task Force on Reactive Nitrogen and the International Nitrogen Initiative. The editors gratefully acknowledge the wider support which the assessment received, in the form of all those attending and hosting the ENA workshops, internal and

external reviewers of chapters and summaries. We particularly thank Agnieszka Eljasz of CEH, Susan Francis and Laura Clark of Cambridge University Press, Ellen Degottv.W.Rekowski and Paola Campus of the European Science Foundation, Peter Coleman of Defra, Anastasios Kentarchos of the European Commission, Matti Johannsson and Tea Aulavuo of the Secretariat to the UN-ECE Convention on Long-range Transboundary Air Pollution for their support through this process.

The European Nitrogen Assessment Sources, Effects and Policy Perspectives Edited by

Mark A. Sutton

NERC Centre for Ecology and Hydrology

Clare M. Howard

NERC Centre for Ecology and Hydrology and University of Edinburgh

Jan Willem Erisman Energy Research Centre of the Netherlands

Gilles Billen

CNRS and University of Paris VI

Albert Bleeker

Energy Research Centre of the Netherlands

Peringe Grennfelt

Swedish Environmental Research Institute (IVL)

Hans van Grinsven

PBL Netherlands Environmental Assessment Agency

Bruna Grizzetti

European Commission Joint Research Centre

C A M B R I D G E U N I V E R SI T Y P R E S S Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo, Delhi, Tokyo, Mexico City Cambridge University Press The Edinburgh Building, Cambridge CB2 8RU, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title:€www.cambridge.org/9781107006126 © Cambridge University Press 2011 © Editorial contributions by Bruna Grizzetti, European Union 2011 © Chapter 17, European Union 2011 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2011 Printed in the United Kingdom at the University Press, Cambridge A catalogue record for this publication is available from the British Library Library of Congress Cataloguing in Publication Data The European nitrogen assessmentâ•›:â•›sources, effects, and policy perspectivesâ•›/â•›[edited by] Mark A. Sutton ... [et al.]. â•…â•… p.â•… cm. Includes bibliographical references and index. ISBN 978-1-107-00612-6 (hardback) 1.╇ Nitrogen compounds–Environmental aspects–Europe. 2.╇ Nitrogen cycle–Europe.â•… 3.╇ Nitrogen fertilizers–Government policy–Europe.â•… I.╇ Sutton, Mark A. TD196.N55E96 2011 363.738–dc22â•…â•…â•… 2010051120 ISBN 978-1-107-00612-6 Hardback Additional resources for this publication at www.cambridge.org/ena Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Contents List of contributors╯╯╯╯╯╯page xi Foreword╯╯╯╯╯╯xxiii Summary for policy makers╯╯╯╯╯╯xxiv Technical summary╯╯╯╯╯╯xxxv

1 Assessing our nitrogen inheritance╯╯╯╯╯╯1 Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti

Part I╇ Nitrogen in Europe:€the present position╯╯╯╯╯╯ 2 The European nitrogen problem in a global perspective╯╯╯╯╯╯9 Jan Willem Erisman, Hans van Grinsven, Bruna Grizzetti, Fayçal Bouraoui, David Powlson, Mark A. Sutton, Albert Bleeker and Stefan Reis 3 Benefits of nitrogen for food, fibre and industrial production╯╯╯╯╯╯32 Lars Stoumann Jensen, Jan K. Schjoerring, Klaas W. van der Hoek, Hanne Damgaard Poulsen, John F. Zevenbergen, Christian Pallière, Joachim Lammel, Frank Brentrup, Age W. Jongbloed, Jaap Willems and Hans van Grinsven 4 Nitrogen in current European policies╯╯╯╯╯╯62 Oene Oenema, Albert Bleeker, Nils Axel Braathen, Michaela Budňáková, Keith Bull, Pavel Čermák, Markus Geupel, Kevin Hicks, Robert Hoft, Natalia Kozlova, Adrian Leip, Till Spranger, Laura Valli, Gerard Velthof and Wilfried Winiwarter 5 The challenge to integrate nitrogen science and policies:€the European Nitrogen Assessment approach╯╯╯╯╯╯82 Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, William J. Bealey, Gilles Billen, Albert Bleeker, Alexander F. Bouwman, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti

Part II╇ Nitrogen processing in the biosphere 6 Nitrogen processes in terrestrial ecosystems╯╯╯╯╯╯99 Klaus Butterbach-Bahl, Per Gundersen, Per Ambus, Jürgen Augustin, Claus Beier, Pascal Boeckx, Michael

Dannenmann, Benjamin Sanchez Gimeno, Ralf Kiese, Barbara Kitzler, Andreas Ibrom, Robert M. Rees, Keith A. Smith, Carly Stevens, Timo Vesala and Sophie Zechmeister-Boltenstern 7 Nitrogen processes in aquatic ecosystems╯╯╯╯╯╯126 Patrick Durand, Lutz Breuer, Penny J. Johnes, Gilles Billen, Andrea Butturini, Gilles Pinay, Hans van Grinsven, Josette Garnier, Michael Rivett, David S. Reay, Chris Curtis, Jan Siemens, Stephen Maberly, Øyvind Kaste, Christoph Humborg, Roos Loeb, Jeroen de Klein, Josef Hejzlar, Nikos Skoulikidis, Pirkko Kortelainen, Ahti Lepistö and Richard Wright 8 Nitrogen processes in coastal and marine ecosystems╯╯╯╯╯╯147 Maren Voss, Alex Baker, Hermann W. Bange, Daniel Conley, Sarah Cornell, Barbara Deutsch, Anja Engel, Raja Ganeshram, Josette Garnier, Ana-Stiina Heiskanen, Tim Jickells, Christiane Lancelot, Abigail McQuatters-Gollop, Jack Middelburg, Doris Schiedek, Caroline P. Slomp and Daniel P. Conley 9 Nitrogen processes in the atmosphere╯╯╯╯╯╯177 Ole Hertel, Stefan Reis, Carsten Ambelas Skjøth, Albert Bleeker, Roy Harrison, John Neil Cape, David Fowler, Ute Skiba, David Simpson, Tim Jickells, Alex Baker, Markku Kulmala, Steen Gyldenkærne, Lise Lotte Sørensen and Jan Willem Erisman

Part III╇ Nitrogen flows and fate at multiple spatial scales 10 Nitrogen flows in farming systems across Europe╯╯╯╯╯╯211 Steve Jarvis, Nick Hutchings, Frank Brentrup, Jørgen Eivind Olesen and Klaas W. van der Hoek 11 Nitrogen flows and fate in rural landscapes╯╯╯╯╯╯229 Pierre Cellier, Patrick Durand, Nick Hutchings, Ulli Dragosits, Mark Theobald, Jean-Louis Drouet,

vii

Contents

Oene€Oenema, Albert Bleeker, Lutz Breuer, Tommy Dalgaard, Sylvia Duretz, Johannes Kros, Benjamin Loubet, Joergen Eivind Olesen, Philippe Mérot, Valérie Viaud, Wim de Vries and Mark A. Sutton

18 Nitrogen as a threat to European air quality╯╯╯╯╯╯405 Jana Moldanová, Peringe Grennfelt, Åsa Jonsson, David Simpson, Till Spranger, Wenche Aas, John Munthe and Ari€Rabl

12 Nitrogen flows and fate in urban landscapes╯╯╯╯╯╯249 Anastasia Svirejeva-Hopkins, Stefan Reis, Jakob Magid, Gabriela B. Nardoto, Sabine Barles, Alexander F. Bouwman, Ipek Erzi, Marina Kousoulidou, Clare M. Howard and Mark A. Sutton

19 Nitrogen as a threat to the European greenhouse balance╯╯╯╯╯╯434 Klaus Butterbach-Bahl, Eiko Nemitz, Sönke Zaehle, Gilles Billen, Pascal Boeckx, Jan Willem Erisman, Josette€Garnier, Rob Upstill-Goddard, Michael Kreuzer, Oene Oenema, Stefan Reis, Martijn Schaap, David Simpson, Wim de Vries, Wilfried Winiwarter and Mark A. Sutton

13 Nitrogen flows from European regional watersheds to coastal marine waters╯╯╯╯╯╯271 Gilles Billen, Marie Silvestre, Bruna Grizzetti, Adrian Leip, Josette Garnier, Maren Voss, Robert Howarth, Fayçal Bouraoui, Ahti Lepistö, Pirkko Kortelainen, Penny Johnes, Chris Curtis, Christoph Humborg, Erik Smedberg, Øyvind Kaste, Raja Ganeshram, Arthur Beusen and Christiane Lancelot 14 Atmospheric transport and deposition of reactive nitrogen in Europe╯╯╯╯╯╯298 David Simpson, Wenche Aas, Jerzy Bartnicki, Haldis Berge, Albert Bleeker, Kees Cuvelier, Frank Dentener, Tony Dore, Jan Willem Erisman, Hilde Fagerli, Chris Flechard, Ole Hertel, Hans van Jaarsveld, Mike Jenkin, Martijn Schaap, Valiyaveetil Shamsudheen Semeena, Philippe Thunis, Robert Vautard and Massimo Vieno 15 Geographical variation in terrestrial nitrogen budgets across Europe╯╯╯╯╯╯317 Wim de Vries, Adrian Leip, Gert Jan Reinds, Johannes Kros, Jan Peter Lesschen, Alexander F. Bouwman, Bruna Grizzetti, Fayçal Bouraoui, Klaus Butterbach-Bahl, Peter Bergamaschi and Wilfried Winiwarter 16 Integrating nitrogen fluxes at the European scale╯╯╯╯╯╯345 Adrian Leip, Beat Achermann, Gilles Billen, Albert Bleeker, Alexander F. Bouwman, Wim de Vries, Ulli Dragosits, Ulrike Döring, Dave Fernall, Markus Geupel, Jürg Herolstab, Penny Johnes, Anne-Christine Le Gall, Suvi Monni, Rostislav Nevečeřal, Lorenzo Orlandini, Michel Prud’homme, Hannes I. Reuter, David Simpson, Günther Seufert, Till Spranger, Mark A. Sutton, John van Aardenne, Maren Voß and Wilfried Winiwarter

Part IV╇ Managing nitrogen in relation to key societal threats 17 Nitrogen as a threat to European water quality╯╯╯╯╯╯379 Bruna Grizzetti, Fayçal Bouraoui, Gilles Billen, Hans van Grinsven, Ana Cristina Cardoso, Vincent Thieu, Josette Garnier, Chris Curtis, Robert Howarth and Penny Johnes

viii

20 Nitrogen as a threat to European terrestrial biodiversity╯╯╯╯╯╯463 Nancy B. Dise, Michael Ashmore, Salim Belyazid, Albert Bleeker, Roland Bobbink, Wim de Vries, Jan Willem Erisman, Till Spranger, Carly J. Stevens and Leon van den Berg 21 Nitrogen as a threat to European soil quality╯╯╯╯╯╯495 Gerard Velthof, Sébastien Barot, Jaap Bloem, Klaus Butterbach-Bahl, Wim de Vries, Johannes Kros, Patrick Lavelle, Jørgen Eivind Olesen and Oene Oenema

Part V╇ European nitrogen policies and future challenges 22 Costs and benefits of nitrogen in the environment╯╯╯╯╯╯513 Corjan Brink, Hans van Grinsven, Brian H. Jacobsen, Ari Rabl, Ing-Marie Gren, Mike Holland, Zbigniew Klimont, Kevin Hicks, Roy Brouwer, Roald Dickens, Jaap Willems, Mette Termansen, Gerard Velthof, Rob Alkemade, Mark van Oorschot and Jim Webb 23 Developing integrated approaches to nitrogen management╯╯╯╯╯╯541 Oene Oenema, Joost Salomez, Christina Branquinho, Michaela Budňáková, Pavel Čermák, Markus Geupel, Penny Johnes, Chris Tompkins, Till Spranger, Jan Willem Erisman, Christian Pallière, Luc Maene, Rocio Alonso, Rob Maas, Jacob Magid, Mark A. Sutton and Hans van Grinsven 24 Future scenarios of nitrogen in Europe╯╯╯╯╯╯ 551 Wilfried Winiwarter, Jean-Paul Hettelingh, Alex F. Bouwman, Wim de Vries, Jan Willem Erisman, James Galloway, Zbigniew Klimont, Allison Leach, Adrian Leip, Christian Pallière, Uwe A. Schneider, Till Spranger, Mark A. Sutton, Anastasia Svirejeva-Hopkins, Klaas W. van der Hoek and Peter Witzke

Contents

25 Coordinating European nitrogen policies between international conventions and intergovernmental organizations╯╯╯╯╯╯570 Keith Bull, Robert Hoft and Mark A. Sutton 26 Societal choice and communicating the European nitrogen challenge╯╯╯╯╯╯585 David S. Reay, Clare M. Howard, Albert Bleeker, Pete Higgins, Keith Smith, Henk Westhoek, Trudy Rood,

Mark R. Theobald, Alberto Sanz-Cobeña, Robert M. Rees, Dominic Moran, Kate Ravilious and Stefan Reis

Glossary╯╯╯╯╯╯602 Index╯╯╯╯╯╯607

ix

This page intentionally left blank

Contributors

John van Aardenne European Commission Joint Research Center Institute for Environment and Sustainability via Enrico Fermi 2749 21027 Ispra (VA) Italy Wenche Aas NILU, Norwegian Institute for Air Research PB 100 2027 Kjeller Norway Beat Achermann Federal Office for the Environment Air Pollution Control and NIR Division Air Quality Management Section CH-3003 Bern Switzerland Rob Alkemade Netherlands Environmental Assessment Agency P.O. Box 303 3720 AH The Netherlands Per Ambus Risø DTU National Laboratory for Sustainable Energy, Technical University of Denmark Biosystems Division Frederiksborgvej 399 4000 Roskilde Denmark Michael Ashmore University of York Environment Department Heslingon YO10 5DD United Kingdom Juergen Augustin Leibniz-Centre for Agricultural Landscape Research (ZALF) Eberswalder Strasse 84 Muencheberg

D-15374 Germany Alex Baker School of Environmental Sciences University of East Anglia Norwich NR4 7TJ United Kingdom Hermann W. Bange IFM-GEOMAR, Leibniz-Institut für Meereswissenschaften Düsternbrooker Weg 20 Kiel D-24226 Germany Sabine Barles Université Paris Est€– LATTS, Institut Français d’Urbanisme 4 rue Alfred Nobel€– Cité Descartes Champs-sur-Marne 77420 France Sébastien Barot IRD-Bioemco, Bioemco ENS 46 rue d’Ulm 75230 Paris Cedex 05 France Jerzy Bartnicki Norwegian Meteorological Institute P.O. Box 43 Oslo NO-0313 Norway Claus Beier RisØ DTU, National Laboratory for Sustainable Energy Ecosystems Research Programme P.O. Box 358 4000 Roskilde Denmark

xi

List of Contributors

Salim Belyazid Belyazid Consulting and Communication AB, Österportsgatan 5a 21128 Malmö Sweden Leon J. L. van den Berg Radboud University Nijmegen Heyendaalseweg 135 6525 AJ Nijmegen The Netherlands Peter Bergamaschi European Commission Joint Research Centre Institute for Environment via Enrico Fermi 2749 and Sustainability 21027 290 Ispra (VA) Italy Haldis Berge Norwegian Meteorological Institute PO 43 Blindern 0313 Oslo Norway Arthur Beusen Netherlands Environmental Assessment Agency P.O. Box 303 3720 AH Bilthoven The Netherlands Gilles Billen University Pierre & Marie Curie 4 place Jussieu 75005 Paris France Albert Bleeker Energy Research Centre of the Netherlands P.O. Box 1 1755 ZG Petten The Netherlands Jaap Bloem Alterra Wageningen University and Research Centre Soil Science Centre P.O. Box 47 6700 Wageningen The Netherlands Roland Bobbink B-Ware Research Centre Radboud University P.O. Box 9010 9500 GL Nijmegen The Netherlands Pascal Boeckx Ghent University Faculty of Bioscience Engineering

xii

Coupure 653 9000 Gent Belgium Fayçal Bouraoui European Commission Joint Research Centre via Enrico Fermi 2749 21027 Ispra (VA) Italy Lex Bouwman Netherlands Environmental Assessment Agency P.O. Box 303 3720 AH Bilthoven The Netherlands Nils-Axel Braathen OECD 2 rue André-Pascal F-75775 Paris Cedex 16 France Cristina Branquinho Universidade de Lisboa, Faculdade de Ciências Centro de Biologia Ambiental, Campo Grande, Bloco C2, 5º Piso, sala 37 1749–016 Lisboa Portugal Frank Brentrup Yara International, Centre for Plant Nutrition and Environmental Research Hanninghof 35 48249 Duelmen Germany Lutz Breuer Institute for Landscape Ecology and Resources Management Research Centre for BioSystems, Land Use and Nutrition Heinrich-Buff-Ring 26 35392 Giessen Germany Corjan Brink Netherlands Environmental Assessment Agency P.O. Box 303 3720 AH Bilthoven The Netherlands Roy Brouwer VU University Amsterdam Institute for Environmental Studies De Boelelaan 1085 1081 HV Amsterdam Netherlands Michaela Budňáková Ministry of Agriculture of the Czech Republic Těšnov 17 117 05 Praha 1 Czech Republic

List of Contributors

Keith R. Bull Centre for Ecology and Hydrology Lancaster Environment Centre Library Avenue Lancaster LA1 4AP United Kingdom Klaus Butterbach-Bahl Karlsruhe Institute of Technology Institute for Meterology and Climate Research Atmospheric Environmental Research Kreuzeckbahnstrasse 19 82467 Garmisch-Partenkirchen Germany Andrea Butturini University of Barcelona Department of Ecology Faculty of Biology avd. Diagonal 645 8028 Barcelona Spain John Neil Cape Centre for Ecology and Hydrology Bush Estate Penicuik EH26 0QB United Kingdom Ana C. Cardoso European Commission Joint Research Centre Institute for Environment and Sustainability via Enrico Fermi 2749 21027 Ispra (VA) Italy Pierre Cellier INRA, UMR EGC 78850 Thiverval-Grignon France Pavel Čermák Central Institute for Supervising and Testing in Agriculture Hroznová Street 2 656 06 Brno Czech Republic Daniel J. Conley Lund University Department of Earth and Ecosystem Sciences Sölvegatan 12 223 62 Lund Sweden Sarah E. Cornell University of Bristol QUEST, School of Earth Sciences

Queens Road Bristol BS8 1RJ United Kingdom Chris J. Curtis University College London Environmental Change Research Centre Gower Street London WC1E 6BT United Kingdom Cornelis Cuvelier European Commission Joint Research Centre P.O. Box 410 21020 Ispra (VA) Italy Tommy Dalgaard Aarhus University Department of Agroecology and Environment P.O. Box 50 8830 Tjele Denmark Michael Dannenmann University of Freiburg Institute of Forest Botany and Tree Physiology Georges Köhler Allee 53/54 79110 Freiburg Germany Frank Dentener European Commission Joint Research Centre via Enrico Fermi 2749 21027 Ispra (VA) Italy Barbara Deutsch Stockholm University Department of Applied Environmental Science Svanthe Arrheniusväg 8 11418 Stockholm Sweden Roald Dickens Department for the Environment Food and Rural Affairs 17 Smith Square London SW1P 3JR United Kingdom Nancy B. Dise Manchester Metropolitan University Department of Environmental and Geographical Sciences John Dalton East Building, Chester Street Manchester M1 5GD United Kingdom

xiii

List of Contributors

Ulrike M. Doering European Commission Joint Research Centre Institute for Environment and Sustainability P.O. Box 290 21020 Ispra (Va) Italy Anthony Dore Centre for Ecology and Hydrology Bush Estate Penicuik EH26 9HF United Kingdom Ulrike Dragosits Centre for Ecology and Hydrology Bush Estate Penicuik EH26 0QB United Kingdom Jean-Louis Drouet INRA UMR INRA/AgroParisTech Environment and Arable Crops 78850 Thiverval-Grignon France Patrick Durand INRA UMR 1069 SAS 35000 Rennes France Sylvia Duretz INRA UMR EGC 78850 Thiverval-Grignon France Anja Engel Alfred Wegener Institute for Polar and Marine Research Am Handelshafen 12 27515 Bremerhaven Germany Jan Willem Erisman Energy Research Centre of the Netherlands P.O. Box 1 1755 ZG Petten the Netherlands Ipek Erzi TUBITAK Marmara Research Centre Environment Institute P.O. Box 21 41470 Gebze Kocaeli Turkey

xiv

Hilde Fagerli Norwegian Meteorological Institute P.O. Box 43 0313 Blindern Norway David Fernall Department for Environment, Food and Rural Affairs Kingspool, Peasholme Green York YO1 2PX United Kingdom Chris R. Flechard Soils, Agro-hydro systems and Spatialization 65 rue de St-Brieuc 35042 Rennes France David Fowler Centre for Ecology and Hydrology Bush Estate Penicuik EH26 0QB United Kingdom James Galloway University of Virginia P.O. Box 400772 Charlottesville VA 22901 United States of America Raja S. Ganeshram University of Edinburgh School of GeoSciences Grant Institute West Mains Road Edinburgh EH16 5NW United Kingdom Josette Garnier UMR Sisyphe UPMC & CNRS, , 4 place Jussieu 75005 Paris France Markus Geupel Federal Environment Agency, Germany Wörlitzer Platz 1 6844 Dessau Germany Ing-Marie Gren Swedish University of Agricultural Sciences Department of Economics 750 07 Uppsala Sweden

List of Contributors

Peringe Grennfelt IVL Swedish Environmental Research Institute Ltd Aschebergsgatan 44 P.O. Box 5302 400 14 Gothenburg Sweden

Jean-Paul Hettelingh National Institute for Public Health and the Environment Coordination Centre for Effects P.O. Box 1 3720 BA Bilthoven The Netherlands

Hans van Grinsven Netherlands Environmental Assessment Agency P.O. Box 303 3720 AH Bilthoven The Netherlands

Kevin Hicks University of York Stockholm Environment Institute Grimston House Heslington YO10 5DD United Kingdom

Bruna Grizzetti European Commission Joint Research Centre via Enrico Fermi 2749 21027 Ispra (VA) Italy Per Gundersen University of Copenhagen Forest and Landscape Denmark Rolighedsvej 23 1958 Frederiksberg Denmark Steen Gyldenkærne Afdeling for Systemanalyse Danmarks Miljøundersøgelser Frederiksborgvej 399 4000 Roskilde Denmark Roy M. Harrison University of Birmingham School of Geography, Earth and Environmental Sci. Edgbaston Birmingham B15 2TT United Kingdom Anna-Stiina Heiskanen Finnish Environment Institute P.O. Box 140 251 Helsinki Finland Josef Hejzlar Institute of Hydrobiology Biology Centre AS CR Na Sadkach 7 370 05 Ceske Budejovice Czech Republic Ole Hertel University of Aarhus National Environmental Research Institute P.O. Box 358 4000 Roskilde Denmark

Peter Higgins University of Edinburgh Holyrood Road Edinburgh EH8 8AQ United Kingdom Klaas W. van Der Hoek National Institute for Public Health and the Environment P.O. Box 1 3720 BA Bilthoven The Netherlands Robert Hoft Convention on Biological Diversity 413, Saint Jacques Street, suite 800 Montreal QC H2Y 1N9 Canada Mike Holland University of Reading EMRC Whitchurch Hill Reading RG8 7PW United Kingdom Clare M. Howard Centre for Ecology and Hydrology Bush Estate Penicuik EH23 4RB United Kingdom Robert W. Howarth Cornell University Department of Ecology and Evolutionary Biology Corson Hall Ithaca NY 14853 United States of America Christoph Humborg Stockholm University

xv

List of Contributors

Department of Applied Environmental Science Svanthe Arrheniusväg 8 10691 Stockholm Sweden Nicholas J. Hutchings University of Aarhus Research Centre Foulum 8830 Tjele Denmark Andreas Ibrom Risø National Laboratory for Sustainable Energy Frederiksborgvej 399 4000 Roskilde Denmark Hans van Jaarsveld Netherlands Environmental Assessment Agency P.O. Box 303 3720 AH Bilthoven The Netherlands Brian H. Jacobsen University of Copenhagen Institute of Food and Resource Economics Rolighedsvej 25 1958 Frederiksberg Denmark Steve Jarvis The European Journal of Soil Science Centre for Rural Policy Research University of Exeter Amory Building, Rennes Drive Exeter EX4 4RJ United Kingdom Michael E. Jenkin Atmospheric Chemistry Services Okehampton EX20 1FB United Kingdom Lars Stoumann Jensen University of Copenhagen Faculty of Life Sciences Department of Agriculture and Ecology Thorvaldsensvej 40 1871 Frederiksberg C Denmark Timothy Jickells University of East Anglia School of Environmental Sciences Norwich NR4 7TJ United Kingdom

xvi

Penny Johnes University of Reading Aquatic Environments Research Centre Whiteknights Reading RG6 6DW United Kingdom Age W. Jongbloed Wageningen UR Livestock Research Edelhertweg 15 8219 PH Lelystad The Netherlands Åsa Jonsson IVL Swedish Environmental Research Institute P.O. Box 5302 400 14 Göteborg Sweden Øyvind Kaste Norwegian Institute for Water Research Jon Lilletuns vei 3 4879 Grimstad Norway Ralf Kiese Karlsruhe Institute for Technology Institute for meteorology and Climate Research Atmospheric Environmental Research Kreuzeckbahnstrasse 19 82467 Garmisch-Partenkirchen Germany Barbara Kitzler Federal Research and Training Centre for Forests, Natural Hazardo and Landscape Seckendorff-Gudent-Weg 8 1130 Vienna Austria Jeroen de Klein Wageningen University and Research Centre Aquatic Ecology and Water Quality Management Group P.O. Box 47 6700 AA Wageningen The Netherlands Zbigniew Klimont International Institute for Applied Systems Analysis Schlossplatz 1 2361 Laxenburg Austria Pirkko Kortelainen Finnish Environment Institute (SYKE) P.O. Box 140 00251 Helsinki Finland

List of Contributors

Marina Kousoulidou Aristotle University of Thessaloniki Department of Mechanical Engineering Laboratory of Applied Thermodynamics 54124 Thessaloniki Greece Natalia Kozlova North-West Research Institute of Agricultural Engineering and Electrification (SZNIIMESH) P.O.Tiarlevo, Filtrovskoje shosse, 3 196625 Saint-Petersburg-Pavlovsk Russian Federation Michael Kreuzer ETH Zurich Institute of Plant, Animal and Agroecosystem Science Universitätstrasse 2 8092 Zurich Switzerland Johannes Kros Alterra, Wageningen University and Research Centre P.O. Box 47 6700 AA Wageningen The Netherlands Markku Kulmala University of Helsinki Department of Physics P.O. Box 64 14 Helsinki Finland Joachim Lammel Yara International Centre for Plant Nutrition and Environmental Research Hanninghof 35 48249 Duelmen Germany Christiane Lancelot Université Libre de Bruxelles Ecologie des Systèmes Aquatiques ESA, CP 221 Boulevard du Triomphe 1050 Bruxelles Belgium Patrick Lavelle CIAT Km 17, Recta Cali-Palmira Apartado Aéreo 6713 Cali Colombia

Anne-Christine Le Gall INERIS Economics and Decision for the Environment, Chronic Risks Division Parc Technologique Alata, BP2 60550 Verneuil en Halatte France Allison Leach University of Virginia P.O. Box 400123 Charlottesville VA 22904 United States of America Adrian Leip European Commission Joint Research Centre Institute for Environment and Sustainability via E Ferminrico 2749 21027 Ispra (VA) Italy Ahti Lepistö Finnish Environment Institute (SYKE) P.O. Box 140 251 Helsinki Finland Jan Peter Lesschen Wageningen University and Research Centre Alterra P.O. Box 47 6700 AA Wageningen The Netherlands Roos Loeb B-ware Research Centre P.O. Box 6558 6503 GB Nijmegen The Netherlands Benjamin Loubet INRA, INA PG UMR Environm & Grandes Cultures 78850 Thiverval-Grignon France Rob Maas Netherlands Environmental Assessment Agency P.O. Box 303 3720 AH Bilthoven The Netherlands Stephen C. Maberly Centre for Ecology and Hydrology Lancaster Environment Centre Library Avenue Lancaster LA1 4AP United Kingdom

xvii

List of Contributors

Luc Maene International Fertilizer Industry Association 28 rue Marbeuf 75008 Paris France Jakob Magid Copenhagen University Department of Agriculture and Ecology Thorvaldsensvej 40 1873 Copenhagen Denmark Abigail McQuatters-Gollop Sir Alister Hardy Foundation for Ocean Science Citadel Hill Plymouth PL1 2PB United Kingdom Philippe Merot INRA 65 rue de Saint-Brieuc, CS84215, 35042 Rennes France Jack J. Middelburg Utrecht University Faculty of Geosciences Budapestlaan 4 3584 CD Utrecht The Netherlands Jana Moldanová IVL Swedish Environmental Research Institute Ltd Box 5303 400 14 Göteborg Sweden Suvi Monni European Commission Joint Research Centre Institute for Environment and Sustainability via Enrico Fermi 2749 21027 Ispra (VA) Italy Dominic Moran Scottish Agricultural College King’s Buildings Edinburgh EH9 6GU United Kingdom John Munthe IVL Swedish Environmental Research Institute P.O. Box 5302 400 14 Gothenburg Sweden

xviii

Gabriela B. Nardoto Universidade de Brasília Faculdade UnB Planaltina Área Universitária 1 Vila Nossa Senhora de Fátima, Planaltina 73.340–710 Brasília Brazil Eiko Nemitz Centre for Ecology an d Hydrology Bush Estate Penicuik EH26 0QB United Kingdom Rostislav Neveceral Czech Hydrometeorological Institute Na Sabatce 17 14000 Praha Czech Republic Nikolaos P. Nikolaidis Technical University of Crete Department of Environmental Engineering University Campus 73100 Chania Greece Oene Oenema Wageningen University and Research Centre Alterra P.O. Box 47 6700 AA Wageningen The Netherlands Jorgen E. Olesen Aarhus University Department of Agroecology and Environment Blichers Alle 20 8830 Tjele Denmark Mark van Oorschot Netherlands Environmental Assessment Agency P.O. Box 303 3720 AH Bilthoven The Netherlands Lorenzo Orlandini European Commission – DG AGRI Rue de la Loi 130–05/20 1000 Brussels Belgium Christian Pallière Fertilizers Europe Avenue E. Van Nieuwenhuyse 6 1160 Brussels Belgium

List of Contributors

Gilles Pinay University of Birmingham School of Geography Birmingham B15 2TT United Kingdom Hanne Damgaard Poulsen Aarhus University Department of Animal Health and Bioscience P.O. Box 50 8830 Tjele Denmark David Powlson Rothamsted Research Harpenden AL5 2JQ United Kingdom Michel Prud’homme International Fertilizer Industry Association 28 rue Marbeuf 75008 Paris France Ari Rabl ARMINES/Ecoles des Mines de Paris 6 av. Faidherbe 91440 Bures sur Yvette France David S. Reay University of Edinburgh School of Geosciences CECS, High School Yards Edinburgh EH8 9XP United Kingdom Robert M. Rees Scottish Agricultural College West Mains Road Edinburgh EH9 3JG United Kingdom Gert Jan Reinds Wageningen University and Research Centre Alterra P.O. Box 47 6700 AA Wageningen The Netherlands Stefan Reis Centre for Ecology and Hydrology Bush Estate Penicuik EH26 0QB United Kingdom

Hannes Isaak Reuter Gisxperts gbr Eichenweg 42 06849 Dessau Germany Michael O. Rivett University of Birmingham Water Sciences Group Birmingham B15 2TT United Kingdom Trudy G. A. Rood Netherlands Environmental Assessment Agency P.O. Box 303 3721 AH Bilthoven The Netherlands Joost Salomez Flemish Government K. Albert II-laan 20 1000 Brussels Belgium Benjamin Sanchez Gimeno CIEMAT Avda. Complutense 22 28040 Madrid Spain Alberto Sanz-Cobena Technical University of Madrid Av/ Complutense s/n, Ciudad Universitaria 28040 Madrid Spain Martijn Schaap TNO Built Environment and Geosciences P.O. Box 80015 3508 TA Utrecht The Netherlands Doris Schiedek National Environmental Research Institute Frederiksborgvej 399 4000 Roskilde Denmark Jan K. Schjoerring University of Copenhagen Department of Agriculture and Ecology Thorvaldsensvej 40 1871 Frederiksberg C Denmark Uwe A. Schneider KlimaCampus, Hamburg University Research Unit Sustainability and Global Change Bundesstrasse 55

xix

List of Contributors

20146 Hamburg Germany Valiyaveetil Shamsudheen Semeena Norwegian Meteorological Institute P. O. Box 43 0313 Blindern Norway Günther Seufert European Commission Joint Research Centre Institute for Environment and Sustainability P.O. Box 050 21027 Ispra (VA) Italy Jan Siemens University of Bonn Institute of Crop Science and Resource Conservation – Soil Sciences Nussallee 13 53115 Bonn Germany Marie Silvestre CNRS€– FR3020 FIRE 4 place Jussieu 75005 Paris France David Simpson Norwegian Meteorological Institute EMEP MSC-W P.O. Box 43 0313 Blindern Norway Ute Skiba Centre fro Ecology and Hydrology Bush Estate Penicuik EH26 0QB United Kingdom Carsten Ambelas Skjøth Aarhus University P.O. Box 358 4000 Roskilde Denmark Caroline Slomp Utrecht University Department of Earth Sciences Budapestlaan 4, 3584 CD Utrecht The Netherlands Erik Smedberg Stockholm University Baltic Nest Institute

xx

Stockholm Resilience Centre 10691 Stockholm Sweden Keith A. Smith University of Edinburgh Institute of Atmospheric and Environmental Science West Mains Road Edinburgh EH9 3JN United Kingdom Lise Lotte Sørensen Risø National Laboratory for Sustainable Energy P.O. Box 49 4000 Roskilde Denmark Till Spranger Federal Ministry for the Environment, Nature Conservation and Nuclear Safety Stresemannstrasse 128–130 10117 Berlin Germany Carly J. Stevens The Open University Department of Life Sciences Walton Hall Milton Keynes MK7 6AA United Kingdom Mark A. Sutton Centre for Ecology and Hydrology Bush Estate Penicuik EH26 0QB United Kingdom Anastasia Svirejeva-Hopkins Potsdam Institute for Climate Impact Research Telegrafenberg A31 14473 Potsdam Germany Mette Termansen University of Aarhus Department of Policy Analysis Frederiksborgvej 399 4000 Roskilde Denmark Mark Theobald Technical University of Madrid/Centre for Ecology and Hydrology Department of Agricultural Chemistry and Analysis Ciudad Universitaria, s/n 28040 Madrid Spain

List of Contributors

Vincent Thieu UMR 7619 Sisyphe CNRS/UPMC 4 place Jussieu 75005 Paris France Philippe Thunis European Commission Joint Research Centre Institute for Environment and Sustainability via Enrico Fermi 2749 21020 Ispra (VA) Italy Chris Tompkins Independent consultant United Kingdom Robert Upstill-Goddard Newcastle University School of Marine Science and Technology Ridley Building Newcastle-upon-Tyne NE47 9BL United Kingdom Laura Valli CRPA Corso Garibaldi 42 42100 Reggio Emilia Italy Robert Vautard LSCE/IPSL laboratoire CEA/CNRS/VSQ Orme des Merisiers 91191 Gif/Yvette Cedex France Gerard L. Velthof Wageningen University and Research Centre Alterra P.O. Box 47 6700 AA Wageningen The Netherlands Timo Vesala University of Helsinki Department of Physics P.O. Box 48 14 Helsinki Finland Valérie Viaud INRA, UMR 1069 SAS 65 rue de Saint-Brieuc 35000 Rennes France

Massimo Vieno University of Edinburgh School of Geosciences The King’s Buildings Edinburgh EH9 3JN United Kingdom Maren Voss Leibniz-Institute of Baltic Sea Research Warnemuende Seestrasse 15 18119 Rostock Germany Wim de Vries Alterra, Wageningen University and Research Centre Centre Soil, Droevendaalsesteeg 4, Wageningen 6708 PB The Netherlands Jim Webb AEA Energy and Environment Gemini Building, Harwell Business Centre Didcot OX11 0QR United Kingdom Henk J. Westhoek Netherlands Environmental Assessment Agency P.O. Box 303 3720 AH Bilthoven The Netherlands Jaap Willems Netherlands Environmental Assessment Agency P.O. Box 303 3720 AH Bilthoven The Netherlands Wilfried Winiwarter International Institute for Applied Systems Analysis Schlossplatz 1 2361 Laxenburg Austria Peter Witzke EuroCARE GmbH Nussallee 21 53115 Bonn Germany Richard F. Wright Norwegian Institute for Water Research Gaustadalleen 21 349 Oslo Norway Sönke Zaehle Max Planck Institute for Biogeochemistry

xxi

List of Contributors

Biogeochemical Systems Department Hans-Knöll-Strasse 10 07745 Jena Germany Sophie Zechmeister-Boltenstern Federal Research and Training Centre for Forests Natural Hazards and Landscape Seckendorff Gudent Weg 8

xxii

1131 Vienna Austria John F. Zevenbergen TNO Defence, Security and Safety Lange Kleiweg 137 2288 GJ Rijswijk The Netherlands

Foreword

Addressing the grand challenges of society depends fundamentally on firm scientific evidence. Today, Europe faces several of these challenges, as outlined in the Europe 2020 strategy adopted by the Commission on 3 March 2010, including climate change, energy and food security, health and an ageing population. Research and innovation are crucial to address these challenges effectively. For that reason, the Commission launched the ‘Innovation Union’ flagship initiative, with the aim to re-focus research and development as well as innovation policy on these grand societal challenges. In this framework we very much welcome the European Nitrogen Assessment. It is fair to say that nitrogen will be a new story for many people. Yet we can here clearly identify a case of science at its best: innovative thinking that enables the development of connections from evidence-based policies to evidence-tested decisions. The Assessment highlights how human production of reactive nitrogen has literally changed the world. Since the invention of the Haber-Bosch process a century ago, humans have been able to double the world’s circulation of nitrogen compounds, resulting in nitrogen fertilizers sustaining around 3 billion people, almost half of the world population. It is therefore obvious that nitrogen is essential, not only to meeting the challenge for food security, but, with the increasing importance of biofuels, also for energy security. Yet with this achievement, originating from European innovation a century ago, has also come an inheritance of environmental effects that cuts across all global ecosystems. As the Assessment reveals, excess reactive nitrogen contributes to climate change; it adversely affects water, air and soil quality, and is putting unsustainable pressure on ecosystems and biodiversity in Europe. Moreover, the surplus of nitrogen compounds leaking into air and water may lead to a substantial health risk for vulnerable human populations.

The Assessment highlights how nitrogen is related to each of the great challenges that European society faces, and the need to develop joined up approaches to address them. In this respect the European Nitrogen Assessment is an important step, building scientific and institutional bridges and sharing different perspectives. It is rewarding to see different environmental disciplines being brought together, and scientists proactively seeking to engage European industry, policy makers and the public. These significant commitments also emphasize the importance of critical mass in the European Research Area. The Assessment is a key output from a large amount of ongoing research in Europe and elsewhere, but in particular from the NitroEurope Integrated Project supported by the European Commission’s 6th Framework Programme and the Nitrogen in Europe (NinE) Research Networking Programme of the European Science Foundation. With the involvement of Action 729 of the COST Programme, the necessary expertise has been gathered to drive the Assessment. The message of 200 leading European experts from different disciplines and perspectives is surely that we need to take steps forward. Only by joining forces to face the societal challenges will European research provide the scientific basis and the evidence needed for solutions. If European innovation has handed us down a nitrogen inheritance, threatening the environment as a price for a solution to nourish the growing world population, it is only right that European science should lead the way in responding to the challenge. Robert-Jan Smits Director General for Research, European Commission Professor Marja Makarow Chief Executive, European Science Foundation

xxiii

Summary for policy makers Lead authors:€Mark A. Sutton and Hans van Grinsven Contributing authors:€Gilles Billen, Albert Bleeker, Alexander F. Bouwman, Keith Bull, Jan Willem Erisman, Peringe Grennfelt, Bruna Grizzetti, Clare M. Howard, Oene Oenema, Till Spranger and Wilfried Winiwarter

Main messages Too much nitrogen harms the environment and the economy • Over the past century humans have caused unprecedented changes to the global nitrogen cycle, converting atmospheric di-nitrogen (N2) into many reactive nitrogen (Nr) forms, doubling the total fixation of Nr globally and more than tripling it in Europe. • The increased use of Nr as fertilizer allows a growing world population, but has considerable adverse effects on the environment and human health. Five key societal threats of Nr can be identified:€to water quality, air quality, greenhouse balance, ecosystems and biodiversity, and soil quality. • Cost–benefit analysis highlights how the overall environÂ� mental costs of all Nr losses in Europe (estimated at €70–€320 billion per year at current rates) outweigh the direct economic benefits of Nr in agriculture. The highest societal costs are associated with loss of air quality and water quality, linked to impacts on ecosystems and especially on human health.

Nitrogen cascade and budgets • The different forms of Nr inter-convert through the environment, so that one atom of Nr may take part in many environmental effects, until it is immobilized or eventually denitrified back to N2. The fate of anthropogenic Nr can therefore be seen as a cascade of Nr forms and effects. The cascade highlights how policy responses to different Nr forms and issues are inter-related, and that a holistic approach is needed, maximizing the abatement synergies and minimizing the trade-offs. • Nitrogen budgets form the basis for the development and selection of measures to reduce emissions and their effects in all environmental compartments. For instance, the

xxiv

European nitrogen budget highlights the role of livestock in driving the European nitrogen cycle.

Policies and management • Existing policies related to Nr have been largely established in a fragmented way, separating Nr forms, media and sectors. Despite the efforts made over many years to reduce Nr inputs into the environment, most of the Nr-related environmental quality objectives and environmental action targets have not been achieved to date. • The five societal threats and N budgets are starting points for a more-holistic management of Nr. The Assessment identifies a package of 7 key actions for overall management of the European nitrogen cycle. These key actions relate to:€Agriculture (3 actions), Transport and Industry (1 action), Waste water treatment (1 action) and Societal consumption patterns (2 actions). • The key actions provide an integrated package to develop and apply policy instruments. The need for such a package is emphasized by cost–benefit analysis that highlights the role of several Nr forms especially nitrogen oxides (NOx), ammonia (NH3) and Nr loss to water, in addition to nitrous oxide (N2O), in the long term.

International cooperation and communication • Tackling Nr necessitates international cooperation. There are various options to implement multi-lateral environmental agreements; a possible inter-convention agreement on nitrogen needs to be further explored. • Communication tools for behavioural change should be extended to nitrogen, such as calculating nitrogen ‘food-prints’. Messages should emphasize the potential health co-benefits of reducing the consumption of animal products to avoid excess above recommended dietary guidelines.

Summary for policy makers

1.╇ Why nitrogen? Concerns and the need for new solutions 1.╇ Nitrogen is an abundant element on earth, making up nearly 80% of the earth’s atmosphere. However, as atmospheric di-Â� nitrogen (N2), it is unreactive and cannot be assimilated by most organisms. By contrast there are many reactive nitrogen (Nr) forms that are essential for life, but are naturally in very short supply. These include ammonia, nitrates, amino acids, proteins and many other forms. Until the mid nineteenth century, limited availability of these Nr compounds in Europe severely constrained both agricultural and industrial productivity [1.1, 2.1].1 2.╇ With increasing population in the late nineteenth century, rates of biological nitrogen fixation were not sufficient for crop needs and Europe became increasingly dependent on limited sources of mined Nr (guano, saltpetre, coal). At the start of the twentieth century, several industrial processes were developed to fix N2 into Nr, the most successful being€the Haber–Bosch process to produce ammonia (NH3) [1.1, 2.1]. 3.╇ Since the 1950s, Nr production has greatly increased, representing perhaps the greatest single experiment in global geoengineering [1.1]. Europe’s fertilizer needs have been met, as well as its military and industrial needs for Nr [3.2, 3.5]. In addition, high temperature combustion processes have substantially increased the formation and release of nitrogen oxides (NOx) [2.4]. While the Nr shortage of the past has been solved, Europe has stored up a nitrogen inheritance of unexpected environmental effects [1.1]. 4.╇ Europe remains a major source region for Nr production, with many of the environmental impacts being clearly visible and well studied. There is a wealth of evidence on sources, fate and impacts of Nr. However, the complexity and extent of the interactions mean that scientific understanding has become scattered and focused on individual sectors. A parallel fragmentation can be seen in environmental policies related to nitrogen, which are typically separated by media (air, land, water, etc.), by issue (climate, biodiversity, waste etc) and by Nr form [4.4, 5.3]. 5.╇ While this specialization has advanced understanding, European science and policies related to nitrogen have to a significant degree lost sight of the bigger picture. The occurrence of Nr in many different Nr forms and media, means that each component should not be considered in isolation. A more comprehensive understanding of the nitrogen cycle is therefore needed to minimize the adverse effects of Nr in the environment, while optimizing food production and energy use [5.3].

2.╇ Role and approach of the European Nitrogen Assessment 6.╇ A key challenge is to synthesize the science and understanding of nitrogen into a form that is useful to governments and society. This involves bringing the different Nr forms, disciplines and stakeholders together. 1

References in this summary (e.g., [1.1, 11.1]) refer to chapter and section numbers of the European Nitrogen Assessment.

7.╇ The European Nitrogen Assessment (ENA) was established in response to these needs. It was coordinated by the Nitrogen in Europe (NinE) programme of the European Science Foundation, drawing on underpinning research from across Europe, but especially the NitroEurope Integrated Project co-funded by the European Commission, with input from the COST Action 729. The Assessment provides a European contribution to the International Nitrogen Initiative (INI) [1.3]. 8.╇ The lead policy audience for the Assessment is the Geneva Convention on Long-range Transboundary Air Pollution (CLRTAP), established under the auspices of the United Nations Economic Commission for Europe (UNECE). Through its Task Force on Reactive Nitrogen, the Convention has formally adopted the Assessment as a contributing activity to its work [1.3]. 9.╇ In addition to supporting CLRTAP, the Assessment is targeted to provide scientific and policy support to the European Union and its Member States, as well as other multi-lateral environmental agreements, including the Global Partnership on Nutrient Management facilitated by UNEP [1.5]. 10.╇ Recognizing these needs, the goal of the European Nitrogen Assessment was established:€to review current scientific understanding of nitrogen sources, impacts and interactions across Europe, taking account of current policies and the economic costs and benefits, as a basis to inform the development of future policies at local to global scales [1.4]. 11.╇ The Assessment process was conducted through a series of five open scientific workshops between 2007 and 2009. Draft chapters were submitted to internal and external peer review [1.3].

3.╇ Disruption of the European nitrogen cycle Fertilizers, energy and transport:€drivers for increased nitrogen inputs 12.╇ Production of Nr is a key input for agriculture and industry, and a persistent side-effect of combustion for energy and transport. Industrial production in Europe of Nr in 2008 was about 34 Tg per year (where 1 Tg = 1 million tonnes) of which 75% is for fertilizer and 25% for chemical industry (production of rubbers, plastics, and use in electronic, metals and oil industry) [3.5]. The trend in mineral fertilizer represents the largest change in overall Nr inputs to Europe over the past century (Figure SPM.1). 13.╇ The combustion of fossil fuels has allowed a substantial increase in industrial production and transportation, reflected in the greatly increased emission of nitrogen oxides, which only over the last 20 years have partly been controlled. By contrast, the total contribution of crop biological nitrogen fixation has decreased significantly. 14.╇ The provision of Nr from the Haber–Bosch process removed a major limiting factor on society, permitting substantial population growth and improving human welfare.

xxv

Summary for policy makers

Figure SPM.1 Estimated trend of anthropogenic reactive nitrogen inputs to the European Union (EU-27) [5.1] (1 Tg equals 1 million tonnes).

Figure SPM.2 Simplified view of the N-cascade, highlighting the capture of atmospheric di-nitrogen (N2) to form reactive nitrogen (Nr) by the Haber–Bosch process€– the largest source of Nr in Europe. The main pollutant forms of Nr (orange boxes) and five environmental concerns (blue boxes) are summarized. Blue arrows represent intended anthropogenic Nr flows; all the other arrows are unintended flows [1.2]. For fuller description including other Nr sources, see [5.2].

However, accounting for natural sources, humans have more than doubled the supply of Nr into the environment globally [1.1], and more than tripled this supply in Europe (Figure SPM.3) [16, supplementary material]. 15.╇ As of the year 2000, Europe creates about 19 Tg per year of Nr, of which 11 Tg per year is from chemical fertilizers, 3.4 Tg per year from combustion sources, 3.5 Tg per year from food and feed import and 1 Tg per year by crop biological N-fixation (BNF) (Figure SPM.3).

The nitrogen cascade 16.╇ Human production of Nr from N2 causes a cascade of intended and unintended consequences. The intended cascade is that each molecule of Nr contributes to soil fertility and increased yields of crops, subsequently feeding livestock and humans, allowing the formation of amino acids, proteins and

xxvi

DNA. In a well managed system, the intention is for the Nr in manures and sewage to be fully recycled back through the agricultural system (blue arrows in Figure SPM.2). 17.╇ Reactive nitrogen, is however, extremely mobile, with emissions from agriculture, combustion and industry leading to an unintended cascade of Nr losses into the natural environment (Figure SPM.2). Once released, Nr cascades through the different media, exchanging between different Nr forms and contributing to a range of environmental effects, until it is finally denitrified back into N2. An important consequence of the cascade is that the environmental impacts of Nr eventually become independent of the sources, so that nitrogen management requires a holistic approach. This is important, both to minimize ‘pollution swapping’ between different Nr forms and threats, and to maximize the potential for synergies in mitigation and adaptation strategies [2.6, 5.2].

Summary for policy makers Europe (EU27), around 1900. N fluxes in TgN/yr

Europe (EU27), around 2000. N fluxes in TgN/yr atmospheric N2 pool

atmospheric N2 pool

2.1

3.5

9.7 crop N2 fix

atmosph. NH3NOxN2O N2 fix industry & traffic

0.6 1.9

atmosph. deposition Fertilizers

1.9

4

9.6 N2 fix industry & traffic

crop production

gaseous losses

3.4

3.8

0.2 2.1 4

crop N2 fix

atmosph. NH3NOxN2O

Livestock & human nutrition

Net atmospheric export

2.4

atmosph. deposition

3.8

1 Fertilizers

Net import of food & feed

17.6 crop production

Livestock & human nutrition

11.2 Soils

Soils Losses to water

2.3

Export by rivers to the sea

gaseous losses

13.5

Losses to 4.5 water

Export by rivers to the sea

Figure SPM.3 Simplified comparison of the European nitrogen cycle (EU-27) between 1900 and 2000. Blue arrows show intended anthropogenic nitrogen flows; orange arrows show unintended nitrogen flows; green arrows represent the nearly closed nitrogen cycle of natural terrestrial systems [16.4 and 16 supplementary material].

A new nitrogen budget for Europe 18.╇ One of the tasks addressed in the European Nitrogen Assessment has been to construct a comprehensive nitrogen budget for Europe (EU-27 for the year 2000), considering each of the major flows in the nitrogen cascade [16.4]. In parallel, the estimates have also been compared with 1900 [16, supplementary material]. By combining all the nitrogen flows, such budgets provide improved perspective on the major drivers and the most effective control options. 19.╇ Figure SPM.3 summarizes the European nitrogen budget in its simplest form [derived from 16.4]. The budget for 2000 shows that overall human perturbation of the nitrogen cycle is driven primarily by agricultural activities. Although the atmospheric emissions of NOx from traffic and industry contribute to many environmental effects, these emissions are dwarfed by the agricultural Nr flows. 20.╇ It is important to note the magnitude of the European Nr flow in crop production, which is mainly supported by Nr fertilizers. The primary use of the Nr in crops, however, is not directly to feed people:€80% of the Nr harvest in European crops provides feeds to support livestock (8.7 Tg per year plus 3.1 Tg per year in imported feeds, giving a total of 11.8╛Tg per year). By comparison, human consumption of Nr is much smaller, amounting to only 2 Tg per year in crops and 2.3 Tg per year in animal products. Human use of livestock in Europe, and the consequent need for large amounts of animal feed, is therefore the dominant human driver altering the nitrogen cycle in Europe [16.4]. 21.╇ These major intended alterations in Nr flows cause many additional unintended Nr flows (Figure SPM.3). Overall, NH3 from agriculture (3.2 Tg per year) contributes a similar

amount to emissions of Nr to the atmosphere as NOx (3.4 Tg per year). Agriculture also accounts for 70% of nitrous oxide (N2O) emissions in Europe, with total N2O emissions of 1 Tg per year. The food chain also dominates Nr losses to ground and surface waters, mainly as nitrates (NO3), with a gross load of 9.7 Tg resulting mainly from losses due to agriculture (60%) and discharges from sewage and water treatment systems (40%) [16.4]. 22.╇ The comparison between 1900 and 2000 shows how each of these flows have increased, including denitrification back to N2. Denitrification is the largest and most uncertain loss, as it occurs at many different stages during the continuum from soils to freshwaters and coastal seas. Although emissions of N2 are environmentally benign, they represent a waste of the substantial amounts of energy put into human production of Nr, thereby contributing indirectly to climate change and air pollution. This is in addition to the impact on climate change of N2O formed especially as a byproduct of denitrification.

Achievements and limitations of current policies 23.╇ Peak production of Nr in Europe occurred in the 1980s, which was linked to agricultural over-production and lack of emissions regulations. Since that time, the introduction of policies and other changes affecting agriculture (including the Common Agricultural Policy, Nitrates Directive and the restructuring of Eastern Europe after 1989), as well as stringent emission controls, e.g., for large combustion plants (EC Large Combustion Plants Directive, UNECE Sofia Protocol and Gothenburg Protocol, etc.) and the EURO standards for

xxvii

Summary for policy makers

Figure SPM.4 Estimated trends in European reactive nitrogen emissions between 1900 and 2000 (EU-27) [5.1].

road transport vehicles, have led to decreases in the emissions (Figure€SPM.4) [4.4]. 24.╇ Overall, emissions of combustion NOx have reduced by ~30% since 1990, but much greater NOx reductions per unit output have been achieved. These have been offset by an increase in traffic and energy consumption. The net emission reduction is therefore a clear example of decoupling, as emissions would have increased by over 30% if no measures had been implemented. The extent of success of the technical measures can be in part attributed to the involvement of a small number of players (e.g., electricity supply industry, vehicle manufacturers) and the fact that the costs of these measures could be easily transferred to consumers [4.5]. 25.╇ Agricultural measures have resulted in only a modest reduction in total agricultural Nr inputs for the EU-27 of ~15% (Figure SPM.1). This small overall reduction is reflected in the trends in NH3 emissions (Figure€ SPM.4). Most of the reductions that have been achieved to date can be attributed to reductions in fertilizer use and livestock numbers, especially in Eastern€ Europe after 1989. Although management improvements will have contributed to reduced emissions (e.g., nitrate leaching and loss to marine areas), there has as yet been little quantitative achievement of measures to reduce N2O and NH3 emissions from agriculture on a European scale. The fact that current€ Nr emission reduction policies in agriculture (e.g., Nitrates Directive, Oslo and Paris Commission for the protection€of the North East Atlantic, UNECE Gothenburg Protocol and National Emissions Ceilings Directive) have only made limited progress can be linked in part to the large number of diverse actors (including many small farms), the diffuse nature of the Nr emission sources, and the challenge of passing any perceived costs onto consumers [4.5]. As a consequence, agriculture is the sector with the largest remaining emission reduction potential. 26.╇ Several instances of pollution swapping in Nr control have been observed. These include the introduction of three way catalysts in vehicles, which increased NH3 and N2O emissions (although overall Nr emissions were still greatly reduced), and the implementation of the Nitrates Directive, prohibiting wintertime manure spreading, which has led to a new peak in springtime NH3 emisssions€[9.2].

xxviii

4.╇ The benefits and efficiency of nitrogen in agriculture Nitrogen fertilizers feed Europe 27.╇ There is no doubt that human production of Nr has greatly contributed to the increase in productivity of agricultural land. Without anthropogenic Nr, a hectare of good agricultural land in Europe, with no other growth limitations, can produce about 2 tonne per ha of cereal annually. With typical additional inputs from biological nitrogen fixation (BNF), it can produce about 4–6 tonne per ha, and with addition of chemical fertilizer about 8–10 tonne per ha. Synthetic Nr fertilizer has been estimated to sustain nearly 50% of the world’s population, and is essential for the EU to be largely self-sufficient in cereals. For pork, poultry and egg production, Europe strongly depends on soybean imports from America [3.1]. 28.╇ Agronomic efficiency provides an indicator of the Nr-benefit to the farmer (kg crop production per kg applied N). Typically, fertilizer rates in the eastern EU Member States are up to four times lower than in the 15 ‘old’ Member States, but agronomic efficiencies are comparable (Figure SPM.5). The use of Nr is profitable as there is a robust financial return of €2–5 on every euro invested in Nr fertilizer, depending on the market price of cereals and fertilizer [3.6].

Grain and meat production considerably differ in their Nr losses to the environment

29.╇ The nitrogen recovery (kg N taken up by a crop per kg applied N) provides a measure of environmental N-loss in crop production. For cereals it varies 30%–60% across Europe, indicating that 40%–70% of the fertilizer Nr applied is lost to the atmosphere or the hydrosphere [3.2]. 30.╇ The nitrogen recovery in animal farming is inherently lower than in crops, with only 10–50% of Nr in feed being retained in liveweight and 5%–40% in the edible weight (Figure SPM.6). Accounting for the additional Nr losses in feed production, the overall efficiency of Nr use for meat production is around half these values. For this reason, the full chain of animal protein production

Summary for policy makers

Cereal yield (kg / ha)

10000 8000

NL

DE

FR SI

IE AT DK HU

6000 RO

4000

SK SE

BG

PL EE

LV

2000 0

BE

CY

05

0

FI

IT

CZ

UK

GR LT ES

PT

100

150

200

Nitrogen application (kg N / ha)

Figure SPM.5 Variation of nitrogen fertilizer use on winter wheat across the European Union (EU 15: blue, EU 12: red) around the year 2000. The variation indicates that there is substantial scope to increase performance and reduce environmental effects [3.2].

generates much more losses to the environment than plant protein production. 31.╇ About one third (7.1 Tg per year in 2000) of the total farm input of Nr to soil comes from animal manures. This represents about two thirds of the Nr from animal feeds, while the fraction of Nr in animal manures that is lost to the environment is typically double that of mineral Nr fertilizer, highlighting the importance of proper measures to maximize the effectiveness of manure reuse [3.2].

Variation in nitrogen use efficiency highlights the potential for solutions 32.╇ The overall efficiency of European agriculture (ratio of N in food produced to the sum of synthetic N fertilizer used plus food and feed imports) is about 30% since 2000 [derived from 16.4, see Figure SPM.3]. The wide variety in N application rates and nitrogen use efficiency across Europe indicates that there is a huge scope to improve resource efficiency and reduce environmental effects (Figure SPM.5). 33.╇ In the EU, protein consumption exceeds recommended intake by 70% [26.3] and the share of animal proteins in this total is increasing. Even a minor change in human diet, with less animal protein consumption (or protein from more efficient animals), would significantly affect the European nitrogen cycle.

5.╇ The key societal threats of excess nitrogen 34.╇ From a longer list of around 20 concerns, the Assessment identifies five key societal threats associated with excess Nr in the environment:€ Water quality, Air quality, Greenhouse balance, Ecosystems and biodiversity, and Soil quality. Together, these threats can be easily remembered by an acronym as the ‘WAGES’ of excess nitrogen, and visualized by analogy to the four ‘elements’ (water, air, fire, earth) and quintessence of classical Greek cosmology (Figure SPM.7). These five threats provide a framework that incorporates almost all issues related to the longer list of concerns associated with excess Nr [5.4].

Figure SPM.6 Range of Nr recovery efficiencies in farm animal production in Europe (kg N in edible weight per kg N in animal feed) [3.4, 10.4, 26.3], see also supplementary material for Chapter 3. A higher recovery efficiency is indicative of a smaller nitrogen footprint. Accounting for the full chain from fertilizer application to Nr in edible produce, overall nitrogen use efficiency in animal production for the EU-27 is around 15%–17% [3, 10, supplementary material]. While intensive systems tend to have a higher Nr recovery, they also tend to have larger Nr losses per ha unless efforts are taken to reduce emissions [10.4].

Nitrogen as a threat to European water quality 35.╇ Water pollution by Nr causes eutrophication and acidification in fresh waters [7.4, 8.8]. Estuaries, their adjacent coastlines and (near) inland seas are also affected by eutrophication from Nr with inputs to the coastal zone being four times the natural background [13.7]. Biodiversity loss, toxic algal blooms and dead zones (fish kill) are examples of effects [8.8]. Nitrate levels in freshwaters across most of Europe greatly exceed a threshold of 1.5 to 2 mg Nr per litre, above which waterbodies may suffer biodiversity loss [7.5, 17.3]. 36.╇ High nitrate concentrations in drinking water are considered dangerous for human health, as they might cause cancers and (albeit rarely) infant methaemoglobinaemia. About 3% of the population in EU-15 is potentially exposed to levels exceeding the standard for drinking water of 50 mg NO3 per litre (11.2 mg Nr per litre) and 6% exceeding 25 mg NO3 per litre [17.3]. This may cause 3% increase of incidence of colon cancer, but nitrate is also considered to be beneficial to cardiovascular health [22.3]. 37.╇ Although aquatic eutrophication has decreased to some extent since the 1980s, agreed international policies have not been fully implemented. In addition, increasing nitrate in groundwaters threatens the long-term quality of the resource, due to long residence times in aquifers [7.5,€17.2]. Achieving substantial progress at the European scale requires integration of sectoral policies, reducing overall inputs of Nr to watersheds [4.5, 13.7, 17.5].

Nitrogen as a threat to European air quality 38.╇ Air pollution by nitrogen oxides (NOx) and ammonia (NH3) causes formation of secondary particulate matter (PM), while emissions of NOx also increase levels of nitrogen dioxide

xxix

Summary for policy makers

42.╇ Overall, European Nr emissions are estimated to have a net cooling effect on climate of −16 mW per m2, with the uncertainty bounds ranging from substantial cooling to a small net warming (−47 to +15 mW per m2). The largest uncertainties concern the aerosol and Nr fertilization effects, and the estimation of the European contributions within the global context [19.6]. The estimate of the Intergovernmental Panel on Climate Change (IPCC) for indirect N2O emissions from Nr deposition is considered to be an underestimate by at least a factor of 2 [6.6, 19.6]. 43.╇ There are many opportunities for ‘smart management’, increasing the net cooling effect of Nr by reducing warming effects at the same time as other threats, e.g., by linking N and C cycles to mitigate greenhouse gas emissions through improved nitrogen use efficiency [19.6].

Nitrogen as a threat to European terrestrial ecosystems and biodiversity Figure SPM.7 Summary of the five key societal threats of excess reactive nitrogen, drawn in analogy to the ‘elements’ of classical Greek cosmology. The main chemical forms associated with each threat are shown [5.4]. Photo sources: Shutterstock.com and garysmithphotography.co.uk.

(NO2) and tropospheric ozone (O3). All of these are causes for respiratory problems and cancers for humans, while ozone causes damage to crops and other vegetation, as well as to buildings and other cultural heritage [18.2, 18.5]. 39.╇ Models estimate that PM contributes to 300–400 thousand premature deaths annually in Europe leading to a reduction in life expectancy due to PM of 6–12 months across most of central Europe. Nr contributes up to 30%–70% of the PM by mass [18.3, 18.5]. However, the individual contributions of NOx- and Nr-containing aerosol to human health effects of air pollution remain uncertain [18.2]. 40.╇ Although NOx emission decreases have reduced peak O3 concentrations, background tropospheric O3 concentrations continue to increase. By comparison to the limited progress in reducing NOx emissions, there has been even less success in controlling agricultural NH3 emissions, which therefore contribute to an increasing share of the European air pollution burden [4.5, 18.6].

Nitrogen as a threat to European greenhouse balance 41.╇ Reactive nitrogen emissions have both warming and cooling effects on climate. The main warming components are increasing concentrations of nitrous oxide (N2O) and tropospheric ozone, which are both greenhouse gases. The main cooling effects are atmospheric Nr deposition presently increasing CO2 removal from the atmosphere by forests, and the formation of Nr containing aerosol, which scatter light and encourage cloud formation [19].

xxx

44.╇ Atmospheric Nr deposition encourages plants favouring high Nr supply or more acidic conditions to out-compete a larger number of sensitive species, threatening biodiversity across Europe. The most vulnerable habitats are those with species adapted to low nutrient levels or poorly buffered against acidification. In addition to eutrophication, atmospheric Nr causes direct foliar damage, acidification and increased susceptibility to pathogens [20.3]. 45.╇ Although there are uncertainties in the relative effects of atmospheric nitrate (NO3−) versus ammonium (NH4+), gaseous ammonia (NH3) can be particularly harmful to vegetation, causing foliar damage especially to lower plants [20.3]. This emphasizes the threat to semi-natural habitats occurring in agricultural landscapes [9.6, 11.5]. While uncertain, Nr deposition is expected to act synergistically with climate change and ground-level ozone [20.2]. 46.╇ Thresholds for atmospheric concentrations and deposition of Nr components to semi-natural habitats are exceeded across much of Europe, and will continue to be exceeded under current projections of Nr emissions. In order to achieve ecosystem recovery, further reductions of NH3 and NOx emissions are needed [20.5]. Due to cumulative effects of Nr inputs and long time-lags, rates of ecosystem recovery are expected to be slow, and in some cases may require active management intervention in the affected habitats [20.5].

Nitrogen as a threat to European soil quality 47.╇ Soil integrates many of the other Nr effects, highlighting their interlinked nature. The major Nr threats on soil quality are soil acidification, changes in soil organic matter content and loss of soil biodiversity. Soil acidification can occur from the deposition of both oxidized and reduced Nr, resulting from NOx and NH3 emissions, reducing forest growth and leading to leaching of heavy metals [21.3]. High levels of Nr deposition to natural peatlands risk losing carbon stocks through interactions with plant species changes, although this effect is poorly quantified [6.6, 19.4].

Summary for policy makers

48.╇ Addition of Nr typically has a beneficial effect in agricultural soils, enhancing fertility and soil organic matter [6.4 , 21.3]. However, Nr losses increase, while some soil fungi and N-fixing bacteria are reduced by high N availability. The interactions between Nr and soil biodiversity, soil fertility and Nr emissions are not well understood [21.3]. 49.╇ European forest soils are projected to become less acidic within a few decades, mainly as a result of reduced SO2 and NOx emissions. Ammonia emissions have only decreased slightly and NHx is increasingly dominating soil acidification effects over large parts of Europe [20.3, 21.4].

6.╇ The economics of nitrogen in the environment Estimated loss of welfare due to nitrogen emissions in Europe 50.╇ The social costs of the adverse impacts of Nr in the European environment are estimated. Expressed as € per kg of Nr emission, the highest values are associated with air pollution effects of NOx on human health (€10–€30 per kg), followed by the effects of Nr loss to water on aquatic ecosystems (€5–€20 per kg) and the effects of NH3 on human health through particulate matter (€2–€20 per kg). The smallest values are estimated for the effects of nitrates in drinking water on human health (€0–€4 per kg) and the effect of N2O on human health by depleting stratospheric ozone (€1–€3 per kg) [22.6]. 51.╇ Combining these costs with the total amount of emissions for each main Nr form, provides a first estimate of the annual Nr-related damage in EU-27 (Figure SPM.8). The overall costs are estimated at €70–€320 billion per year, of which 75% is related to air pollution effects and 60% to human health. The total damage cost equates to €150–€750 per person, or 1–4% of the average European income [22.6] and is about twice as high as the present ‘Willingness to Pay’ to control global warming by carbon emissions trading [22.6]. 52.╇ Environmental damage related to Nr effects from agriculture in the EU-27 was estimated at €20–€150 billion per year. This can be compared with a benefit of N-fertilizer for farmers of €10–€100 billion per year, with considerable uncertainty about long-term N-benefits for crop yield [22.6]. 53.╇ Apart from the uncertainties inherent in valuing the environment, including the use of ‘willingness to pay’ approaches for ecosystem services, the main uncertainties in these estimates concern the relative share of Nr in PM to human health effects and of Nr to freshwater eutrophication effects [22.6].

Future European nitrogen mitigation and scenarios 54.╇ Internalizing the environmental costs for N-intensive agriculture in North Western Europe provides economically optimal annual Nr application rates that are about 50 kg per ha (30%) lower than the private economic optimum rate for the

Figure SPM.8 Estimated environmental costs due to reactive nitrogen emissions to air and to water in the EU-27 [22.6].

farmer. This highlights the importance of increasing nitrogen use efficiency and accounting for external effects on the environment in providing N-recommendations to farmers [22.6]. 55.╇ The results also highlight the small overall cost due to N2O emissions compared with NOx, NH3 emissions and Nr losses to water (Figure SPM.8). Although unit costs of N2O, at €6–€18 per kg Nr emitted, are similar to the other issues, N2O emissions are much smaller (para. 21), so that total European damage costs due to N2O are much less than from the other Nr forms. Based on the ‘willingness to pay’ approach and current values, this indicates that the highest policy priority be put on controlling European NOx and NH3 emissions to air and Nr losses to water, as compared with the control of N2O emissions. It is important to target measures that have maximum synergy, reducing emissions of all Nr forms and impacts simultaneously. However, where some measures involve limited trade-offs between Nr (‘pollutant swapping’), Figure SPM.8 indicates that further control of NOx, NH3 and Nr to water would be justified economically even if a proportionate percentage increase in N2O emission were to occur. 56.╇ Estimated costs of technical measures to reduce emissions of NOx, NH3 and N2O are available in the IIASA GAINS model. Based on these estimates, future scenarios up to 2030 compare current reduction plans with maximum feasible reduction and a cost optimization approach. This comparison indicates substantial scope for further reductions in NOx and NH3 emissions, supporting the case for revision of the Gothenburg Protocol [24.6]. Although not assessed here, preliminary indications suggest that costs of NH3 abatement measures (€ per kg€Nr) are cheaper than previously estimated, being the subject of ongoing review.2 United Nations Economic Commission for Europe (2010), Options for Revising the 1999 Gothenburg Protocol to Abate Acidification, Eutrophication and Ground-level Ozone:€Reactive Nitrogen (ECE/ EB.AIR/WG.5/2010/13).

2

xxxi

Summary for policy makers 6000 5000

NH3_R26 NH3_R45

Gg N/yr

4000

Figure SPM.9 Nitrogen emission scenarios for the EU-27, following the Representative Concentration Pathways (RCP) for three different storylines on radiative forcing. The storyline names indicate the radiative forcing exerted in 2100, between 2.6 (R26), 4.5 (R45) and 8.5 (R85) W per m2 [24.6].

NH3_R85 NOX_R26

3000

NOX_R45 NOX_R85

2000

N2O_R26 1000 0 2000

2020

2040

2060

2080

57.╇ Future long-term scenarios emphasize the possibility for major reductions in NOx emissions (by 75% or more for 2000 to 2100), due to improved technologies combined with projected decreases in energy use for some scenarios (Figure SPM.9). By contrast, the anticipated trends for NH3 and N2O are much less clear. A high CO2 scenario representing unrestricted development (+8.5 W/m2 radiative forcing) indicates an increase in NH3 emissions, which does not occur with the more optimistic climate scenarios (+2.6 and +4.5 W/m2 radiative forcing). But even these scenarios highlight a long-term outlook where NH3 quickly becomes the dominant form of Nr emission to the atmosphere, and a key challenge for control policies [24.6]. 58.╇ The long term outlook for scenarios of Nr use and emissions must also consider the possible extent of future renewable energy production. There is potential for substantial synergy in increased forest cover, where the main Nr input is atmospheric deposition, allowing increased scavenging of air pollutants and a contribution to carbon sequestration [9.4, 19.4]. By contrast, the increased use of fertilizer Nr to support intensively managed bioenergy and biofuel crops can involve significant tradeoffs, requiring that additional N2O, other Nr and N2 losses be balanced against the carbon benefits (para. 22) [2.4, 24.5].

7.╇ The potential for integrated approaches to manage nitrogen A holistic view to managing the nitrogen cascade 59.╇ Given the range of adverse environmental effects in the Nr cascade, the most attractive mitigation options are those that offer simultaneous reductions of all N pollutants from all emitting sectors and in all environmental compartments. 60.╇ An integrated approach to Nr management holds the promise of decreasing the risks of inconsistency, inefficiency and pollution swapping. Efforts at integration should recognize the varying level of success in Nr policies (para. 23–26) aiming to ensure balance in mitigation efforts between sectors.

xxxii

2100

Integration puts higher demands on interdisciplinarity and consensus building between science, policy and stakeholders [4.6, 23.4]. 61.╇ Integrated policies are also justified within sectors, such as agriculture, because of the large number of actors and the connection between sources, sectors and effects [23.4]. The Common Agricultural Policy of the EU provides a potentially powerful incentive to improve sustainability of agricultural production.

Seven key actions for better management of the nitrogen cascade 62.╇ Seven key actions in four sectors provide a basis for further developing integrated approaches to N management [23.5].

Agriculture (1)╇ Improving nitrogen use efficiency in crop production This includes improving field management practices, genetic potential and yields per Nr input, with the potential to reduce losses per unit of produce, thereby minimizing the risk of pollution swapping [3.3, 22.6, 23.5]. (2)╇ Improving nitrogen use efficiency in animal production As with crops, this includes management practices and genetic potential, with an emphasis on improving feed conversion efficiency and decreasing maintenance costs, so reducing losses per unit of produce and the extent of pollution swapping [3.4, 10.3, 23.5]. (3)╇ Increasing the fertilizer N equivalence value of animal manure Increasing fertilizer equivalence values requires conserving the Nr in manure during storage and land application (especially reducing NH3 emissions where much Nr is lost), while optimizing the rate and time of application to crop demand [3.4, 10.3, 23.5].

Transport and Industry (4)╇ Low-emission combustion and energy-efficient systems These include improved technologies for both stationary

Summary for policy makers

combustion sources and vehicles, increasing energyefficiency and use of alternative energy sources with less emission, building on current approaches [4.5, 23.5, 24.6].

Waste water treatment (5)╇ Recycling nitrogen (and phosphorus) from waste water systems╇ Current efforts at water treatment for Nr in Europe focus on denitrification back to N2. While policies have been relatively successful [4.6], this approach represents a waste of the energy used to produce Nr (para. 22). An ambitious long-term goal should be to recycle Nr from waste waters, utilizing new sewage management technologies [12.3, 23.5].

Societal consumption patterns (6)╇ Energy and transport saving╇ Against the success of technical measures to reduce NOx emissions per unit consumption, both vehicle miles and energy use have increased substantially over past decades. Dissuasion of polluting cars and far-distance holidays, and stimulation of energy-saving houses and consumption patterns can greatly contribute to decreasing NOx emissions [23.5]. (7)╇ Lowering the human consumption of animal protein European consumption of animal protein is above the recommended per capita consumption in many parts of Europe. Lowering the fraction of animal products in diets to the recommended level (and shifting consumption to more N-efficient animal products) will decrease Nr emissions with human health co-benefits, where current consumption is over the optimum [23.5, 24.5, 26.3]. 63.╇ Key Action 4 involves technical measures that are already being combined with public incentives for energy saving and less polluting transport (Key Action 6), linking Nr, air pollution and climate policies (cf. Figure SPM.9). Similarly, each of the Key Actions in the food chain (1–3, 7) offers co-benefits with climate mitigation and the management of other nutrients, including phosphorus. Given the limited success so far in reducing agricultural Nr emissions, more effort is needed to link the Key Actions, both to learn from the successes and to ensure equitability between sectors.

8.╇ Challenges for society and policy Nitrogen in multilateral environmental agreements and future research 64.╇ International treaties, such as Multilateral Environmental Agreements (MEAs), have done much to protect the global environment, promoting intergovernmental action on many environmental issues, but none has targeted nitrogen management policy holistically [4.3, 25.2]. 65.╇ A new international treaty targeted explicitly on nitrogen could be a powerful mechanism to bring the different elements of the nitrogen problem together. While a new convention would be complex to negotiate and could compete with

existing structures, a joint protocol between existing conventions could be effective and should be explored [25.3, 25.4]. 66.╇ New coordinating links on nitrogen management between MEAs should be further developed, including the Global Partnership on Nutrient Management facilitated by the United Nations Environment Programme, the Task Force on Reactive Nitrogen of the UNECE Convention on Longrange Transboundary Air Pollution and the links with other UNECE Conventions. There is the opportunity for the UNECE Committee on Environmental Policy to develop nitrogen management links between UNECE Conventions, while the European Union and its Member States have important roles to play in harmonization and coordination [25.4]. 67.╇ Such coordination actions will require ongoing support from the scientific community, especially given the many remaining uncertainties inherent in developing the long-term vision of a holistic approach. Research programmes should put a higher priority on quantifying the nitrogen links between the traditional domains of disciplines, media and environmental issues, providing data and models that can underpin future negotiations and policies.

Societal choice, public awareness and behavioural change 68.╇ European society is facing major choices regarding food and energy security, and environmental threats including climate change, water, soil and air quality and biodiversity loss. These issues are intricately linked to the nitrogen cycle and have a strong global context, with the decisions of European individuals on life-style and diet having a major role to play [26.3]. 69.╇ In Europe, different scenarios and models suggest a strong 75% decline of NOx, while emissions of NH3 and N2O display an uncertain future outlook (Figure SPM.9) [24.6]. The constraints that have so far limited reductions in Nr emissions from agriculture include many stakeholders, an open farming system with diffuse losses, the desire to maintain high outputs for European agro-economy and food security, and possible concerns about how to transfer anticipated costs to consumers (para. 25). Changes in agricultural practices to achieve substantial reductions of European Nr emissions in the coming decades therefore require awareness and broad support from policy, industry, farmers, retailers and consumers [23.3, 26.3]. 70.╇ The comparison between combustion and agricultural Nr emissions highlights the need to engage the public. This should emphasize mutual responsibility along the whole food-supply chain, support the basis for transferring any mitigation costs to the consumer, and emphasize that the substantial costs of environmental impacts fully justify taking action [4.5, 23.5, 26.3]. 71.╇ At present, public and institutional awareness of the global nitrogen challenge is very low. The comparison with carbon and climate change highlights how the nitrogen story is multifaceted, cutting across all global-change

xxxiii

Summary for policy makers

themes. This complexity is a barrier to greater public awareness, pointing to the need to distil easy messages that engage the public [5.4, 26.4]. 72.╇ Simple messages for nitrogen include contrasting its huge benefits for society against the environmental threats, and emphasizing the need to extend existing footprinting

xxxiv

approaches, for example to calculate ‘nitrogen foodprints’. Perhaps the strongest message to the public is that there are substantial health benefits to be gained by keeping consumption of animal products within recommended dietary limits. It is an opportunity to improve personal health and protect the environment at the same time [23.5, 24.5, 26.3].

Technical summary Lead authors:€Mark A. Sutton and Gilles Billen Contributing authors:€Albert Bleeker, Jan Willem Erisman, Peringe Grennfelt, Hans van Grinsven, Bruna Grizzetti, Clare M. Howard and Adrian Leip

Part I╇ Nitrogen in Europe:€the present position Nitrogen inheritance 1. Gaseous di-nitrogen (N2) constitutes 78% of the earth’s atmosphere. It is a rather inert chemical, being nearly unavailable for the biological cycle. The other nitrogen forms are much more reactive; these include nitrate (NO3−), ammonium (NH4+) and ammonia (NH3), gaseous nitrogen oxides (NOx), nitrous oxide (N2O) and many other inorganic and organic nitrogen forms. Collectively, they are termed ‘reactive nitrogen’ (Nr). They are normally scarce in natural environments, with their low availability limiting the productivity of natural ecosystems. This was also the case for agricultural production before 1900, which long remained dependent on the recycling of Nr in human waste and manure, and the capacity of legumes to fix atmospheric N2 biologically. 2. With a growing human population through the nineteenth century and the need for more Nr, Europe increasingly operated a ‘fossil nitrogen economy’, dependent on the addition of nitrogen fertilizers from mined sources, including from guano, coal and saltpetre. The ‘nitrogen problem’ of the time was that these sources were fast becoming insufficient to meet Europe’s escalating need for fertilizer Nr, and its military need for Nr in explosives [1.1].1 3. The situation changed dramatically shortly after 1900, with the invention of the Haber–Bosch process. This allowed the cheap industrial production of ammonia from di-nitrogen and hydrogen, permitting mass production of synthetic Nr fertilizers. By the 1930s, the European shortage of Nr had become a problem of the past, with Nr use in agriculture strongly increasing from the 1950s [1.1, 2.2]. 4. The deliberate production and release of Nr in the Haber– Bosch process can be considered as perhaps the greatest single experiment in global geo-engineering that humans have ever made [1.1]. In Europe, human production of Nr fully met its 1

References in this summary refer (e.g., [1.1, 2.2]) to chapter and section numbers of the European Nitrogen Assessment.

objectives to underpinning food and military security, while supplying a vital feedstock for many industrial processes [3.2, 3.5]. What was not anticipated was that this experiment would lead to a ‘nitrogen inheritance’ of unintended consequences [1.1], with Nr leaking into the environment in multiple forms, causing an even larger number of environmental effects [1.1, 2.6]. Simultaneously, the increasing extent of fossil fuel combustion for transportation and electricity production has led to a massive unintentional additional release of Nr into the atmosphere, mainly as nitrogen oxides (NOx) [2.4]. 5. At the global scale, together with crop biological nitrogen fixation, these processes have altered the nitrogen cycle to an unprecedented extent, and much more than that of carbon or phosphorus. Humans introduce more Nr into the biosphere than all natural processes together [2.5, 13.2, 16.4]. Europe (EU-27) is a hot spot in this sense, producing 10% of global anthropogenic Nr, even though its surface covers less than 3% of the total world continental area [2.5, 13.2].

Benefits, threats and current policies 6. The value of the benefits brought to the European economy by the production of fertilizers and the combustion of fuel is substantial. For example, the economic benefit of applying Nr fertilizers to wheat in the EU is estimated at around €8 billion per year. Accounting for benefits to other crops, livestock production, downstream food processing and many industrial benefits (including mining and chemical synthesis), the total benefits of Nr production will be very much larger [3.6]. By contrast, the formation of NOx through high temperature combustion processes has no economic benefit, as control efforts focus on denitrifying Nr rather than its use [5.1]. 7. Against these benefits must be listed the many effects on the environment and human health, as a result of Nr pollution. Several of these threats have been addressed by governmental policy measures related to abating atmospheric pollution or limiting nitrogen contamination of groundwater and surface water resources. These include several European directives, such as the Nitrates Directive, Water Framework Directive, Groundwater Directive, Ambient Air Quality Directive, National Emissions Ceilings Directive, Urban Waste Water Treatment Directive,

xxxv

Technical Summary

Figure TS.1 Simplified view of the nitrogen cascade, highlighting the major anthropogenic sources of reactive nitrogen (Nr) from atmospheric di-nitrogen (N2), the main pollutant forms of Nr (orange boxes) and nine main environmental concerns (blue boxes). Estimates of anthropogenic N fixation for the world (Tg /yr for 2005, in black) are compared with estimates for Europe (Tg /yr for 2000, in blue italic). Blue arrows represent intended anthropogenic Nr flows; all the other arrows are unintended flows [5.2].

Marine Strategy Framework Directive, Integrated Pollution Prevention and Control (IPPC) and the Habitats Directive [4.4]. 8. The policy responses also include European commitments to multi-lateral environmental agreements, including the United Nations Economic Commission for Europe (UNECE) Convention on Long-Range Transboundary Air Pollution, the UN Framework Convention on Climate Change, the UN Convention on Biological Diversity, and the Oslo and Paris, Helsinki and Barcelona Conventions for the protection of the North East Atlantic, the Baltic Sea and the Mediterranean Sea, respectively [4.3]. 9. Review of the current policies highlights a tendency to address individual Nr species from specific source sectors (agriculture, traffic, industry), media (air, freshwater, marine), and for specific issues (climate, urban air pollution, biodiversity, water quality, etc.). Until now, there has been little focus on developing policies that recognize the full extent and complexity of the nitrogen cycle [4.3, 4.4]. 10. Trends in environmental pollution show that significant progress has been made in reducing emissions of NOx to air. These policies have benefited from the availability of measures targeted at few stakeholders (e.g. electricity generation companies, industry, vehicle manufacturers), but have nevertheless been offset by increases in overall transport use and energy consumption. In the same way, significant progress has been

xxxvi

made in reducing water pollution due to water treatment policies that have engaged water companies [4.6]. 11. By contrast, from a European perspective, Nr pollution from agriculture has shown only modest reductions in response to policies over the past 20 years [4.6]. Particular challenges faced in controlling Nr losses from agriculture are the many (often small) stakeholders, a highly diverse open system, and the perception of difficulty in passing anticipated costs to consumers [4.6]. In addition, the low price of Nr fertilizer, combined with its clear benefits to agricultural production, does not provide a strong incentive for farmers to use less than the (private) economic optimum [3.3].

Nitrogen cascade and the need for integration 12. The lack of a holistic approach to developing nitrogen policies can partly be explained by the sectoral approach taken in government departments, and can also be linked to the tendency to scientific specialization. Nitrogen research communities have developed to become highly fragmented, providing a major challenge to integrate across all nitrogen forms, proÂ� cesses and scales [5.2]. 13. The ‘nitrogen cascade’ (Figure TS.1) provides a useful concept to demonstrate the benefits of a holistic understanding

Technical Summary

of the nitrogen cycle [2.6, 5.2]. Anthropogenic fixation of N2 to Nr raises the energy state of the nitrogen, with the energy being gradually dissipated as the Nr is converted through many different forms, until it is eventually denitrified back to N2, the thermodynamically stable form of nitrogen in most environments. Even though Nr may be emitted from different proÂ� cesses and sectors, once released into the environment, the origin gradually becomes less relevant, causing multiple environmental effects. Each molecule of Nr emitted may cause several effects before eventual denitrification [2.6, 5.2]. 14. The cascade concept highlights the potential for tradeoffs and synergies in managing Nr. For example, if losses of Nr to one pathway are reduced, this can easily increase losses of another Nr form (e.g., some meaures to reduce nitrate leaching may increase ammonia emissions and vice versa). Such tradeoffs are sometimes termed ‘pollution swapping’. By contrast, because of the cascade, some measures may have benefits in reducing multiple forms and impacts of Nr [1.2]. 15. It is concluded that a more holistic approach to managing the nitrogen cycle would have benefits to ensure more effective control of the different forms of Nr pollution and impacts. Such a development must be based on more-effective integration across environmental disciplines, providing the foundation to link traditionally separate policy domains [5.3].

Approach of the European Nitrogen Assessment 16. Developing a more joined up approach to managing the nitrogen cycle necessarily proceeds gradually, and this has been encouraged through the European Nitrogen Assessment process. The overall goal of the Assessment was established as:€ to review current scientific understanding of nitrogen sources, impacts and interactions across Europe, taking account of current policies and the economic costs and benefits, as a basis to inform the development of future policies at local to global scales [1.4]. 17. In developing this vision of gradual integration, based on analysis of the present position (Part I), the Assessment first examines the nitrogen turn-over processes in the biosphere (Part II), and then addresses nitrogen flows at different spatial scales (Part III). 18. One of the key conclusions of the first part of the Assessment is that the complexity of the nitrogen cycle needs to be distilled to highlight the priority concerns. This is important to limit the number of interactions when developing integrated approaches [5.4]. 19. Recognizing these issues, the Assessment process established a comprehensive list of around 20 problems related to nitrogen. The list was first distilled down to nine ‘main environmental concerns’, setting the agenda for the Nitrogen in Europe (NinE) programme, as shown in Figure TS.1 [5.4]. 20. In a second stage, the list was reduced to five ‘key societal threats’ of excess nitrogen, identified as:€Water quality, Air quality, Greenhouse balance, Ecosystems and biodiversity, and Soil quality. These five threats provide a framework that automatically includes many of the other issues, balancing the complexity of the nitrogen cycle with the need for simplification [5.4].

21. The short-listing of the five key threats also provides a useful tool to communicate the nitrogen challenge to society. Together the five threats make an acronym as the ‘WAGES’ of excess nitrogen, while they can be also envisaged in direct analogy to the ‘elements’ of classical Greek cosmology (Figure TS.2) [5.4]. 22. The Assessment applies the framework of five key societal threats to summarize the scale of the nitrogen challenge facing Europe (Part IV). Finally, the threats are brought together to examine the future perspective for European nitrogen policies (Part V). The Assessment used a network approach, where expert teams were formed for each chapter based on open invitations, including discussions of outlines during workshops held for each of the five parts (I–IV). The chapters take a variety of approaches across the Assessment, reflective of variation in data availability (e.g., limited data for some parts of Europe) and the nature of the issues being assessed. Draft chapters were subjected to internal and external peer review before being finalized.

Part II╇ Nitrogen processing in the biosphere 23. In recent decades substantial advances have been achieved in our understanding of the processes that govern nitrogen cycling in terrestrial environments (including natural and agricultural ecosystems), in aquatic environments (including freshwater, estuarine and marine ecosystems) and in the atmosphere. Each of these environments has been considered, integrating the processes of all relevant nitrogen forms.

Reactive nitrogen turnover in terrestrial ecosystems 24. The understanding of N cycling in terrestrial ecosystems has undergone a paradigm shift since 1990. Until then, the perception was that:€(1) Nr mineralization is the limiting step in N cycling; (2) plants only take up inorganic Nr; and (3) plants compete poorly for Nr against microbes and use only the Nr which is ‘left over’ by microbes. Since then studies have shown that plants compete effectively for Nr with microÂ� organisms and take up organic N in a broad range of ecosystems [6.4]. 25. On the ecosystem scale, soils are the main reservoir for Nr. This is more pronounced for agricultural systems than for forest systems, with more than 90%–95% of Nr being stored in the soil. Nitrogen stocks of managed systems are typically depleted and with retention processes negatively affected [6.2, 6.4]. 26. In cereal farming, the use of only mineral Nr fertilizers, instead of animal manures or composts, as well as the simplification of the crop rotation scheme that this has made possible, has in some cases resulted in a decline of soil organic matter. In the long-term this practice of using only mineral fertilizers has decreased the buffer capacity of the soil towards inorganic N inputs, thus increasing its propensity to Nr leaching [6.4]. 27. Nitrogen fixation in non-agricultural legumes or in other N-fixing organisms remains difficult to quantify, hampering a

xxxvii

Technical Summary

Figure TS.2 Summary of the five key societal threats of excess reactive nitrogen, drawn in analogy to the ‘elements’ of classical Greek cosmology. The main chemical forms associated with each threat are shown [5.4]. Photo sources: Shutterstock.com and garysmithphotography.co.uk.

better understanding of the importance of biological N2 fixation for most terrestrial ecosystems [6.3]. 28. Nitrogen-enriched terrestrial ecosystems lose significant amounts of N via nitrate leaching and gaseous emissions (N2, N2O, NO, NH3) to the environment. Estimates of denitrification to N2 remain highly uncertain, due to difficulties in measurement and a high degree of temporal and spatial variability. There remain substantial uncertainties in the average fraction of Nr applied to fields that is emitted as N2O, ranging from 1% to 3.5%–4.5% of fertilizer N applied, using bottom-up and topdown estimates, respectively. Further research is needed to better understand the relative contribution of direct and indirect N2O emissions [6.5]. 29. In forests, the C:N ratio of the forest leaf litter or top mineral soil is a good indicator of Nr status related to nitrate leaching. At C:N above 25, mineral Nr is usually retained, whereas below 25, nitrate leaching increases with increasing Nr deposition [6.5] (Table TS.1).

Reactive nitrogen turnover and transfer along the aquatic continuum 30. Major sources of Nr in the aquatic environment include households and sewage discharges together with diffuse pollution losses from agricultural practices. 31. Nitrate retention by riparian wetlands is a frequent justification for conservation and restoration policies of these systems. However, their use for mitigating NO3 contamination of river systems must be treated with caution, since their effectiveness is difficult to predict, and side effects observed include

xxxviii

increased dissolved organic matter and N2O emissions, together with loss of biodiversity [7.5]. 32. Release of dissolved organic nitrogen has often been neglected, while it can play a significant role, particularly in upland semi-natural catchments, but is not determined in most routine European water quality monitoring programmes [7.3]. 33. The effects of increased Nr loadings to aquatic environments include acidification and loss of biodiversity in seminatural environments, and eutrophication in more disturbed ecosystems. Standing waters are particularly sensitive to both acidification and eutrophication, since the longer residence time in these systems leads to greater interaction between the biota and changing water chemistry [7.4]. 34. The richest submerged plant communities in lakes have been observed to be associated with winter nitrate concentrations not exceeding 2 mg N/l, and this has been proposed as an appropriate target concentration for enriched shallow European lakes to reach ‘good ecological status’ [7.5]. 35. Although phosphorus is often the main limiting element controlling primary production in freshwater systems, Nr has been reported as a factor limiting or co-limiting biological production in some eutrophicated lakes, and control of both Nr and P loading is needed in impacted areas, if ecological quality is to be restored [7.4]. 36. The importance of storage and denitrification in aquifers is a major uncertainty in the global N cycle, and controls in part the response of catchments to land use or management changes. In some aquifers, the increase of N concentrations will continue for decades even if efficient mitigation measures are implemented now [7.5]. 37. Nitrogen inputs from human activities have led to ecological deterioration in large parts of the coastal oceans along European coastlines, including harmful algal blooms and anoxia. The riverine Nr-loads are the most pronounced Nr source to coasts and estuaries, while atmospheric Nr deposition and N2 fixation also contribute significantly [8.8]. 38. A large imbalance of Nr with respect to silica inputs causes the development of severe harmful algal blooms. Especially affected by eutrophication are the major European estuaries (e.g., Rhine, Scheldt, Danube and the coastlines receiving their outflow), North Sea, Baltic Sea, and Black Sea, as well as some parts of the Mediterranean coastline [8.10]. 39. Marine biodiversity is reduced under high nutrient loadings, affecting nutrient recycling negatively. Recovery of communities may not be possible if eutrophication and anoxia persist for long time periods of several years [8.9]. 40. The European coastal zone plays a major role in denitrification of Nr to N2. Export of Nr to the sea is estimated at 4.5 TgN per year, most of which will be denitrified to N2. Globally, coastal denitrification is estimated at 61 TgN per year, including 8 Tg per year in estuaries. Comparison with independant estimates of estuarine and sediment denitrification implies that coastal systems impert 54–197 TgN per year from the open ocean globally, compensating the Nr losses due to sediment denitrification [8.8].

Technical Summary Table TS.1 Characteristics of coniferous forest ecosystems with low, intermediate and high N status, as grouped according to total Nr input [6.5]

Nitrogen status

Low N status (N-limited)

Intermediate

High N status (N-saturated)

Input (kg N per ha per yr)

0–15

15–40

40–100

Needle N% (in spruce)

< 1.4

1.4–1.7

1.7–2.5

C:N ratio (g C per g N)

> 30

25–30

< 25

Soil N flux density proxy (litterfall + throughfall) (kg N per ha per yr)

< 60

60–80

>80

Proportion of input leached (%)

<10

0–60

30–100

Reactive nitrogen turnover in the atmosphere 41. The main Nr compounds emitted to the atmosphere by anthropogenic activities are NH3 (~3.2 Tg N per year in EU-27), mainly from agriculture, and NOx including both NO and NO2 (~3.5 Tg N per year). European NOx emissions arise mainly from transport (50%), power generation (25%) and other combustion sources (21%). Emissions of nitrous oxide (N2O) in Europe are much smaller (1 Tg N per year), mainly arising from soils, especially in agriculture [9.2, 16.3]. 42. These Nr forms have different fates in the atmosphere. The atmospheric chemistry of NH3 is well known, undergoing irreversible reaction with sulphuric acid (H2SO4) and reversable reaction with nitric acid (HNO3) and hydrochloric acid (HCl). Previous assumptions of instantaneous equilibrium in the atmosphere are now believed to be incorrect due to kinetic constraints. Together with the effects of mixed aerosol chemistry (e.g., organic layers), these effects are not yet fully parametrized in models [9.3]. 43. Very little is known either quantitatively or qualitatively about atmospheric organic nitrogen compounds, although they can contribute up to half of wet-deposited Nr. Sources may include emissions of amines, amides, urea, amino acids. These represent an extra contribution to Nr deposition and eutrophication, missing from current estimates [9.3]. 44. The biosphere–atmosphere exchange of NH3 and NOx is dependent on a combination of surface (canopy, soil, management) and environmental conditions. In the case of NH3, the environmental dynamics of emissions need to be better quantified, given the expectation that climate change may increase future emissions. Reductions in SO2 are reducing rates of NH3 dry deposition, tending to increase its atmospheric lifetime [9.4]. 45. The potential for ‘pollution swapping’ has been illustrated by the effect of implementing policies to reduce nitrate leaching. By prohibiting winter manure application to fields, more manure is spread in spring, leading to a new peak in observed NH3 concentrations [9.2]. 46. Ammonia has substantial impact near its sources of emission due to high dry deposition rates to natural ecosystems, so that it may significantly affect natural ecosystems in agricultural areas. By contrast, NOx has little impact close to the sources, due to low dry deposition rates, until it is converted into nitric acid (about 5% per hour). Long-range transport of both components occurs in the form of aerosol phase compounds,

which are transported more than 1000 km. Abatement strategies need to take account of these differences when assessing the impact of Nr deposition on sensitive ecosystems [9.4, 9.6].

Part III╇ Nitrogen flows and fate at multiple spatial scales 47. In managing our environment, humans have created complex ecosystem mosaics, the structure of which largely determines the flows and fate of Nr from local to European scales.

Nitrogen flows at farm and landscape scales 48. Farms represent the operational units at which local decisions on the use of nitrogen fertilizer are taken in accordance with policies to encourage food production and sustain farm incomes [10.1]. 49. The basis of good nitrogen management in agriculture is to increase the temporal and spatial coincidence between Nr availability in the soil and Nr uptake by crop, thus increasing the N use efficiency (ratio of Nr produced in final agricultural goods to Nr introduced as fertilizer) and minimizing Nr flows into water and the atmosphere [10.2]. 50. Current management drivers often cause farms to be ‘open’ with substantial N losses, as the objective of high production often surpasses that of minimizing the emissions to the environment. Livestock farming presents particular problems with large potential losses associated with the management of manure. Animal excrements, historically the major fertilization resource for cropland, have often acquired the status of wastes to be disposed. The geographical dissociation of crop and livestock farming increases these problems [10.3]. 51. Farm nitrogen budgets typical of north west Europe highlight the main N loss pathways according to farm type. Although there is a great variation in farm type, illustrative budgets show how the N losses per unit area from beef and organic dairy farms are typically lower than from more intensively managed pig and dairy farms. By contrast, the total N loss per unit N in products is highest for beef, dairy and organic dairy farms (2.7, 2.55, 1.92, respectively) as compared with pork farms (0.8) [10.4]. 52. The fate and environmental effects of Nr losses from agricultural systems are strongly dependent on the structure of the surrounding terrestrial and hydrological landscape. For

xxxix

Technical Summary

this purpose, the landscape is considered as that scale in which adjacent farms, fields, forests, nature areas and water courses interact, as affected by Nr dispersion through air and water, and by human transfers [11.1]. 53. Landscape features, like the patchiness of land uses or presence of woods close to emission sources, can play a sigÂ� nificant role in the fate of emitted ammonia, minimizing net emissions and helping to avoid longer distance transport. At the same time, such landscape features can affect the nature of surface runoff and deep infiltration, while the presence of active wetlands determine the occurrence of denitrification of leached nitrate, with possible secondary N2O emissions [11.3]. These interactions highlight the potential to exploit spatial relationships at the landscape scale to minimize Nr losses and their environmental effects [11.6]. 54. Integrated models are being established to assess the overall effect of landscape-scale decisions in relation to the local spatial distribution of multiple Nr effects (e.g., on biodiversity, greenhouse balance, water quality). The verification of spatially explicit landscape-scale models provides the basis to develop ‘landscape planning’ approaches to minimize Nr threats [11.6]. When designing and implementing environmental measures, greater attention should be given to the landscape scale in order to take into account local dispersion and buffering processes so as to maximize the efficiency of the measures.

Nitrogen flows at the city scale 55. Cities, although they take only ~2% of land, dramatically affect the N cycle. Cities import Nr from rural areas the agricultural products needed to feed dense populations, then disperse the wastes resulting from its consumption:€to surface water (2.3 Tg N per year in wastewater in the EU-27), to air (0.015 Tg N as NH3 and N2O emissions from wastewater and solid waste treatment plants), and to soils (1.5 Tg N in sludge and solid wastes) [12.1, 16.4]. 56. Because of their dense transportation networks, industrial facilities and energy production infrastructures, cities are the main sites of the 3.5 Tg N per year NOx emission to the atmosphere through the burning of fossil and other fuels [12.3, 16.4]. 57. A case study for Paris highlights the amplification of the N cycle from 1800. Today, the major part of Nr output is attributed to fuel combustion for transport and energy (50 Gg Nr per year mainly as NOx). Sewage water treatment plant denitrify 32€Gg Nr to N2, with 12 Gg Nr released to water courses, and only 12 Gg Nr returned to soils [12.3]. 58. By comparison to the present situation, late nineteenth century Paris recycled 50% of the Nr in solid/liquid waste for use in fertilizers, a system which came to the end with the advent of dilute sewage streams (e.g., the flushing toilet) and the Haber–Bosch process. The illustration highlights the potential for future sewage processing systems to process Nr (and phosphorus) for re-use as fertilizer, rather than wasting the Nr resource through denitrification [12.3, 23.5].

xl

Regional scale nitrogen flows through watersheds and the atmosphere 59. Many large watersheds such as the Danube, Rhine and the Scheldt cover several countries and, together with coastal water transfers and atmospheric dispersion, allow substantial transboundary transport of Nr pollution across Europe. 60. Regional watersheds represent territorial units for water resources management in relationship with both agricultural and urban activities. The strong regional specialization of agriculture and the concentration of urban habitat in Europe result in some basins exporting large amounts of Nr as food and feed (autotrophic basins), while others import them (heterotrophic basins). The difference between the two (i.e., net autotrophy) is illustrated in Figure TS.3. This map highlights the imbalance of the regional nitrogen cycle between watersheds, some of which import much more nitrogen (as food and animal feed) than their crop production can use [13.2]. 61. Throughout Europe, net anthropogenic Nr inputs (fertilizer application, food and feed import, crop N2 fixation and atmospheric deposition) represent 3700 kg Nr per km² annually (watershed ranges 0–8400 kg per km2), which is five times the rate of natural N2 fixation [13.2]. Of the anthropogenic Nr input, ~80% is stored (in soils, sediments or groundwater) or lost to the atmosphere along the drainage network as Nr or N2. Only ~20% reaches the basin outlet and the marine coastal zones, at rates four times background. In such coastal areas with limited silica availability, Nr causes harmful algal blooms [8.9, 13.6]. 62. The chain of processes leading to atmospheric emissions of Nr, atmospheric dispersion, chemical transformation and deposition is extremely complex and currently observations only address part of this chain. More observations are needed especially of gaseous nitric acid (HNO3), ammonia (NH3) and coarse nitrate aerosol concentrations. Concentrations of all compounds should be measured at the same site if the mass-balance of Nr is to be assessed, pointing to the need for integrated site measurements in monitoring networks [14.4]. 63. Atmospheric models are routinely used to quantify the spatial patterns and trans-boundary fluxes of Nr across Europe, including inputs to coastal seas. Differences among European models can be 30% in some areas, and substantially more for specific locations. The major uncertainties indicate the need for further information on atmosphere–biosphere fluxes of Nr with sensitive ecosystems, dry deposition of particles, sub-grid fluxes of NHx compounds, and effects of topography on wetdeposition. A balanced program of observations and models is critical to future understanding of atmospheric transport of Nr at local to global scales [14.6]. 64. Considering the Nr inputs to the environment from multiple sources, and the inter-related flows between ecosystems, watersheds and the atmosphere, there is an obvious need for closer cooperation between existing monitoring activities. Strengthened monitoring programs and data integration are needed at the international level, which should improve harmonization between terrestrial, aquatic and atmospheric mon-

Technical Summary

Autotrophy-Heterotrophy, kgN/km²/yr > 5000 autotrophic 5000 – 1000 1000 – -1000 balanced -1000 – -5000 -5000 – -10000 heterotrophic < -10000

itoring communities (terminology, methodology, units) and the exchange of information [synthesis of 13, 14].

European scale nitrogen flows 65. European nitrogen flows and budgets have been estimated for terrestrial ecosystems (agriculture, forest and other ecosystems) and for all systems combined (also including urban, transport, industrial and aquatic flows). 66. For the year 2000, a comparison of four mass balance models estimated Nr inputs to agriculture in the EU-27 at 23–26 Tg N, being mainly due to fertilizer and animal manure. For emissions from agriculture, the comparison showed similar relative estimates for NH3 (2.8–3.9 Tg N) and N2O (0.35–0.46 Tg N), but diverging results for soil NOx (0.02–0.20 Tg N). The largest absolute uncertainties were for NH3 [15.6]. 67. Inputs of Nr to soils from both fertilizers and manure increased between 1970 and 2010 by ~20% for the EU-27. Although cattle numbers decreased, this trend was more than

Figure TS.3 Difference between nitrogen autotrophic and heterotrophic contributions for territories in the EU-27. Autotrophy is defined as the amount of nitrogen in crop and grass produced by agriculture. Heterotrophy is the amount of nitrogen feed ingested by livestock and food consumed by humans. The map shows overall net autotrophy (positive values) and net heterotrophy (negative values) [synthesis from 13.2 and 16.4].

offset by increased Nr excretion rates per cow (e.g., increased milk production per cow) [15.5]. These increases are consistent with overall estimated increases in NH3 and N2 emissions and Nr leaching by ~10% between 1970 and 2000, with emissions decreasing slightly since peak values around 1985. Emissions per unit agricultural area have increased by 20%–30% as a result of intensification over the period in western Europe [15.5]. 68. The estimated distribution of overall Nr losses to the environment is shown in Figure TS.4. Emissions to the atmosphere reflect the distribution of livestock production (dominating NH3 emissions) and human population centres (dominating NOx emissions) across Europe. The distribution of Nr inputs to aquatic systems is dominated by nitrate losses, which are largest in areas with high livestock density and precipitation excess, while more localized peaks are associated with urban waste-waters [16.3]. 69. An overall budget of the present N cycle has been established (around 2000 for EU-27) and is summarized in

xli

Technical Summary

Figure TS.4 Distribution of reactive nitrogen emissions across Europe (kg N per km2 for 2000) including emissions to air as NOx, NH3 and N2O, and total losses to aquatic systems, including nitrate and other Nr leaching and wastewaters [16.3].

Figure€TS.5. The budget highlights the central role of crop production and livestock farming. The annual Nr brought to agricultural soils at 27.5 Tg N (consisting of 11.2 Tg N as synthetic fertilizers, 7.1 Tg N as manure, 2.4 Tg N as atmospheric deposition and 1.0 Tg N through biological nitrogen fixation and 5.8€Tg N as crop residues) is in surplus over the requirements of crop production (17.6 Tg N). The annual Nr surplus of 9.9 Tg contributes to substantial Nr leaching to surface and groundwater (6 Tg N), denitrification to N2 (4.5 Tg N), volatilization as NH3 (1.6 Tg N) and emission of N2O and NO (0.5 Tg N). The overall balance for agricultural soils implies a small annual depletion of soil Nr stocks, although this term is considered to be very uncertain since it is calculated by difference of several uncertain estimates. It represents the regional average of a net loss of Nr in soils of autotrophy-dominated regions with arable farming, and a net gain of Nr in soils of heterotrophy�dominated regions with intensive livestock farming [16.4]. 70. In order to provide an annual consumption of animal products by humans of 2.3 Tg N, livestock farming in Europe uses five times as much Nr from crops and imported feed (11.8 Tg), driving the overall European agricultural N cycle. By comparison, direct consumption by humans of crops grown in Europe represents only 2.0 Tg N per year. The handling of livestock excreta also leads to direct gaseous emissions of 1.5 Tg N per year. Overall, in order to produce 4.3 Tg N of food

xlii

annually for the European population (not including 0.4 Tg N in imported food and feed), three times as much Nr is emitted to the environment, which corresponds to a nitrogen use efficiency of 30% (as compared with a global average of 50%) [16.4]. 71. To these emissions should be added 3.7 Tg N per year of wastewater discharged to surface waters and 3.4 Tg N per year of NOx emitted through fossil fuel combustion by the energy, industry and transportation sectors [9.2, 16.3]. 72. As a whole, at around the year 2000, the EU-27 is a net commercial importer of animal feed and human food of 3.5 Tg N per year (mainly due to feed). Conversely, Europe is a net exporter of Nr to the environment:€ by atmospheric transport (2.4 Tg N per year) and by river export to marine systems (4.5 Tg N per yr). The largest single sink for Nr is denitrification to N2 in soils, river sediments and the sediments of European shelf regions, estimated at 9.3 Tg N per year, which is one of the most uncertain estimates [16.4, 8.8]. A net transfer from anthropogenic to natural systems is also implied, as estimated annual losses (0.7 Tg N) are less than half the estimated inputs (1.6 Tg N) mainly from atmospheric Nr deposition. 73. A similar budget has been reconstructed for the same territory (the present EU-27) in the beginning of the twentieth century (Figure TS.6). The estimates are necessarily less certain than for 2000, and the hypotheses leading to this

Technical Summary

Europe (EU27), around 2000. N fluxes in TgN/yr

Atmospheric N2 pool Net atmosph. export

2.4

9.7

6

3.5

3.1

N2fix indust & traffic

4

3.4 3.8 Nat N2fix

Wood exp.

Atm depos

0.2

3.8

0.3

2

crop N2fix

Atmospheric NH3, NOx, N2O

0.4

7

11.8

Fertilizers

2.3

Net import of food & feed

5

2

11.2

1.0

Crop production

Human nutrit.

Livestock farming

17.6

5

5.8

1

3.1 wwt

3.6 Semi-nat soils

0.2

1

2.4

13.6

Agricult soils

1.4 NH3,NOx & N2O emission

2.1

8

1.5 0.1

landfill

4.5 6.2 1.8

Leachg & runoff

Denitrification

3

2.7

4.6 0.6

4.5

Export by rivers to the sea

Figure TS.5 The N cycle at the scale of EU-27 [simplified from 16.4] for the year 2000. Fluxes in green refer to ‘natural’ fluxes (to some extent altered by atmospheric Nr deposition), those in blue are intentional anthropogenic fluxes, those in orange are unintentional anthropogenic fluxes. The numbered green circles indicate a package of seven key actions for overall integrated management of the European nitrogen cycle (see para. 111) [23.5].

retrospective reconstruction are described in the ENA supplementary material [16]. 74. By comparison with present-day amounts, the use of mineral fertilizers around the year 1900 (mainly from Chilean salpetre, guano and coal) was very small. The primary source of new Nr in agriculture was biological N fixation by legume fodder crops, typically grown once every three years in triennial rotations. The nitrogen fixed by legume crops was brought to arable soils by the incorporation of crop residues and application of animal manures. Losses of Nr from agricultural soils, though already significant, were only one third of the current level (Figure TS.6). 75. Annual atmospheric deposition of Nr was around 1.9 Tg N in 1900, roughly half of which was deposited each to agricultural and semi-natural land. This was roughly half of the atmospheric deposition in 2000 at 3.8 Tg N, of which only 37% is deposited to semi-natural land, reflecting the overall reduction in area of semi-natural land (Figure TS.6). As with 2000, the overall difference between Nr removals and gains for

agricultural soils is not considered significant, while the average gains to semi-natural soils including forests (1.1 Tg N) were larger than the removals (0.6 Tg N), but less so than in 2000.

Part IV╇ Managing nitrogen in relation to key societal threats 76. For each of the five key societal threats of excess reactive nitrogen, the Assessment examined the scale of the concern, including, where possible, information on temporal trends and the progress made through any existing policy measures.

Water quality 77. Anthropogenic increase of Nr in water poses direct threats to humans and aquatic ecosystems. High nitrate concentrations in drinking water are considered dangerous for human health, as they might cause cancers and (albeit rarely) infant methaemoglobinaemia. There is also evidence for benefits of nitrate

xliii

Technical Summary Europe (EU27), around 1900. N fluxes in Tg N /yr

Atmospheric N2 pool

2.1

N2fix indust & traffic

0.6

2

Crop N2fix

Atmospheric NH3, NOx, N2O

4

1.9 Nat N2fix

Wood exp.

0.2

0.2

Atm depos

1.9

Fertilizers

0.2

9.6

7.6

Crop production

1.1

semi-nat soils

0.3 0.3

1.2 0.9

Agricult soils

1 NH3,NOx & N 2O emission

1.2

6

1

0.6

2

0.8

1.2 1.2 Leachng & runoff

Denitrification

Human nutrit.

Livestock farming

0.9

1.2

1.2 2.3

Export by rivers to the sea

Figure TS.6 Reconstruction of the N cycle at the scale of current EU-27 for the year 1900 (16.4, supplementary material). Same colour code is used as in Figure TS.5.

for cardiovascular health and protection against infections. In aquatic ecosystems the Nr enrichment produces eutrophication, which is responsible for toxic algal blooms, water anoxia, fish kills and biodiversity loss [8.8, 17.3]. 78. In addition to high Nr concentrations in European waterbodies, increasing nitrate in groundwaters threatens the long-term quality of the resource, as nitrate may have long residence time in the aquifers, and it can be expected that past fertilizer strategies will impact for many decades the quality of European groundwaters [7.5, 17.2]. 79. About 3% of the population in EU-15 using drinking water from groundwater resources is potentially exposed to concentrations exceeding the standard for drinking water of 50 mg NO3/l (11.2 mg N/l), with 5% of the population is chronically exposed to concentrations exceeding 25 mg NO3/l (5.6 mg N/l), which may double the risk of colon cancer for above median meat consumers [17.3]. 80. A value of 1.5 mg N/l has been considered as the total Nr limit above which freshwater bodies may develop loss of

xliv

biodiversity and eutrophication. Except in Scandinavia and in mountainous regions, this level is already exceeded in most European freshwater bodies (Figure TS.7) [7.5, 17.3]. 81. With Nr inputs to the coastal zone at four times the natural background (para. 61), large areas along Europe’s coastline are suffering severe eutrophication problems, with anoxia and/ or the proliferation of undesirable or toxic algae. Particularly affected are the south-eastern continental coast of the North Sea, the Baltic Sea (except the Gulf of Bothnia), the coasts of Brittany, the Adriatic Sea and the western coastal Black Sea [8.10, 13.7, 17.3]. 82. Although eutrophication is decreasing, existing international policies have not been fully implemented, and even under favourable land use scenarios, Nr export to European waters and seas is anticipated to remain a problem in the near future. Achieving substantial progress at the European scale requires integration of sectoral policies, reducing overall inputs of Nr to watersheds, e.g., through changes in agriculture and other N flows [4.5, 13, 17.5].

Technical Summary

Figure TS.7 Indication of the nitrogen threat to water quality. Potential risk of eutrophication for surface freshwater based on estimated total Nr concentrations. The three classes of risk are:€low, <0.5 mg/l; medium 0.5–1.5 mg/l; high >1.5 mg/l as total Nr concentration in water [17.3.3].

Air quality 83. Emissions of NOx and NH3 contribute to several negative effects on human health and ecosystems. In addition to effects of NO2, secondary pollutants play key roles. These include ground level ozone (O3), formed photochemically in the presence of NO2 and volatile organic compounds (VOC), and inhalable particulate matter (PM), formed from oxidation of NO2 to HNO3 and reaction with NH3 to form ammonium nitrates. NOx, O3 and PM cause or aggravate asthma, reduced lung functions and bronchitis. Chronic exposure may increase the probability of respiratory or cardiovascular mortality and cancers [18.2]. 84. Direct NO2 and ozone damage to vegetation has been recognized for a long time, as well as to materials, buildings and objects of cultural heritage. There is a difficulty of ascribing health effects to NO2 per se at ambient levels rather than considering NO2 as a surrogate for a traffic-derived air pollution mixture [18.2]. 85. The role of particulate ammonium and nitrate in human health effects is still under discussion. Current approaches assume damage on a mass basis for PM with a median diameter less than 2.5 μm (PM2.5). Nr compounds contribute up to 30–70% of PM2.5 mass in Europe [18.5]. Overall, models estimate a loss of statistical life expectancy due to PM of 6–12 months across most of central Europe (Figure TS.8). There has been a low success in controlling NH3 emissions in Europe

which needs to be further assessed, in particular in connection with the development of new agricultural policies [18.6, 4.5]. 86. In the EU-27 countries, 60% of the population lives in areas (mainly urban) where the annual EU limit value of NO2 is exceeded. Levels have decreased since 1990, although the downward trends have been smaller or even disappeared after 2000. Although episodic O3 levels have decreased since 1990 due to VOC and NOx control, continental background concentrations have increased, with O3 levels remaining a threat to human health and ecosystems [18.5, 4.5].

Greenhouse balance 87. European anthropogenic Nr emissions have a complex effect on climate by altering global radiative forcing. They directly affect the greenhouse gas balance through N2O emissions and indirectly affect it by increasing tropospheric O3 levels, altering methane (CH4) fluxes, and by altering biospheric CO2 sink (including atmospheric Nr deposition and O3 effects). Aerosol formed from NOx and NH3 emissions also has a cooling effect [19]. 88. A first assessment has been made of the overall effect (between 1750 and 2005) of European Nr emissions on radiative forcing. The main warming effects of European anthropogenic Nr emissions are estimated to be from N2O (17 (15€– 19) mW/m2) and from the reduction in the biospheric CO2 sink by tropospheric O3 (4.4 (2.3€– 6.6) mW/m2). The main cooling

xlv

Technical Summary Figure TS.8 Indication of the nitrogen threat to air quality. Across Europe Nr typically accounts for up to 0.3–0.7 of total particulate matter load with a median diameter less than 2.5 μm (PM2.5) on a mass basis [18.5]. Assuming that health effects are proportionate to mass [18.2.2] the map indicates the loss in statistical human life expectancy (in months) attributable to total PM2.5 [18.6, supplementary material].

<1 1-2 2-4 4-6 6-9 9 - 12 > 12

effects are estimated to be from increasing the biospheric CO2 sink by atmospheric Nr deposition (−19 (−30 to −8) mW/m2) and by light scattering effects of Nr containing aerosol (−16.5 (−27.5 to −5.5) mW/m2) (Figure TS.9) [19.6]. 89. Overall, European Nr emissions are estimated to have a net cooling effect, with the uncertainty bounds ranging from substantial cooling to a small net warming (−15.7 (−46.7 to +15.4) mW/m2) [19.6]. 90. The largest uncertainties concern the aerosol and Nr fertilization effects, and the estimation of the European contributions within the global context. Published estimates suggest that the default N2O emission factor of 1% used by the Intergovernmental Panel on Climate Change (IPCC) for indirect emissions from soils following Nr deposition is too low by at least a factor of two [6.6, 19.6]. 91. Industrial production of Nr can be considered as having permitted increased livestock and human populations (and associated food, feed and fuel consumption). The expected substantial net warming effect of these wider Nr interactions remains to be quantified. Although individual components of Nr emissions have cooling effects, there are many opportunities for ‘smart management’ linking N and C cycles. These can help mitigate greenhouse gas emissions, while reducing the other Nr-related environmental threats [19.7].

Terrestrial ecosystems and biodiversity 92. Atmospheric Nr deposition is a significant driver of biodiversity loss in terrestrial ecosystems. Rates of Nr deposition have substantially exceeded critical load thresholds for natural and semi-natural areas since the increase in agriculture-, energyand transport-related emissions from the 1950s, resulting in a considerable loss of biodiversity in Europe [5.1, 8.2, 20.4].

xlvi

93. Nr deposition affects vegetation diversity through direct foliar damage, eutrophication, acidification, and pathogen susceptibility. The most vulnerable habitats are those with species adapted to low nutrient levels or sensitive to acidification, and include grassland, heathland, wetlands and forests [20.3]. First estimates of the overall reduction in biodiversity due to Nr deposition across Europe have been made (Figure TS.10), while the reductions in Nr- sensitive species will be even greater [20.4]. 94. Although it is not yet clear to what extent oxidized Â�versus reduced Nr (e.g., NO3−, NH4+) have different effects on biodiversity, gaseous ammonia (NH3) can be particularly harmful to vegetation, especially lower plants, through direct foliar damage [20.3]. Changes in plant communities may also indirectly affect faunal biodiversity. It is likely that Nr deposition acts synergistically with other stressors, in particular climate change, acid deposition and ground-level ozone, but these synergies are poorly understood [20.2]. 95. Because of cumulative effects of Nr inputs, it is likely that biodiversity has been in decline for many decades due to Nr deposition. This implies that recovery after a reduction in deposition is likely to be slow, and in some cases may require active management intervention in the habitats affected [20.5].

Soil quality 96. The major Nr threats on soil quality for both agricultural and natural soils are soil acidification, changes in soil organic matter content and quality and loss of soil biodiversity linked to eutrophication. Application of N fertilizers and manure and atmospheric Nr deposition cause soil acidification, which leads to a decrease in crop and forest growth and increases leaching of components negatively affecting water quality, such as heavy metals [21.3].

Technical Summary

426 [382 to 469]

Fossil fuel & landuse change CO 2 biospheric CO2

Long-lived greenhouse gases

(incl. atmos. fertilization & O3 effect)

CH4 (decreased atmospheric lifetime & and decreased soil uptake)

N2 O

-74 [-86 - -62] -19 [-30 - -8 ] 4.4 [2.3 - 6.5] 24.5 [22-27 ] -4.6 (-6.7 - -2.4 ) 0.13 [0.03 - 0.24 ] 17.0 [14.8 - 19.1] 17.0 [14.8 - 19.1] 7.5 [4.5 - 10.5 ]

Halocarbons

<- 8 <- 1

Stratospheric Ozone

5.0 [2.0 - 8.0] 2.9 [0.3 - 5.5]

Tropospheric Stratospheric water vapour from CH4

3.6 [1.0 to 6.1] Land use

-38.3 [-76.6 - 0.0]

Surface albedo Black carbon on snow Sulphates Total aerosol

Direct effect

(SO2 oxidation & aerosol neutralization)

Cloud albedo effect

9.9 [0 - 19.8 ] -26.5 [-16.5 to -36.5] -5.4 [-9.4 to -1.4 ]

Nitrate

-11.1 [-18.1 - -4.1 ] -11.1 [-18.1 - -4.1 ]

?

? <2

Linear contrails

409.7 [336.9 - 557.8] -15.7 [-46.7 - +15.4]

Total Anthropogenic -80

-60

-40

-20

0

20

100 200 300 400 500 600 -2

European contribution to global radiative forcing [mW m ] Figure TS.9 Estimate of the change in global radiative forcing (RF) due to European anthropogenic reactive nitrogen (Nr) emissions to the atmosphere [19.6]. Red bars:€positive radiative forcing (warming effects); light green bars:€positive radiative forcing due to direct/ indirect effects of Nr; blue bars:€negative radiative forcing (cooling effects); dark green bars:€negative radiative forcing due to direct / indirect effects of Nr. For biospheric CO2, the dark green bar represents the additional CO2 sequestered by forests and grasslands due to Nr deposition, while the light green bar represents the decrease in productivity due to effects of enhanced O3 caused by NOx emissions. For CH4 the positive (not visible) and negative contributions represent the effects of Nr in reducing CH4 uptakes by soil and the decreased atmospheric lifetime, respectively. Other contributions include the positive effect of tropospheric ozone from NOx and the direct and indirect cooling effects of ammonium nitrate and sulphate containing aerosol.

98. Nitrogen stocks of agricultural land are often depleted compared with semi-natural systems, because of soil disturbance, crop removals and increased Nr losses (para. 25). Addition of nitrogen generally has a positive effect on the quality of agricultural soils, by enhancing soil fertility, soil organic matter and conditions for crop growth, especially when added with carbon in manures [6.4, 21.3]. 99. Some soil fungi and N fixing bacteria are reduced by high N availability, although the effect of N on diversity of soil organisms and the effects of changes of soil biodiversity on soil fertility, crop production and Nr emissions are not fully understood [21.3].

100. In widespread pyrite containing soils, nitrate removal from groundwater by pyrite oxidation increases concentrations of cations, heavy metals and sulphate, causing problems for its use as drinking water [21.2]. 101. Model simulations indicate that most of the European forest soils could recover from their acidified state within a few decades, as a result of recent and possible future reductions in SO2 and NOx emissions. Although NH3 emissions have only decreased slightly and can contribute to soil acidification, the effects of NH3 on plant and soil biodiversity by eutrophication appear to be of more concern [21.4, 20.3].

xlvii

Technical Summary Figure TS.10 Indication of the nitrogen threat to terrestrial ecosystems and biodiversity. The map shows modelled similarity in species composition in European forests between un-impacted conditions, and under long-term enhanced N deposition (1990 scenario) [20.4].

Part V╇ European nitrogen policies and future challenges Costs and benefits of nitrogen 102. Cost–benefit analysis can provide guidance for the setting of policy priorities to abate Nr emissions with an integrated perspective. Social cost estimates have been derived based on available estimates of ‘willingness to pay’ for human life and health, ecosystem services and greenhouse gas emission reduction [22]. 103. Total annual N-related damage cost in the EU-27 is estimated at around €70–€320 billion (i.e., €150–750 per capita per year), representing 1%–4% of the average European income [22.6]. Comparison of the overall costs, for each Nr and threat, is summarized in Table TS.2, which shows that health damage and air pollution cause the largest costs. A provisional ranking of damage from N-emissions in terms of cost, expressed as € per kg N released to the environment, is provided in Table TS.3. 104. Although the unit social costs of N2O effects on climate (€5–€15 per kg N) are similar to several of the other effects (Table TS.3), because total N2O emissions are much smaller, the overall cost to Europe of N2O emissions (also from increased UV radiation) is much smaller than the other effects (Table TS.2). The further effects of Nr emissions on radiative balance (para. 87) are not expected to alter this overall conclusion, although future work is needed to extend the analysis.

xlviii

105. The estimated social benefit of Nr for the farmer is €1–€3 per kg added N-fertilizer equivalent [22.3]. The total marginal environmental costs associated with N-emissions tend to exceed the marginal benefits for the farmer. Internalizing the environmental costs of N-fertilization would lower the optimal N-rate for arable production in north-west Europe by about 50€kg per ha per year [22.6]. 106. The results provide support for N-policies on air pollution and human health, and on reducing ammonia emissions from agriculture, as the social benefits of abatement tend to exceed the additional costs. While strategies that give simultaneous emission reductions are attractive, the results point to the need to prioritize abatement of NOx, NH3 and Nr loss to water over the abatement of N2O emissions [22].

Integrated approaches to nitrogen management 107. The European Union has many policy measures aimed at decreasing unwanted N emissions from combustion, agriculture and urban wastes. Such sector-, media- and individual polllutant-based approaches have had varying success [4.5, 5.2, 22.1]. However, even under favourable land use scenarios the Nr export to European waters and seas is expected to remain problematic in the near future, as are atmospheric NH3 and N2O emissions from agricultural activities, leading to ongoing threats to water quality, air quality, greenhouse balance, ecosystems and biodiversity and soil quality [17–21]. 108. The lack of a holistic approach to the nitrogen cascade leads to risks of contradictory effects of policies dealing with

Technical Summary Table TS.2 Estimates of overall social damage costs in the European Union (EU-27) as a result of environmental Nr-emissions (billion € per year at 2000). Values are shown here rounded to the nearest 5 billion € to avoid over precision, explaining differences with the sums. The calculated value for N2O effects on human health is 1–2 billion € per year [22.6]

NOx emission to air

NH3 emission to air

Nr loss to water

N2O emission to air

Total

Human health

35–100

5–70

0–20

<5

40–190

Ecosystems

5–35

5–35

15–50

—

25–115

a a

Climate

—

—

—

5–10

5–10

Total

40–135

10–105

15–70

5–15

70–320

a

T he value for health effects is proportionately smaller than the value for ecosystems as not all leaching is associated with health effects (e.g., denitrified during the path from soil to sea).

Table TS.3 Estimated cost of different Nr-threats in Europe per unit Nr emitted [22.6]

Effect

Emitted nitrogen form

Emission/ loss to

Estimated cost € per kg Nr emitted

Human health (particulate matter, NO2 and O3)

NOx

Air

10–30

Ecosystems (eutrophication, biodiversity)

Nr (inc. nitrate)

Water

5–20

Human health (particulate matter)

NH3

Air

2–20

Climate (greehouse gas)

N2O

Air

5–15

Ecosystems (eutrophication, biodiversity)

NH3 and NOx

Air

2–10

Human health (drinking water)

Nr (inc. nitrate)

Water

0–4

Human health (increased ultraviolet radiation from ozone depletion)

N2O

Air

1–3

different aspects of the problem (para. 12) [4.6, 5.2]. The promise of integrated approaches to N management is that these are more effective (larger decreases in unwanted emissions) and/or more efficient (less side effects, less complexity) than the set of policies focused on individual sources and Nr forms [23.2]. 109. A conceptual framework developed here distinguishes five dimensions of integration:€ (i) vertical dimension, i.e., cause–effect relationships of N species; (ii) horizontal dimension, i.e., integration of all N species via for example N budgets; (iii) integrating N management with the management of other biogeochemical cycles, (iv) integrating stakeholders views, and (v) regional integration [23.2]. 110. The toolbox for developing integrated approaches to N management includes systems analyses, communication, integrated assessment modeling, N budgeting, stakeholder dialogue and chain management [15, 16, 22, 23.3, 26]. Integrated approaches are especially applicable to agriculture, because of its multiple sources, multiple N species, and multiple interÂ�related actors [23.4]. 111. It remains a challenge to define the optimum level of integration for various situations and cases. A package of seven key actions in four sectors is envisaged that should contribute to further developing integrated approaches to N management (Figure TS.5) [23.5].

Agriculture (1) Improving nitrogen use efficiency in crop production. (2) Improving nitrogen use efficiency in animal production.

(3) Increasing the fertilizer N equivalence value of animal manure.

Transport and industry (4) Low-emission combustion and energy-efficient systems.

Waste water treatment (5) Recycling nitrogen (and phosphorus) from waste water systems.

Societal consumption patterns (6) Energy and transport saving. (7) Lowering the human consumption of animal protein.

Future scenarios 112. Scenarios of nitrogen use follow the approaches currently used for air pollution, climate, and ecosystem projections. Short-term projections (to 2030) are developed using a ‘baseline’ path of development, which considers abatement options that are consistent with European policy. For medium-term projections (to 2050) and long-term projections, the European Nitrogen Assessment applies a ‘storyline’ approach similar to that used in the IPCC Special Report on Emission Scenarios (SRES) [24.4]. Beyond 2050 in particular, such storylines also take into account technological and behavioral shifts [24.5]. 113. All scenarios agree in projecting a decrease in NOx emissions, while agricultural nitrogen use is expected to remain the leading cause of Nr release to the environment. However,

xlix

Technical Summary Figure TS.11 Total release of NOx, NH3 and N2O to the atmosphere from the EU-27 as estimated by the GAINS model, including scenarios to 2030. The main lines show estimates according to current legislation (‘Climate & Energy package’). Estimates for 2020 are given according to cost optimization by GAINS and according to the maximum feasible reduction based on technical measures only, as included in the GAINS model. The small increase in N2O emissions under the optimized and maximum reduction scenarios is related to estimated pollution trade-offs incorporated into the model [24.6]. Such potential trade-offs highlight the need to prioritize approaches that lead to reductions across the nitrogen cascade [23.5].

4000 3500

Emissions (Gg N/yr)

3000 2500 2000 1500 1000 500 0 2000

2005

2010

2015

2020

2030

GAINS NOx

cost optimization

maximum reduction

GAINS NH3

cost optimization

maximum reduction

GAINS N2O

cost optimization

maximum reduction

integrated assessment modelling including available technical measures indicates that cost-optimized NOx and NH3 emissions for 2030 are substantially smaller than current reduction plans (Figure TS.11), highlighting the case for further emission reductions [24.6]. 114. Major reductions in agricultural Nr emissions will occur only if the extent of agricultural production changes, for example linked to changing human populations or per capita consumption patterns. Such a scenario is examined based on a healthier ‘low meat’ diet leading to lower Nr losses [23.5, 24.5, 26]. The scenario, consisting of 63% less meat and eggs, would reduce NH3 emissions from animal production by 48%. The associated land use changes need to be further explored. For example, a possible increase in intensively fertilized biofuel crops would lead to ‘pollution swapping’, with N2O and Nr losses to water [24.5].

The role of international conventions 115. International treaties, such as multilateral environmental agreements (MEAs), including conventions and their protocols, have done much to protect the global environment through promoting intergovernmental action on many environmental issues. MEAs and inter governmental organizations (IGOs) between them have targeted most environmental problems, but none has targeted nitrogen management policy holistically [4.3, 25.2]. 117. Scientific and technical cooperation between MEAs has proved especially important in identifying the many links between reactive nitrogen threats, with the international scientific community able to provide an important role in

l

2025

harmonizing information (para. 64) and promoting coordination as a foundation for action [25.3]. 118. A new international treaty targeted explicitly on nitrogen could be a powerful mechanism to bring the different elements of the nitrogen problem together. While a new convention targeted on nitrogen would be complex to negotiate and could compete with existing structures, a joint protocol between existing conventions could be effective and should be explored [25.3, 25.4]. 119. The immediate recommendation is to exploit established mechanisms and institutions to develop new coordinating links on nitrogen management between MEAs and IGOs, including the Global Partnership on Nutrient Management facilitated by the United Nations Environment Programme, the Task Force on Reactive Nitrogen of the UNECE Convention on Long-range Transboundary Air Pollution and the links with other UNECE Conventions. There is the opportunity for the UNECE Committee on Environmental Policy to develop the nitrogen management links between the UNECE Conventions, while the European Union and its Member States have important roles to play in harmonization and coordination [25.4].

The role of societal choice and public awareness 120. Public and institutional awareness of the many benefits and threats of nitrogen remains very low. Public understanding is not helped by the complexity of the nitrogen cycle, while it has been insufficiently emphasized how nitrogen links many global change challenges. Increased public awareness has the potential to improve the efficacy of nitrogen

Technical Summary

Figure TS.12 Per capita protein consumption by source in the Netherlands and the EU-27 between 1960 and 2007 (PBL-calculations based on FAO). Reproduced with permission of PBL [26.3].

policies, while reducing the risk of antagonisms with other issues [5.4, 23.2, 26.1]. 121. Food production, consumption and wastage represent key sectors where societal choice can greatly influence N use efficiency, benefits and threats. As an example, because of the low conversion efficiency of plant to animal products, the production of animal proteins releases at least seven times more reactive nitrogen into the environment than the production of the same amounts of plant proteins [26.3]. At the same time, many European citizens are increasingly eating more animal products than is necessary for a healthy diet (Figure TS.12). Even a limited reduction of the share of meat and milk in the European diet would substantially affect the overall N budget of Europe (Figure TS.5) [23.5, 24.5, 26.3].

122. Although this issue belongs to the field of personal choice, public initiatives, as for instance through encouraging healthy eating and reducing food waste in institutional and school catering, can play a significant role in changing behaviours [23.4, 26.3]. 123. Awareness of nitrogen communication tools for analysts, media and European citizens should be increased, for example considering ‘nitrogen footprints’ alongside those for carbon. Succinct messages that convey the nitrogen challenge facing Europe are required. Only by joining efforts between policy makers, producers and the public will we be able to take effective action in managing our ‘nitrogen inheritance’ [1.1, 25.4].

li

Chapter

1

Assessing our nitrogen inheritance Lead author: Mark A. Sutton Contributing authors: Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti

1.1╇ A new challenge from a past solution Human perturbation of the nitrogen cycle represents a major example of global geo-engineering. Historically, the limited availability of reactive nitrogen compounds has provided a key constraint to human activities. Although the element nitrogen is extremely abundant, making up 78% of the Earth’s atmosphere, it exists mainly as unreactive di-nitrogen (N2). By contrast, to be useable by most plants and animals, reactive nitrogen (Nr) forms are needed. These include oxidized and reduced nitrogen compounds, such as nitric acid, ammonia, nitrates, ammonium and organic nitrogen compounds, each of which is normally scarce in the natural environment. The two main historical needs for reactive nitrogen have been to provide fertilizers to increase food production and as a basis for the manufacture of munitions. Biological nitrogen fixation has always added new reactive nitrogen into the system, but the inputs have been barely sufficient for human needs. As a result, traditional agricultural production was highly dependent on effective recycling of nitrogen in manures. By the end of the nineteenth century, an increasing human population combined with expanding military needs required that large amounts of extra reactive nitrogen be added into circulation. These demands were met by increased mining of reactive nitrogen deposits, including Chile saltpetre and guano, supplemented by the extraction of reactive nitrogen from coal and peat (Vincent, 1901; Clow and Clow, 1952; Watt, 2003; Sutton et al., 2008). The western world had effectively become a ‘fossil nitrogen economy’, as both food and military security depended critically on these nitrogen sources (Erisman et al., 2008; Sutton et al., 2009). Increasing dependence on these fossil nitrogen reserves was, naturally, not sustainable. The ‘nitrogen problem’ of the time was that many of the mined nitrogen supplies were becoming exhausted, and that these would soon be insufficient to meet the needs of a rapidly growing world population. As Sir William Crookes famously pointed out to the British Association, if sufficient wheat were to be produced to feed the world, new efforts would be urgently needed to find commercially viable ways of fixing atmospheric di-nitrogen into reactive nitrogen (Crookes, 1898; Leigh, 2004). Potentially, the atmosphere represented a nearly inexhaustible supply from which to

manufacture reactive nitrogen, limited only by the energy costs of chemical production. Efforts at industrial nitrogen fixation were intensified, including development of the cyanamide process and the arc process. Both of these were extremely energy expensive. However, by 1908 Fritz Haber in Germany had filed his patent for the direct ‘synthesis of ammonia from its elements’, in a new process with greatly reduced energy costs (Haber, 1920; Smil, 2001; Leigh, 2004). Following commercial upscaling of Haber’s method by Carl Bosch, large-scale chemical production of reactive nitrogen became economic. The ‘nitrogen problem’ of the early twentieth century rapidly became a thing of the past (Partington, 1925), and by the 1950s, the Haber–Bosch process had replaced fossil reserves as the main source of additional reactive nitrogen. The scale of the Haber–Bosch achievement cannot be overestimated. It represents perhaps the greatest single experiment in global geo-engineering that humans have ever made, underpinning present day food and military security. By allowing the human population to expand, it can equally be considered as laying the foundation for all other aspects of global change. Thus it is only through synthetic nitrogen fertilizers that humanity has been able to reach 6 billion people, around half of whom would not be alive without it (Erisman et al., 2008). Future projections of the human population depend even more strongly on increasing global production of nitrogen fertilizers. Together with human-driven increases in crop biological nitrogen fixation, this great effort of geo-engineering has more than doubled the global production of reactive nitrogen compared with pre-industrial levels (Galloway et al., 2008). This extreme change has knock-on effects in other element cycles, both through direct biogeochemical interactions, and because increased nitrogen has allowed the human population to grow, thereby fuelling additional resource use and global change. It is as a result of this great increase in nitrogen fixation that we now reap our ‘nitrogen inheritance’. In learning to produce additional reactive nitrogen, humans have largely failed to manage its implications for the natural environment. Fundamentally, agricultural practices have a low nitrogenuse-efficiency, especially under increasing nitrogen inputs. As a result, losses of reactive nitrogen to the environment

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

1

Assessing our nitrogen inheritance

have increased greatly, including nitrate pollution of water courses and emissions of both ammonia and nitrous oxide to the atmosphere, with impacts on biodiversity and climate change. In parallel with these changes, humanity has found itself releasing reactive nitrogen directly into the atmosphere through high-temperature combustion processes in industry and transport, which convert atmospheric di-nitrogen to nitrogen oxides. These nitrogen compounds are produced unintentionally and react to form ozone and particles in the air we breathe, damaging human health. Even with this brief listing of issues, it is clear that human alteration of the nitrogen cycle is having consequences that could not have been foreseen a century ago. Just as society has realized the climate implications of fossil fuel combustion, we now appreciate that our nitrogen inheritance is not all good. With reactive nitrogen, humanity has managed to feed the world, but at the same time has created a complex web of impacts occurring through air, land and water that threaten our global environment.

1.2╇ Challenges for a nitrogen assessment Research on alteration of the nitrogen cycle is not new. Together with the environmental consequences of excess reactive nitrogen, we also inherit a wealth of scientific literature that covers the wide diversity of nitrogen processes, sources and impacts. Key scientific resources include the assessments brought together by the Scientific Committee on Problems of the Environment (SCOPE) (Söderlund and Svensson, 1976; Blackburn and Sorenson, 1988) and by the Royal Society of London (Stewart and Rosswall, 1982), including a recent assessment of the nitrogen cycle in agriculture (Mosier et al., 2004). While these studies provide an important foundation, it must be recognized that there has never before been a

comprehensive continental-scale assessment covering all the main effects of nitrogen. This may be partly a reflection of the extreme complexity of the nitrogen cycle, with scientific specialization tending to separate the different research communities. It is easy to see, for example, how research into the atmospheric chemistry of oxidized nitrogen species can be sufficient to develop scientific communities that have little awareness of research into carbon-nitrogen interactions in forestry, of the biodiversity implications of nitrogen deposition, or of the consequences of nitrogen-phosphorus interactions for ‘dead zones’ in coastal waters. In parallel with the trend toward scientific specialization, environmental policies have gradually begun to address the threats posed by excess reactive nitrogen. Such policy domains have developed in a piecemeal fashion until now, generally focusing on individual threats. With the needs of existing policies feeding research agendas, this has led to further separation of the many nitrogen research communities. The consequence of these trends is that, in Europe at least, we now inherit a huge scientific expertise on different aspects of the nitrogen problem, together with a disparate set of nitrogen-relevant policies. In principle, there is therefore a huge resource to develop mitigation strategies to the different environmental threats. By contrast, the limited degree of coordination between these activities can mean that they are far from optimal. After the production and release of a new reactive nitrogen molecule, it may be transformed many times, having multiple effects in the environment before eventually being immobilized or denitrified back to di-Â�nitrogen. This idea of a ‘nitrogen cascade’ including many nitrogen forms and impacts means that substantial interactions can be expected between different policies related to nitrogen (Figure 1.1). In principle, there are serious risks, for example, as measures to reduce nitrogen water pollution

Figure 1.1 Simplified summary of the nitrogen cascade illustrating the example losses, transform­ ation and effects of reactive nitrogen (Nr) fertilizers in the environment. Similar images can be drawn to illustrate the cascade of Nr forms following fossil fuel emissions and crop biological nitrogen fixation.

2

Mark A. Sutton

can increase reactive nitrogen pollution in the air, illustrating the potential for ‘pollution swapping’. By contrast, there are also major opportunities to exploit synergies and to identify Â�win–win situations, where several forms of nitrogen pollution are reduced at the same time, while also minimizing other environmental threats. With these issues in mind, it is clear that the challenges faced by a nitrogen assessment are rather different to the challenges of recent climate assessments. Rather than focusing on a single central question of whether human alteration of the nitrogen cycle is leading to environmental change, the present assessment takes the undoubted evidence of nitrogen impacts as a starting point. The central challenge of the present assessment is to draw together the different aspects of ‘nitrogen change’ to develop a more coherent understanding of how they fit together. This must be the foundation for identifying options for better nitrogen management, and for explaining the key messages to society. In this respect, the present assessment can be seen as initiating a process of integration and communication between nitrogen scientists of different disciplines, between scientists, industry and policy makers, and finally between scientists and the public at large. There is no doubt that the complexity of nitrogen forms, processing and impacts in the environment hinders public understanding. The scope of this assessment must therefore include considering how the nitrogen challenge facing humanity can be communicated more effectively.

1.3╇ Approach of the European Nitrogen Assessment In developing a first continental nitrogen assessment, it is clear that Europe has a key role to play. Given its role in the history of industrial nitrogen fixation, there is a matching responsibility on Europe to consider the full consequences of excess nitrogen in the environment. It is thus appropriate that this assessment coincides with the centenary of Fritz Haber’s discovery (Erisman et al., 2008).

In most areas of Europe, there is a surplus of reactive nitrogen, both from agricultural fertilizer inputs and from combustion-based nitrogen oxide emissions. The main focus of the European Nitrogen Assessment (ENA) is therefore on quantifying and managing the environmental impacts. We recognize, by contrast, that, in other parts of the world, a shortage of reactive nitrogen still limits food production. The ENA will thus complement other continental nitrogen assessments, together with which, it will provide a platform to start addressing the wider global picture. In refining the European Nitrogen Assessment, a close interaction has been developed with key stakeholders, including science communities, policy makers, industry representatives and environmental managers (Figure 1.2). The ENA itself has been developed through a cluster of several European networks, under the lead of the Nitrogen in Europe (NinE) framework research programme of the European Science Foundation (ESF). NinE has worked to integrate scientific Â�understating across nine main interlinked environmental concerns of excess nitrogen, drawing on both national and EU activities (NinE, 2010; Bleeker et al., 2008). Major contributions to the ENA have been provided through the COST 729 Action on ‘Managing Nitrogen in the Atmosphere-Biosphere System in Europe’, which examines nitrogen impacts and policy interactions including the development of integrated assessment approaches (COST, 2010). In addition, underpinning research, especially on nitrogen-greenhouse gas interactions, has been provided by the NitroEurope Integrated Project (NEU, 2010; Sutton et al., 2007, 2009), funded by the European Commission 6th Framework Programme. Together, these activities provide key inputs to the International Nitrogen Initiative (INI), a joint project of the Scientific Committee on Problems of the Environment (SCOPE) and the International Geosphere Biosphere Programme (IGBP). The INI is structured around the activities of six regional centres, with the ENA representing a contribution of the INI European Centre. By having a global oversight, the INI provides

Figure 1.2 The place of the European Nitrogen Assessment in relation to major research networks, integrating activities and policy frameworks.

3

Assessing our nitrogen inheritance

a coÂ�ordinating forum with which other continental nitrogen assessments are being developed. By addressing the linkages between issues, the ENA is specifically designed to support the work of several international conventions. These include: • the Convention on Long-range Transboundary Air Pollution (CLRTAP) of the United Nations Economic Commission for Europe (UNECE), • the UN Framework Convention on Climate Change (FCCC), • the UN Convention on Biological Diversity (CBD), • the Helsinki Convention on the Protection and Use of Transboundary Watercourses and International Lakes (Water Convention) of the UNECE, • the UNECE Espoo Convention on Environmental Impact Assessment in a Transboundary Context (TEIA Convention), • the Marine Conventions for the North Atlantic and North€Sea, Baltic Sea and Mediterrean Seas:€respectively, the Oslo and Paris Commission (OSPARCOM), the Helsinki Commission (HELCOM) and the Barcelona Convention. Given the scale of activity in each of these different conventions, coupled with the substantial legislative programme of the European Union, the challenge of communication between conventions is substantial. The CLRTAP has a long-standing experience of developing multi-pollutant multi-effect stratÂ� egies, such as in the Gothenburg Protocol. At the same time, many of the environmental effects of reactive nitrogen are connected to long-range transport in the atmosphere. Recognizing this challenge, in 2007 the CLRTAP established the Task Force on Reactive Nitrogen (TFRN) (UNECE, 2007). This Task Force works to investigate the linkages between nitrogen issues across the CLRTAP and encourage communication of the nitrogen-related interactions between conventions, as a base to develop more effective mitigation strategies. Considering the obvious affinities, the TFRN adopted the European Nitrogen Assessment into its work plan, thereby providing an important bridge to policy makers in the CLRTAP and the other international conventions. The European Nitrogen Assessment process represents a four year effort over 2007 to 2011. The outline structure and objectives of the ENA were initially developed through a workshop with key stakeholders in Schaagen, the Netherlands, in January 2007, being refined through subsequent meetings linked to NinE, COST 729, NitroEurope and the CLRTAP. Based on the structure established, the ENA process was developed as a series of five main workshops taking place through 2008–2009, linked to five main sections of the assessment. For each of the workshops, background documents were invited as a basis to inform working group discussions, from which draft chapters of the ENA were prepared. Following successive tuning between different parts of the assessment, revised chapters were reviewed within the wider ENA team. Finally, each of the chapters was subjected to international peer review prior to publication.

4

1.4╇ Overall goal and structure of the European Nitrogen Assessment Based on the developments described above, the overall goal of the European Nitrogen Assessment was established as being to review current scientific understanding of nitrogen sources, impacts and interactions across Europe, taking account of current policies and the economic costs and benefits, as a basis to inform the development of future policies at local to global scales. In taking this approach, it was recognized that the ENA required the involvement of a wide variety of actors, focused especially in bringing together scientists of different disciplines, together with economists and experts in policy development. Given the magnitude of this challenge, the ENA was developed to establish a process of gradual integration between communities. The following main workshops were held as the basis for the matching parts of this assessment. I╇ Nitrogen in Europe:€the present position. The focus of this part is to take stock of the current nitrogen challenges faced by Europe. The scene is set by considering both the European environmental threats in the global context (Chapter€2; Erisman et€al., 2011) and the significant benefits of reactive nitrogen production (Chapter 3; Jensen et al., 2011). The multiplicity of current European policies relevant to nitrogen in the environment is then reviewed (Chapter 4; Oenema et€ al., 2011a), followed by a reflection of the developing approach taken by the ENA to link different science and policy areas more closely (Chapter 5; Sutton et al., 2011). II╇ Nitrogen processing in the biosphere. The aim of this part is to review recent progress in scientific understanding of the nitrogen cycle and to highlight the major uncertainties. The focus on processes deliberately emphasizes the fundamental interactions between the many forms of nitrogen as these occur in different environmental compartments. Recognizing the specialized nature of the nitrogen research communities, this part makes a first step toward integration by including consideration of all different nitrogen forms into each chapter, while retaining the distinctive expertise of research communities on terrestrial ecosystems (Chapter 6; Butterbach-Bahl et€al., 2011a), freshwater aquatic ecosystems (Chapter 7; Durand et€ al., 2011), coastal and marine ecosystems (Chapter 8; Voβ et al., 2011) and, finally, nitrogen processing in the atmosphere (Chapter 9; Hertel et al., 2011). III╇ Nitrogen flows and fate at multiple spatial scales. The next step of integration, as addressed in this section, is to scale up nitrogen processes through the range of different spatial domains. For this part of the assessment, science communities were increasingly linked between environmental compartments in order to assess the key interdisciplinary concerns in managing the fate of nitrogen in the environment. The section starts with consideration of nitrogen management at the farm scale, considering the variation in typical nitrogen flows across Europe (Chapter 10; Jarvis et al., 2011). The assessment then increases in scale to consider how adjacent sources and sinks of nitrogen interact within rural landscapes (Chapter 11; Cellier et al., 2011) and urban landscapes, with a focus on

Mark A. Sutton

example cities (Chapter 12; Svirejeva-Hopkins et al., 2011). The regional scale transport of reactive nitrogen is then addressed, contrasting the transfers through regional scale watersheds into coastal marine areas (Chapter 13; Billen et al., 2011) to the atmospheric transport and deposition of reactive nitrogen (Chapter 14; Simpson et al., 2011). The approach of developing nitrogen budgets provides a means to consider how different components of the nitrogen cycle fit together (Chapter 15; De Vries et€al., 2011), followed by the integrated picture of nitrogen fluxes across Europe (Chapter 16; Leip et al., 2011). IV╇ Managing nitrogen in relation to key societal threats. Given the multi-dimensional complexity of reactive nitrogen forms and impacts on the environment, significant effort was given to distilling out the key threats, as discussed in Chapter€5 (Sutton et al., 2011). The idea was to identify a short list of key threats, to which the main concerns of nitrogen could be linked, as a means of visualizing the problem and encouraging more effective communication. The resulting chapters in this section review the consequences of nitrogen in Europe for five key threats:€ water quality (Chapter 17; Grizzetti et al., 2011), air quality (Chapter 18; Moldanová et al., 2011), greenhouse balance (Chapter 19; Butterbach-Bahl et al., 2011b), terrestrial ecosystems and biodiversity (Chapter 20; Dise et al., 2011) and soil quality (Chapter 21; Velthof et al., 2011). Each of the chapters aims to give evidence of how the threat has developed over time, to highlight the future prospects and to indicate the focus of current mitigation approaches. V╇ European nitrogen policies and future challenges. The final part of the assessment brings together the key threats to consider how nitrogen might be managed more effectively in the future. The first challenge is to see how different nitrogen threats may be inter-related, including assessment of the costs and benefits of different nitrogen forms in the environment (Chapter 22; Brink et al., 2011). Together with preceding material, this provides a basis to review the options for integrated approaches to manage nitrogen in the environment (Chapter 23; Oenema et al., 2011b). Linked future nitrogen scenarios are then addressed, bringing together each of the main nitrogen threats (Chapter 24; Winiwarter et al., 2011). Two last chapters explore how the developing messages for reactive nitrogen can be communicated. The first considers how to develop coordination between different European policies and international conventions in which nitrogen plays an important role (Chapter 25; Bull et al., 2011). The second considers how the scientific perspective can be distilled to communicate the nitrogen challenge to society at large (Chapter 26; Reay et al., 2011). In reviewing the overall product, it is clear that the ENA has contributed substantially to bringing the different research, policy and other stakeholder communities together. In this respect the ENA should be considered as much an ongoing process, as it is a book. The present volume thus represents a milestone along the road, rather than a final destination. The nitrogen story provides a clear example of the great benefits of global geo-engineering. At the same time it provides a warning, demonstrating the complexity and extent of the unanticipated environmental effects. While even more reactive nitrogen will be needed in future, we should aim to pass on an

inheritance that allows for a more-sustainable management of this precious resource.

References Billen, G., Silvestre, M., Grizzetti, B. et al. (2011). Nitrogen flows from European watersheds to coastal marine waters. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Blackburn, T. H. and Sorenson, J. (1988). Nitrogen Cycling in Coastal Marine Environments. (eds.) SCOPE report no. 33. J. Wiley and Son, Chichester. Bleeker, A., Reis, S., Britton, C., Erisman, J. W. and Sutton, M. A. (2008). Nitrogen in Europe:€activities addressing the European nitrogen cycle. (Actividades relacionadas con el ciclo del Nitrógeno en Europa.) Seguridad Y Medio Ambiente, 111, 6–7, 22–31 (in English and Spanish). Brink, C., van Grinsven, H., Jacobsen, B. H. et al. (2011). Costs and benefits of nitrogen in the environment. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Bull, K., Hoft, R., Sutton, M. A. (2011). Co-ordinating European nitrogen policies between directives and international conventions. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C.€M.€Howard, J. W. Erisman et al. Cambridge University Press. Butterbach-Bahl, K., Gundersen, P., Ambus, P. et al. (2011a). Nitrogen processes in terrestrial ecosystems. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Butterbach-Bahl, K., Nemitz, E., Zaehle, S. et al. (2011b). Nitrogen as a threat to the European greenhouse balance. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Cellier, P., Durand, P., Hutchings, N. et al. (2011). Nitrogen flows and fate in rural landscapes. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Clow, A. and Clow, N. L. (1952). The Chemical Revolution: A Contribution to Social Technology. Reprinted 1970. Books for Libraries Press, Freeport, New York. COST (2010). www.cost729.org Crookes, W. (1898). Presidential Address to the British Association for the Advancement of Science 1898. Chemical News, 78, 125. De Vries, W., Leip, A., Reinds, G. J. et al. (2011). Geographic variation in terrestrial nitrogen budgets across Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W.€Erisman et al. Cambridge University Press. Dise, N. B., Ashmore, M., Belyazid, S. et al. (2011). Nitrogen as a threat to European terrestrial biodiversity. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Durand, P., Breuer, L., Johnes, P. et al. (2011). Nitrogen processes in aquatic ecosystems. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Erisman J. W., Sutton, M. A., Galloway, J. N., Klimont, Z. and Winiwarter, W. (2008). How a century of ammonia synthesis changed the world. Nature Geoscience, 1, 636–639. Erisman, J. W., van Grinsven, H., Grizzetti, B. et al. (2011). The European nitrogen problem in a global perspective. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press.

5

Assessing our nitrogen inheritance Galloway, J. N., Townsend A. R., Erisman J. W. et al. (2008). Transformation of the nitrogen cycle:€recent trends, questions and potential solutions. Science, 320, 889–892. Grizzetti, B., Bouraoui, F., Billen, G. et al. (2011). Nitrogen as a threat to European water quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Haber, F. (1920). The synthesis of ammonia from its elements. Nobel Lecture (1920), available at www.nobelprize.org/nobel_prizes/ chemistry/laureates/1918/haber-lecture.pdf Hertel, O., Reis, S., Ambelas Skjøth, C. et al. (2011). Nitrogen processes in the atmosphere. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Jarvis, S., Hutchings, N., Brentrup, F., Olesen, J. and van der Hoek, K. (2011). Nitrogen flows in farming systems across Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Jensen, L. S., Schjoerring, J. K., van der Hoek, K. et al. (2011). Benefits of nitrogen for food fibre and industrial production. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Leigh, G. J. (2004). The Worlds’ Greatest Fix: A History of Nitrogen and Agriculture. Oxford University Press. Leip, A., Achermann, B., Billen, G. et al. (2011). Integrating nitrogen fluxes at the European scale. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Moldanová, J., Grennfelt, P., Jonsson, Å. et al. (2011). Nitrogen as a threat to European air quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Mosier, A. R., Syers, J. K. and Freney, J. R. (eds.) (2004). Agriculture and the Nitrogen Cycle, SCOPE report no. 65. Island Press, Washington, DC. NEU (2010). www.nitroeurope.eu Oenema, O., Bleeker, A., Braathen, N. A. et al. (2011a). Nitrogen in current European policies. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Oenema, O., Salomez, J., Branquinho, C. et al. (2011b). Development of integrated approaches to nitrogen management. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Partington (1925). Textbook of Inorganic Chemistry, 2nd edition, Macmillan, London. NinE (2010). www.nine-esf.org Reay, D. S., Howard, C. M., Bleeker, A. et al. (2011). Societal choice and communicating the European nitrogen challenge. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press.

6

Simpson, D., Aas, W., Bartnicki, J. et al. (2011). Atmospheric transport and deposition of nitrogen in Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Smil, V. (2001). Enriching the Earth: Fritz Haber, Carl Bosch, and the Transformation of World Food Production. MIT Press, Cambridge, MA. Stewart, W. D. P. and Rosswall, T. (eds.) (1982). The Nitrogen Cycle, The Royal Society, London. Sutton, M. A., Nemitz, E., Erisman, J. W. et al. (2007). Challenges in quantifying biosphere-atmosphere exchange of nitrogen species. Environmental Pollution, 150, 125–139. Sutton, M. A., Erisman, J. W., Dentener, F. and Moeller, D. (2008). Ammonia in the environment:€from ancient times to the present. Environmental Pollution, 156, 583–604. Sutton, M. A., Reis, S. and Butterbach-Bahl, K. (2009). Reactive nitrogen in agro-ecosystems:€integration with greenhouse gas interactions. Agriculture, Ecosystems and Environment, 133, 135–138. Sutton, M. A., Howard, C. M., Erisman J. W. et al. (2011) The need to integrate nitrogen science and policies. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Svirejeva-Hopkins, A., Reis, S., Magid, J. et al. (2011). Nitrogen flows and fate in urban landscapes. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Söderlund, R. and Svensson, B.H. (1976). The global nitrogen cycle. In:€Nitrogen, Phosphorus and Sulphur: Global Cycles, eds. B.€H.€Svensson and R. Söderlund, SCOPE report no. 7, Ecological Bulletin, Stockholm, 22, 23–73. UNECE (2007). Establishment of a Task Force on Reactive Nitrogen. Executive Body of the Convention on Long-Range Transtoundary Air Pollution, Decision 2007/1, ECE/EB.AIR/ 91/Add.1. Velthof, G., Barot, S., Bloem, J. et al. (2011). Nitrogen as a threat to European soil quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Vincent, C. (1901). Ammonia and Its Compounds, Their Manufacture and Uses (trans. M. J. Salter ). Scott, Greenwood & Co., London. Voß, M., Baker, A., Bange, H. W. et al. (2011). Nitrogen processes in coastal and marine systems. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Watt, J. G. (2003). A brief history of the Chilean nitrates industry. CIM Bulletin, 96, 84–88. Winiwarter, W., Hettelingh, J. P., Bouwman, L. et al. (2011). Future scenarios of nitrogen in Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press.

Part

I

Nitrogen in Europe:€the present position

Chapter

2

The European nitrogen problem in a global perspective Lead author: Jan Willem Erisman Contributing authors: Hans van Grinsven, Bruna Grizzetti, Fayçal Bouraoui, David Powlson, Mark A. Sutton, Albert Bleeker and Stefan Reis

Executive summary Nature of the problem • Reactive nitrogen has both positive and negative effects on ecosystems and human health. Reactive nitrogen is formed through the use of fossil fuels releasing large amounts of nitrogen oxides into the atmosphere and through the production of ammonia by the Haber–Bosch process and using it in agriculture to increase our food, feed and fuel production. While the use of nitrogen as a fertilizer and chemical product has brought enormous benefits, losses of fertilizer nitrogen and combustion nitrogen to the environment lead to many side effects on human health, ecosystem health, biodiversity and climate.

Approaches • The European nitrogen problem is placed in a global perspective, showing the European nitrogen fixation, transport and environmental impacts compared with different regions of the globe.

Key findings/state of knowledge • Humans, largely through agriculture, but also through burning of fossil fuels, have had a huge impact on the nitrogen budget of the Earth. Europe is one of the leading producers of reactive nitrogen, but it is also the first region in the world where the issue was recognized and in some parts of Europe the reactive nitrogen losses to the environment started to decrease. Europe is a nitrogen hotspot in the world with high nitrogen export through rivers to the coast, NOx and particulate matter concentrations and 10% of the global N2O emissions. • The consequences of nitrogen losses in Europe are visible and are on the average more pronounced than in the rest of the world. Nitrogen contributes to all environmental effects to some extent. • There is a clear policy on reducing nitrogen oxide emissions that led to reductions by implementation of end of pipe technology. Europe is ahead compared to the rest of the world with NOx policies. • Fertilizer production and use decreased in Europe in the early 1990s, in particular, due to the economic recession in the Eastern part of Europe. Currently, the fertilizer use in Europe is about 12 Mton, which is 4 Mton lower than in the 1980s, but increasing again. The nitrogen use efficiency of nitrogen in the EU, defined as the net output of N in products divided by the net input is about 36%. This is lower than the world average (50%) as fertilization rates in Europe are much higher.

Major uncertainties/challenges • More quantification of the effects is needed to establish cause–effect relationships. Most is known about the exceedances of critical limits, but more quantitative results are needed on impacts, including biodiversity loss, ground water pollution and eutrophication of ecosystems; eutrophication of open waters and coastal areas resulting in algal blooms and fish kills; increased levels of NOx and aerosols in the atmosphere resulting in human health impacts and climate change; and the increased emissions of the greenhouse gas nitrous oxide resulting in climate change. The effects of nitrogen affecting the other biogeochemical cycles such as carbon and phosphorus need to be quantified on different scales. • The complexity of multi-pollutant–multiple-effect interactions is a major hurdle to improving public awareness.

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

9

The European nitrogen problem in a global perspective

2.1╇ Introduction Nature and its biodiversity could only exist because of the availability, even if limited, of reactive nitrogen (Nr) in the system, which is defined as all nitrogen compounds except for N2. This reactive nitrogen was provided by limited natural sources such as lightning, biomass burning and biological nitrogen fixation. Because of the limited availability, nature became very effective in conserving and re-using reactive nitrogen compounds. Nitrogen, together with other nutrients and water, is the limiting factor for the production of food. Mankind has sought for different ways to increase the crop production necessary for food to sustain a growing population. This has led to the development of synthetic fertilizer production based on the Haber–Bosch process (Smil, 2001; Erisman et al., 2008). This additional availability of reactive nitrogen has led to increased crop production and to the intensification of agriculture. The large increase in population is due to intensification and extension of agricultural land, but also due to the availability of fertilizers. A recent estimate of the current human population supported by synthetic fertilizer is 48%, 100 years after the invention of the synthesis of ammonia from its elements (Erisman et al., 2008; Figure€2.1). To maximize crop production, the availability of cheap fertilizer in the industrialized world led to excessive use of nitrogen, resulting in a large nitrogen surplus and increased nitrogen losses. As the use of fossil fuel in the industrial revolution expanded, fertilizer production increased similarly (see Figure 2.1). The industrial revolution was accelerated by the combustion of fossil fuels producing heat and power, but also polluting gases, such as carbon dioxide, sulphur dioxide and nitrogen oxides. The use of fossil fuels led at the same time to an increase in the production of fertilizer through the Haber–Bosch process, and to a replacement of manpower by machines increasing the productivity and yield per hectare,

7000

6000

world population

50

world population (no Haber–Bosch N)

5000

40

average fertilizer input (kg N/ha/year) meat production (kg/person/year)

4000

30

3000 20 2000 10

1000

0 1900

10

0 1920

1940

1960

1980

2000

% World population, fertilizer input, meat production

World population (millions)

% world population fed by Haber–Bosch N

further accelerating excess nitrogen. Furthermore, the availability of fossil fuels made globalization possible, transporting food, feed, goods and products all over the world, and depleting nutrients in one area and concentrating nutrients in another area, e.g. in intensive livestock production (Galloway et al., 2008). These leakages from agriculture, industry and transport, in their turn, have led to a cascade of N through the global environment causing a number of different environmental effects:€loss of biodiversity, eutrophication of waters and soils, drinking water pollution, acidification, greenhouse gas emissions, human health risks through exposure to oxidized nitrogen (NOx), ozone (O3) and particulates, and destruction of the ozone layer. Europe has benefited to a large extent from the increase in nitrogen, both economically as well as socially (see Jensen et al., 2011, Chapter 3 this volume). Agriculture has contributed to a large extent to GDP development and, apart from some poorer areas in Europe, hunger is no longer a key issue. The situation has, however, developed into overuse of nitrogen in agriculture as a straightforward ‘cheap’ insurance against low yields with all the concomitant negative side effects. Therefore the focus is to deal with the unwanted downside:€ optimizing use while minimizing adverse effects. This chapter of the European Nitrogen Assessment (ENA) provides an overview of the European nitrogen problem in a global perspective. The chapter reviews existing knowledge, bringing different studies together to assess the European nitrogen situation relative to different priorities in other areas in the world. The specific processes and effects are addressed in more detail in the following chapters in this book. This chapter starts with an introduction on reactive nitrogen formation in nature, agriculture and through fossil fuel combustion. Then the nitrogen fluxes are described, including the losses to air and water, followed by a section describing the negative effects of nitrogen in Europe in a global perspective.

Figure 2.1 Trends in human population and nitrogen use throughout the twentieth century (Erisman et al., 2008). Of the total world population (solid line), an estimate is made of the number of people that could be sustained without reactive nitrogen from the Haber–Bosch process (dashed line), also expressed as a percentage of the global population (short dashed line). The recorded increase in average fertilizer use per hectare of agricultural land (blue symbols) Â� and the increase in per capita meat production (green symbols) are also shown.

Jan Willem Erisman

2.2╇ Reactive nitrogen

Table 2.1╇ Examples of nitrogen components and their oxidation state

Reactive nitrogen, Nr, is defined here as all other nitrogen forms in our system apart from N2. This includes oxidized nitrogen, mainly NO, NO2, NO3; reduced forms of nitrogen:€NH4+, NH3 and organic nitrogen:€ proteins, amines, etc., with different states of oxidation (Table 2.1) Natural sources of the formation of Nr include volcanoes, biological nitrogen fixation in natural soils and lightning (Figure 2.2, Box 2.1) (Smil, 2001; Reid etâ•›al., 2005; Schlesinger, 2009; Sutton et al., 2008 and Hertel et al., 2011 (Chapter 9 this volume), Simpson et al., 2011 (Chapter 14 this volume), where details are provided) and weathering of rocks (Holloway and Dahlgren, 2002). Senescence of plants, wildlife and forest fires are natural processes that result in the re-distribution of Nr in the biosphere. Most of the abiotic natural sources of Nr are in oxidized forms, although wildlife and volcanos also emit reduced forms (Galloway et al., 2003; Schlesinger, 2009). In Europe natural sources of Nr are estimated to create annually 2–3 Mton Nr (van Egmond et al., 2002; Galloway et al., 2004). Anthropogenic activities enhancing the formation of Nr include cultivated Biological

Oxidation state

Example

Component name

−3

NH3

Ammonia

−2

NH2NH2

Hydrazine

Reduced forms

−1

HNNH

Diimide

0, non-reactive

N2

Di-nitrogen

Oxidized forms +1 +2

NO

Nitrogen oxide

+3

HNO2

Nitrous acid

+4

NO2

Nitrogen dioxide

+5

HNO3

Nitric acid

Nitrogen Fixation (BNF) in agriculture, N2 fixation through the Haber–Bosch process, the burning of fossil fuels and forest fires (Figure 2.2).

Figure 2.2 The Nr cycle and the main fluxes (picture by Anne-Christine LeGall).

11

The European nitrogen problem in a global perspective

Of the major nutrients needed for biomass production, nitrogen is most commonly the limiting one, at least in terrestrial systems (De Vries et al., 2011, Chapter 15 this volume). Phosphorus and potassium are the other major limiting nutrients. Higher plants can mostly only use nitrogen after it has been converted to reactive forms such as nitrate (NO3−) or ammonium (NH4+). There are basically three ways of accomplishing this: • decay of organic matter by microbes in soils (dead organisms, leaves, manure, etc.) and eventually release biologically available forms of N; • Biological Nitrogen Fixation (BNF):€nitrogen-fixing organisms (e.g. bacteria) ‘fix’ atmospheric N2 into biologically available forms of reactive nitrogen; • production and application of reactive nitrogen as inorganic fertilizer.

2.2╇ Human intervention in the nitrogen cycle The nitrogen cycle with an explanation of the most important processes is given in Box 2.1. Humans increase the creation of Nr by three processes. The first process is the combustion of fossil fuels for energy production, which generates nitrogen oxides from the oxidation of N2 or fossil organic N in the fuel. Second, the production of fertilizers and chemicals (e.g. nylon, explosives), mainly through the Haber–Bosch process, which creates NH3 by the reaction of N2 and H2. And third, the planting of nitrogen-fixing crops (e.g. legumes) which convert N2 to NH3 incorporated in the organic matter. The primary emission of oxidized nitrogen is through the formation of nitrogen oxides (NO and NO2) and nitrous oxide (N2O). Through reactions in the atmosphere other oxidized nitrogen compounds are formed. Oxidized nitrogen has several sources, but is formed mainly by combustion processes where fuel N is oxidized or atmospheric N2 is oxidized at high temperatures. These processes occur in industries, fuel combustion for transportation and energy production. Among the other sources of oxidized nitrogen, soils are most important (Skiba et al., 1994; 1997). The Haber–Bosch process has facilitated the production of agricultural fertilizers on an industrial scale, dramatically increasing global agricultural productivity in most regions of the world. The number of humans supported per hectare of arable land has increased from 1.9 to 4.3 persons between 1908 and 2008 (Erisman et al., 2008). This increase was only possible because of nitrogen from the Haber–Bosch process. An additional use of fertilizer is the production of crops and biomass for bioenergy and biofuels. Currently, bioenergy contributes 10% to the global energy use, while biofuels contribute 1.5% and the influence on global fertilizer use is still marginal. However, present climate and energy policies tend to stimulate biofuel production so that the influence of Haber–Bosch nitrogen will tend to grow, depending on which soils and crops are used and in how far N-efficiencies in food production can be increased (Erisman et al., 2009). The environmental impacts

12

of inefficient use of fertilizer and livestock breeding systems result in groundwater pollution, airborne emission of ammonia and nitrous oxides, contributing to excess nitrogen cascading through the environment with negative impacts on human health and ecosystem services. Sources of reduced nitrogen include emissions from the fertilizer industry and/or other industry applying ammonia as a Box 2.1 ╇ The global nitrogen cycle and the main processes The nitrogen cycle consists of the following main processes. •

Biological Nitrogen Fixation. 2N2 + 3H2 → 2NH3. This is a natural process performed by a number of diazotrophs, such as Anabaena (a cyanobacterium), and Rhizobium (the symbiotic bacterium found in legume root nodules).

•

Nitrogen fixation. Humans influence the nitrogen cycle through industrial N fixation. Conversion of N2 into NH3 by the Haber–Bosch process.

•

Nitrification. This is the oxidation of ammonia to oxyanions. The initial oxidation to nitrite, 2NH3 + e− + 3O2 → 2NO2− + 2H2O + 2H+, is performed by bacteria such as Nitrosomonas and the next step oxidation to nitrate, 2NO2− + O2 → 2NO3− is performed by Nitrobacter.

•

Denitrification. Some bacteria, such as Pseudomonas, are able to use nitrate as a terminal electron acceptor in respiration:€2NO3− + 12H+ + 10e− → N2 + 6H2O.

•

Assimilation. Plants assimilate nitrogen in the form of nitrate and ammonium. The nitrate assimilated is first reduced to ammonium, and then combined into organic forms, generally via glutamate. Animals generally assimilate nitrogen by first breaking protein down into amino acids.

•

Decay (ammonification) and excretion. When plants and animals decay, putrefying bacteria produce ammonia from the proteins they contain. Animals also produce breakdown products such as ammonia, urea, allantoin and uric acid from excess dietary nitrogen. These compounds are also targets of ammonification by bacteria.

•

Annamox reaction. This is conversion of nitrite and ammonium to pure nitrogen gas (N2) in seas and oceans, hot springs, hydrothermal vents, and many freshwater wetland ecosystems, which than escapes to the atmosphere. The reaction mechanism is triggered by a newly discovered bacterium, called Brocadia anammoxidans.

•

Volatilization. Turns fertilizers and manures on the soil surface into gases like NH3, N2O and N2 that also join the atmospheric pool.

•

Weathering of rocks. The process where stored nitrogen in rocks is released by wind, rain and erosion.

•

Runoff. Carries the nitrogen in fertilizers and manure and the nitrogen in the soil into rivers and streams causing a concern for water quality.

•

Leaching. Carries nitrates deep into the soil so that plants can no longer use them, producing a dual concern; for lost fertility and for water quality, as nitrates enter the groundwater and wells that provide drinking water.

Jan Willem Erisman

chemical, from fertilizer application, and plant senescence, from livestock manure (in the field, housing systems and application of manure), cars equipped with three way catalysts to reduce NOx, industrial emissions applying urea or ammonia Selective Catalytic Reduction (SCR) to reduce emissions, households using ammonia as a cleaning agents and excretion by cats and dogs. Total denitrification of nitrate by microbes in soils, groundwater and surface water is considerable. Particularly when nitrate concentrations are comparatively high, denitrification is incomplete and small quantities of N2O are released in the process. N2O is also released during nitrification. N2O is also a by-product in the Haber–Bosch process of synthetic fertilizer production.

2.3╇ Nitrogen use in agriculture Europe is one of the world’s largest and most productive suppliers of food and fibre. In 2004 Europe produced 21% of global meat production and 20% of global cereal production. About 80% occurred in Europe, defined here as the 25 European countries, EU25 (IPCC, 2007). The productivity of European agriculture is generally high, in particular in Western Europe:€average cereal yields in the EU are more than 60% higher than the global average (EFMA, 2010). The major sources of Nr in agriculture are Biological Nitrogen Fixation and the use of inorganic and organic fertilizers. This section will outline the major sources for European conditions comparing this with global context. Important for the environmental aspect of agricultural nitrogen is the nitrogen use efficiency. Finally, the influence of bioenergy on the agricultural nitrogen is described.

2.3.1╇ Biological Nitrogen Fixation (BNF) Biological nitrogen fixation (BNF) is a vital biological process which allows atmospheric molecular di-nitrogen (N2) to be converted into mineral nitrogen (NH3) that can then be assimilated by living organisms. This process is carried out by specific N-fixing bacteria that are either free-living in soil or water or associated with the root nodules of legume plants. Global terrestrial BNF is estimated at 120 Tg N/yr, of which a little less than half is fixed in oceans, the rest on land (Smil, 2001). BNF in Europe is estimated at 14.8 Tg N/yr in natural soils (Galloway et al., 2004). Crop BNF, the human induced BNF, was estimated by Galloway et al. (2004) to be 30 Tg N on the global level and for Europe 3.9 Tg N/yr, while Velthof et al. (2009) estimated for Europe a value of 5 Tg N/yr. The distinction between natural and crop BNF is difficult to make when fertilizers are added or atmospheric deposition substantially contributes to nitrogen inputs as this can suppress BNF (van Kessel and Hartley, 2000).

2.3.2╇ Nitrogen in mineral and organic fertilizer Mineral fertilizer Nitrogen is an essential element for plant growth, being a component of chlorophyll, amino acids, proteins and enzymes

and increased nitrogen application leads to higher crop production (see e.g. Olson and Kurtz, 1982). Sufficient supply of nitrogen is required for plant metabolism, and addition of N will essentially increase the efficiency of photosynthesis to produce carbohydrates. Higher inputs of nitrogen have increased yields as shown in Figure 2.3 where the changes in yield and fertilizer intensity are plotted for 1961–1990. In the USA and Europe the yield has decreased during recent years showing that the efficiency of the added nitrogen has become less. Nitrogen fertilizers are manufactured by combining atmospheric N2 with hydrogen from methane or gasified coal to produce ammonium nitrate, ammonium sulphate, or urea. Nitrogen fertilizer data throughout the world shows that the annual use rate is increasing (FAO, 2010c; Davidson 2009). For Europe there is no inventory of fertilizer data before 1960. Therefore, we used application rates in kilograms of fertilizer per hectare for the years 1910, 1920, 1935 and 1950 as reported by Moïssey Postan and Rich (1952) to construct the European time series. Figure 2.4 illustrates the trends in amounts of applied fertilizer in Europe compared to the whole world. There is an overall increase in fertilizer application on a global scale. There is some anti-correlation with the gas prices explaining the reductions in application. Throughout the 1930s, European consumption of nitrogenous fertilizers remained above half of the world’s total while the continent’s arable land accounted for only 12% of all cropland. The applications were heavily concentrated in€Germany, Benelux, England and France and applications outside the Northwestern part of the continent remained marginal. After a decrease of production during the Second World War their synthesis and applications began to grow substantially only during the early 1950s. The big drop in 1990–1992 has two main reasons:€ (1) the collapse of the Eastern European countries economy, which had a dramatic effect; and (2) the impact of the McSharry reform (new CAP in 1990). This was the introduction of the mandatory set aside, and the farming community has ‘heavily overreacted’ to this new policy. Then we have seen a slight recovery in the following years (until 1996), shown in Figure 2.4. Europe accounts currently for about 10% of the global use of N fertilizers. After the political changes in Eastern Europe around 1989, the economic breakdown resulted in a strong decrease in fertilizer application, which is reflected in Figure 2.4. The mean national application rate varies very much in Europe from the lowest level of 42 kg/ha of agricultural land in Portugal to 243 kg/ha on grassland in the Netherlands.

Manure and livestock The process of livestock production does not directly lead to the creation of new Nr, but to a concentration of Nr and a redistribution of it over different spatial scales. This occurs through the intake of feed, international and local transport of feed and the emissions to air and soil/water. Therefore, livestock production is very important for Nr as an external

13

The European nitrogen problem in a global perspective Figure 2.3 Overall yield (Y) and fertilizer intensities (FI) for the developing regions, and other countries and regions between 1960 and 1990. The fertilizer intensity is the fertilizer input (expressed as the amount of N + P2O5 + K 2O) as a fraction of total biomass production (FAO, 2010a).

5 4.5 4

Yield (ton cereals/ha)

3.5 3 2.5 2 1.5 1 0.5 0 0

20

40

60

80

Fertilizer intensity (ton/1000 ton NPK/P) Developing countries incl. China Sub-Saharan Africa Europe World

input in different regions or an internal cycling of nitrogen with appreciable losses to the environment. Europe has a large share of animals and animal breeding farms in the world (Table 2.2; Figure 2.5). Meat production in Europe in 2004 was about 20% of the world production, milk production about 30% (source FAOstat:€FAO, 2010c). The number of animals is shown in Table 2.2. The regions of most intensive livestock production in Europe include Denmark, the Netherlands, Belgium, Brittany, Spain, Poland, the UK and the Po€Valley (Italy). The density of pigs per unit area is among the highest in the world together with some regions of China and the USA as can be seen in Figure 2.5. Cattle are more equally distributed over the world and are linked to the grassland areas (land bound). Nitrogen cycling within European livestock production is very important with the total N excretion from animals almost equalling the use of mineral fertilizer in Europe (Oenema et al., 2007). Although livestock production does not directly lead to the production of new Nr, there are relevant indirect interactions. The nitrogen supply for animal feeds is dependent on new Nr supply both through fertilizers and BNF. Thus increasing livestock production, results in additional demands for nitrogen fixation. Livestock manure production in the world was plotted by Davidson (2009) and is shown in Figure 2.4. European

14

North Africa Latin America USA

data are not available for the period before 1960. Data were obtained from Buijsman (1986) and OECD to compile a similar trend for Europe, plotted in Figure 2.4, with the global and European fertilizer use for the same period. The data are highly uncertain. There is a difference between the ratio of manure to fertilizer use globally and that in Europe. Currently, about equal amounts of nitrogen are applied in manure and fertilizers in Europe. Globally, however, the amount of nitrogen applied in manure was an order of magnitude higher than that applied in fertilizer, currently decreased to a factor of two. Apparently much manure is produced globally on unfertilized (grass)lands, whereas in Europe most agricultural land is fertilized. The inputs in agriculture in Europe are further specified and presented in De Vries et al., 2011 (Chapter 15 this volume) and Leip et al., 2011 (Chapter 16 this volume).

2.3.3╇ Nitrogen use efficiency During the last decade the EU Common Agricultural Policy (CAP) has been reformed to reduce overproduction, reduce environmental impacts and improve rural development. This is not expected to greatly affect agricultural production in the short term (OECD, 2001). Excessive application of nitrogen leads to an imbalance as not all applied nitrogen can be taken

12000

200000 World fertilizer

180000

Manure Production World

160000

10000

Europe fertilizer

140000

Manure Production Europe

8000

120000 6000

100000 80000

4000

60000 40000

2000

20000 0

Table 2.2╇ Livestock numbers in Europe and the world in 2005

Livestock (1000 head)

Europe

World

Poultry

1 329 162

15 146 608

Pig

164 794

917 635

Cattle

100 508

1 310 611

Small ruminant

142 476

1 722 175

Source: From Steinfeld et al., 2006

up by the crops. The nitrogen surplus or nitrogen balance is an indicator for the agricultural pressure on the environment. The gross nutrient balance is calculated by subtracting the sum of the total nitrogen output in harvested crops and forage from the total nitrogen input calculated as the sum of total fertilizer N (inorganic fertilizers, organic fertilizers:€organic inputs from non-agricultural sources:€ urban compost and sewage sludge spread on agricultural land), livestock manure production, manure stocks (stock levels, imports and exports of livestock manure), biological nitrogen fixation and atmospheric deposition of nitrogen compounds (EEA, 2005b; OECD, 2001; Campling et al., 2005). Nitrogen use efficiency (NUE) can be defined in different ways. Usually in agriculture it is the nitrogen in the product leaving the farm divided by the nitrogen input to the farm. The nitrogen use efficiency (%), generally increases with a decrease in N input and N surplus as calculated by the nitrogen balance. The efficiency is less than 50% in countries with an N surplus above 80 kg/ha/yr (the Netherlands, Belgium, Denmark and UK), between 50% and 70% in countries with an N surplus between 50–80 kg/ha/yr and more than 70% in countries with an N surplus below 50 kg/ha/yr, except for Portugal and Spain (OECD, 2006).

Figure 2.4 Global and European livestock manure and fertilizer nitrogen consumption (Kton N). Global data are obtained from Davidson (2009). European data are constructed from animal numbers and excretion factors by Buijsman (1986) and the INTEGRATOR model (de Vries et al., 2009).

20 10

19 90

19 70

Year

19 50

19 30

19 10

18 90

18 70

18 50

0

Total N fertilizer and manure in Europe EU27 (Kton N)

Total N fertilizer and manure in the world (kton N)

Jan Willem Erisman

The worldwide database compiled by FAO shows that the NUE has decreased exponentially in all countries except Western Europe and the United States (Figure 2.6). The driver for these trends is the increasing amount of N fertilizer applied in all world regions except Europe. In the analysis, the grain production was computed as the sum of all cereal crops and maize production. Grain production has increased linearly since 1960 in the United States and Western Europe (Hatfield and Prueger, 2004). These changes in grain production have caused a slight increase in NUE in the past decade. However, these trends may hide the effect of manure, which is applied in large amounts. In addition, the NUE of fertilizer in the United States and Western Europe is low because of over-application (see also Figure 2.3). Another approach for defining the NUE is to consider the consumed amounts of calories and of protein as effectively used by humans. Ultimately, nitrogen for food production aims to provide the necessary proteins. Animal protein is much more inefficient in terms of NUE than plant proteins. The more efficiently protein is obtained by humans, the higher the NUE (van Grinsven et al., 2003). Estimates for different regions of the world (Figure 2.7, van Grinsven et al., 2003) show major regional differences in protein consumption per capita. The �differences mainly result from variation in the fraction of protein in diet provided by animal products and in the type of animal product. In the developed countries more animal protein is consumed than in developing countries, where, especially in the low protein countries almost all proteins are consumed through vegetable food products. Rough estimates can be made of the amount of nitrogen that was used in agriculture to produce the consumed amount of protein. This yields a nitrogen consumption efficiency for the different regions in the world (Figure 2.7). In western societies about 60% of the harvested crop, or whole animal is converted

15

Figure 2.5 Global pig (top) and cattle (below) density in 2005 (FAO, 2010b).

16

Jan Willem Erisman

Figure 2.6 Nitrogen use efficiency for grain production relative to N fertilizer use for 1961–2002 for selected regions in the world (Hatfield and Prueger, 2004). Data source is http://faostat.fao.org/faostat.

to food products. Without correcting for over consumption of proteins as compared to the recommended intake of protein (60 g per day capital or 3.5 kg N per year capita; recommended by the World Health Organisation (WHO, 2007), the nitrogen efficiency of consumption varies from 10% in the USA to 28% in China (van Grinsven et al., 2003). Present Chinese consumption is far less N-efficient compared to the data in Figure 2.7, because animal protein consumption has increased substantially since 1995.

2.3.4╇ Biomass and food production and future fertilizer consumption in Europe Biomass is the oldest resource of energy used by mankind and has been the main source of energy until a century ago (Smil, 2004). Because of the inherently low efficiency of the photosynthetic process, no form of energy supply has such low power densities, and hence such high land demands (and fertilizer), as does the production of phytomass (Smil, 2004). In principle, there is globally enough annual growth of new biomass to cover up to four times the human annual energy use (Dornburg et al., 2007). However, in order to grow, collect and use biomass in a sustainable way to satisfy the human energy requirements, a well regulated and optimized process is needed. The European Fertilizer Manufacturer Association (EFMA) reports that in Europe the most ambitious Action Plans for biofuels production are still those of France and Germany. However, the UK, Sweden, Italy and Greece now also have ambitious objectives for biofuels production, closely followed by Austria and Denmark, the original ‘pioneer’ countries in this domain (EFMA, 2010). There are sufficient domestic resources to meet the EU targets set for the year 2010 but if more stringent goals are set for bioenergy in the future, it will be challenging to find sufficient resources in Europe and biomass imports from outside the EU (Fagernäs et al., 2006; Londo and Deurwaarder, 2007). There is a major challenge to reach the targets in a sustainable way

and there is much discussion on the availability of different biomass sources for bioenergy application, especially in relation to the additional use of fertilizer and the effect on greenhouse gas emissions (N2O). These developments are already affecting the food area as we see that energy crops are grown on former cropland and grassland, even without using the energy crop premium. According to EFMA (2010), in the coming decade, the production of biofuels will contribute to the 4.7% increase of nitrogen consumption in the EU-27 between 2009 and 2019. These prospects do not take into account possible new generations of bioenergy which might present an additional potential. For the longer term, recent scenario’s predict a much higher increase of fertilizer application as the result of increased food demand, biofuels production and limited land availability (Erisman et€al., 2009).

2.4╇ Energy, transport and industry The major link of energy, transport and industry to the nitrogen issue is the direct emission into the atmosphere of nitrogen oxides (NOx) from combustion of fossil fuels. The gases disperse, react and are eventually lost through deposition to the earth surface as gas or aerosol. Nitrogen oxides contribute to a variety of adverse effects, such as the formation of tropospheric ozone, the deposition of acidifying and eutrophying substances and the formation of secondary aerosols (mainly ammonium nitrates). While aerosols having an impact in Europe are largely formed from European emissions, background ozone levels in particular are significantly affected by NOx emissions throughout the Northern Hemisphere (with substantial contributions from emissions in Asia and North€America). There are other links with energy, transport and industry to the nitrogen issue, which are much less well quantified. Energy (coal or natural gas) is needed to produce nitrogen fertilizers and thus related to energy. Furthermore, through the use of fossil fuels the labour by man and draught animals has been replaced by machines and agriculture could be expanded and intensified leading to higher production in total and per ha. Furthermore, increased transportation of fertilizer, feed, food, fuel and other products has led to a redistribution of Nr over the world, while emitting NOx on the way.

2.4.1╇ NOx formation processes

NOx is mainly formed by two processes:€thermal NOx, when nitrogen and oxygen in the combustion air combine with one another at the high temperatures in a flame, and fuel NOx by the reaction of nitrogen bound in the fuel with oxygen in the combustion of air. A third and generally less important source of NOx formation is prompt NOx that forms from the rapid reaction of atmospheric nitrogen with hydrocarbon radicals (Dean and Bozzelli, 1999). Large combustion plants in power generation contribute to NOx emissions from high stacks,

17

The European nitrogen problem in a global perspective

4.5

Figure 2.7 Protein consumption per capita as direct nitrogen intake in kg/yr in 1995 for different regions of the world and the nitrogen efficiency of protein consumption relative to N inputs (%) (van Grinsven et al., 2003).

30 Total plant

Total animal

Nitrogen efficiency (%) 25

4 3.5

20

3 2.5

15

2 10

1.5 1

Nitrogen efficiency (%)

Protein consumption per capita in kg/yr

5

5

0.5 0

0 World

US

EU-15 Netherlands Benin

China

Denmark Germany

Region

while road transport sources are mainly line sources. Urban traffic, residential and commercial combustion as well as offroad sources can be classified as area sources. NOx emissions in Europe have fallen markedly in the last decades, mainly due to stringent emission controls applied to large combustion plants (EC Large Combustion Plants Directive, EEA, 2005a) and the EURO standards for road transport vehicles. At the same time, overall emission control due to effect-based regulations (EC National Emissions Ceiling (NEC) Directive and Gothenburg Protocol) have led to reductions in other sectors which have contributed to this decline (see Hertel et al., 2011, Chapter 9 this volume). Total non-transport NOx emissions in Europe are currently about 2000 kton/yr (EEA, 2005a). Emissions of NOx from public electricity and heat production in the EU fell by 45% over the period 1990 to 2004. If the structure of power production had remained unchanged from 1990 then by 2004 emissions of NOx would have increased by 33% above their 1990 levels, in line with the additional amount of electricity and heat produced. This decoupling of NOx emissions and electricity and heat production over the period 1990 to 2004 has been due to the following (EEA, 2005a). • The introduction of low-NOx combustion technology and flue gas treatment, which led to a 49% reduction. • Efficiency improvements, which resulted in a 14% reduction. • The switch in the fuel mix, away from coal and fuel oil towards natural gas, which led to an 8% reduction. • The lower share of nuclear and non-thermal renewable energy (i.e. excluding biomass) in 2004 compared to 1990, which actually increased emissions by 3%. The overall effect was a 45% reduction in NOx emissions in 2004 compared to 1990 levels. The total transport emission of NOx from Europe is currently about 8000 kton/yr. The specific emissions of air

18

pollutants from passenger and freight transport decreased for most modes of transport, more so for passenger transport than for freight transport (EEA, 2007). The highest reduction of specific emissions can be found in the road sector, following the increasingly stricter emission standards. Rail only slightly improved its performance over the past decade. Inland waterway freight transport stabilized its emissions per tonne-kilometre, while maritime passenger and freight transport increased their specific emissions over the past decade.

2.4.2╇ Additional NOx from bioenergy use

Biofuels and bioenergy are forms of energy (heat, power, transport fuels or chemicals) based on different forms of biomass. Recently, the EU adopted new targets for sustainable energy and greenhouse gas (GHG) emission reductions:€20% GHG reductions and 20% contribution of sustainable energy sources, including a target of 10% share of biofuels in the transportation sector in 2020 (EU, 2009). It is clear that biomass as transport fuel (biofuels), electricity and heat production (bioenergy) and Substitute Natural Gas (SNG or Green gas) will be a major component necessary to reach the targets. By 2050, it is estimated that biomass and waste utilization could rise from 9.0 to 13.5 EJ/a (215–320 Mtoe) (EU Biomass Action Plan, 2007). Increased biomass production potentially requires more fertilizer inputs, which will accelerate the nitrogen cycle (see Section 2.7). Additional fertilizer use will also cause additional Nr losses. Furthermore, bioenergy emits NOx into the atmosphere when combusted without de-NOx installations such as SCR. Additional emissions of NOx might be expected because the fuel-N is higher compared to that in fossil fuels and/or no de-NOx installations will be used for small scale applications and because more energy (combustion) is needed to produce one unit of electricity or transport. The

Jan Willem Erisman

Figure 2.8 Global emissions of nitrogen oxides (NOx) (EDGAR, 2010).

direct nitrogen emissions from different options to produce heat and power were compared by Pehnt (2006). Power generating systems excluding biomass are considerably better than the ‘reference mix’ which is based on fossil fuels, but biomass systems are well above the reference mix. An exception to this is systems with co-combustion of forest wood. This is due, in particular, to the fact that the NOx emissions of small combustion plants tend to be higher due to the lower temperature and efficiency. A special case is the biogas system. The nitrogen emissions of this system are more than the reference mix owing to the ammonia emissions resulting from the animal manure of the agricultural system prior to combustion (Pehnt, 2006).

2.4.3╇ European NOx emissions in a global perspective Figure 2.8 shows global NOx emissions for the year 2000 (EDGAR, 2010). Europe contributed about 14% of global NOx emissions in the year 2000, which was lower than that by North America (17%) and Asia (12% for S/SE Asia and 14% for East Asia, respectively). By far the largest sectoral contribution (Figure 2.9) to European NOx emissions stems from mobile combustion sources, contributing an estimated 62% (road transport 30.7%, other mobile sources 16.2% and international shipping 15%), followed by stationary combustion (15.7% from

large combustion plants and 4.5% residential and commercial combustion). Current inventories, such as the EDGAR inventory used here, often do not include data on natural and biogenic sources of emissions (or if so only partially), as can be seen in Figure 2.9 showing a very low contribution from biogenic and natural sources. Figure 2.10 indicates the estimated atmospheric transport distance of NOx emissions of Europe and North America (Sanderson et al., 2008) showing that Europe is substantially impacting parts of Asia and North America, and vice versa Europe is mostly influenced by emissions from North America. A few percent of the NOx emissions from North America are reaching Europe. The Taskforce on Hemispheric Transport of Air Pollution states that on average 75% of the NOx emissions in Europe is deposited within Europe, with small fractions falling on North America (1%) (Sanderson et al., 2008); South Asia (2%); East Asia (2.5%), and the remainder deposited in the oceans, and Russia.

2.5╇ Global and European nitrogen budget Globally, it is estimated that about 57% of anthropogenic nitrogen fixation results from the manufacture of nitrogencontaining fertilizers, 29% from cultivation of nitrogen�fixing crops, and 14% from burning fossil fuels (see Table 2.3, Erisman et€al., 2005). Fixation occurs in marine systems

19

The European nitrogen problem in a global perspective

Figure 2.9 NOx contribution by different regions of the world (a) and different sources (b), based on EDGAR data for the year 2000 (EDGAR, 2010).

as well, but those rates are highly uncertain. Van Egmond et€al. (2002) presented the estimated input and output flows for Europe. Updated values for Europe are presented by Leip et€al., 2011 (Chapter 16 this volume). The export is lower than the import of Nr. The remaining part is stock increase in vegetation, soils and water, but the largest part is denitrified to the atmosphere. The atmospheric emissions consist of oxidized and reduced forms of N. The total NOx emission in EU27 is currently 11€Mton NOx and its distribution is shown in Simpson et al., 2011 (Chapter 14 this volume). The total emissions of reduced

20

nitrogen to the atmosphere amount to 4 Mton NH3 (EU27) (Hertel et al., 2011, Chapter 9 this volume). The total emission of oxidized and reduced nitrogen in Europe is not much different, but the spatial and temporal variation is different and the chemical behaviour in the atmosphere is different. Therefore they have a different footprint, with oxidized nitrogen being transported over much larger distances. As a result, NOx emissions are much more of a global problem (also linked to O3 background) than the NH3 emissions. Total deposition of oxidized nitrogen in Europe (EU27) in 2006 was 1.7 Mton N and for reduced nitrogen 2.3 Mton, the total nitrogen deposition being 4 Mton N per year and the distribution is given in Hertel et al., 2011 (Chapter 9, this volume). The highest nitrogen deposition occurs in central Europe. Deposition of oxidized N is significant in the UK, the Netherlands, Germany and the Po valley (Italy). In addition, these areas present the highest deposition of reduced N, as a result of the intensive livestock production. In the regions of Europe with intensive agriculture spatial variability of nitrogen emissions and the deposition climate is high. More than half of the fertilizer that is produced in Europe is exported as fertilizers or agricultural products, mainly to the USA and Asian countries. Most of the nitrogen emitted into the air is deposited again on the land surface even though about one third is exported outside Europe, mainly as NOx and particles. Riverine transport to outside Europe is somewhat higher than atmospheric transport. The difference in the nitrogen balance in different regions can be derived from Figure 2.11, where the nitrogen budgets for the continents are given (Galloway et al., 2004). There are large differences, with BNF being the largest source in Latin America, Oceania and Africa and in all other regions it is fertilizer input. Also the major outputs differ:€ in Africa, Asia and Latin America most output is through riverine transport, in Oceania most of the output occurs via the atmosphere and in Europe and Northern America most of the Nr is exported through fertilizer and products. Global and European Nr production in 2000 and the different fluxes are presented in Table 2.3. Based on the numbers in Table 2.3 it can be calculated that the nitrogen efficiency, defined as the product output divided by the total inputs, in agriculture in Europe in the year 2000 was 36% compared to a global average of 50%. The nitrogen consumption by humans in Europe was 5.5 kg N per person compared to 4.3 kg N per person globally. The gross input necessary for consumption in Europe was 75 kg N per person compared to 45 kg N per person globally. The agricultural system and consumption in Europe therefore uses and wastes much more nitrogen than the global system. Apart from the creation of new Nr, there is a fair amount that is transported over the globe and within Europe, concentrating Nr in certain regions where its use is not always efficient. In 2005, ~45 Tg N of the ~190 Tg N of Nr created was traded internationally (Figure 2.12). Over the preceding decade, global trade of N-commodities increased twofold faster than the rate of Nr fixation (Galloway et al., 2008).

Jan Willem Erisman Figure 2.10 The transport distance of NOx emissions of Europe and North America over the globe. Shown is the percentage change in deposition of NOy in each NOx emission perturbation experiment relative to the control run, using multi-model annual mean deposition fluxes (Sanderson et al., 2008).

Nr input and output (Tg)

120 Input 100

Table 2.3╇ Global and European current inputs of N to the biosphere and per person. In brackets:€the percentage of the total budget (Erisman et al., 2005)

Global

Europe

Global

Europe

Tg N (%)

Tg N (%)

Tg N per person

Tg N per person

90 (24)

14.8 (28)

15.9

55.0

0.1 (0)

0.9

0.4

80 60

Biological N fixation

40

Lightning

20 0 Africa

–20 Output –40

Asia

Europe

Latin North Oceania America America

Region of the world BNF Imports

Fertilizer C-BNF Fossil fuel Atmospheric Exports Riverine

Figure 2.11 Nitrogen input and output (Tg) for different regions of the world (data from Galloway et al., 2004).

Unlike aquatic or atmospheric transport, where Nr is diluted to varying degrees, commerce typically results in injection of Nr to ecosystems in more concentrated doses. Regions that consume N-containing products, such as meat and milk, are often far removed from regions that produce the commodity and thus do not have to bear the environmental cost of the production.

2.6╇ Consequences of the nitrogen cascade There are many benefits of nitrogen, especially through the Haber–Bosch production. These are discussed by Jensen et€al. 2011 (in Chapter 3, this volume). Here the focus is on the

5 (1)

Total

95 (25)

14.9 (28)

16.8

55.4

Haber–Bosch N fertilizer & industry

85 (23)

21.6 (41)

15.0

80.3

Biological N fixation in agriculture

33 (9)

3.9 (7)

5.8

14.5

Animal feed imports

—

7.6 (14)

21 (6)

6.1 (11)

Total

140 (37)

Natural N fixation in oceans

140 (37)

Combustion in industry and transportation

Total

375

28.3 3.7

22.7

39.2 (74)

24.5

145.7

—

24.7

53.2

66.1

201.1

adverse effects in Europe in the global context. Nitrogen in its various chemical forms plays a major role in a great number of environmental issues (see Box 2.2). It contributes to acidification and eutrophication of soil, groundwater and surface

21

The European nitrogen problem in a global perspective

waters, decreasing ecosystem vitality and biodiversity and causing groundwater pollution through nitrate and aluminium leaching. Nitrogen compounds play an important role in carbon sequestration, global change, and formation of ozone, oxidants and aerosols, potentially posing a threat to human health and affecting visibility. Each of the emissions takes part in the cycling of N causing a number of different effects with its consequent linkages. For example, reactive N emitted to the atmosphere from fossil fuel combustion, in sequence can cause tropospheric ozone levels to increase, visibility to decrease and atmospheric acidity to increase. Once deposited from the atmosphere, reactive N can acidify soils and waters, over-�fertilize forests, grassland and coastal ecosystems, and can then be re-emitted to the atmosphere as nitrous oxide contributing to global warming and stratospheric ozone depletion. The Box 2.2╇ Most important adverse effects of reactive nitrogen (modified from Cowling et al., 1998)

Direct effects on humans Respiratory disease in people caused by exposure to high concentrations of: •

ozone

•

other photochemical oxidants

•

fine particulate aerosol

•

(on rare occasions) direct toxicity of NO2

Nitrate contamination of drinking water Increase allergenic pollen production, and several parasitic and infectious human diseases Blooms of toxic algae and decreased swimability of water bodies

Direct effects on ecosystems Ozone damage to crops, forests, and natural ecosystems Acidification effects on forests, soils, ground waters, and aquatic ecosystems Eutrophication of freshwater lakes and coastal ecosystems inducing hypoxia Nitrogen saturation of forest soils Biodiversity impacts on terrestrial and aquatic ecosystems Inducing damage by plagues and diseases

Effects on other societal values Odour problems associated with animal agriculture Acidification effects on monuments and engineering materials Regional hazes that decrease visibility at scenic vistas and airports Accumulation of hazes in arctic regions of the globe Depletion of stratospheric ozone by NO2 from high-altitude aircraft Global climate change induced by emissions of N2O Regional climate change induced by aerosol cooling Enhanced deterioration of archaeological artefacts

22

environmental changes will continue as long as Nr remains in circulation, for reactive N once created, and then lost to the environment, can be transported to any part of the Earth system, no matter where it was introduced. This sequence of effects has been termed the nitrogen cascade. In principle every pollutant can cause a cascade of effects, however nitrogen stands out because it can occur in many very mobile compounds that can cause a wide range of effects.

2.7╇ Effects of nitrogen on the European environment While some environmental problems are strictly local, like soil and groundwater pollution or exposure to high concentrations, N-related problems include the regional to global scales. The emissions of N2O readily spread across the atmosphere and have a global contribution. NOx has a continental character and can be transported over long distances between continents; NH3 is also continental but less than NOx and has smaller intercontinental exchange. The scale of N problems in estuaries and coastal seas depends on the extent of the river basin feeding them. The scales are important for the abatement strategy.

2.7.1╇ Nitrogen leaching in soil and groundwater Water quality is a major concern throughout Europe and other regions of the globe. Nitrate pollution of groundwater poses a recognized risk for its use as drinking water, while eutrophication of surface water due to excessive nutrient loads can lead to algal growth, oxygen deficiencies, and fish kills. Agriculture puts the largest pressure on groundwater and also on surface water pollution (EEA, 2005a). During the 1990s the nitrate concentrations slightly decreased in some European rivers, while they have remained constant in groundwater and high in some regions. Although some improvements have been carried out to reduce the nutrient input from wastewater discharge, diffuse pollution of agricultural origin remains a major threat for waters in the EU (EEA, 2005a). In the period 2000–2003, in EU15 nearly 40% of the groundwater monitoring stations (average values) exceed 25 mg NO3/l, and almost 50% of the surface water monitoring stations presents values greater than 10 mg NO3/l (EC, 2007a). These values are based on the information reported by EU Member States, and they are affected by the inhomogeneous distribution of sampling stations. The European Community Nitrates Directive (Council Directive 91/676/EEC) aims to control N losses and requires Member States of the European Union to identify areas contributing to N pollution of groundwater and surface water (EC, 2007a). In these areas agriculture may also be restricted. For example, the application of fertilizers should balance the needs of the crops, and the application of manure should not exceed 170 kg N/ha. Nitrate concentrations in drinking water should not exceed 50€mg/l (EC Drinking Water Directive, EC, 2007a).

Jan Willem Erisman Figure 2.12 Amounts of N contained in internationally traded products:€(A) fertilizer (31 Tg N), (B) grain (12 Tg N), and (C) meat (0.8 Tg N). Data are for 2004 and are in units of thousand of tons. Minimum requirements for drawing a line are 50€000 tons N, 20 000 tons N and 10 000 tons N for fertilizer, grain and meat respectively (UNEP, 2007; Galloway et al., 2008).

Exceedances of the nitrate standards are a common problem across Europe, particularly from shallow wells. It is often a problem in rural water supplies. For example, in Belgium 29% of 5000 wells examined had concentrations in excess of the limit value (OECD, 1997) and in Bulgaria it was estimated that, in the early 1990s, up to 80% of the population was exposed to nitrate concentrations that exceeded the limit value (OECD, 1995). In about a third of the groundwater bodies for which information was available nitrate concentrations exceeded the recommended limit. In general, there has been no substantial improvement in the nitrate situation in European groundwater and hence nitrate pollution remains a significant problem (EEA, 2003). The same is true for other parts of the globe, where nitrogen leaching to groundwater and subsequent riverine and watershed increase in nitrates are recognized as an increasing issue (UNEP, 2007).

2.7.2╇ Wastewater discharge to surface water In surface waters, the overall trend is that N concentrations have remained relatively stable throughout the 1990s and are highest in those Western European countries where agriculture is most intensive. Also in Europe’s seas the nitrate (nitrogen) concentrations have generally remained stable. A few stations in the Baltic, Black and North Seas, though, have demonstrated a slight decrease in nitrate concentrations (EEA, 2003). Although the most important, agriculture is not the only contributor of nitrogen in European streams. Other inputs of nitrogen come from the atmospheric deposition, household scattered dwellings, and from the direct discharges from sewerage, wastewater treatment plants and industries. The

23

The European nitrogen problem in a global perspective

nitrogen input from direct discharges from sewerage, wastewater treatment plants and industries constitute a threat for surface waters. According to what is reported by Member States, in the year 2000 in EU15, about 80% of wastewaters received adequate treatment before reaching the water bodies and the number of ‘big cities’ (agglomeration with waste water discharges greater than 150 000 population equivalents) without sufficient treatments has declined from 27 in 1999 to 17 in 2003 (EC, 2007b). However, the percentage of population connected to wastewaters treatment in Southern and Eastern Europe and in the accession countries is relatively low (EEA, 2005a) and information is often missing or not easily accessible (Mulligan et al., 2006). The load can be estimated based on the map of population density, emissions factor per population equivalent, and national statistics of population connected to sewerage system and level of wastewater treatment (Grizzetti and Bouraoui, 2006) and is given in Billen et€al., 2011 (Chapter 13 this volume) and Grizzetti et al., 2011 (Chapter 17 this volume). According to this estimate, the regions affected by higher nitrogen losses to surface waters include Belgium, the Netherlands, the Po valley (Italy), the Brittany region (France). Most of these areas are already totally or partially designated as Nitrates Vulnerable Zones to meet the EU Nitrate Directive. The EU makes progress in controlling point sources of pollution from industry and households through wastewater treatment. The Urban Waste Water Treatment Directive aims at 75% removal of the N load to the treatment plants in sensitive areas. However, by the end of 1998, still some 37 out of 527 cities had no treatment at all, including Brussels, Milan and Porto, while 57 others, including Aberdeen, Athens, Barcelona, Dublin, Florence, Liège and Marseille, were discharging a large part of their effluents untreated. The situation is generally improving and some of these cities made the necessary investments (EEA,€2005a). The GEMSTAT database of UNEP (www.gemstat.org) contains currently over 600 000 stations measuring nutrients in ground, surface and estuary waters. The data is used in different assessments to determine the watershed nutrient loads. In order to compare the European situation with the rest of the world the Nitrogen loading indicator is used (Figure 2.13). This indicator provides a measure of potential water pollution by explicitly mapping out the extent of both natural and anthropogenic nitrogen loading to the land and aquatic systems (Green et al., 2004). Global, continental, regional, and coastline-specific estimates of nitrogen loadings onto the continental land mass are derived by applying a mass balance assessment of nitrogen loads to the landscape providing an accounting of nitrogen sources, uptake, transport and leakages to terrestrial and riverine systems. In Europe the water pollution from nitrogen is mainly the result of fertilizer and lifestock production with the latter being dominant (Figure 2.13). Only in India and the southern parts of Latin America lifestock production is the dominant contributor. In Northern America fertilizers is dominant and in the rest of the world fixation dominates.

24

2.7.3╇ Eutrophication and acidification of terrestrial ecosystems The deposition of Nr is far above levels that the ecosystems are able to absorb and handle without adverse consequences for its vitality. Many ecosystems have changed from N limited systems to N saturated systems where N is not limiting any longer. Heathlands, e.g. in the Netherlands and Denmark, have turned into grasslands and forests once dominated by blueberries and lingon-berries have now a large occurrence of grasses (Bobbink et€al., 2010). Long-term high N deposition loads to ecosystems will also lead to N leaching into groundwater and surface water runoffs. A substantial fraction, not uncommonly in the order of 30% of the deposition may in this way be leached and transported to marine areas and contribute significantly to the marine eutrophication (EEA, 2005a). Direct deposition of nitrogen to sea surfaces is also of significant importance for the overall N input to marine ecosystems (see Voß et al., 2011, Chapter 8 this volume). About one third of the overall N input to the Baltic Sea, which is suffering from severe algae blooms every summer, is caused by N deposition (Billen et al., 2011, Chapter€13 this volume). The issues of acidification and eutrophication have been effectively, but not sufficiently, tackled by policy measures in the EU since the 1980s (EEA, 2005a). Several international agreements under the Convention on Long-range Transboundary Air Pollution (LRTAP) have been reached to reduce emissions. With respect to air pollution control, the EU has adopted emission and fuel quality standards for its Member States. In addition, many European countries have adopted national standards and other types of regulation reflecting the seriousness of pollution and national environmental quality priorities (Oenema et al., 2011, Chapter 4 this volume). An impact indicator that has been extensively used in Europe to assess the policy responses is the proportion of ecosystems where ‘critical loads’ of acidity and eutrophication are exceeded. The critical loads and critical levels refer to thresholds, which can serve as a tool to assess the occurrence of effects in natural ecosystems due to acid deposition. A critical load is a quantitative estimate of an exposure to one or more pollutants below which significant harmful effects do not occur according to current knowledge. Critical loads are ecosystem specific and show a large variation over Europe. Current European policies are anticipated to substantially improve the environmental conditions for Europe’s nature. The area affected by acidification is expected to decrease from 25% in 1990 to less than 5% in 2010. The eutrophication indicator shows the percentage of unprotected ecosystems improving from 55% to 41% (Hettelingh et€al., 2008). This underlines that eutrophication is a far larger problem in Europe than acidification and needs further abatement/attention. Critical loads are developed for other areas in the world, such as Asia and the USA.

2.7.4╇ Eutrophication of marine ecosystems Pollution of coastal seas occurs by the influx of nitrates and DON (dissolved organic N) through€– often transboundary€– rivers and by atmospheric deposition. Spatially explicit,

Jan Willem Erisman

Figure 2.13 Nitrogen loading onto the land mass and aquatic systems as a source for delivery to the coastal zone; a measure of potential water pollution. Total and inorganic nitrogen loads as deposition, fixation, fertilizer, livestock loads, human loads and total distributed nitrogen to the land and aquatic system. Map prepared by Water Systems Analysis Group, University of New Hampshire.

quantitative assessments of N inputs to coastal waters and marine ecosystems are not developed in most large-scale assessment reports. However, there are published studies of N inputs for individual estuaries in some regions as well as spatially explicit regional and global river N export models that provide considerable information. One of the first global syntheses of measurements of river nitrogen export, by nitrogen form, was by Meybeck (1982). Since then, several databases have been created documenting measured nitrogen export from rivers for specific regions and globally (Peierls et al., 1991; Meybeck and Ragu, 1995; Smith et al., 2003; LOICZ; UNEP/GPA, 2006). The creation of these databases has highlighted the large variation among rivers, both in terms of nitrogen flux density (kg N/km watershed/yr) and nitrogen load (kg N/watershed/yr), and made it possible to develop a more refined understanding of patterns of nitrogen export at local, regional and global scales. There is considerable spatial variation at local, regional and global scales in the magnitude of nitrogen loading (amount per watershed) as well as nitrogen yield (amount per unit area of watershed) from watersheds to coastal systems (Figure 2.14), with many hotspots around the world. It is clear from these maps that Europe forms a hot spot in the world with about the highest increases in nitrogen transport to the river mouth. These hot spots are the result of the growing nitrogen surplus, especially in agriculture. The source contribution

varies very much among the different river deltas. Also the environmental influence on transboundary outputs is variable. The amount of nutrients entering the oceans tend to vary significantly over time and from region to region (see Figure€2.15; UNEP/GPA, 2006), as do the actions to control the problem. Nutrient enrichment between 1960 and 1980 in the developed regions of Europe, North America, Asia and Oceania resulted in major changes in coastal ecosystems. Estuaries and bays are most affected, but eutrophication is also apparent over large areas of semi-enclosed seas, including the Baltic, North Adriatic and Black Seas in Europe, the Gulf of Mexico and the Seto Inland Sea in Japan (UNEP/ GPA, 2006).

2.7.5╇ Global warming:€N2O emissions and other effects of nitrogen Although the absolute quantities are small, the increasing N2O production plays an important role in the global warming issue since N2O is a powerful greenhouse gas. Europe’s emission is estimated at 0.8 Mton N2O-N€– 65% of which is due to ecosystem denitrification (EDGAR, 2010). The greenhouse gas targets for Europe defined in the Kyoto Protocol is a reduction of 8% compared to 1990 (EDGAR, 2010). These targets have to be met during the period 2008–2012. Europe contributes 10.8% of the global N2O emissions. Nitrous oxide has emerged as such a major GHG issue from agriculture and there has been some

25

The European nitrogen problem in a global perspective

Figure 2.14 Increase in nitrogen transport to river mouth between 1980 and 2000 (Reid et al., 2005).

debate about the validity of the emission factors used within IPCC (see e.g. Crutzen et al., 2008). More work is needed to provide consistent factors and use them for abatement strategies (Davidson, 2009). Apart from N2O there are indications that other chemical forms of nitrogen are emitted that could have a major impact on the global warming potential (GWP). Nitrogen trifluoride is about 17â•›000 times more potent than carbon dioxide. Its estimated worldwide release into the atmosphere this year is equivalent to the total global-warming emissions from Austria (Wen-Tien Tsai, 2008). Other impacts of nitrogen on the GHG emissions and the net GWP include the effect on carbon sequestration in waters, soils and plants; the effect on aerosol formation causing a direct and indirect cooling effect (through clouds) on the radiation balance and the effect on the emissions of other GHG, such as methane. De Vries et al. (2008) for example estimated that the effect of nitrogen deposition on the net GHG emissions for European forests yielded a net reduction in GWP through the additional sequestration of CO2. Recent debate has focused on the response of forests to this effect. The reported amounts of carbon stored per kg N added show a large range from 40 to 400 kg C per kg N deposition (Högberg, 2007; Magnani et al., 2007; De Vries et al., 2008; Reay et al., 2008). Meanwhile, further efforts are being directed to understand the overall effect of Nr on greenhouse gas balance, including the interactions with nitrous oxide, methane, ozone and aerosols (see ButterbachBahl et al., 2011, Chapter 19 this volume). The nitrogen cycle links with several other cycles, the most important being phosphorus and carbon, acidity and sulphur. For some issues, the complex role of the nitrogen cycle is well appreciated and discussed in Sutton et al., 2011 (Chapter 5 this

26

volume). Climate change is one example where these multifaceted interactions are understood, as the roles of N2O and tropospheric O3 (enhanced due to increased NOx emissions) are well understood as a contributing factor in greenhouse gas emissions. However, for other issues, there is a poor understanding of the role of the nitrogen cycle, including its place in the process of carbon sequestration and the interactions among the nitrogen, carbon and phosphorus cycles (Gruber and Galloway, 2008).

2.7.6╇ Effects of nitrogen on human health Excess nitrogen inputs to land, air and water can influence human health and welfare in both direct and indirect ways. Some such connections are well known. For example, exposure to high levels of NOx in urban areas or along roads cause human health problems, N-driven increases in tropospheric O3 pose direct health threats to humans (Levy et al., 2005) and cause substantial losses in agricultural productivity (Reilly et€al., 2007); the combination of these effects likely has a multibillion dollar cost. Nr in the air also contributes to the formation of fine particulates, which are in turn a substantial health threat in polluted regions such as urban areas (Wolfe and Patz, 2002). Excess nitrate in drinking water may also pose risks for some types of cancer and reproductive problems, though epidemiological data on these links remains too sparse to draw firm conclusions and there is considerable debate and a lack of consensus on the interpretation of medical evidence (van Grinsven et al., 2006; Ward et al., 2005). Nitrate intake through drinking water is only part of the total dietary intake; with the main dietary intake of nitrate for many people being from vegetables and meats.

Jan Willem Erisman

Figure 2.15 Changes in nitrogen concentrations for significant global watersheds (percentage) and by region (concentration):€1979–1990 and 1991–2005 (UNEP GEMS, 2006).

Jakszyn and González (2006) concluded that:€‘The Â�available evidence supports a positive association between nitrite and nitrosamine intake and gastric cancer, between meat and processed meat intake and gastric and oesophagal cancer and between preserved fish, vegetable and smoked food intake and gastric cancer, but is not conclusive’. Van Grinsven et al. (2006) concluded that there are both experimental and epidemiologic studies that indicate possible chronic health effects associated with consumption of elevated levels of drinking water nitrate, although there is no consistency across all studies. Therefore, the uncertainties associated with risk estimates are considerable, and hamper the design of cost-effective specific preventive measures for sensitive subpopulations or regions. Moreover, the enhanced risk of nitroso compounds (NOC)-induced toxicity as a result of high drinking water nitrate in combination with other individual risk factors, such as inflammatory diseases, emphasizes the importance of changing the limit values only when such risks have been carefully evaluated. At this moment this is not the case. Likewise, uncertainties do not allow an estimate of the health losses related to methemoglobinemia due to

drinking water nitrate. Evidence is emerging for possible benefits of nitrate/nitrite as a potential pharmacological tool for cardiovascular health (Wink and Paolocci, 2008). Although it is not yet possible to estimate net health loss due to nitrate, it is possible to make estimates of potential exposure. Based on data reported to the European Commission about the implementation of the Drinking Water Directive and data on the present nitrate levels in groundwater at drinking water extraction depths, the population in ten west European countries potentially exposed to drinking water exceeding the 50 mg/l nitrate standard, or the 3 mg/l nitrite standard, was estimated at over 9 million (2.7%). Other feedbacks remain poorly known but are potentially important and costly, including the possible effects of excess nutrients on human infectious and parasitic diseases (Townsend et al., 2003). Diseases that show signs of change following N (and/or P) caused eutrophication include mal� aria, West Nile virus, cholera and schistosomiasis (Townsend and McKenzie, 2007). These effects are more relevant for other parts in the world. In Europe and in parts of Asia and

27

The European nitrogen problem in a global perspective

the USA the exposure of humans to NOx and PM and the intake of NO3 is the main threat. Nonetheless, the facts that tropical regions will experience marked increases in nutrient loading and also contain the greatest diversity of human parasitic and infectious diseases highlights the need to understand these connections (Townsend and McKenzie, 2007). Finally, it is important to note that a healthy immune system requires adequate nutrition, thus one of the most critical links between fixed nitrogen and many tropical diseases may be via its greater supply in fertilizer to undernourished regions (Sanchez and Swaminathan, 2005).

2.7.7╇ Conclusions The nitrogen cascade effect is expected to be relevant in Europe. Through long-range atmospheric transport, river transport or groundwater transport the effects extend from regional to continental (acidification, eutrophication, carbon sequestration, aerosols) and even global dimensions (N2O). The cascade depends on the nitrogen status of a region:€this is defined as the amount of excess nitrogen in the system (or region) causing effects at different levels in the cascade of N causing a number of different effects. If the nitrogen excess increases, the number of effects in the cascade likely will increase (the cascade length increases). At the same time the area that is affected by nitrogen pollution increases (higher contribution to long-range transport or N2O emissions). While the linkages in the cascade effect still require to be quantified at the different scales, the available information already highlights its importance. Only at the beginning of the cascade the form of Nr is of importance. In the next stages of the cascade, it will be transformed either in the oxidized or reduced form and the origin is of little importance, whether it comes through the atmosphere or directly from manure or through mineralization or nitrification in the soil. Within the global context, Europe can be regarded as an excess nitrogen area, in contrast to developing regions such as Africa where nitrogen is limited in food production. Europe was one of the first regions where nitrogen became an environmental issue, with hotspots in the Netherlands, Denmark, France and Italy. Other areas in the world currently experience similar issues, such as parts of the USA, China, India and Latin America. It is expected that the nitrogen situation will become worse. Knowledge on the European nitrogen fixation rates, the transport through environment and cascading effects as described in this European Nitrogen Assessment might serve as input for these other regions. Of enormous significance is that excess nitrogen is linked to many of the major global and regional challenges that policymakers face today, such as globalization, strong development of growing economies, increase in human population, political stress, environmental aspects, etc. A prerequisite to reducing these problems is the development of a sound scientific base to help identify policy options. Furthermore, these issues need be recognized at scientific and political levels. The focus on food production in developed and developing countries should take environmental impacts of

28

nitrogen into consideration. For the future it is envisaged that the focus on the production of biofuels and the increased use of fertilizer will yield similar issues. The basis for a successful approach was laid down in the Nanjing Declaration on nitrogen management (Erisman, 2004). A comprehensive overview of N-related policies in Europe is given in (Oenema et al., 2011, Chapter 4 this volume).

Acknowledgements This chapter was prepared with the support of the NinE Programme of the European Science Foundation, the NitroEurope IP (funded by the European Commission) and the COST Action 729. Anne-Christine LeGall is acknowledged for drawing and providing Figure 2.2. The authors gratefully acknowledge the contribution of the EDGAR team at the EC Joint Research Centre in Ispra and the Integrated Project of Climate Change and Impact Research:€ The Mediterranean Environment (Project No. 036961€ – CIRCE) for providing detailed global emission maps of Nitrogen Oxides.

References Billen, G., Silvestre, M., Grizzetti, B. et al. (2011). Nitrogen flows from European watersheds to coastal marine waters. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Bobbink, R., Hicks, K., Galloway, J. et al. (2010). Global assessment of nitrogen deposition effects on terrestrial plant diversity:€a synthesis, Ecological Applications, 20, 30–59. Buijsman, E. (1986). Historical Trend in the Ammonia Emission in Europe (1870–1980), Report R-86-9, Institute for Meteorology and Oceanography, University of Utrecht, The Netherlands. Butterbach-Bahl, K., Nemitz, E., Zaehle, S. et al. (2011). Nitrogen as a threat to the European greenhouse balance. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Campling P., Terres J. M., Walle S. V., Orshoven J. V. and Crouzet,€P. (2005). Estimation of nitrogen balances from agriculture for EU-15:€spatialisation of estimates to river basins using the CORINE Land Cover. Physics and Chemistry of the Earth, 30, 25–34. Cowling, E., Erisman, J. W., Smeulders, S. M., Holman, S. C. and Nicholson, B. M. (1998). Optimizing air quality management in Europe and North America:€justification for integrated management of both oxidised and reduced forms of nitrogen. Environmental Pollution, 102, 599–608. Crutzen, P., Mosier, A. R., Smith, K. A. and Winiwater, W. (2008). N2O release from agro-biofuel production negates global warming reduction by replacing fossil fuels. Atmospheric Chemistry Physics, 8, 389–395. Davidson, E. (2009). The contribution of manure and fertilizer nitrogen to atmospheric nitrous oxide since 1860. Nature Geoscience, 2, 659–662. De Vries, W., Solberg, S., Dobbertin, M. et al. (2008). Ecologically implausible carbon response? Nature, 451, E1–E3. De Vries, W., Kros, J., Reinds, G. J. et al. (2009). INTEGRATOR:€a modelling tool for European-wide assessments of nitrogen and greenhouse gas fluxes in response to changes in land cover, land management and climate. Calculation procedures, application

Jan Willem Erisman methodology and examples of scenario results, Alterra Report (in press), Alterra Wageningen UR:€Wageningen, The Netherlands. De Vries, W., Leip, A., Reinds, G. J. et al. (2011). Geographic variation in terrestrial nitrogen budgets across Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Dean, A. M. and Bozzelli, J. W. (1999). Combustion chemistry of nitrogen. In:€Gas-Phase Combustion Chemistry, ed. W. C. Gardiner, Jr., Springer, New York, pp. 125–341. Dornburg, V., Faaij, A., Verweij, P. et al. (2007). Biomass Assessment:€Global Biomass Potentials and Their Links to Food, Water, Biodiversity, Energy Demand and Economy. Climate Change Scientific Assessment and Policy Analysis, PBL, The Netherlands. EDGAR (2010). Integrated Project of Climate Change and Impact Research:€The Mediterranean Environment, Project No. 036961– CIRCE. http://www.mnp.nl/edgar/model/v32ft2000edgar/ EFMA (2010). Forecast of Food, Farming and Fertilizer Use in the European Union 2009–2019. EFMA, Brussels, Belgium (www. fertilizerseurope.com). Erisman, J. W. (2004). The Nanjing declaration on management of reactive nitrogen. BioScience, 54, 286–287. Erisman, J. W., Domburg, N., de Vries, W. et al. (2005). The Dutch N-cascade in the European perspective. Science in China, Series C, Life Sciences, 48, 827–842. Erisman, J. W., Galloway, J. N., Sutton, M. S., Klimont, Z. and Winiwater, W. (2008). How a century of ammonia synthesis changed the world. Nature Geoscience, 1, 636–639. Erisman, J. W., van Grinsven, H., Leip, A., Mosier, A. and Bleeker, A. (2009). Nitrogen and biofuels; an overview of the current state of knowledge. Nutrient Cycling in Agroecosystems, 86, 211–223. European Commission (2007a). COM(2007) 120 final. Report from the Commission to the Council and the European Parliament on implementation of the Council Directive 91/676/EEC concerning the protection of waters against pollution caused by nitrates from agricultural sources for the period 2000–2003. 11 pp. European Commission (2007b). SEC(2007) 363. Annex to the Communication from the Commission to the European Parliament and the Council ‘Towards Sustainable Water Management in the European Union’ First stage in the implementation of the Water Framework Directive 2000/ 60/EC. 4th Commission Report (Executive Summary) on the Implementation of the Urban Waste Water Treatment Directive. European Environment Agency (EEA) (2003). Europe’s Environment:€The Third Assessment. Environmental Assessment report no. 10. Copenhagen. European Environment Agency (EEA) (2005a). The European Environment:€State and Outlook 2005. Copenhagen. European Environment Agency (EEA) (2005b). Agriculture and Environment in EU-15:€the IRENA Indicator Report. EEA Report no. 6/2005. Copenhagen. European Environment Agency (EEA) (2007). NEC Directive Status report 2006. Technical report no.15/2007. Copenhagen. EU (2007). Biomass Action Plan. Brussels. EU (2009). Directive 2009/28/EC of the European Parliament and of the Council of 23 April 2009 on the promotion of the use of energy from renewable sources and amending and subsequently repealing Directives 2001/77/EC and 2003/30/EC.

Fagernäs, L., Johansson, A., Wilén, C. et al. (2006). Bioenergy in Europe: Opportunities and Barriers. Espoo VTT Tiedotteita, Finland. FAO (2010a). (http://www.fao.org/docrep/w5146e/w5146e00.htm). FAO (2010b). (http://www.fao.org/agriculture/lead/tools/densities/en/). FAO (2010c). (http://faostat.fao.org/faostat). Galloway, J. N., Aber, J. D., Erisman, J. W. et al. (2003). The nitrogen cascade. BioScience, 53, 341–356. Galloway, J. N., Dentener, F. J., Capone, D. G. et al. (2004). Nitrogen cycles:€past, present and future. Biogeochemistry, 70, 153–226. Galloway, J. N., Townsend, A. R., Erisman, J. W. et al. (2008). Transformation of the nitrogen cycle:€Recent trends, questions, and potential solutions. Science, 320, 889–892. GEMS Water Triennial Reports (2006). United Nations Environment Programme GEMS/WATER (http://www.cciw.ca/gems/ intro.html). Green, P. A., Vörösmarty, C. J., Meybeck, M. et al. (2004). Preindustrial and contemporary fluxes of nitrogen trough rivers:€a global assessment based on topology. Biogeochemistry, 68, 71–105. Grizzetti, B. and Bouraoui, F. (2006). Assessment of Nitrogen and Phosphorus Environmental Pressure at European Scale. Report EUR 22526 EN. Grizzetti, B., Bouraoui, F., Billen, G. et al. (2011). Nitrogen as a threat to European water quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Gruber, N. and Galloway, J. N. (2008). An earth-system perspective of the global nitrogen cycle. Nature, 451, 293–296. Hatfield, J. L. and Prueger, J. H. (2004). Nitrogen over-use, under-use, and efficiency. In:€Proceedings of the 4th International Crop Science Congress, 26 Sept–1 Oct 2004, Brisbane, Australia, CD-ROM, www. cropscience.org.au. Hertel, O., Reis, S. and Ambelas Skjøth, C. (2011). Nitrogen processes in the atmosphere. In:€The European Nitrogen Assessment, ed. M.€A.€Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Hettelingh, J.-P., Posch, M. and Slootweg, J. (2008). Status of the critical load database and impact assessment. In:€Critical Load, Dynamic Modelling and Impact Assessment in Europe, ed. J.-P.€Hettelingh, M. Posch and J. Slootweg, CCE Status Report 2008, www.pbl.nl/cce. Högberg, P. (2007). Nitrogen impacts on forest carbon. Nature, 474, 781–782. Holloway, J. M. and Dahlgren, R. A. (2002). Nitrogen in rock:€occurrences and biogeochemical implications. Global Biogeochemical Cycles, 16, 1118. IPCC (2007). Fourth Assessment Report on Climate Change: Impacts, Adaptation and Vulnerability for Researchers, Students, Policymakers. Cambridge University Press. Jakszyn, P. and González, C. A. (2006). Nitrosamine and related food intake and gastric and oesophagal cancer risk:€a systematic review of the epidemiological evidence. World Journal of Gastroenterology, 12, 4296–4303. Jensen, L. S., Schjoerring, J. K., van der Hoek, K. et al. (2011). Benefits of nitrogen for food fibre and industrial production. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J.€W. Erisman et al. Cambridge University Press. Leip, A., Achermann, B., Billen, G. et al. (2011). Integrating nitrogen fluxes at the European scale. In:€The European Nitrogen Assessment,

29

The European nitrogen problem in a global perspective ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Levy, J. I., Chemerynski, S. M. and Sarnat, J. A. (2005). Ozone exposure and mortality:€an empiric bayes metaregression analysis. Epidemiology, 16, 458–468. Londo, M. and Deurwaarder, E. (2007). Developments in EU biofuels policy related to sustainability:€overview and outlook. Biofuels, Bioproducts and Biorefining, 1, 292–302. Magnani, F., Mencuccini, M., Borghetti, M. et al. (2007). The human footprint in the carbon cycle of temperate and boreal forests. Nature, 447, 848–850. Meybeck, M. (1982). Carbon, nitrogen, and phosphorus transport by world rivers. American Journal of Science, 282, 401–450. Meybeck, M. and Ragu, A. (1995). River Discharges to the Ocean:€an Assessment of Suspended Solids, Major Ions, and Nutrients. United Nations Environment Program, Nairobi. Moïssey Postan, M. and Rich, E. E. (1952). The Cambridge Economic History of Europe:€Trade and Industry in the Middle Ages, Cambridge University Press. Mulligan, D., Bouraoui, F., Grizzetti, B., Aloe, A. and Dusart, J. (2006). An Atlas of Pan-European Data for Investigating the Fate of Agrochemicals in Terrestrial Ecosystems. Report EUR 22334 EN. OECD (Organisation for Economic Co-operation and Development) (1995). OECD (Organisation for Economic Co-operation and Development) (1997). OECD (Organisation for Economic Co-operation and Development) (2001). Environmental Indicators for Agriculture Methods and Results, Volume 3. OECD Publishing, Paris. OECD (2006). Key Environmental Indicators. OECD Environment Directorate, Paris, France. http://www.oecd.org/ dataoecd/32/20/31558547.pdf Oenema, O., Oudendag, D. and Velthof, G. L. (2007). Nutrient losses from manure management in the European Union. Livestock Science, 112, 261–272. Oenema, O., Bleeker, A., Braathen, N. A. et al. (2011). Nitrogen in current European policies. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Olson, R. A. and Kurtz, L. T. (1982). Crop nitrogen requirements, utilization and fertilization. In:€Nitrogen in Agricultural Soils, ed. F.€J. Stevenson, pp. 567–599. American Society of Agronomy, Ann Arbor, MI. Pehnt, M. (2006). Dynamic life cycle assessment (LCA) of renewable energy technologies. Renewable Energy, 31, 55–71. Peierls, B. L., Caraco, N. F., Pace, M. L. and Cole, J. J. (1991). Human influence on river nitrogen. Nature, 350, 386–387. Reay, D. S., Dentener, F., Smith, P., Grace, J. and Feely, R. (2008). Global nitrogen deposition and carbon sinks. Nature Geoscience, 1, 430–437. Reid, W. V., Mooney, H. A., Cropper, A. et al. (2005). Millennium Ecosystem Assessment: Ecosystems and Human WellBeing:€Synthesis. Island Press, Washington, DC. Reilly, S., Paltsev, B., Felzer, X. et al. (2007). Global economic effects of changes in crops, pasture, and forests due to changing climate, carbon dioxide, and ozone. Energy Policy, 35, 5370–5383. Sanderson, M. G., Dentener, F. J., Fiore, A. M. et al. (2008). A multimodel study of the hemispheric transport and deposition of oxidised nitrogen. Geophysics Research Letters, 35, L17815. Sanchez, P. A. and Swaminathan, M. S. (2005). Public health. Cutting world hunger in half. Science, 307, 357–359.

30

Schlesinger, W. H. (2009). On the fate of anthropogenic nitrogen. Proceedings of the National Academy of Sciences of the USA, 104, 203–208. Simpson, D., Aas, W., Bartnicki, J. et al. (2011). Atmospheric transport and deposition of nitrogen in Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Skiba, U., Fowler, D. and Smith, K. A. (1994). Emissions of NO and N2O from soils. In:€Non-CO2 Greenhouse Gases, ed. J. van Ham, L.€J.€Janssen and R. J. Swart. Kluwwer, Dordrecht, pp. 153–158. Skiba, U., Fowler, D. and Smith, K. A. (1997). Nitric oxide emissions from agricultural soils in temperate and tropical climates:€sources, controls and mitigation options. Nutrient Cycling in Agroecosystems, 48, 139–153. Smil, V. (2001). Cycles of Life: Civilization and the Biosphere. Scientific American Library, New York. Smil, V. (2004). World history and energy. In:€Encyclopedia of Energy, ed. C. Clevaland, Volume 6. Elsevier, Amsterdam,€pp. 549–561. Smith, S. V., Swaney, D., Talaue-McManus, L. et al. (2003). Humans, hydrology, and the distribution of inorganic nutrient loading to the ocean. BioScience, 53, 235–245. Steinfeld, H., Gerber, P., Wassenaar, T. et al. (2006). Livestock’s Long Shadow:€Environmental Issues and Options, LEAD/FAO, Rome. Sutton, M. A., Erisman, J. W., Dentener, F. and Möller, D. (2008). Ammonia in the environment:€from ancient times to the present. Environmental Pollution, 156, 583–604. Sutton, M. A., Howard, C. M., Erisman, J. W. et al. (2011). The need to integrate nitrogen science and policies. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Townsend, A. R., Howarth, R. W., Bazzaz, F. A. et al. (2003). Human health effects of a changing global nitrogen cycle. Frontiers Ecology and Environment, 1, 240–246. Townsend, A. R. and McKenzie, V. J. (2007). Parasitic and infectious disease responses to changing global nutrient cycles. Ecohealth, 4,€384–396. UNEP/GPA (2006). The State of the Marine Environment:€Trends and processes. unep/gpa, The Hague. UNEP (2007). Global Environmental Outlook€– 4 . UNEP, Nairobi, Kenya. van Egmond, K., Bresser, T. and Bouwman, L. (2002). The European nitrogen case. Ambio, 31, 72–78. van Grinsven, J. J. M., van Schijndel, M. W., Schotten, C. G. J. and van Zeijts, H. (2003). Integrale analyse van stikstofstromen en stikstofbeleid in Nederland: Een nadere verkenning. Integrated analysis of nitrogen flows and nitrogen policy in the Netherlands further explored in Dutch€– 81 pp.€– 2003 Onderzoeksrapport€– RIVM rapport 500003001. van Grinsven, H. J., Ward, M. H., Benjamin, N. and de Kok, T. M. (2006). Does the evidence about health risks associated with nitrate ingestion warrant an increase of the nitrate standard for drinking water? Environmental Health:€A Global Access Science Source 2006,€5, 26. van Kessel, C. and Hartley, C. (2000). Agricultural management of grain legumes:€has it led to an increase in nitrogen fixation? Field Crops Research, 65, 165–181. Velthof, G., Oudendag, D., Witzke, H. P. et al. (2009). Integrated Assessment of Nitrogen Losses from Agriculture in EU-27 using MITERRA-EUROPE. Journal Environmetal Quality, 38, 402–417. Voss, M., Baker, A. and Bange, H. W. (2011). Nitrogen processes in coastal and marine systems. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press.

Jan Willem Erisman Ward, M. H., de Kok, T. M., Levallois, P. et al. (2005). Workgroup report:€drinking-water nitrate and health:€recent findings and research needs. Environmental Health Perspective, 113, 1607–1614. Wen-Tien Tsai (2008). Environmental and health risk analysis of nitrogen trifluoride (NF3), a toxic and potent greenhouse gas. Journal of Hazardous Materials, 159, 30, 257–263. WHO (2007). Protein and Amino Acid Requirements in Human Nutrition : Report of a Joint FAO/WHO/UNU Expert Consultation,

WHO Technical Report Series no. 935. World Health Organization, Geneva. Wink, D. A. and Paolocci, N. (2008). Mother was right: eat your vegetables and do not spit ! When oral nitrate helps with high blood pressure. Hypertension, 51, 1–3. Wolfe, A. H. and Patz, J. A. (2002). Reactive nitrogen and human health:€acute and long-term implications. Ambio, 31,€120–125.

31

Chapter

3

Benefits of nitrogen for food, fibre and industrial production Lead authors: Lars Stoumann Jensen and Jan K. Schjoerring Contributing authors: Klaas W. van der Hoek, Hanne Damgaard Poulsen, John F. Zevenbergen, Christian Pallière, Joachim Lammel, Frank Brentrup, Age W. Jongbloed, Jaap Willems and Hans van Grinsven

Executive summary Nature of the issue • Reactive nitrogen (Nr) has well-documented positive effects in agricultural and industrial production systems, human nutrition and food security. Limited Nr supply was a key constraint to European food and industrial production, which has been overcome by Nr from the Haber–Bosch process. • Given the huge diversity in Nr uses, it becomes a major challenge to summarize an overall inventory of Nr benefits. This full list of benefits needs to be quantified if society is to develop sound approaches to optimize Nr management, balancing the benefits against the environmental threats.

Approaches • When reviewing trends in European Nr production rates, including those from chemical and biological fixation processes, and the consumption of this Nr in human activities, agriculture is by far the largest sector driving Nr creation. • Particular attention has been given to relationships between N application rates, productivity and quality of products from major crops and livestock types, including consideration of the mechanisms underlying variations in N response/outputs and the derived impacts on land use and land requirements.

Key findings/state of knowledge • The economic value of N benefits to the European economy is very substantial. Almost half of the global food can be produced because of Nr from the Haber–Bosch, and cereal yields in Europe without fertilizer would only amount to half to two-thirds of those with fertilizer application at economically optimal rates. • There is a wide variety in N responses at field level. For cereals, nitrogen productivity, also termed the agronomic efficiency, averages 41 kg grain per kg applied fertilizer N across the EU countries, with significant variation between the member states. Variation reflects differences in crop type, farm type, cropping practices, area, region, soil fertility and climate. • Farmers have an economic incentive to apply only the economically optimal rate of fertilizer N, but there is no strong incentive to increase N use efficiency as the economic return on using fertilizer N is very robust, especially in high value crops. However, recent initiatives to reduce environmental impacts of Nr losses have led to an increase of N use efficiency in both crop and livestock production. • Increasing fertilizer prices and climate change will create new incentives to increase N use efficiency. There are ample options to achieve this via N-conserving field practices such as catch crops, reduced soil tillage, better estimation of crop N requirements and improved timing and placement of N inputs. Also modifications to livestock diets, enhanced recycling of livestock wastes, prevention of ammonia loss from animal housing and field manure application can enhance benefits per unit applied Nr. Plant materials with improved composition of major storage compounds and novel feed additives, e.g. proteins from bio-fuel production, can also improve feed N responses per unit mass Nr used.

Recommendations • Legislative drivers to reduce Nr use, including mineral fertilizer, must take account of the nitrogen benefits in agricultural production needed to maintain food and energy security, given the limited options to increase arable land area. • New technological tools should be implemented to improve nitrogen-efficiency and the overall benefits of Nr use.

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

32

Lars Stoumann Jensen and Jan K. Schjoerring

3.1╇ Introduction Nitrogen is an essential component of many compounds found in living cells within plants, animals and humans. All nitrogen in animals and humans originates in one way or another from plants or microbes because only they have the ability to convert mineral forms of reactive nitrogen (Nr), such as nitrate and ammonium, into organic nitrogenous compounds such as amino acids and nucleotides, which are the building blocks of proteins and nucleic acids essential for life. The availability of these basic mineral forms of Nr is a key factor determining the productivity of crops for food, feed, fibre and bio-energy and hence for all human activities (Sutton et al., 2011, Chapter€1, this volume). The main paths for production of these mineral Nr forms are fertilizer manufacture, especially through the Haber– Bosch process, and biological nitrogen fixation in€crops.

3.1.1╇ What are the benefits of reactive nitrogen? The provision of reactive nitrogen through mineral fertilizers has contributed greatly to the increased production of agricultural products needed to feed the increasing global population (Erisman et al., 2008) and hence to food security. In 1900, world agriculture was able to sustain around 1.6 billion people on 850 million ha of agricultural land using mainly extensive cultivation practices without mineral fertilizers. The same combination of agronomic practices extended to today’s 1.5 billion ha cropland would feed around 3 billion people, i.e. no more than around 50% of the present population at the generally inadequate per capita level of year 1900 diets. Today, synthetic fertilizer N has been estimated to be the basis for the production of almost 50% of the food consumed by mankind (Smil, 2000; Erisman et al., 2008). In this sense, the use of reactive nitrogen provides huge benefits for man, but in order to maximize these benefits, the (a)

efficiency of use of nitrogen inputs should be optimized. It should also be noted that in this chapter, benefits are defined relatively broadly, including economic as well as social, health and political (stability) values. For the major cereal food crop in Europe, wheat, it can be estimated that the agronomic benefit obtained by application of N fertilizer amounts to a yield increase from 86 to 150 million tons of grain per year (assumptions:€27 300 million ha of wheat in the EU-27, average yield of 5.5 t grain/ha, at 112 kg fertilizer N/ha; EFMA, 2009), based on an average yield without mineral N fertilizer in ecological farming at 60%–70% of yield with mineral N fertilizer (Offermann and Nieberg, 2000). Fertilizer nitrogen has also played a beneficial role in avoiding natural terrestrial ecosystems from being converted to cropping systems (Tilman et al., 2002). At the global scale, land use changes due to replacement of forest or natural grasslands with agricultural cropland contribute significantly (6%–17%) to greenhouse gas emissions because large amounts of carbon dioxide fixed or stored in soil organic matter are released upon cultivation. In comparison, the greenhouse gas emissions from production and use of mineral fertilizers are relatively small, constituting 0.8% and 1.3%, respectively (Figure 3.1a). The estimated contribution of European agriculture to total greenhouse gas emissions is only around 10% and land use changes in Europe have been estimated to act as a net sink for greenhouse gases (Figure 3.1b). However, this is solely driven by afforestation, with cropped land being a small net source of CO2, although at a declining rate (Kitou et al., 2009). If less intensified agriculture becomes predominant in Europe, implying significantly lower or completely abandoning nitrogen fertilization, it may result in land use changes either within Europe or elsewhere in the world to compensate for the decrease in crop yields. Thus, von Witzke and Noleppa (2010) demonstrated that increasing production of

GLOBAL = 49 bn t CO2eq (agriculture contributes 26.3%)

(b)

EU-27 = 5 bn t CO2eq (agriculture contributes 10%)

Production of mineral N fertilizer (0.8%)*

Energy, Waste, Industry, etc. (74%)

1.1%* 1.1%*

N2O from mineral N fertilizer use (1.3%)*

3,2% N2O from organic N sources (3.8%) Other agricultural GHGs, mainly CH4 (8.4%)

90%

4.6%

In Europe, LULUCF is a sink of 432 mio t, but almost solely by afforestation

Land use change for agriculture (LULUCF) (12% avg., range 6% - 1 7%) Figure 3.1 Estimated greenhouse gas emissions from production and use of mineral N fertilizers (in CO2-equivalents) together with other agricultural activities and land use change for agriculture at (a) the global and (b) EU-27 scale (all 27 member states of the European Union as of 2007). Order of contributions in (b) the same as in (a). From Brentrup and Palliere (2008), based on IPCC (2007), Bellarby et al. (2008), UNFCCC (2008) and * author calculations.

33

Benefits of nitrogen for food, fibre and industrial production

agricultural commodities in the EU would significantly reduce the current EU net food imports which have increased over the past decade, and hence also the associated import of ‘virtual land use’ around the world. From the same point of view, agricultural intensification may be viewed as a greenhouse gas mitigation mechanism (Burney et al., 2010) and a measure for preserving natural habitats (Balmford et al., 2005). However, it has also been argued that extensification of European agriculture would have little bearing on the proportion of native land areas being converted into cropland. Rudel et al. (2009) analysed trends in crop yields and cultivated land areas for ten global regions and found that agricultural intensification was not generally accompanied by a decline or stasis in cropland area at a national scale during the period 1990–2005. They argued that many other factors influence conversion of native land to cropland, including trade and market prices, economic development and national policies and regulations. However, there is little doubt that avoiding new cultivation of major areas of native land is crucial with respect to reducing the anticipated increase in atmospheric carbon dioxide levels originating from land use change (Tilman et al., 2002; Cassman et€al., 2003). Economically and environmentally sound nitrogen fertilization practices on the generally fertile and productive soils in the majority of European countries can contribute to this (Brentrup and Pallière, 2008) even if other factors will also play a significant role. The pressure on native land may also be accentuated with the increased focus on replacing fossil energy with that generated on the basis of biomass. The land area required to meet the EU target for bio-ethanol in vehicle fuels by 2020 (10% blending by volume, total consumption 101 million t gasoline) would be 9.5 million ha if optimally fertilized wheat was the source, but 16.7 million ha (out of a total arable area of 98 million ha) if not fertilized with N, supposing 60%–70% yield reduction as cited above. Extensification of agricultural production in Europe in parallel with an increased European demand for bioenergy may thus increase the pressure on land resources elsewhere in the€world. Fertilizer nitrogen inputs also affect the level of soil organic matter (SOM), albeit only in a long-term perspective and often relatively moderately (Raun et al., 1998). Soil

organic matter is one of the most important factors for soil fertility. This is the case because soil organic matter directly affects nutrient availability via mineralization of organically bound N, P and S, via adsorption of cations and via complexation of trace elements. In addition, soil organic matter indirectly affects soil water dynamics, stability of soil aggregates, resilience against erosion and other deterioration processes. In the Broadbalk continuous winter wheat long-term experiment, which was started in 1843 on a silty clay loam in the UK, soil C in the plot annually applied 144 kg N/ha together with P and K presently amounts to 1.12% C, corresponding to about 25% more SOM than in the unfertilized control soil only containing 0.85% C (Johnston et al., 2009). The fertilization with mineral N has naturally resulted in higher biomass production, higher yields and greater organic matter returns in stubble and roots than on the unfertilized plot. Application of animal manure will also enhance soil organic matter and often to a greater extent than fertilizer N alone (Johnston et al., 2009). Appropriate nitrogen inputs contribute significantly to maximizing the utilization of other costly inputs or resources for soil and crop management such as other nutrients, pesticides, labour, energy and capital as well as crop genetic potential (cultivars). An example of the interactions between genetic potential, N, P and K application can be seen in Table € 3.1. Nitrogen application clearly interacted with P application, resulting in larger N use efficiency when P was also applied. The use of improved barley varieties has increased the yield, but only when N, P and K were supplied together as evidenced by the 2003–2006 data (Johnston and Poulton, 2009). Finally, benefits of nitrogen also emerge via the very€important use of nitrogen products in the manufacture of explosives, nylon and acrylic fibres‚ methacrylate and other plastics‚ foamed insulation and plastics‚ electronics, metal plating‚ gold mining‚ animal feed supplements‚ herbicides‚ and many pharmaceuticals (Maxwell, 2004). Other uses of reactive nitrogen compounds involve ammonia for the abatement of atmospheric NOx and SO2 emissions as well as a refrigerant for cooling, especially in connection with food storage. Ammonium-phosphates and -sulphates are components of metallurgy for welding and fire fighting. Despite these important applications, the

Table 3.1╇ Improvements in crop nitrogen use efficiency of spring barley cultivars with a gradually higher yield potential, grown in the long-term Hoosfield Barley experiment at Rothamsted, UK (Johnston and Poulton, 2009)

Spring barley varieties (period grown) (kg grain / kg N applied)

Treatment

34

N applied (kg/ha)/yr

Chevalier (1852–1871)

Plumage Archer (1952–1961)

Optic (2003–2006)

N

48

43

34

18

NK

48

46

36

24

NP

48

62

53

60

NPK

48

61

52

83

Lars Stoumann Jensen and Jan K. Schjoerring

consumption of reactive nitrogen for industrial use only constitutes around one third of the total European budget, the dominating uses being crop and livestock production (see further details in€Section 3.5).

The use of nitrogen fertilizers in the EU-27 countries, i.e. all 27 member states of the European Union as of 2007, increased substantially from the 1950s to the early 1980s (Figure 3.2). From the mid 1980s, the consumption of N fertilizers started to decline. A major decline occurred in the early 1990s, mainly due to the collapse of the economy of the eastern European countries (later new EU member states), but also due to the McSharry reform of the European Union Common Agricultural Policy (CAP) in 1992 (EU Glossary, 2010). This introduced mandatory set aside, causing the farming community in the old EU member states to take a proportion of the farmland out of production. Since then, nitrogen consumption in EU-27 has stabilized around 11 Tg, where it is forecasted to stay with only a small increase (3%–4%) until 2019 (EFMA, 2009). This level of consumption corresponds to an annual average quantity of around 85–90 kg N per ha of arable land. When permanent crops (fruit and vineyards) and fertilized grasslands are included, the average annual nitrogen consumption per unit surface area of total agricultural land amounts to around 65€kg€N/ha.

making it the most popular N source in developing countries where it is by far the dominating fertilizer, constituting around two thirds of total fertilizer N consumption (Figure 3.3b). In Europe, the share of urea N is much lower, only 22% (including 6% from UAN). This is partly due to the fact that the availability of nitrogen for plant uptake can be delayed since urea must first be transformed into ammonium and subsequently to the final nitrate form. This delay may particularly be a problem in the cold spring weather typical of NW parts of Europe, whereas in areas bordering the Mediterranean Sea urea consumption is traditionally higher. Another disadvantage of urea is that it implies a large risk for N loss to the atmosphere by ammonia volatilization which may exceed 20% of the applied N (Sommer et al., 2004). As a consequence, urea is generally much more difficult to manage properly than ammonium and nitrate-based fertilizers. The principal straight nitrogen fertilizer in Europe is calcium ammonium nitrate (CAN) which is well suited for most European soils, crops and climatic conditions, since half of the nitrogen is in the nitrate form, being immediately available to plants, and the other half is in the ammonium form. Ammonium nitrate (AN) use in Europe has declined since 2004 due to security regulations associated with its potential use as explosive. Other straight nitrogen fertilizers include ammonium sulphate, calcium nitrate and anhydrous ammonia. The latter is a highly concentrated nitrogen fertilizer (82% N) mainly used in North America. It requires specific logistics because of safety precautions necessitating special transport, handling and application equipment for injection into the soil. In Europe, it represents less than 1% of total nitrogen fertilizer used (EFMA, 2003).

Major nitrogen fertilizer forms used in the EU

Nitrogen use for major crops in the EU

The current use of nitrogen in western Europe (European Union countries in 2004 + Norway and Switzerland) distributed across different types of nitrogen fertilizers is shown in Figure 3.3a. Approximately 80% of the nitrogen is applied in straight N fertilizers, while 20% is applied in multi-nutrient compound fertilizers together with phosphorus and/or potassium. Urea is the most concentrated solid nitrogen fertilizer, containing 46% N on a weight basis. For this reason, urea has advantages in terms of distribution, storage and handling costs,

Wheat and barley are the dominating cereal crops in the EU, covering nearly 40% of the area (Table 3.2). Fertilized grassland covers about 30% of the area. Oilseed rape is a large and emerging nitrogen user among intensive crops. Compared with other crops, oilseed rape, sugar beet and wheat are the crops with the highest N application rates (Table€3.2). However, there is a considerable variation between countries as evidenced by the wide minimum– maximum range in Table 3.2. Assessed in terms of total

3.2╇ Trends in European N use for crop production 3.2.1╇ Fertilizer N consumption and crops

Base year 2009

2007

Nutrients (million tonnes)

N + 4.1

+ 3.0

+ 7.7

– 4.4

+ 3.9

– 9.7

Figure 3.2 Total fertilizer nutrient consumption in the EU-27 countries in the period 1927–2009, and predicted trend in fertilizer nutrient use until 2019. Two Â�different predictions are given, based on either the reference years 2007–2009 or 2005–2007 (EFMA, 2009).

K2O

P2O5

35

Benefits of nitrogen for food, fibre and industrial production

(a)

(b)

fertilizer-N use, wheat is the most important crop in the EU, followed by grassland and barley. Although oilseed rape and sugar beet have the highest N rates, their crop area is smaller. Based on this information the economic benefit of N will be discussed later in more detail for wheat, oilseed rape and grassland. Emphasis will be given to regional differences within Europe.

Expected future trends in fertilizer Nitrogen consumption for major crops in the EU Forecasts of developments in European (for the EU-27 countries) fertilizer demand over the next 10 years are created annually by Fertilizers Europe (formerly European Fertilizer Manufacturers Association, EFMA). The forecasts are based on national prognoses, in a standardized, upward procedure, where fertilizer consumption is evaluated by assessing area and nutrient application rates for each crop in all member states, and held against economic developments, i.e. market prices for agricultural inputs and products. On average over the last three cropping years (2007–2009) in the EU-27, fertilizers containing 10.5 Gt of N have been applied to 135.3 million ha of farmland each year (46.1 million farmable hectares are not fertilized, which include 36.5 million hectares of unfertilized grasslands). By 2018/2019, the forecasters expect the fertilizer N consumption to reach 11.0 Gt of nitrogen (an increase in N consumption of +4.1%, Figure€ 3.2), applied to 133.6 million ha (EFMA, 2009). However, this expected increase is not evenly distributed geographically, most of the increase (+17%) is expected in the new EU member states (termed EU-12), with very marked increases in e.g. Bulgaria (+25%) and Romania (+55%). In the old EU member states (termed EU-15), the majority of countries predict a decline (−1% to −12%) in fertilizer N consumption until 2019, with only Spain and Sweden expecting increases above +10% (EFMA, 2009). For the major crops produced in Europe, the expectation is that N demand for wheat, coarse grains (other cereals) and oilseeds will increase, whereas fertilizer N input for sugar beets, potatoes, fodder crops and grasslands are expected to decrease until 2019 (Figure 3.4).

36

Figure 3.3 Current consumption of different sources of fertilizer nitrogen in (a) EU-25 plus Switzerland and Norway and (b) developing �countries (EFMA, 2003). AN = Ammonium nitrate, CAN = Calcium ammonium nitrate, UAN = Urea ammonium nitrate, N compounds = compound fertilizer, containing N together with P and/or K as well as other macro- and micronutrients.

3.2.2╇ Organic manure N inputs Nitrogen in manure is applied in the form of stored manure collected in animal housing and manure deposited by grazing animals in the field. The input of manure-N to crops in the European countries varies from 15 to 225 kg/ha of agricultural land per year (Figure 3.5). In comparison, the average use of mineral fertilizer amounts to between 15 and 140 kg/ha of agricultural land per year (Figure 3.5). Total national nitrogen inputs (incl. manure N) to agricultural land range from about 40 kg/ha of agricultural land in Romania to 365 kg/ha in the Netherlands. The amount of manure-N applied to specific crops in the EU is not well known. It varies between countries, depending on livestock systems, manure type, crops, rotations and their distribution. Manure from cattle, sheep and goats (ruminants) produced during grazing is deposited on grasslands, while manure from animal housing and storage is mainly applied to fodder or roughage crops (grass, silage maize). Pig and poultry manure is generally used on non-fodder (feed, food, fibre, fuel) crops, as these are generally grown on farms without ruminant livestock. Velthof et al. (2009) distinguish between three types of nonfodder crops, viz. those with high manure application rates (potato, sugar beet, barley), crops with moderate rates (wheat, rye, oat, grain maize) and crops where generally no manure is given (fruits, citrus, oil crops).

3.2.3╇ Biological N fixation Globally, leguminous crops, mainly soybean and peanut, cover about 10% of agricultural land (Smil, 1999). Galloway et al. (1995) and Smil (1999) estimated global N2 fixation for cultivated agricultural systems, i.e. excluding the extensive tropical savannas, at 43 Tg (range 32–53 Tg) and 33 Tg (range 25–41 Tg) annually. Herridge et al. (2008) calculated N2 fixation by the coarse grain legume–rhizobia symbioses at 21 Tg N annually and by the forage and fodder legume–rhizobia symbioses to range from 12 to 25 Tg, annually. In Europe, the main N fixing crop species is clover grown together with grass as a crop for feed purposes (grazing and

Lars Stoumann Jensen and Jan K. Schjoerring Table 3.2╇ Average annual, minimum, maximum and cumulative fertilizer N-use for European crops in EU-27 (EFMA 2005–2007).

Crop

Average kg/ha

Range (min–max) kg/ha

Crop area million ha

N use = crop area × avg. N-rate (million kg)

Oilseed rape

148

50–195

6.1

884

Sugar beet

123

50–160

1.9

228

Wheat

113

25–200

25.9

2902

Grain maize

106

26–200

9.0

958

Potato

98

40–185

2.2

218

Barley

88

15–145

13.9

2011

Grassland

69

10–170

30.5

2075

Silage maize

65

10–126

4.7

304

Rye, triticale, oats, rice

64

10–110

8.7

549

Figure 3.4 Forecast changes from 2009 to 2019 in fertilizer N, P and K use by crop sector (taking account of both projected changes in area and yields of the crops) in the EU-27 overall (EFMA, 2009).

roughage). Using various modelling approaches, de Vries et al. and Leip et al. (2011, Chapters 15 and 16, this volume) estimated that the biological N fixation in European agriculture is in the range of 0.8–1.4 Tg N annually, out of a total annual N input of 20.8–26.2 Tg. Hence biological N fixation only accounts for 3%–5% of the N inputs to agricultural land in Europe.

3.2.4╇ Efficiency of nitrogen inputs Mineral fertilizer Mineral fertilizer N can in principle be applied at the time and location that is optimal for crop uptake. This should lead to potentially high N use efficiencies. However, in practice many factors may reduce the actual N use efficiency. There are many ways to define and measure N use efficiency. Here, two different approaches are applied:

(1) the apparent recovery efficiency (NUEa), which is the increase in N uptake (or total biomass) divided by the amount of N applied [N-uptakeN-fertilizer-rate(X)€– N-uptake No-N-fertilizer] / N-fertilizer-rate(X); (2) the direct recovery efficiency (NUEd), which is the amount of labelled N that is taken up in a crop (usually only aboveground material) following application of 15N labelled fertilizer N-uptake/15N-fertilizer-rate(X).

15

The direct recovery efficiencies are generally smaller than the apparent recovery efficiencies because some of the applied N is incorporated into the microbial biomass N and subsequently becomes incorporated into the soil organic matter.

37

Benefits of nitrogen for food, fibre and industrial production 400

Nitrogen use in European agriculture 350 300 Grazing-N

Kg N/ha

250

Manure-N 200

Fertilizer-N

150 100 50 0 m Ro

an

ia B

g ul

ar

ia

t La

via P

tu or

l e a d n n a ia nia ary ga oni ai tri ec vak an de a g t us re Sp Pol we o ithu un l A G Es S H S L

Ita

y s d d ic om ia m ce ark an lan giu and an ubl en d an nm p rl ov erm Ire Bel ng Fr e l i e e S R K D G th h d Ne ec nite z C U

ly

Fi

nl

Figure 3.5 Average annual nitrogen inputs in fertilizer and manure (applied and deposited during grazing) to agricultural land in the EU (Luxembourg, Malta and Cyprus omitted). Fertilizer data for 2007–2008 based on EFMA (2009). Manure data are based on data for 2000 from Velthof et al. (2009).

Currently, worldwide recovery of N fertilizer (NUEa) in cereal crops is on average 30%–50%. Higher values, exceeding 60%, have been reported for winter wheat, e.g. in Denmark and in the UK (Sylvester-Bradley and Kindred, 2009) when crop management is optimal and N applications are balanced against expected yield and soil fertility status. In contrast, average N fertilizer recovery in cereals in China may be as low as 30%–35%, mainly due to unbalanced and excessive fertilization, leading to large N surplus and N losses (Vitousek et al., 2009). Typical nitrogen recovery efficiencies (NUEa) in research plots are about 40%–50% for cereals when defined based on grain N yield, increasing to 60%–70% when based on total aboveground N uptake (Chien et al., 2009). Direct recovery efficiencies (NUEd) of mineral fertilizer N applied in autumn in temperate humid climates have been measured at 11%–42% for winter wheat in Great Britain (Powlson et al., 1986a). For springtime applications, NUEd increases to 42%–78% illustrating the effect of improved timing of the application and synchrony with crop N uptake (Powlson et al., 1986b; Pilbeam, 1996). Experiments with 15N-labelled fertilizers applied to wheat have shown higher direct recovery efficiencies (NUEd) in humid than in dry environments. However, the retention of residual 15 N in the soil increases with increasing climate dryness (Figure 3.6). In a review of a large number of wheat experiments across various climates, ranging in annual rainfall/evaporation ratios from 0.14 (Israel) to 1.32 (Alabama, USA), Pilbeam (1996) found a significant positive correlation between the annual rainfall/evaporation ratio and N uptake efficiency (Figure 3.6). Hence, owing to unfavourable growth conditions in a drier climate, N uptake efficiencies tend to be lower than under humid

38

conditions, but the larger proportion of unused fertilizer N apparently remains in the soil after harvest. The loss of fertilizer N seemed independent of climate and averaged 20%, with the main loss pathways thought to be dominated by leaching loss in humid climates and gaseous loss in arid climates. However, under practical conditions, fertilizers prone to gaseous loss, e.g. urea, may lose up to 20%–30% by NH3 volatilization immediately following application in warm climates. Postharvest losses of residual fertilizer N are usually small (less than 5%), indicating that the soil nitrate pool which is susceptible to leaching during autumn and winter in humid environments mainly originates from mineralization of organic nitrogen. Thus nitrate leaching typically represents an indirect rather than a direct loss of applied fertilizer N, having first been converted to organic matter. Fertilizer nitrogen use efficiency in the EU-27 countries, i.e. the member states of EU as of 2007, varies between countries both due to soil and climatic differences, but also because dominant crop species and fertilization practices differ significantly from one country to another. If cereals are taken as a main common denominator, a wide range of yields are observed, both within the EU-15 (old member states) and the EU-12 (more recent members). In Figure 3.7, cereal yield is expressed as a function of the annual nitrogen application rate per ha of arable land in each of the EU countries. The quantity of grain produced for an additional quantity of nitrogen, commonly referred to as the agronomic efficiency (slope of regressions in Figure 3.7), appears somewhat higher for the EU-15 than for the EU-12 countries, but the difference is not significant. The average agronomic efficiency for all EU-27 countries is around 41 kg grain per kg N applied.

Lars Stoumann Jensen and Jan K. Schjoerring 70

70

(b) 60

50

50

% Fertilizer N recovered in soil

% Fertilizer N recovered in crop

(a) 60

40 30

40 30

20

20

10

10

0

0 0

0.25

0.5

0.75

1

1.25

0

1.5

0.25

Annual rainfall: evapotranspiration

0.5

0.75

1

1.25

1.5

Annual rainfall: evapotranspiration

Figure 3.6 Relationship between the precipitation/evapotranspiration ratio and the recovery of 15N-labelled fertilizer N in (a) the crop and (b) the soil at harvest for wheat grown in different locations (redrawn from Pilbeam, 1996).

9000 EU-15 ( ): Yield = 48*N + 333 R 2 = 0.54

8000

Cereal yield (kg / ha)

7000 6000

BE IE DK

AT

5000 4000

RO

PL EE

2000

GB

CZ

SI

IT

BG

3000

FR DE

HU SE

SK

NL

EU-12 ( ): Yield = 29*N + 1740 R 2 = 0.42

GR LT ES

FI

PT

LV

EU-27 (all): Yield = 41*N + 950 R 2 = 0.58

CY

1000 0 0

50

100

150

200

Nitrogen application (kg N / ha) Figure 3.7 Annual cereal crop yield vs. nitrogen application in the EU-27 countries. The slope of the regression lines, i.e. kg grain harvested per kg fertilizer N applied, is the apparent fertilizer N use efficiency, also termed the agronomic N efficiency. The slope does not differ significantly between EU-12 and EU-15 countries. Data from EFMA (2007).

In order to further improve the recovery efficiency of mineral N fertilizers focus should be on (i) improved synchrony between fertilizer N and crop demand, i.e. the timing, (ii) sitespecific fertilization to take into account spatial heterogeneity on field-level, i.e. the rate and place, and (iii) possibilities for taking into account year-to-year weather variations affecting crop growth and soil N mineralization and (iv) reduce risk of

fertilizer N loss, e.g. through rapid soil incorporation (if possible) or the use of inhibitors of urea hydrolysis (to minimize ammonia volatilization, most relevant for surface application) or nitrification (to reduce leaching of nitrate). The most obvious way to improve the synchrony between crop N demand and N supply is to split the N application into several single dressings. Doing so allows the total N supply to

39

Benefits of nitrogen for food, fibre and industrial production

be readjusted according to the actual growing conditions during the year of cultivation. Analysis of soil and plants can further help the farmer to exactly target the N application to the crop requirements. Slow-release fertilizers have been developed which contain N in forms that delay the initial availability or extend it over time, ideally to match the uptake by the crop. These ferti�lizers typically consist of urea-aldehyde polymers (ureaformaldehyde, isobutyliden-diurea or crotonolidendiurea) which are compounds with a very low solubility in water. This is in contrast to the so-called controlled-release fertilizers, which are produced through modification of urea, enabling them to release nitrogen over a given period (up to 12 months) through a coated surface or through an encapsulating membrane. It must be noted that the costs of controlled-release or stabilized fertilizers are significantly higher than those of conventional fertilizers. Thus, their main uses have so far been restricted to high value crops, specific cultivation systems and non-agricultural higher-value sectors (horticulture, nurseries, greenhouses, etc.). Urease inhibitors are used to reduce ammonia volatilization from urea (Chien et al., 2009). More than 14 000 mixtures of compounds with a wide range of characteristics have been tested and many patented as urease inhibitors. Lately, focus has been on the urease inhibitor N-(n-butyl) thiophosphoric triamide (NBPT) and other inhibitors of the phosphoro� amide family, including 4-methyl-2-nitrophenyl phosphoric triamide and 2-nitrophenyl phosphoric triamide. These compounds have been widely tested and in many cases reported to significantly increase N recoveries from urea (Watson et al., 2008; Turner et al., 2010). However, urease inhibitors cannot completely control NH3 loss when urea is surface applied to soils because the inhibitory effect depends on soil physical and chemical characteristics and also on environmental conditions. The urease inhibitors available so far can prevent urea hydrolysis for at most 1 or 2 weeks, during which time the fertilizer should ideally be incorporated into the soil by water (rain or irrigation) or mechanical methods. In addition, the price of NBPT may exceed the payback and hence limit the economic incentive for using it. Stabilized fertilizers are associated with nitrification inhibitors such as ammonium thiosulphate, thiourea, dicyandiamide, nitrapyrin and 3,4-dimethylpyrazole phosphate. Nitrification inhibitors are chemical compounds that delay bacterial oxidation (nitrification) of ammonium nitrogen. The objective is to preserve applied ammonium nitrogen in its original form, which is stable in the soil, and to slow its conversion to nitrate. This temporarily lessens the proportion of nitrate in the soil thereby reducing potential leaching losses or formation of N2O (Irigoyen et al., 2006; Akiyama et al., 2010). Crop parameters influencing N uptake and the dry matter production per unit absorbed N are obviously also important for maximizing fertilizer N recoveries. In-season crop monitoring through ground-based reflectance sensors, leaf chlorophyll meters or aerial/satellite imaging can be used to target crop N requirements. This can be accompanied by selection of

40

N-efficient cultivars. Specific examples of traits which are of particular value for increasing nitrogen recovery in feed wheat cultivars are (Figure 3.8):€ (i) increased root length density at depth, (ii) a high capacity for N accumulation in the stem, potentially associated with a high maximum N-uptake rate, (iii) low leaf lamina N concentration, (iv) more efficient postanthesis remobilization of N from stems to grain, but less efficient remobilization of N from leaves to grain, both potentially associated with delayed senescence, and (v) reduced grain N concentration. In cultivars for bread-making, high nitrogen use efficiency may in addition be associated with specific grain protein composition.

Organic manures In organic manures, a significant proportion of the N is organically bound. This organic N pool mineralizes slowly into ammonium and subsequently nitrate in the soil and only a part becomes available for plant production in the year of application. In addition, the portion of Nr in organic manures that is not organically bound is present as ammonium, implying risk for a high loss through ammonia volatilization. Consequently, crop use efficiencies of manure N are normally lower than that of mineral fertilizer N. The lower the proportion of organic N in the manure (as in liquid manures, e.g. slurry or urine), the faster and greater crop uptake (utilization) of manure N can be expected, as long as large losses of ammonia are avoided. The organically bound N in manure contributes to the N pool in the soil and this becomes plant-available via mineralization in subsequent years (Schröder et al., 2005; Vellinga et al., 2010). Proper determination of the true N use efficiency therefore requires long-term trials. The plant-availability of N in manure within the year of application is often expressed as the ‘Nitrogen Fertilizer Value’ (NFV) or ‘Mineral Fertilizer Equivalent’ (MFE). These parameters are determined by reference to crop mineral fertilizer N response, with 100% representing equivalent crop utilization of manure and mineral fertilizer N within the first cropping season after application. Owing to the content of organic N in manure, MFE values normally range between 20% and 80%, depending on the type of manure (proportion of organic N), the crop, the application time (autumn application resulting in higher leaching losses than spring application) and application methods (surface application resulting in higher ammonia losses and lower MFE value than injection into the soil). With strict regulations on manure application methods in some EU countries (e.g. Denmark and the Netherlands), fertilizer efficiency of manure N has generally increased in recent years as evident from the MFE values in Table 3.3 and also reported by Birkmose (2009). Consequently, the use of N in mineral fertilizer has declined, in Denmark by as much as 50% since 1990. A monitoring programme shows that the Danish nitrate leaching in the same period was reduced by 41% (Grant et al., 2009), and the emission of ammonia by 42% (Gyldenkærne and Mikkelsen, 2007). However, in order to achieve such improvements, substantial investments have been

Lars Stoumann Jensen and Jan K. Schjoerring

Maximize photosynthetic capacity per unit N

Optimize grain protein and N remobilization

• • • •

• Optimize N remobilization

Leaf and stem N storage Vertical distribution of canopy N RuBisCo catalytic properties C4 metabolism

Maximize N capture

• • • •

Distribute roots deeper Increase specific root length Optimize root-to-shoot ratio N transporter systems

Figure 3.8 Parameters important for crop N benefits (modified after Foulkes et al., 2009 and Reynolds et al., 2009).

efficiency and stay green

• Optimize grain N% • Optimize grain protein composition Maximize nitrate assimilation

• Glutamine synthetase activity • Alanine aminotransferase activity • Organic acid metabolism

made in low emission stables, manure storage facilities and low emission spreading equipment. The long-term effects of management practices on soil quality will also have an important influence on the N use efficiency of the entire agro-ecosystem. Soil organic matter content is a key measure of soil quality and soils that sequester carbon also sequester N, resulting in greater indigenous N supply and a reduction in N fertilizer requirements. Therefore, management practices which increase soil organic matter will generally provide efficiency benefits over the long term.

3.3╇ Nitrogen effects on crop productivity and quality Nitrogen is one of the most important limiting factors for biomass productivity in terrestrial ecosystems. The stimulating effects of N on plant growth are due to a direct role of N as a building block in proteins, nucleic acids and pigments (chlorophyll). Roughly three quarters of all N in the leaf is contained within the chloroplasts (Dalling, 1985), predominantly as a constituent of the enzyme Rubisco which catalyses fixation of carbon dioxide. The N supply affects the biosynthesis of the phytohormone cytokinin, which functions as a growth promoter. It is via cytokinin that ample nitrogen supply stimulates growth and early establishment of the leaf area which is required for light (energy) interception, photosynthesis and biomass production (Marschner, 1995; Wang and Below, 1996).

3.3.1╇ Crop yield responses to nitrogen Nitrogen fertilizers have a decisive influence on the yield of arable crops. Since the days of von Liebig in the nineteenth century up to the present day, countless experiments have been carried out to determine the crop yield response to N. Response curves generally show an increase in crop production with increasing N supply up to a certain level, provided other production factors such as water and other nutrients are sufficiently available (see also Table 3.1). Crop N demand is the product of plant dry weight and the minimum N concentration

in the dry matter needed to obtain maximum growth. The fact that yield response to N application typically follows a convex curve reflects that yield responses per extra unit of N applied become smaller and smaller as the N quantity increases. This diminishing return is due to the fact that the efficiency of light interception within the canopy decreases as more and more leaves get shaded by the above leaves in the canopy. Various crops show a negative response to N at high levels because of effects like lodging (cereals), increased incidence of pests, decreased quality (sugar content in sugar beet, oil content in oilseed rape). Wheat field trials in north-western Europe (Belgium, UK, Ireland, Denmark, France, Germany, the Netherlands) show maximum grain yields in the range of 9–11â•›ton/ha. However, actual farm grain yields in this region are often 2–3 ton/ha lower. Yields in eastern and north-eastern Europe are about the same, at around 4 t/ha. Especially yields in southern Europe under rain-fed conditions are much lower (trials:€2–5 t/ha), whereas actual yields are about 2.5 t/ha. However, Â�published results of field trials for wheat, oilseed rape and grassland in southern and eastern Europe are rather rare (Shiel et€ al., 1999; Sidlauskas and Bernautas, 2003; Barlóg and Grzebisz, 2004; Lopez-Bellido et al., 2007; see also supplementary material to Chapter 22 of this volume, Brink et€al., 2011). Different crop production models (mathematical functions) have been developed and tested for calculation of the relationship between crop yield and nitrogen supply or fertilization rates. This of course implies uncertainties in terms of model choice, annual nitrogen response variations and parameter estimation (Henke et al., 2007). A number of these models has been analysed, showing advantages and disadvantages (e.g. for corn:€Cerrato and Blackmer, 1990; winter wheat:€Webb et€al., 1998; Makowski et al., 1999; Gandorfer, 2006). The most commonly used crop yield response functions used in northwest European agriculture are the linear with plateau (LP), the quadratic (Q) and the exponential (EXP) type. An example of these, fitted to the same data, can be seen in Figure 3.9a. The quadratic and exponential functions can be combined with linear functions to construct quadratic functions with plateau

41

Benefits of nitrogen for food, fibre and industrial production Table 3.3╇ Average Mineral Fertilizer Equivalent (MFE) value of N in manure for different countries in the EU as affected by animal manure type. Values represent estimates from field experiments in five different EU countries as used by advisory systems for fertilization planning (ten Berge and van Dijk, 2009)

Type of manure Cattle slurry

Nitrogen MFE value (% of manure total N) Crop, application time

NL

FL

DE

DK

FR

Arable land, spring, maize/ pot./beets

50–55

55

70

55–70

55

Arable land, spring, winter wheat

40

55

70

45–55

Grassland, before 1st cut

45–50

55

70

45–50

50–60

Excreted on pasture Pig slurry

25

Arable land, spring, maize/ pot./beets

70–75

65

60

70–75

60–75

Arable land, spring, winter wheat

55

65

60

65–70

60–70

Grassland, before 1st cut

45–55

65

60

60

50–65

Solid cattle manure

Arable land, spring, maize/ pot./beets

30

30

60

45

15–30

Solid chicken manure

Arable land, spring, maize/ pot./beets

50–55

55

50

65

45–65

Liquid fraction

Arable land, spring, maize/ pot./beets

85–90

80–90

90

after separation

Arable land, spring, winter wheat

70

80–90

85–90

Grassland, before 1st cut

65–75

80–90

75–80

Solid fraction after separation, cattle

Arable land, spring, maize/ pot./beets

25

25

55

Solid fraction after separation, pigs

Arable land, spring, maize/ pot./beets

50

35

55

Compost

Arable land, spring, maize/ pot./beets

10

10

10–15

Note:€NL = the Netherlands, FL = Flanders (Belgium), DE = Germany, DK = Denmark, FR = France.

(QP) or linear-exponential functions (LEXP), which have a distinct maximum yield level, whereas the EXP function has no maximum, and the quadratic function yield declines at high N input rates. Differences in economic optimal N application rate (EONR, see further details below) resulting from applying different crop response functions can be substantial (e.g. 60 kg/ha in Figure 3.9a). Although the degree of model fit to yield data should be an important criterion, it is not always clear why a certain model is given preference over other models (Cerrato and Blackmer, 1990). The yield at zero N rate (Figure 3.9) is caused by the crop response to N originating from atmospheric deposition, biological fixation (in case of legume–grass mixtures) and from soil mineral N (SMN), which is also a result of the history of the plots (N-input and uptake efficiency of preceding crops). In a number of countries, among others the UK, Germany,

42

Belgium, the Netherlands and Denmark, the estimation of SMN is an integral part of the fertilizer recommendation systems. Figure 3.9b illustrates the influence of long-term absence of fertilizer or manure N input on the soil N supply capacity. In the nil fertilizer N plot of the long-term (>150 years) trials, crop N uptake and removal in grain was only 26 kg N/ha, whereas in the short-term trials with normal fertilization in preceding years this was around three times higher, 74 kg N/ ha, reflecting a higher SMN and N mineralization capacity of the soil. In regions with very low fertilizer N-input, soil mineral N may often be the major source of N for crop yield and uptake. It is clear from Figure 3.9b, that if the N use efficiency is calculated as apparent N recovery (NUEa) at economic optimal fertilizer N rate (EONR) with reference to the unfertilized plot ([N removal at EONR€– N removal without fertilizer]/ EONR),

Lars Stoumann Jensen and Jan K. Schjoerring

(a)

Grain yield (t/ha)

8

Q: EONR=239

LP: EONR=179

6 LP Q EXP

4

EXP: EONR=206

152 kg N/ha

74 kg N/ha

Crop N removal at economic optimum N fertilizer rate (EONR)

6 Avg. of 154 1–y trials Long-term trial (>150y)

4 26 kg N/ha

2

2 0

8

167 kg N/ha

(b)

10

Grain yield (t/ha)

10

Crop N removal with no N fertilizer applied

0 0

40

80

120

160

200

240

280

N fertilizer rate (kg/ha)

0

50

100

150

200

250

300

350

N fertilizer rate (kg/ha)

Figure 3.9 (a) Examples of mathematical functions commonly used to express crop yield responses to increasing annual nitrogen application (LP:€linear with �plateau, Q:€quadratic, EXP:€exponential) and the resulting different economically optimal fertilizer N rates (EONRs) for winter wheat in Germany (after Gandorfer, 2006). (b) Winter wheat grain yield response to increasing fertilizer N application in short-term (1-year) trials (average of 154 individual experiments) and in a long-term trial with the same N application rates for many years (>150 years). Numbers (kg N/ha/yr) indicate N removal with grain at either no fertilizer N application or at EONR (indicated by arrows), respectively (calculations based on Broadbalk long-term trial in Rothamsted, UK, and Yara field trials, F. Brentrup personal communication).

the NUEa is somewhat lower (51%) from the short-term (1-year) trials than from the long-term trial (66%). However, this is because a small part of the mineral N applied to a crop in previous years will be immobilized as soil organic N and thus be available for subsequent years’ crop. Therefore, interpretation of NUE from short-term trials should be done with consideration of this long-term effect of fertilization level. Field trials show a large year-to-year variation in yield response, which leads to different economic optimal N application rates. As an example for winter wheat in Germany, economic optimal N application was found to vary by as much as 85 kg/ha (Henke et al., 2007). Similarly, in a review of a very large number of N response trials in winter wheat and spring barley from different combinations of season, site, and cultivar in the UK (Sylvester-Bradley and Kindred, 2009), economic optimal N rates at a N price:grain price ratio of 5 were found to vary by more than 200 kg/ha for winter wheat and up to 150€ kg/ha for spring barley (Figure 3.10). These data confirm that there is no direct relationship between crop yield and economic optimum N fertilizer rate. The presence of this wide range affects both the N-recommendation and the farmers’ decision to adjust the N-rate, as future weather conditions are always unknown. Farmers can react to this challenge, for example, by splitting their nitrogen application into several dressings. Other sources of variation in crop response are soil conditions and crop cultivar. For horticultural crops such as vegetables, yield response to N application also varies considerably, both between fields and years of the same crop, but in particular between species. As illustrated in Figure 3.11, yield typically increases with N application rate until a plateau, where no further yield increase is achieved, but also no yield decline in contrast to cereals. It is obvious that optimal N levels vary greatly between a crop like white cabbage, with a very large N uptake capacity, deep roots and a long growing season, and a crop like lettuce, with

a rapid growth, but shallow roots and a short growing season. In order to avoid excessive N application in vegetable cropping, and hence leaching or gaseous N losses to the environment, monitoring of soil mineral N supply is crucial in order to adjust the fertilizer N input accordingly. However, usually the cost of fertilizer is negligible compared to the often very high value of a vegetable crop, giving relatively little incentive for the farmer to limit fertilization to the economic optimum.

3.3.2╇ Assessment of economic optimal N-rate The costs of nitrogen in proportion to the total production cost of agricultural crops such as cereals ranges from 20% to 30% of the variable production cost (Zimmer, 2008), thus constituting a significant share of the total costs. For highvalue horticultural crops, however, the N fertilizer costs may amount to only a few percent; for instance for edible potato it ranges between 2% and 9% of the variable production costs, and between 1% and 4% of the total production costs. Generally speaking, a farmer will strive for maximization of profit. Farmers should try to meet the economically optimal N application rate, but in practice they tend to assure adequate input of nitrogen, as this increases the chance of economical return on investments for other production factors. This may lead to unnecessary application of nitrogen fertilizer, e.g. in the USA amounting to 20%–35% of total N at a cost up to 50€€/ha (Sheriff, 2005). Overuse of N-fertilizer is also promoted when farmers or their contracts with purchasers set yield targets before the growing season. In this case, the N application rate is adjusted to the target yield without being able to properly consider other production factors like availability of other nutrients, water, pest control, etc., which have to be adequate to achieve the target yield. As Figure 3.10 shows, there is no relationship between yields and economic optimum fertilizer rates.

43

Benefits of nitrogen for food, fibre and industrial production

(a) W. wheat

Grain yield (t/ha)

(b) Sp. barley

Available N (kg/ha)

Available N (kg/ha)

Figure 3.10 Fitted responses of grain yield to available N (soil N supply through mineralization plus fertilizer N applied per year) for (a) winter wheat (129 response curves) and (b) spring barley (47 response curves); from different combinations of season, site, and cultivar in the UK. Economic N optima (at fertilizer N:grain price ratio = 5) for each response curve are indicated by small triangles, mean of all economic optima with large triangle (Sylvester-Bradley and Kindred, 2009).

Relative yield (%)

Spinach

Available N (fertilizer+soil) 0-30 cm (kg/ha)

Lettuce (8 field trials)

Available N (fertilizer+soil) 0-30 cm (kg/ha)

Cauliflower (7 field trials)

Available N (fertilizer+soil) 0-60 cm (kg/ha)

White cabbage (6 field trials)

Available N (fertilizer+soil) 0-90 cm (kg/ha)

Figure 3.11 Examples of vegetable N responses for various vegetable crops in field trials from Germany (Feller et al., 2001).

There is no standard approach to determine the economic value of N at the farm level. The parameter to determine the appropriate N level is the ‘economic optimal N-rate’ (EONR). This is the N-rate where the marginal financial return of the harvested crop equals the marginal cost of N,

44

i.e. where the slope of the tangent of the yield response curve is equal to the reciprocal of the price ratio (see Figure 3.12). Nitrogen fertilizer application should not target the maximum crop yield, rather it has to target the economic optimum crop yield.

Lars Stoumann Jensen and Jan K. Schjoerring 10

Figure 3.12 Illustration of the concept of economically optimal N rate (EONR). The slope of the tangent to the yield response curve represents the marginal yield increase due to additional annual fertilizer N application. EONR is where the slope of this tangent is equal to the fertilizer N to grain price ratio.

Max yield (Ymax) EONR tangent

Grain yield (t/ha)

8

6

Chord: mean N-productivity

4

2

N rate for Ymax

EONR

0 0

50

100

150

200

250

300

350

400

N rate (kg/ha)

Fertilizer N:crop price ratio (euro/kg fertilizer per euro/kg product)

8

Figure 3.13 Trend of the price ratio of fertilizer nitrogen (calcium ammonium nitrate; LEI, 2009) over crop price for wheat and oilseed rape, using data for the UK, Denmark, Germany, Czech Republic and Spain (EUROSTAT, 2009) and for milk (data from the Netherlands; LEI, 2009).

7 Wheat 6 5 4 Oilseed rape 3 2 Milk 1 0 1994

1996

1998

2000

2002

2004

2006

The economic optimum N-rate (EONR) depends on the ratio of prices of mineral fertilizer N and of crops. The prices of fertilizers and crops may be somewhat correlated, as increased crop prices will increase the demand for fertilizer, and conversely, increasing energy prices will tend to increase crop prices. However, the general tendency has been an increase in this price ratio over the past couple of decades (Figure 3.13). This trend is largely a result of increasing energy prices, while crop prices have been stable or decreasing and means that the costs of N fertilizer are becoming an increasingly important control on rates of N application. The economic return on N (ERoN) is not a standard param� eter and is defined here as the ratio of the slope of the chord

2008

2010

of the response curve (the mean N-productivity, kg grain per kg N) connecting yield at N = 0 and the maximum yield (see Figure€3.12), and the fertilizer:crop price ratio. Jenkinson (2001) was one of the few who assessed the economic value of fertilizer N in this way. Using the results of the Broadbalk continuous wheat experiment at Rothamsted, he concluded that the investment of £66 on fertilizer generated an extra grain yield of £367, which corresponds to an ERoN value 5.6. Yield response curves depend on soil, climate, crop variety, management practices, and consequently also values of EONR and ERoN derived from such curves. In Figure 3.14, we have estimated the gross economic return on nitrogen fertilizer (ERoN) for winter wheat and oilseed rape

45

Benefits of nitrogen for food, fibre and industrial production 16

(a) Oilseed rape

7

Gross economic return on N

Germany (Sieling & Kage, 2008)

6

UK (Berry & Spink, 2009)

5

Lithuania (Sidlauskas & Bernautas, 2003)

4

(euro/euro)

(euro/euro)

Gross economic return on N

8

3 2 1 0

1

3

5

7

9

11

Fertilizer:crop price ratio

(b) Winter wheat

14 12

Germany (Henke et al., 2007)

10

Spain (Lopez-BellidoRJ et al., 2007)

UK (Dampney et al., 2006)

8 6 4 2 0

1

3

5

7

9

11

Fertilizer:crop price ratio

Figure 3.14 Gross economic return on nitrogen fertilizer (ERoN) for oilseed rape (a) and winter wheat (b) as a function of fertilizer:crop price ratio (€/kg N per €/ kg crop). Results based on N-response curves from field trials and at actual mean national N-rates (EFMA, 2009).

as a function of different price ratios between fertilizer and crop. The calculations are based on selected yield data for cultivation of winter wheat in Germany (Henke et al., 2007; Rathke et al., 2005; Sieling and Kage, 2008), Spain (Lopez-Bellido et al., 2007) and the UK (Dampney et al., 2006), and for oilseed rape in Germany, the UK (Berry and Spink, 2009) and Lithuania (Sidlauskas and Bernautas, 2003). For oilseed rape, the price ratio increased from 3 to 4 between 2000 and 2008, and for winter wheat from 5 to 7. As seen from Figure 3.14a, for oilseed rape, typical current values for ERoN are 2–5 €/€ in Germany and the UK and 1–2 €/€ in Lithuania, where there is a low yield potential. At an ERoN below 1, it is not cost-effective to apply N to the crop. For winter wheat, ERoN ranges from 3–7 €/€ in Germany and the UK to 1–2 €/€ in Spain (Figure 3.14b). Excluding non-production related issues, such as environmental considerations, the objective of farmers is to apply N at the economic optimum N rate. As also mentioned above, the maximum biological yield should not guide the decision about the fertilizer application rate, since associated ERoN values are substantially lower when targeting the maximum yield, and there is a risk that a farmer will lose money on the N-investment. Differences between actual N-rates and EONR, and between EONR and the N-rate for maximum yield amount to around 50 kg/ha, and at the present price level represent a value of about 40 €/ha. This is a marked proportion (8%–10%) of the total direct production costs for oilseed rape (400–600 €/ha) and wheat (300–500 €/ha) according to Zimmer (2008). Since it is often difficult for the farmer to estimate and actually match EONR for his individual fields, farmers tend to add N beyond EONR in order to secure high yields, which on the other hand can be harmful for the environment (see also Brink et al., 2011, Chapter 22, this volume). This again emphasizes the importance in supporting farmers with improved and more accurate decision support systems for estimating EONR.

46

3.3.3╇ Nitrogen effects on quality of harvested products The supply of nitrogen has a profound influence on the content of a large number of macro-molecules and secondary metabolites in plants which are important for their quality characteristics in relation to use for food, feed, fibre and bio-energy. In particular, the relationship between N application rate and grain protein content in cereal crops has received consid� erable attention. High levels of N application result in increased grain protein content due to greater synthesis and accumulation of storage proteins. Particularly the content of gluten proteins, consisting of gliadins and glutenins, which together constitute more than 85% of the total protein content of wheat grains, is positively correlated with N-fertilization. Gluten proteins are the major determinant of the baking quality of wheat flour, affecting water absorption and mixing stability of the dough, its CO2 retention capacity and the bread volume (Shewry, 2009). Increasing applications of nitrogen fertilizer to wheat result in an increased proportion of gliadin proteins and increased dough extensibility (Godfrey et al., 2010). At the Broadbalk continuous winter wheat long-term experiment, the grain %N, protein composition and dough properties from plots receiving 35 t/ha farmyard manure per year, containing approximately 250 kg/ha of total N, was similar to that from the plot receiving 144 kg/ha per year N in inorganic fertilizer, indicating that much of the applied manure N was unavailable in the year of application (Godfrey et al., 2010). The minimum protein content required for bread making wheat is typically taken as 13% on a dry weight basis. However, farmers do often not get a substantial increase in payment for high-quality wheat with high protein content. Actually, the prices for baking quality wheat are in many cases not much higher than for feed quality wheat. In this connection it must also be emphasized that the quantity of protein is in itself not a sufficient parameter for characterization of wheat baking

Lars Stoumann Jensen and Jan K. Schjoerring

quality. Also the amino acid composition of the gluten proteins must be taken into account. The genetic constitution has a dominating influence on the composition of proteins and a large variability exists between different wheat genotypes. Thus, high quality wheat genotypes in general produce good bread wheat over a wide range of protein contents whereas poor bread wheat genotypes produce poor bread quality even if the protein contents are elevated by fertilization. In terms of feed quality of cereals, high rates of N application may lead to a relative decline in the protein quality due to limitations in the amounts of essential amino acids. This is especially the case for lysine which is recognized as one of the most important essential amino acids because it is frequently the first limiting amino acid for optimal utilization of protein in monogastric animals and humans. Consequently, intensive research has been dedicated to increase the lysine concentration in cereals. It is evident that the lysine concentration of grain protein cannot be improved by fertilization which has stimulated plant breeders to focus on genotypes with a high lysine production. Several barley and maize genotypes have been developed in which the lysine concentration has increased by more than 50%, but abnormal phenotypes are usually developed in these lysine-rich genotypes, resulting in reduced yields€(Shewry, 2009). The ideal grain protein concentration of malting barley for production of European lager beer is 10.7% of dry matter, with a permitted range of 9.5%–11.5%. Higher protein levels result in lower starch content, less alcohol and risks of cloudy beer, whereas yeast activity may be limited by N shortage at lower grain protein levels (Pettersson and Eckersten, 2007). This optimization of N fertilization for production of malting barley is a delicate balance, because reduced N application decreases yields, but favours the desirable low protein content of the grains. In addition, grain size, grain weight, extract yield and wort viscosity, all positive quality parameters in malting barley, are reduced when the rate of N application is increased. Growing barley with excellent malting performance is also complicated by the fact that low rainfall and high temperatures favour protein synthesis and might lead to excessive protein contents even if the N-application is kept at a low level. Important quality parameters for crops grown for bioenergy purposes are summarized by Karp and Shield (2008). There is limited knowledge on how increasing nitrogen supply affects the composition and proportions of ligno-Â�cellulosic compounds in plant cell walls. Wheat straw consists of 35%–40% cellulose, 20%–30% hemi-cellulose and 20%–25% lignin (Mosier et al., 2005). N fertilization seems to cause a small decline in ligno-cellulose per unit straw dry matter (Porteaus et al., 2009), but the effect is not marked. This may be related to the fact that N stimulates the biosynthesis of phenylalanine and tyrosine which are precursors for lignin biosynthesis, thereby counteracting the general decline in C/N ratio in response to increased tissue N status. In poplar trees, N fertilization decreases wood density, cell wall thickness and lignin content (Pitre et al., 2007).

3.4╇ Trends in European N use in livestock production 3.4.1╇ Livestock productivity in EU-27 and feed resources Since the Second World War, animal production in Europe has undergone a substantial increase. Expansion of fertilizer use and imported feedstuffs from outside of Europe has contributed to the increased production. Nowadays, the 27 Member States of the European Union are self-sufficient for milk and€meat. Characteristics of European livestock industry are shown in Table 3.4. First of all, Europe produces 26% of world milk production, achieving this from only 2% of the world’s grasslands. High fertilizer use per hectare of grassland and high milk production levels per dairy cow in Europe are the responsible driving forces. Second, the EU production of pig and poultry meat and eggs totals 40 million ton, representing 16% of the global production of these products. Pigs and poultry are fed especially with crop products and so they rely on arable land. The EU share of global production for pig and poultry meat is double that of global arable land (17.9% vs. 8.6%, Table 3.4). This illustrates how substantial imports of protein rich oil cakes and meals from other parts of the world are responsible for the high level of European pig and poultry production. European livestock consumed in total 473 million ton of feedstuffs in 2007 (Fefac, 2009). Roughages and cereals grown and consumed on farm of origin contributed 48% and 13%, respectively, to this feed base. Compound feed and other feed materials contributed the remaining 32% and 7%, respectively. One third of the consumed compound feed was imported, mainly consisting of oil cakes and meals, and feed cereals (Fefac, 2009).

3.4.2╇ N use in the diets of pigs and poultry It has long been known that protein is an essential dietary component for all animals. Later, it was realized that it was not protein per se, but amino acids as the constituents of proteins that played the essential role (Lewis and Southern, 2001). This basic knowledge on the importance of amino acids in relation to productivity (growth and reproduction) has been central for improvements in nitrogen utilization efficiency (NUE) in monogastric animals for more than three decades. It was also recognized that some of the 20 different amino acids present in proteins were essential and have to be fed to the pigs and poultry, whereas some are non-essential and need not be provided in the diets because these amino acids are synthesized by the animals. Unfortunately, cereals and protein feedstuffs also contain an overload of non-essential amino acids in relation to the animals’ requirements. The factors affecting pig and poultry productivity and NUE are summarized in Table 3.5, providing an overview of the main issues to be addressed in order to improve NUE in pigs and poultry. Historically, the first approach in modern pig and poultry farming was to feed protein-sufficient diets to animals. This resulted in an improvement in the overall

47

Benefits of nitrogen for food, fibre and industrial production Table 3.4╇ Global and EU-27 data for land use, animal numbers, and animal products for the year 2007

Unit

World

EU-27

Arable land

106 ha

1â•›411

121

8.6%

Grassland

”

3â•›378

69

2.0%

Cattle

10

1â•›361

90.3

6.6%

6

EU-27 share of world

Dairy cows

”

245

Pigs

”

921

161

17.5%

Poultry

”

17â•›887

1â•›341

7.5%

Dairy milk

10 ton

571

148

25.9%

6

24.3

9.9%

Beef meat

”

62.3

8.2

13.2%

Pig meat

”

99.5

22.7

22.8%

Poultry meat

”

88.0

10.9

12.4%

Eggs

”

59.3

6.4

10.8%

Source:€FAO (2009).

Table 3.5╇ Main factors affecting the productivity and nitrogen excretion and utilization in pigs and poultry

Factors Dietary means

Methods

Productivity

N excretion

N utilization (NUE)

Protein balanced diets

Higher productivity

High

Low

Balanced dietary amino acid supply

↔

↓

↑

Substitution of protein with industrial amino acidsa

↔

↓

↑

Increasing bioavailability of amino acids

↔

↓

↑

According to physiological requirement

↔

↓

↑

Use of increased number of diets

↔

↓

↑

Optimization of feeding systems, e.g. avoid waste

↔

↓

↑

Breeding

Selection programmes

More efficient (kg gain / kg feed intake; litter size)

↓

↑

Management

Compilation of above mentioned methods/ tools

High productivity, welfare and product quality

↓

↑

Feeding strategy

a

The effect depends on which essential amino acids are available on the feed market, but the number of industrial (crystalline) amino acids increases.

nutrient efficiency, as the productivity of the pigs was markedly increased and because the maintenance requirement for protein/amino acids was lowered due to the higher daily performance. Increased knowledge on the specific need for the different essential amino acids was followed by tools for balancing the dietary nutrient contents. Thus, practical diets for pigs and poultry were composed to fulfil the specific needs for amino acids by optimizing the use of the available feedstuffs. The first limiting amino acid in cereals is lysine calling for the need for protein supplements from alternative sources such as soybean, rape seed, sunflower meal, etc. Industrially

48

produced amino acids may also be used in crystalline form, but in this case it is also necessary to address the next limiting amino acids (normally methionine, threonine, tryptophan, isoleucine, valine, histidine and others). Nowadays, crystalline amino acids are widely used in order to reduce the overall protein content in diets for monogastric animals, but the exact use depends on economy and the demand for reduction in environmental emissions. Besides addressing the protein and amino acid content (profile) of feedstuffs, efficient N use requires that the amino acids are bio-available, because unavailable protein/ amino acids will be excreted and not utilized. In summary,

Lars Stoumann Jensen and Jan K. Schjoerring Intake (100%) Indigestible (18%)

Digestible (82%) Maintenance (10%)

Feces (18%)

Not used (37%)

Production (35%)

Urine (47%) Excretion (65%)

diet formulation needs information on amino acid content and availability in each single feedstuff in order to optimize a balanced diet for the animals according to their requirement. Most countries use feed evaluation systems and optimization programmes for balancing proper diets. Lowering the proportion of feed with a low protein digestibility in the diet in favour of cereals and other feedstuffs with a higher protein digestibility will result in a better balance of dietary protein. A schematic representation of N flow in growing pigs is presented in Figure 3.15, which shows that only onethird of the N provided with the feed is retained in the growing pig. The proportion not used (37%) is mainly due to an imbalance in the amino acids provided. A temporally varying feeding strategy is another tool used to improve NUE in pig and poultry production. In principle, each single animal should be fed exactly what it needs each single day. Although this is not possible in practice, it is now widespread to use more than one diet through the whole production period. This means that several diets are used sequentially for feeding a pig from weaning to slaughter. Two diets, viz. a starter diet and a weaner diet, may be used for the young pig and two or more diets for the growing pig. The nutrient content of each diet is adjusted to the physiological need for that period (Dourmad et al., 1999). Phase feeding systems are also used in poultry production in order to fit the nutrient supply to the animals’ requirement. Another€ – often forgotten€– factor in order to improve NUE, is to minimize the loss of feed caused by inappropriate feeding equipments and feeding systems. Finally, the effect of animal breeding should be considered. In modern animal husbandry, the animals have undergone genetic selection for productivity. This means that modern breeds demand less quantity of feed for producing one unit of product, implying an improvement in NUE. Animal management factors are also very important in order to ensure a high NUE. These include the use of efficient feeding equipment and systems, appropriate shifts in diets at different production stages as well as practices to ensure good hygiene and health status of the animals. Improvements in NUE for pig and poultry farming have been achieved over the last two decades in many European countries. Denmark can be considered as one of the leading countries in this respect, reflecting environmental pressures to improve NUE and N excretion rates. As shown in Table 3.6, the N intake in Danish finisher pigs, covering the period from 30 kg body weight until slaughter, was gradually reduced

Figure 3.15 Scheme of nitrogen flow in growingfinishing pigs from 25 to 110 kg.

by 30% from 1985 to 2009. At the same time, the N excretion decreased from 72 to 41 g per kg body weight gain. As a consequence, NUE has significantly increased from 28% to 42% by means of the tools mentioned in Table 3.5. These numbers demonstrate that feeding management, genetic breeding and diet composition are important tools to improve NUE. However, in the majority of European pig production systems, much still needs to be done to achieve this improvement in NUE (Jongbloed and Lenis, 1998). Corresponding results have been obtained in poultry, where genetic improvements have contributed to the improvements in NUE together with dietary changes. As already discussed, feedstuffs not only vary in N, amino acid content and digestibility, but also in amino acid content expressed per kg N. For economic reasons, it is impossible to formulate diets without oversupplying certain amino acids, particularly when many by-products from the food-processing industry are used.

3.4.3╇ Dairy farming Dairy farming systems combine animal production and grassland production. A useful way to analyse such systems is to consider NUE and the related input–output balance on a farm scale. At this scale, inputs of Nr to the farm include bought fertilizer and feedstuffs, while the outputs consist of sold milk and meat. Since animal manure is produced on the farm it does not figure directly in a farm input–output balance or in the calculation of overall NUE, although N losses are represented indirectly by reducing outputs. There is an extensive literature on nutrient balances for dairy farming in European countries (see also Jarvis et al., 2011, Chapter 10, this volume). The nutrient balance studies can be divided in to three groups: (A) nutrient balances on farming systems (see below), (B) nutrient balances on animal production (section 3.4.4), (C) nutrient balances on grassland production (section 3.4.5). Nutrient balances from group B and C facilitate the interpretation of the farm scale balances in group A. A group of 20 intensive cattle farms in Portugal, with zerograzing, had over a period of three years an average NUE for milk and meat of 33%. The intensive production on these farms was characterized by an average surplus of 502 kg N/ha (Fangueiro et al., 2008). In principle, such intensive systems without grazing can achieve higher NUE, but suffer from very high N surpluses.

49

Benefits of nitrogen for food, fibre and industrial production Table 3.6╇ The development in nitrogen intake, excretion and utilization in an average Danish finisher pig from 1985 to 2009. Calculations based on standard values according to Poulsen et al. (2006), which are updated annually (Poulsen, 2009). The values are given per kg gain for a finisher pig (30 kg to slaughtering). The body weight at slaughter has gradually increased from less than 100 (1985) to 107 kg (2009)

1985

1990

1995

2000

2005

2009

Intake, kg N/animal

7.1

6.5

5.2

5.1

5.3

5.2

Excretion, kg N/animal

5.1

4.5

3.3

3.2

3.2

3.0

Excretion, g/kg gain in animal weight

72

65

47

45

44

41

Nitrogen Use Efficiency (NUE), % of N intake utilized

28

30

37

38

40

42

80

Figure 3.16 N use efficiency on European dairy farms expressed as annual N output in milk and meat per annual N input in fertilizer + feed. Slope of the curves indicates gross NUE (based on data from Bleken et€al., 2005).

70

N output (kg/ha)

60 y = 0.1302x + 21.23

50

R 2 = 0.66

40 30 20

Milk y = 0.1114x + 16.41

10

Milk and meat

R 2 = 0.57

0 0

50

100

150

200

250

300

350

N-input fertilizer +feed (kg/ha)

By contrast, extensive cattle farms in Finland achieved, over a one year period, an almost similar average NUE for milk and meat of 25%, but a much smaller average surplus of 109 kg N/ ha (Virtanen and Nousiainen, 2005). Finally, 21 intensive dairy farms in Ireland were studied during a period of four years. The first year, the average NUE for milk and meat was 18% and the surplus was 277 kg N/ha. Due to lower fertilizer application rates, the average NUE increased to 20% and the surplus decreased to 232 kg N/ha in the last three years (Treacy et al., 2008). Bleken et al. (2005) reviewed a large number of European dairy efficiency studies and found a gross NUE of only 13% for the combination of milk and meat produced or 11% when only the milk was considered (Figure 3.16). There are many options to improve the NUE on dairy farms. Feeding management addresses the crude protein content of the ration, for example by decreasing the N fertilizer application rate on grassland or by implementing fodder maize or other crops in the ration. Manure management involves minimizing ammonia losses during animal housing and manure storage, application of manure under favourable weather conditions during the crop growing season, and minimizing the contact time between manure and the atmosphere. Shortening of the grazing period also has a positive effect on the NUE due to the low N fertilizer value of excreta deposited in the meadow,

50

although housing can increase the percentage loss of N as ammonia to the atmosphere (Webb et al., 2005). Intensification of production per animal means that less animals are needed for the same production level of milk at the farm scale. Feed intake on the farm scale can therefore be reduced and this leads to higher NUE and lower N surplus. Replacing grass with more on-farm grown fodder crops can lead to rations more balanced to the animal needs and this will improve NUE and lower the N surplus at farm level. Intensification of milk production at the farm level with external feedstuffs will also typically improve NUE, but it will, however, increase farm N surplus, and hence the risk of environmental losses. Furthermore, the environmental N losses occurring from the field production of the external feedstuffs are also not included in the balance (Kohn et al., 1997; Schröder et al., 2003; Bleken et al., 2005).

3.4.4╇ Animal production and nitrogen use€efficiency From the previous section, it can be seen that there is potentially a tension between improving NUE and minimizing nitrogen surplus. In seeking to maximize the benefits of N, strategies are therefore desirable that benefit both indicators. In this way reduction of N excretion (and hence N surplus) may be associated with improved NUE.

Lars Stoumann Jensen and Jan K. Schjoerring Table 3.7╇ Nitrogen excretion and nitrogen use efficiency for Dutch livestock categories. Figures are converted to animals kept housed for 365 days per year. Data from CBS (2009)

Nitrogen excretion (kg/year)

Nitrogen Use Efficiency (% of intake)

1990

2008

1990

2008

Female cattle for replacement < 1 year

40.1

34.9

14.3%

15.5%

Female cattle for replacement > 1 year

93.1

73.7

6.2%

6.6%

141.7

127.6

19.4%

26.3%

Female cattle for replacement < 1 year

44.3

39.5

13.1%

14.0%

Female cattle for replacement > 1 year

95.9

76.7

6.1%

6.4%

157.0

144.2

17.8%

23.3%

Veal calves on milk

10.6

10.7

51.2%

49.8%

Veal calves on fodder maize

30.8

27.4

28.9%

28.6%

Male beef cattle < 1 year

28.9

26.0

28.0%

30.7%

Ruminant animals Dairy:€Regions with high fodder maize ration

Dairy cows; lactating cows Dairy:€Regions with low fodder maize ration

Dairy cows; lactating cows Beef cattle:

Male beef cattle > 1 year

72.6

53.8

10.9%

15.7%

110.7

84.9

10.1%

11.4%

Sheep (including lambs)

25.0

14.4

9.3%

13.5%

Goats (including kids)

19.9

16.0

15.9%

25.2%

Suckling cows

Horses

58.4

1.9%

Monogastric animals Fattening pigs

14.3

12.9

29.8%

35.6%

Sows (including piglets)

33.8

30.8

28.0%

36.8%

Broilers

0.61

0.53

41.0%

49.9%

Laying hens < 18 weeks

0.38

0.34

22.9%

26.9%

Ducks for meat

1.12

0.76

34.3%

49.3%

Turkeys

1.98

1.71

37.5%

44.8%

NUE is calculated as the ratio of nitrogen in milk and meat over nitrogen intake with roughage and concentrates. The milk production per cow for the two indicated years was 6050 and 7926 litres per year. The calculation method made use of country-wide average feed intake levels for all individual animal categories concerning concentrates, ensiled grass and fodder maize. For ruminants, the amount of consumed grass was assumed to close the energy demand of the animals.

Excretion rates can be calculated as part of an input–output balance approach on animal level, using the simple equation Excretion = feed intake – animal products. Assessments of the N excretion rates and the corresponding N use efficiencies for the main animal categories in the Netherlands show that dairy cattle N excretion rates in regions with a high share of fodder maize in the ration are lower than those in regions with low share (Table 3.7). An increasing

milk yield does not necessarily lead to increased N excretion rates, partly caused by decrease of the amount of applied fertilizer N in the same period, resulting in a lower protein content of the consumed grass. However, according to Witzke and Oenema (2007) comparing data from EU member states, there is a reasonably close positive relationship between milk yield and N excretion rates, with a standard variation across Europe of ±23%. It should be underlined that these N excretion rates are in fact the resultant of different animal rations with varying protein contents between the EU member states.

51

Benefits of nitrogen for food, fibre and industrial production

Ruminants generally have a lower NUE than pigs and poultry, as seen from Table 3.7. Finally, young animals have a higher NUE than older animals. Using the same principle for NUE, an intensive Italian survey was undertaken for estimating N excretion rates. Based on about 10 000 cows with an average milk yield of 8366 litre of milk per year, an average N excretion of 116 kg per year was found. Compared to the Netherlands, the Italian rations were mainly based on corn silage with lower feed N concentrations, thereby reducing N excretion (Xiccato et al., 2005). Feeding trials with high yielding cows on a research farm of the Swedish University of Agricultural Sciences showed that varying the crude protein content of the ration between 135 and 184 g crude protein per kg feed resulted in NUE for milk production between 18% and 40%, with the lower protein content increasing NUE (Nadeau et al., 2007). Another approach is to use a model based on feed intake according to the energy requirements for maintenance, meat and milk production; this will enable improved optimization of NUE and, hence, lowering of N excretion rates (Vérité and Delaby, 2000; Peyraud and Delaby. For an overview of N recovery efficiencies in EU-27, the USA and the Netherlands see supplementary material for Chapter€3. 2004; Dämmgen et al., 2009).

3.4.5╇ Economic value of N in dairy farming To illustrate the economic value of N in livestock production, the case of dairy farming is used here as an example.

Grassland productivity and N response Grassland productivity is affected by climatic factors such as rainfall and temperature and depends on the specific farm management. Nitrogen is one of the key factors to improve the productivity of grasslands, assuming no other nutrients are limiting. Soil N supply (SNS) is that originating from other sources than fertilizer. Recommendations for fertilization of grassland and arable land take the SNS into account, implying that the amount of inorganic fertilizer or animal manure can be lowered accordingly. An analysis of Dutch nitrogen fertilizer experiments on grassland during the period 1934–1994 showed that increasing fertilizer applications resulted in higher N uptake by the grass, but at the same time the N use efficiency (NUE) decreased. The analysis also showed that grazing leads to a higher SNS and to a lower NUE, compared to cutting. These effects were stronger with pure grazing than with mixed grazing and cutting (Vellinga and André, 1999). For reasons of uneven distribution and high local N loadings with urine, N in the manure which is deposited directly in the meadow has a much lower mineral fertilizer equivalent (MFE) value than N in manure spread on the field provided appropriate abatement techniques of ammonia loss are implemented. Today, the optimum N fertilization rate is based on economic and environmental targets. Until about 1990, the economic criterion was a marginal N response of 7.5 kg dry matter of herbage per kg N fertilizer applied. This criterion implied economically optimal fertilizer rates up to around 400 kg N/ ha grassland (Prins, 1983) when harvested as cut sward. Much lower values (around 200 kg N/ha) are found for grazed grass swards (Deenen and Lantinga, 1993; Lantinga et al., 1999;

52

Nevens and Reheul, 2003). Nowadays environmental targets are directed to lower protein contents of herbage and to meeting the Nitrate Directive, both leading to lower fertilization levels in the Netherlands (Vellinga et al., 2004; Oenema et al., 2011, Chapter 4 this volume). Information on the geographical distribution and corresponding productivity of European grasslands has been published recently (Smit et al., 2008). The potential grass yield varies strongly between regions in Europe and reflects areas with different natural productivity levels (Peeters and Kopec, 1996; Smit et al., 2008). The potential production of herbage dry matter (DM) can be divided into three classes: • 10–15 ton/ha:€North-western Europe (Atlantic coastal area), • 5–10 ton/ha:€North and east Europe, • 0–5 ton/ha:€Southern Europe (semi-arid Mediterranean, not irrigated). Grass yield trials in NW Europe (cut grass only) show annual DM yields up to about 16 ton/ha at N rates of 300–500 kg/ha (Figure 3.17). For cut grass, annual yields without N fertilizer application range between 2 and 6 ton/ha and the maximum yield between 8 and 18 ton/ha.

Effect on milk production Based on datasets of 19 dairy farm groups (Bos et al., 2003; Raison et al., 2006; Bleken et al., 2005; Aarts et al., 2008) there is a fairly good correlation between the fertilizer N rate applied to fodder crops (mainly pasture) and the milk production (R2 = 0.57, same dataset as in Figure 3.16). A survey of 139 dairy farms in the Atlantic area, ranging from Ireland to Portugal, shows a ratio of milk production per unit N applied to fodder crops ranging from about 29 kg milk per kg N on extensive farms with grazing to 547 on intensive farms without grazing (Raison et€al., 2006). Data suggest that intensive farms are more efficient with fertilizer N (more milk per kg N applied) but this is also an effect of a higher proportion of feed concentrates and imported feedstuffs in total N-input. The N-losses for production of these concentrates or imported fodder are not accounted for in the farm balance, and hence the true NUE for the overall production of milk may be lower (see also Section 3.4.3).

Economic return on N for milk production The economic return on N for milk production (ERoN; see also Section 3.3.2) were derived for grass yield response curves in the UK and in the Netherlands (Figure 3.18). The fertilizer N rate used in the calculations was that needed to obtain grass yields supporting a maximum annual milk production of 10.3€ton/â•›ha for the Dutch case and 5.9 ton/ha for the UK case. Because present-day intensive dairy farming uses considerable amounts of feed concentrates, a fixed annual value of 100 kg/ha N as feed concentrates was used, typical for intensive dairy farming. ERoN values for price levels since 2000 (fertilizer:milk price ratio increased from 2 to 3 and is still rising, see Figure€3.13), range from 2–7 € per kg milk/€ per kg N when targeting maximum milk yields per ha, and exceed 15 €/€ at present N-fertilizer levels in the Netherlands and the UK. ERoN values tend to be higher than those for arable agriculture, because,

Lars Stoumann Jensen and Jan K. Schjoerring 20

Figure 3.17 Examples of grass yield response curves for cut grass in response to annual N �fertilizer inputs.

Grass yield (ton/ha DM)

18 16

R 2 = 0.97

14 12 10 8 6

R 2 = 0.77

4 Schils et al. (1999) Netherlands Shiel et al. (1999) UK

2 0 0

100

200

300

400

500

600

700

N-fertilizer (kg/ha) 30

Figure 3.18 Gross economic return on nitrogen fertilizer (ERoN) application to grass for milk production as function of fertilizer N:milk price ratio (€/kg N per €/kg milk) for average fertilizer N rates (indicated in legend) and use of feed concentrates in the Netherlands and the UK. N response based on trials in the Netherlands and in the UK (see Figure 3.17).

UK Fert-70 kg/ha N, Feed-100 kg/ha N

25

20 (euro/euro)

Gross economic return on N

NL Fert-170 kg/ha N, Feed-100 kg/ha N

15

10

5

0 0

1

2

3

4

5

6

7

8

9

Fertilizer-N:milk price ratio

in contrast to wheat and oilseed rape, grass yield continues to increase up to N-rates beyond 400 kg/ha. The high ERoN for fertilizer N in milk production compared to crop production indicates how large the incentive is for the dairy farmers to apply large, and likely also excessive, amounts of fertilizer N for grasslands. Although farmers in south-eastern Europe generally use lower N fertilization rates than in north-western Europe, with significant possibilities for increasing their milk production level, the low ERoN under their production conditions (climate, soils, etc.) typically provides little incentive for them to increase fertilizer N use.

3.5╇ Industrial uses of dinitrogen gas and reactive nitrogen based compounds Industrial uses of nitrogen cover a range of different applications. The gas dinitrogen is used to maintain for instance an inert atmosphere, while reactive nitrogen forms (especially

ammonia and nitric acid) are used as ingredients in the chemical industry (rubbers, plastics including nylon, melamine), in the electronics industry for etching and pickling (nitric acid), in production of primary metals via leaching (nitric acid) and for cleaning catalysts used in for instance petroleum refining. In the food industry, ammonia is used for refrigeration of foods, and in the medical field it is used to refrigerate medical samples. Industrial dinitrogen gas uses worldwide are summarized in Table 3.8, while the following sections summarize they ways in which major reactive N compounds are used (Maxwell, 2004). In Europe, industrial and other uses than fertilizers consume 23% of total European ammonia production (Table 3.9) but 35% of total European ammonia is exported, so industrial and other non-fertilizer uses of ammonia constitute roughly one third of the total ammonia consumption. From Table 3.9 it can also be seen that western Europe is a net importer of ammonia, the majority coming from eastern Europe and central Asia.

53

Benefits of nitrogen for food, fibre and industrial production Table 3.8╇ Uses of industrial dinitrogen gas worldwide (Maxwell, 2004)

Application

Market share (%)

Chemical industry

33

Oil and gas extraction

14

Electronics

13

Primary metals

11

Petroleum refining

10

Food industry

5

Glass

2

Rubber and plastics Miscellaneous

1 11

3.5.1╇ Ammonia Ammonia is one of the best known bulk chemicals in the world, and the major synthetic nitrogen products made from ammonia are shown in Figure 3.19. Ammonia is predominantly used as feedstock for the production of fertilizers. It is produced by the Haber–Bosch process, invented by Fritz Haber in 1908, and turned into an industrial scale process by Carl Bosch in the years after. In the Haber–Bosch process, hydrogen from natural gas is combined catalytically with free nitrogen gas in the air at high temperature and pressure, yielding ammonia (Domene and Ayres, 2001). According to Yara (2009), 83% of all ammonia produced globally was used for fertilizer production. Beside from the direct fertilizer use of ammonia, it can also be used as a reactant with respective acids to produce ammonium nitrate, ammonium phosphate and ammonium sulphate. Reacting liquid ammonia with carbon dioxide at about 190 ºC and elevated pressure according to the so-called Basaroff reactions yields the fertilizer urea (Maxwell, 2004). The remainder of the ammonia is used in various other proÂ� cesses. The most important compound made using ammonia as a reactant is nitric acid. In three steps, ammonia is converted to nitric acid based on the so-called Ostwald process, using precious metals as catalysts at elevated pressure and temperature (Buchel et al., 2000). Nitric acid in turn is predominantly used to make explosives, like ammonium nitrate, nitroglycerine, trinitrotoluene and nitrocellulose. Ammonia is further used in the production of the cyclic amide caprolactam, a feedstock for the production of nylon-6, amines, polyacrylonitrile, hydrazine, polyurethanes, resins based on phenol or melamine, formaldehyde, nitriles, sodium nitrate, sodium cyanide and many others. Nitrogen compounds are particularly used in technologies for cleaning flue gases after fossil fuel combustion. This includes the reduction of nitrogen oxides using ammonia, both catalytically as well as non-catalytically (Caton and Xia, 2004; Wojciechowska and Lomnicki, 1999; Baukal, 2003). Ammonia is also used for the removal of sulphur dioxide, although the technology is rather new and as such it is not widely used. With this technique, an electron beam passes through the flue gas

54

and ammonia reacts with sulphur dioxide to yield ammonium sulphate (Chmielewski et al., 2002). Despite its toxicity, ammonia is regaining its position as a refrigerant due to the environmental concerns associated with chlorofluorocarbons. It has favourable thermodynamic properties, especially a low boiling point and high heat of evaporation, and is widely available for low prices (Redwood, 2010; Stoecker, 1998). In the metal industry, ammonia is used for the extraction of metals such as copper, gold and tungsten from their respective ores (Wohler, 2009). In the extraction process, the metal ores are suspended in an ammonia solution and subsequently heated, thereby creating the corresponding metal-amines, which can be isolated. Ammonia can also be used for annealing/nitriding of steel (Ross, 1988) and as a corrosion inhibitor after conversion to quaternary ammonium compounds (Sastri, 1998).

3.5.2╇ Ammonium nitrate Ammonium nitrate is predominantly used as a fertilizer and as an ingredient for explosives and propellants. In the United States, industrial explosives (including ammonium nitrate) account for approximately 4% of total reactive nitrogen output (Domene and Ayres, 2001). The major N containing explosives apart from ammonium nitrate are TNT, PETN, Tetryl, Nitroguanidine and Nitroglycerin. The principal non-military use of explosives is in coal mining, followed by quarrying, surface mining and construction work. All of the nitrogen contained in explosives is released directly into the atmosphere the moment they are used, mainly as free dinitrogen, but in the case of ammonium nitrate the majority is released as NO. Detonation of nitroglycerin also releases a significant proportion as N2O, as much as 97 kg per ton of nitroglycerin, according to theoretical model calculations (Domene and Ayres, 2001). Ammonium nitrate mixed with a suitable fuel, mostly fuel oil and as such abbreviated as ANFO, is a well known blasting agent used in the mining industry and for construction purposes, in which it has largely replaced dynamite (Persson et al., 1993; Tatiya, 2005; Monroe and Hall, 2006). It is used in a 95:5 weight ratio of prilled ammonium nitrate to fuel oil. Another well known mixture is AMMONAL, which consists of 60 wt% ammonium nitrate, 20 wt% trinitrotoluene (TNT) and 20 wt% aluminium. Unfortunately, owing to the large scale availability of both ingredients of ANFO, the mixture has also been used for the construction of so-called improvised explosive devices (Turkington, 2009).

3.5.3╇ Urea Urea is predominantly used as a fertilizer, since it has the highest nitrogen content of all known solid nitrogenous fertilizers (Schepers and Raun, 2008). Combined with formaldehyde, it forms a resin that is used in adhesives and plastics in general (called urea formaldehyde resins). The production of the bulk chemical melamine, a feedstock for predominantly plastics, is based on the ring closure of three urea molecules at elevated temperatures. On a smaller scale, urea is used as an ammonia source for removal of nitrogen oxides in flue gases, as an

Lars Stoumann Jensen and Jan K. Schjoerring Table 3.9╇ European potential nitrogen supply and demand balances in 2008 (FAO, 2008)

Europe, total

Central Europe

Western Europe

(million ton N and % of supply) NH3 max. prod. capacity (as N)

37.5

NH3 actual prod. (as N)

33.6

100%

4.8

100%

9.7

100%

19.0

100%

N fertilizer consumption

14.4

43%

2.7

55%

8.6

88%

3.1

17%

Non-fertilizer N demand & others

7.7

23%

0.6

13%

5.2

54%

1.9

10%

+11.5

+34%

1.5

+32%

−4.1

−42%

+14.0

+74%

Balance (+:€export, −:€import)

Monoammoniumphosphate(MAP) NH4H2PO4

6.2

Diammoniumphosphate(DAP) (NH4)2HPO4

10.3

Eastern Europe + Central Asia

Hydrogen cyanide HCN

21.0

Ammoniumsulphate(MAP) (NH4)2SO4

Ammoniumnitrate NH4NO3

NH3

H3PO4

CH4

H3PO4

C2H6 H2SO4 O2+H2O

NaOCl HCOH CH3OH

ClCH2CH2Cl

Ammonia NH3

CO2

Other or organics anics

Heat eat & catal catalyst s

Fig. 3.19 Synthetic nitrogen products made from ammonia (modified from Maxwell, 2004).

55

Benefits of nitrogen for food, fibre and industrial production

intermediate product in the pharmaceutical industry, as well as a reactant for the production of urea nitrate.

3.6╇ Economic value of reactive N use to the European economy There are various ways to approach the economic value of reactive N. Taking a global perspective, Erisman et al. (2008) argued that nearly 50% of the world human population in 2008 could be fed thanks to Haber–Bosch derived Nr applied as fertilizer. The global revenue on sale of fertilizers in 2005 amounted to nearly 30 billion USD (25 billion €; Yara, 2009). For the EU-27 countries it was estimated by Yara (2009) that the increase of wheat production in 2008 due to use of mineral N fertilizer was 64 million tons. This estimation is based on a comparison with the wheat yields achieved in ecological farming without mineral N fertilizer. The fertilizer-derived increase in wheat production represents a net economic gain (grain value minus fertilizer costs) of 7.6 billion € per year for the entire EU, or 280 €/ha. However, this net gain is sensitive to the relatively volatile world market prices for grain and fertilizer, as seen during the 2008–9 food crisis and subsequent financial breakdown, and assumptions on the potential yields in absence of mineral N fertilizer. At the level of a farm or a crop, the cost of N fertilizer is just one of several production factors. As described in previous sections, the economic return on investment in N (ERoN) is a very robust measure of importance for the farm economy and, hence, for the farmer decisions. Judging from Figures 3.14 and 3.18, the following current ERoN values can be summarized. The farmer will make a profit from N inputs if ERoN is above Product

ERoN (€ product / € fertilizer N)

Winter Wheat:

2–7

Oilseed Rape:

1–5

Milk:

10–15

one and the range in ERoN depends on (i) actual N fertilization level and (ii) shape of the response curve. A lower maximum yield for oilseed rape, wheat and grasslands is commonly found in south-eastern compared to north-western Europe. This is due especially to water limitation and implies a tendency for ERoN to be relatively low in south-eastern Europe compared to northwestern Europe, where climatic conditions favour higher potential yields under economically optimal fertilizer N input. The ERoN ranges presented in this chapter mean that for most farmers there is a huge economic profit from use of Nr, especially in relation to livestock production. The high ERoN for fertilizer N in milk production compared to crop production indicates how large the incentive is for the dairy farmers to apply large, and likely also excessive, amounts of fertilizer N for grasslands. In addition to chemical fertilizer, manure and biological nitrogen fixation are other sources of N that can be affected by farm management. The economy of N at the farm level is therefore quite complex. Costs of purchasing and handling of various N sources are quite different and change in time, e.g. depending

56

on the price of energy (natural gas) and environmental policies (see Oenema et al., 2011, Chapter 4, this volume). Compiling a comprehensive, robust inventory of the economic benefits to society of reactive nitrogen is not a simple matter. As indicated above, a coarse estimate may be that about half the value of European agricultural production may be considered as dependent on Nr supply. However, in a review of yield differences between organic and conventional farming in Europe, Offermann and Nieberg (2000) found that organic cereal yields are typically 60%–70% of those under conventional management, vegetable yields are often just as high as under conventional management and pasture and grassland yields in the range of 70%–100% of conventional yields. The derived consequences for economic profit or benefit are quite complicated as Offermann and Nieberg (2000) also state that the majority of the studies evaluated report an increase of labour needs, on average in the range of 10%–20% (but higher for vegetables), the cost of which has to be accounted for. Therefore the economic benefits of Nr use in agriculture are not easy to estimate. In the case of industry, the overall economic value includes nearly all explosives (including the economic value of military security; Erisman et al., 2008), the value of coal and other products mined with explosives, and the wide diversity of other nitrogen-containing chemical compounds. For industrial uses, however, i.e. especially explosives and plastics, there are alternatives for using Nr, and therefore the real value of Nr becomes very difficult to assess. For agricultural production, there is no simple substitute to Nr at the scale of its current level of use, but also the Nr contribution to agriculture is challengeable (Bruges, 2007). Although we have estimated that between 30%–50% of the current food production, population and GDP may be derived from use of Nr, to some extent Nr has also replaced labour. Historically, human development has been driven by the big transition in which labour force for agriculture was transferred to industry and services. The continued productivity in agriculture was ensured partly with fossil energy for machinery and Nr, partly with modern pest control agents and breeding for improved crop genotypes. Another issue is that economic benefits in the modern definition include the externalities, i.e. the negative effects (or benefits) of Nr for which there is no market. This issue is discussed at length in Chapter 22 of this volume (Brink et al., 2011). The real societal price of food is that including the external costs, or alternatively formulated, is the price of food produced without any external effects. Including externalities of Nr use (and of P, pesticides, fossil fuels, etc.) in the price may then enable transfer of part of the labour back to food production to maintain food production at lower external inputs. However, this approach would not be easily applicable in a market based economy. Given the many uncertainties in the assessment of the economic value of reactive N use to the European economy, the coarse estimate at the beginning of Section 3.6 may be as valid as any estimate derived from more refined calculations. Based on this and the additional data and arguments presented in this chapter it can be concluded that the overall benefits of N use are very substantial.

Lars Stoumann Jensen and Jan K. Schjoerring

3.7╇ Perspectives and recommendations

3.8╇ Conclusions

The need to maintain food and energy security under an increasing world population poses major challenges to supply the quantity and quality of commodities (including biofuels), given the few options to increase arable land area. With its resource of relatively fertile and productive soils, Europe has a clear capability for contributing to this, and it may be argued that Europe also has a moral obligation to do so. However, increased land use changes elsewhere in the world may not exclusively be due to an eventually diminishing agricultural production in Europe if inputs of reactive nitrogen are significantly reduced, but these possible secondary effects of reducing European fertilizer N rates must be taken into account. At the same time, environmental concerns, including agricultural responses to climatic change, as well as the need to feed the growing global population, represent a major challenge for further improvement of nitrogen benefits, i.e. to increase the use efficiency of the reactive nitrogen applied. The following recommendations for policy decisions and research priorities can be made. • Initiatives, whether voluntary or legislative, to reduce the use or surplus of nitrogen in agriculture, including inorganic fertilizer N, should take account of the need to maintain the nitrogen benefits in agricultural production€– food, feed and biomass productivity should be maintained while improving N use efficiency. • Modified field management practices for N conservation, modifications to livestock diets and recycling of wastes can enhance benefits per unit Nr used, and should be strongly promoted as best available technology (BAT). • New developments, combined with stimulatory incentives for farmers, should promote innovative technological tools to improve resource-efficiency and the overall benefits of N€use: (i) management strategies involving N-conserving field practices (e.g. catch crops, reduced soil tillage, better timing of N inputs, etc.), (ii) modifications to livestock diets for decreasing N excretion rates, (iii) enhanced manure N use efficiency through improved environmental technologies for management, recycling and field application of manures, (iv) improved accounting of field level N responses depending on cropping practices, soil fertility and climate.

• Although considerable uncertainty exists in the assessment of overall benefits of reactive nitrogen, particularly as regards the economic value of Nr in industrial production, it can be concluded that Nr is very much a key factor for achievement of food security and social welfare in Europe. • Maintaining food and energy security under an increasing world population poses major challenges to supply the quantity and quality of commodities (food, feed, fibre and fuels). Changing the input of reactive nitrogen significantly to European agriculture may influence conversion of natural land areas to cropped land elsewhere in the world. • Future legislative actions to reduce the use or surpluses of nitrogen in agriculture should take account of the need to maintain benefits for food security and farm economy in Europe. • There is still a large potential for increased nitrogen efficiency in European agriculture by better management strategies, improved recirculation of nitrogen in waste materials, adoption of new fertilizer technologies, crop monitoring tools and new crop cultivars, all demanding improved skills of the individual farmers and their advisory€service. • The economically optimal N application rate for crops varies significantly across field, farms and regions, depending on crop type, crop N response, farm type, soil type and climate. • Crop N use efficiency can be increased by improving prediction of the economically optimal N rate, but at the current relatively low ratio between nitrogen fertilizer costs and crop prices, farmers often have relatively little economic incentive to restrict N application, so long as environmental effects are considered as externalities. • Nitrogen use efficiency for livestock production can be greatly improved, especially with optimized feed protein and amino acid composition, but also by animal breeding. Although intensification of livestock production with external feedstuff may increase N use efficiency, it should be noted that this may lead to larger local surplus of N (and other nutrients), necessitating application of environmental technologies for waste and manure processing to avoid increased environmental load. • For dairy farming, nitrogen use efficiency can be improved by adjusting the nitrogen content of the feed to the requirements of the cattle and by minimizing the ammonia losses from animal housing and during manure application.

• New research initiatives should focus on: (i) breeding plant species and crop varieties with improved nitrogen use efficiency through increased root length density at depth, high capacity for N accumulation in the stem, high maximum N-uptake rate and N remobilization during grain-filling, (ii) improved composition of major feed crops and novel feed additives, e.g. proteins from bio-fuel production waste and other means of increasing feed N responses per unit mass Nr used, (iii) new technologies for improving fertilizer application and sensing of crop N demand, including tools for improved utilization of N in agricultural and urban waste materials to increase overall N use efficiency.

Acknowledgements The authors of this gratefully acknowledge support from the European Science Foundation for the NinE programme and the COST Action 729. In addition, the lead authors acknowlÂ� edge financial support from The Danish Research Council for Technology and Production (274–08–0439), The Strategic Research Council (09–067246), The Hofmansgave Foundation, The Danish Ministry of Food, Agriculture and Fisheries

57

Benefits of nitrogen for food, fibre and industrial production

(3304-FVFP-09-B-004) and the European Commission (NitroEurope Integrated Project). Christian Pallière acknowledges the support of Fertilizers Europe and Joachim Lammel and Frank Brentrup the support of Yara International.

Supplementary materials Supplementary materials (as referenced in the chapter) are available online through both Cambridge University Press: www.cambridge.org/ena and the Nitrogen in Europe website: www.nine-esf.org/ena.

References Aarts, H. F. M, Daatselaar, C. H. G. and Holshof, G. (2008). Bemesting, Meststofbenutting en Opbrengst van Productiegrasland en Snijmaïs op melkveebedrijven. Plant Research International B.V., Wageningen. Akiyama, H., Yan, X. Y. and Yagi, K. (2010). Evaluation of effectiveness of enhanced-efficiency fertilizers as mitigation options for N2O and NO emissions from agricultural soils:€meta-analysis. Global Change Biology, 16, 1837–1846. Balmford, A., Green, R. E. and Scharlemann, J. P. W. (2005). Sparing land for nature:€exploring the potential impact of changes in agricultural yield on the area needed for crop production. Global Change Biology, 11, 1594–1605. Barlóg, P. and Grzebisz, W. (2004). Effect of timing and nitrogen fertilizer application on winter oilseed rape (Brassica napus L.) I. Growth dynamics and seed yield. Journal of Agronomy and Crop Science, 190, 305–313. Baukal, C. E. (2003). Industrial Combustion Pollution and Control. CRC Press, Boca Raton, FL. Bellarby, J., Foereid, B., Hastings, A. and Smith, P. (2008). Cool Farming:€Climate Impacts of Agriculture and Mitigation Potential. Greenpeace International, Amsterdam, The Netherlands. (available at http://www.greenpeace.org). Berry, P. M. and Spink, J. H. (2009). Canopy Management and Late Nitrogen Applications to Improve Yield of Oilseed Rape, HGCA Project Report No. 447. Birkmose, T. S. (2009). Nitrogen recovery from organic manures:€Improved slurry application techniques and treatment€– the Danish scenario. In: Proceedings 656, International Fertilizer Society, York, UK. Bleken, M. A., Steinshamn, H. and Hansen, S. (2005). High nitrogen costs of dairy production in Europe:€worsened by intensification. Ambio, 34, 598–606. Bos, J., Pflimlin, A., Aarts, H.F.M. and Vertès, F. (eds.) (2003). Nutrient management at farm scale. In: First Workshop of the EGF working group ‘Dairy farming systems and environment’, Quimper, France 23–25 June 2003. Plant Research International, Wageningen. Brentrup, F. and Palliere, C. (2008). GHG emissions and energy efficiency in European Nitrogen fertilizer production and use. In: Proceedings 639, International Fertilizer Society, York, UK. Brink, C., van Grinsven, H., Jacobsen, B. H. et al. (2011). Costs and benefits of nitrogen in the environment. In:€The European Nitrogen Assessment, ed. M.A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Bruges, J. (2007). The Big Earth Book:€Ideas and Solutions for a Planet in Crisis. Alastair Sawday, London. Buchel, K. H., Moretto, H. H. and Woditsch, P. (2000). Industrial Inorganic Chemistry, 2nd edn, Wiley-VCH, New York.

58

Burney, J. A., Davis, S. J. and Lobella, D. B. (2010). Greenhouse gas mitigation by agricultural intensification. Proceedings of the National Academy of Sciences of the United States of America, 107, 12052–12057. Cassman, K. G., Dobermann, A., Walters, D. T. and Yang, H. (2003). Meeting cereal demand while protecting natural resources and improving environmental quality. Annual Reviews of Environmental Resources, 28, 315–358. Caton, J. A. and Xia, Z. (2004). The selective non-catalytic removal of nitric oxides from engine exhaust streams:€comparison of three processes. Journal of Engineering for Gas Turbines and Power, 126, 234–240. CBS (2009). Animal Manure and Minerals 1990–2008* (Dierlijke mest en mineralen 1990–2008*). Centraal Bureau voor de Statistiek, Den Haag/Heerlen (in Dutch). Cerrato, M. E. and Blackmer, A. M. (1990) Comparison of models for describing corn yield response to nitrogen fertilizer. Agronomy Journal, 82, 138–143. Chien, S. H., Prochnow, L. I. and Cantarella, H. (2009). Recent developments of fertilizer production and use to improve nutrient efficiency and minimize environmental impacts. Advances in Agronomy 102, 267–322. Chmielewski, A. G., Tyminski, B., Pawelec, A. and Licki, J. (2002). Industrial plant for flue gas treatment with high power electron accelerators. In: AIP Conference Proceedings, 680 (1), International Conference on the application of accelerators in research and Industry, Denton, Texas, USA. Dalling, M. J. (1985). The physiological basis of nitrogen redistribution during grain filling in cereals. In: Exploitation of Physiological and Genetic Variability to Enhance Crop Productivity, American Society of Plant Physiologists, Chicago, IL, pp. 55–71. Dämmgen, U., Haenel, H.-D., Rösemann, C. et al. (2009). Improved national calculation procedures to assess energy requirements, nitrogen and VS excretions of dairy cows in the German emission model GAS-EM. Landbauforschung€– vTI, Agriculture and Forestry Research, 3, 233–252. Dampney, P., Dyer, C., Goodlass, G. and Chambers, B. (2006). Crop responses to Different N Fertilizer Materials. Component report for Defra report NT 2605. Defra, London. de Fries, W., Leip, A., Reinds, G. J. et al. (2011). Geographic variation in terrestrial nitrogen budgets across Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Deenen, P. J. A. G. and Lantinga, E. A. (1993). Herbage and animal production responses to fertilizer nitrogen in perennial ryegrass swards. I. Continuous grazing and cutting. Netherlands Journal of Agricultural Science, 41, 179–203. Domene, L. A. F. and Ayres, R. U. (2001). Nitrogen’s role in industrial systems. Journal of Industrial Ecology, 5, 77–103. Dourmad, J. Y., Seve, B., Latimier, P. et al. (1999). Nitrogen consumption, utilisation and losses in pig production in France, The Netherlands and Denmark. Livestock Production Science, 58, 261–264. EFMA (2003). Understanding nitrogen and its use in agriculture. European Fertilizer Manufacturers Association (now Fertilizers Europe), report, 64 pp. EFMA (2007). Forecast of food, farming and fertilizer use in the European Union 2007–2017. European Fertilizer Manufacturers Association (now Fertilizers Europe), report, 7 pp. EFMA (2009). Forecast of food, farming and fertilizer use in the European Union 2009–2019. European Fertilizer Manufacturers Association (now Fertilizers Europe), report, 20 pp.

Lars Stoumann Jensen and Jan K. Schjoerring Erisman, J. W., Sutton, M. A., Galloway, J., Klimont, Z. and Winiwarter, W. (2008). How a century of ammonia synthesis changed the world. Nature Geoscience, 1, 636–639. EU Glossary (2010). Common agricultural policy (CAP). Available at http://europa.eu/scadplus/glossary/agricultural_policy_en.htm (accessed 4 Aug 2010). Eurostat (2009). Agricultural statistics, Annual selling prices of crop products per 100 kg; wheat and rape 1995–2008. Accessed at http:// epp.eurostat.ec.europa.eu/portal/page/portal/agriculture/data/ database June 2010. Fangueiro, D., Pereira, J., Coutinho, J., Moreira, N. and Trindade, H. (2008). NPK farm-gate nutrient balances in dairy farms from Northwest Portugal. European Journal of Agronomy, 28, 625–634. FAO (2008). Current World Fertilizer Trends and Outlook to 2012. Food and Agriculture Organization of the United Nations, Rome. FAO (2009). FAOSTAT Database Collections. Food and Agriculture Organization of The United Nations. Rome. http://faostat.fao.org/. Fefac (2009). Feed and Food: Statistical Yearbook 2007. Fefac, Brussels. Feller, C., Fink, M., Maync, A. et al. (2001) Düngung im Freilandgemüsebau. Institut für Gemüse- und Zierpflanzenbau Großbeeren/Erfurt e.V. Foulkes, M. J., Hawkesford, M. J., Barraclough, P. B. et al. (2009). Identifying traits to improve the nitrogen economy of wheat:€Recent advances and future prospects. Field Crops Research, 114, 329–342. Galloway, J. N., Schlesinger, W. H., Levy, H. II, Michaels, A. and Schnoor, J. L. (1995). Nitrogen fixation:€atmospheric enhancement–environmental response. Global Biogeochemical Cycles, 9, 235–252. Godfrey, D., Hawkesford, M. J., Powers, S. J., Millar, S. and Shewry, P. R. (2010). Effects of crop nutrition on wheat grain composition and end use quality. Journal of Agricultural and Food Chemistry, 58, 3012–3021. Grant, R., Pedersen, L. E., Blicher-Mathiesen, G. et al. (2009). Landovervågningsoplande 2007, Report no. 709 from DMU/NERI (in Danish, with English summary). Gyldenkærne, S. and Mikkelsen, M. H. (2007). Projection of the Ammonia Emission from Denmark from 2005 until 2025, NERI Technical Report No. 239. National Environmental Research Institute, University of Aarhus. Henke, J., Breusted, G., Sieling, K. and Kage, H. (2007). Impact of uncertainty on the optimum nitrogen fertilisation rate and agronomic, ecological and economic factors in an oilseed rape based crop rotation. Journal of Agricultural Science, 145, 455–468. Herridge, D. F., Peoples, M. B. and Boddey, R. M. (2008). Global inputs of biological nitrogen fixation in agricultural systems. Plant and Soil, 311, 1–18. IPCC (2007). Intergovernmental Panel on Climate Change:€Technical Summary. In:€Climate Change 2007:€The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change [S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, I. B. Averyt, M. Tignor and HL. Miller (eds.)]. Cambridge University Press. Irigoyen, I., Lamsfus, C., Aparicio-Tejo, P. and Muro, J. (2006). The influence of 3,4-dimethylpyrazole phosphate and dicyandiamide on reducing nitrate accumulation in spinach under Mediterranean conditions. Journal of Agricultural Science, 144, 555–562. Jarvis, S., Hutchings, N., Brentrup, F., Olesen, J. and van der Hoek, K. (2011). Nitrogen flows in farming systems across Europe. In:€The

European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J.€W. Erisman et al. Cambridge University Press. Jenkinson, D.S. (2001). The impact of humans on the nitrogen cycle, with focus on temperate arable agriculture. Plant and Soil, 228, 3–15. Johnston, A. E. and Poulton, P. R. (2009). Nitrogen in agriculture:€an overview and definitions of nitrogen use efficiency. In: Proceedings 651, International Fertilizer Society, York, UK. Johnston, A. E., Poulton, P. R. and Coleman, K. (2009). Soil organic matter:€its importance in sustainable agriculture and carbon dioxide fluxes. Advances in Agronomy, 101, 1–57. Jongbloed, A. W. and Lenis, N. P. (1998). Environmental concerns about animal manure. Journal of Animal Science, 76, 2641–2648. Karp, A. and Shield, I. (2008). Bioenergy from plants and the sustainable yield challenge. New Phytologist, 179, 15–32. Kitou, E., Fernandez, R., Gugele, B., Goettlicher, S. and Ritter, M. (2009). Annual European Community Greenhouse Gas Inventory 1990–2007 and Inventory Report 2009. European Commission, DG Environment European Environment Agency. http://www.eea. europa.eu/publications/european-community-greenhouse-gasinventory-2009/. Kohn, R. A., Dou, Z., Ferguson, J. D. and Boston, R. C. (1997). A sensitivity analysis of nitrogen losses from dairy farms. Journal of Environmental Management, 50, 417–428. Lantinga, E. A., Deenen, P. J. A. G. and van Keulen, H. (1999). Herbage and animal production responses to fertilizer nitrogen in perennial ryegrass swards. II. Rotational grazing and cutting. Netherlands Journal of Agricultural Science, 47, 243–261. LEI (2009). Agricultural prices in the Netherlands. Milk- and fertilizer (CAN) prices 1995–2008. Accessed June 2010 at http://www.lei. wur.nl/NL/statistieken/Agrarische+prijzen/. Leip, A., Achermann, B. Biller, G. et al. (2011). Integrating nitrogen fluxes at the European scale. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Lewis, A. J. and Southern, L. L. (2001). Swine Nutrition. CRC Press, Boca Raton, FL. Lopez-Bellido, R. J., Lopez-Bellido, L., Benıtez-Vega, J. and LopezBellido, F. J. (2007). Tillage system, preceding crop, and nitrogen fertilizer in wheat crop:€II. Water utilization. Agronomy Journal, 99, 59–65. Makowski, D., Wallach, D. and Meynard, J. M. (1999). Models of yields, grain protein and residual mineral nitrogen responses to applied nitrogen for winter wheat. Agronomy Journal, 91, 377–385. Marschner, H. (1995). Mineral Nutrition of Higher Plants, Academic Press, London. Maxwell, G. R. (2004). Synthetic Nitrogen Products: A Practical Guide to the Products and Processes. Kluwer Academic/New York. Monroe, C. E. and Hall, C. (2006). Explosives in Metal Mining and Quarry Operations, Merchant Books. Mosier, N., Wyman, C., Dale, B. et al. (2005). Features of promising technologies for pretreatment of lignocellulosic biomass. Bioresource Technology, 96, 673–686. Nadeau, E., Englund, J.-E. and Gustafsson, A. H. (2007). Nitrogen efficiency of dairy cows as affected by diet and milk yield. Livestock Science, 111, 45–56. Nevens, F. and Reheul, D. (2003). Effect of cutting or grazing grass swards on herbage yield, nitrogen uptake and residual soil nitrate at different levels of N fertilisation. Grass and Forage Science, 58, 431–449.

59

Benefits of nitrogen for food, fibre and industrial production Oenema, O., Bleeker, A., Braathen, N. A. et al. (2011). Nitrogen in current European policies. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Offermann, F. and Nieberg, H. (2000). Economic Performance of Organic Farms in Europe. In:€Organic Farming in Europe: Economics and Policy, Volume 5. University of Hohenheim, Department of Farm Economics, Stuttgart, Germany. Peeters, A. and Kopec, S. (1996). Production and productivity of cutting grasslands in temperate climates of Europe. The Proceedings of the 16th General Meeting European Grasslands Federation, pp. 59–73. Persson, P. A., Holmberg, R. and Lee, J. (1993). Rock Blasting and Explosives Engineering. CRC Press, Boca Raton, FL. Pettersson, C. G. and Eckersten, H. (2007). Prediction of grain protein in spring malting barley grown in northern Europe. European Journal of Agronomy, 27, 205–214. Peyraud, J. L. and Delaby, L. (2004). Can we minimize nitrogen excretion in dairy herds while maintaining performances? In:€Animal Production in Europe:€The Way Forward in a Changing World. Volume. 1. International Society for Animal Hygiène, SaintMalo, France. Pilbeam, C. J. (1996). Effect of climate on the recovery in crop and soil of N-15- labelled fertilizer applied to wheat. Fertilizer Research 45, 209–215. Pitre, F. E., Pollet, B., Lafarguette, F. et al. (2007). Effects of increased nitrogen supply on the lignification of poplar wood. Journal of Agriculture and Food Chemistry, 55, 10306–10314. Porteaus, F., Hill, J., Ball, A. S. et al. (2009). Effect of free air carbon dioxide enrichment (FACE) on the chemical composition and nutritive value of wheat grain and straw. Animal Feed Science and Technology, 149, 322–332. Poulsen, H. D. (2009). Danish Standard Values on Nutrient Content in Manure€– 2009. (in Danish). Available at http://www.agrsci. dk/ny_navigation/institutter/institut_for_husdyrsundhed_ velfaerd_og_ernaering/husdyrernaering_og_miljoe/normtal, Aug 2010. Poulsen, H. D., Lund, P., Sehested, J., Hutchings, N. and Sommer, S.€G. (2006). Quantification of nitrogen and phosphorus in manure in the Danish normative system. In:€12th Ramiran International Conference:€Technology for recycling of manure and organic residues in a whole-farm perspective, DIAS report no. 123 from Danish Institute of Agricultural Sciences, vol. II, pp. 105–107. Powlson, D. S., Hart, P. B. S., Pruden, G. and Jenkinson, D. S. (1986a). Recovery of 15N-labelled fertilizer applied in autumn to winter wheat at four sites in eastern England. Journal of Agricultural Science, 107, 611–620. Powlson, D. S., Pruden, G., Johnston, A. E. and Jenkinson,€D.€S. (1986b). The nitrogen cycle in the Broadbalk wheat experiment:€recovery and losses of 15N-labelled fertilizer applied in spring and inputs of nitrogen from the atmosphere. Journal of Agricultural Science, 107, 591–609. Prins, W. H. (1983). Limits to nitrogen fertilizer on grassland. Ph. D. Thesis, Wageningen Agricultural University. Raison, C., Pflimlin, A. and Le Gall, A. (2006). Optimisation of environmental practices in a network of dairyfarms of the Atlantic Area. In: Proceedings of the Final Seminar of the Green Dairy Project Interreg. Atlantic Area III B No.100, Rennes, 13–14 December 2006,€pp.€43–65. Rathke, G. W., Christen, O. and Diepenbrock, W. (2005). Effects of nitrogen source and rate on productivity and quality of winter

60

oilseed rape (Brassica napus L.) grown in different crop rotations. Field Crops Research 94, 103–113. Raun, W. R., Johnson, G. V., Phillips, S. B. and Westerman, R. L. (1998). Effect of long-term N fertilisation on soil organic C and total N in continuous wheat under conventional tillage in Oklahoma. Soil and Tillage Research, 47, 323–330. Redwood, I. I. (2010). Theoretical and Practical Ammonia Refrigeration. General Books LLC. Reynolds, M., Foulkes, M. J., Slafer, G. A. et al. (2009). Raising yield potential in wheat. Journal of Experimental Botany, 60, 1899–1918. Ross, R. B. (1988). Handbook of Metal Treatments and Testing, 2nd edition. Springer, New York. Rudel, T. K., Schneider, L., Uriarte, M. et al. (2009). Agricultural intensification and changes in cultivated areas, 1970–2005. Proceedings of the National Academy of Sciences of the United States of America, 106, 20675–20680. Sastri, V. S. (1998). Corrosion Inihibitors: Principles and Applications. J.€Wiley, New York. Schepers, J. S. and Raun, W. R. (2008). Nitrogen in Agricultural Systems. Chicago, IL. Schils, R. L. M., van der Meer, H. G., Wouters, A. P., Geurink, J. H. and Sikkema, K. (1999). Nitrogen utilization from diluted and undiluted nitric acid treated cattle slurry following surface application to grassland. Nutrient Cycling in Agroecosystems, 53, 269–280. Schröder, J. J., Aarts, H. F. M., ten Berge, H. F. M., van Keulen, H. and Neeteson, J. J. (2003). An evaluation of whole-farm nitrogen balances and related indices for efficient nitrogen use. European Journal of Agronomy 20, 33–44. Schröder, J. J., Jansen, A. G. and Hilhorst, G. J. (2005). Long-term nitrogen supply from cattle slurry. Soil Use and Management, 21, 196–204. Sheriff, G. (2005). Efficient waste? Why farmers over-apply nutrients and the implications for policy design. Review of Agricultural Economics, 27, 542–557. Shewry, P. R. (2009). Improving the protein content and composition of cereal grain. Journal of Cereal Science, 46, 239–250. Shiel, R. S., El Tilib, A. M. A. and Younger, A. (1999). The influence of fertilizer nitrogen, white clover content and environmental factors on the nitrate content of perennial ryegrass and ryegrass/white clover swards. Grass and Forage Science, 54, 275–285. Sidlauskas, G. and Bernotas, S. (2003). Some factors affecting seed yield of spring oilseed rape (Brassica napus L.) Agronomy Research, 1, 229–243. Sieling, K. and Kage, H. (2008). The potential of semi-dwarf oilseed rape genotypes to reduce the risk of N leaching. Journal of Agricultural Science, 146, 77–84. Smil, V. (1999). Nitrogen in crop production:€an account of global flows. Global Biogeochemical Cycles, 13, 647–662. Smil, V. (2000). Feeding the World: A Challenge for the Twenty-First Century. MIT Press, Cambridge, MA. Smit, H. J., Metzger, M. J. and Ewert, F. (2008). Spatial distribution of grassland productivity and land use in Europe. Agricultural Systems, 98, 208–219. Sommer, S. G., Schjoerring, J. K. and Denmead, O. T. (2004). Ammonia emission from mineral fertilizers and fertilized crops. Advances in Agronomy, 82, 558–622. Stoecker, W. (1998) Industrial Refrigeration Handbook, McGraw-Hill, New York. Sutton, M. A., Howard, C. M., Erisman, J. W. et al. (2011). Assessing our nitrogen inheritance. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press.

Lars Stoumann Jensen and Jan K. Schjoerring Sylvester-Bradley, R. and Kindred, D. R. (2009). Analysing nitrogen responses of cereals to prioritize routes to the improvement of nitrogen use efficiency. Journal of Experimental Botany, 60, 1939–1951. Tatiya, R. R. (2005). Surface and Underground Excavations:€Methods, Techniques and Equipment. Taylor and Francis, London. ten Berge, H. F. M. and van Dijk, W. (2009). Management of Nitrogen inputs on farms within the EU regulatory framework. In: Proceedings 654, The International Fertilizer Society, York, UK. Tilman, D., Cassman, K. G., Matson, P. A., Naylor, R. and Polasky,€S. (2002). Agricultural sustainability and intensive production practices. Nature, 418, 671–677. Treacy, M., Humphreys, J., McNamara, K., Browne, R. and Watson,€C.€J. (2008). Farm-gate nitrogen balances on intensive dairy farms in the south west of Ireland. Irish Journal of Agricultural and Food Research, 47, 105–117. Turkington, R. (2009). Chemicals Used for Illegal Purposes: J. Wiley, New York. Turner, D. A., Edis, R. B., Chen, D. et al. (2010). Determination and mitigation of ammonia loss from urea applied to winter wheat with N-(n-butyl) thiophosphorictriamide. Agriculture, Ecosystems and Environment, 137, 261–266. UNFCCC (2008). United Nations Framework Convention on Climate Change:€Greenhouse Gas Inventory Data (online). http://unfccc.int Vellinga, T.V. and André, G. (1999). Sixty years of Dutch nitrogen fertilizer experiments, an overview of the effects of soil type, fertilizer input, management and of developments in time. Netherlands Journal of Agricultural Science, 47, 215–241. Vellinga, T. V., André, G., Schils, R. L. M. and Oenema, O. (2004). Operational nitrogen fertilizer management in dairy farming systems:€identification of criteria and derivation of fertilizer application rates. Grass and Forage Science, 59, 364–377. Vellinga, T. V., André, G., Schils, R. L. M., Kraak, T. and Oenema, O. (2010). Accounting for residual effects of previously applied nitrogen fertilizer on intensively managed grasslands. Grass and Forage Science, 65, 58–75. Velthof, G. L., Oudendag, D., Witzke, H. P. et al. (2009). Integrated assessment of nitrogen losses from agriculture in EU-27 using MITERRA-EUROPE. Journal of Environmental Quality, 38, 402–417. Vérité, R. and Delaby, L. (2000). Relation between nutrition, performances and nitrogen excretion in dairy cows. Annales de Zootechnie, 49, 217–230.

Virtanen, H. and Nousiainen, J. (2005). Nitrogen and phosphorus balances on Finnish dairy farms. Agricultural and Food Science, 14, 166–180. Vitousek, P. M., Naylor, R., Crews, T. et al. (2009). Nutrient imbalances in agricultural development. Science, 324, 1519–1520. von Witzke, H. and Noleppa, S. (2010). EU Agricultural Production and Trade:€Can More Efficiency Prevent Increasing “Land-Grabbing” Outside Of Europe? Report published by OPERA,€ Research Centre of Università Cattolica del Sacro Cuore. Available at http://www.opera-indicators.eu/eng/home. html Wang, X. T. and Below, F. E. (1996). Cytokinins in enhanced growth and tillering of wheat induced by mixed nitrogen source. Crop Science, 36, 121–126. Watson, C. J., Akhonzada, N. A., Hamilton, J. T. G. and Matthews, D. I. (2008). Rate and mode of application of the urease inhibitor N-(n-butyl) thiophosphoric triamide on ammonia volatilization from surface-applied urea. Soil Use and Management, 24, 246–253. Webb, J., Seeney, F. M., and Sylvester-Bradley, R. (1998). The response to fertilizer nitrogen of cereals grown on sandy soils. Journal of Agricultural Science, 130, 271–286. Webb, J., Anthony, S. G., Brown, L. et al. (2005). The impact of increasing the length of the cattle grazing season on emissions of ammonia and nitrous oxide and on nitrate leaching in England and Wales. Agriculture, Ecosystems and Environment, 105, 307–321. Witzke, H. P. and Oenema, O. (2007). Assessment of Most Promising Measures: Integrated Measures in Agriculture to Reduce Ammonia Emissions€– Task 3. Report, Alterra€– Wageningen University Research, The Netherlands. Wohler, F. (2009). Handbook of Mineral Analysis. Bibliobazaar. Wojciechowska, M. and Lomnicki, S. (1999). Nitrogen oxides removal by catalytic methods. Clean Technologies and Environmental Policy, 1, 237–247. Xiccato, G., Schiavon, S., Gallo, L., Bailoni, L. and Bittante, G. (2005). Nitrogen excretion in dairy cow, beef and veal cattle, pig, and rabbit farms in Northern Italy. Italian Journal of Animal Science, 4,€103–111. Yara (2009). Yara Fertilizer Industry Handbook. www.yara.com Zimmer, Y. (2008). Cash Crop Report 2008. vTI Braunschweig. Available at http://www.agribenchmark.org.

61

Chapter

4

Chapter

Nitrogen in current European policies Lead author: Oene Oenema Contributing authors: Albert Bleeker, Nils Axel Braathen, Michaela Budňáková, Keith€Bull, Pavel Čermák, Markus Geupel, Kevin Hicks, Robert Hoft, Natalia Kozlova, Adrian€Leip, Â� Till Spranger, Laura Valli, Gerard Velthof and Wilfried Winiwarter

Executive summary Nature of the problem • Europe, and especially the European Union (EU), has many governmental policy measures aimed at decreasing unwanted reactive nitrogen (Nr) emissions from combustion, agriculture and urban wastes. Many of these policy measures have an ‘effects-based approach’, and focus on single Nr compounds, single sectors and either on air or waters. • This chapter addresses the origin, objectives and targets of EU policy measures related to Nr emissions, considers which instruments are being used to implement the policies and briefly discusses the effects of the policy measures.

Approaches • The chapter starts with a brief description of the basic elements of governmental policy measures. • A review of the main international conventions and EU policies related to emissions of Nr to air and water is then provided. • Finally the chapter provides a semi-quantitative assessment of the effectiveness and efficiency of European policy measures.

Key findings/state of knowledge • International conventions and other treaties have played a key role in raising awareness and establishing policy measures for Nr emissions abatement in EU through so-called Directives and Regulations. • There are many different EU Directives, often addressing individual Nr compounds from individual sectors (e.g. NOx emissions from combustion; NH3 emissions from agriculture, pollution of groundwater and surface water by nitrates from agriculture, discharge of total nitrogen from urban sewage to surface waters). • Many EU Directives have been revised following review and evaluation. There are increasing efforts to cluster single EU Directives into larger Framework Directives. • Compliance with, and effectiveness of, the Directives differs between sectors; it decreases in the order (i) reducing NOx emissions from combustion sources, (ii) reducing nitrogen (and especially Phosphorus) discharges to waters from industries and households, and (iii) reducing NH3 emissions and NO3 leaching from agriculture. • There is not much literature on the differences in the effectiveness and efficiencies of Directives; a number of factors seem to be involved in effectiveness and efficiency, but these have not yet been analysed in a coherent manner.

Major uncertainties/challenges • There is a huge diversity in Nr emission sources and pathways, while the number of policy instruments is limited. There is need to find the optimal mix of policy instruments targeted to the emission sources as well as the stakeholders involved. • It has been indicated that some EU Directives addressing emissions of nitrogen compounds from specific sources have antagonistic effects. The magnitude of these effects is not yet well known. • There is a delay in the environmental and ecological responses following the introduction of Directives; these are due to legislative delays, lack of enforcement and control, constraints in practice and because of biogeochemical hysteresis effects; these effects are not yet well understood quantitatively. • In general, only modest reductions in Nr emissions from agriculture have been achieved to date; this reflects the need for more effective and efficient policy measures and/or greater enforcement of current policies.

Recommendations • To examine further the differences between sectors of the factors that contribute to the effectiveness and efficiency of policy measures for the abatement of Nr emissions. • To explore further the effectiveness and efficiency of more integrated N management and integrated policy measures for the abatement of adverse impacts of Nr emissions. The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

62

Oene Oenema

4.1╇ Introduction This chapter discusses the nature and effects of governmental policies in Europe aimed at decreasing the unwanted emissions of reactive nitrogen (Nr) compounds into the wider environment. Policy is commonly defined as ‘a plan of actions to guide decisions’. Governmental policy is usually a response to unwanted developments or problems in society. Such policy is thus intended to change the developments in a desired direction and/or to solve problems, in this case related to excess Nr in the environment. Governmental policies are based on the premise that humans as individuals and/or as organizations change their behaviour and activities in response to such policies. This premise originates from the fact that humans prefer to live in communities (families, bands, tribes, chiefdoms, states), and that they accept vertical hierarchy (Diamond, 1997; Patterson, 2001). They are expected to follow rules from the top (in this case government) in return for services provided by government. The historian Fernard Braudel (1979, pp. 458–599) insightfully described the development of modern states in Europe and the main tasks of their governments:€(i) to secure obedience, (ii) to exert control over the market, which serves as a mechanism of exchange between the supply and demand of goods and services, and (iii) to strengthen the culture of the society. Evidently, governmental policies are directed to achieving the main tasks of the governments. Key governmental policies usually relate to national defence, food security, economic development, education, health care, spatial planning, infrastructure, traffic, etc. Environmental policy is a relatively new branch of governmental policy, with the theory borrowed initially from economic policy (Tinbergen, 1952). The general aim of environmental policy is to contribute to social welfare by protecting the environment through correcting societal failures, decreasing pollution, halting biodiversity loss and maintaining natural resources. The United Nations Conference on the Human Environment in Stockholm in 1972 is generally seen as having been a key step for increased political awareness in Europe about environmental problems created in part by N (UNEP, 1972), and subsequently for the establishment of environmental policies by governments. One of the main aims of the Conference was to put the issue of acid rain on the international agenda. Nitrogen oxides (NOx) and sulphur dioxide (SO2) are the main contributors to acid rain (Finlayson-Pitts and Pitts, 2000). They are formed during combustion processes and were linked initially to the acidification of Scandinavian lakes and streams. The 1972 Conference ultimately led to the establishment, in 1979, of the UNECE Convention on Long-range Transboundary Air Pollution (CLRTAP) (UNECE, 2010), which has been ratified by most countries in Europe. International treaties and conferences also played major roles in the establishment of water-related environmental policies. The first Convention on the Protection of the Marine Environment of the Baltic Sea was signed in Helsinki in 1974 (HELCOM, 2010). In 1992, a new convention was signed, aimed at protecting the Baltic Sea from all sources of pollution derived from land, shipping and atmospheric deposition

(HELCOM, 2010). The OSPAR Convention on the Protection of the Marine Environment of the North-East Atlantic was also signed in 1992 (OSPAR, 2010). One of the recommendations was the ‘substantial reduction (about 50%) of inputs of N and P into marine areas of the North-East Atlantic where these inputs are likely, directly or indirectly, to cause pollution’, between 1985 and 1995, using N (and P) balances as monitoring tools. The HELCOM and OSPAR Conventions have resulted in various national and EU policies on the protection of groundwater and surface waters, as discussed below. Justification of governmental policy to decrease Nr emissions is mainly based on the significant human health effects and biodiversity losses associated with increased amounts of various reactive N compounds in air, surface waters and groundwaters, and terrestrial ecosystems sensitive to eutrophication and acidification (Erisman et al. 2011, Chapter 2 this volume). Hence, the ultimate objective of governmental policy is ‘to decrease Nr emissions to a level where the value of marginal damages to human health and biodiversity is (approximately) equal to the marginal cost of achieving further reductions’ when considered from a cost–benefit point of view. An alternative formulation is ‘the ultimate objective of policies is to decrease Nr emission to levels that do not give rise to significant negative impacts on, and risks to human health and environment’. However, defining the objective of governmental policy is value-laden and often the subject of fierce political debate (Hajer, 1995; Baker et al., 1997). This debate is further complicated by the complexity of the cause–effect relationships of N compounds emissions and the multi-dimensional outcome of governmental policy, which affects different stakeholders, often with opposite interests, in different ways. This in turn often leads to compromises and delays in the implementation of governmental policy (Bressers and Huitema, 2001; Driessen and Leroy, 2007). The main sources of reactive N compound emissions distinguished by current governmental policy are: (i) combustion (mainly NOx by industry, power plants and€traffic); (ii) waste waters (mainly dissolved and particulate N in discharges by industry and households); and (iii) agriculture (mainly NH3 and N2O to air, NO3 to groundwater and dissolved and particulate N to surface waters). The lack of full understanding of different emission sources, Nr compounds and loss pathways, and of different receptors with different sensitivities to Nr compounds (Hatfield and Follett, 2008) has led to a strong compartmentalization and (regional) differentiation of governmental policies. There are thus policies for specific sectors (energy, industry, households, waste waters and agriculture), N compounds (NOx, NO3, NH3, etc.), regions (countries, sensitive areas, vulnerable zones, etc.), and compartments or receptors (atmosphere, nature conservation areas, forests, groundwater, surface waters, soil, etc.). These complexities in part also reflect the compromises of fierce debates and diverging interests between stakeholders, for example, between industry and nature conservation organizations, and between the Departments of Economic

63

Nitrogen in current European policies Table 4.1 Possible policy instruments, with some examples

Regulatory instruments

Economic instruments

Communicative instruments

–╇ p  ublic land use planning (zoning/ spatial planning) –╇ pollution standards and ceilings –╇ fertilization limits –╇ best available technique requirement

–╇ taxes –╇ subsidies (including price support) –╇ import/export tariffs –╇ tradable emission rights and quotas

–╇ extension services –╇ education and persuasion –╇ co-operative approaches

Development, Traffic and Agriculture on the one hand and the Departments for Environment and Nature Conservation on the other (Driessen and Leroy, 2007). The purpose of this chapter is to provide (i) some concepts of governmental policies, (ii) an overview of governmental policies in Europe (mainly EU) that influence N flows and emissions, and (iii) a preliminary assessment of the various policies, with the aim of identifying interactions between policies and critical success factors.

4.2╇ Concepts of governmental policy Basically, there are four principle drivers in organizing and governing societies, namely: • culture (human values, traditions, fashion and cultural habits); • market power and expertise (the ‘invisible hand’ of the free market); • public policy measures (state coercion, i.e. regulation pressure by governments); and • civic society pressure (pressure from non-governmental organizations (NGOs) and societal pressure and lobby groups). Public or governmental policy is a response to the identification of a societal problem, where culture, markets and civic society pressure collectively fail to solve that problem. Governmental policy aims at modifying human individual behaviour so as to achieve societal (public) objectives, i.e. to contribute to the total welfare of society (Tinbergen, 1952; Baumol and Oates, 1988). The fact that ‘public policy’ addresses societal objectives does not mean that everybody in the society equally accepts this policy and its consequences. There is often a strong divide in societies between those who believe in the cleansing mechanism of the market and in the ability of humans to act responsibly, and who therefore prefer a minimum of governmental policy, and those who emphasize the failures of markets and the need to help the less endowed in society, and therefore favour more extensive governmental policy. Policy instruments are the tools to implement the policy in practice. There are different type of instruments, the choices of which depend on the nature of the problem, the objectives of the policy and the competences and characteristics of the addressees (Baumol and Oates, 1988; Gunningham and Grabosky, 1998). Instruments can be divided into three categories:€ (i) regulatory or command-and-control instruments, (ii) economic or market-based instruments and (iii) communicative or persuasive instruments (Table 4.1).

64

Regulatory instruments (regulation) involve a restriction on the choice of agents, methods and actions. Regulations are compulsory measures imposing requirements on producers to achieve specific levels and standards of environmental quality, including environmental restrictions, bans, permit requirements, maximum rights or minimum obligations. They are the most common policy instrument used in EU environmental policy (e.g. Nitrates Directive). Economic instruments (stimulation) are meant to stimulate preferred production pathways. They are common in agricultural policy, for example, in the EU Common Agricultural Policy (CAP). Environmental taxes and tradable rights/quotas have only been implemented in a few countries. Subsidies are increasingly used as a policy instrument to promote environmentally friendly practices and the introduction of new technology. Communicative instruments (persuasion) include public projects to address environmental issues and measures to improve information flows to promote good practices and environmental objectives. This information can be provided to both producers, in the form of technical assistance and extension, and to consumers, e.g. via labelling. Technical assistance and extension are meant to provide users with information and technical assistance to implement environmentally friendly practices. This category also includes so-called voluntary approaches, e.g. codes of good agricultural practice (Sutton et€al., 2007). Whether those addressed by policy then change their behaviour and contribute to achieving the objectives depends on the instrument and the decision environment of those addressed. A decision environment can be defined as ‘the collection of information, alternatives, values, and preferences available at the time of the decision’. An ideal decision environment would include all possible information, all of it accurate, and every possible alternative at the time. This is usually not the case and explains why the implementation of a policy in practice is far from complete. In short, compliance with a policy will depend on the knowledge and information held by the addressee (‘capability’), the availability of the appropriate tools and means (‘ability’) and on the persuasion (‘willingness’) of the addressee to implement the policy (Figure 4.1). The theoretical and empirical bases of governmental policy measures are still relatively small. This holds also for policy measures related to the abatement of unwanted Nr emissions. The relationships between ‘policy objectives€ – policy instruments€ – change in human behaviour€ – human health, ecological impacts and possible side-effects’ are complex, and to

Oene Oenema

Policy Monitoring

Tool box – Regulation – Stimulation – Persuasion

Instruments

Competences – Capability – Ability – Willingness

Change in humans’ behaviour

Societal objectives

Humans’ objectives

Figure 4.1 Simple representation of the intended working of governmental policy.

Driving forces

International conventions have played major roles in the establishment of governmental policies aimed at decreasing emissions of Nr to, and Nr concentrations in, the environment. Conventions and their protocols relevant to this chapter are summarized in Table 4.2 and further discussed in the supÂ� plementary information (Section 4.4). Intergovernmental organizations (IGOs), whilst not specifically legislative bodies, influence policy internationally (see Table 4.3). They are distinguished from treaties by virtue of their ‘international legal personality’. Further discussion on the inter-relationships of international conventions and IGOs and their interests in N control may be found in Bull et al. (2011, Chapter 25, this volume).

4.4╇ Policy measures affecting nitrogen in European Union

Responses

Pressures

4.3╇ International conventions and intergovernmental organizations

Impact

State

Figure 4.2 The Driving forces€– Pressures€– State€– Impact€– Responses framework (DPSIR) for assessing cause–effect relationships and for developing a policy response (Source:€EEA, 1995.)

some extent based on trial and error. Further, the toolbox for implementing governmental policy measures is relatively small; choices have to be made between regulatory instruments, economic instruments and communicative/voluntary instruments, or a mix of these three. The available theoretical and empirical bases often do not help indicate, a priori, which combination of instruments will be most effective and efficient. The development of the so-called DPSIR framework (see Figure 4.2) and related frameworks by the Organisation for Economic Co-operation and Development (OECD) and the European Environmental Agency (EEA) in the 1990s has improved the understanding of the cause–effect Â�relationships of environmental pollution (see, for example, OECD, 1991; EEA, 1995). It has also provided a framework for responding to environmental problems via policy measures. According to the DPSIR framework, there is a chain of causal links starting with ‘driving forces’ (economic sectors, human activities) through ‘pressures’ (emissions, waste) to ‘states’ (physical, chemical and biological) and ‘impacts’ on ecosystems, human health and functions, eventually leading to political ‘responses’ (policy definition, prioritization, target setting, indicators).

In the following sections, current EU policy measures dealing with N are briefly summarized. Policies related to air and water are discussed first, followed by policies related to agriculture, biofuel and nature conservation. The final section (Section€ 4.4.6) provides a comprehensive overview. To facilitate access to the various EU policies documents, reference is made to the most recent websites (all policies are referenced as EC, 2010a–y). EU environmental policy is mostly established by means of Directives, imposing environmental objectives to be achieved by the Member States. EU Directives fix the framework in which Member States must create national legislation directed to industries/civilians in order to attain the environmental quality objectives laid down in the Directives. In contrast, EU agricultural policy is mostly established through so-called Regulations. These Regulations are directly binding for Member States and, depending on the issue, producers/stakeholders/ industries. Hence, EU Directives provide more flexibility than EU Regulations for Member States’ implementation. Note that EU Directives are commonly based on ‘regulatory instruments’ (Table 4.1) and that EU Regulations are often based on a mixture of ‘economic instruments’ and ‘regulatory instruments’. Understanding EU policy measures dealing with N emissions abatement requires insight into the understanding and perception by scientists and policy makers of the cause–effect relationships of these emissions. Many current policy measures dealing with N emissions reflect a simple ‘source€ – receptor/ effect’ model of understanding. Combustion (mainly NOx by industry, power plants and traffic), waste waters (mainly dissolved and particulate N in discharges by industry and households) and agriculture (diffuse emissions of NH3 and N2O to air and NO3− to waters) are seen as the main N sources, while atmosphere, surface waters and groundwater are seen as the direct receptors. Thus, many policy measures focus on decreasing N compound emissions from specific sources and/or on decreasing N compound concentrations in receiving bodies (receptors) to below critical concentration levels.

65

Nitrogen in current European policies Table 4.2╇ Conventions and protocols addressing nitrogen emissions

1974 Helsinki Convention (HELCOM) on the Protection of the Baltic Sea in Helsinki 1974 OSPAR Convention (PARCOM) on the Protection of the North-East Atlantic 1976 Barcelona Convention on the Protection of the Mediterranean Sea 1979 UNECE Convention on Long-range Transboundary Air Pollution (CLRTAP) 1988 Sofia protocol on Nitrogen oxide (NOx) emissions 1992 Bucharest Convention on the Protection of the Black Sea 1992 Convention on Biological Diversity 1992 Convention on Transboundary Waters and International Lakes 1994 United Nations Framework Convention on Climate Change (UNFCCC) 1997 Kyoto Protocol 1999 Gothenburg Protocol on acidification, eutrophication and ground-level ozone Table 4.3╇ IGOs with linkages to nitrogen

1945 Food and Agriculture Organization (FAO) 1948 World Health Organization (WHO) 1950 World Meteorological Organization (WMO) 1972 United Nations Environment Programme (UNEP) 1988 Intergovernmental Panel on Climate Change (IPCC) 1996 Arctic Council

4.4.1╇ EU policy measures related to atmospheric Nr

Table 4.4 provides an overview of the three main EU Directives on nitrogen in the atmosphere. Following extensive reviews, the 1988 Directive on Large Combustion Plants (LCP; EC, 2010a), the 1996 Directive on Integrated Pollution Prevention and Control (IPPC; EC, 2010b), the 2000 Waste Incineration Directive (WID; EC, 2010c) and the 2005 Directive on Emission from Ignition Engines in Heavy-duty Vehicles (HDV; EC, 2010d), were incorporated into the 2008 Directive on Industrial Emissions concerning Integrated Pollution Prevention and Control (IPPC) (EC, 2010b). This 2008 IPCC Directive is now one of the cornerstones of EU Directives dealing with atmospheric Nr, and sets requirements and standards for NOx emissions from all kinds of combustion sources (Table 4.4). The IPPC Directive employs an integrated approach to the management of all types of pollution from industrial installations, including those for the intensive rearing of poultry or pigs. It requires these installations to have a permit and to minimize all kinds of pollution (including reactive N compounds emissions) by using Best Available Techniques (BAT). An essential part of the IPPC Directive is that the listed activities require a permit to operate, the approval and renewal of which is subject to cross-compliance with other European Community legislation.

66

The second cornerstone of EU Directives dealing with atmospheric Nr is the 2001 National Emission Ceilings Directive (NEC; EC, 2010e). This Directive sets upper limits (ceilings) for each Member State for the total emissions in 2010 and 2020 of the four pollutants responsible for acidification, eutrophication and ground-level ozone pollution (SO2, NOx, VOCs and NH3), but leaves it largely to the Member States to decide which measures to take in order to comply (Table 4.4). The Directive aims at achieving the long-term objectives of not exceeding critical levels and loads by establishing national emission ceilings, taking the years 2010 and 2020 as benchmarks. This Directive is currently (2010) under revision. The 1996 Framework Directive on Ambient Air addresses ambient air quality assessment and management (EC, 2010f). It includes a series of daughter directives, which set the numerical limit values for atmospheric pollutants. For example, the 1999 Air Quality Directive relates to limit values for, among others, nitrogen oxides (NOx), ozone (O3) and particulate matter (PM10) in ambient air. The main emphasis is human health in urban areas and on air pollutants from combustion sources. The most recent version of the Ambient Air Quality Directive was approved in 2008. It contains limit values for NOx, O3 and PM2.5, but not for NH3. Ozone is included as nitrogen oxides (NO and NO2) are important O3 precursor substances, and because of adverse effects of high O3 concentration on human health and crop growth. Particulate matter is included because of its close link to the N cycle (see Hertel et al., 2011; Chapter€9 this volume), being formed as a result of the processing of ammonia, nitrogen oxide and other N-containing substances, and its effects on human health. The 2008 Ambient Air Quality Directive is now one of the three cornerstone Directives dealing with atmospheric Nr in the EU-27 (EC, 2010f).

4.4.2╇ EU policy measures related to N in water€bodies A number of EU policy measures exist which address the issue of Nr emissions and concentrations in water bodies, these are detailed below and summarized in Table 4.5. The 2000 Water Framework Directive (EC, 2010h) embraces all EU legislation for the protection of inland surface waters, transitional waters, coastal waters and groundwater. The Water Framework Directive (WFD) requires all waters to reach ‘good ecological status’ by 2015. It will do this by establishing a river-basin district structure within which demanding environmental objectives will be set, including ecological targets for surface waters and good chemical and quantitative status for groundwater bodies. It requires the implementation of measures from 11 other EU Directives, including the 1976 Bathing Water Directive (EC, 2010i), the 1990 Urban Waste-water Treatment Directive (EC, 2010j), the 1985 Environmental Impact Assessment Directive (EC, 2010k), the 1991 Nitrates Directive (EC, 2010l), the 1996 IPPC Directive (EC 2010b), the 1998 Drinking Water Directive (EC, 2010m) and the 2006 Groundwater Directive (EC, 2010n). The WFD includes an indicative list of main pollutant substances, including substances

Oene Oenema Table 4.4╇ Overview of main EU Directives related to N emissions to, and concentrations in, the atmosphere (see also EC, 2010g)

Directive

Description / objectives

Limit values

2008/50/EC

Ambient air quality: definitions, threshold values, targets and assessment, in relation to sulphur dioxide, nitrogen dioxide, particulate matter, lead, benzene and carbon monoxide.

• Critical level for NOx for vegetation (average over 1 year): 30 μg m−3 • Limit values for NOx for human health (averaged over 1 yr): 40 μg m−3 • Limit values for NOx for human health (averaged over 1 hr):€ 200 μg m−3 • Alert thresholds for NOx for human health (averaged over 3 hr):€ 400 μg m−3 • Target and limit values for PM2.5 in urban areas (average over 3 yr): 20–25 μg m−3.

2008/1/EC

Integrated Pollution, Prevention and Control (IPPC):€to prevent and control emissions from industrial activities into air, water or soil, in relation to polluting substances, including nitrogen

• Installations need a permit • Installations need to comply with environmental quality standards described in other Directives • Installations need to apply best available techniques (BATs)

2001/81/EC

National Emission Ceilings (NEC):€to limit emissions to protect the environment and human health against risks of adverse effects from acidification, eutrophication and groundlevel ozone, by establishing national emission ceilings, taking the years 2010 (and 2020) as benchmarks

• National emission ceilings for SO2, NOx, VOC and NH3, for each country to be attained by 2010, expressed in kilotonnes (Gg) • In regard of the long term objectives ‘not exceeding critical levels and loads and of effective protection of all people against recognized health risks from air pollution’ no ceilings have been yet set for 2020 though the Directive envisages ongoing review

which contribute to eutrophication (in particular, nitrates and phosphates). The WFD allows Member States the flexibility to define specific ambitions, targets and time frames, albeit under the constraints of proper underpinning and justifications. The most important linked Directives of the WFD as regards Nr emissions to groundwater and surface waters are the 1991 Urban Waste Water Directive and the 1991 Nitrates€Directive. The 1991 Urban Waste Water Directive (UWWD; EC, 2010j) concerning urban waste water treatment was adopted in 1991 to protect the water environment from the adverse effects of discharges of urban waste water and from certain industrial discharges. The UWWD has requirements for sewerage (or collection systems) to be established and sets standards for sewage treatment. The general principle of the Directive is to provide treatment of sewage from the largest discharges first, and to protect sensitive waters. It sets secondary treatment as the normal standard, but requires tertiary treatment where discharges affect sensitive areas identified under the Directive. It also requires that discharges from urban waste water treatment plants to sensitive areas do not contain more than 10–15 mg N per litre, depending on the size of the communities, and that the waste water treatment system removes 70%–80% of the initial amount of Nr in the sewage. The main objective of the 1991 Nitrates Directive is ‘to reduce water pollution caused or induced by nitrates from agricultural sources and prevent further such pollution’ (EC, 2010l). This Directive requires Member States to take the following steps:€ (i) water monitoring (with regard to nitrate concentration and trophic status); (ii) identification of waters that are polluted or at risk of pollution; (iii) designation of vulnerable zones (areas that drain into identified waters); (iv) the establishment of codes of good agricultural practices and

action programmes (a set of measures to prevent and reduce nitrate pollution); and (v) the review at least every four years of the designation of vulnerable zones and action programmes. Waters must be identified as polluted or at risk of pollution if nitrate concentrations in groundwater and surface waters contain or could contain more than 50 mg/l per litre if no action is taken, or if surface waters, including freshwater bodies, estuaries, coastal and marine waters are found to be eutrophic or in the near future may become eutrophic if no action is taken. The action programmes must contain mandatory measures relating to:€(i) periods when application of animal manure and fertilizers to land is prohibited; (ii) capacity of and facilities for storage of animal manure; and (iii) limits to the amounts of animal manure (170 kg/ha/yr) and fertilizers applied to land, which should ensure a balanced fertilization. The 2008 Marine Strategy Directive (EC, 2010p) aims to achieve good environmental status of the EU’s marine waters by 2020 and to protect the resource base upon which marinerelated economic and social activities depend. It covers the following marine regions:€ (a) the Baltic Sea; (b) the North-East Atlantic Ocean; (c) the Mediterranean Sea; and (d) the Black Sea. It contains an indicative list of characteristics, pressures and impacts which have to be monitored and assessed regularly, and for which environmental targets have to be set. The list of pressures and impacts includes inputs of fertilizers and other nitrogen- and phosphorus-rich substances (from point and diffuse sources, including agriculture, aquaculture and atmospheric deposition). Each Member State has to draw up a programme of cost-effective measures to address adverse characteristics, pressures and impacts. Impact assessments, including detailed cost–benefit analysis of the measures proposed, are required prior to the introduction of new measures.

67

Nitrogen in current European policies Table 4.5╇ Overview of main EU Directives related to N emissions and concentrations in water bodies (see also EC, 2010o)

Directive

Description / objectives

Requirements/Limit values

2000/60/EC

Water Framework Directive (WFD): to establish a framework for the protection of inland surface waters, transitional waters, coastal waters and groundwater from pollution and depletion

•

Urban Waste Water Treatment Directive (UWWD): to protect the environment from the adverse effects of waste water discharges from urban areas and certain industrial sectors

•

91/271/EEC

91/676/EEC

Nitrates Directive (ND): concerning the protection of waters against pollution caused by nitrates from agricultural sources

• • •

• •

•

• • •

• 2008/56/EC

Marine Strategy Framework Directive:€establishes a framework to take the necessary measures to achieve or maintain good environmental status in the marine environment by the year 2020 at the latest

• •

• •

2006/118/EC

Groundwater Directive: establishes a regime which sets underground water quality standards and introduces measures to prevent or limit inputs of pollutants into groundwater

•

•

•

The 2006 Groundwater Directive (EC, 2010n) complements the Water Framework Directive and requires Member States to:€ (i) establish groundwater quality standards by the end of 2008; (ii) carry out pollution trend studies; (iii) reverse pollution trends so that environmental objectives are achieved by 2015; (iv) operate measures to prevent or limit inputs of pollutants into groundwater; (v) make reviews of technical provisions of the Directive in 2013 and every six years thereafter; (vi) comply with good chemical status criteria (based on EU standards of nitrates and pesticides and on threshold values established by Member States).

68

Maintaining/establishing good ecological status in surface water bodies and good chemical and quantitative status in groundwater bodies Establishment of river basement management plans Designation of â•›‘protected areas’ For ‘limit values’ and ‘measures required’ reference is made to other Directives All agglomerations must be provided with collecting systems for urban waste water Identification of sensitive areas Requirements for discharges from urban waste water treatment plants to sensitive areas:€(i) a reduction of total Nr of 70%–80% of the influent; and (ii) maximum annual mean total N concentrations of 1.5–10 mg/l, depending on size of the urban area Establishment of a code of good agricultural practice, including balanced N fertilization, to be implemented by farmers on a voluntary basis Designation of Nitrate Vulnerable Zones Establishment of action programmes with mandatory measures in vulnerable zones, including N application limits Water quality trigger criteria:€(i) 50 mg nitrate per litre in groundwater and surface waters, and (ii) eutrophic status of surface waters Application limit for nitrogen from animal manure:€170 kg/ha/yr Determination of a set of characteristics for good environmental status Establishment of a comprehensive set of environmental targets for marine waters to guide progress towards achieving good environmental status Identification and implementation of measures needed to achieve or maintain good environmental status There are no prescribed limit values Groundwater quality standards for nitrate and active substances in pesticides, including their relevant metabolites, degradation and reaction products Threshold values for all pollutants and indicators of pollution which characterize groundwater as being at risk of failing to achieve good groundwater chemical status Establishes the 50 mg/l for nitrate as a binding maximum quality threshold

4.4.3╇ EU Common Agricultural Policy and its€reforms. The Common Agricultural Policy (CAP) of the EU was established in 1958 by the EEC. The CAP has contributed greatly to the modernization and productivity of agriculture and to food security in the EU (Ritson and Harvey, 1997). Indirectly, it has also contributed to increased inputs of N in agriculture via N fertilizers and to the import of animal feed from outside the EU, as well as to increased N losses from agriculture to the environment (Romstad et al., 1997). Following the recognition and increased awareness of the effects of surpluses of agricultural products and environmental

Oene Oenema

burden associated with the intensification of agricultural production, the CAP went through a series of reforms, notably in 1984 (implementation of milk quota), 1992 (set-aside regulations), 1997 (agenda 2000) and 2003 (fundamental change in the EU support to agriculture:€EC, 2010q; EC, 2010t; Meester et al., 2005; Blandford and Hill, 2006). In 2003, it was agreed that the CAP has two pillars:€(i) market policies and (ii) rural development policies. In 2008, agreement was reached to further modernize, simplify and streamline the CAP and remove restrictions on farming (the so-called ‘Health Check’). This agreement includes the phasing-out of the milk quota system, the abolition of set-aside regulations and a further shift from direct aid for production support to the Rural Development Programme (EC, 2010q; EC, 2010s.). The reforms of the CAP continue to have a significant influence on N use and its loss to the environment. ‘Cross-compliance’ is a main policy vehicle to implement the CAP reform. In this context, cross-compliance is the requirement that farmers in receipt of payments under the CAP are also shown to be meeting other relevant European Community legislation. In June 2003, cross-compliance became an obligatory element of the first pillar of CAP, thereby coupling existing environmental policies and other policies to agricultural income support, as implemented in the so-called ‘Single Farm Payments’ to farmers. There are two major aspects of cross-compliance in the Single Farm Payment (EC, 2010q):€(i) Compliance with 19 Statutory Management Requirements (SMRs) covering the environment, food safety, animal and plant health and animal welfare, including the provisions of the relevant directives; and (ii) Compliance with a requirement to maintain land in Good Agricultural and Environmental Condition (GAEC). Definitions of GAEC are specified at the national or regional level and address soil organic matter, soil erosion, maintenance of the land(scape) and avoidance of the deterioration of natural habitats. A few of the SMRs directly or indirectly address N inputs and N emissions in agriculture. These include, for example, the 1991 Nitrates Directive, the 1986 Sewage Sludge Directive, the 1992 Directive on the conservation of natural habitats and of wild flora and fauna (Habitats Directive), and the 1979 Directive on the conservation of wild birds (Birds Directive). Such cross-compliance with other environmental regulations has the potential to encourage the reduction of Nr losses from agriculture. However, this is not always the case. For example, emerging requirements for animal housing to meet new animal welfare standards (EC, 2010r) will in many cases contribute to increased emissions of NH3 and N2O. This interaction highlights the need to consider environmental regulation in the context of other societal pressures. The second pillar of the CAP is the Rural Development Policy, which for the period 2007 to 2013 is set out in Council Regulation No. 1698/2005 (EC, 2010t). Under this regulation, rural development policy is focused on three themes (known as ‘thematic axes’) plus the LEADER approach:€ (i) improving the competitiveness of the agricultural and forestry sector; (ii) improving the environment and the countryside;

(iii)€ improving the quality of life in rural areas and encouraging diversification of the rural economy; and (iv) mainstreaming the LEADER approach ‘Links between Activities Developing the Rural Economy’ (LEADER, ‘Liaison Entre Actions de Développement de l’Economie Rurale’). To help ensure a balanced approach to the rural development policy, Member States and regions are obliged to spread their rural development funding between all these thematic axes. Within each of the first three axes, various support mechanisms have been described in articles 20 to 35 for Axis 1, in articles 36 to 51 for axis 2 and in articles 52 to 59 for axis 3, which help with improving the agronomic and environmental performances of agricultural activities in the rural areas. These measures may include the setting up of advisory services, supporting modernization of agricultural holdings, supporting operations related to access to farm and forest land, land consolidation and improvement, energy supply and water management, and agri-environmental payments. Clearly, the Rural Development Policy can contribute to measures that decrease Nr losses from agriculture to the environment.

4.4.4╇ EU nature conservation policies The policy framework for preventing biodiversity loss in the EU is provided by the Birds and Habitats Directives, which are being implemented through Natura 2000, an EU-wide network of protected areas, which now covers some 18% of the territory of the EU. The 1979 Birds Directive (EC, 2010v) requires Member States to designate Special Protection Areas (SPAs) for endangered bird species. Currently, over 4000 SPAs have been designated, covering 8% of EU territory. The 1992 Habitats Directive (EC, 2010w) aims to protect other wildlife species and habitats. Each Member State is required to identify Special Areas of Conservation (SACs) and to put in place a special management plan to protect them. The SPAs and SACs together make up the Natura 2000 network. Member States are required to improve the ecological coherence of Natura 2000 by maintaining, and where appropriate developing, features of the landscape which are of major importance for wild fauna and flora. The Birds and Habitats Directives imply restrictions on human activities within and around the Natura 2000 areas. Widely established restrictions include infrastructural, industrial and agricultural activities in and near to Natura 2000 sites. The Directives also have implications for activities taking place that are not on the site itself. In addition, the Birds and Habitats Directives establish lists of designated species and habitats, with a commitment to monitoring the performance of these across the whole of the EU. This represents an important part of the overall objective of these Directives, though it should be noted that there is a lack of measures to protect such habitats and species outside of the Natura 2000 network. In principle, the Birds and Habitats Directives are drivers to safeguard biodiversity and to lower NH3 and NOx emissions, by virtue of the precautionary approach. However, this is still an area of ongoing development in learning to implement the

69

Nitrogen in current European policies

N input limits

SOURCES

Wastes

Combustion

Agriculture

Figure 4.3 Schematic overview of the N control mechanisms of European policy measures. For the N emission sources, there are N input limits and N compounds emission limits, for the N receptors, there are N compounds concentrations limits and N compounds exposure limits, including critical loads.

N emission limits

Air

N concentration limits

Water

RECEPTORS

N exposure limits Humans

Flora

Fauna

existing legislation, and in evaluating its limitations (COST 729, 2009).

4.4.5╇ EU bio-energy policy Current EU energy policy focuses on increasing the security of energy supply and reducing greenhouse gas (GHG) emissions, as set out in 2007 in ‘The Renewable Energy Road Map’ (COM(2006)848; EC, 2010u). In the Road Map, a mandatory target was set for achieving a 20% share of renewables in energy consumption in the EU by 2020 and a mandatory minimum target of 10% of all energy in transport from biofuels. The recent Directive on the promotion of the use of energy from renewable sources (2009/28/EC; EC, 2010y) amended the 2003 Biofuel Directive (2003/30/EC). Though ‘nitrogen’ is not mentioned explicitly in any of the energy policy documents (except for N2O as a greenhouse gas), the EU policy on bioenergy will have influence on N use in agriculture, as bio-energy crops require Nr for their growth and release various N compounds to the broader environment during and following their growth and utilization. The EU policy on bioenergy will also have influence on the total agricultural area used for the production of food, feed and fibres.

4.4.6╇ Summary of nitrogen control by European€policies In summary, N flows and emissions in Europe are regulated by a broad variety of policy measures. These policy measures regulate N flows and emissions via (i) input control (e.g. N application limits in agriculture), (ii) emission control (e.g. Nr emission limits, discharge limits), (iii) concentration limits for Nr in air and water bodies, and (iv) Nr exposure limits and critical N loads (Figure 4.3). Input controls exist only for agriculture, via application limits for Nr from animal manure and fertilizers to agricultural land, and via provisions for the protein content of animal feed. Such limits do not apply for combustion and wastes. Emission controls exist for all major N compounds, for example via the national emission ceilings for NOx and NH3, NOx emission limits for stationary and mobile combustion

70

sources, discharge limits for industry and sewage treatment plants. Further, NH3 emissions abatement measures exists for animal housing, manure storages and manure application, and N fertilizer application to land. Table 4.6 provides a summary of quantitative EU limit values for various N compounds in air and water. In air, there are limit values for NOx (NO and NO2) and for substances that are formed in part through the presence of NOx in air, viz., ozone (O3) and fine particles (PM2.5 and PM10). Currently, there are no limit values for NH3 concentrations in air. In water, there are limit values for NO3−, NO2−, NH4+, and Ntotal. There are no limit values for N compounds in soil. Exposure limits for humans and N-sensitive flora and fauna are defined either via concentration limits or via input limits, such as critical loads. A critical load is defined by the CLRTAP (UNECE, 1999) and the NEC Directive (EC, 2010e) as ‘a quantitative estimate of an exposure to one or more pollutants below which significant harmful effects on specified sensitive elements of the environment do not occur, according to present knowledge’. Critical N-loads for ecosystems are determined following specific methodologies and criteria for mapping critical levels/ loads and geographical areas (ICP Modelling and Mapping, 2004). Critical loads form the basis for setting emission limits and ceilings (Sliggers and Kakebeeke, 2004).

4.5╇ Assessment of environmental policies in€Europe Assessments of environmental policy usually include analyses of its compliance, expressed in terms of implementation of mandatory obligations, its effectiveness, expressed in terms of achieving policy objectives, and its efficiency, expressed in terms of the economic costs of its implementation. In addition, assessments may address the possible technical, technological, socio-economic, institutional and societal changes brought about by environmental policy. Assessment of compliance is usually the first step; it simply records whether the obligations of the policy (e.g. abatement measures, designation of specific areas, and monitoring and reporting obligations) have been satisfied. However, the effectiveness and efficiency of planned environmental policy may be

Oene Oenema Table 4.6╇ Summary of limit values for N compounds concentrations in air and water as set by EU policies

Effects

Indicators

Limit values / targets

Regulatory reference

Respiratory diseases of humans

NOx

40 μg m (annual mean) 200 μg m−3 (hourly mean) 400 μg m−3 (threshold, 3 hrs)

Ambient Air Quality Directive (2008/50/EC)

Ibid

PM2.5

20–25 μg m−3 (average of 3 yrs)

Ibid

Ibid

PM10

40–50 μg m (average of 3 yrs)

Ibid

Ibid

O3

180–240 μg m (hourly mean)

Ibid

Ibid

AOT40

120 μg m (hourly mean)

Ibid

Plant damage

NOx

30 μg m−3 (annual mean)

Ibid

Adverse effects on humans from nitrates

NO3−

50 mg/l in groundwater

Drinking Water Directive (98/83/EC) Nitrates Directive (91/676/EEC) Groundwater Directive (2006/118/EC)

Adverse effects on humans by nitrites

NO2−

0.5 mg/l in water used for drinking water. Further, [NO3−]/50 + [NO2−]/3 ≤ 1

Drinking Water Directive (98/83/EC)

Adverse effects on humans from ammonium

NH4+

Indicator value:€0.5 mg/l

Drinking Water Directive (98/83/EC)

Eutrophication of surface waters

NO3−

25–50 mg/l

Nitrates Directive (91/676/EEC)

Contamination of groundwater

NO3−

50 mg/l

Nitrates Directive (91/676/EEC) Goundwater Directive (2006/118/EC)

Eutrophication of surface waters

Ntotal

2–10 mg/l; discharge from sewage treatment plants

Urban Waste Water Directive (91/271/ EEC)

a

−3

−3

−3

a

−3

AOT40 stands for accumulated exposure over a threshold ozone concentration of 40 ppb.

assessed in advance (ex-ante) through simulation modelling and stakeholder consultation. Such ex-ante assessments provide a view of the effectiveness and efficiency of environmental policy prior to implementation (often assuming 100% compliance), and are instrumental for achieving political agreement for ratification and implementation. By contrast, retrospective (ex-post) assessments are usually based on analyses of data obtained through various monitoring programmes, censuses, inquiries and reviews. Assessments of environmental policy are sometimes heavily debated and also criticized. First, there are differences of views about the appropriateness of the objectives and targets that must be achieved, e.g. emission targets, concentration targets or ecological targets. Figure 4.4 shows that there is a large ‘separation’ between emissions targets and the human health and ecological impacts targets, there are also many possible interactions. Second, there is debate about the accuracy of, and uncertainties in, the data and the cause–effect relationships. For example, the NOx and NH3 emission estimates in Europe are thought to have an uncertainty range of 30% and 50%, respectively (EEA, 2005). Third, there are often discussions about the economic cost–benefit analyses and the effects on the competitiveness of sectors. Experiences over the past 20 years indicate that environmental policies in Europe have contributed to decreasing Nr losses to air, surface waters and groundwaters in Europe, but

that critical loads are still exceeded and that the environmental and ecological status of many groundwater bodies, surface waters and natural areas are still below the set targets (Erisman et al., 2011, Chapter 2 this volume). Many of these set targets reflect ecological targets and political compromises; few targets (if any) have been set at levels ‘where the value of marginal damages to human health and biodiversity is (approximately) equal to the marginal cost of achieving further reductions’, which would yield most societal benefit (see also Brink et al., 2011, Chapter 22 this volume). By contrast, there has been a tendency to go in one of two directions:€either to specify environmental targets based on their technical and political achievability or to set objectives for avoidance of adverse impacts. Many European environmental policies are based on regulatory instruments, with frequent use of BAT requirements and emission standards, and these appear to have a relatively low economic efficiency (OECD, 2007). EU environmental Directives leave little room for the use of more flexible economic instruments (e.g. taxes or trading systems for NOx emissions, taxes or trading systems for N-input to the agricultural sector). Economic instruments are not necessarily prohibited, but the Directives limit the flexibility these instruments could have offered (OECD, 2007). So far, policy measures aimed at decreasing Nr species emissions have achieved larger responses from combustion sources than from urban sources or from agricultural sources especially.

71

Nitrogen in current European policies

Emissions targets

Concentrations targets

Ecological process rate targets

Human health & ecological impacts targets

Acidification [NOX] in air Eutrophication NH3

[NH3] in air

N2O

[N2O] in air

Ground-level ozone (O3) formation

NO3

[NO3] in water bodies

Fine-particle matter (PMX) formation

NH4

[NH4] in drinking water

[Nt] surface waters Corrosion Other substances

Biodiversity loss

Food production

Toxification Radiation interception

Nt

Human health

Global Change

NOX

Climate change

Material loss

Interactions

Figure 4.4 Illustration of the major links between the multiple forms of reactive nitrogen emission and the resulting impacts on different concentration, process and impact targets.

This can be shown by the fact that relative emission reductions have been achieved in the following order:€ NOx emissions > Ntotal emissions from urban areas > NO3− leaching from agriculture > NH3 emissions from agriculture (see Erisman et al., 2011, Chapter 2, Figure 2.5). Emission reductions may also follow from changes in economic activities. For example, significant reductions in total NOx emissions to air in the EU-15 between 1990 and 2006 may be considered, to a large extent, influenced by environmental policies. By contrast, the decreases in total NOx emissions to air in the 12 new Member States of the EU (EU-12) between 1990 and 2006 mainly follow from the changes in the political and economic systems after 1989 (EEA, 2005; 2008), rather than the implementation of specific environmental policies. It can be seen from Figure 4.5 that the reductions in NOx emissions have been less successful than SO2 emission reductions, which is largely due to increased vehicle mileage offsetting the benefits of low NOx emission technologies (NEGTAP, 2001). By comparison, there has been only a small effect of environmental policies in reducing ammonia emissions. Figure 4.5 shows that NH3 emissions for the EU-15 only decreased by 10% between 1990 and 2006, while the larger decrease of 49% for the EU-12 was the result of

72

the political and economic changes following 1989, rather than due to specific environmental policies in the period. Within the EU-15, the differences in the effects of policies are also large. The extent to which, for example, the objectives of the policies to reduce ammonia emissions and nitrate leaching from agriculture have been achieved is variable across Member States (EEA, 2008). These variable results are ascribed to: • differences between Member States in their perceptions of EU Directives; • differences in economic sectors and systems and environmental conditions; • legislative delays and implementation delays; • economic costs of the measures and lack of enforcement; • continued economic growth, which has ‘neutralized’ some of the improvements in ‘eco-efficiency’ at the system level (e.g. increased car fleets offsetting projected NOx reductions from low emission vehicle technology); • ineffectiveness of some measures; • antagonisms between some of the measures; and • hysteresis effects, due to buffering reactions within the systems.

Oene Oenema EU-15 SO2

Emissions (Gg)

15000

NOX

NH3

12000 9000 6000 3000

NH3 emissions, relative to 1990 (%)

150

18000

Netherlands

EU-15

100

50

0

1990

1995

2000

2001

2002

2003

2004

2005

2006

Year

0 1995

2000

2001

2002

2003

2004

2005

2006

Year

EU-12

Latvia

Slovenia

EU-12 new

100

18000 SO2

15000

150

NH3 emissions, relative to 1990 (%)

1990

Emissions (Gg)

Spain

NOX

NH3

12000 9000 6000

50

0

1990

1995

2000

2001

2002 Year

2003

2004

2005

2006

Figure 4.6 Relative changes in total NH3 emissions to air in the EU-15, The Netherlands and Spain (top) and in the EU-12 (new), Latvia and Slovenia (�bottom) between 1990 and 2006 (source:€EEA, 2008).

3000 0 1990

1995

2000

2001

2002

2003

2004

2005

2006

Year

Figure 4.5. Changes in total NOX and NH3 emissions to air in EU-15 (top) and EU-12 (bottom) between 1990 and 2006. Emissions of SO2 to air are shown for comparison (source:€EEA, 2008).

4.5.1╇ Changes in NOx emissions from combustion

Combustion is a major source of NOx emissions and the basis for emissions abatement policy in Europe have been the 1988 Sofia Protocol on NOx emissions (Table 4.2) and the related EU Directives (Table 4.2). The transport and energy sectors are the main sources (Erisman et al., 2011, Chapter 2 this volume; EEA, 2008) and emissions of NOx in the EU-27 have decreased on average by about 31% between 1990 and 2005. Basically, reductions have occurred in all economic sectors and most countries have reported lower emissions of NOx in 2005 compared to 1990. The exceptions to this are Austria (7% increase), Cyprus (19%), Greece (6%), Portugal (13%) and Spain (26%). The three sectors ‘responsible’ for the vast majority of the decreased NOx emissions are road transport (contributing 53% of the total reduction in NOx emissions), energy industry (contributing 29%) and industry (energy) (contributing 15%). The significant reduction in NOx emissions from road transport (38% between 1990 and 2005) has been achieved despite the general increase in activity within this sector (EEA, 2008). Emissions of NOx have also declined in the energy industry (38% between 1990 and 2005), despite again an increase in activity (EEA, 2008). The decoupling of NOx emissions, transport and electricity and heat production has been due to (EEA, 2007; EMEP, 2007): • the introduction of catalytic converters in car engines; • the introduction of low-NOx combustion technology and flue gas treatment, which led to a 49% reduction;

• efficiency improvements, which resulted in a 14% reduction; • the switch in the fuel mix, away from coal and fuel oil towards natural gas, which led to an 8% reduction; • the lower share of nuclear and non-thermal renewable energy (i.e. excluding biomass) in 2004 compared to 1990, which actually increased emissions by 3%.

4.5.2╇ Changes in N losses from agriculture Agriculture in Europe contributes, on average, to about 80%–90% of the total emissions of NH3 into the atmosphere, to roughly 40%–60% of the Nr to surface waters, and to about 50%–70% of the emissions of N2O to the atmosphere (EEA, 2005; Oenema et al., 2007, 2009). Most of the NH3 originates from animal manure in stables, from manure storage systems and from the application of animal manure to agricultural land. Between 1990 and 2006, emissions of NH3 decreased by 12% in the EU-15 and by 47% in the EU-12 (Figure 4.5). In the EU-15, abatement policy and decreases in NH3 emissions were the greatest in the Netherlands and the least in Spain (Figure€4.6). While the Netherlands is estimated to have had a 50% reduction in NH3 emissions between 1990 and 2006, NH3 emissions in Spain increased by 25% due to an expansion of the animal livestock sector. For the new Member States (EU-12), the contraction of the livestock herd and the decreased use of mineral fertilizer after 1989 resulting in decreases in NH3 emissions were greatest in Latvia (~70%) and least in Slovenia (~20%). Decreases in NH3 emissions in Hungary following the political and economic changes have been described by Horvath and Sutton (1998). There are a number of countries that report a decreasing trend of mean NO3− concentrations in shallow groundwaters following the implementation of the EU Nitrates Directive.

73

Nitrogen in current European policies

74

However, the decreases are modest and a significant number of monitoring stations show increasing NO3− concentrations (EC, 2007, 2010). Similarly, while 55% of the monitoring stations in surface waters in rural areas of the EU-15 had a decreasing trend in NO3− concentrations during the period 1996–2003, 31% of monitoring stations had stable NO3− concentrations and 14% of the stations showed increasing NO3− concentrations (EC, 2007, 2010x). Changes in NO3− concentrations have been related to changes in Nr surpluses. Surpluses of N of the soil surface balance in EU countries have on average decreased since 1990, in part in response to structural changes in agriculture following changes in the common agricultural policy, in part also in response to environmental policies, such as the Nitrates Directive. In the EU-15, mean Nr surplus decreased from 65 kg per ha in 1990 to 50 kg per ha in 2000 (EEA, 2005a). Surpluses (range 150–250kg per ha) and decreases in surpluses (range 30–50 kg per ha) were largest for the Netherlands, Belgium and€Germany. The variable and slow responses of Member States to environmental policies in agriculture have been ascribed to (Romstad et al., 1997; Smith et al., 2007; MNP, 2007; Oenema et al., 2009; Mikkelsen et al., 2010): (i) the large differences in farming systems and environmental conditions in the EU-27 combined with the complexity of the N cycle; (ii) a variable interpretation by Member States of the targets and measures in environmental directives and regulations; (iii) hesitation in implementing measures, due to the perceived high costs to farmers and perceived low effectiveness; (iv) hesitation in introducing mechanisms to monitor compliance by farmers, due to the perceived high costs; (v) legislative delays; (vi) failure by farmers to implement measures, due to withinsystem constraints, perceived and actual costs, and the time needed for learning; and (vii) potential antagonisms between measures aimed at decreasing NH3 emissions and those aimed at decreasing NO3 leaching.

ammonia concentrations (Bleeker et al., 2009) and the envirÂ� onmental and ecological status of lakes, rivers and streams in rural areas have improved little yet (Stälnacke et al., 2004; Mourad et al., 2006).

Moreover, the recovery of the environmental and Â�ecological status of lakes, rivers and streams often takes more time than expected from the measures implemented and associated decrease of emissions. The same point has been made for atmospheric Nr compounds, including the question of why atmospheric ammonia levels did not decrease as fast as expected following implementation of emission reduction policies in Western Europe (Bleeker et al., 2009). Both of these findings point to the complexity of the systems and our limited understanding of the biogeochemical connectivity of systems and controls. There are ‘hysteresis’ effects and feedback mechanisms that are often overlooked and that lead to slow responses. This seems also to be the case for the new Member States in central Europe where fertilizer N inputs and livestock numbers decreased drastically following the political changes in the early 1990s, while the atmospheric

The treatment of urban waste water has also contributed to significant decreases in the Nr load to coastal waters and to the improvement of surface water quality in Europe in general. However, there are large spatial and temporal variations, and some contribution may have come from lower emissions from agriculture due to the implementation of the Nitrates Directive (EEA, 2005b). The 2005 OSPAR Assessment of Riverine Inputs (all sources) and Direct Discharges (urban waste water) for the period 1990–2002 noted significant decreases in total inputs of both N (up 32%) and P (up 135%) to the Arctic Waters and a significant reduction in total inputs of N (down 12%) in the Greater North Sea (OSPAR, 2005). Similarly, a downward trend in total riverine and direct point-source inputs of N and P has been observed for the Baltic Sea during the period 1994–2006, but again with large spatial and temporal variations (HELCOM, 2009). However, the overall policy target of a 50% reduction in

4.5.3╇ Changes in N losses from urban waste€waters The Urban Waste Water Treatment Directive (91/271/EEC; EC, 2010j) regulates discharges of municipal waste water from towns and larger villages and specifies which kind of treatment must be installed. The Directive requires that all European agglomerations (settlements) with a size of more than 2000 population equivalents (p.e.) are equipped with collecting and treatment systems for their waste waters. The basic level of treatment is so-called secondary treatment (i.e. removal of organic pollution). In sensitive areas (68% of the EU-27 territory), a more stringent treatment is required, for example, removal of a minimum of 75% of the N and P loads. Most EU Member States have designated their whole territory as a sensitive area, but some (e.g. United Kingdom, Spain, Hungary) have designated only a small area as sensitive (EC, 2009b). By the end of 2005, waste water collecting systems were in place for 93% of the total polluting load (in 83% of the agglomerations) (EEA, 2005b). Secondary treatment was in place for 87% of the load and was reported to work adequately for 78% of it. More stringent treatment was in place for 72% of the load and was reported to work adequately for 65% of it. The European Commission has concluded that considerable progress has been achieved in implementing the Directive, but that key challenges remain to align waste water treatment over the entire EU with the provisions of the Directive and the ‘good status’ environmental objective under the Water Framework Directive (EC, 2009). In particular, the secondary treatment and the more stringent treatment need to be improved, especially in the new Member States (EC, 2009b).

4.5.4╇ Changes in N pollution of marine waters€by€50%

Decrease in N and P losses (%)

Oene Oenema 100

have not decreased to the same extent. As a result, agriculture increasingly becomes a relatively large contributor to the loading of surface waters with N and P (EEA, 2005b).

N P

80 60

4.6╇ Assessment of factors crucial for effective nitrogen emission abatement

40 20 0 BE

DE

DK

NL

NO

SE

CH

Country Figure 4.7 Percentage reductions in anthropogenic discharges/losses of nitrogen and phosphorous to ‘eutrophication problem areas’ around the Eastern North Atlantic (source of data:€OSPAR, 2008b).

N and P input into marine surface waters (see Section 4.4.2) has not yet been achieved. The 2008 OSPAR Eutrophication Assessment (OSPAR, 2008a) shows that eutrophication is still a problem in many coastal areas of the Greater North Sea. The 2008 Report on the Implementation of PARCOM Recommendations 88/2 and 89/4 (OSPAR, 2008b) concludes that Contracting Parties contributing to N and P inputs to eutrophication problem areas have mostly achieved the 50% reduction target for discharges and losses of phosphorus (P), but not for N (see Figure 4.7). Modelling studies suggest that nutrient input reductions beyond the 50% target will be needed in some areas to eliminate all eutrophication problems (OSPAR, 2008a). Agriculture is the biggest contributor to discharges and losses of N to eutrophication problem areas (OSPAR, 2008b). Combustion in power plants and traffic (including road traffic and increasing emissions from maritime shipping in the North Sea and the Atlantic) are the main contributors to airborne NOx inputs to the OSPAR maritime area (OSPAR, 2005), while agriculture is the main contributor to atmospheric deposition of reduced nitrogen (mainly NH3). Eutrophication by N and P is also a major problem in the Baltic Sea (HELCOM, 2005, 2009). Total loads entering the Baltic Sea (as riverine and direct point-source discharges) amounted to 891 Gg N and 51 Gg P in 1990, and it was agreed to decrease these inputs by 50% by 1995 (HELCOM, 2010). In the Baltic Sea Action Plan, the maximum allowable nutrient input targets were set at 41% of the 1990 load of P and approximately 68% of that of N. Both targets have not yet been achieved; by 2006 the reduction for P was 45% and for N only 30%. Eutrophication by N and P inputs is less of a problem in the Mediterranean than in the North Sea and the Baltic Sea. In fact, the Mediterranean is one of the most oligotrophic regional seas in the world (Karydis and Chatzichristofas, 2003). Eutrophication is limited to coastal zones, especially in the western and northern half of the Mediterranean. However, N and P inputs to the Mediterranean marine environment have increased steadily over the past 20 years (UNEP, 2009). Summarizing, EU policy to treat municipal and industrial waste waters have been effective in decreasing N (and especially P) loadings to surface waters, though further improvements are needed (EC, 2009b). Diffuse N and P losses from agriculture

4.6.1╇ Differences between sectors So far, the most successful Nr emissions abatement policies have been on (see Section 4.5 and Erisman et al. 2011, Chapter 2 this volume):€(i) reducing NOx emissions to air from power plants and stationary combustion sources through catalytic converters, (ii) reducing emissions of NOx from mobile combustion sources to air (catalytic converters for gasoline cars, combustion optimization and NOx destruction by Selective Catalytic Reduction (SCR) with urea for diesel cars), and (iii) reducing N (and especially P) discharges to surface waters from industrial sources and households through sewage treatment plants. Though less spectacular than the decreases in SO2 emissions to air (see Figure 4.5), emission reductions for NOx to air and for Nr from human sewage to surface waters are larger than the emission reductions achieved for NH3 and NO3− from agriculture. The question is therefore:€‘why are certain policies more effective than others?’. So far, there has been little cross-sector comparison on the effectiveness and efficiency of policy measures aimed at decreasing Nr emissions. The success of the emissions abatement policies for NOx from combustion and Nr from human sewage may be ascribed to one or a combination of the following factors: (i) use of economic instruments (subsidies and taxes) to facilitate the implementation of the policy, which results in a high degree of compliance; (ii) availability of relatively straightforward and effective technologies to reduce the emissions effectively with few major side-effects; (iii) the limited number of addressees who must take action to implement the measures; (iv) the scale of investments required and the degree to which these are shared; (v) the cost of the compliance measures are relatively small and/or can be transferred to others; and (vi) enforcement and control, leading to a high degree of compliance with the policy measures. Theory and practice suggest that economic instruments or a mix of economic and regulatory and persuasive instruments tend to be more effective for the implementation of policy than a single regulatory or persuasive instrument (Gunningham and Grabosky, 1998; OECD, 2007). Subsidies, premiums and taxes often provide a strong incentive to adopt the provisions. Compliance with the obligations of a policy requires that all relevant stakeholders are informed and have the necessary knowledge, tools and will to implement the provisions. Subsidies on cars with catalytic converters to decrease NOx emissions, and

75

Nitrogen in current European policies

EU financial support for building sewage treatment plants, are indeed effective instruments for implementation of these emissions abatement technologies (OECD, 2007). The larger the number of addressees (stakeholders) of the policy, the larger the transaction costs of the policy and the less resource the government can allocate to supporting individual addressees. While cars with catalytic converters are driven by numerous drivers, few of these drivers know about the details of converter operation, as these are implemented by the car industry, which encompasses only few stakeholders. Similarly, while all humans in Europe produce Nr-containing wastes, few of them are involved in sewage collection and treatment. By contrast, all individual farmers in the EU (the percentage of farmers to the total work force ranges from 2% to 25% between the Member States) have to comply with the measures of the Nitrates Directive (especially those in Nitrate Vulnerable Zones) and other EU directives relevant to agriculture (see Section 4.4). The scale of investments in hardware and software needed to comply with policy obligations may differ greatly. Collection and treatment of urban sewage waters requires huge investments, but is done for a multitude of arguments, of which Nr emission abatement is only one, and the costs of the investments are transferred to and shared by numerous tax payers. Catalytic converters do not require much investment by the car industry (relative to other investments), although research costs may be significant. By contrast, building low-NH3-emission housing systems, manure storage systems and manure application techniques can require relatively large investments by individual farmers, though the Rural Development Programme may provide funds for subsidizing infrastructural modernizations (see Section 4.4.3). In the case of high-investment activities, such as new animal housing systems, much of the cost may be associated with other requirements, such as new animal welfare standards. For other techniques, such as low emission manure application, additional costs may be largely offset by saving more nitrogen in the system, thereby reducing fertilizer requirements (Webb et al., 2010). Compliance with the Nitrates Directive requires in principle relatively little investment, apart from the obligation of sufficient manure storage. However, the application limit of 170 kg N per ha per year can be a serious constraint to intensive livestock farms; they may have to export animal manure elsewhere (with or without prior processing) or will have to decrease livestock density. The costs of the catalytic converters or sewage treatment plants are all transferred to consumers (or tax payers), and therefore can be implemented easily by the car industry and communities, respectively. By contrast, farmers represent in many cases small businesses which have themselves to bear the cost of the measures for abating NH3 emissions and N leaching; they can less easily pass on costs to those further down the food production chain. For example, in a globalizing market for agricultural products, farmers in the EU may lose competitive power relative to farmers with less stringent environmental policies, unless other safeguards are put in place (such as the Rural Development Programme). There are nevertheless precedents for requiring investment in agriculture to meet policy requirements, such as animal welfare legislation. Such

76

environmental and welfare requirements come with associated costs which must, in the end, be born by governments and/ or consumers, or will have to be covered by increased income through up-scaling (larger farms). Summarizing, the relatively variable and slow implementation of environmental EU policy and measures in agriculture to decrease Nr emissions may be ascribed to: (1) ongoing incentives to maintain agricultural production levels and the limited ability of farmers to transfer the costs of environmental protection to consumers; (2) huge differences in farming systems and environmental conditions in the EU-27 and the complexities that arise when making the requirements of existing EU Directives farm-specific; (3) delays by Member States to implement measures in agriculture, fuelled by strong farm lobby groups, due to the perceived costs to farmers and the perceived low effectiveness; (4) delays by Member States to introduce effective control mechanisms to monitor compliance by all farmers, due to the difficulties in setting up such control systems as well as the perceived cost to a Member State; (5) failure by farmers to implement measures, due to within system constraints, perceived costs and the time needed for learning; and (6) the possibility for, and fear of, antagonisms between measures, due to lack of integration of measures aimed at abating NO3 leaching and measures aimed at abating NH3 and N2O emissions. Table 4.7 summarizes the results of a qualitative assessment of factors influencing the abatement of Nr emissions from different sectors. Various factors are different for agriculture compared to combustion and urban wastes, although it is unclear how much each of these contributes to differences in implementation of, and compliance with, the policies. Evidently, further studies are needed.

4.6.2╇ Differences between regions and EU Member States There are differences in the ways EU Member States and their regional governments implement environmental policies. These may relate to differences in the political need and political will, but also to differences in culture, environmental conditions, economic developments, institutional organization and in the availability of competent policy officers at regional and local levels. Such differences may change over time, for example, as a result of elections and changes in the political orientation of governments. Developments of civic society and pressure groups may also exert influence on the compliance to environmental policy (see Section 4.2). For example, farmers’ lobby groups were strong in delaying the implementation of the Nitrates Directive in the Netherlands during the 1990s, while green lobby groups greatly contributed to increasing the political pressure by the European

Oene Oenema Table 4.7╇ Qualitative assessment of factors that affect the implementation of EU policies to decrease Nr emissions:€NOx emissions from combustion, NH3 emissions and NO3 leaching from agriculture, and Ntot discharges from urban wastes

Factors

Combustion NOx to air

Policy instruments

Agriculture

Urban wastes

NH3 to air

NO3 to waters

Ntot to waters

Mixed

Regulation

Regulation

Mixed

Number of stakeholders

Few

Many

Many

Few

Technology level

Advanced

Modest

Modest

High

Level of standardization in production

High

Low

Low

High

Number of techniques involved

Few

Many

Many

Few

Development costs

High

High

High

High

Implementation costs

Modest

Modest for animal feeding and manure application; high for animal housings and manure storages

Low for optimizing fertilizer applications; high for adjusting farming systems

High

Who bears costs?

Manufacturers, but transferred effectively to consumers

Farmer

Farmer + public sector (RDP)

Water companies, but effectively transferred to consumers

Management activities & knowledge involved

Essentially no activities required by car drivers

Many activities, requires both proper techniques and information and knowledge

Many activities, requires information and knowledge

Many activities, requires both proper techniques and information and knowledge

Influence of climate & soil conditions

Absent

Large

Large

Negligible

Potential side-effects (apart from costs)

Increased N2O and NH3 emissions

Increased N2O emissions and energy use; fertilizer savings

Yield loss; fertilizer saving; increased / decreased NH3 emissions

Increased N2O emissions and energy use

Commission on the Netherland’s government to fully implement the Nitrates Directive (Bavel et€ al., 2004). Within the context of the Nitrates Directive, changes in legislation are often under pressure of infringement procedures launched by the European Commission, indicating that enforcement of legislation is a key point. Scandinavian countries seem to have made most effort to comply with environmental policy. The effects of air pollution were already felt in the Scandinavian lakes and forests in the 1960s and 1970s, because these were highly sensitive to acidification and eutrophication. Though the origin of the air pollution largely came from outside Scandinavia, societal awareness of the effects led to the organization of the 1972 United Nations Stockholm Conference and to the foundation of CLRTAP (as discussed in Sections 4.1 and 4.3). These impacts also contributed to political will in Scandinavia to protect the environment from their own pollution sources. Western Europe has a high density of industrial and agricultural activities, with high emission densities. It has stakes in both continuation of economic activities and protection of the environment, and hence in the need to decrease the emission densities of economic activities. Southern Europe, in many locations, has a lower emission density than Western

Europe and an environment less sensitive to acidification than Scandinavia. Also, economic development and water harvesting are a societal priority in southern Europe. Finally, the 12 new Member States in central Europe had centralized political and economic systems until the early 1990s, with relatively low political priority for protecting the environment. These countries are now catching up following their accession to the EU in 2004 or 2007.

4.7╇ Conclusions • Environmental policy is a relatively new subject that emerged in the 1970s and 1980s. International agreements have given a strong impetus to the establishment of policy measures related to Nr emissions. The theoretical and empirical bases of policy measures related to Nr emissions are still small. • The toolbox for environmental policy instruments comprises regulatory instruments, economic instruments and communicative/voluntary instruments. Initially, there was a strong focus and emphasis on regulatory instruments; now there is increasing evidence that each environmental policy must have a specific mix of instruments, depending

77

Nitrogen in current European policies

•

•

•

•

•

78

on the capability, ability and willingness of the addressee to implement the environmental policy effectively and efficiently. Policy measures aimed at decreasing Nr emissions in the EU are effects-based or target-based, i.e. the policy measures aim to prevent well-defined human health effects or ecological effects or aim to meet specific threshold/target/ limit values. The policy measures aimed at decreasing Nr emissions in the EU have been implemented through Directives, which have to be addressed by all Member States through national legislation, and to a lesser extent Regulations, which have to implemented directly by all Member States. There is a large number of Directives, many of which have been revised following review and evaluation. There is also an increasing trend towards clustering specific Directives within Framework Directives. The Common Agricultural Policy (CAP) of the EU has a large influence on EU agriculture and indirectly also on Nr use and Nr emissions. Through a series of reforms of the CAP, there is increasing integration of agricultural, environmental and rural development objectives in agriculture, but the number of Directives and Regulations remains large. The EU Directives aimed at decreasing Nr emissions from the various sources have been developed and implemented while our understanding of the functioning of N in the biosphere, atmosphere and hydrosphere are still limited and evolving. Policies have been developed initially for single Nr compounds (NO3−, NH3, N2O and NOx), for single sectors (households, industries, traffic, crop production, animal production), for single environmental compartments (air, water, nature, humans), and for various specific impacts (e.g. human health, food security, climate change, eutrophication, acidification, biodiversity loss); this is partly because of our limited understanding of the complex N cycle, and partly because of the departmentalization of governments, These multicompound, multi-sector, multi-receptor, multi-impact approaches have contributed to a ‘wealth’ of policies, with some having interactive effects (both synergistic and antagonistic). As a result, there is an increasing quest for integrating environmental policy measures. Most successful Nr emission abatement policy measures, in terms of abatement of Nr emissions, have been on (i) reducing NOx emissions to air from power plants, stationary combustion sources and transport through catalytic converters, and (ii) reducing N (and especially P) to surface waters from industrial sources and households through sewage treatment plants. The success of these emission abatement policy measures has been ascribed to the availability of relatively straightforward technologies to reduce emissions, the limited number of addressees, the use of mixes of instruments and the level of governmental enforcement and control. However, there is not much literature on the comparison between

•

•

•

•

Directives or between sectors of the effectiveness and efficiencies of the various Directives related to Nr emissions abatement. Less successful, so far, have been policies on reducing Nr emissions from agriculture. In principle, the technologies and measures to reduce these emissions are available, but there are various reasons to explain why these have not been adopted and/or have not been effective. One of these reasons is the diversity and complexity of the farming systems involved and the complex, diffuse Nr pathways, which have resulted in many different regulatory obligations, but which are not equally effective for all farms. Further studies are needed to find out the optimal mix of packages of measures and incentives to decrease the diffuse Nr losses to air, soil and water. Based in part on the successful reduction of SO2 and NOx emissions from the energy, industry and transport sectors through technological measures, there is some belief that technology will reduce all unwanted emissions from all sectors. However, management and (changes in) economic activities may be equally important factors. So far, Scandinavian countries have done most on the implementation of environmental measures for nitrogen, perhaps because they felt the effects of air pollution on surface waters and forests most intensively. Current EU Directives on agriculture consider the threats from NO3 leaching, NH3 emissions (and N2O emissions) separately. However, when not combined with an integrated approach to N management, the policy measures may have the risk of antagonistic effects.

Acknowledgements This chapter was prepared with the support of the NinE Programme of the European Science Foundation, the NitroEurope IP (funded by the European Commission) and the COST Action 729. The authors are also especially grateful to Andrea Weiss for providing information on OSPAR, to Mark Sutton for many critical comments and suggestions and to two anonymous reviewers for helpful comments.

Supplementary materials Supplementary materials (as referenced in the chapter) are� available online through both Cambridge University Press:€ www.cambridge.org/ena and the Nitrogen in Europe website: www.nine-esf.org/ena.

References Baker, S., Kousis, M., Richardson, D. and Young, S. (eds.) (1997). The Politics of Sustainable Development. Routledge, London. Baumol, W. J. and Oates, W. E. (1988). The Theory of Environmental Policy. Cambridge University Press, Cambridge. Bavel, M. van, Frouws, J. and Driessen, P. (2004). The Netherlands and the Nitrates Directive. Ostrich or strategist? Wageningen University, Wageningen.

Oene Oenema Blandford D. and Hill, B. (eds.) (2006). Policy Reform and Adjustment in the Agricultural Sectors of Developed Countries. CAB International, Wallingford, UK. Bleeker, A., Sutton, M. A., Acherman, B. et al. (2009). Linking ammonia emission trends to measured concentrations and deposition of reduced nitrogen at different scales. In:€Atmospheric Ammonia, eds. M. Sutton, S. Reis and S. M. H. Baker, Springer, New York, pp. 123–180. Braudel, F. (1979). Civilization and Capitalism 15th–18th Century, Volume II. The Wheels of Commerce. William Collins, London. Bressers, H. T. A. and Huitema, D. (2001). What the doctor should know; politicians are special patients. The impact of the policy making process on the design of economic instruments. In:€Market-Based Instruments for Environmental Management: Politics and Institutions, ed. M. S. Andersen and R.-U. Sprenger. Edward Elgar, Cheltenham, UK. Brink, C., van Grinsven, H., Jacobsen, B. H. et al. (2011). Costs and benefits of nitrogen in the environment. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Bull, K., Hoft, R. and Sutton, M. (2011). Co-ordinating European nitrogen policies between directives and international conventions. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. CONSSO (2002). Progress Report, Fifth International Conference on the Protection of the North Sea 20–21 March 2002. Committee of North Sea Senior Officials (CONSSO). Bergen, Norway. http:// www.regjeringen.no/md/html/nsc/progressreport2002/Progress_ Report.pdf. COST 729 (2009). Nitrogen Deposition and Natura 2000:€Science€and Practice in Determining Environmental Impacts. Summary for Policy Makers. http://cost729.ceh.ac.uk/webfm_send/18 (accessed 4 January 2009). Diamond, J. (1997). Guns, Germs, and Steel:€The Fates of Human Societies. W.W. Norton, New York. Driessen, P. and Leroy, P. (2007). Environmental Policy. Analysis and Perspectives. Coutinho, Bussum, NL. EC (2007). Report from the Commission to the Council and the European Parliament. On implementation of Council Directive 91/676/EEC concerning the protection of waters against pollution caused by nitrates from agricultural sources for the period 2000– 2003 {SEC(2007)339} COM (2007) 120 final. EC (2009). Fifth Commission Summary on the Implementation of the Urban Waste Water Treatment Directive Brussels, Commission Staff Working Document SEC (2009) 1114 final, 23€pp. http://ec.europa.eu/environment/water/water-urbanwaste/ implementation/implementationreports_en.htm EC (2010a). Directive on Large Combustion Plants (LCP; 2001/80/ EC) http://ec.europa.eu/environment/air/pollutants/stationary/ lcp.htm EC (2010b). Directive on Integrated Pollution Prevention and Control (IPPC; 1996/61/EC) http://ec.europa.eu/environment/air/ pollutants/stationary/ippc/index.htm EC (2010c). Directive on Waste Incineration (WID; 2000/76/EC) http:// ec.europa.eu/environment/air/pollutants/stationary/wid.htm EC (2010d). Directive on Emission from Ignition Engines in Heavy-duty Vehicles (HDV; 2005/55/EC) http://ec.europa.eu/ enterprise/sectors/automotive/documents/directives/directive2005–55-ec_en.htm EC (2010e). Directive on National Emission Ceilings (NEC, 2001/81/ EC) http://ec.europa.eu/environment/air/pollutants/ceilings.htm

EC (2010f). Framework Directive on Ambient Air (96/62/EC / 2008/50/EC) http://ec.europa.eu/environment/air/quality/ legislation/existing_leg.htm EC (2010g). Existing legislation related to air quality in the EU. http:// ec.europa.eu/environment/air/legis.htm EC (2010h). Water Framework Directive (WFD; 2000/60/EC) http://ec.europa.eu/environment/water/water-framework/ index_en.html EC (2010i). Bathing Water Directive (BWD; 2006/7/EC) http:// ec.europa.eu/environment/water/water-bathing/index_en.html EC (2010j). Urban Waste-water Treatment Directive (91/271/EEC) http://ec.europa.eu/environment/water/water-urbanwaste/ index_en.html EC (2010k). Environmental Impact Assessment Directive (85/337/ EEC) http://ec.europa.eu/environment/eia/full-legal-text/85337. htm EC (2010l). Council Directive 91/676/EEC concerning the protection of waters against pollution caused by nitrates from agricultural sources. http://ec.europa.eu/environment/water/water-nitrates/ index_en.html EC (2010m). Drinking Water Directive (98/83/EC) http://ec.europa. eu/environment/water/water-drink/index_en.html EC (2010n). Groundwater Directive (2006/118/EC) http://ec.europa. eu/environment/water/water-framework/groundwater.html EC (2010o). Water in EU legislation. http://ec.europa.eu/ environment/water/index_en.htm EC (2010p). A Marine Strategy Directive to save Europe’s seas and oceans (2008/56/EC) http://ec.europa.eu/environment/water/ marine/index_en.htm EC (2010q). Agriculture; meeting the needs of farmers and consumers. http://europa.eu/pol/agr/index_en.htm EC (2010r). Animal Welfare legislative references. http://ec.europa.eu/ food/animal/welfare/references_en.htm EC (2010s). CAP Health Check. http://ec.europa.eu/agriculture/ healthcheck/index_en.htm EC (2010t). Rural Development legislations. http://ec.europa.eu/ agriculture/rur/leg/index_en.htm EC (2010u). EU Energy Policy. http://ec.europa.eu/agriculture/ bioenergy/index_en.htm EC (2010v). Birds Directive (79/409/EEC) http://ec.europa.eu/ environment/nature/legislation/birdsdirective/index_en.htm EC (2010w). Habitats Directive (92/43/EEC) http://ec.europa.eu/ environment/nature/legislation/habitatsdirective/index_en.htm EC (2010x). Report from the Commission to the Council and the European Parliament. On implementation of Council Directive 91/676/EEC concerning the protection of waters against pollution caused by nitrates from agricultural sources based on Member State reports for the period 2004–2007 COM(2010)47 final. http://ec.europa.eu/environment/water/water-nitrates/pdf/ com_2010_47.pdf EC (2010y). Directive 2009/28/EC of the European Parliament and of the Council of 23 April 2009 on the promotion of the use of energy from renewable sources and amending and subsequently repealing Directives 2001/77/EC and 2003/30/EC. http://europa.eu/ legislation_summaries/energy/renewable_energy/en0009_en.htm EEA (1995). Europe’s Environment:€the Dobris Assessment. European Environment Agency, Copenhagen. EEA (2005a). The European Environment, State and Outlook 2005. European Environment Agency, Copenhagen.

79

Nitrogen in current European policies EEA (2005b). Source Apportionment of Nitrogen and Phosphorus Inputs into the Aquatic Environment. European Environmental Agency, Copenhagen. EEA (2007). EN09 Emissions (CO2, SO2 and NOx) from public electricity and heat production€– explanatory indicators. (http:// ims.eionet.europa.eu/Sectors_and_activities/energy/indicators/ EN09%2C2007.04) EEA (2008). Annual European Community LRTAP Convention Emission Inventory Report 1990–2006, Submission to EMEP through the Executive Secretary of the UNECE. European Environmental Agency, Copenhagen. EMEP (2007). Transboundary Acidification, Eutrophication and Ground Level Ozone in Europe in 2005. Norwegian Meteorological Institute, Oslo. Erisman, J. W., van Grinsven, H., Grizzetti, B. et al. (2011). The European nitrogen problem in a global perspective. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J.€W. Erisman et al. Cambridge University Press. Finlayson-Pitts, B. J. and Pitts, J. N. (2000). Chemistry of the Upper and Lower Atmosphere:€Theory, Experiments and Applications. Academic Press, London. Gunningham, N. and Grabosky, P. (1998). SMART Regulation: Designing Environmental Policy. Clarendon Press, Oxford. Hajer, M. (1995). The Politics of Environmental Discourse: Ecological Modernization and the Policy Process. Oxford University Press. Hatfield, J. L. and Follett, R. F. (eds.) (2008). Nitrogen in the Environment:€Sources, Problems and Management, 2nd Edition. Elsevier, Amsterdam. HELCOM (2005). Nutrient Pollution to the Baltic Sea in 2000. Baltic Sea Environment Proceedings No. 100. Helsinki Commission, Baltic Marine Environment Protection Commission, Helsinki. http://www.helcom.fi/publications/bsep/en_GB/ bseplist/ HELCOM (2009). Eutrophication in the Baltic Sea, an integrated thematic assessment of the effects of nutrient enrichment in the Baltic Sea region. Baltic Sea Environment Proceedings No. 115. Helsinki Commission, Baltic Marine Environment Protection Commission, Helsinki, 152 pp. http://www.helcom.fi/publications/ bsep/en_GB/bseplist/ HELCOM (2010). Helsinki Commission/Baltic Marine. Environment Protection Commission. http://www.helcom.fi/ (accessed 15–02–2010). Hertel, O., Reis, S., Ambelas Skjøth, C. et al. (2011). Nitrogen processes in the atmosphere. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Horvath L. and Sutton M. A. (1998). Long term record of ammonia and ammonium concentrations at K-puszta, Hungary. Atmospheric Environment, 32, 339–344. ICP Modelling and Mapping (2004). Manual on Methodologies and Criteria for Modelling and Mapping Critical Loads and Levels and Air Pollution Effects, Risks and Trends. (Available at:€http:// icpmapping.org/cms/zeigeBereich/5/manual-und-downloads. html. Accessed 4 January 2009) Karydis, M. and Chatzichristofas, F. (2003). Coastal eutrophication in the Mediterranean marine environment:€a short review. In: Proceedings 8th International Conference on Environmental Science and Technology, Lemnos Island, Greece, 8–10 September 2003, pp.€426–430. Meester, G., Oskam, A. and Silvis, H. (eds.) (2005). EU Policy for Agriculture, Food and Green, From Policy to Practice. Wageningen Academic Publishers, Wageningen (in Dutch).

80

Mikkelsen, S. A., Iversen, T. M., Jacobsen, B. H. and Kjær, S. S. (2010). Denmark–European Union:€reducing nutrient losses from intensive livestock operations, in:€Livestock in a Changing Landscape. Volume 2. Experiences and Regional Perspectives, eds. P.€Gerber, H. Mooney, J. Dijkman, S. Tarawali and C. de Haan. Island Press, Washington, pp 140–153. MNP (2007). Efficacy of the Manure Policy in the Past (1998–2005), Present (2006–2007) and Future (2008–2015), Report 500124001, Environmental Assessment Agency MNP, Bilthoven, The Netherlands (in Dutch). Mourad, D. S. J., van der Perk, M. and Piirimae, K. (2006). Changes in nutrient emissions, fluxes and retention in a North-Eastern European lowland drainage basin. Environmental Monitoring and Assessment, 120, 415–448. NEGTAP (National Expert Group on Transboundary Air Pollution) (2001). Transboundary Air Pollution:€Acidification, Eutrophication and Ground-Level Ozone in the UK. NEGTAP, London. OECD (1991). Environmental Indicators: A Preliminary set. OECD, Paris. OECD (2007). Instrument Mixes for Environmental Policy. OECD, Paris. Oenema, O., Oudendag, D. A., Witzke, H. P. et al. (2007). Integrated Measures in Agriculture to reduce Ammonia Emissions. Alterra, Wageningen, The Netherlands. Oenema, O., Witzke, H. P., Klimont, Z., Lesschen, J. P. and Velthof, G.€L. (2009). Integrated assessment of promising measures to decrease nitrogen losses from agriculture in EU-27. Agriculture, Ecosystems€and Environment, 133, 280–288. OSPAR (2005). 2005 Assessment of Data Collected Under the OSPAR Comprehensive Study on Riverine Inputs and Direct Discharges for the Period 1990–2002. Publication No.€2005/233. OSPAR Commission. OSPAR (2008a). Second OSPAR Integrated Report on the Eutrophication Status of the OSPAR Maritime Area, OSPAR Publication No. 372/2008. OSPAR Commission. OSPAR (2008b). Nutrients in the OSPAR Convention area: Assessment of the Implementation of PARCOM Recommendations 88/2 and 89/4. OSPAR publication No. 344/2008, OSPAR Commission. OSPAR (2010). Protecting and Conserving the North-East Atlantic and its Resources. http://www.ospar.org (accessed 15 February 2010). Patterson, M. (2001). Understanding Global Environmental Politics. Domination, Accumulation, Resistance. Palgrave, Basingstoke, UK. Ritson, C. and Harvey, D. (eds.) (1997). The Common Agricultural Policy. CAB International, Wallingford, UK. Romstad, E., Simonsen, J. and Vatn, A. (1997). Controlling Mineral Emissions in European Agriculture: Economics, Policies and the Environment. CAB International, Wallingford, UK. Sliggers, J. and Kakebeeke, W. (2004). Clearing the Air: 25 Years of the Convention on Long-range Transboundary Air Pollution. United Nations, New York. Smith, R. J., Glenn, G. A., Parkinson, R. and Richards, J. P. (2007). Evaluating the implementation of the Nitrates Directive in Denmark and England using an actor-orientated approach. European Environment, 17, 124–144. Stälnacke, P., Vandsemb, S. M., Vassiljev, A., Grimvall, A. and Jolankai, G. (2004). Changes in nutrient levels in some Eastern European rivers in response to large-scale changes in agricultural production. Water Science and Technology, 49, 29–36. Sutton, M. A., Erisman, J. W. and Oenema. O. (2007). Strategies for controlling nitrogen emissions from agriculture:€regulatory, voluntary and economic approaches. In:€Fertilizer

Oene Oenema Best Management Practices: General Principles, Strategies for their Adoption and Voluntary Initiatives vs Regulations. International Fertilizer Industry Association, Paris, France, pp.€245–259. Tinbergen, J. (1952). On the Theory of Economic Policy. NorthHolland, Amsterdam. UNECE (1999). Protocol to the 1979 Convention on Long-Range Transboundary Air Pollution to Abate Acidification, Eutrophication and Ground-Level Ozone. http://www.unece.org/env/lrtap/ multi_h1.htm UNECE (2010). Convention on Long-Range Transboundary Air Pollution. http://www.unece.org/env/lrtap

UNEP (1972). Report of the United Nations Conference of the Human Environment. United Nations Environment Programme, Stockholm, 1972. UNEP (2009). State of the Environment and Development in the Mediterranean, United Nations Environment Programme/ Mediterranean Action Plan (UNEP/MAP- Plan Bleu), Athens, Greece. Webb, J., Pain, B., Bittman, S. and Morgan, J. (2010). The impacts of manure application methods on emissions of ammonia, nitrous oxide and on crop response:€a review. Agriculture, Ecosystems and Environment, 137, 39–46.

81

Chapter

5

Chapter

The challenge to integrate nitrogen science and policies:€the European Nitrogen Assessment approach Lead author: Mark A. Sutton Contributing authors: Clare M. Howard, Jan Willem Erisman, William J. Bealey, Gilles Billen, Albert Bleeker, Alexander F. Bouwman, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti

Executive summary Nature of the problem • Anthropogenic releases of reactive nitrogen (Nr) can disturb natural systems and affect human health and welfare in many different ways. Scientific and policy views of the nitrogen cycle have typically addressed these problems from separate perspectives, looking in each case at only part of the overall issue. • Given the multi-faceted nature of the nitrogen cycle, it is a major challenge to develop a more-integrated understanding of how different areas of nitrogen science and policies fit together.

Approaches • Observations from the first part of the European Nitrogen Assessment (ENA Part I) are summarized, considering the distinctive character of Nr in Europe, the benefits and threats, and the current policies. Approaches to developing the following parts of the Assessment are discussed with an emphasis on how to draw out the key issues.

Key findings • Recognizing the multi-pollutant, multi-phase complexity of the nitrogen cycle, it is concluded that it is essential to focus on a limited set of priority issues to allow effective communication between nitrogen scientists and policy makers. • A pathway is developed for prioritization of the key environmental concerns of excess Nr. Starting with around twenty environmental effects, the list is reduced down, first to nine main concerns, and then to five key societal threats. • The five key threats of excess Nr in Europe are identified as:€Water quality, Air quality, Greenhouse gas balance, Ecosystems and biodiversity, and Soil quality. These headings€– which are easily remembered by the acronym ‘WAGES’€– provide a basis for summarizing societal concern about excess Nr in later chapters of the Assessment. • The selection of five key threats represents a conscious balancing of complexity and simplification. The division also lends itself to developing communication models, as illustrated by its analogy to the classical cosmology of Empedocles and Aristotle.

Major challenges • Ongoing efforts must focus on linking scientific communities between nitrogen form (N2O, NH3, NOx, NO3−, etc.), environmental compartment (air, water, plant, soil, etc.), and spatial scale (farm, landscape, region, continent, etc.), tracing the Nr cascade from the main source sectors, especially agriculture and combustion. • For policy makers, major infrastructural constraints limit the connection of different threats and media (climate, air pollution, water pollution, etc.) through spatial scales (local, regional, global policies). Political positions often require a deliberate separation between issues, making it harder to negotiate joined-up approaches. • Ongoing efforts are needed to simplify further the nitrogen story. Multiple communication models should be used, matching the needs of different audiences.

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

82

Mark A. Sutton

Recommendations • Ongoing research at the scale of nitrogen processes (ENA Part II) is essential as this provides the foundation for developing mechanistic understanding. • Additional efforts are needed to quantify nitrogen flows through different spatial scales (ENA Part III). A new emphasis on rural and urban landscapes provides an important link between the plot and regional scales, while nitrogen budgeting should be further developed as an integration tool identifying the main Nr pathways. • The key societal threats of excess Nr (ENA Part IV) are not equally well quantified. While understanding of the water and air threat is rather mature, the threat of Nr on greenhouse gas balance is still at the stage of early quantification. Further efforts are needed to quantify fully the impacts of all the threats. • The first multi-pollutant approaches to Nr cost–benefit analysis, future scenarios and integrated Nr management (ENA Part V) should be further developed to underpin the development of integrated abatement strategies. • Efforts at communicating Nr to policy makers should highlight how integrated Nr management can help meet multiple environmental targets. For the general public, efforts should emphasize the responsibility we all have to manage our own nitrogen footprint.

5.1╇ Introduction As introduced in Chapter€1, the European Nitrogen Assessment (ENA) has been structured in five parts, each of which deals with a different stage of the ENA process (Sutton et€al., 2011, Chapter 1 this volume). The previous chapters in Part I provide the background upon which we here develop the vision and approach for the rest of the Assessment. It is thus clear from the continental context explained in Chapter€2, that, compared with many other parts of the world, Europe is characterized by an excess of reactive nitrogen (Nr) leading to many environmental problems (Erisman et€al., 2011, Chapter 2 this volume). This is true equally of the Nr which is deliberately produced for food production, as for the Nr that is produced inadvertently through high temperature combustion processes. This concern about excess Nr provides a central theme running through the Assessment, which may be contrasted with some other parts of the world (such as parts of Africa and South America) where there is still a societal shortage of Nr. Based on the analysis of Chapter€3 (Jensen et€al., 2011, this volume), it is equally evident that there are huge societal benefits associated with fixing atmospheric N2 into Nr. The focus of Jensen et€ al. is on adding up the benefits of deliberate N2 fixation in fertilizer production and biological nitrogen fixation, which are together essential for healthy functioning of the European economy. If the most obvious benefit is the supply of Nr-containing fertilizers to agriculture, it is equally obvious that modern society would not be possible without the concomitant production of Nr explosives (essential for all mining activity in the world), many plastics (such as nylon) and a huge diversity of other Nr-containing chemicals. By contrast, we must exclude formation of Nr due to the combustion of fossil fuels from the list of benefits, since the Nr produced is immediately lost to the environment, while control efforts (such as catalytic converters) focus on its chemical denitrification back to N2, rather than capture and use of the Nr produced. In economic terms, it is obvious that a better management of nitrogen has substantial benefits. The aim to improve nitrogen use efficiency means that less Nr fertilizers are needed, potentially saving costs and the energy-use associated with their production (Jensen et€ al., 2011). However, it is equally clear that Nr as a commodity is currently rather cheap, its price

being largely set by the fuel costs associated with its production in the Haber–Bosch process. As a result, no policies appear to have been needed in recent decades to ensure adequate supply of Nr in Europe. Against this, it is clear from the analysis of Oenema et€al. (2011a, Chapter€4 this volume) that there has been a plethora of different government policies related to the release of different Nr forms into the environment. As Nr is emitted, its conversion into many different chemical forms and accumulation of stocks in each of air, land and water, makes it obvious that several environmental effects should be expected. One of the clear conclusions of Oenema et€al. (2011a) is that policies designed to address these environmental effects have in most cases been conducted separately according to Nr form and environmental media in which the pollution form occurs (e.g., freshwater pollution, urban air pollution, marine pollution). There is thus a lack of a joined-up view of how Nr is being lost into the environment and how holistic approaches may be implemented to manage the nitrogen cycle. Based on Chapter€4, it appears that one of the most problematic sectors for managing Nr threats on the environment is agriculture. Oenema et€ al. (2011a) explain how policies to reduce atmospheric emissions from large combustion sources, such as power stations, and from vehicles have been relatively successful. By focusing on a few key actors (e.g., large power generating companies, vehicle manufacturers), with a clear ability to transfer any costs to consumers, it has been possible to achieve a high take-up of technical measures. The challenges in those sectors have been the offsetting of reductions, achieved through technical mitigation measures, by increased consumption patterns (e.g., increased vehicle miles per person), as well as by some chemical trade-offs (e.g., catalytic converters increasing N2O and NH3 emissions). By comparison, the challenges to control Nr loss from agriculture have been rather harder. In this sector, Nr emissions are generated by many independent actors (individual farmers), operating a high diversity of processes, in a rather open system (i.e., open to air, soil and water). The sector, especially with the smaller operators, is frequently characterized by conservatism, reflected in a caution to take up new technical approaches, particularly if there is uncertainty on how any perceived costs might be transferred to consumers.

83

The challenge to integrate nitrogen science and policies

in piecemeal fashion, and that is far too complex to be easily explained to the general public. In this chapter, we reflect on these challenges to develop the basis for subsequent Parts II–V of the European Nitrogen Assessment. We outline a vision for integration both across nitrogen science domains and across policy domains, as well as the links between the two. In particular, we focus on how to communicate the issues in a way that balances complexity with an easily understandable list of priority concerns.

5.2╇ Integrating nitrogen science 5.2.1╇ Linking nitrogen forms, processes and scales

Figure€5.1 Temporal changes in annual reactive nitrogen (Nr) inputs to Europe (EU-27) and the emissions to the environment. Top panel:€crop biological nitrogen fixation, N in imported feed and food, land application of mineral fertilizers (mainly from N fixation by the Haber–Bosch process) and N fixed in industry and traffic combustion processes. Bottom panel:€leaching and run-off to water, especially as nitrate, emission to air as ammonia (NH3) mainly from agriculture, emission to air as nitrogen oxides (NOx) mainly from industry and traffic. Inferred from Leip et€al. (2011, Chapter 16 this volume) and Bouwman et€al. (2011).

Such differences between sectors are clearly reflected in the trends of Nr pollution to air and water as summarized by Oenema et€ al. (2011a), with only modest reductions in the agricultural emissions. If these effects of recent policies over the last two decades are put into historical context, it is clear that European use and emissions of Nr from all sources are still greatly in excess of natural rates. The massive increase in fertilizer and manure use in Europe since 1860 is shown in Chapter€2 by Erisman et€ al. (2011). This can be compared with the amounts Nr inputs through N fixation (crops and NOx formation) mineral fertilizer application feed and food imports, and the consequent emissions of Nr through leaching and run-off, and as ammonia (NH3) and nitrogen oxides (NOx) emission to air (Figure€5.1). Overall, there has been a major up-Â�regulating of the European nitrogen cycle by human activities. Recent policies over the last two decades have made some progress in reducing both inputs and emissions, but represent only the first step in developing optimized strategies for Nr management. The message that emerges is that there is a huge diversity of Nr pollutant forms, including NOx, nitrous oxide (N2O), NH3, NO3–, leading to many secondary pollutants (including many organic nitrogen forms in water and in air), and an even longer list of environmental effects. The problem of Nr in the environment provides a degree of complexity that few scientists are able cover in full, that policy makers have so far tackled

84

The complexity and many facets of the nitrogen cycle are clearly reflected in the structure and relationships of the scientific communities that have developed to study it. This is illustrated in Figure€5.2, for the research area of atmosphere– biosphere exchange of Nr and interacting compounds. In this area, in recent decades, the degree of specialization has become such that the individual compounds have become the focus for whole research communities, and with the main focus of integration being between components listed in the vertical in Figure€5.1. Examples are recent major collaborations, such as the GRAMINAE project, which focused on the exchange of ammonia, also considering interactions with nitric acid and inorganic aerosol dynamics (Sutton et€ al., 2001, 2009a), and the GREENGRASS project, which investigated the integration between N2O, CH4 and CO2 exchange processes with European grasslands (Soussana et€al., 2007). At a similar scale, the NOFRETETE project addressed the dynamics of N2O and NO fluxes with European forest soils (Pilegaard et€ al., 2006). Given the different measurement tools relevant for each of these chemical species and diverse set of biogeochemical proÂ� cesses (e.g., plant processes, soil processes, atmospheric chemistry interactions), it therefore becomes a major challenge to integrate our understanding of all of the different components listed in Figure€ 5.2. Such integration efforts have become an important focus in recent years, such as within the NitroEurope and ACCENT research communities (Sutton et€al., 2007; Fowler et€al., 2009), which have started to develop the links across the whole suite of pollutant forms shown in Figure€5.2. It is evident, however, that these efforts are part of addressing an even larger challenge to integrate communities, not just between Nr forms, but between environmental media and spatial contexts. At present, there is still only limited connection between many Nr research communities, such as between interests in stratospheric chemistry, biosphere–atmosphere exchange, freshwater and marine pollution. In aquatic research for instance, the community dealing with nitrate groundÂ�Â�water contamination is rather distinct from that concerned with nitrogen river fluxes causing coastal zone eutrophication. A similar tension is repeated across many domains within the nitrogen cycle, as the tendency to specialization of recent decades is reflected by the need to assess the wider perspective. From the perspective of environmental management, it is clear that much more linkage between science communities

Mark A. Sutton

Net GHG

Nitric Acid (HNO3)

Ozone (O3)

Carbon dioxide (CO2)

Wet deposition + – NH4 and NO3

Aerosol NH4NO3

Nitrogen dioxide (NO2)

Methane (CH4)

Particle Organic Nitrogen (PON)

Ammonia (NH3)

Nitric oxide (NO)

Nitrous oxide (N2O)

Volatile Organic Nitrogen (VON)

across the N cycle is needed (Galloway et€ al., 2008; Sutton et€al., 2009b). This is essential to deal with issues of trade off ’s between different Nr forms and linked biogeochemical cycles. Such management issues apply over different temporal and spatial scales. For example, ways need to be understood of how Nr migrates through individual natural or agricultural ecosystems, and subsequently moves from one ecosystem type to another, either through human transfers (such as agricultural products) or by dispersion through water courses and the atmosphere. With the centre of gravity of recent European nitrogen research tending toward individual Nr forms at a series of different scales, the question that arises is how fast and how far should a greater level of integration be developed. Here it must be recognized that the process of scientific integration is a slow one, and that an effort such as the European Nitrogen Assessment represents only the first steps on the path toward integration. It is with this in mind that the following group of chapters (Part II) retain a current focus on the point and process scale reflecting the expertise of the current scientific communities (Butterbach-Bahl et€al., 2011a; Durand et€al., 2011; Voß et€al., 2011; Hertel et€al., 2011, Chapters 6–9 this volume). The ‘integration target’ for these chapters is to ensure that all the Nr forms and their major interactions are considered, focusing on what we know about Nr processing in each of these systems, as well as the major knowledge gaps. While the complexity of these systems will always make it hard to take a fully integrated approach at the process level, such a focus is nevertheless essential as the foundation for understanding the component mechanisms underlying the system. In developing the ENA though a series of workshops as explained in Chapter€1 (Sutton et€al., 2011, this volume), the degree of integration achieved in Part II of the Assessment, provided the basis to develop the subsequent stages. Hence Part III of the Assessment develops the next steps of integration, in examining the nitrogen cycle of Europe through successive spatial scales.

Figure€5.2 Specialization in Nr and greenhouse gas fluxes over recent decades has led to separation of research communities on biosphere– atmosphere exchange according to chemical compound. The vertical arrows indicate first stages of emerging integration, while the vertical blocks indicate a stronger separation between research communities.

5.2.2╇ Integrating through the nitrogen cascade A useful concept in upscaling nitrogen processes that has informed recent research is the ‘nitrogen cascade’ (Galloway et€al., 2003). In the classical concept of the nitrogen cycle, the emphasis is on recycling between atmospheric di-nitrogen (N2) and the many Nr forms. In a natural system, the nitrogen cycle is seen as being in balance, as nitrogen fixation from N2 to Nr is ultimately matched by denitrification returning Nr to N2. Under anthropogenic modification, the amounts of Nr in circulation are increased in multiple directions. The concept of the nitrogen cascade emphasizes a rather different view of the same system under human influence. Substantial energy is needed to fix nitrogen from N2 into Nr forms, be it fuels needed for industrial production of ammonia or the photosynthetic energy needed for biological nitrogen fixation. The fixation process therefore raises nitrogen from a low to a high energy state, providing a starting point for the subsequent cascade. The important point of the cascade is that, as this energy is gradually dissipated, the Nr converts between a multiplicity of different forms, with different environmental consequences at every stage. Each molecule of Nr can therefore be expected to be involved in several environmental effects before it is finally denitrified back to unreactive N2. The complexity of the nitrogen cascade means that any graphical description is necessarily a simplification. An extremely simplified version is shown in Chapter€ 1 (Sutton et€al., 2011), illustrating the cascade of fertilizer Nr produced by the Haber–Bosch process. This can be compared with Figure€5.3, which provides a more complete€– but still very simplified€– summary of the cascade, accounting for each of the main anthropogenic influences on nitrogen fixation. For simplicity, the only numbers shown in Figure€5.3 are the amounts of N2 fixed to Nr by human activities, comparing global estimates (Galloway et€al., 2008; Erisman et€al., 2011) with European estimates, as developed in Chapter€16 for the EU-27 by Leip et€al. (2011, this volume). At both Global and European

85

The challenge to integrate nitrogen science and policies

Figure€5.3 Simplified view of the nitrogen cascade, highlighting the major anthropogenic sources of reactive nitrogen (Nr) from atmospheric di-nitrogen (N2), the main pollutant forms of Nr (orange boxes) and nine main environmental concerns (boxes outlined with blue). Estimates of N fixation for the world (Tg /yr for 2005, in black; Galloway et€al., 2008) are compared with estimates for Europe (Tg /yr for 2000, in blue italic; Leip et€al., 2011, Chapter€16 this volume). Energy is needed to fix N2 to Nr, which is gradually dissipated through the cascade with eventual denitrification back to N2. Blue arrows represent intended anthropogenic Nr flows; all the other arrows are unintended flows.

levels, industrial production is by far the largest source of new Nr. Industrial production accounts for 63% and 70%, of total anthropogenic nitrogen fixation at Global and European scales, respectively, of which use of the Haber–Bosch process to make fertilizers is the largest component. Compared with the global average, Europe has a higher contribution of Nr production from combustion processes (power generation, transport, other industry) (13%, 21%, respectively), reflecting the fact that Europe is an industrialized region with a high intensity of energy use and motorized transport. By contrast, compared with the global average, Europe has a smaller contribution from crop biological nitrogen fixation (24%, 8%, respectively), owing to the high use of mineral fertilizers in Europe. It is clear from the magnitude of these numbers that understanding the pathways and fate of industrially produced nitrogen fertilizers must be a priority for Europe. Secondly, the Nr formed from high temperature combustion is of particular interest as it represents a completely unintended production of Nr. As can be seen from Figure€ 5.3, each of the forms of Nr produced can be inter-converted, having several environmental effects, before eventually being denitrified to N2 at the end of the cascade. For the purpose of this summarized view, nine main environmental concerns are shown in Figure€5.3, as further discussed in Section 5.4.

86

It must be acknowledged that, to date, the nitrogen cascade represents a purely conceptual framework. So far, models have yet to be constructed that fully trace the pathway of anthropogenically fixed Nr though all forms and stages, showing how many times on average each Nr atom contributes to different environmental effects. The cascade concept nevertheless provides a stimulus for integrating Nr research at different spatial scales, as well as for identifying control points (Erisman et€al., 2001; Galloway et€al., 2008; see Section 5.5). Based on the need to track Nr through the different environmental compartments and impacts in the cascade, Part III of the Assessment emphasizes the building up of understanding between the different scales. Each of the chapters in Part€III necessarily addresses the lateral flows of Nr, including, as relevant, those by direct human transfers (movement of fertilizer, manure, feed and food), by atmospheric transport and by water flow in catchments. Figure€5.4 illustrates the conceptual upscaling of Part III, putting into context the following parts of the assessment. The components scale up from the farm level (Jarvis et€al., 2011, Chapter€10 this volume) to complete multimedia Nr integration across Europe (de Vries et€al., 2011; Leip et€al., 2011, Chapters 15 and 16 this volume). Traditionally, much effort has been put into addressing regional scale transfers in watersheds and in the atmosphere,

Mark A. Sutton

Upscaling & integration

Part V

European Nitrogen policies & future challenges Water quality

Air quality

Greenhouse balance

Ecosystems & biodiversity

Soil quality

Part IV

Figure€5.4 Main structure of the following parts of the European Nitrogen Assessment, highlighting the upscaling elements addressed in Part III and the five key societal threats of excess nitrogen addressed in Part IV.

Integrating nitrogen fluxes at the European scale Geographic variation in terrestrial nitrogen budgets across Europe Nitrogen flows from European watersheds to coastal marine waters Nitrogen flows and fate in rural landscapes

Atmospheric transport & deposition of nitrogen in Europe Nitrogen flows and fate in urban landscapes

Part III

Nitrogen flows in farming systems across Europe

Processes & mechanisms

Nitrogen processing in the biosphere

and these scales are addressed by the chapters of Billen et€al. (2011, Chapter€ 13 this volume) and Simpson et€ al. (2011, Chapter€14 this volume). Much is known at these scales, which represent relatively mature, though still challenging, areas of research. As complex areas, there remain many unknowns, including tracing the fate of nitrogen through compartments and the chemical exchanges between inorganic and the many organic Nr forms. Nevertheless, by comparison, full assessment of the nitrogen cascade at the landscape scale has been much less studied compared with regional Nr transfers in either air or water. A full view of the Nr cascade in landscapes integrates spatially explicit transfers between source and sink landscape elements (e.g., farms, fields, roads, forests, mountains) and between each of the air, land and water compartments. The landscape scale is particularly important as it is the linking scale between the plot and region, being the scale in which many local decisions on Nr management take place. Two chapters address these new scales of integration, contrasting Nr flows in rural and urban landscapes (Cellier et€al., 2011; SvirejevaHopkins et€al., 2011, Chapters 11 and 12 this volume). If the integration of nitrogen science across Nr forms and spatial scales already represents a complex challenge, it cannot be forgotten that nitrogen also interacts with many other biogeochemical cycles. Given the complexity and resources, it is important to limit the scope of this assessment to focus on the nitrogen interactions, rather than assess all biogeochemical cycles simultaneously. In order to optimize the scientific effort, the approach taken here centres on the interactions between nitrogen forms, while not forgetting the main links with other element cycles where these occur. These main links tend to be different according to the environmental compartment being considered. In atmospheric chemistry, the main biogeochemical links are with sulphur chemistry and organic carbon chemistry (e.g., in aerosol and photochemical oxidant transformation processes, Hertel et€ al., 2011; Simpson et€al., 2011). In terrestrial ecosystems, key relationships exist

Part II

between nitrogen supply, turnover of organic matter and net carbon storage, both through influences on primary production and decomposition (Butterbach-Bahl et€ al., 2011a) and with phosphorus through the use of manures in agriculture (Jarvis et€al., 2011). Finally, in freshwater and marine systems, the relationships between nitrogen and phosphorus are especially important, as well as interactions with silica (Durand et€al., 2011; Billen et€al., 2011). The Assessment thus addresses the interactions with key other element cycles according to the priorities in the different environmental compartments.

5.3╇ The challenge to integrate nitrogen policies Similar to the scientific perspective, Oenema et€ al. (2011a, Chapter€4 this volume) explain how current environmental policies in Europe have taken a rather fragmented approach to the nitrogen cycle. The reasons for this appear to relate both to this historical separation between the supporting scientific communities, and the fact that such policies have been driven by perceived environmental problems in different contexts. This is equally reflective of the way in which policy portfolios are typically separated in government departments (e.g., between water and air, between urban and rural, between agriculture and nature, etc.). It can even be the case that political positions require a deliberate separation between issues, making it harder to negotiate joined-up approaches. An example here is the desire of some parties to the UN Framework Convention on Climate Change to ensure that climate related policy issues are not addressed in other conventions. This makes it challenging to address multi-effect interactions in international conventions, which is compounded by the fact that the relevant multi-lateral environmental agreements are made up of different memberships (e.g., UN, UN-ECE, EU-27, etc.). In response to the present position, there is a strong case to be considered for developing more joined-up, integrated

87

The challenge to integrate nitrogen science and policies

approaches to managing nitrogen in the environment (see also Section 5.5 in relation to Part V of the Assessment). Aside from the procedural difficulties, a key challenge that emerges is to find the optimum level of integration. For example a balance should be identified between the simplicity of single-issue approaches, versus the inefficiencies that occur as a result of not considering the major interactions. Put in another way, it must be recognized that joined-up approaches take longer to develop and the advantages (in optimization, improved delivery, synergies, avoidance of trade-offs, etc.) must be shown to outweigh the risk of additional complexity (Oenema et€ al., 2011b; Bull et€al., 2011, Chapters 23 and 25 this volume). In particular, the benefits of integration must be shown to be achievable in order to address the potential concern that the complexity of integration becomes an excuse for inaction. However, an integrated approach can also lead to simpler policies if integration is used to weigh and prioritize the many N-issues. In this context, it is extremely important to identify the priority issues that should be integrated. If a short list of key issues can be established, this would provide a foundation on which to identify an achievable level of integration, at least between the key issues. The short-list can then inform the discussion on to what extent specific multi-sector, multi-issue policies on nitrogen are needed, or to what extent existing policies should be further developed to be nitrogen aware, what may be called ‘nitrogen proofing’.

5.4╇ Distilling complexity to integrate and communicate nitrogen It is not surprising that something as multi-faceted as the nitrogen cascade should be associated with an extremely large number of environmental concerns. As explained above, establishing a short-list of priority issues is therefore important as a basis for informing the integration of environmental and other

policies. At the same time, such a short-list is needed in order to communicate the nitrogen problem more effectively to the general public. The Nitrogen in Europe (NinE) networking programme (funded by the European Science Foundation), in its aims to link together for the first time the environmental problems related to Nr, established a short list of priorities. The identification of the priorities was carried out in two stages. Starting with a list of around 20 environmental issues related to nitrogen, NinE agreed a list of 9 major environmental concerns, as a basis for communication of its efforts to link issues across the nitrogen cycle. The outcome is clearly expressed in the NinE logo, including the mnemonic acronym ‘ACT AS GROUP’ (Figure€5.5). While this first listing of nine concerns was useful to illustrate the challenge faced by the NinE network, it was recognized it needed further simplification to allow effective communication to a wider audience. In the second stage of simplification, the full list of around 20 environmental concerns was ranked by a group of European experts as a contribution of the activities NinE, NitroEurope and COST Action 729 to an expert workshop under the Convention on Long-range Transboundary Air Pollution (Saltsjöbaden-3, Gothenburg, April 2007), with the short-list being tested through further consultation with experts and other stakeholders during the COST Action 729 Workshop on integrated assessment modelling for nitrogen (Laxenburg, November 2007). The full lists of nitrogen effects, as developed in this process (Erisman et€al., 2007) are shown in Tables€5.1 to 5.3. In estimating scores of the ‘relevance and link to nitrogen’, the experts used a scale of 1 (highest relevance) to 5 (unimportant). This prioritization combined consideration of the relevance of the issue to society and the extent to which nitrogen contributed to that issue.

Stratospheric chemistry and ozone

88

Acidification

Greenhouse gas

of soils & waters

& global warming

Terrestrial

Ozone

eutrophication & biodiversity

vegetation & health

Coastal

Urban

& marine eutrophication

air quality & health

Aquatic

Particles

eutrophication & water quality

health, visibility & global dimming

Figure€5.5 Graphical representation of the main nitrogen concerns addressed by the Nitrogen in Europe (NinE) networking programme of the European Science Foundation. The nine environmental concerns together provide the mnemonic ‘ACT AS GROUP’.

Mark A. Sutton Table€5.1╇ Summary of the direct effects of excess nitrogen on humans in relation to currently used indicators, the current existence of limit values (legally binding and/or broadly established for scientific assessment) in Europe, and the link to the nitrogen cascade. The relevance and link to N provides a prioritization estimated by an expert group for international action to mitigate the effects of excess nitrogen (based on Erisman et€al., 2007)

Direct effects on humans

Indicators

Limit?

Link to N cascade

Relevance and link to N

– ozone

O3 conc. values including SOMO35

Yes

NOx emission

3

– other photochemical oxidants

Organic NO3, PAN

No

NOx emissions

5

– fine particulate aerosol

PM10, PM2.5

Yes

NH3, NOx emissions

1

– direct toxicity of NO2

NO2

Yes

NOx emissions

2

Nitrate contamination of drinking water

NO3 conc (aq.)

Yes

NO3 leaching

2

Increase allergenic pollen production, and several parasitic and infectious human diseases

—

No

N fertilizer and N deposition

5

Blooms of toxic algae and decreased swimmability of in-shore water bodies

Chlorophyll A NO3 (aq.)

No

Run-off, N deposition

1

Respiratory disease in people caused by exposure to high concentrations of:

Relevance and link to nitrogen qualitatively incorporates the societal priority of the issue and the N contribution to that issue:€(1) highest relevance, (2) high relevance, (3) significant relevance, (4) some relevance, (5) unimportant. Indicators:€SOMO35:€sum of ozone concentration above 35 parts per billion in air; PAN, peroxyacetyl nitrate; PM10, PM2.5:€particulate matter in air having a median diameter larger than 10 μm and 2.5 μm, respectively. Table€5.2╇ Summary of the effects of excess nitrogen on ecosystems in relation to currently used indicators, the current existence of limit values (legally binding and/or broadly established for scientific assessment) in Europe and the link to the nitrogen cascade. The relevance and link to N provides a prioritization for future international action to mitigate the effects of excess nitrogen (based on Erisman et€al., 2007)

Direct effects on ecosystems

Indicators

Limit?

Link to N cascade

Ozone damage to crops, forests, and natural ecosystems

O3 flux, AOT40

Relevance and link to N

Yes

NOx emission

2

Acidification effects on forests, soils, ground waters, and aquatic ecosystems

Critical loads

Yes

N deposition

2

Eutrophication of freshwaters, lakes (incl. Biodiversity)

BOD, NO3 (aq) Critical loads

Yes No

Run-off, N deposition

3

Eutrophication of coastal ecosystems inducing hypoxia (incl. Biodiversity)

BOD, NO3 (aq) Critical loads

Yes No

Run-off, N deposition

1

Nitrogen saturation of soils (incl. effects on GHG balance)

Critical loads

Yes

N deposition

1

Biodiversity impacts on terrestrial ecosystems (incl. Pests and diseases)

Critical loads, critical level (NH3 in air)

Yes

N deposition

1

Relevance and link to nitrogen qualitatively incorporates the societal priority of the issue and the N contribution to that issue:€(1) highest relevance, (2) high relevance, (3) significant relevance, (4) some relevance, (5) unimportant. Indicators:€AOT40:€accumulated ozone concentration above a threshold of 40 parts per billion in air; BOD:€biological oxygen demand in water.

Tables€ 5.1–5.3, show that the following issues were given the highest scores: (1) Respiratory disease caused by fine particulate matter in the atmosphere. (2) Blooms of toxic algae and decreased swimmability of in-shore water bodies. (3) Eutrophication of coastal ecosystems inducing hypoxia (incl. their biodiversity).

(4) Nitrogen saturation of soils (incl. effects on GHG balance). (5) Biodiversity impacts on terrestrial ecosystems (including pests and diseases). (6) Global climate warming induced by excess nitrogen. (7) Regional climate cooling induced by aerosol.

89

The challenge to integrate nitrogen science and policies Table€5.3╇ Summary of the effects of excess N on other societal values in relation to currently used indicators, the current existence of limit values (legally binding and/or broadly established for scientific assessment) in Europe, and the link to the nitrogen cascade. The relevance and link to N provide a prioritization for future international action to mitigate the effects of excess nitrogen (based on Erisman et€al., 2007)

Effects on other societal values

Indicators

Limit?

Link to N cascade

Relevance and link to N

Odour problems associated with animal agriculture

NH3 concentration

No

NH3 emission

5 (in Europe)

Effects on monuments and engineering materials

Precipitation acidity, O3, PM10, PM2.5 concentrations

Yes

NOx, NH3

3

Regional hazes that decrease visibility at scenic vistas and airports

PM2.5

No

NOx, NH3

4 (for Europe)

Depletion of stratospheric ozone

NOx, N2O concentrations

No

NOx, N2O

3

Global climate warming induced by excess nitrogen

N2O, CH4, CO2 concentrations

No

N2O (direct & indirect sources), CH4, CO2

1

Regional climate cooling induced by aerosol

PM2.5 concentration

No

NOx, NH3

1

Relevance and link to nitrogen qualitatively incorporates the societal priority of the issue and the N contribution to that issue:€(1) highest relevance, (2) high relevance, (3) significant relevance, (4) some relevance, (5) unimportant. Indicators:€PM10, PM2.5:€particulate matter in air having a median diameter larger than 10 μm and 2.5 μm, respectively.

Based on these and recognizing some overlap, a group of five key societal threats of nitrogen was identified: • Air quality (including respiratory disease concerns), especially as affected by NOx, O3 and particulate matter. • Water quality (including ecosystems and human health concerns). • Greenhouse balance (including effects on trace gases and atmospheric aerosol). • Ecosystems and biodiversity (including pests and diseases), especially as affected by atmospheric Nr deposition. • Soils quality (including effects on nitrogen saturation and acidification). In regard of the last threat, feedback from stakeholders at the Laxenburg workshop, strongly argued for the inclusion of soils, given the need to include soil acidification and to consider soils as an integrator of different pressures. Subsequent reflection of these five key societal threats of excess nitrogen has shown several interesting features. Firstly, with these headings, many of the other environmental concerns become automatically incorporated into the overall framework. For example, the air quality threat of Nr includes both ozone and particulate matter, while the soils threat includes acidification and alteration of soil organic matter storage. Secondly, this subdivision into five threats has allowed analysis of the negative effects caused by nitrogen together with some potential benefits. This is the case for the effect of nitrogen on greenhouse balance, where warming effects due to Nr (e.g., nitrous oxide, ozone) are at least partly offset by several cooling effects (e.g., aerosol, carbon sequestration). In addition to highlighting the key issues for policy makers, the key societal threats also lend themselves to developing wider communication approaches. The selection of five issues highlights the complexity of the nitrogen problem (i.e. it is multi-issue), while focusing on a list which is sufficiently short to remember. The threats can, for example, be considered as the WAGES of excess nitrogen, being a mnemonic for the

90

five key threats to:€Water, Air, Greenhouse balance, Ecosystems and€Soils. Somewhat more surprising is the observation that this list of five threats also falls neatly into another ancient communication framework. The Greek philosopher Empedocles is famous for having presented the fundamental components of matter as four ‘elements’:€water, air, fire and earth (Wright, 1995), to which Aristotle subsequently added aether as the quintessence, or fifth element (de Caelo I.2, Guthrie, 1986; Wilderg, 1988). Figure€5.6 illustrates the analogy between these elements of the Greek cosmos and the five key threats of excess nitrogen. In this model of nitrogen in the environmental macrocosm, the allocation of water, air and soil is straightforward, while greenhouse balance is linked to fire, with ecosystems and biodiversity, placed as the quintessence. Figure€5.6 also shows how it is possible to apply the system to highlight the key Nr forms for each threat. Such an assignment is naturally open to much debate, and must be considered loosely. However, any such controversy should not hinder the use of this model to communicate nitrogen issues, just as the longstanding debate in identifying Empedocles’ elements (Wright, 1995) did not Â�prevent€– or even encouraged€– acceptance of his approach. Part IV of the ENA considers each of these five societal threats of excess nitrogen in turn, water quality (Grizzetti et€al., 2011, Chapter€17), air quality (Moldanová et€al., 2011, Chapter€18), greenhouse balance (Butterbach-Bahl et€al., 2011b, Chapter€19), ecosystems and biodiversity (Dise et€al., 2011, Chapter€20) and soil quality (Velthof et€ al., 2011, Chapter€ 21). A deliberately sectoral approach is taken in each of these chapters, providing the basis to show the key issues that need to be linked in developing more integrated approaches. As far as possible, trends in Nr threats over time and across Europe are explored to show how the problem has arisen and to highlight the outlook in the light of existing policies. Although these five chapters were initially developed in parallel using a common framework, it quickly became apparent that different approaches were needed according to each

Mark A. Sutton

A

FIRE

B

Hot

Dry

AETHER QUINTESSENCE

AIR

GREENHOUSE BALANCE

AIR QUALITY

EARTH

Moist

Dr y

Hot

Moist

Cold

SOIL QUALITY

ECOSYSTEMS & BIODIVERSITY

Cold WATER QUALITY

WATER

C

Nitrous oxide (N2O) & Nitrogen-GHG interactions Atmospheric chemistry

Nitrogen oxides (NOx) particles (PM2.5) & ozone (O3) in air

Storage of organic matter

Ammonia & organic N in Ecosystems

Soil Organic Nitrogen (SON)

Aqueous Transformations

Human health Aqueous nitrate (NO3–) & other dissolved N

Figure€5.6 Analogy of nitrogen concerns in the environment to the Empedoclean–Aristotelian framework of the macrocosm:€(A) visualization of Empedocles’ elements, indicating their common properties, together with aether, the Aristotelian ‘quintessence’; (B) the five key societal threats of reactive nitrogen visualized in macrocosmic framework; (C) key chemical forms and issues typifying each of the key societal threats. In each model, the diagonals represent shared properties of the adjacent elements.

issue. Thus some of the key threats like water quality and air quality represent mature areas, where data on spatial patterns and trends are well established. By contrast, the threat of Nr on greenhouse gas balance represents a much less well developed research and policy area. In this case, the chapter provides a first examination, drawing together the evidence needed to guide assessment of the net effects and the future policy development. The use of these five threats was also tested as a communication tool by the NinE programme in collaboration with the BBC ‘Green Room’ (NinE, 2008; Sutton, 2008). In exploring the idea of the ‘NitroNet’, the web of interlinked challenges related to nitrogen, members of the public visiting the web-site were asked in the ‘NitroNet Poll’ how they would rate the different societal threats, on a scale of 1 to 5, from unimportant to important. The results, summarized in Figure€5.7, show

that, while there were some significant differences between mean scores, members of the public had a wide range of views over which problems were the priority. The clear message was that all the issues need to be addressed. Feedback also showed that while some respondents accepted the challenge to prioritize between issues, others rejected the whole idea, considering that it is impossible to set such priorities. Whatever view one takes, the very debate on the validity of such a comparison again serves its main purpose of encouraging people to start thinking about how to manage the multiple threats of excess nitrogen. An important communication tool is the use of visual images. For example, a further simplification of the Greek cosmological analogy (Figure€ 5.6) provides the basis to summarize visually the five key societal threats of excess nitrogen (Figure€5.8). Of the five threats, it is notable that the two that scored highest in

91

The challenge to integrate nitrogen science and policies

completely eradicated, being replaced by a thick algal slime. Of course there are many stages between the conditions shown by these two photographs, but the comparison powerfully demonstrates the way in which reactive nitrogen supply is having major effects on terrestrial biodiversity. The second (lower) image shows the results of nutrient enrichment in coastal marine waters. Again, excess reactive nitrogen encourages algal growth, forming harmful ‘blooms’ which reduce oxygen availability and threaten fish and other species populations. Such algal blooms can have negative effects on bathing water quality, reducing water visibility and causing high levels of foam, as shown here, resulting from released gelatinous substances.

a

Water quality

bc

Air quality Greenhouse gas balance

abd

Ecosystems & biodiversity

ad

Soil quality

bc 0

1

2

3

4

5

Score

Figure€5.7 Outcome of scores from members of the public regarding the relative priorities between the five key societal threats of excess nitrogen, from the NitroNet Poll of NinE (2008) in cooperation with the BBC ‘Green Room’. Members of the public were asked to allocate a score for each issue ranging from 1 (low priority) to 5 (high priority). The error bars are standard errors (n = 175), with the different letters (a to d) indicating significant Â�differences (P = 0.05) based on paired t-tests (a shared letter indicates no significant difference between bars).

Figure€5.8 Summary of the five key societal threats of excess reactive nitrogen, with the visualization based on the analogy presented in Figure€5.6. Photo€sources: Shutterstock.com and garysmithphotography.co.uk.

the NitroNet Poll€– Water quality, and Ecosystems and biodiversity€– are also highly amenable to visual images. To illustrate this, Figure€5.9 shows two examples of the effects of excess reactive nitrogen in the environment. The first (top) shows some of the consequences of high levels of atmospheric ammonia deposition on the epiphyte flora of birch woodland in the United Kingdom. On the left, at a clean site, the birch trunk shows a rich diversity of lichens and bryophytes. On the right, at a site near an intensively managed livestock farm, the natural flora has been

92

5.5╇ Integrating European nitrogen policies and future challenges The analysis of key societal threats in Part IV of the Assessment provides the platform to develop more integrated science and policy approaches. With the main issues clearly identified, Part V of the Assessment brings these threats together and relates them to the benefits of Nr for food security and industrial production. As the same time, such approaches can be informed by the development of more integrated approaches, such as the establishment of comprehensive nitrogen budgets and maps for Europe (de Vries et€al., 2011; Leip et€al., 2011). Three chapters in Part V address different elements of integration followed by two chapters on how to communicate the outcomes with policy makers and the general public. It is evident that the simple comparison of the ‘NitroNet Poll’ (Figure€5.7) represents only a first step, and it is a major challenge to develop more rigorous approaches to bring together the nitrogen issues and set priorities. In fact, the NitroNet Poll captures significant differences of opinion, as the respondents were answering based on a wide mix of perspectives and degrees of knowledge. In developing a more formal approach to compare the issues, economic methods may be used. In the subsequent assessment, Chapter€22 applies economic approaches to assess the environmental costs and societal benefits of reactive nitrogen in Europe (Brink et€ al., 2011, Chapter 22 this volume). Using willingness-to-pay approaches, the authors estimate societal damage costs as euro per kg Nr emission for each of Nr to water, NH3 to air, NOx to air and N2O to air. The quantified uncertainty bounds estimated by Brink et€al. highlight the major challenges of such an approach. Nevertheless, they provide a foundation for discussion with policy makers, showing the substantial financial benefits of mitigating Nr emissions. The different elements of managing Nr are brought together by Oenema et€al. (2011b, Chapter€23 this volume). Based on the foregoing contributions, they analyze what it means to develop ‘integrated approaches’ to nitrogen management. They examine several dimensions of integration, linking scales, issues, stakeholders etc. Based on these reflections, they identify a package of key actions that together provide an integrated perspective for overall management of anthropogenic Nr emissions and their effects. Such a short list builds on the key actions previously identified by Galloway et€al. (2008) and warrants further consideration as a foundation for developing future European

Mark A. Sutton Figure€5.9 Visual illustrations of the effect of excess nitrogen on the natural environment. Top:€effect of atmospheric ammonia on epiphyte biodiversity in birch woodland in the UK:€left, clean conditions showing a rich array of lichens and bryophytes (photo:€Ian Leith); right, replacement of the natural epiphyte flora under high ammonia by a thick algal slime (photo:€Mark Sutton). Bottom:€nitrogen input into coastal seas in excess over silica, can cause severe algal blooms, in this case with Phaeocystis globosa, leading to a build up of gelatinous foam on a Dutch beach (photo: Gilles Billen).

policies. While some of the key actions are already being implemented in existing policies, it is equally clear that most of them require much more attention. In developing future perspectives, it is vital to have a clear idea on the trajectories of Nr emissions and effects that can be expected. This places an important role on scenario development, considering both future economic development and current plans for Nr mitigation. Such a first assessment for the major Nr emissions in Europe is brought together by Winiwarter et€al. (2011, Chapter€24 this volume). A major achievement is the combination of both

short- and medium-term scenarios, though further efforts will be needed to develop scenarios of integrated packages for Nr management, including the long-term (e.g., 2100). In developing the European Nitrogen Assessment, it has become clear how nitrogen issues cut across all global change threats. At the same time, the connected nature of the nitrogen cycle has clearly not been fully addressed by policy makers or recognized by the general public. A key task for the Assessment must therefore be to consider how better to communicate the nitrogen issues to these audiences.

93

The challenge to integrate nitrogen science and policies

Bull et€al. (2011, Chapter€25 this volume) address the issue of how to communicate the Nr challenge with policy makers, highlighting the possibilities for more effective coordination between multi-lateral environmental agreements. They demonstrate the complexity of the international landscape for nitrogen, involving agreements between many different sets of national parties (e.g., UN, UN-ECE, EU, other groupings). The key challenge they raise is to develop mechanisms that ensure joined up approaches to nitrogen management, linking all of the key societal threats. They assess the possibility and relative merits different options, ranging from coordination actions, ‘nitrogen proofing’ existing policies, to the establishment of an international convention on nitrogen. An intermediate option is to develop the basis for a multi-media protocol on nitrogen, drawing on the work of the existing international conventions. The analysis of Bull et€al. (2011) reflects the tension to ensure streamlined approaches that avoid substantial additional burden on the current conventions, while maximizing the synergies. In particular, it remains a challenge to develop a framework that develops sufficient ‘gravity’ to ensure that the different Nr related problems are drawn together. Finally, Reay et€al. (2011, Chapter€26 this volume) address the issue of how to communicate the European nitrogen challenge to the general public. They highlight how insufficient recognition has been given to the different barriers to optimizing future human use of nitrogen and its environmental consequences. They assess these barriers and relate them to the key societal levers, drawing on experience from the societal and policy challenge to manage climate change. In particular, they highlight the role of societal choice both in raising awareness of the nitrogen challenge, and in making a significant contribution towards meeting mitigation targets. Patterns of societal consumption€– one of the seven key actions listed by Oenema et€al. (2011b, Chapter€23 this volume)€– are identified as a key focus relevant for the nitrogen cycle, especially related to diets, food choice and food waste. Based on the dominance of the food chain in the nitrogen cascade (Figure€5.3), human food choices have major effects on the overall amounts of Nr processed and lost to the environment. Reay et€ al. (2011) discuss how a ‘segmented strategy’ can be used to reach different stakeholders, including the use of proven communication tools, with involvement of the social sciences.

5.6╇ Conclusions A key emerging feature of the European Nitrogen Assesment (ENA) is the challenge to develop holistic approaches that integrate across science disciplines, across policy domains and build the links between the science and policy communities. Until now the multi-media, multi-impact nature of the nitrogen cycle has been much broader than the individual science communities, and several steps are needed to provide the basis to address the whole. This chapter reflects at the end of Part I of the ENA on how to develop subsequent parts of the Assessment. Based on the particular characteristics of the European nitrogen problem, on the major benefits of reactive nitrogen to the European

94

economy and on the many existing policies that address parts of the nitrogen cycle, it is evident that Europe has a long history and experience of actively managing its nitrogen cycle. At the same time, it is clear that the complexity of the system has hindered the development of a broad perspective that would aim to optimize European nitrogen management. The need for such a perspective is fully justified by the multiple, non-linear interacting impacts illustrated by the nitrogen cascade. In seeking to optimize future European nitrogen management, the first steps must be to agree on the scientific foundations. For this reason, the following parts of the ENA focus on our understanding of nitrogen processes (Part II) and how these can be upscaled from the farm to the European level (Part€III). The next step must be to establish the basis for prioritization of the key issues, which is necessary to set the framework for judging the optimum level of integration. Given the complexity of integration, the components need to be kept as simple as possible to be understandable and implementable in policy. Without such an issue building, the complexity of nitrogen management risks becoming seen in policy circles as an excuse for inaction. Based on these reflections, this chapter reports a distillation of the many different nitrogen threats facing society. In a first cut, the Nitrogen in Europe (NinE) programme identified a network of nine main concerns of excess Nr. Such a grouping is well fitted to the scientific community, emphasizing the need to cooperate between media and disciplines. However, it was concluded that this listing remains too complicated to be the basis for developing integrated approaches with policy makers or to communicate the issues with society. The chapter therefore describes a second distillation, into five key societal threats of excess nitrogen. This listing of five key threats was derived from a prioritization and clustering of around twenty environmental concerns, with the number five reflecting a deliberate balancing between complexity and simplification. The five key societal threats:€ Water quality, Air quality, Greenhouse gas balance, Ecosystems and biodiversity, and Soil quality turn out to be well suited to developing communication models. In mnemonic form these threats can be seen as the ‘WAGES of excess nitrogen’, and be easily illustrated by analogy to each of the ‘elements’ of classical Greek philosophy. Part IV of the Assessment analyzes each of the five key threats, clearly highlighting the main reasons why society should be concerned about excess Nr in the European environment. The initial aim was to ensure that these five chapters were closely streamlined in approach, highlighting the magnitude, spatial distribution, temporal trends and current efforts to manage each issue. However, it quickly became clear that the knowledge base for the five threats is very different, with the result that the greenhouse gas and soils threat chapters focus much more on problem quantification, while the water, air, and ecosystems chapters are able to provide more detail on trends and patterns. Finally, the key societal threats provide the basis to inform the development of more holistic approaches to nitrogen

Mark A. Sutton

management in Part V of the Assessment. A first examination of the costs of Nr pollution on the European environment and the benefits of Nr mitigation is conducted, scenarios of future Nr use and pollution are brought together, and integrated approaches to Nr management are developed. In particular, a package of seven key actions is identified, which would together provide the basis for integrated management of the European Nr resource, minimizing the environmental threats. These messages provide the foundation for further communication and application by policy makers and by society at large. Here major challenges remain. In the context of policy development, it is evident that more holistic approaches are needed, but much more work is needed by policy makers to agree the optimum degree of policy integration, and the right framework within which it should be conducted. There are major opportunities for closer working between existing multi-lateral environmental agreements in Europe, such as the different conventions of the UN-ECE and across EU policy domains. Such action needs to develop sufficient ‘gravity’ to pull together the key nitrogen concerns, while being sufficiently streamlined in order to have a chance of making effective progress. The bottom line of Section V of the Assessment is the challenge to involve European citizens in recognizing and taking action on nitrogen. Major efforts still need to be devoted to simplifying the nitrogen story to make it understandable to the general public, developing the key messages. One of the key hooks identified is the importance of personal food choice to the whole nitrogen cascade. This illustrates the need for future efforts to quantify the impacts and mitigation potential, as well as to quantify the co-benefits of eating patterns that are both healthy for the individual and for the environment.

Acknowledgements The authors gratefully acknowledge funding support from the European Commission for the NitroEurope Integrated Project and the COST Action 729 programme, the European Science Foundation for the NinE programme and the UK Department for Environment Food and Rural Affairs for support of the UN-ECE Task Force on Reactive Nitrogen, together with underpinning support from the UK NERC Centre for Ecology and Hydrology. We are grateful to David Leaver for support in implementing the NitroNet Poll.

Supplementary materials Supplementary materials (as referenced in the chapter) are� available online through both Cambridge University Press:€ www.cambridge.org/ena and the Nitrogen in Europe website: www.nine-esf.org/ena.

References Billen, G., Silvestre, M., Grizzetti, B. et€al. (2011). Nitrogen flows from European watersheds to coastal marine waters. In:€The European Nitrogen Assessment, eds. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press.

Bouwman, A. F., Klein Goldewijk, K., van der Hoek, K. W. et€al. (2011). Exploring global changes in nitrogen and phosphorus cycles in agriculture induced by livestock production for the period 1900–2050. Proceedings of the National Academy of Sciences of the USA (submitted). Brink, C., van Grinsven, H., Jacobsen, B. H. et€al. (2011). Costs and benefits of nitrogen in the environment. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Bull, K., Hoft, R. and Sutton, M.A. (2011). Co-ordinating European nitrogen policies between directives and international conventions. In:€The European Nitrogen Assessment, ed. M.€A.€Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Butterbach-Bahl, K., Gundersen, P., Ambus, P. et€al. (2011a). Nitrogen processes in terrestrial ecosystems. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Butterbach-Bahl, K., Nemitz, E., Zaehle, S. et€al. (2011b). Nitrogen as a threat to the European greenhouse balance. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Cellier, P., Durand, P., Hutchings, N. et€al. (2011). Nitrogen flows and fate in rural landscapes. In:€The European Nitrogen Assessment, eds. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. de Vries, W., Leip, A., Reinds, G. J. et€al. (2011). Geographic variation in terrestrial nitrogen budgets across Europe. In:€The European Nitrogen Assessment, eds. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Dise, N., Ashmore, M., Belyazid, S. et€al. (2011). Nitrogen as a threat to European terrestrial biodiversity. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Durand, P., Breuer, L., Johnes, P. J. et€al. (2011). Nitrogen processes in aquatic ecosystems. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Erisman, J. W., De Vries, W., Kros, H. et€al. (2001). An outlook for a national integrated nitrogen policy. Environmental Science and Policy, 4, 87–95. Erisman, J. W., Spranger, T., Sutton, M. A. et€al. (2007). Working Group 5:€Nitrogen€– integrated environmental policies. In:€Air Pollution and its Relations to Climate Change and Sustainable Development€– Linking Immediate Needs with Long-Term Challenges “Saltsjöbaden 3” (12–14 March 2007, Gothenburg, Sweden). http://asta.ivl.se/Workshops/Saltsjobaden3/Conclusions/ WG5.pdf (last accessed 15 September 2010). Erisman, J. W., van Grinsven, H., Grizzetti, B. et€al. (2011). The European nitrogen problem in a global perspective. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Fowler, D., Pilegaard, K., Sutton, M. A. et€al. (2009). Atmospheric composition change:€ecosystems€– atmosphere interactions. Atmospheric Environment, 43, 5193–5267. Galloway, J. N., Aber, J. D., Erisman, J. W. et€al. (2003). The nitrogen cascade. BioScience, 53, 341–356. Galloway, J. N., Townsend, A. R., Erisman, J. W. et€al. (2008). Transformation of the nitrogen cycle:€recent trends, questions and potential solutions. Science, 320, 889–892. Grizzetti, B., Bouraoui, F., Billen, G. et€al. (2011). Nitrogen as a threat to European water quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press.

95

The challenge to integrate nitrogen science and policies Guthrie, W. K. C. (1986). Aristotle: On the Heavens, Loeb Classical Library, Harvard University Press,€Cambridge, MA. Hertel, O., Reis, S., Ambelas Skjøth, C. et€al. (2011). Nitrogen turnover processes in the atmosphere. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Jarvis, S., Hutchings, N., van der Hoek, K., Brentrup, F. and Olesen, J. (2011). Nitrogen flows in farming systems across Europe. In:€The European Nitrogen Assessment, eds. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Jensen, L. S., Schjoerring, J. K., van der Hoek, K. et€al. (2011). Benefits of nitrogen for food fibre and industrial production. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Leip, A., Achermann, B., Billen, G. et€al. (2011). Integrating nitrogen fluxes at the European scale. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Moldanová, J., Grennfelt, P., Jonsson, Å. et€al. (2011). Nitrogen as a threat to European air quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. NinE (2008). The NitroNet Poll. Nitrogen in Europe programme of the European Science Foundation. www.nine-esf.org/?q=nitronet_poll (last accessed 15 September 2010). Oenema, O., Bleeker, A., Braathen, N. A. et€al. (2011a). Nitrogen in current European policies. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Oenema, O., Salomez, J., Branquinho, C. et€al. (2011b). Integrated approaches to nitrogen management. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Pilegaard, K., Skiba U., Ambus A. et€al. (2006). Factors controlling regional differences in forest soil emission of nitrogen oxides (NO and N2O). Biogeosciences, 3, 651–661. Reay, D. S., Howard, C.M., Bleeker, A. et€al. (2011). Societal choice and communicating the European nitrogen challenge. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Simpson, D., Aas, W., Bartnicki, J. et€al. (2011). Atmospheric transport and deposition of nitrogen in Europe. In:€The European Nitrogen Assessment, eds. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Soussana, J. F., Allard, V., Pilegaard, K. et€al. (2007). Full accounting of the greenhouse gas (CO2, N2O, CH4) budget of nine European

96

grassland sites. Agriculture, Ecosystems and Environment, 121, 121–134. Sutton, M. (2008) Snared in a homemade ‘NitroNet’. The Green Room, BBC News, 8 July 2008, http://news.bbc.co.uk/ 1/hi/sci/tech/7496036.stm (last accessed 15 September 2010). Sutton, M. A., Milford, C., Nemitz, E. et€al. (2001). Biosphere–atmosphere interactions of ammonia with grasslands:€experimental strategy and results from a new European initiative. Plant and Soil, 228,€131–145. Sutton, M. A., Nemitz, E., Erisman, J. W. et€al. (2007). Challenges in quantifying biosphere–atmosphere exchange of nitrogen species. Environmental Pollution, 150, 125–139. Sutton, M. A., Nemitz, E., Milford, C. et€al. (2009a). Dynamics of ammonia exchange with cut grassland:€synthesis of results and conclusions of the GRAMINAE Integrated Experiment. Biogeosciences (GRAMINAE Special Issue), 6, 2907–2934. Sutton, M. A., Oenema, O., Erisman, J. W. et€al. (2009b) Managing the European Nitrogen Problem:€A Proposed Strategy for Integration of European Research on the Multiple Effects of Reactive Nitrogen. Centre for Ecology and Hydrology / Partnership for European Environmental Research, Edinburgh, UK. http://www.clrtap-tfrn. org/european-research-strategy (last accessed 15 September 2010). Sutton, M. A., Howard, C. M., Erisman J. W. et€al. (2011). Assessing our nitrogen inheritance. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Svirejeva-Hopkins, A., Reis, S., Magid, J. et€al. (2011). Nitrogen flows and fate in urban landscapes. In:€The European Nitrogen Assessment, eds. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Velthof, G. et€al. (2011). Nitrogen as a threat to European soil quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Voß, M., Baker, A., Bange, H. W. et€al. (2011). Nitrogen processes in coastal and marine systems. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Wildberg, C. (1988). John Philoponus’ Criticism of Aristotle’s Theory of Aether, De Gruyter,€Berlin. Winiwarter, W., Hettelingh, J. P., Bouwman, L. et€al. (2011). Future scenarios of nitrogen in Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C.M. Howard, J. W. Erisman et€al. Cambridge University Press. Wright, M. R. (1995). Empedocles:€The Extant Fragments. Gerald Duckworth, London.

Part

II

Nitrogen processing in the biosphere

Chapter

6

Nitrogen processes in terrestrial ecosystems Part II

Lead authors: Klaus Butterbach-Bahl and Per Gundersen Contributing authors: Per Ambus, Jürgen Augustin, Claus Beier, Pascal Boeckx, Michael Dannenmann, Benjamin Sanchez Gimeno, Andreas Ibrom, Ralf Kiese, Barbara Kitzler, Robert M. Rees, Keith A. Smith, Carly Stevens, Timo Vesala and Sophie Zechmeister-Boltenstern

Executive summary Nature of the problem • Nitrogen cycling in terrestrial ecosystems is complex and includes microbial processes such as mineralization, nitrification and denitrification, plant physiological processes (e.g. nitrogen uptake and assimilation) and physicochemical processes (leaching, volatilization). In order to understand the challenges nitrogen puts to the environment, a thorough understanding of all these processes is needed.

Approaches • This chapter provides an overview about processes relating to ecosystem nitrogen input and output and turnover. On the basis of examples and literature reviews, current knowledge on the effects of nitrogen on ecosystem functions is summarized, including plant and microbial processes, nitrate leaching and trace gas emissions.

Key findings/state of knowledge • Nitrogen cycling and nitrogen stocks in terrestrial ecosystems significantly differ between different ecosystem types (arable, grassland, shrubland, forests). • Nitrogen stocks of managed systems are increased by fertilization and N retention processes are negatively affected. • It is also obvious that nitrogen processes in natural and semi-natural ecosystems have already been affected by atmospheric Nr input. • Following perturbations of the N cycle, terrestrial ecosystems are increasingly losing N via nitrate leaching and gaseous losses (N2O, NO, N2 and in agricultural systems also NH3) to the environment.

Major uncertainties/challenges • Due to their complexity, ecosystem nitrogen stocks and nitrogen cycling processes are not well studied, as compared to those of carbon. However, strong ecosystem feedbacks to global changes have to be expected, especially with regard to nitrate leaching, C sequestration and emissions of the primary and secondary greenhouse gases N2O and NO.

Recommendations • In view of the still limited knowledge on nitrogen and carbon interactions at ecosystem and landscape scales and effects of global changes (climate, N deposition, landuse, land management) on C and N cycling, multi-disciplinary research needs to be initialized, encouraged and supported. Interdisciplinary and multi-scale studies should focus on simultaneous and comprehensive measurements of all major Nr fluxes at site and landscape scales including plant uptake/release of organic and inorganic N compounds as well as microbial Nr conversion. • Based on an in-depth understanding of nitrogen cycling processes, best management options need to be developed to minimize negative environmental impacts of global change on N cycling, C/N interactions and biosphere–atmosphere–hydrosphere exchange processes.

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

99

Nitrogen processes in terrestrial ecosystems

6.1╇ Introduction Nitrogen (N) is a key nutritional element for any life form on earth. Living organisms use N to form a number of complex organic compounds such as amino and nucleic acids, chitin and proteins. In unperturbed terrestrial systems N is a factor which limits net primary ecosystem production (Vitousek and Howarth, 1991). However, this situation has changed dramatically during the last decades. In most regions of Europe there is little limitation of biomass production by N, due to intensive use of fertilizers in agricultural systems and increasing N deposition to natural and semi-natural systems. The effects of increased N availability include changes in biosphere–atmosphere exchange (such as increased aerosol formation and emissions of the greenhouse gas N2O from terrestrial systems), eutrophication of terrestrial and aquatic systems (with consequences for species composition and richness, carbon sequestration, surface and drinking water quality), or acidification of soils and water bodies following the deposition of reactive nitrogen (Nr). Nitrogen cycling in terrestrial ecosystems and landscapes is mainly driven by microbiological and plant processes, with physico-chemical processes such as diffusion, emission, volatilization, leaching or erosion leading to displacement of N on site, regional and global scales (Galloway et al., 2003; Erisman et al., 2008). To better understand how N is affecting ecosystem functioning and to predict future ecosystem responses to increased N availability, it is necessary to have detailed knowledge of the processes involved in N cycling in terrestrial systems. This chapter provides an overview of current knowledge on N stocks in terrestrial ecosystems (Section 6.2), sources of N inputs (Section 6.3), N cycling at the ecosystem scale (Section 6.4), N loss pathways (Section 6.5) and N effects at the ecosystem scale (Section 6.6). The chapter has a particular focus on forest ecosystems, while agricultural systems are more thoroughly considered in Jarvis et al., 2011 (Chapter€10 this volume).

6.2╇ Nitrogen in terrestrial ecosystems 6.2.1╇ Nitrogen as a key element in biogeochemistry Nitrogen is a key element for global biogeochemistry and its cycling is closely linked to the carbon cycle. Nitrogen availability often limits net primary production in agricultural as well as natural and semi-natural ecosystems (Vitousek and Howarth, 1991; De Vries et al., 2006). Nitrogen bound in organic compounds is an essential part of all proteins and enzymes and thus, N is driving the key metabolic processes involved in growth and energy transfer. Furthermore, N is a part of chloroÂ� phyll, the green pigment of the plant that is responsible for photosynthesis. A large fraction of N in primary producers is utilized directly in capturing energy in photosynthesis (Evans, 1989). Nitrogen has different oxidation states from −3 in NH3 to +5 in NO3− and a series of microbial processes such

100

as nitrification (autotrophic, heterotrophic), denitrification or anaerobic ammonium oxidation (anammox) have evolved to either gain energy from the oxidation process or to use oxidized N compounds as an alternative electron acceptor when growing anaerobically. In contrast to global C cycling, where the largest fluxes are associated with the net primary production of terrestrial and marine systems, the global biogeochemical cycle of N is dominated by microbial processes in soils, sediments and water bodies (Seitzinger et al., 2006). This is also a major difference from the global cycling of phosphorus, which becomes available to the biosphere mostly through mineral weathering. Depending on ecosystem type and land use, N cycling and N storage in soils and vegetation varies considerably. In agricultural systems, N cycling is dominated by N fertilization and crop removal, while in natural and semi-natural systems N cycling is largely affected by climatic, edaphic and landscape conditions and the sum of N inputs via N deposition and biological N fixation. Across differing climates, shrublands in Europe have been predominantly N-limited systems, i.e. plant production was suboptimal due to shortages in available N. In such systems N availability is mostly hampered by poor soil properties, e.g. high sand content and thus reduced ion exchange capacities and low organic matter content, which negatively affects retention of reactive nitrogen (Nr; here and in the following:€all organic N forms plus inorganic N forms except N2) in the system. Also the human use of these systems (e.g. grazing, fodder collection) over centuries has depleted the nutrient reservoirs. At the present time, shrublands and heathlands exposed to high N deposition are often showing indications of N saturation, such as changes in species composition or nitrate leaching (Schmidt et al., 2004). Wetlands are also mostly N-limited systems, due to accelerated losses of Nr via the denitrification pathway. Forests are naturally N-limited systems, at least for European climatic conditions, whereas tropical rainforests are often N-rich systems. This situation has changed markedly during the last decades due to atmospheric N deposition (De Vries et al., 2007). Forest foliage is more effective at receiving N deposition compared to other vegetation types, resulting in an increased N deposition. Signs of N saturation of forests have been widely reported and include accelerated growth and significant Nr losses via nitrate leaching and N trace gas emissions (Dise et al., 2009; Pilegaard et al., 2006).

6.2.2╇ Distribution of nitrogen stocks in the soil and plant system Based on a detailed database of soil properties, Batjes (1996) estimated global amounts of soil N to be 133–140 Pg N for the upper 100 cm of the soil profile. By comparison, about 10 Pg N is held in the plant biomass and about 2 Pg N in the microbial biomass (Davidson, 1994). This shows that on an ecosystem scale, soils are the main reservoir for N. We have compiled data for representative European terrestrial ecosystems on N pools and fluxes in representative

Klaus Butterbach-Bahl and Per Gundersen

Hyytiälä, Finland Pine Forest (Boreal)

N input

1.5

0

2.6

2

0

40

BNF 0

170

N-Fertils.

Höglwald, Germany Spruce Forest (Temperate)

N-Deposit.

40

Maulde, Belgium arable land, wheat (Temperate)

Plant N stock

175 Harvest

not considered

90

192

100

22

0 1?

0.3

9000

1570

ca. 550 (gross mineral.)

20

26

0

5-6

4-5

N-leaching ??

25 NH3

110 (net mineral)

? N2

4660

3.8 N2O+NO

Soil N Stock

1500

17

190

Plant N uptake

175

Plant N litter

not considered

??

<0.5

Figure 6.1 Nitrogen fluxes, pools and cycling at three different monitoring sites of the EU-funded integrated project NitroEurope. Although these sites are currently among the best investigated, uncertainties remain high and fluxes of, e.g. denitrification N2 losses or gross (net) rates of microbial turnover are scarcely measured if at all. BNF:€biological N2 fixation.

ecosystem types (supplementary material, Chapter 6). Using three well-studied sites selected from those listed in supplementary material (Chapter 6), the major differences in pools and fluxes between agricultural and forest systems as well as between N saturated and N limited forests can be seen (Figure 6.1). Our data (supplementary material, Chapter 6) show that the dominance of soil N as a reservoir (1350–9000€ kg N ha−1 on mineral soil) is more pronounced for agricultural systems, with greater than 90%–95% of Nr being stored in the soil as compared to forest systems, where N storage in biomass can be up to 30%–50% (Figure 6.1). Most of the N in soil, is bound to organic material (soil organic matter) and in natural systems only a very tiny portion is available as inorganic N in the form of NH4 or NO3: 1–10 kg N ha−1 in natural systems and 10–200 kg N ha−1 in fertilized systems (supplementary material, Section€ 6.6 and Figure€ 6.1). Even though N cycling is largely dependent on microbial processes (Section 6.4) the microbes only hold 40–1300 kg N ha −1, with the largest in fertilized arable systems (supplementary Â�material, Chapter 6).

6.3╇ Inputs of nitrogen into ecosystems 6.3.1╇ Input by biological nitrogen fixation In preindustrial times biological N2 fixation was the major pathway of Nr creation in terrestrial ecosystems and in more pristine regions this may still be the case (Cleveland et al., 1999). Even though dinitrogen (N2) is the major component of the earth’s atmosphere and nitrogen is essential for all forms of life, N2 cannot be used directly by biological systems to synthesize the chemicals required for growth, maintenance and reproduction. The general chemical reaction for the fixation of N (N2 + 3H2 → 2NH3) is identical for both the chemical (Haber– Bosch) and the biological processes. The triple bond of N must be broken€ – which requires substantial amounts of energy€ – and three atoms of hydrogen must be added to each of the N atoms. Living organisms use energy derived from the oxidation (‘burning’) of carbohydrates to reduce molecular nitrogen (N2) to ammonia (NH3). The capacity of biological N fixers to convert N2 to organic N is substantial, more than enough to maintain N pools in

101

Nitrogen processes in terrestrial ecosystems Table 6.1 Reported ranges for biological N2 fixation in natural and managed ecosystems.

N fixation rate kg N ha−1 yr−1

Source

Boreal forests and boreal woodland

1.5–2

Cleveland et al. (1999)

Temperate forests and forested floodplains

6.5–26.6

Cleveland et al. (1999)

Natural grasslands

2.3–3.1

Cleveland et al. (1999)

Mediterranean shrublands

1.5–3.5

Cleveland et al. (1999)

Common bean

30–50

Smil (1999)

Faba bean

80–120

Smil (1999)

Soya bean

60–100

Smil (1999)

Trifolium pratense and T. repens mixture

31–171

Carlsson and Huss-Danell (2003)

Medicago sativa with different grasses

65–319

Ecosystem type

Agricultural ecosystems

Managed grasslands

Crop

ecosystems and to replenish N losses. In many systems, N fixers drive the accumulation of fixed N on long time scales, bringing N supply close to equilibrium with other potentially limiting resources (Vitousek et al., 2002). Significant biological N2 fixation on ecosystem scale is most often associated with symbiotic N2 fixation, with the classic example of the Â�bacterium Rhizobium infecting the roots of leguminous plants (e.g. peas and beans, soya bean, clover, peanut). Leguminous symbiotic N2 fixation can be up to approximately 200 kg N ha−1 yr−1 (Table€ 6.1). Therefore, legumes are often used in agricultural systems, representing a major direct source of food and forage for livestock. In low-input and organic farming systems, leguminous crops are included in rotational designs in order to provide N for other crops within the rotation (Ball et al., 2005). Total global biological N fixation by agricultural crops is most likely in the range of 40–60 Tg N yr−1 (Herridge et al., 2008). Also in natural systems numerous genera of non-leguminous angiosperms, such as Alnus, Casuarina, Coriaria, Myrica, etc., exist that are capable of supporting symbiotic biological N2 fixation. These associations may achieve fixation rates as high as 100€kg N ha−1€y−1 and may occur as climax vegetation or as pioneer species in adverse soil environments. Another major pathway of biological N2 fixation is by heterotrophic bacteria in soils and sediments. Heterotrophic fixation during the decomposition of plant litter is thought to be important in terrestrial ecosystems and may account for 1–5 kg N ha−1 yr−1. The net contribution of heterotrophic N fixation to ecosystem N budgets may be greater in wetland soils. N budgets of flooded rice suggest that 50–100 kg N ha−1 yr−1 may be added by N fixation (Cassman and Harwood, 1995), and heterotrophic fixers contribute a substantial proportion of this total (Eskew et€al., 1981). Similarly, high rates of heterotrophic fixation may support plant production in some natural wetlands. While understanding of the symbiotic system in a few legume crop plants is relatively advanced, much less is known about N fixation in non-agricultural legumes or in other N-fixing

102

organisms, such as symbiotic cyanobacteria or free-living heterotrophic bacteria. In many ecosystems, the contribution of such organisms is assumed to be significant and needs to be accounted for. However, extreme difficulties in measuring rates of N2 fixation accurately at ecosystem scale have so far hampered a better understanding of the importance of biological N2 fixation for most terrestrial ecosystems.

6.3.2╇ Nitrogen input via atmospheric deposition Nitrogen is deposited from the atmosphere in both ‘wet’ (in precipitation) and ‘dry’ (as gases processes; and particles) for further information see Hertel et al. (2011) (Sections 9.4–9.6, this volume). In short vegetation, wet deposition constitutes the largest part of the atmospheric N input. Whereas forest canopies are efficient sinks for atmospheric gases and particles. Hence, N deposition on forests is up to a factor of 2–3 times larger than on open land (Fowler et al., 2004). As forests are the most exposed system, this chapter focuses on issues related to N input to forests. Deposition to forests ranges from less than 5 kg N ha−1 yr−1 in Northern Europe to greater than 60 kg N ha−1 yr−1 in Central and Western Europe (Dise et al., 2009) and the range for short vegetation is about half, i.e. 3–30 kg N ha−1 yr−1. At remote sites in the southern hemisphere wet deposition is less than 1 kg N ha−1 yr−1 (Vitousek et al., 1997), hence total pre-industrial inputs to forests were probably below 2 kg N ha−1 yr−1, assuming a small contribution from dry deposition. In large parts of Europe ammonium is the dominant form of atmospheric N input. Ammonium originates from ammonia emitted primarily from animal husbandry. Nitrate originates from N oxides emitted by fossil fuel burning, including automotive exhaust. The nitrate fraction of the atmospheric input is mainly below 10–15 kg N ha−1 y−1 and is relatively uniform across sites (Faegerli and Aas, 2008). Also there may be an additional wet deposition input of dissolved organic nitrogen (DON) with a magnitude of approximately 1 kg N ha−1 yr−1 (e.g. 0.5–1.1 kg N ha−1 yr−1 at the

Klaus Butterbach-Bahl and Per Gundersen Figure 6.2 Biological N2-fixing agents in agricultural and terrestrial natural systems (Herridge et€al., 2008).

Plant-associated •legume-rhizobia (symbiotic) •Azolla-cyanobacteria (symbiotic) •cereal-associative bacteria •cereal-endophytic bacteria Free-living •cyano bacteria •heterotrophic bacteria •autotrophic bacteria

Plant-associated •legume-rhizobia (symbiotic) •cereal-associative bacteria •cereal-endophytic bacteria Free-living •cyano bacteria •heterotrophic bacteria •autotrophic bacteria

Plant-associated •legume-rhizobia (symbiotic) •nonlegume-Frankia (symbiotic) •Azolla-cyanobacteria (symbiotic) •cycad-cyanobacteria (symbiotic) •cereal-associative bacteria •cereal-endophytic bacteria Free-living •cyano bacteria •heterotrophic bacteria •autotrophic bacteria

Höglwald Forest, Germany (Rothe et al., 2002), which is not yet accounted for in current inventories (Cape et€al., 2004)). Although European models of total N deposition may reproduce the general continental pattern in deposition and produce reliable regional total N deposition estimates (Faegerli and Aas, 2008), local variability in N deposition can be large and is not accounted for in these models (Dragosits et al., 2002). Particularly in fragmented and complex forest land with many edges towards open land the N deposition at the edges is several times higher than in the interior of the forest (Spangenberg and Kölling, 2004). This ‘edge effect’ is mainly caused by the increased turbulence created by the edge, and may extend 50–100 m into the forest. Larger ‘edge effects’ (extending up to 200 m into the forest) may be found close to local ammonia sources from animal farms (Spangenberg and Kölling, 2004). The sum of throughfall and stemflow N is often used as a proxy for total deposition of N to forests although this is most probably an underestimation due to retention by leaf and bark processes (Lovett and Lindberg 1993; Kreutzer et al., 2009) and retention by microorganisms on canopy surfaces. In forest, most of the evaluations of N budgets in Europe are based on N fluxes in throughfall. Thus much of our knowledge on N deposition and its effects in forests relies on the assumption that throughfall (+ stemflow) represents the total N input. Often only the inorganic N components of the throughfall are considered since the sources of organic N are not well known. In many cases the contribution from stemflow is also ignored even in broadleaf stands where the contribution may be significant (Rothe et al., 2002). When inorganic N in wet deposition is compared to canopy throughfall at the same site, a net uptake in the canopy is indicated at wet deposition rates below 5–10 kg N ha−1 yr−1 (Kristensen et al., 2004). This may be due to ion exchange

reactions in the canopy and uptake by epiphytes in the canopy. These uptakes may also occur at higher deposition rates, but only a few investigations have quantified the uptake. Kreutzer et al. (2009) estimated an uptake of 2–3 kg N ha−1 yr−1 at the Höglwald Forest site, Germany, which included also the potential uptake of N gasses directly through stomata (Figure 6.1). Assessments based on canopy budget models give somewhat higher uptake rates (Staelens et al., 2008), but these models do not include potential transformations to organic N and do rely on assumptions that need further testing. It is however, uncertain how much of the uptake actually ends up inside the leaf. In the canopy, microbes, lichens and other epiphytes may sequester N from wet and dry deposition. Also decomposing leaves, twigs and branches with high C:N ratio may sequester N. In a large-scale 3-yr experiment adding 18 kg N ha−1 yr−1 to the canopy in mist using a helicopter, Gaige et al. (2007) found up to 70% retention in the canopy, but part of it may have been assimÂ� ilated by epiphytes or microorganisms on the canopy surface. A detailed analysis of the fate of the N addition (including a 15N trace) in this N-deficient forest showed that only 1%–3% of the added N was recovered in the bole wood and a similar amount in the foliage (Dail et al., 2009). Conversely, 25%–56% of the N addition was recovered in bark and branches, but unfortunately it was not possible, with the methods used, to separate out how much of this was localized in tree tissue or in mosses, lichens and bacteria living on the bark (Dail et al., 2009). Thus bark and epiphytes may play an important role in canopy N retention. Besides retaining deposition N, the microbes and lichens may take up ammonium and nitrate and release organic N. Furthermore, nitrification of ammonium to nitrate has been observed in the canopy (Papen et al., 2002). The N retained in epiphytes may be counted in the litterfall N flux as part of the foliage, bark and other organic

103

Nitrogen processes in terrestrial ecosystems

material. However at the Solling site, a forest in Northern Germany, throughfall was collected, stored and filtered to measure the flux of elements from suspended matter in the throughfall. It appeared that this material contributed a considerable fraction of the nutrient input to the soil: for N, 10–14 kg ha−1 yr−1 (Gundersen et al., 1998c). The nutrient composition of the material was comparable to that found in the very fine litter particles with C:N ratios of 11–14 (g g−1) but differed from that in needle litter (C:N 38). This suspended matter input to the soil is not included in throughfall sampling (that is usually filtered) nor in conventional litterfall sampling that collects material above a certain mesh-size, typically 0.5 mm. To get a better quantification of the total deposition to forest and to more quantitatively know the potential underestimation of deposition N by the throughfall methods more studies (including 15N labelling) are needed.

6.3.3╇ Fertilizer and manure input into agricultural systems The agricultural use of N in the form of synthetic fertilizers has increased dramatically over the past 50 years (Erisman et€al., 2008; Galloway et al., 2004, see also Erisman et al., 2011, (Chapter 2 this volume) and Jensen et al., 2011 (Chapter 3 this volume)), in Europe and across the world. In 1950, global annual consumption of fertilizer N was less than 4 Mt, but had increased to 32€Mt by 1970 and to greater than 80 Mt by the 1990s (Roy and Hammond, 2004). Over the EU-27 as a whole, the main source of N input to agricultural land is now mineral fertilizer, with livestock manure a close second. Mineral fertilizer N use in Europe increased from 4.6 Mt in 1960 to 11.8 Mt by 1995 (van Egmond et al., 2002), when it accounted for 50% or more of total N input in Denmark, Germany, Greece, France, Luxembourg, Finland and Sweden. However, in Belgium and the Netherlands livestock manure was still responsible for more than 50% of N inputs (Vall and Vidal, 1999). In addition to fertilizer and manure N, a further 2.2 Mt is introduced annually into agroÂ�ecosystems through biological N fixation, by forage and grain legume crops (van Egmond et al., 2002). The average N application rate to European agricultural land (fertilizer plus manure) increased in parallel with the increased use of fertilizers, reaching an average of 123 kg N ha–1 by the 1990s (van Egmond et al., 2002; note that this study excluded the area of former Soviet Union). Since then, modest decreases have occurred in some countries, driven mainly by regulations to limit water pollution by leached nitrate; recent steep fertilizer price increases may well further reduce N use. Further information on fertilizer and manure use in agricultural systems and shifts in the use of different types of mineral fertilizers (e.g. increased use of urea) is provided in Jarvis et al. (2011) (Chapter 10, this volume).

6.4╇ Internal nitrogen cycling in terrestrial ecosystems 6.4.1╇ Nitrogen cycling processes

104

Nitrogen cycling in terrestrial ecosystems is characterized by a variety of N transformations involving both organic and inorganic (ammonium and nitrate) N species and uptake/immobilization of N by microbes and plants, as shown in Figure 6.3.

A growing temperate forest has a net N demand of approximately 5–10 kg N ha−1 yr−1 (Scarascia-Mugnozza et al., 2000), while N cycling due to fine root and leaf production and turnover varies mostly from 60–100 kg N ha−1 yr−1 (Kreutzer et€al., 2009). Compared to this, microbial N turnover via mineralization, nitrification and immobilization may be up to a magnitude higher (Corré et al., 2003; Kreutzer et al., 2009). Losses to the environment in form of leaching or gaseous volatilization are€– under N mass balance considerations€– mostly small and in a range of a few kg per hectare. Litter production and decomposition (and the balance between these two processes) are the major drivers of storage and release of Nr from the soil organic matter pool. During the decomposition process soil organic matter is cleaved from large polymers to largely bio-available monomers which are accessible to both plants and microbes. Microbes can further degrade these organic monomers to form ammonium (ammonification or N mineralization). Ammonium as well as organic N can also be oxidized to nitrate (nitrification). Both ammonium and nitrate can either be taken up by plants or immobilized by microorganisms. Thus, plants and microorganisms might compete for both bio-available organic N compounds (e. g. amino acids) and mineral N under N-limited conditions. Furthermore, N cycling in terrestrial ecosystems is characterized by internal loops, which occur at extremely variable timescales, from seconds to decades and can be mediated by both microorganisms, soil fauna (grazing of microorganisms) or plants (e.g. uptake of N and subsequent death of plants or root exudation). All mechanisms of internal N cycling favour ecosystem N retention, and thus are of crucial ecological significance (Corré et al., 2003). The common view of the terrestrial N cycle has undergone a considerable change in the past two decades (for detailed reviews see Schimel and Bennett, 2004; Chapman et al., 2006:€Jackson et€al., 2008; Rennenberg et al., 2009). Until the 1990s the perception and understanding of N cycling in terrestrial ecosystems was dominated by the paradigms that (1) N mineralization is the limiting step in N cycling, (2) plants take up only inorganic N, (3) plants poorly compete for N against microbes and therefore use only the N which is ‘left over’ by microbes (Schimel and Bennett, 2004). This perception led to the definition of net N mineralization as the sum of N which is exceeding the microbial demand and, thus, is available for plant uptake and pathways of N losses from the ecosystem. Consequently, net rate assays like the buried bag technique (Eno et al., 1960) have been widely used in order to measure plant available N. Since the late 1980s, researchers have increasingly become aware that net rates may be a poor approximation of the real status of N cycling in soils (Davidson et al., 1991; Hart et al., 1994), especially with regard to losses such as N trace gas emissions. Abundant published studies utilizing the 15N pool dilution technique for the determination of gross rates of ammonification, nitrification and inorganic N immobilization highlighted the limitations of net rate assays and provided a new and more advanced insight into actual N turnover. The 15 N pool dilution studies mostly revealed significantly higher rates of gross ammonification and nitrification while little or no ammonium and nitrate accumulated (Davidson et al., 1992; Hart et al., 1994; Neill et al., 1999; Verchot et al., 2001; Ross

Klaus Butterbach-Bahl and Per Gundersen

death of plant roots, leaves and branches

N2

root exudation

plant N pools dead SOM polymers, e.g. proteins

depolymerization

denitrification

ammonification

monomers, e.g. amino acids

microbes

death of microbes Micro-/meso-faunal grazing

microbial immobilization

NH4+

NO3-

nitrification

leaching dinitrogen fixation

Figure 6.3 Scheme of major ecosystems processes of N cycling, including internal N retention pathways. Dashed lines:€plant processes; solid lines:€microbial processes; red dashed and solid lines:€competitive processes between plants and microorganisms. Blue lines:€hydrological transport pathways. SOM:€soil organic matter. Taken from Rennenberg et al. (2009).

et€al., 2004). This illustrated the complex dynamics of microbial production and consumption of inorganic N which cannot be evaluated by net rate assays. Since the 1990s, many studies have shown that in quite a broad range of ecosystems, including arctic tundra, boreal and temperate forests, desert ecosystems and low productivity grasslands, plants are able to take up organic N, e.g. amino acids (Näsholm et al., 2009; Rennenberg et al., 2009). Further studies suggested that mycorrhyzal fungi can claim an important role in the acquisition of organic N by mycorrhyzal plants (see review by Meyer et al., 2009), with approximately 90% of all plants supporting mycorrhyzation. The increasing awareness of the role of DON in plant nutrition in natural ecosystems such as heathlands or forests and the surprisingly competitive strength of plants also led to a shift in paradigms concerning N mineralization. Compared to the old paradigm, which was centred around a simple Nr mineralization term (ammonification of organic N) as the ratelimiting step, the new, more holistic definition of N mineralization takes into account two steps:€(1) the depolymerization of organic macromolecules to bioavailable DON and (2) mineralization/ammonification of bioavailable DON to ammonium. The new paradigm of terrestrial N cycling leaves the core concept unchallenged, i.e. that microorganisms are responsible for breaking down complex organic material into plant-available form. However, it widens the definition of Nr mineralization into the steps of depolymerization into monomers and subsequent ammonification to ammonium. The new view on Nr cycling is centred around depolymerization as the rate-limiting step (Figure 6.3).

A summary of the synthetic analysis of 15N pool dilution studies of Booth et al. (2005) showed that soil organic matter quality€– and here the C:N ratio€ – is a major tool for understanding differences between ecosystem types (e.g. at the same soil organic matter concentrations, grassland soils with relatively low C:N ratios show high N mineralization rates, while forest soils with relatively high C:N ratios showed lower N mineralization rates) and that N mineralization in agricultural or other soils is not necessarily stimulated by fertilization. The study also showed that soil C losses due to agricultural practices, changes in vegetation, or climatic factors, may increase the ability of nitrifiers to compete with heterotrophic microorganisms for NH4, which in consequence will lead to higher N losses due to leaching or volatilization. Thus, a key role in potential N retention and N loss from terrestrial ecosystems has been attributed to the balance of inorganic N immobilization by microbes and (autotrophic) nitrification (Tietema and Wessel, 1992; Stockdale et al., 2002). Subsequently the ratio between nitrification and immobilization and the ratios of nitrification to N mineralization (relative nitrification) and ammonium immobilization to N mineralization (relative ammonium immobilization) have recently been used as indices for potential N loss or the N retention capacity of ecosystems (Stockdale et al., 2002; Accoe et al., 2004; Dannenmann et al., 2006). However, experimental assays for the determination of gross rates of N turnover are mostly based on root-free soil and thus ignore plant–microbe interactions. Therefore, it remains unclear to which extent the understanding of major controls of the terrestrial N cycling is transferable to actual N cycling in ecosystems. It was, for instance, indicated that dissimilatory

105

Nitrogen processes in terrestrial ecosystems

nitrate reduction to ammonium (DNRA, see later in this section) may also be an important process involved in terrestrial ecosystem N cycling (Silver et al., 2001). Please note that in the context of ecosystem Nr cycling denitrification is not considered, but that it is discussed later as a pathway of Nr loss.

Decomposition and mineralization Litter production and decomposition are the major drivers of terrestrial N turnover. Nitrogen enters the soil organic matter pool after internal N cycling via the plants (Figure 6.3). Depolymerization of organic matter is carried out by extracellular enzymes of fungi and bacteria (Jackson et al., 2008). The bioavailable N resulting from the cleavage of macromolecules can be used either by plants or microorganisms. Microorganisms can further convert bioavailable DON into ammonium (N mineralization or ammonification). Ammonification is performed by unspecific heterotrophic microorganisms both under aerobic and anaerobic conditions (Jarvis et al., 1996). Depolymerizing and ammonifying heterotrophic microbes are supposed to be C-limited, and thus should be negatively affected by very low C:N ratios. However, this might be compensated by root exudation, root turnover and mycorrhyzal turnover adding both labile N and C sources to the soil (Bais et al., 2006; Hogberg and Read, 2006). These root-derived C sources are supposed to claim a key role in C substrate supply to N mineralizing microbes. The strong dependence of depolym�erizing and ammonifying heterotrophic microorganisms on labile C sources illustrates the tight coupling of the C and N cycles.

Nitrification Nitrification is the biological oxidation of ammonium (NH4+) or ammonia (NH3) via hydroxylamine to nitrite (NO2−) or nitrate (NO3−). Ecologists have identified nitrification to be a key process in ecosystem N cycling (Figure 6.4) in view of

Nitrate ammonification

N2O

its relevance for ecosystem nutrient loss, atmospheric chemistry (with N2O losses during nitrification being in the range of 0.1–10‰; Breuer et al., 2002; Ingwersen et al., 1999), streamwater quality and soil acidification. Nitrification of ammonium increases the probability that the converted Nr is lost from the ecosystems, since the end-products nitrite/nitrate are susceptible to losses by leaching along hydrological pathways or further reduction to gaseous NO, N2O and N2 via denitrification (Figure 6.4). On the other hand the likelihood that nitrified Nr will be immobilized by microbes or taken up by plants is somewhat lower in comparison to that of ammonium, since N can be only incorporated into the cell walls in reduced form (Glass et€al., 2002). Furthermore, nitrate is more susceptible to leaching, since the anion exchange capacity of soils is markedly lower than that for cations, thus, promoting retention of ammonium rather than of nitrate in the soil matrix. Both the high probability of loss along hydrological as well as gaseous pathways and the unfavourable energetics of consumption by plants and microbes make nitrate a critical N species with regard to Nr loss from terrestrial ecosystems. Increasing nitrification has opened the N cycle of many terrestrial ecosystems towards increased N losses. Nitrification can be performed both along heterotrophic and autotrophic pathways. Autotrophic nitrification is performed in two steps by different groups of microorganisms (Costa et al., 2006; Wrage et al., 2001). In a first step, an ammonia oxidizer oxidizes ammonium or ammonia to hydroxylamine (NH2OH) and further to nitrite (NO2−). Nitrosomonas europaea is the best known autotrophic ammonia oxidizer. In a second step a nitrite-oxidizer (e.g. Nitrobacter) convert nitrite to nitrate. While autotrophic nitrifiers use the oxidation of NH4+ or NO2− as an energy source for carbon dioxide (CO2) fixation, heterotrophic nitrifiers use organic Nr both as an energy source and a C-source. Heterotrophic nitrifiers may be able to oxidize both ammonium and organic N compounds. The enzymatics Figure 6.4 Nitrification is a central process for the formation of several N forms in the soil. Microbial sources of N2O in soils are highlighted (Baggs, 2008).

+

NH4

NO2–

NO

N2O

N2

Denitrification

NO3–

Nitrification

N2O NH3

106

NH2OH

Nitrifier denitrification

NO2–

NO

N2O

?

N2

Klaus Butterbach-Bahl and Per Gundersen

of heterotrophic nitrification may€– despite the intermediates and products being the same€– differ from autotrophic nitrification (Wrage et al., 2001; Conrad, 2002). Detailed reviews of nitrification pathways involving enzymes and their regulation were provided by Wrage et al. (2001) and Conrad (2002). Nitrification is in general an aerobic process. Consequently, the oxygen availability in the soil exerts a major influence (Conrad, 2002). The optimum soil water content for nitrification is€– depending on the soil texture€– in the range of approximately 30%–60% water-filled pore space (Bouwman, 1998). However, autotrophic nitrification is affected by soil pH, the optimum pH for nitrification is thought to be at pH 5.5–6.5 (Machefert et al., 2002). Another potential environmental control on nitrification is soil temperature (Machefert et al., 2002), although nitrification can also take place in frozen soil (Freppaz et al., 2007). Plants may influence nitrification via consumption of ammonium, thus removing the substrate for nitrification. However, there might also be a direct effect of plants on nitrification activity via toxic compounds in root exudates (Castaldi et al., 2009). Root exudates of some tropical pasture grasses contain nitrification inhibitors that block both the monooxygenase and the hydroxylamine-oxidoreductase pathways in Nitrosomonas (Subbarao et al., 2006).

Microbial immobilization of N in the soil Nitrogen is crucial for the growth and activity of heterotrophic soil microorganisms. Various organic and inorganic (ammonium, nitrate) N species can be taken up by both bacteria and fungi. Ammonium immobilization appears to be largely related to and controlled by gross ammonification and available C. Heterotrophic microorganisms in the soil prefer ammonium over nitrate (Rice and Tiedje, 1989; Recous et al., 1990; Jansson et al., 1955) due to energy costs (Tiedje et al., 1982). Consequently a meta-analysis of available literature revealed a distinct, negative relationship between ammonification and the ratio of nitrate assimilation to total assimilation (Booth et al., 2005). Nevertheless, microbial nitrate immobilization is a significant process in a wide range of terrestrial ecosystems (Booth et al., 2005) especially in relatively undisturbed soils of semi-natural ecosystems (Dannenmann et al., 2006; Myrold and Posavatz, 2007), but also exceptionally in agricultural soils (Burger and Jackson, 2003). Nitrate assimilation has been shown to be dependent on available C (Azam et al., 1988; Trinsoutrot et al., 2000; Nishio et al., 2001; Booth et al., 2005), since C fuels the growth and activity of heterotrophic microorganisms. Additionally, a meta-analysis of 15N-pool-dilution studies also revealed a dependence of nitrate immobilization on microbial biomass N and gross nitrification as the process supplying the substrate (Booth et al., 2005). This might be explained by microsite variability (Sexstone et al., 1985; Parkin, 1987; Schimel and Bennett, 2004) in the heterogeneous soil system which is likely to include microsites with high available C and low NH4+, where NO3– will readily be assimilated by microorganisms (Schimel and Bennett, 2004; Booth et al., 2005). This

‘microsite hypothesis’ is confirmed by the observation that nitrate immobilization is larger in undisturbed soils than in physically disrupted soils (Booth et al., 2005). Furthermore, nitrate immobilization has been shown to be especially large close to decomposing residues high in C (Cliff et al., 2002; Moritsuka et al., 2004).

Plant uptake The acquisition of resources depends both on the phenology of the plants and on the supply of nutrients. Mobile resources including water and N may have a strong seasonal pattern of availability in many ecosystems. Plants tend to take up most of their N during vegetative growth (until flowering). However, distinct uptake patterns may be observed according to plant functional groups (annuals versus perennials) or even according to species. Co-existing plant species tend to decrease their competition for a limiting resource by taking up N at distinct times or soil depths, or using distinct N sources in a way that the composition of the plant community may be related to partitioning of differentially available forms of a single limiting resource. Plants predominantly take up inorganic N forms and here mostly NH4+ in presence to NO3− to cover their N demand for growth. However, it was shown recently that in N-poor and cold ecosystems amino acids and other organic monomers are readily taken up by plants as well, thus, plant N uptake may by-pass microbial N mineralization. However, when N is not strongly limiting as in agricultural systems or N-depositionaffected forests, inorganic N remains the major nutrient for plants (Harrison et al., 2007; Jackson et al., 2008). Plants take up N via transport systems in the plasma membrane of root cells. According to the growth requirements, N uptake is regulated by root system architecture and mechanisms that regulate the activity of the transport systems. High-affinity transport systems regulate N uptake at soil DIN concentrations between 1 μmol l−1 and 1 mmol l−1, while low affinity transport systems become significantly active above concentrations of approx. 0.5 mmol (Jackson et al., 2008). Plant N uptake can be mediated by mycorrhyzal symbiosis even though the magnitude and ecological significance of this is still a matter of debate (Meyer et€al., 2009). Nitrogen acquisition by plants depends much on the plant species and on the respective growth potential but also on Â�abiotic factors such as temperature (which affects root activity as well as N availability), soil pH and competition by microbes (Jackson et al., 2008). Soil pH is one of the factors that can influence the relative proportion of nitrate and ammonium in the soil. When both inorganic forms are present at similar concentrations most plant species take up ammonium preferentially. However, many plants are sensitive to high ammonium concentrations, showing ‘ammonium syndrome’. The toxicity of ammonium nutrition is universal, although the threshold at which it is observed is species-dependent. As a general rule fast growing species are more ammonium-sensitive than slowgrowing species. Ammonium toxicity has been associated with

107

Nitrogen processes in terrestrial ecosystems

ionic unbalance, cytoplasmatic pH changes and futile cycles (Britto and Kronzucker, 2002). For N-saturated spruce forests, it has been shown that increased concentrations of NH4+ can inhibit NO3− uptake, which is most likely due to the energy dependence of nitrate uptake (Jackson et al., 2008). Thus, plant nitrate immobilization ceases, followed by high rates of nitrate leaching. The actual uptake of N by plants is regulated by internal factors such as C and N metabolites and external factors like ammonium, nitrate and organic N compound concentrations in soil, light, temperature and soil pH.

Dissimilatory nitrate reduction to ammonium In addition to nitrate assimilation, dissimilatory nitrate reduction to ammonium (DNRA) represents a second microbial pathway of ecosystem nitrate retention. Dissimilatory nitrate reduction to ammonium is an anaerobic process, catalyzed by fermentative bacteria and reducing NO3− via nitrite (NO2−) to ammonium (NH4+). DNRA has been shown to occur predominantly in anaerobic sludge and sediments (Tiedje et al., 1982; Ambus et al., 1992; Bonin, 1996; Nijburg et al., 1997). By providing NH4+ for plant uptake and microbial immobilization and by reducing the size of the NO3− pool available for denitrification and nitrate leaching, DNRA has the potential to play an important role in ecosystem N retention (Silver et al., 2001). Both DNRA and denitrification appear to be favoured by similar soil conditions (low redox potential, high NO3− and labile C availability). The ratio of C to electron acceptors seems to control the partitioning of nitrate to denitrification and DNRA in such a way that DNRA is favoured when this ratio is high (Tiedje, 1988). DNRA has been recognized as a significant process mainly in wetland ecosystems. Silver et€al. (2001) found DNRA to be an important process in moist tropical forest soils with high clay content. DNRA might also be a significant process in temperate grassland soils (Müller et€ al., 2004, 2007). The significance of DNRA in aerobic upland soils however, remains largely unclear and urgently needs to be investigated.

6.4.2╇ Controls on nitrogen cycling Factors affecting nitrogen cycling Nitrogen cycling in ecosystems is strongly affected by environmental conditions such as climate, soil properties, vegetation type and management activities. Temperature and moisture are major controllers on temporal and spatial scales and microbial N cycling is high if Â�neither temperature nor moisture are limiting. As an example, it has been shown that mineralization as well as nitrification respond positively to increasing soil moisture as long as the soil is not saturated and, thus becomes anaerobic (Stark and Firestone, 1995; Breuer et al., 2002, Corré et al., 2003). The Nr mineralization has been shown to increase with rising temperature up to approximately 30 °C (Shaw and Harte, 2001), this also applies for nitrification. However, the temperature

108

response of nitrification will depend upon NH4 availability. As assimilation of NH4 also increases with temperature and even faster than mineralization (Binkley et al., 1994), nitrification can be substrate-limited at higher temperatures (Binkley et€al., 1994). Soil properties such as texture and clay content can affect N turnover in soils in several ways. On the one hand fine textured soils have a higher water holding capacity than coarse (sandy) textured soils and tend to have higher soil organic carbon concentrations. At the same precipitation level fine textured soils will show higher soil moisture values and become more often and more intensively anaerobic after intensive rainfall events. In consequence, gaseous N losses and (in this case especially important from an environmental point of view), N2O and NO emissions are higher from finetextured soils due to the more frequent stimulation of the predominantly anaerobic process of denitrification. On the other hand, the reduced ion exchange capacity and the improved soil drainage of sandy soils promotes nitrate leaching. Other important factors affecting N cycling are soil organic carbon content (SOC), C:N ratio and soil pH. The pace of N cycling has been found to be closely linked to increasing SOC (Li et€al., 2005). Moreover, the soil C:N ratio has been identified in several studies as a major indicator for leaching and gaseous N losses on the ecosystem scale. For example studies by Gundersen et al. (1998a) and Klemedtsson et al. (2005) show that at C:N ratios of less than 20–25 N2O losses and nitrate leaching increase dramatically from the forest ecosystem whereas losses are usually negligible from soils above this C:N (see later). It is widely accepted that soil pH has a significant effect on soil Nr cycling and the associated production and emission of N trace gases from soils, as it influences the three most important processes that generate N2O and NO:€ nitrification, denitrification and DNRA (Stevens et al., 1998; Kesik et al., 2006). N2O as well as NO emissions have shown the tendency to increase under low soil pH conditions. This is most likely due to a combination of factors including increased N trace gas ‘leakage’ rates during nitrification and denitrification, shifts in microbial communities or contribution of chemo-denitrification to N trace gas production for ecosystems with soil pH values of less than 4.0 (Yamulki et al., 1997; Ormecci et€al., 1999; Kesik et al., 2006). Vegetation type also exerts a major effect on ecosystem N cycling. For example litter quality is a major controlling factor for mineralization. Furthermore, other vegetation parameters such as canopy structure, leaf geometry or root distribution have been reported to affect N trace gas exchange, soil nitrate leaching and rates of microbial N turnover via effects on soil aeration, soil moisture distribution or substrate availability (Brumme et al., 1999; Butterbach-Bahl et al., 2002; Rothe et al., 2002). Bearing in mind the still incomplete list of factors affecting ecosystem N cycling, it becomes obvious that human management is of outstanding importance and for many ecosystems the crucial factor. Drainage, fertilization, irrigation, tillage practices, amendment of soils with manure or lime, crop rotations, soil compaction, grazing or human-induced N deposition

Klaus Butterbach-Bahl and Per Gundersen

are altering ecosystem state variables at a multitude of temporal and spatial scales via effects on substrate availability, litter quality, aeration, microbial community composition, soil moisture and even temperature profiles.

Effects of land use and management on nitrogen cycling Land-use changes have a tremendous effect on N cycling. Conversion of natural land into arable land is not only characÂ� terized by losses of ecosystem C stocks, but is mostly also accompanied by significant losses of ecosystem N stocks along hydrological pathways, gaseous volatilization or through erosion (Tiessen et al., 1982; McLauchlan, 2006). Based on a review of reported data, Murty et al. (2002) concluded that the conversion of forest soils to agricultural use resulted in an average loss of 15% N. This is less than the average loss of SOC, indicating that the C:N ratio narrows when agriculture is introduced (Murty et al., 2002; McLauchlan, 2006). However, intensification of land use results not only in a loss of N stocks and a narrowing of C:N ratios, but mostly goes along with an increase in soil pH due to liming, compaction of the soil and an increase in soil temperature and moisture amplitudes due to the reduction of the aboveground biomass cover. Not taking into account other management strategies such as fertilization, irrigation or drainage, this already has major effects on ecosystem N cycling. Mineralization rates are mostly slowed down in temperate agricultural systems as compared to native grasslands or forests (Booth et al., 2005). However, the likeliness of nitrate leaching losses increases, since nitrification is stimulated by external N inputs (Watson and Mills, 1998) and denitrification is often C-limited. Recent studies also show that soil microbial communities involved in N cycling are adapting to different land uses, which is accompanied by changes in the composition of species and functional groups (Coloff et al., 2008). However, it remains mostly unclear if these changes are accompanied by losses in functionality. Agricultural management can also support an accumulation of N and recovery of N cycling. Introduction of leguminous plants in the crop rotation (Tiessen et al., 1982) or application of organic fertilizers (Griffin et al., 2005) can support an accretion of N stocks. In summary, there are several studies available which show that land use and land management are of utmost importance for N cycling and biosphere–hydrosphere–atmosphere exchange at site and regional scales and that for example, intensification of agriculture has resulted in dramatic changes in regional and global N cycling (Galloway et al., 2003; Gundersen et al., 2006; Coloff et al., 2008).

Effects of climate change on nitrogen cycling Predicted climate changes are likely to feedback on ecosystem N cycling and biosphere–hydrosphere–atmosphere exchange processes. Based on field manipulation experiments at four different shrubland sites in Denmark, UK, the Netherlands and Spain, Beier et al. (2008) found that N mineralization was relatively insensitive to temperature increases (approximately 1 K). The lack of direct connection between temperature and

N mineralization in this study suggests that changes in the N cycle appear to occur more slowly and to have little direct influence on the C cycle (Beier et al., 2008). Contrary results were reported by Rustad et al. (2006) on the basis of a metadata analysis for 32 research sites in North America, representing four broadly defined biomes, including high (latitude or altitude) tundra, low tundra, grassland, and forest. They found that experimental warming in the range of 0.3–6.0â•›K significantly stimulated rates of soil respiration (+20%) and N mineralization (+46%) as well as plant productivity (+19%). The differences between the analyses could be due to the confounding effect of soil water availability. Temperature stimulation effects can only be expected if moisture is not limiting. The study by French et€al. (2009) on climate change effects on agricultural soils provides some indication which microbialÂ�mediated processes that play an important role in the N cycle may also be influenced as a result of climate change. Modelling studies dealing with climate change effects on N trace gas exchange or nitrate leaching of forest ecosystems in North America and Europe indicate that the stimulation of microbial N turnover in soils with climate change may be accompanied by increased losses of nitrate and increased emissions of NO rather than of N2O (Kesik et al., 2006a, b; Campbell et al., 2009), mostly due to large increases in N mineralization and nitrification.

6.4.3╇ Spatial and temporal variability of nitrogen cycling Describing N cycling on spatial and temporal scales remains a challenge, much of which is associated with the fact that small areas (hotspots) and brief periods (hot moments) often account for a high percentage of N turnover and in these cases especially denitrification activity and N losses (Groffman et€al., 2009). While the knowledge on factors affecting N cycling such as substrate availability, temperature, oxygen or moisture is pretty well established (see above), the understanding of how these factors are interacting on spatial and temporal scales remains low. The best known are, for example, emission pulses of N trace gases following re-wetting of dry soil in e.g., Mediterranean and subtropical climates (Davidson et€al., 1993), fertilization or irrigation/drainage events (Dobbie et€al., 1999) or freeze-thaw-related N2O emissions from agricultural or natural soils (Christensen and Tiedje, 1990; Wolf et al., 2010). Such pulse emissions, lasting often only for a few days, may contribute up to 80% of total annual emissions. Also the understanding that certain components of landscapes, for example riparian zones, could be viewed as hotspots of denitrification and zones of intensive N exchange between biosphere–hydrosphere and atmosphere (McClain et al., 2003; Groffman et al., 2009) is still increasing. A better characterization of N cycling processes and quantification of N exchange across spheres therefore requires research at landscape scales and a better understanding of how biological, climatic, soil and hydrological factors converge to create hotspots and hot moments at these scales (Groffman et al., 2009).

109

Nitrogen processes in terrestrial ecosystems

6.5╇ Outputs of nitrogen from ecosystems Several natural processes remove N from terrestrial ecosystem, which together with low inputs is the reason why N limitation may persist in many natural ecosystems (Vitousek and Howarth, 1991). The most important pathways are volatilization in fires, denitrification and N leaching. Moreover, erosion processes can contribute to N limitations and redistribution of nutrients. These N outputs may be modified by exploitation and management of the ecosystem. In the following we discuss the natural removal processes and how these are changed by human activities.

6.5.1╇ Fire Wildfires have coexisted with human activities and have shaped the landscapes of the Mediterranean but also those of temperate climates. Although fire has always existed as a natural phenomenon, the use of fire for activities such as grazing, agriculture, and hunting has significantly modified fire regimes, primarily in the Mediterranean region. More recently, the increase in population density in Europe, and the extensive use of natural and forest regions for recreation has increased the number of man-made fires (Barbosa et al., 2009). According to the EU Fire Database, almost 600 000 ha of forest land burn every year (2000–2005); of these c. 500 000 ha yr−1 occurred in Mediterranean countries (Miranda et al., 2009). Although the number of fires has steadily increased in the last decades, the total burned area has not increased in Europe. This is mostly due to a fire suppression policy and the efficiency of (and the large investment in) fire-fighting equipment (Miranda et al., 2009). Burning is a major pathway of N loss for ecosystems exposed to high fire frequencies. Depending on fire intensity (heat), 50%–99% of aboveground biomass N is lost as N2 (pyrodenitrification) or as NOx, N2O and NH3. Part of the litter layer N may also be lost, whereas the N pool in the mineral soil usually is unaffected (Alexis et al., 2007). Several uncertainties exist with regard to regional N mass balances but the largest are related to the characteristics of the fuel and fuel consumption. According to the CORINAIR emission inventory, forest fires cause 0.2% of NO2, and 1.3% of NH3 emissions, in Europe. Fire has significant short-term effects on soil N availability and therefore on water quality through soil heating effects and mid-term effects (1–3 years) on decomposition rates. The N deposition may also enhance nitrate runoff to stream waters one to three years after fire events (Johnson et al., 2007; Gimeno et al., 2009). Fire can have substantial long-term effects (decades to a century) on ecosystem C and N by causing changes in vegetation, often by facilitating the occupancy of the burned site with N-fixing vegetation (Wan et al., 2001).

6.5.2╇ Gaseous losses Total N losses by denitrification Denitrification is defined as the dissimilatory reduction of nitrate, nitrite, nitric oxide (NO) or nitrous oxide (N2O) to N2O

110

and N2 by microbes. The ability to carry out denitrification is not only found in bacteria, but also in some fungi and archaea (see review by Hayatsu et al., 2008; Zumft, 1997). In general, denitrifying enzyme activity occurs and is expressed under oxygen-limiting conditions, i.e. (if available) nitrogen oxides are used as an alternative electron acceptor. However, several bacteria have been shown to reduce nitrite or nitrate to gaseous nitrogen and a combined inhibitor and isotope tracer study by Bateman and Baggs (2005) suggests that aerobic denitrification occurred at 20% water-filled pore space. In addition to microbial denitrification, chemo-denitrification may also occur, i.e. in the presence of Fe2+ (formed through weathering of minerals) and an alkaline pH, nitrate can be chemically reduced (Van Cleemput, 1998; Samarkin et al., 2010). N2 can also be produced via the anammox process (anaerobic ammonium oxidation). This process, discovered in 1995 combines (under strict anaerobic conditions) ammonium and nitrite directly into N2 (Hayatsu et al., 2008). So far it has not been possible to show that this process is of significance in soils (instead only in bioreactors, wastewater plant and landfills). This is despite the fact that anammox sequences of the relevant bacteria (e.g. Candidatus Brocadia anammoxidans) have been detected in permafrost soil, agricultural soil and in samples associated with nitrophilic or nitrogen-fixing plants production (Humbert et al., 2010). It seems that the diversity of anammox bacteria is higher in terrestrial systems in comparison to marine systems, which is possibly a consequence of the larger variability of soil habitats and specific ecological requirements (Humbert et al., 2010). Microbial denitrification is the main pathway of Nr loss in terrestrial ecosystems. It is remarkable that the importance of denitrification is more constrained at regional and global scales as compared to the site scale. In an analysis involving major watersheds in the USA and Europe, Van Breemen et al. (2002) concluded that only 23% of Nr applied to the landscapes was found in riverine exports. Even if one accounts for changes in N storage due to land use change and biomass increment or export of Nr with food, feed and wood products this still leaves 37% of Nr left to be denitrified within the landscape. This estimate of the magnitude of landscape denitrification is well in-line with estimates on a global scale by Seitzinger et al. (2006), who estimated that approximately 40% of the global addition of 270 Tg Nr yr−1 to terrestrial ecosystems is removed via soil Â�denitrification. Based on a calculation of 15N:14N ratios, Houlton and Bai (2009) concluded that approx. 28 Tg of N2 are lost annually via denitrification to N2 in the terrestrial soil beneath natural vegetation, with an N2:N2O ratio ranging from 2.2 to 4.6 at the global scale. Estimates of denitrification rates for different ecosystem types vary largely in dependence of ecosystems type, soil properties and management (Barton et al., 1999; Hofstra and Bouwman, 2005), but also with regard to the method applied for the quantification of N losses via denitrification (Groffman et al., 2006; Hofstra and Bouwman, 2005). Based on a detailed literature review on denitrification in soils, Barton et al. (1999) showed that, on average, denitrification rates in agricultural

Klaus Butterbach-Bahl and Per Gundersen

The challenge to estimate denitrification losses at the site scale is closely linked with the difficulties in measuring denitrification rates in soils at a background N2 concentration of 78%. The commonly used C2H2 blockage technique, i.e. using C2H2 to block the further reduction of N2O to N2 during denitrification, has been shown to fail in the presence of O2 (Bollmann and Conrad, 1997), since C2H2 catalyses the conversion of NO (another intermediate of denitrification) to nitrite/nitrate which is then metabolized by soil microbial processes. Therefore, this method can lead to an underestimation of N losses. The most reliable methods for quantifying denitrification in soils are the use of stable isotope techniques or the soil core gas flow technique (Butterbach-Bahl et al., 2002). However, these methods also have severe drawbacks with regard to the experimental complexity, representativeness of soil cores, sensitivity of measurements and cost of experiments (Groffman et al., 2006).

Nitrous oxide Figure 6.5 Box-plots comparing annual denitrification rates in agricultural and forest soils. The centre vertical line marks the median, the edges of the box (hinges) mark first and third quartiles, and the central 50% of annual rates are within the range of the box. Vertical lines show the range of values that fall within 1.5 (midrange) of the hinges. Asterisks are values outside inner margins (1.5 (|median-hinge|)). (Figure taken from Barton et al., 1999.).

soils are one order of magnitude higher than in natural soils (Figure€6.5). They concluded that ‘most annual denitrification rates reported in the literature appear to be fairly low, with over half of the rates in forest soils being less than 1 kg N ha−1. yr−1 (with a mean of 1.9 kg N ha−1.yr−1). Rates of denitrification in agricultural soils tend to be higher than in forest soils, with 85% of rates reported being greater than 1 kg N ha−1.yr−1, and a mean rate of 13 kg N ha−1.yr−1.’ Numerous soil, site, and management factors have been reported (Barton et al., 1999; Groffman et€al., 2009) to affect the denitrification process in situ. The literature indicates that the highest rates of denitrification can be expected in N-fertilized soils, or where site management increases soil nitrate availability. Where nitrate is non-limiting, denitrification rates appear to be highest in irrigated loam soils. The review suggests that it is ‘difficult to predict denitrification rates based on our current understanding, and that plot studies should still be conducted if soil N balances are required’. Hofstra and Bouwman (2005) showed that soil-core-based estimates are a factor of two lower than those based on mass balance approaches. The situation becomes even more complex if the huge variability of denitrification across temporal and spatial scales is taken into account (Groffman et al., 2009). At the present site scale, estimates of denitrification are highly uncertain, despite more than eight decades of research on the process. It depicts partly our current lack of understanding of denitrification in soils and partly the problem of variability that will remain and will always induce large uncertainty at field scale.

From a mass balance perspective, global N2O losses from terrestrial ecosystems are small and in the magnitude of 10 Tg N€yr−1. However, N2O is important from the perspective of stratospheric ozone destruction and climate protection (ButterbachBahl et al., 2011; Chapter 19, this volume). On a global scale, the main sources of N2O are associated with soil emissions (e.g. Smith, 1997) and, more specifically, mainly with emissions from agricultural soils and tropical rainforest soils. Even though it is known that on a global scale soils are a major source for N2O (approximately 60% of all sources), there is still an ongoing debate regarding the source strength of individual ecosysÂ� tems for N2O, the potential of soils to function also as sinks for atmospheric N2O (Chapuis-Lardy et al., 2007) and about losses of N2O following Nr application/deposition. This is due to the variability of N2O emissions from soils on temporal and spatial scales, our shortcomings with measuring techniques (mainly chamber-based, with significant drawbacks with regard to spatial representation), the lack of continuous year-round measurements and the still limited understanding of microÂ� bial processes driving soil N2O formation. It is remarkable that, based on a thorough literature review, IPCC (2007) assumes that 1.0% of fertilizer N applied to fields is directly emitted in the form of N2O, while the top-down approach by inverting the global atmospheric N2O budget yielded loss rates of N2O from newly created Nr that are in the range of 3.5%–4.5% (Crutzen et al., 2008). A re-evaluation of the top-down approach of Crutzen et€al. (2008) by Davidson (2009) indicates that 2.5% of fertilizer N-production has been converted to N2O, either directly following fertilizer application or indirectly following Nr cascading through downwind/downstream ecosystems. This means that research is still needed to get a better understanding of direct and indirect N2O emissions following Nr input to terrestrial systems. Major uncertainties here are specifically the ratio of N2 to N2O emissions and how this ratio is affected by environmental conditions and involved microbial processes. Based on the work of Seitzinger et al. (2006), approximately 40% of the 270 Tg Nr that is brought annualy into terrestrial ecosystems is

111

Nitrogen processes in terrestrial ecosystems

denitrified in soils (100 Tg). Since N2O emissions from terrestrial ecosystems are in the magnitude of 10 Tg yr−1, the N2:N2O ratio should at least be on average 10 (assuming that all terrestrial emissions are emissions from soils). Good experimental background data exists which shows that the N2:N2O ratio is highly variable (Simek et al., 2002).

positive correlation between N deposition and the magnitude of NO emissions from coniferous forest soils, e.g. as shown by Pilegaard et al. (2006). NO is a highly reactive N trace gas and reacts quickly€– already partly within the canopy€– with O3 to form NO2. The NO2 can be readily taken up and metabolized by plants. Hence high soil NO emissions are often correlated with substantial NO2 re-deposition to plant and soil surfaces (Sparks, 2009). Soil NO emissions and re-deposition on and uptake of NO2 by plant leaves can be regarded as a potentially important process of nutrient dispersal (Butterbach-Bahl et€al., 2004; Sparks, 2009). In summary, little is known about the regulation of NO turnover in soils and ecosystems, and what is known is based on empirical observations but is not understood theoretically on the basis of microbial and plant metabolism (Conrad, 2002). To improve understanding there is a need for (a) continuous measurements (year-round) for understanding temporal variability and to improve loss estimates and (b) the study of microbial and plant processes involved in production and consumption of N trace gases in soils and ecosystems.

Nitric oxide Soil emissions of NO from heavily fertilized areas can reach the same magnitude as the anthropogenic NO release in urban areas (on a per area and time basis) (Ludwig et al., 2001) and annual loss rates have been reported to be as high as 12–52 kg N ha−1 yr−1 for fertilized (280 kg N ha−1 yr−1) and irrigated maize fields in southwestern France (see also Figure 6.6). Estimates of NO emissions from agricultural soils for EU-15 states are vary within a range of 49 to 190 kt NO-N for the year 2000, depending on the approach used (Butterbach-Bahl et al., 2009). For forests in the EU-15, soil NO emissions have been estimated to be 75 kt NO-N yr–1 (Kesik et al., 2006a, b). In total, NO emissions from soils in EU-15 have been calculated to be 4%–6% of annual anthropogenic NO/NO2 emissions (Butterbach-Bahl et€al., 2009). The drivers of soil NO emissions are in principle the same as for N2O, i.e. soil moisture and temperature, and N availability, but also soil properties (texture, soil organic carbon, pH, etc.) and management practices such as tillage and irrigation have been shown to affect the timing and magnitude of soil NO emissions (Ludwig et al., 2001). In contrast to N2O, the main microbial process responsible for NO production in soils seems to be nitrification rather than denitrification (Conrad, 2002). Thus, optimum moisture conditions in NO emissions from soils have often been found to be in the range of 50%–70% water-filled pore space. Since soil–atmosphere NO exchange is the product of simultaneously occurring production and consumption processes the role of denitrification for NO exchange is often associated with the uptake of NO produced from nitrification which is already within the soil matrix, i.e. NO consumption (Conrad, 2002). The close relationship between N availability and NO emissions is demonstrated by the close cultivated land

forest

Ammonia For agricultural systems and under certain environmental conditions (high pH environments, application of organic and inorganic fertilizers) NH3 losses are also very important from a mass balance point of view (Jarvis et al., 2011, Chapter 10 this volume). In the case of synthetic fertilizers, urea has the greatest potential for loss of NH3 to the atmosphere by volatilization, especially from alkaline soils. Major losses of NH3 to the atmosphere also arise from animal manures, with estimates of the fraction of their N content lost in this way ranging from 20% to 33% (Bouwman et al., 1997; IPCC, 1997). The calculated median NH3 loss from global application of synthetic N fertilizers (78 million tons N per year) and animal manure (33 million tons N per year) amount to 14% (10%–19%) and 23% (19%–29%), respectively (Bouwman et al., 2002). Across the EU-27, Oenema et al. (2009) estimated NH3 losses from agriculture in the year 2000 of 2873 Gg N. The principal measures to reduce these NH3 emissions include avoidance of excessive

grassland/woodland

others

tundra (2) deserts and semi-deserts (1)

min

swamps and marshes (3)

25% 10%

75%

median

90%

max

tropical fields/2* forests (17) tropical savanna/woodland (41) chaparral, thom forest (12) temperate grass/field (16) tropical montane (2) tropical deciduous (5) tropical evergreen (6) temperate, N affected (8) temperate (4) boreal (1) paddy rice (1) tropical (23) temperate (67) boreal (2) 0,01

112

0,05 0,1 0,5 1 mean annual NO emission

5 10 [kg N ha–1 yr–1]

50 100

Figure 6.6 Published data on NO emissions from soils from various ecosystem and land use types. (Figure taken from Ludwig et al., 2001). With kind permission from Springer Science+Business Media.

Klaus Butterbach-Bahl and Per Gundersen Table 6.2 Characteristics of coniferous forest ecosystems with low, intermediate and high N status (Gundersen et al., 2006). Nitrogen input is not a good indicator of N status, but the ranges given are typical for low, intermediate and high N status ecosystems

Nitrogen status Input (kg N ha yr ) –1

–1

Low N status (N-limited)

Intermediate

High N status (N-saturated)

0–15

15–40

40–100

Needle N% (in spruce)

< 1.4

1.4–1.7

1.7–2.5

C:N ratio (g C g N−1)

> 30

25–30

< 25

Soil N flux density proxy (litterfall + throughfall) (kg N ha–1 yr–1)

< 60

60–80

>80

Proportion of input leached (%)

<10

0–60

30–100

N in agricultural production systems, i.e. feeding animals and fertilizing fields and pastures according to requirements, covering manure storage and use of low emission application systems rather than surface spreading. The use of N fertilizers with high losses (such as ammonium bicarbonate and urea) is to be€avoided.

6.5.3╇ Leaching Ammonium and dissolved organic nitrogen The mobility of N in soils largely depends on the form of dissolved N (NH4+, NO3− or DON). Ammonium is absorbed on the soil cation exchange complex and is thus quite immobile in the soil profile. As a result, NH4+ concentrations are generally low in seepage water and very low in streams. Ammonium usually contributes less than 5% to the total dissolved N concentration in soil water except in extremely NH4+-loaded soils (Dise et al., 2009; De Vries et al., 2007). Concentrations of DON are below 0.6 mg N l−1 and often even below 0.1 mg N l−1 in both seepage water from well-aerated soils (Michalzik et al., 2001) and streams (Campbell et al., 2000; Perakis and Hedin, 2002). No relation to concentrations of dissolved inorganic N was found in these studies (Michalzik et al., 2001). In pristine forest streams, DON constitutes the dominant N leaching loss, in the order of 1–3 kg N ha−1y−1, since nitrate concentrations were very low in those streams (Campbell et al., 2000; Perakis and Hedin, 2002). A recent analysis by Brookshire and others (2007) showed that elevated atmospheric N deposition (spanning an input gradient of 5–45 kg N ha−1 yr−1) increased DON output from temperate forested watersheds in the Appalachian Mountains of the USA. However, over a similar gradient across 12 European sites no significant relation between throughfall N and forest floor DON leaching was found (Park and Matzner, 2006). Nitrogen addition experiments in the USA showed a consistent increase in DON effluxes following the increase of N input (Pregitzer et€al., 2004; McDowell et al., 2004) whereas no such effects were shown in European N addition experiments (Gundersen et al., 1998b; Raastad and Mulder, 1999; Sjöberg et€al., 2003).

Nitrate leaching and N status indicators Nitrate is the constituent in seepage and stream water that responds most to input changes or disturbances of the plant cover. Nitrate is highly mobile in soils and input and release

in excess of plant and microbial uptake requirements will be transported through the soil profile and leached. In consecutive compilations of input–output budgets for N in European forests nitrate leaching increased with increasing N input (MacDonald et al., 2002; De Vries et al., 2007; Dise et al., 2009). However, elevated nitrate leaching hardly ever occurs below a throughfall N deposition of 8–10 kg N ha−1 yr−1 and always occurs above 25 kg N ha−1 yr−1. Within this N input interval both full retention and no retention in forests can be found. This variability in the response to N deposition is to a large extent determined by the ‘N status’ or N availability of the system; N, poor systems have a high retention and N-rich systems have a low retention (Table€6.2). The type of N input (ammonium or nitrate) also influences retention. Those sites dominated by ammonium retain a higher proportion of the input (Emmett et al., 1998) probably due to absorption of ammonium on the soil cation exchange complex and subsequent biological processing. In agricultural soils, leaching losses can increase rapidly when rates of fertilizer N exceed plant N demand (Vinten et€al., 1994), but losses can also be influenced by rainfall, cultivation and soil management (Shepherd et al., 2001; Misselbrook et al., 1996). From N input manipulation experiments and analyses of available European datasets from coniferous forests, three classes of forest N status can be distinguished, based on N concentrations and fluxes (Table 6.2). The forest floor C:N ratio is a good indicator of N status, at least for coniferous forests, but a more broadly applicable indicator is the C:N ratio of the top mineral soil (Gundersen et al., 2009). A negative relationship between C:N ratio and nitrate leaching has been found and below a C:N ratio threshold of 25 all sites had elevated nitrate leaching (Gundersen et al., 1998a), even though other studies have found a less clear relationship (De Vries et al., 2007). MacDonald et al. (2002) found an average retention of 68% and 35% of the input at sites above and below a forest floor C:N ratio of 25, respectively. An important mechanism behind the shift of the balance between retention and leaching seems to be the onset of net nitrification at forest floor C:N ratios around 24–27 (or 1.4%–1.6 N in organic matter) (Aber et al., 2003; Kriebitzsch, 1978). This emphasizes the importance of soil C content on N retention, where C-rich soils may have large N retention potentials, especially if the C:N ratio is greater than 25. This threshold for N retention compares well with the

113

Nitrogen processes in terrestrial ecosystems

threshold (C:N >25) for growth response to N fertilization in conifers (Hyvönen et al., 2008). N content in needles correlates with the forest floor C:N ratio and is also a good indicator of N status (Kristensen et al., 2004). At N contents in needles below 1.4%, no nitrate leaching occurs and the system appears N-limited (Table 6.2), whereas above 1.4% N in needles leaching often occurs. This level corresponds surprisingly well with the threshold of 1.3%–1.4% N above which conifer stands show no growth response to N fertilizer additions (Sikström et al., 1998). In accordance with the definition of N saturation, nitrate leaching occurs if the soil flux density of mineral N (defined as N deposition plus net mineralization) exceeds the capacity of N uptake by plants. Net mineralization, which is a measurable parameter, includes (by definition) the microbial demand, i.e. the microbial immobilization. Datasets combining N deposition and net mineralization suggest a threshold in N flux density of 100 kg N ha−1 yr−1 for elevated nitrate leaching (Andersson et al., 2002; Fisk et al., 2002). Net mineralization is not routinely measured, but may be strongly correlated to the N flux with litterfall. Hence, total aboveground N input to the soil (throughfall N plus litterfall N) excluding belowground root litter input can be taken as a proxy for N flux density. Using this parameter the threshold is around 60 kg N ha−1 yr−1 for conifers (Table€6.2). Retention of deposited N in soils mainly occurs in the forest floor (Nadelhoffer et al., 1999). This retention may slowly change the organic matter quality and decrease the C:N ratio of forest floors with time. There is no documentation whether this has occurred, but an indication is given by decreasing C:N (increasing forest floor per cent N) with increasing N deposition (Aber et al., 2003; Kristensen et al., 2004). On the other hand, Moldan et al. (2006) found no change in C:N ratio after a decade of chronic N addition to a coniferous forest, but an increased accumulation of both C and N in the organic mor layer. The European soil inventory of Vanmechelen et al. (1997) showed that currently approximately 40% of the c. 4000 sites had forest floor C:N ratios below the threshold of 25, below which elevated nitrate leaching often occurs. With current N loads, many forest sites may move towards N saturation. At this condition, the ecosystem is very responsive to changes in N deposition (Gundersen et al., 1998b). Reductions in N deposition will, with only a short time delay, translate into a proportional decrease in nitrate leaching. This was shown in experiments where N inputs in throughfall water were removed by roofs. In these experiments, nitrate leaching was reduced and N content in older needles decreased, indicating reversibility of the N-saturated condition (Boxman et al., 1998a; Bredemeier et al., 1998).

6.6╇ Effects of nitrogen 6.6.1╇ Nitrogen enrichment and saturation Since N is considered the major limiting element in terrestrial ecosystems, increased N deposition would be expected to be retained in the ecosystem and stimulate growth. However, N

114

deposition in large areas in Europe exceeds the normal growth requirement of forests and other (semi-natural) ecosystems and they may thus over time become enriched in N. Central to our understanding of N enrichment and its effects is the concept of ‘nitrogen saturation’, that describes a series of temporal changes in ecosystem functioning in response to increased N input. Nitrogen saturation can be defined in several ways (Ågren and Bosatta, 1988; Aber et al., 1989). The most widely used definition is in conditions where ‘availability of mineral N may exceed the combined nutritional demands of plants and microbes’ (Aber et al., 1989) which then can be determined as elevated nitrate leaching from the rooting zone. In this situation, what also occurs is that N (derived from deposition) leaves terrestrial ecosystems and may potentially affect downstream ecosystems through the ‘N cascade’ (Galloway et€ al., 2003). Ågren and Bosatta (1988) have defined an N-saturated system as ‘an ecosystem where N losses approximate or exceed the inputs of N’ which implies an accumulation in the system close to zero. From a theoretical point of view, this may be the most proper use of the term ‘saturation’. However, this type of ‘true saturation’ may be seen as the end point of the N-saturation continuum where there is practically no biological control over N retention. Agricultural systems are often N-saturated according to the first definition, since it is usually economical to add more fertilizer N than can be taken up by the crop. On the other hand, they will never reach true saturation since a major fraction of the N input will be removed in harvested crops. The development of N saturation by increased N inputs involves a complex interaction of the processes in the N cycle (Aber et al., 1989, 1998). The progression from N limitation to N excess and the potential effects of N deposition may be explained as follows. In the first phase primary production increases, plants and microbes effectively absorb added N and the N content of plants increases. Retranslocation of N from senescent foliage (and roots) may decrease leading to higher N contents in litter materials and thus increased litterfall N flux. The internal cycling of N is accelerated through increases in litterfall N, net mineralization and tree N uptake. As N availability is increased, the composition of the forest floor vegetation may gradually change towards more nitrophilic species and other essential resources (P, K, Ca, Mg, or water) may at least periodically limit growth. In the accelerated N cycle, net nitrification becomes important and nitrate starts to appear in soil water. When elevated nitrate leaching becomes chronic, soil acidification resulting from N transformations becomes significant. Destabilization and potential forest decline from excess N deposition have been shown in case studies where the nutritional imbalance was important (Roelofs et al., 1985). Recent synthesis efforts support this general scheme although the understanding of processes and interactions has become more complex and detailed (Aber et al., 1998, 2003; Emmett et€al., 1998; Gundersen et al., 1998b). The timing of the changes in ecosystem N function at a certain N deposition load is still not well understood, nor is the cumulative load required to change the N status of a low-N ecosystem known.

Klaus Butterbach-Bahl and Per Gundersen

6.6.2╇ Biodiversity and vegetation change Nitrogen deposition has the potential to impact on biodiversity in a wide range of ways (see also Dise et al., 2011, Chapter€20, this volume). Aboveground impacts on vegetation include Â�direct toxicity to sensitive species (Britto and Kronzucker, 2002), reduced resistance to environmental stresses such as frost (Caporn et al., 2000), increased susceptibility to pests and disease (Brunstig and Heil, 1985) and soil-mediated effects of acidification and eutrophication (Stevens et al., 2006). The degree to which N deposition has either a positive or negative impact on an individual species depends on the tolerances and requirements of the respective species. At a community level, the impact is usually negative, resulting in a change in the species composition from that typically found at a given location. Additionally, declines in species richness have been widely reported in a number of vegetation communities, both from experimental studies (Bobbink et al., 1998; Suding et€al., 2005), studies using natural deposition gradients (Stevens et€al., 2004; Maskell et al., 2009) and studies of species change over time (Dupré et al., 2010). Changes in species composition of the vegetation are likely to have impacts on N cycling. Species may have differential abilities to use N or a preference for either nitrate or ammonium. Consequently, changing species composition may impact on the relative rates or form of N uptake (Britto and Kronzucker, 2002). Other feedbacks between plant community change and the N cycle may result from different concentrations of N stored in plant tissues, plant growth rates and life-span and ease of decomposition of dead materials, all resulting in changes in the availability of N or residence times in N pools.

6.6.3╇ Carbon cycle Effects on plant growth and aboveground C sequestration In agricultural systems, N additions from fertilizer usually lead to increased plant growth even though the system is N-saturated as defined above, but in this case other nutrients (P, K, and micronutrients) are added as well. When NPK fertilizers are used in forest ecosystems strong growth responses are usually observed there as well (Jarvis and Linder, 2000), but when N alone is added growth response is more modest and dependent on N status prior to fertilization (Hyvönen et al., 2008). Growth responses to repeated annual N fertilization in northern Europe decrease with soil C:N ratio, with an indicated threshold for a growth response at C:N ratio 25 (Hyvönen et al., 2008). Under high loads of N fertilization the growth response can be reversed, as other nutrients become limiting (Högberg et al., 2006). Experiments more closely simulating chronic N inputs from deposition (lower doses split in several additions over the growing season or the whole year) show quite variable growth responses:€ the European NITREX sites showed no growth response to N addition (Emmett et al., 1998), whereas a longterm experiment at four sites in Michigan, USA, did show an increase in woody biomass (500 kg C ha−1 yr−1) after adding 30 kg N ha−1 yr−1 over 10 years (Pregitzer et al., 2008). In an

Nr-enriched site, a growth increase after trees were relieved from excess N deposition has also been observed (Boxman et€ al., 1998a). The experiments simulating chronic N inputs were criticized for adding N to the soil, rather than to the canopy as happens with N deposition (Sievering, 1999). Recently it was shown that canopy uptake of N stimulated plant growth at a low-Â�deposition (3 kg N ha−1 yr−1 total) site in the USA (Sievering et€al., 2007) and a modelling analysis indicate that direct canopy N uptake, by-passing the competition for nutrients exerted by soil microbes, could have a significant effect on the sensitivity of tree growth (Dezi et al., 2009). However, a study by Dail et al. (2009) revealed a limited canopy N uptake (see Section 6.3.2). Thus the experimental evidence is somewhat inconclusive. Recent Â�multi-factorial analysis of forest growth at nearly 400 plots across Europe showed significant responses to N deposition at 1%–2% growth increase per kg N ha−1 yr−1 when analysed at stand level (Solberg et al., 2009) as well as at individual tree level (Laubhann et al., 2009). A similar analysis of deposition effects on forest C storage, using, North American data showed comparable results (Thomas et al., 2010). Stronger responses were seen for sites with high C:N ratios (Solberg et al., 2009) as also observed in the fertilizer experiments (Hyvönen et al., 2008). Forest growth has increased in Europe by more than 60% over the past 50 years (Ciais et al., 2008), resulting in a strong increase in C sequestration. N deposition has been suggested as the main driving factor for this increase in forest growth (Karjalainen et al., 2008). On the one hand, it has been argued that N only makes a minor contribution to this sink since most of the N is retained in soils at a C:N ratio of approx. 30 (Nadelhoffer et al., 1999). Recently, Magnani et al. (2007) suggested that a large fraction of ecosystem C sequestration potential in temperate and boreal forests could be attributed to the effects of atmospheric N deposition. This study has fuelled a new debate on the effect of elevated N deposition on C sequestration (de Vries et al., 2008, 2009; Sutton et al., 2008). Sutton et al. (2008) suggested that the proposed sensitivity of about 200 kg C per kg N (Magnani et al., 2008) could be an artefact resulting from the parallel effects of other environmental factors and should be reduced to 50–75:1. De Vries et al. (2009) compiled data from a range of different approaches (observational, experimental and modelling) to analyse the impact of N deposition on C sequestration. The results of the various studies were in close agreement with regard to the effect of N addition for aboveground C sequestration (i.e. 15–30 kg C:1 kg N), but more variable for soil values (i.e. 5–35 kg C:1 kg N). All together these data indicate a total C sequestration range of 30–70 kg C per kg N deposition. The discrepancy in estimated C–N sensitivity of forest ecosystems between different studies could stem from several sources of error. On the one hand, studies based on regional N deposition datasets (Magnani et al., 2007) ignore the fact that higher dry N deposition rates are generally observed over forest canopies, as a result of their greater roughness. By underÂ�estimating N deposition, they would overestimate the response to a unit dose of added N. On the other hand, ecosystem manipulation studies could also be affected by artefacts, as they neglect the potentially important role of canopy N uptake and often apply

115

Nitrogen processes in terrestrial ecosystems

doses (up to 100 kg N ha−1 yr−1) well in excess of natural N deposition over most forest ecosystems (Dise et al., 2009). Whatever the sensitivity, the question is whether this apparent C sequestration from N deposition will be sustained in the long-term if forest ecosystems become N-saturated. At true N saturation (i.e. no or very low N accumulation) the ecosystem may also be C-saturated (no C sequestration) at least in the soil compartment (trees in managed systems may still grow and sequester N, but with no further response to N). The long-term fertilizer experiments in Sweden (Hyvönen et al., 2008) show that C sequestration per unit of added N decreases with dose (i.e. the response levels of above c. 30 kg N ha−1 yr−1). It remains to be seen if the same is true for increasing N deposition or for cumulative loads. An important question could be if there is an optimum level of N deposition for C sequestration or if a critical threshold exists for the effect of N on C sequestration.

Effect on soil C processes and C sequestration Although there is observational as well as experimental evidence for increased wood production, there is no evidence for an increase in leaf litter production due to elevated N deposition. Despite the relatively unsophisticated sampling of this ecosystem flux, it has not been part of the European forest monitoring programme and no data compilation seems to be available which could allow an analysis of the effect of N on leaf litter fluxes. N addition experiments show no response in leaf litter mass but instant increases in N concentrations of litter and thereby in the aboveground litterfall N flux (Gundersen et al., 1998b; Pregitzer et al., 2008). As the litterfall N flux increases, N mineralization also increases (Gundersen et al., 1998b; Nave et al., 2009). The N content in all compartments increases and major internal N fluxes increase. The responses of soil C processes are more complex and less well understood. Early stages of litter decomposition may also respond positively to elevated N deposition, since microorganisms on high C:N ratio litter material need to immobilize N for the decomposition (Berg, 2000). At later stages, on the other hand, when the easily decomposable organic matter has been processed, decomposition may actually decrease with N availability (Berg and Matzner, 1997). Thus with a constant leaf litter C input, SOM is expected to accumulate due to reduced decomposition at elevated N deposition. However, the mechanism whereby N addition is accompanied by a decrease in decomposition and increase in soil C stocks is still unclear (Janssens et al., 2010). A prevailing hypothesis is the lower production of lignolytic enzymes and phenol oxidases (see review in Janssens et al., 2010). There is growing evidence of reduced soil respiration from chronic N addition experiments (Burton et al., 2004; Hagedorn et al., 2003; Janssens et€ al., 2010) in parallel with a decline in soil microbial biomass (Treseder, 2008). Since soil respiration, however, includes not only a component from decomposition but also respiration components from roots and mycorrhyzae, we cannot conclude that reduced soil respiration indicates increased SOM-C accumulation (C sequestration), as it could also indicate reduced root respiration.

116

A critical gap in our knowledge is the contribution of belowground litter input (roots, mycorrhyzae and exudates) to SOM formation versus that from aboveground plant litter (leaves, etc.) (Rasse et al., 2005). Likewise the responses of these belowground components to elevated N deposition are not well known, but total belowground C allocation is known to decline with increasing productivity and N availability (Palmroth et al., 2006). Root biomass (Boxman et al.,1998b; Nadelhoffer, 2000) and turnover (Majdi, 2004) may decrease with N availability. Likewise ectomycorrhyzal mycelial growth was also negatively affected by N additions after N fertilization (Nilsson and Wallander, 2003; Parrent and Vilgalys, 2007), as an effect of N deposition (Nilsson et al., 2000) and with increasing N availability in natural nutrient fertility gradients (Nilsson et al., 2005). Alternatively, accumulation of SOM by reduced decomposition could be outweighed by reduced contributions from other sources of SOM-C (roots, mycorrhyzae and exudates). The integrated response of these processes to elevated N in the form of increased SOM accumulation may be difficult to measure due to the size and the variability in this pool (Yanai et al., 2003). However, long-term forest N fertilization experiments from Sweden and Finland did reveal an increase in SOM-C from N addition (Hyvönen et al., 2008) and for the first time a significant response of SOM-C was shown for a chronic N addition experiment at four forest sites in Michigan, USA (Pregitzer et al., 2008). A long-term N addition experiment on heathland also revealed increased SOM-C accumulation (Evans et al., 2006). Further measurements of increased organic layer thickness over 40 years in an intensive network of sites across Sweden (Berg et al., 2009) indicate that C sequestration in organic layers could be a widespread phenomenon. A metadata analysis of studies on the response of CO2 flux from N additions in multiple terrestrial and wetland ecosystem types by Liu and Graever (2009) showed a large variation of net ecosystem CO2 exchange (NEE) for non-forest ecosystems (grassland, wetland and tundras), thus leading to a statistically insignificant effect.

6.6.4╇ N leaching associated effects Nitrate leaching is an acidifying process in soils and an important process in acidification of lower soil horizons (Velthof et€ al. 2011, Section 21.4 this volume). The input may be as nitric acid or as ammonia/ammonium that can be nitrified in the soil and release protons. Soils have an ability to Â�neutralize acids through the supply of base cations from weathering and cation exchange reactions. Increased acidification of forest soils has been observed during recent decades (FalkengrenGrerup et al., 1987; Wesselink et al., 1995) with pH declining by up to 1 unit, which in part may be caused by air pollution including deposition of N compounds. Both proton (H+)producing and proton-consuming processes including N species occur in soils, but a net acidification only occurs when nitrate is leached from the system (Gundersen and Rasmussen, 1990). Each 14 kg N ha−1 yr−1 of nitrate leached is equivalent to the production of 1 kmol H+ ha−1 yr−1. Depending on the acid status of the soil, base cations and/or Al will be leached

Klaus Butterbach-Bahl and Per Gundersen

with the nitrate. An increasing fraction of the acidity in acidsensitive surface waters is related to nitrate (Stoddard et al., 1999; Durand et al., 2011, Chapter 7, this volume). In Europe, approximately 30% of monitored forest sites leach between 7 and 50 kg N ha−1 yr−1 (De Vries et al., 2007; MacDonald et al., 2002), equivalent to an acid production of 0.5–3.5 kmol H+ ha−1 yr−1. At the majority of these sites this was buffered by Al release (Dise et al., 2001; De Vries et al., 2007) and nitrate and aluminium are usually positively correlated in acid soil and surface waters. In the long-term this may lead to significant nutrient loss, impairment of base cation uptake by Al toxicity and potentially reduced forest production (Velthof et al., 2011, Chapter 21, this volume) as well as Al toxicity in surface water (Havas and Rossland, 1995).

6.6.5╇ N-driven vulnerability to disturbances Nitrogen deposition may play a significant role in increasing forest susceptibility to wildfires, as documented in mixed coniferous forests of southern California (USA). High levels of ozone and nitrogenous compounds derived from regional urbanization and industrialization cause specific changes in forest tree C, N, and water balances that enhance individual tree susceptibility to drought, bark beetle attack, and disease, and when combined contribute to the whole ecosystem susceptibility to wildfire (Grulke et al., 2009). Similar findings have been documented for deserts and coastal sage scrub formations in southern California where N deposition and frequent fire promotes increased grass biomass from invasive species and increases the risk of fire further as it provides the fuel for subsequent fires (Brooks et al., 2004; Rao et al., 2009).

6.7╇ Summary This section summarizes the main conclusions on the terrestrial N cycling processes and their importance. Major uncertainties and gaps in knowledge are also highlighted.

Nitrogen pools and N availability (1) On the ecosystem scale, soils are the main reservoir for N. This is more pronounced for agricultural systems, with more than 90%–95% of Nr being stored in the soil as compared to forest systems, where N storage in soil is 50–70%. (2) N availability varies with temperature and humidity gradients in Europe. (3) The contemporary global biogeochemical cycle of N in terrestrial ecosystems is dominated by microbial processes in soils.

Nitrogen inputs in non-agricultural systems (4) An understanding of biological N2 fixation in a few legume crop plants is relatively advanced, but much less is known about biological N2 fixation in non-agricultural legumes or in other N2-fixing organisms. The difficulties of measuring rates of biological N2 fixation accurately at the ecosystem scale have so far hampered a better understanding of the

importance of biological N2 fixation for most terrestrial ecosystems. (5) The estimation of N deposition inputs at the site scale is affected by neglecting the input of dissolved organic nitrogen (DON) and by the uncertainty in atmosphere canopy interaction of N species.

Nitrogen cycling ╇ (6) The understanding of N cycling in terrestrial ecosystems has undergone a paradigm shift since 1990. Until then, the perception was that (i) N mineralization is the limiting step in N cycling, (ii) plants take up inorganic N, and (iii)€plants poorly compete for N against microbes and use only the N which is ‘left over’ by microbes. Consequently, net N mineralization has been assessed to measure plantavailable N. Since then studies have shown that plants effectively compete for N with microorganisms and take up organic N, in a broad range of ecosystems. ╇ (7) Nitrogen mineralization/ammonification is the dominant control of gross nitrification. ╇ (8) Microbial nitrate immobilization is a significant process of Nr-retention in a wide range of terrestrial ecosystems that depend largely on gross nitrification. ╇ (9) The importance of other recently recognized processes (DNRA, anammox, nitrifier denitrification) for Nr-cycling in ecosystems is not well developed. (10) Denitrification rates in soils are highly uncertain, despite more than eight decades of research, partly because of our lack of understanding and partly due to the large spatial and temporal variability. (11) Factors controlling Nr fluxes include moisture content (water-filled pore space) and soil temperature, soil properties, such as clay content, carbon content, C:N ratio and pH and vegetation factors, which are all affected by land use change and climatic change.

Nitrogen outputs (12) Burning is a major pathway of Nr loss for ecosystems exposed to high fire frequencies. The influence of Nr deposition on fuel N build-up should be considered in order to estimate wildfire NOx emissions. This effect is neglected at present. (13) Major outputs in ecosystems not exposed to fires are nitrate leaching and gaseous losses (N2O, NO, N2 and in agricultural systems also NH3) to the environment. N2 emissions are most uncertain due to the difficulties to quantify N2 emissions and to constrain what is driving denitrification on site and regional scales. (14) The C:N ratio of the forest floor or the top mineral soil is a good indicator of N status related to NO3– leaching. At C:N above 25, mineral N is usually retained, whereas below 25, NO3– leaching often occurs and increases with increasing N deposition.

Nitrogen effects (15) Atmospheric Nr input has caused N saturation in terms of a decline in the soil C:N ratio in the forest floor, associated

117

Nitrogen processes in terrestrial ecosystems

with an increase in (i) Nr (nitrate) leaching and in gaseous N losses (N2O, NO, N2), (ii) N availability and related biodiversity change, and (iii) the occurrence of pests and diseases. (16) Atmospheric Nr input has led to an increase in aboveground C sequestration but the impact on soil C sequestration is less clear. A critical gap in our knowledge is the impact of Nr deposition on belowground litter input (roots, mycorrhyzae and exudates) and through that on soil C sequestration.

Future studies on terrestrial N cycling (17) Interdisciplinary and multi-scale studies should focus on simultaneous and comprehensive measurements of all major Nr fluxes at site and landscape scale, including plant uptake/release of organic and inorganic N compounds as well as microbial Nr conversion. (18) Linking plant physiological and soil microbial Nr cycling, as well as soil hydrological Nr transport, to more reliable estimates of ecosystem N fluxes will be a major research challenge for coming years. In particular, this will include further development of methodological approaches and experimental assays that allow direct assessment of N turnover processes in the larger context of intact plant– soil-systems, where competitive mechanisms between microorganisms and plants persist. This will require an intensified interdisciplinary cooperation between plant scientists and soil ecologists/soil microbiologists.

Acknowledgements This chapter was prepared with the support of the NinE Programme of the European Science Foundation, the NitroEurope IP (funded by the European Commission), the COST Action 729 and the Villcum Foundation, Denmark.

Supplementary materials Supplementary materials (as referenced in the chapter) are available online through both Cambridge University Press:€ www.cambridge.org/ena and the Nitrogen in Europe website:€ www.nine-esf.org/ena.

References Aber, J. D., Nadelhoffer, K. J., Steudler, P. and Melillo, J. M. (1989). Nitrogen saturation in northern forest ecosystems. BioScience, 39, 378–386. Aber, J. D., McDowell, W., Nadelhoffer, K. et al. (1998). Nitrogen saturation in temperate forest ecosystems:€hypothesis revisited. BioScience, 48, 921–934. Aber, J. D., Goodale, C. L., Ollinger, S. V. et al. (2003). Is nitrogen deposition altering the nitrogen status of northeastern forests? BioScience, 53, 375–389. Accoe, F., Boeckx, P., Busschaert, J., Hofman, G. and Van Cleemput,€O. (2004). Gross N transformation rates and net N mineralisation rates related to the C and N contents of soil organic matter fractions in grassland soils of different age. Soil Biology and Biochemistry, 36, 2075–2087. Ågren, G. I. and Bosatta, E. (1988). Nitrogen saturation of terrestrial ecosystems. Environmental Pollution, 54, 185–198.

118

Alexis, M. A., Rasse, D. P., Rumpel, C. et al. (2007). Fire impact on C and N losses and charcoal production in a scrub oak ecosystem. Biogeochemistry, 82, 201–216. Ambus, P., Mosier, A. and Christensen, S. (1992). Nitrogen turnover rates in a riparian fen determined by 15N dilution. Biology and Fertility of Soils, 14, 230–236. Andersson, P., Berggren, D. and Nilsson, I. (2002). Indices for nitrogen status and nitrate leaching from Norway spruce (Picea abies (L.) Karst.) stands in Sweden. Forest Ecology and Management, 157, 39–53. Azam, F., Mahmood, T. and Malik, K. A. (1988). Immobilisationremineralisation of NO3-N and total N balance during decomposition of glucose, sucrose and cellulose in soil incubated at different moisture regimes. Plant and Soil, 107, 159–163. Baggs, E. M. (2008). A review of stable isotope techniques for N2O source partitioning in soils:€recent progress, remaining challenges and future considerations. Rapid Communication in Mass Spectrometry, 22, 1664–1672. Bais, H. P., Weir, T. L., Perry, G., Gilroy, S. and Vivanco, J. M. (2006). The role of root exudates in rhizosphere interactions with plants and other microorganisms. Annual Review of Plant Biology, 57, 233–266. Ball, B. C., Bingham, I., Rees, R. M., Watson, C.A. and Litterick, A. (2005). The role of crop rotations in determining soil structure and crop growth conditions. Canadian Journal of Soil Science, 85, 557–577. Barbosa, P., Camia, P., Kucera, J. et al. (2009). Assessment of forest fire impacts and emissions in the European Union based on the European Forest Fire Information System. In:€Wildland Fires and Air Pollution. ed. A. Bytnerowicz, M. Arbaugh, A. Riebau and C.€Andersen, Elsevier: New York, pp. 197–208. Barton, L., McLay, C. D. A., Schipper, L. A. and Smith, C. T. (1999). Annual denitrification rates in agricultural and forest soils:€a review. Australian Journal of Soil Research, 37, 1073–1093. Bateman, E. J. and Baggs, E. M. (2005). Contributions of nitrification and denitrification to N2O emissions from soils at different waterfilled pore space. Biology and Fertility of Soils, 41, 379–388. Batjes, N. H. (1996). Total carbon and nitrogen stocks in the soils of the world. European Journal of Soil Science, 47, 151–163. Beier, C., Emmett, B. A., Peñuelasc, J. et al. (2008). Carbon and nitrogen cycles in European ecosystems respond differently to global warming. Science of the Total Environment, 407, 692–697. Berg, B. (2000). Litter decomposition and organic matter turnover in northern forest soils. Forest Ecology and Management, 133, 13–22. Berg, B. and Matzner, E. (1997). Effect of N deposition on decomposition of plant litter and soil organic matter in forest systems. Environmental Reviews, 5, 1–25. Berg, B., Johansson, M.-B., Nilsson, Å., Gundersen, P. and Norell,€L. (2009). Sequestration of Carbon in humus layers in Swedish forests€– direct measurements. Canadian Journal of Forest Research, 39, 962–975. Binkley, D., Stottlemyer, R., Suarez, F. and Cortina, J. (1994). Soil nitrogen availability in some arctic ecosystems in northwest Alaska:€responses to temperature and moisture. Ecoscience, 1, 64–70. Bobbink, R., Hornung, M. and Roelofs, J. G. M. (1998). The effects of air-borne nitrogen pollutants on species diversity in natural and semi-natural European vegetation. Journal of Ecology, 86, 717–738. Bollmann, A. and Conrad, R. (1997). Acetylene blockage technique leads to underestimation of denitrification rates in oxic soils

Klaus Butterbach-Bahl and Per Gundersen due to scavenging of intermediate nitric oxide. Soil Biology and Biochemistry, 29, 1067–1077. Bonin, P. (1996). Anaerobic nitrate reduction to ammonium in two strains isolated from costal marine sediment:€A dissimilatory pathway. FEMS Microbiology Ecology, 19, 27–38. Booth, M. S., Stark, J. M. and Rastetter, E. (2005). Controls on Nitrogen cycling in terrestrial ecosystems:€a synthetic analysis of literature data. Ecological monographs, 75, 139–157. Bouwman, A. (1998). Nitrogen oxides and tropical agriculture. Nature, 392, 886–887. Bouwman, A. F., Boumans, L. J. M. and Batjes, N. H. (2002). Estimation of global NH3 volatilization loss from synthetic fertilizers and animal manure applied to arable lands and grasslands. Global Biogeochemical Cycles, 16, 1024–1038. Bouwman, A. F., Lee, D. S., Asman, W. A. H. et al. (1997). A global high-resolution emission inventory for ammonia. Global Biogeochemical Cycles, 11, 561–587. Boxman, A. W., van der Ven, P. J. M. and Roelofs, J. G. M. (1998a). Ecosystem recovery after a decrease in nitrogen input to a Scots pine stand at Ysselsteyn, the Netherlands. Forest Ecology and Management, 101, 155–163. Boxman, A. W., Blanck, K., Brandrud, T.-E. et al. (1998b). Vegetation and soil biota response to experimentally-changed nitrogen inputs in coniferous forest ecosystems of the NITREX project. Forest Ecology and Management, 101, 65–79. Bredemeier, M., Blanck, K., Dohrenbusch, A. et al. (1998). The Solling roof experiments€– site characteristics, experiments and results. Forest Ecology and Management, 101, 281–293. Breuer, L., Kiese, R. and Butterbach-Bahl, K. (2002). Temperature and moisture effects on nitrification rates in tropical forest soils. Soil Science Society of America Journal, 66, 834–844. Britto, D. T. and Kronzucker, H. J. (2002). NH4+ toxicity in higher plants:€a critical review. Journal of Plant Physiology, 159, 567–584. Brooks, M. L., D’Antonio, C. M., Richardson, D. M. et al. (2004). Effects of invasive alien plants on fire regimes. Bioscience, 54, 677–688. Brookshire, E. N. J., Valett, H. M., Thomas, S. A. and Webster, J. R. (2007). Atmospheric N deposition increases organic N loss from temperate forests. Ecosystems, 10, 252–262. Brumme, R., Borken, W. and Finke, S. (1999). Hierachical control on nitrous oxide emission in forest ecosystem. Global Biogeochemical Cycles, 13, 1137–1148. Brunstig, A. M. H. and Heil, G. W. (1985). The role of nutrients in the interactions between a herbivorous beetle and some competing plant species in healthlands. Oikos, 44, 23–26. Burger, M. and Jackson, L. E. (2003). Microbial immobilisation of ammonium and nitrate in relation to ammonification and nitrification rates in organic and conventional cropping systems. Soil Biology and Biochemistry, 35, 29–36. Burton, A. J., Pregitzer, K. S., Crawford, J. N., Zogg, G. P. and Zak, D. R. (2004). Simulated chronic NO3 deposition reduces soil respiration in northern hardwood forests. Global Change Biology, 10, 1080–1091. Butterbach-Bahl, K., Rothe, A. and Papen, H. (2002). Effect of tree distance on N2O and CH4-fluxes from soils in temperate forest ecosystems. Plant and Soil, 240, 91–103. Butterbach-Bahl, K., Kahl, M., Mykhayliv, L. et al. (2009). A European wide inventory of soil NO emissions using the biogeochemical models DNDC/ Forest DNDC. Atmospheric Environment, 43, 1392–1402.

Butterbach-Bahl, K., Kock, M., Willibald, G. et al. (2004). Temporal variations of fluxes of NO, NO2, N2O, CO2 and CH4 in a tropical rain forest ecosystem. Global Biogeochemical Cycles, 18, doi:10.1029/2004GB002243. Butterbach-Bahl, K., Nemitz, E., Zaehle, S. et al. (2011). Nitrogen as a threat to the European greenhouse balance. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Campbell, J. L., Hornbeck, J. W., McDowell, W. H. et al. (2000). Dissolved organic nitrogen budgets for upland, forested ecosystems in New England. Biogeochemistry, 49, 123–142. Campbell, J. L., Rustad, L. E., Boyer, E. W. et al. (2009). Consequences of climate change for biogeochemical cycling in forests of northeastern North America. Canadian Journal of Forest Soils, 39, 264–284. Cape, J. N., Anderson, M., Rowland, A. P. and Wilson, D. (2004). Organic nitrogen in precipitation across the United Kingdom. Water, Air, and Soil Pollution:€Focus, 4, 25–35. Caporn, S. J. M., Ashenden, T. W. and Lee, J. A. (2000). The effect of exposure to NO2 and SO2 on frost hardiness in Calluna vulgaris. Environmental and Experimental Botany, 43, 111–119. Carlsson, G. and Huss-Danell, K. (2003). Nitrogen fixation in perennial forage legumes in the field. Plant and Soil, 253, 353–372. Cassman, K. G. and Harwood, R. R. (1995). The nature of agricultural systems€– food security and environmental balance. Food Policy, 20, 439–454. Castaldi, S., Carfora, A., Fiorentino, A. et al. (2009). Inhibition of net nitrification activity in a Mediterranean woodland:€possible role of chemicals produced by Arbutus unedo. Plant and Soil, 15, 273–283. Chapman, S. K., Langley, J. A., Hart, S. C. and Koch, G. W. (2006). Plants actively control nitrogen cycling:€uncorking the microbial bottleneck. New Phytologist, 169, 27–34. Chapuis-Lardy, L., Wrage, N., Liemetay, A., Chotte, J. L. and Bernoux, M. (2007). Soils, a sink for N2O? A review. Global Change Biology, 13, 1–17. Christensen, S. and Tiedje, J. M. (1990). Brief and vigorous N2O production by soil at spring thaw. Journal of Soil Science, 41, 1–4. Ciais, P., Schelhaas, M. J., Zaehle, S. et al. (2008). Carbon accumulation in European forests. Nature Geoscience, 1, 425–429. Cleveland, C. C., Townsend, A. R., Schimel, D. S. et al. (1999). Global patterns of terrestrial biological nitrogen (N2) fixation in natural ecosystems. Global Biogeochemical Cycles, 13, 623–645. Cliff, J. B., Bottomley, P. J., Gaspar, D. J. and Myrold, D. D. (2002). Exploration of inorganic C and N assimilation by soil microbes with time of flight secondary ion mass spectrometry. Applied and Environmental Microbiology, 68, 4067–4073. Colloff, M. J., Wakelin, S. A., Gomez, D. and Rogers, S. L. (2008). Detection of nitrogen cycle genes in soils for measuring the effects of changes in land use and management. Soil Biology & Biochemistry, 40, 1637–1645. Conrad, R. (2002). Microbiological and biochemical background of production and consumption of NO and N2O in soil. In:€Trace Gas Exchange in Forest Ecosystems, ed. R. Gasche, H. Papen and H. Rennenberg. Dordrecht:€Kluwer Academic Publishers, pp.€3–33. Corré, M. F., Beese, F. O. and Brumme, R. (2003). Soil nitrogen cycle in high nitrogen deposition forest:€changes under nitrogen saturation and liming. Ecological Applications, 13, 287–298. Costa, E., Perez, J. and Kreft, J. U. (2006). Why is metabolic labour divided in nitrification? Trends in Microbiology, 14, 213–219.

119

Nitrogen processes in terrestrial ecosystems Crutzen, P. J., Mosier, A. R., Smith, K. A. and Winiwarter, W. (2008). N2O release from agro-biofuel production negates global warming reduction by replacing fossil fuels. Atmospheric Chemistry and Physics, 8, 389–395. Dail, D. B., Hollinger, D. Y., Davidson, E. A. et al. (2009). Distribution of nitrogen-15 tracers applied to the canopy of a mature sprucehemlock stand, Howland, Maine, USA. Oecologia, 160, 589–599. Dannenmann, M., Gasche, R., Ledebuhr, A. and Papen, H. (2006). Effects of forest management on soil N cycling in beech forests stocking on calcareous soils. Plant and Soil, 287, 279–300. Davidson, E. A. (1994). Climate change and soil microbial processes:€secondary effects are hypothesised from better known interacting primary effects. In:€Soil Response to Climate Change, ed. M. D. A. Rounsevell and P. J. Loveland. Springer,€Berlin: pp.€156–168. Davidson, E. A. (2009). The contribution of manure and fertilizer nitrogen to atmospheric nitrous oxide since 1860. Nature Geoscience, 2, 659–662. Davidson, E. A., Hart, S. C. and Firestone, M. K. (1992). Internal cycling of nitrate in soils of a mature conferous forest. Ecology, 73, 1148–1156. Davidson, E. A., Hart, S. C., Shanks, C. A. and Firestone, M. K. (1991). Measuring gross nitrogen mineralisation, immobilisation, and nitrification by 15N isotopic pool dilution in intact soil cores. Journal of Soil Science, 42, 335–349. Davidson, E. A., Matson, P. A., Vitousek, P. M. et al. (1993). Processes regulating soil emissions of NO and N2O in a seasonally dry tropical forest. Ecology, 74, 130–139. De Vries, W., Reinds, G. J., Gundersen, P. and Sterba, H. (2006). The impact of nitrogen deposition on carbon sequestration in European forests and forest soils. Global Change Biology, 12, 1151–1173. De Vries, W., van der Salm, C., Reinds, G. J. and Erisman, J. W. (2007). Element fluxes through European forest ecosystems and their relationships with stand and site characteristics. Environmental Pollution, 148, 501–513. De Vries, W., Solberg, S., Dobbertin, M. et al. (2008). Ecologically implausible carbon response. Nature, 441, E1–E3. De Vries, W., Solberg, S., Dobbertin, M. et al. (2009). The impact of nitrogen deposition on carbon sequestration by terrestrial ecosystems. Forest Ecology and Management, 258, 1814–1823. Dezi, S., Medlyn, B., Tonon, G. and Magnani, F. (2009). Effect of nitrogen deposition on forest carbon sequestration:€model analysis. Global Change Biology, 16, 1470–1486. Dise, N. B., Matzner, E., Armbruster, M. and MacDonald, J. (2001). Aluminum output fluxes from forest ecosystems in Europe:€a regional assessment. Journal of Environmental Quality, 30, 1747–1756. Dise, N. B., Rothwell, J. J., Gauci, V., van der Salm, C. and de Vries, W. (2009). Predicting dissolved inorganic nitrogen leaching in European forests using two independent databases. Science of the Total Environment, 407, 1798–1808. Dise, N. B., Ashmore, M., Belyazid, S. et al. (2011). Nitrogen as a threat to European terrestrial biodiversity. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Dobbie, K. E., McTaggart, I. P. and Smith, K. A. (1999). Nitrous oxide emissions from intensive agricultural systems:€variations between crops and seasons, key driving variables, and mean emission factors. Journal of Geophysical Research, 104, 26891–26899. Dragosits, U., Theobald, M. R., Place, C. J. et al. (2002). Ammonia emission deposition and impact assessment at the field scale:€a case

120

study of sub-grid spatial variability. Environmental Pollution, 117, 147–158. Dupré, C., Stevens, C. J., Ranke, T. et al. (2010). Changes in species richness and composition in European acidic grasslands over the past 70 years:€the contribution of cumulative atmospheric nitrogen deposition. Global Change Biology, 16, 344–357. Durand, P., Breuer, L., Johnes, P. et al. (2011). Nitrogen processes in aquatic ecosystems. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Emmett, B. A., Gundersen, P., Kjønaas, O. J. et al. (1998). Predicting the effects of atmospheric nitrogen deposition in conifer stands:€evidence from the NITREX ecosystem-scale experiments. Ecosystems, 1, 352–360. Eno, C. (1960). Nitrate production in the field by incubating the soil in polyethylene bags. Soil Science Society of America Proceedings, 24, 277–279. Erisman, J. W., Sutton, M. A., Galloway, J., Klimont, Z. and Winiwarter, W. (2008). How a century of ammonia synthesis changed the world. Nature Geoscience, 1, 636–639. Erisman, J. W., van Grinsven, H., Grizzetti, B. et al. (2011). The European nitrogen problem in a global perspective. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Eskew, D. L., Eaglesham, A. R. J. and App, A. A. (1981). Heterotrophic (N-15)2 fixation and distribution of newly fixed nitrogen in a riceflooded soil system. Plant Physiology, 68, 48–52. Evans, C. D., Caporn, S. J. M., Carroll, J. A. et al. (2006). Modelling nitrogen saturation and carbon accumulation in heathland soils under elevated nitrogen deposition. Environmental Pollution, 143, 468–478. Evans, J. R. (1989). Photosynthesis and nitrogen relationships in leaves of C3 plants. Oecologia, 78, 9–19. Faegerli, H. and Aas, W. (2008). Trends of nitrogen in air and precipitation:€Model results and observations at EMEP sites in Europe, 1980–2003. Environmental Pollution, 154, 448–461. Falkengren-Grerup, U., Linnermark, N. and Tyler, G. (1987). Changes in acidity and cation pools of south Swedish soils between 1949 and 1985. Chemosphere, 16, 2239–2248. Fisk, M. C., Zak, D. R. and Crow, T. R. (2002). Nitrogen storage and cycling in old- and second-growth northern hardwood forests. Ecology, 83, 73–87. Fowler, D., O’Donoghue, M., Muller, J. B. A. et al. (2004). A chronology of nitrogen deposition in the UK between 1900 and 2000. Water, Air and Soil Pollution, 4, 9–23. French, S., Levy-Booth, D., Samarajeewa, A. et al. (2009). Elevated temperatures and carbon dioxide concentrations:€effects on selected microbial activities in temperate agricultural soils. World Journal of Microbiology Biotechnology, 25, 1887–1900. Freppaz, M., Williams, B. L., Edwards, A. C., Scalenghe, R. and Zanini, E. (2007). Labile nitrogen, carbon and phosphorous pools and nitrogen mineralisation and immobilisation rates at low temperatures in seasonally snow-covered soils. Biology and Fertility of Soils, 43, 519–529. Gaige, E., Dail, D. B., Hollinger, D. Y. et al. (2007). Changes in canopy processes following whole-forest canopy nitrogen fertilisation of a mature spruce-hemlock forest. Ecosystems, 10, 1133–1147. Galloway, J. N., Aber, J. D., Erisman, J. W. et al. (2003). The nitrogen cascade. BioScience, 53, 341–356. Galloway, J. N., Dentener, F. J., Capone, D. G. et al. (2004). Nitrogen cycles:€past, present, future. Biogeochemistry, 70, 153–226.

Klaus Butterbach-Bahl and Per Gundersen Gimeno, B. S., Yuan, F., Fenn, M. E. and Meixner, T. (2009). Management options for mitigating nitrogen (N) losses from N-saturated mixed-conifer forests in California. In:€Wildland Fires and Air Pollution. ed. A. Bytnerowicz, M. Arbaugh, A. Riebau and C. Andersen, Elsevier, New York: pp. 425–455. Glass, A. D. M., Britto, D. T., Kaiser, B. N. et al. (2002). The regulation of nitrate and ammonium transport systems in plants. Journal of Experimental Botany, 53, 855–864. Griffin, T. S., He, Z. and Honeycutt, C.W. (2005). Manure composition affects net transformation of nitrogen from dairy manures. Plant and Soil, 273, 29–38. Groffman, P. M., Altabet, M. A., Böhlke, J. K. et al. (2006). Methods for measuring denitrification:€diverse approaches to a difficult problem. Ecological Applications, 16, 2091–2122. Groffman, P. M., Butterbach-Bahl, K., Fulweiler, R. W. et al. (2009). Incorporating spatially and temporally explicit phenomena (hotspots and hot moments) in denitrification models. Biogeochemistry, 93, 49–77. Grulke, N. E., Minnich, R. A., Paine, T. D. et al. (2009). Air pollution increases forest susceptibility to wildfires:€A case study in the San Bernardino Mountains in southern California. In: Wildland Fires and Air Pollution, ed. A. Bytnerowicz, M. Arbaugh, A. Riebau and C. Andersen, New York:€Elsevier, pp. 365–403. Gundersen, P. and Rasmussen, L. (1990). Nitrification in forest soils:€effects of nitrogen deposition on soil acidification and aluminium release. Reviews of Environmental Contamination Toxicology, 13, 1–45. Gundersen, P., Callesen, I. and de Vries, W. (1998a). Nitrate leaching in forest ecosystems is controlled by forest floor C/N ratio. Environmental Pollution, 102, 403–407. Gundersen, P., Emmett, B. A., Kjønaas, O. J., Koopmans, C. and Tietema, A. (1998b). Impact of nitrogen deposition on nitrogen cycling:€a synthesis of NITREX-data. Forest Ecology and Management, 101, 37–55. Gundersen, P., Boxman, A. W., Lamersdorf, N., Moldan, F. and Andersen, B. R. (1998c). Experimental manipulation of forest ecosystems€– lessons from large roof experiments. Forest Ecology and Management, 101, 339–352. Gundersen, P., Schmidt, I. K. and Raulund-Rasmussen, K. (2006). Leaching of nitrate from temperate forests: effects of air pollution and forest management. Environmental Reviews, 14, 1–57. Gundersen, P., Sevel, L., Christiansen, J. R. et al. (2009). Do indicators of nitrogen retention and leaching differ between coniferous and deciduous forests in Denmark? Forest Ecology and Management, 258, 1137–1146. Hagedorn, F., Spinnler, D. and Siegwolf, R. (2003). Increased N deposition retards mineralization of old soil organic matter. Soil Biology and Biochemistry, 35, 1683–1692. Harrison, K. A., Bol, R. and Bardgett, R. D. (2007). Preferences for different nitrogen forms by coexisting plant species and soil microbes. Ecology, 88, 989–999. Hart, S. C., Nason, G. E., Myrold, D. D. and Perry, D. A. (1994). Dynamics of gross nitrogen transformations in an old-growth forest:€the carbon connection. Ecology, 75, 880–891. Havas, M. and Rossland, B. O. (1995). Response of zoopankton, benthos and fish to acidification:€an overview. Water, Air and Soil Pollution, 85, 51–62. Hayatsu, M., Tago, K. and Saito, M. (2008). Various players in the nitrogen cycle:€diversity and functions of the microorganisms

involved in nitrification and denitrification. Soil Science and Plant Nutrition, 54, 33–45. Herridge, D. F., Peoples, M. B. and Boddey, R. M. (2008). Global inputs of biological nitrogen fixation in agricultural systems. Plant and Soil, 311, 1–8. Hertel, O., Reis, S., Skjøth, C. A. et al. (2011). Nitrogen processes in the atmosphere. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erismar et al. Cambridge University Press. Hofstra, N. and Bouwman, A. F. (2005). Denitrification in agricultural soils:€summarizing published data and estimating global annual rates. Nutrient Cycling in Agroecosystems, 72, 267–278. Högberg, P. and Read, D. J. (2006). Towards a more plant physiological perspective on soil ecology. Ecology and Evolution, 21, 548–554. Högberg, P., Fan, H., Quist, M., Binkley, D. and Tamm, C. O. (2006). Tree growth and soil acidification in response to 30 years of experimental nitrogen loading on boreal forest. Global Change Biology, 12, 489–499. Houlton, B. Z. and Bai, E. (2009). Imprint of denitrifying bacteria on the global terrestrial biosphere. Proceedings of the National Academy of Sciences of the USA, 106, 21713–21716. Humbert, S., Tarnawski, S., Fromin, N. et al. (2010). Molecular detection of anammox bacteria in terrestrial ecosystems:€distribution and diversity. ISME Journal, 4, 450–454. Hyvönen, R., Persson, T., Andersson, S. et al. (2008). Impact of longterm nitrogen addition on carbon stocks in trees and soils in northern Europe. Biogeochemistry, 89, 121–137. Ingwersen, J., Butterbach-Bahl, K., Gasche, R., Richter, O. and Papen,€H. (1999). Barometric process separation:€New method for quantifying nitrification, denitrification, and nitrous oxide sources in soils. Soil Science Society of America Journal, 63, 117–128. IPCC (1997). Revised 1996 Guidelines for National Greenhouse Gas Inventories, Bracknell, UK: IPCC. IPCC (2007). 2006 Guidelines for National Greenhouse Gas Inventories, Vol. 4, Chap. 11, Hayama, Japan: IGES. Jackson, L. E., Burger, M. and Cavagnaro, T. R. (2008). Roots, nitrogen transformations and ecosystem services. Plant Biology, 59, 341–363. Janssens, I. A., Dieleman, W., Luyssaert, S. et al. (2010). Reduction of forest soil respiration in response to nitrogen deposition. Nature Geoscience, 3, 315–322. Jansson, S. L., Hallam, M. J. and Bartholomew, W. V. (1955). Preferential utilization of ammonium over nitrate by microorganisms in the decomposition of oat straw. Plant and Soil, 4, 382–390. Jarvis, P. G. and Linder, S. (2000). Constraints to growth of boreal forests. Nature, 405, 904–905. Jarvis, S. C., Stockdale, E. A., Shepherd, M. A. and Powlson, D.S. (1996). Nitrogen mineralisation in temperate agricultural soils:€processes and measurement. Agronomy, 57, 187–235. Jarvis, S., Hutchings, N., Brentrup, F., Olesen, J. and van der Hoek, K. (2011). Nitrogen flows in farming systems across Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Jensen, L. S., Schjoerring, J. K., van der Hoek, K. et al. (2011). Benefits of nitrogen for food fibre and industrial production. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Johnson, D. W., Murphy, J. D., Walker, R. F., Miller, W. W. and Glass, D. W. (2007). Wildfire effects on forest carbon and nutrient

121

Nitrogen processes in terrestrial ecosystems budgets:€comparison of reconstructed versus measured values. Ecological Engineering, 31, 183–192. Karjalainen, T., Schuck, A., Prietzel, J. et al. (2008). Causes and Consequences of Forest Growth Trends in Europe, Results of the RECOGNITION Project, EFI Report No. 21. Joensuu:€European Forest Institute. Kesik, M., Blagodatsky, S., Papen, H. and Butterbach-Bahl, K. (2006a). Effect of pH, temperature and substrate on N2O, NO and CO2 production by Alcaligenes faecalis p. J. Applied Microbiology, 101, 655–667. Kesik, M., Brüggemann, N., Forkel, R. et al. (2006b). Future scenarios of N2O and NO emissions from European forest soils. Journal of Geophysical Research, 97, 9777–9783. Klemedtsson, L., von Arnold, K., Weslien, P. and Gundersen, P. (2005). Soil N ratio as scalar parameter to predict nitrous oxide emissions. Global Change Biology, 11, 1–6. Kreutzer, K., Butterbach-Bahl, K., Rennenberg, H. and Papen, H. (2009). The complete nitrogen cycle of a N-saturated spruce forest ecosystem. Plant Biology, 11, 694–700. Kriebitzsch, W. U. (1978). Stickstoffnachlieferung in sauren Waldböden Nordwestdeutschlands. Scripta Geobotanica (Goltze, Göttingen, Germany), 14, 1–66. Kristensen, H. L., Gundersen, P., Callesen, I. and Reinds, G. J. (2004). Relationships between soil nitrate concentrations and environmental factors. Ecosystems, 7, 180–192. Laubhann, D., Sterba, H., Reinds, G. J. and de Vries, W. (2009). The impact of changes in atmospheric deposition and climate on forest growth in European monitoring plots:€An empirical tree growth model. Forest Ecology and Management, 258, 1751–1761. Li, C., Frolking, S. and Butterbach-Bahl, K. (2005). Carbon sequestration can increase nitrous oxide emissions. Climatic Change, 72, 321–338. Liu, L. and Greaver, T. L. (2009). Nitrogen addition stimulates emissions of biogenic greenhouse gases in terrestrial and wetland ecosystems. Ecology Letters, 12, 1103–1117. Lovett, G. M. and Lindberg, S. E. (1993). Atmospheric deposition and canopy interactions of nitrogen in forests. Canadian Journal of Forest Research, 23, 1603–1616. Ludwig, J., Meixner, F. X., Vogel, B. and Förstner, J. (2001). Soil-air exchange of nitric oxide:€an overview of processes, environmental factors, and modeling studies. Biogeochemistry, 52, 225–257. MacDonald, J. A., Dise, N. B., Matzner, E. et al. (2002). Nitrogen input together with ecosystem nitrogen enrichment predict nitrate leaching from European forests. Global Change Biology, 8, 1028–1033. Machefert, S. E., Dise, N. B., Goulding, K. W. T. and Whitehead, P. G. (2002). Nitrous oxide emission from a range of land uses across Europe. Hydrology and Earth System Sciences, 6, 325–337. Magnani, F., Mencuccini, M., Borghetti, M. et al. (2007). The human footprint in the carbon cycle of temperate and boreal forests. Nature, 447, 848–850. Magnani, F., Mencuccini, M., Borghetti, M. et al. (2008). Ecologically implausible carbon response? Reply. Nature, 451, E3–E4. Majdi, H. (2004). Root and needle litter decomposition responses to enhanced supplies of N and S in a Norway spruce forest in southwest Sweden. Plant Biosystems, 138, 225–230. Maskell, L. C., Smart, S. M., Bullock, J. M., Thompson, K. and Stevens, C. J. (2009). Nitrogen deposition causes widespread loss of species richness in British habitats. Global Change Biology, 16, 671–679. McClain, M. E., Boyer, E. W., Dent, C. L. et al. (2003). Biogeochemical hotspots and hot moments at the interface of terrestrial and aquatic ecosystems. Ecosystems, 6, 301–312.

122

McDowell, W. H., Magill, A. H., Aitkenhead-Peterson, J. A. et al. (2004). Effects of chronic nitrogen amendment on dissolved organic matter and inorganic nitrogen in soil solution. Forest Ecology and Management, 196, 29–41. McLauchlan, K. (2006). The nature and longevity of agricultural impacts on soil carbon and nutrients:€a review. Ecosystems, 9, 1364–1382. Meyer, A., Grote, R., Polle, A. and Butterbach-Bahl, K. (2009). Simulating mycorrhiza contribution to forest C- and N cycling – the MYCOFON model. Plant and Soil, 327, 493–517. Michalzik, B., Kalbitz, K., Park, J.-H., Solinger, S. and Matzner, E. (2001). Fluxes and concentrations of dissolved organic carbon and nitrogen€– a synthesis for temperate forests. Biogeochemistry, 52, 173–205. Miranda, A. I., Marchi, E., Ferretti, M. and Millán, M. (2009). Forest fires and air quality issues in southern Europe. In:€Wildland Fires and Air Pollution. ed. A. Bytnerowicz, M. Arbaugh, A. Riebau and C. Anderse, New York: Elsevier, pp. 209–231. Misselbrook, T. H., Shepherd, M. A. and Pain, P. F. (1996). Sewage sludge applications to grassland:€Influence of sludge type, time and method of application on nitrate leaching and herbage yield. Journal of Agricultural Science, 126, 343–352. Moldan, F., Kjønaas, O. J., Stuanes, A. and Wright, R. F. (2006). Increased nitrogen in runoff and soil following thirteen years of experimentally-increased nitrogen deposition to a coniferousforested catchment at Gårdsjön, Sweden. Environmental Pollution, 144, 610–620. Moritsuka, N., Yanai, J., Mori, K. and Kosaki, T. (2004). Biotic and biotic processes of nitrogen immobilisation in the soil-residue interface. Soil Biology and Biochemistry, 36, 1141–1148. Müller, C., Kaleem Abbasi, M., Kammann, C. et al. (2004). Soil respiratory quotient determined via barometric process separation combined with nitrogen-15 labeling. Soil Science Society of America Journal, 68, 1610–1615. Müller, C., Rütting, T., Kattge, J., Laughlin, R. J. and Stevens, R. J. (2007). Estimation of parameters in complex 15N tracing models by Monte Carlo sampling. Soil Biology and Biochemistry, 39, 715–726. Murty, D., Kirschbaum, M. U. F., McMurtrie, R. E. and McGilvray, A. (2002). Does conversion of forest to agricultural land change soil carbon and nitrogen? A review of the literature. Global Change Biology, 8, 105–123. Myrold, D. D. and Posavatz, N. R. (2007). Potential importance of bacteria and fungi in nitrate assimilation in soil. Soil Biology and Biochemistry, 39, 1737–1743. Nadelhoffer, K. J. (2000). The potential effects of nitrogen deposition on fine-root production in forest ecosystems. New Phytologist, 147, 131–139. Nadelhoffer, K., Emmett, B. A., Gundersen, P. et al. (1999). Nitrogen deposition makes a minor contribution to carbon sequestration in northern forests. Nature, 398, 145–148. Näsholm, T., Kielland, K. and Ganeteg, U. (2009). Uptake of organic nitrogen by plants. New Phytologist, 182, 31–48. Nave, L. E., Vance, E. D., Swanston C. W. and Curtis, P. S. (2009). Impacts of elevated N inputs on north temperate forest soil C storage C/N, and net N-mineralization. Geoderma, 153, 231–240. Neill, C., Piccolo, M., Melillo, J., Steudler, P. and Cerri, C. (1999). Nitrogen dynamics in Amazon forest and pasture soils measured by 15N pool dilution. Soil Biology and Biochemistry, 31, 567–572. Nijburg, J. W., Coolen, M. J. L., Gerards, S., Gunnewiek, P. J. A. K. and Laanbroek, H. J. (1997). Effects of nitrate availability and the presence of Glyceria maxima on the composition and activity of

Klaus Butterbach-Bahl and Per Gundersen the dissimilatory nitrate-reducing bacterial community. Applied and Environmental Microbiology, 63, 931–937. Nilsson, L. O. and Wallander, H. (2003). Production of external mycelium by ectomycorrhyzal fungi in a Norway spruce forest was reduced in response to nitrogen fertilisation. New Phytologist, 158, 409–416. Nilsson, L. O., Baath, E., Falkengren-Grerup, U. and Wallander, H. (2000). Growth of ectomycorrhyzal mycelia and composition of soil microbial communities in oak forest soils along a nitrogen deposition gradient. Oecologia, 153, 375–384. Nilsson, L. O., Giesler, R., Baath, E. and Wallander, H. (2005). Growth and biomass of mycorrhyzal mycelia in coniferous forests along short natural nutrient gradients. New Phytologist, 165, 613–622. Nishio, T., Komada, M., Arao, T. and Kanamori, T. (2001). Simultaneous determination of transformation rates of nitrate in soil. Japan Agriculture Research Quarterly, 35, 11–17. Oenema, O., Witzke, H. P., Klimont, Z., Lesschen, J. P. and Velthof, G. L. (2009). Integrated assessment of promising measures to decrease nitrogen losses from agriculture in EU-27. Agriculture, Ecosystems€and Environment, 133, 280–288. Ormeci, B., Sanin, S. L. and Peirce, J. J. (1999). Laboratory studies of NO flux from agricultural soil:€effects of soil moisture, pH and temperature. Journal of Geophysical Research, 104, 1621–1629. Palmroth, S., Oren, R., McCarthy, H. R. et al. (2006). Aboveground sink strength in forests controls the allocation of carbon below ground and its [CO2]-induced enhancement. Proceedings of the National Academy of Sciences of the USA, 103, 19362–19367. Papen, H., Gessler, A., Zumbusch, E. and Rennenberg, H. (2002). Chemolithoautotrophic nitrifiers in the phyllosphere of a spruce ecosystem recieving high atmospheric nitrogen input. Current Microbiology, 44, 56–60. Park, J. H. and Matzner, E. (2006). Detrial control on the release of dissolved organic nitrogen (DON) and dissolved inorganic (DIN) from the forest floor under chronic N deposition. Environmental Pollution, 143, 178–185. Parkin, T. B. (1987). Soil microsites as a source of denitrification variability. Soil Science Society of America Journal, 51, 1194–1199. Parrent, J. L. and Vilgalys, R. (2007). Biomass and compositional responses of ectomycorrhyzal fungal hyphae to elevated CO2 and nitrogen fertilisation. New Phytologist, 176, 164–174. Perakis, S. S. and Hedin, L. O. (2002). Nitrogen loss from unpolluted South American forests mainly via dissolved organic compounds. Nature, 415, 416–419. Pilegaard, K., Skiba, U., Ambus, P. et al. (2006). Factors controlling regional differences in forest soil emission of nitrogen oxides (NO and N2O). Biogeosciences, 3, 615–661. Pregitzer, K. S., Zak, D. R., Burton, A. J., Ashby, J. A. and MacDonald, N. W. (2004). Chronic nitrate additions dramatically increase the export of carbon and nitrogen from northern hardwood ecosystems. Biogeochemistry, 68, 179–197. Pregitzer, K. S., Burton, A. J., Zak, D. R. and Talhelm, A. F. (2008). Simulated chronic nitrogen deposition increases carbon storage in Northern Temperate forests. Global Change Biology, 14, 142–157. Raastad, I. A. and Mulder, J. (1999). Dissolved organic matter (DOM) in acid forest soils at Gårdsjön (Sweden):€natural variabilities and effects of increased input of nitrogen and of reversal of acidification. Water, Air and Soil Pollution, 114, 199–219. Rao, L. E., Allen, E. B. and Meixner, T. (2009). Risk-based determination of critical nitrogen deposition loads for fire spread in southern California deserts. Ecological Applications, 20, 1320–1335.

Rasse, D. P., Rumpel, C. and Dignac, M. F. (2005). Is soil carbon mostly root carbon? Mechanisms for a specific stabilisation. Plant and Soil, 269, 341–356. Recous, S., Mary, B. and Faurie, G. (1990). Microbial immobilisation of ammonium and nitrate in cultivated soils. Soil Biology and Biochemistry, 22, 913–922. Rennenberg, H., Dannenmann, M., Gessler, A. et al. (2009). Nitrogen balance in forest soils:€nutritional limitation of plants under climate change stresses. Plant Biology, 11, 4–23. Rice, C. W. and Tiedje, J. M. (1989). Regulation of nitrate assimilation by ammonium in soils and in isolated soil microorganisms. Soil Biology and Biochemistry, 21, 597. Roelofs, J. G. M., Kempers, A. J., Houdijk, A. L. F. M. and Jansen, J. (1985). The effect of air-borne ammonium sulphate on Pinus nigra var. maritima in the Netherlands. Plant and Soil, 84, 45–56. Ross, D. S., Lawrence, G. B. and Fredriksen, G. (2004). Mineralisation and nitrification patterns at eight northeastern USA forested research sites. Forest Ecology and Management, 188, 317–335. Rothe, A., Huber, C., Kreutzer, K. and Weis, W. (2002). Deposition and soil leaching in stands of Norway spruce and European beech:€results from the Höglwald research in comparison with other case studies. Plant and Soil, 240, 33–45. Roy, A. H. and Hammond, L. L. (2004). Challenges and opportunities for the fertilizer industry. In:€Agriculture and the Nitrogen Cycle (SCOPE Book 65), ed. A. Mosier, K. J. Syers and J. R. Freney, Washington DC:€Island Press, pp. 233–243. Rustad, L. E., Campbell, J. L., Marion, G. M. et al. (2006). A meta-analysis of the response of soil respiration, net nitrogen mineralisation and aboveground plant growth to experimental ecosystem warming. Oecologia, 126, 543–562. Samarkin, V. A., Madigan, M. T., Bowles, M. W. et al. (2010). Abiotic nitrous oxide emission from the hypersaline Don Juan Pond in Antarctica. Nature Geoscience, 3, 341–344. Scarascia-Mugnozza, G., Bauer, G. A., Persson, H., Matteucci, G. and Masci, A. (2000). Tree biomass, growth and nutrient pools. In:€Carbon and Nitrogen Cycling in European Forest Ecosystems, ed. E. D. Schulze. Berlin:€Springer, pp. 49–62. Schimel, J. P. and Bennett, J. (2004). Nitrogen mineralisation:€challenges of a changing paradigm. Ecology, 85, 591–602. Schmidt, I. K., Emmett, B. A., Tietema, A. et al. (2004). Soil solution chemistry and element fluxes in three European heathlands and their responses to warming and drought. Ecosystems, 7, 638–649. Seitzinger, S., Harrison, J. A., Böhlke, J. K. et al. (2006). Denitrification across landscapes and waterscapes:€a synthesis. Ecological Applications, 16, 2064–2090. Sexstone, A., Revsbech, N., Parkin, T. and Tiedje, J. (1985). Direct measurement of oxygen profiles and denitrification rates in soil aggregates. Soil Science Society of America Journal, 49, 645–651. Shaw, M. R. and Harte, J. (2001). Response of nitrogen cycling to simulated climate change:€differential responses along a subalpine ecotone. Global Change Biology, 7, 193–210. Shepherd, M. A., Hatch, D. J., Jarvis, S. C. and Bhogal, A. (2001). Nitrate leaching from reseeded pasture. Soil Use and Management, 17, 97–105. Sievering, H. (1999). Reply to Nitrogen deposition makes a minor contribution to carbon sequestration in temperate forests. Nature, 400, 629–630. Sievering, H., Tomaszewski, T. and Torizzo, J. (2007). Canopy uptake of atmospheric N deposition at a conifer forest:€part I – canopy N budget, photosynthetic efficiency and net ecosystem exchange. Tellus, 59B, 483–492.

123

Nitrogen processes in terrestrial ecosystems Sikström, U., Nohrstedt, H. Ö., Petterson, F. and Jacobson, S. (1998). Stem-growth response of Pinus sylvestris and Picea abies to nitrogen fertilisation as related to needle nitrogen concentration. Trees, 12, 208–214. Silver, W. L., Herman, D. J. and Firestone, M. K. (2001). Dissimilatory nitrate reduction to ammonium in upland tropical forest soils. Ecology, 82, 2410–2416. Simek, M., Jisova, L. and Hopkins, D. W. (2002). What is the so-called pH-optimum of denitrification? Soil Biology and Biochemistry, 34, 1227–1234. Sjöberg, G., Bergkvist, B., Berggren, D. and Nilsson, S. I. (2003). Long term N addition effects on the C mineralisation and DOC production in mor humus under spruce. Soil Biology and Biochemistry, 35, 1305–1315. Smil, V. (1999). Nitrogen in crop production:€an account of global flows. Global Biogeochemical Cycles, 13, 647–662. Smith, K. A. (1997). The potential for feedback effects induced by global warming on emissions of nitrous oxide by soils. Global Change Biology, 3, 327–338. Solberg, S., Dobbertin, M., Reinds, G. J. et al. (2009). The impact of changes in atmospheric deposition and climate on forest growth in European monitoring plots:€an empirical stand growth model. Forest Ecology and Management, 258, 1735–1750. Spangenberg, A. and Kölling, C. (2004). Nitrogen deposition and nitrate leaching at forest edges exposed to high ammonia emissions in Southern Bavaria. Water, Air and Soil Pollution, 152, 233–255. Sparks, J. P. (2009). Ecological ramifications of the direct foliar uptake of nitrogen. Oecologia, 159, 1–13. Staelens, J., Houle, D., De Schrijver, A., Neirynck, J. and Verheyen, K. (2008). Calculating dry deposition and canopy exchange with the canopy budget model:€review of assumptions and application to two deciduous forests. Water, Air and Soil Pollution, 191, 149–169. Stark, J. M. and Firestone, M. K. (1995). Mechanisms for soil moisture effects on activity of nitrifying bacteria. Applied and Environmental Microbiology, 61, 218–221. Stevens, R. J., Laughlin, R. J. and Maline, J. P. (1998). Soil pH affects the processes reducing nitrate to nitrous oxide and di-nitrogen. Soil Biology and Biochemistry, 30, 1119–1126. Stevens, C. J., Dise, N. B., Gowing, D. J. and Mountford, J. O. (2006). Loss of forb diversity in relation to nitrogen deposition in the UK:€regional trends and potential controls. Global Change Biology, 12, 1823–1833. Stevens, C. J., Dise, N. B., Mountford, J. O. and Gowing, D. J. (2004). Impact of nitrogen deposition on the species richness of grasslands. Science, 303, 1876–1879. Stockdale, E. A., Hatch, D. J., Murphy, D. V., Ledgard, S. F. and Watson, C. J. (2002). Verifying the nitrification to immobilisation ratio (N/I) as a key determinant of potential nitrate loss in grassland and arable soils. Agronomie, 22, 831–838. Stoddard, J. L., Jeffries, D. S., Lükewille, A. et al. (1999). Regional trends in aquatic recovery from acidification in North America and Europe. Nature, 401, 575–578. Subbarao, G. V., Ishikawa, T., Ito, O. et al. (2006). A bioluminescence assay to detect nitrification inhibitors released rom plant roots:€a case study with Brachiaria humidicoa. Plant and Soil, 288, 101–112. Suding, K. N., Collins, S. L., Gough, L. et al. (2005). Functional- and abundance-based mechanisms explain diversity loss due to N fertilisation. Proceedings of the National Academy of Sciences of the USA, 102, 4387–4392.

124

Sutton, M. A., Simpson, D., Levy, P. E. et al. (2008). Uncertainties in the relationship between atmospheric nitrogen deposition and forest carbon sequestration. Global Change Biology, 14, 2057–2063. Thomas, R. Q., Canham, C. D., Weathers, K. C. and Goodale, C. L. (2010). Increased tree carbon storage in response to nitrogen deposition in the US. Nature Geoscience, 3, 13–17. Tiedje, J. M. (1988). Ecology of denitrification and dissimilatory nitrate reduction to ammonium. In:€Biology of Anaerobic Microorganisms, ed. A. J. B Zehnder.€John Wiley and Sons, New€York: pp. 179–244. Tiedje, J. M., Sexton, A. J., Myrold, D. D. and Robinson, J. A. (1982). Denitrification:€ecological niches, competition and survival. Antonie van Leeuwenhoek, 48, 569–583. Tiessen, H., Stewart, J. W. and Bettany, J. R. (1982). Cultivation effects on the amounts and concentrations of carbon, nitrogen, and phosphorus in grassland soils. Agronomy Journal, 74, 831–835. Tietema, A. and Wessel, W. W. (1992). Gross nitrogen transformations in the organic layer of acid forest ecosystems subjected to increased atmospheric nitrogen input. Soil Biology and Biochemistry, 24, 943–950. Treseder, K. K. (2008). Nitrogen additions and microbial biomass:€a meta-analysis of ecosystem studies. Ecology Letters, 11, 1111–1120. Trinsoutrot, I., Recous, S., Mary, B. and Nicolardot, B. (2000). C and N fluxes of decomposing 13C and 15N Brassica napus L.:€effects of residue composition and N content. Soil Biology and Biochemistry, 32, 1717–1730. Vall, P. M. and Vidal, C. (1999). Nitrogen in Agriculture. ec.europa.eu/ agriculture/envir/report/en/nitro_en/report.htm Van Breemen, N., Boyer, E. W., Goodale, C. L. et al. (2002). Where did all the nitrogen go? Fate of nitrogen inputs to large watersheds in the northeastern U.S.A. Biogeochemistry, 57/58, 267–293. Van Cleemput, O. (1998). Subsoils:€chemo- and biological denitrification, N2O and N2 emissions. Nutrient Cycling in Agroecosystems, 52, 187–194. Van Egmond, N. D., Bresser, A. H. M. and Bouwman, A. F. (2002). The European nitrogen case. Ambio, 31, 72–78. Vanmechelen, L., Groenemans, R. and Van Ranst, E. (1997). Forest Soil Conditions in Europe: Results of a Large-Scale Soil Survey. Brussels: EC, UN/ECE, Ministry of the Flemish Community. Velthof, G., Barot, S., Bloem, J. et al. (2011). Nitrogen as a threat to European soil quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Verchot, L. V., Holmes, Z., Mulon, L., Groffman, P. M. and Lovett, G. M. (2001). Gross vs net rates of N mineralisation and nitrification as indicators of functional differences between forest types. Soil Biology and Biochemistry, 33, 1889–1901. Vinten, A. J. A., Vivian, B. J., Wright, F. and Howard, R. S. (1994). A comparative study of nitrate leaching from soils of differing textures under similar climatic and cropping conditions. Journal of Hydrology, 159, 197–213. Vitousek, P. M. and Howarth, R. W. (1991). Nitrogen limitation on land and in the sea:€how can it occur? Biogeochemistry, 13, 87–115. Vitousek, P. M., Aber, J. D., Howarth, R. W. et al. (1997). Human alteration of the global nitrogen cycle:€Sources and consequences. Ecological Applications, 7, 737–750. Vitousek, P. M., Cassman, K., Cleveland, C. et al. (2002). Towards an ecological understanding of biological nitrogen fixation. Biogeochemistry, 57, 1–45.

Klaus Butterbach-Bahl and Per Gundersen Wan, S., Hui, D. and Luo, Y. (2001). Fire effects on nitrogen pools and dynamics in terrestrial ecosystems:€a meta-analysis. Ecological Applications, 11, 1349–1365. Watson, C. J. and Mills, C. L. (1998). Gross nitrogen transformations in grassland soils as affected by previous management intensity. Soil Biology and Biochemistry, 30, 743–753. Wesselink, L. G., Meiwes, K.-J., Matzner, E. and Stein, A. (1995). Longterm changes in water and soil chemistry in spruce and beech forests, Solling, Germany. Environmental Science and Technology, 29,€51–58. Wolf, B., Zheng, X., Brüggemann, N. et al. (2010). Grazing-induced reduction of natural nitrous oxide release from continental steppe. Nature, 464, 881–884.

Wrage, N., Velthof, G. L., Beusichem van, M. L. and Oenema, O. (2001). Role of nitrifier denitrification in the production of nitrous oxide. Soil Biology and Biochemistry, 33, 1723–1732. Yamulki, S., Harrison, R. M., Goulding, K. W. T. and Webster, C. P. (1997). N2O, NO and NO2 fluxes from a grassland:€effect of soil pH. Soil Biology and Biochemistry, 29, 1199–120. Yanai, R. D., Stehman, S. V., Arthur, M. A. et al. (2003). Detecting change in forest floor carbon. Soil Science Society of America Journal, 67, 1383–1593. Zumft, W. G. (1997). Cell biology and molecular basis of denitrification. Micobiology and Molecular Biology Reviews, 61, 533–616.

125

Chapter

7

Nitrogen processes in aquatic ecosystems Lead authors: Patrick Durand, Lutz Breuer and Penny J. Johnes Contributing authors: Gilles Billen, Andrea Butturini, Gilles Pinay, Hans van Grinsven, Josette Garnier, Michael Rivett, David S. Reay, Chris Curtis, Jan Siemens, Stephen Maberly, Øyvind Kaste, Christoph Humborg, Roos Loeb, Jeroen de Klein, Josef Hejzlar, Nikos Skoulikidis, Pirkko Kortelainen, Ahti Lepistö and Richard Wright

Executive summary Nature of the problem • Freshwater ecosystems play a key role in the European nitrogen (N) cycle, both as a reactive agent that transfers, stores and processes N loadings from the atmosphere and terrestrial ecosystems, and as a natural environment severely impacted by the increase of these loadings.

Approaches • This chapter is a review of major processes and factors controlling N transport and transformations for running waters, standing waters, groundwaters and riparian wetlands.

Key findings/state of knowledge • The major factor controlling N processes in freshwater ecosystems is the residence time of water, which varies widely both in space and in time, and which is sensitive to changes in climate, land use and management. • The effects of increased N loadings to European freshwaters include acidification in semi-natural environments, and eutrophication in more disturbed ecosystems, with associated loss of biodiversity in both cases. • An important part of the nitrogen transferred by surface waters is in the form of organic N, as dissolved organic N (DON) and particulate organic N (PON). This part is dominant in semi-natural catchments throughout Europe and remains a significant component of the total N load even in nitrate enriched rivers. • In eutrophicated standing freshwaters N can be a factor limiting or co-limiting biological production, and control of both N and phosphorus (P) loading is often needed in impacted areas, if ecological quality is to be restored.

Major uncertainties/challenges • The importance of storage and denitrification in aquifers is a major uncertainty in the global N cycle, and controls in part the response of catchments to land use or management changes. In some aquifers, the increase of N concentrations will continue for decades even if efficient mitigation measures are implemented now. • Nitrate retention by riparian wetlands has often been highlighted. However, their use for mitigation must be treated with caution, since their effectiveness is difficult to predict, and side effects include increased DON emissions to adjacent open waters, N2O emissions to the atmosphere, and loss of biodiversity. • In fact, the character and specific spatial origins of DON are not fully understood, and similarly the quantitative importance of indirect N2O emissions from freshwater ecosystems as a result of N leaching losses from agricultural soils is still poorly known at the regional scale. • These major uncertainties remain due to the lack of adequate monitoring (all forms of N at a relevant frequency), especially€– but not only€– in the southern and eastern EU countries.

Recommendations • The great variability of transfer pathways, buffering capacity and sensitivity of the catchments and of the freshwater ecosystems calls for site specific mitigation measures rather than standard ones applied at regional to national scale. • The spatial and temporal variations of the N forms, the processes controlling the transport and transformation of N within freshwaters, require further investigation if the role of N in influencing freshwater ecosystem health is to be better understood, underpinning the implementation of the EU Water Framework Directive for European freshwaters.

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

126

Patrick Durand, Lutz Breuer and Penny J. Johnes

7.1╇ Introduction The scope of this chapter is to document the extent of knowledge regarding the fate of nitrogen (N) in European freshwater systems (wetlands, standing and running waters, the hyporheic zone and groundwaters). Another aim is to highlight those areas where knowledge is currently sparse and future research is required to underpin the development of sound evidencebased environmental policy for the wide range of nutrient enriched waters across Europe. The key processes and controls of nitrogen turnover in freshwaters are discussed to understand the observed trends and the impacts of these processes on the ecological status and societal value of European freshwaters. The definition of groundwater used in this chapter includes the vadose zone beyond the reach of the root system of terrestrial vegetation, running waters are considered down to the limit of their tidal influence, wetlands are restricted to riparian areas receiving surface and groundwater but including the hyporheic zone, and standing waters include all freshwater lakes, ponds, pools and reservoirs. This chapter first describes the main factors controlling nitrogen cycling in freshwater systems, the distribution of N forms in waters and their origin, and the role of N in the ecology of those systems. The specific characteristics of N cycling in different types of freshwater systems are then highlighted.

7.2╇ Factors controlling N cycling in freshwaters 7.2.1╇ The water cycle Fresh water is defined as water with less than 0.5 g/l of dissolved salts. Some 3% of the water on Earth is freshwater. About two thirds of it is frozen in polar caps and glaciers, most of the remainder is present as groundwater, and only 0.3% is surface water. Freshwater ecosystems occupy over 3% of the Earth’s surface. In Europe, freshwaters cover 1% of the surface and wetlands 0.8% (European Environment Agency, 2005). The classical figure of the water cycle (Figure€ 7.1) illustrates that one main feature of aquatic ecosystems is their

interconnectivity. Water infiltrating from terrestrial ecosystems recharges groundwater. In flat, low lying areas the groundwater table reaches the surface, determining the extent of wetlands. These wetlands are often located close to the streams or lakes (riparian wetlands). Streams and lakes are fed by ground�water discharge, but also by surface overland flow and subsurface interflow in variable proportions. The direction of fluxes between groundwater, wetlands, streams and lakes varies according to hydrological conditions. Owing to natural or artificial obstacles to flow, surface waters create standing water bodies that are generally connected to the hydrological network. They can be located at the source of the streams or along the main course, and vary widely in extension and depth. Standing waters also occur in topographic depressions in the landscape including, for example, kettle holes in some post-glacial landscapes which are often less well connected to the running water network. The main hydrological driver is the excess rainfall, defined here as the amount of water available for groundwater recharge or runoff after interception and evapotranspiration. In Europe, this excess rainfall varies from a few mm/yr in the driest Mediterranean zones to more than 1000 mm in NorthWestern coasts. In most of Europe, it is between 150€mm and 500€mm (Figure€7.2). The seasonality of lotic (running water) ecosystems is largely determined by the hydrological regime. In Europe, three major hydrological regimes exists (Figure€ 7.3):€ (1) the oceanic temperate regime, with moderate variations in mean rainfall distribution over the year, and increased evapotranspiration during summer, leading to low summer discharge and winter flooding; (2) the Mediterranean regime with low rainfall and high evapotranspiration in summer leading to extremely low summer discharge and flooding in spring and autumn; and (3) the snowmelt controlled regime (mountainous and Nordic regions) with high discharge during spring or early summer. The first two regimes are characterized by low discharge during the periods with the highest temperature and light intensity, while in the snowmelt regime, the most productive period occurs before the summer flood, under sub-optimal light and temperature conditions.

Figure€7.1 The global water cycle (courtesy of Sandra Süß).

127

Nitrogen processes in aquatic ecosystems

Lower reach systems (stream order >7) are still deeper and wider. They are no longer influenced by substantial lateral hydraulic dilution and show a decline of autochthonous primary production because of increased depth and turbidity which limits primary production capacity. They are again often heterotrophic, with stream metabolism strongly dependent on organic material brought in from upstream reaches. Sedimentation of fine material is possible and sediments are often rich in organic material.

7.2.3╇ Residence times

Figure€7.2 Map of excess rainfall in Europe, based on long term rainfall data and computed actual evapotranspiration (Mulligan et€al., 2006) (© 2006, JRC, European Commission).

7.2.2╇ Stream order The fate of nitrogen in freshwater depends strongly on the stream network geometry, which can be described in a synthetic and meaningful way by the stream order concept. The Strahler stream order system (Strahler, 1952) was proposed to define stream sizes based on a hierarchy of tributaries. Headwaters are first-order streams. When two first-order streams connect they form a second-order stream and so forth. The ecological functioning of water bodies, and therefore the N cycling, varies according to Strahler’s order as follows. Streams of Strahler’s order 1 to 3 are characterized by shallow depth and narrow width, steep slope and a relatively high contribution of lateral inputs of water with respect to the volume of the reach. Most inputs of energy are in the form of coarse organic material from riparian vegetation. Shading by riparian trees is common in these reaches, limiting light availability for stream flora. The overall metabolism of the system is typically heterotrophic, dominated by fungi and shredder invertebrates. Mid-reach river systems are wider, deeper and less strongly influenced by dilution. They receive more light so that autoÂ� chthonous primary production can occur either through macrophyte (typically stream order 4–5) or planktonic (typically stream order 5–7) development. The overall metabolism of the system becomes autotrophic. Organic matter, either of autochthonous origin or transferred from upstream systems, is dominated by fine particulate organic material and supports a community of collector or grazer invertebrates.

128

In all aquatic ecosystems, N cycling is controlled by the energy sources (light, organic matter and reduced inorganic compounds such as sulphur (S) or ferrous minerals), redox conditions (oxygen availability) and the nutrient loads. The main factor differentiating N turnover rates in the different types of freshwater systems, however, is the residence time of the water. The mean residence time of a well mixed reservoir is defined as the ratio between the volume of water (V) and the flux that goes through it (Q). Depending on the lines of flux, the actual residence time may vary considerably within a given water body:€ for example, in groundwater and lakes the residence time increases with depth, and in streams it is higher near the banks and the bed than in the middle of the stream. For running waters residence times are often defined for reaches (the portion of the stream between two confluences). For lakes and wetlands, residence time may be very short along the primary flow channels, but very long in areas less well connected to the primary flow channels. This is particularly evident in lakes formed over flooded river valleys, and in wetlands with a clear point of inflow and outflow. Typical residence times increase from order 1 streams (minutes to hours) to larger streams and wetlands (weeks), standing waters (weeks to decades) and groundwaters (decades to centuries). The highest variability in residence time is found in standing waters (from days to decades) and for groundwater (from months to thousands of years). Residence time also varies over time within a system. Bearing in mind the general definition of residence time as V/Q, since the flux is far more variable than the volume, the variability of residence time is controlled by the variations in discharge from the catchment. Therefore, �rivers with low flows in summer have residence times similar to riverine lakes, and numerous lakes and wetlands are subjected to flushing episodes during high flow events.

7.2.4╇ Nitrogen delivery to freshwaters Nitrogen can reach freshwaters through a number of pathways:€by atmospheric deposition on the catchment or directly on the water body; by leaching from diffuse sources within the catchment, such as those resulting from fertilizer and manure application; by sediment erosion of N rich soils and surface applications of manure in catchments; and by direct input from point sources such as sewage treatment works. A further source of reactive N (Nr) is nitrogen fixation. Some prokaryotes, such as some cyanobacteria, also possess the nitrogenase enzyme that allows atmospheric nitrogen to be converted into ammonia

Patrick Durand, Lutz Breuer and Penny J. Johnes

Figure€7.3 Examples of the three major hydrological regimes of European rivers. Blue bars:€rainfall; yellow bars:€evapotranspiration; brown lines:€temperature; dark lines:€specific discharge.

and thus can exploit dissolved nitrogen gas in freshwaters. A few freshwater diatoms such as Epithemia and Rhapalodia also possess this ability via endosymbiotic inclusions believed to be derived from a cyanobacterium related to Cyanothece (Prechtl et€al., 2004), although this mechanism is not widely found in European freshwaters. The nutrient load delivered to aquatic ecosystems from diffuse catchment and atmospheric sources depends strongly on the hydrological processes, particularly the relative importance of different water pathways in the transfer of the various N forms from terrestrial to aquatic systems (Figure€7.4). Overland flow is responsible for the transport of particulate forms of N and, to a lesser extent, of dissolved organic N (DON) and ammonium. In agricultural areas, dissolved inorganic N (DIN) concentrations in overland flow are usually low compared to those of subsurface water, causing dilution of DIN in streamwater during flood events (Durand and Juan Torres, 1996; Durand et€al., 1999; Kemp and Dodds, 2001). This is not always true in situations where subsurface waters are

low in DIN and where surface accumulation of N occurs, for example:€during snowmelt events when atmospheric N deposition has accumulated in the snowpack; shortly after fertilizer applications; in intensively farmed outdoor stock enterprises; in Mediterranean forested zones during floods (Bernal et€ al., 2006; Johnes, 2007a). Most N leaving the soil in the form of nitrate will be transported to adjacent water bodies by water that has infiltrated in the soil, but the conditions controlling this transfer vary depending on the relative importance of shallow pathways (interflow, return flow, shallow groundwater seepage) and of deep infiltration (Creed et€ al., 1996; Molenat et€ al., 2002). DON is mostly delivered during storm events, with N-rich soil porewater flushed to the hyporheic zone and adjacent surface waters. Less DON is transported to groundwater stores in permeable areas with deep aquifers, where the pathway is dominated by nitrate flux. In these areas unsaturated vertical flow can be very long and lead to significant accumulation of nitrate in the vadose zone. The temporal variation of nitrate fluxes

129

Nitrogen processes in aquatic ecosystems Figure€7.4 Schematic of the different flows in a headwater catchment. The colours of the arrows symbolize the age of water from the most recent (light blue) to the oldest (purple) (courtesy of Sandra Süß).

from groundwater to rivers is damped, and the mean residence time of water and solutes in these areas can be in the order of several decades (Wade et€al., 2006).

7.3╇ Nitrogen forms and sources in freshwater 7.3.1╇ Particulate and dissolved N components Nitrogen cycling in aquatic ecosystems is complex and involves a variety of N forms and associated oxidation states. Both the oxidized and reduced inorganic N species (NO2−, NO3−, NH4+, NH3) and organic N fractions (DON, PON) are commonly found in all freshwater, estuarine and coastal waters across Europe. Nitrate, nitrite, ammonium and DON are directly available for plant uptake, supporting production in both the algal and higher plant communities. In addition the gaseous forms (N2, N2O, NO) are exchanged with the atmosphere. N speciation in streams varies along a gradient of N enrichment. Nitrate is the dominant form in highly enriched rivers, whereas DON is the dominant form in less enriched rivers. DON can also be an important secondary constituent of the TN load even in the most enriched rivers. Figure€7.5 presents nitrogen species concentrations in different European streams ranging from oligotrophic to hypertrophic. Data were collated for the present work from all available national databases and both published and unpublished research on the relative proportion of total N present in the various N species forms in European rivers. Data were only included where at least the inorganic N species and total N had been determined at high sampling frequency (typically weekly to daily sampling frequency). For some sites only total organic N (TON) concentrations were reported (shown yellow on Figure€ 7.5) alongside inorganic N species concentrations. For the majority of sites, however, the full species range had been determined, including both DON and PON fractions. A total of 84 separate annual records for 57 streams were found which fulfilled these criteria. Data have been plotted along a gradient of increasing total N concentrations, from the lowest concentrations typically found in boreal headwater streams to the highest concentrations found in rivers draining the intensively farmed lowland regions of Europe (Figure€7.5).

130

Figure€7.5 Concentrations of N species in 57 European streams and rivers (an explanation of data sources is given in the text. Stream ID, sampling locations and observation periods are provided in the Supplementary Material of Chapter 7).

Two clear patterns emerge from Figure€ 7.5:€ nitrate concentrations increase in absolute terms and also as a proportion of total N along a gradient from ultra-oligotrophic to hyper-trophic waters, and mean annual DON concentrations also increase along this gradient but as a decreasing proportion of total N. Concentrations of nitrate range from <0.002 mg NO3−-N/l in the least disturbed catchments to over 14 mg NO3−-N/l in the most intensively farmed catchments. DON concentrations are initially substantially higher than nitrate concentrations in undisturbed sites, with mean annual DON concentrations ranging from <0.15 mg N/l in low nutrient status upland waters (Willetts et€al., 2004; Kortelainen et€al., 2006; Skoulikidis and Amaxidis, 2009) to over 3 mg N/l in the highly enriched rivers of lowland Britain and the Netherlands, and a maximum of 5 mg N/l in the River Ter in Spain, with maximum daily concentrations in excess of 9 mg N/l (Johnes and Burt, 1991). This pattern, with nitrate concentrations increasing at a proportionally higher rate than DON concentrations as nutrient enrichment occurs is highlighted in Figure€7.6. Ultra-oligotrophic systems, where total N concentrations are less than 1 mg N/l, are highly N limited, and inorganic N is taken up rapidly by the biota to sustain production. DON,

Patrick Durand, Lutz Breuer and Penny J. Johnes

(a)

y R

(b)

y R

Figure€7.6 Proportion of total N present in the form of (a) nitrate-N and (b) DON in European rivers along a gradient of N enrichment (an explanation of data sources is given in the text).

as a by-product of microbial breakdown of organic matter, is released to adjacent waters through flushing of soil porewaters during and after rainfall, providing a substrate for microbial metabolism in situ and a mechanism for N transport downstream. In these systems, DON typically constitutes over 60% of total N, and nitrate concentrations are typically less than 40% of total N. Above 1 mg N/l there is surplus inorganic N being introduced to the system as a result of increasing anthropogenic disturbance of the catchment and/or increasing atmospheric N deposition rates. These increasingly productive systems often generate more DON which, together with background DON, is flushed downstream together with surplus inorganic N leached from N enriched agricultural and forest soils, or delivered from point source sewage discharges and intensive livestock enterprises. This leads to a rising trend in DON concentrations, but a decrease in the proportion of total N in the form of DON; and a rising trend in nitrate concentrations with an associated increase in the proportion of total N in the form of nitrate. Thus, as catchments become enriched through anthropogenic enhancement of N inputs, the system becomes less efficient at processing these inputs, and as system inefficiency rises, inorganic N concentrations rise disproportionately.

In lowland, intensively farmed agricultural catchments DON contributes up to 30% of the total N. In upland waters, with lower TN loading rates, and a higher proportion of histosol soils in their catchments, the DON fraction can constitute up to 90% of the TN load (Kortelainen et€ al., 1997; Willetts et€ al., 2004). Overall in European streams, between 10% and 80% of the total N load, and 11% to 100% of the total dissolved N load may be in the form of DON (Mattsson et€al., 2009), for which no specific European legislation exists. DON is also not determined in most European routine water quality monitoring programmes. Evidence from a range of sources indicating a rising trend in dissolved organic carbon (DOC) concentrations in many upland waters (Evans et€ al., 2005), possibly due to Â�climate change, raises additional concerns since a rise in DON concentrations could also be expected. There is evidence from experimental and field monitoring studies that low molecular weight DON (amino acids, polyamines) is directly available for plant and algal uptake, and is a key resource in N limited oligo- to mesotrophic estuaries and freshwaters (Maberly et€al., 2002; Fong et€al., 2004). Further experimental data show that both PON and high molecular weight DON are available for microbial assimilation in both terrestrial and aquatic environments (Antia et€al., 1991; Chapin et€al., 1993; Seitzinger and Sanders, 1997; Lipson and Näsholm, 2001; Jones et€al., 2005). Further, PON can act as a food resource for aquatic filter-feeding organisms, and as an important component of C metabolism in aquatic ecosystems. Both DON and PON provide a key pathway for nitrogen transfer downstream, playing an important role in sustaining the nutrient spiral within rivers, lakes, estuaries and wetlands. Point sources of nitrogen, including discharges from industrial sources, sewage treatment works and overflow of septic tank systems can be locally significant as sources of N loading to adjacent waters, but are rarely significant contributors to the total N loading delivered to waters from their catchments where atmospheric deposition and agriculture are fully taken into account. These are the primary sources of N enrichment of European waters (Moss et€al., 1996; Johnes and Butterfield, 2002). For EU 27, total N excretion by livestock in 2000 was about the same as total fertilizer use (Oenema et€ al., 2007). Particular problems exist with intensive agricultural regions where heavily fertilized grassland and high stocking densities have led to excessive N loading to these components of the terrestrial ecosystem. It is also a problem in many upland areas of Europe where stocking densities have risen on marginal land with steep slopes, thin soils and high rates of rainfall and snowmelt where the intrinsic nutrient export potential is high. The apportionment of N sources of nitrogen in the major European catchments is described in more detail in Billen et€ al., 2011 (Chapter€13 this volume).

7.3.2╇ Gaseous N emissions from freshwater ecosystems Through the processes of nitrification and denitrification, Nr input may enhance the biogenic production of the greenhouse

131

Nitrogen processes in aquatic ecosystems

gas nitrous oxide (N2O) in standing waters like lakes and ponds (Mengis et€al., 1997), in wetlands (Johansson et€al., 2003) and in groundwaters, streams, and rivers (Seitzinger and Kroeze, 1998; Hasegawa et€al., 2000). N2O can also be leached directly into these systems, particularly from agricultural soils, producing super-saturation of N2O in the soil leachate (Dowdell et€al., 1979). Natural aquatic systems are generally considered relatively minor sources of N2O, e.g. 3%–6% of the direct agricultural emissions in the Seine basin (Garnier et€al., 2009). In fact, aquatic systems with low N availability may act as a sink for N2O (Mengis et€ al., 1997; Blicher-Mathiesen and Hoffmann, 1999; Dhondt et€al., 2004). Whether a system acts as a source or a sink depends on local circumstances including the availability of Nr, temperature and pH. Natural wetlands are likely to be relatively minor sources of N2O, due to the low availability of Nr common to these systems (Regina et€al., 1996). Relatively high nitrous oxide fluxes have been quantified for constructed and/or riparian wetlands receiving high Nr inputs. However, the evidence base is incomplete and other authors claim that N2O emissions from constructed wetlands are relatively insignificant at the landscape scale (Søvik et€ al., 2006). There is clearly a need for further research on the role of wetlands at landscape scale, whether natural or constructed, on N2O emission rates. Based on measurement and budgeting approaches in the Seine basin, the indirect N2O emissions (including the ones from wastewater treatment plants, water mirror, and riparian zones) have been evaluated as 13%–17% of the direct emissions (Garnier et€al., 2009). Lakes can be sources of sinks of N2O, but the importance of N2O emissions from lakes also remains poorly quantified. Many lakes have deep anoxic zones that are undersaturated with N2O and this is generally considered to be the result of N2O consumption by denitrifying bacteria (Mengis et€ al., 1997). In the oxic zone of lakes, incomplete denitrification can result in super-saturations of N2O and subsequent outgasing to the atmosphere. This source may be especially significant in the littoral zone of eutrophic and hyper-eutrophic lakes where variable redox conditions and plentiful supplies of organic carbon (C) lead to increased N2O production from nitrification and/or incomplete denitrification (Huttunen et€al., 2003). In headwater streams, emissions are predominantly a result of the physical outgasing of N2O-supersaturated water entering them via leaching and surface run-off. Such allochthonous N2O inputs are generally lost through outgasing within a few hundred metres of entry to open water courses (Garnier et€al., 2009). However, emissions of N2O from acid upland soils and peats were found to be a negligible proportion of deposition inputs of N in an intensive study of four headwater moorland catchments in the UK (Curtis et€al., 2006). In lowland streams and rivers, in situ production of N2O may become more important, with this autochthonous N2O being produced via nitrification and denitrification in aquatic sediments and occasionally in the water column (Dong et€al., 2004). Globally, most of the N2O emitted from aquatic systems is associated with agriculture (Kroeze et€al., 1999; Bouwman et€al., 2005). Numerous studies have now reported super-saturation of dissolved N2O in agricultural drains, aquifers, streams and

132

rivers (MacMahon and Dennehy, 1999; Hasegawa et€al., 2000; Dong et€al., 2004; Clough et€al., 2006; Beaulieu et€al., 2008), but the true importance of N2O losses from this indirect pathway compared to direct emissions from fertilized fields remains an area of considerable uncertainty (Haag and Kaupenjohann, 2001). Clearly more experimental and process-based modelling studies may be needed to better understand the processes underlying N2O formation in fresh water systems.

7.4╇ The role of N in freshwater ecosystems 7.4.1╇ Role as a nutrient The primary role of N in freshwater ecosystems is as one of the key nutrients, along with carbon, phosphorus (P) and silica (Si), required to support primary production by higher plants and algae. This process of photoautotrophy requires energy from sunlight and C, N, Si and P. Nitrogen also plays a role in determining the food web structure and relative productivity of any water body through microbial, algal and plant uptake of N in the form of both inorganic N species and DON. Any change in the rate of supply of nitrogen to a water body, or the relative abundance of C, N, Si and P, will lead to changes in the productivity of the water body and its microbial metabolism, with associated secondary effects in terms of microbial, plant and animal community species composition and relative abundance, and the structure and balance of the aquatic food web. The role of N in freshwater ecosystems is clearly illustrated through examination of the changes to ecosystem structure and function in response to nutrient enrichment. This process is termed eutrophication, defined under the EU Urban Wastewaters Treatment Directive (91/271/EEC) as the enrichment of water by nutrients, especially compounds of nitrogen and/or phosphorus, causing an accelerated growth of algae and higher forms of plant life to produce an undesirable disturbance to the balance of organisms present in the water and to the quality of the water concerned. Typical responses to the eutrophication process can range from losses in biodiversity resulting from shifts in higher plant community species composition and relative abundance; increases in phytoplankton productivity leading, in extreme examples, to dominance of N fixing cyanobacteria in the primary producer community; and shifts in food web structure. See Grizzetti et€al., 2011 (Chapter€17, this volume) for more detail. In the phytoplankton (algal) community, because of the relative abundance of C and N relative to P in natural surface waters, productivity is most commonly assumed to be limited by the availability of P rather than N or C (Hecky and Kilham, 1988; Huszar et€al., 2006). This is based on the application of Liebig’s Law of the Minimum which proposes that the biological response to resource abundance is controlled by the single most limiting resource. However, this is complicated by the interacting effects of ecological factors (Talling, 1979) and because within the phytoplankton community, different species may be limited by different factors (Hecky and Kilham, 1988; Maberly et€al., 2002). Consequently, it is likely that more than one nutrient can control the yield or rate of growth even in the phytoplankton community.

Patrick Durand, Lutz Breuer and Penny J. Johnes

In the higher plant community, rooted macrophytes obtain a large part of their nutrients from the sediment (Barko et€al., 1991), and N and P are required in a ratio of around 30:1 (Duarte, 1992; Verhoeven et€al., 1996). Higher plants require almost twice as much N as phytoplankton, relative to P. In the sediment and the sediment porewater the availability of P is much higher than in the surface water, particularly in nutrient enriched systems, and in contrast to N. Accordingly, rooted and benthic species experience a higher P availability than free floating plant species such as Lemna spp., and the free-floating phytoplankton community that derive their nutrient resources from the water column and, in the case of the cyanobacteria, through N fixation from atmospheric sources. For higher plants, therefore, N is more likely to be limiting to productivity than P, and this is confirmed by a number of research studies (Anderson and Kalff, 1986; Barko et€al., 1988) including recent work which has shown that high N concentrations seem to be correlated with a low species diversity in the macrophyte community (James et€al., 2005) and a low macrophyte abundance in lakes (Gonzalez Sagrario et€ al., 2005), see also Chapter€ 17 (Grizzetti et€al., 2011, this volume) for more details. Thus the role of N and P in limiting production from oligotrophic to eutrophic status is complex:€different elements of the biota respond to P and N loads delivered to waters in different ways, mediated by the access of each biotic group to each nutrient store within the water body. Submerged aquatic plants, and benthic algal communities in clear water lakes and streams growing on P-enriched sediments are likely to be N-limited. Emergent plant production along lake and river margins will reflect the relative abundance of N and P in the bed and bank sediments. The free-floating phytoplankton community in lakes and slow-flowing lowland river reaches will depend on the relative availability of nutrients in the water column. In all cases, the role of nutrients in limiting biological production may be over-ridden by other factors such as reduced light penetration in turbid and highly coloured waters, or shear stress and abrasion associated with high flow environments. The impact of nitrogen enrichment in European freshwaters is likely, therefore, to lead to changes in community structure and composition, as well as an increase in gross productivity in the system.

7.4.2╇ Role in water acidification In undisturbed lake and stream catchments, most N export is in the form of organic N. Spatially representative long term data bases from 21 managed and 42 unmanaged headwater boreal catchments covered by peatlands and forests across Europe, demonstrate that the long term N load is dominated by organic N in pristine and managed catchments (Kortelainen et€ al., 1997; Mattsson et€al., 2003; Kortelainen et€al., 2006). In northern European lakes and streams, concentrations of inorganic N are generally very low compared with organic N. The presence of elevated nitrate in undisturbed lake and stream catchments can indicate anthropogenic N deposition where direct catchment sources are absent (intensive agriculture, industrial or urban areas, point sources). Nitrate may originate as NOx or reduced N (through nitrification) but in either case, leached

nitrate in these catchments indicates a net input of acidity. Data from various studies suggest that NO–3 concentrations in acidsensitive lakes increased from their very low levels during the 1970s and 1980s (e.g. Grennfelt and Hultberg, 1986; Brown, 1988; Henriksen and Brakke, 1988). Later studies confirmed that nitric acid is making a major contribution to acidification in many parts of Europe and North America (Murdoch and Stoddard, 1992; Allott et€al., 1995; Wright et€al., 2001). The resulting decrease of acid neutralizing capacity and pH can significantly affect the productivity and biodiversity of aquatic ecosystems. Even where acidification is not, or no longer, chronic, episodic acidification, e.g. during snowmelt can threaten sensitive biota. Impacts have been reported on a wide range of species, including algae, macrophytes, microand macroinvertebrates, batrachians, salmonids and riverine birds (Ormerod and Durance, 2009). Acidification can also affect functions such as organic matter decomposition (Merrix et€al., 2006). While total N deposition is now declining in many parts of Europe, nitrate leaching continues to slow the chemical recovery of freshwaters from acidification, due to N saturation of the terrestrial ecosystems (Butterbach-Bahl et€al., 2011, Chapter€6 this volume). Long-term trends are notoriously difficult to detect for nitrate because nitrate concentrations vary strongly over different timescales, ranging from diurnal and seasonal (Reynolds and Edwards, 1995) to climatically driven variations linked to 5–10 yearly fluctuations in the North Atlantic Oscillation index (Monteith et€ al., 2000). Furthermore, this overall lack of general trend at the European scale was suggested to be the result of opposing factors:€ the increasing N saturation status of catchments in the context of declining N deposition in some parts of Europe (Wright et€al., 2001). Data from 54 European lake and stream sites included in the ICP Waters Programme (UNECE International Cooperative Programme on Assessment and Monitoring of Acidification of Rivers and Lakes) show that N concentrations in runoff increase as N deposition levels rise above certain thresholds (Figure€7.7). At sites with inorganic N (TIN) concentrations in precipitation below 0.25 mg N/l, runoff TIN concentrations did not exceed 0.1 mg N/l. This was typically observed at remote sites in Scandinavia and Scotland with natural or semi-natural vegetation cover. In areas with precipitation concentrations in the range of 0.25–0.7 mg N/l, TIN in runoff reached 0.4 mg N/l. This included the more polluted locations in Scandinavia together with sites in the Czech Republic, UK, and Italy. Above this deposition level, runoff concentrations spanned a wide range from below 0.3 mg N/l to more than 4 mg N/l. The latter illustrates that catchment TIN losses are highly variable among sites, depending on landscape characteristics, site history and different types of disturbance. Sites included in this group were located in Germany, the Czech Republic and in Italy. It is clear that reductions in total N deposition will have to be achieved to further reduce nitrate concentrations in seminatural catchments in Europe. What is not yet known is the degree to which nitrate leaching could increase if deposition levels are held constant. Recent evidence from an analysis of 20 years of data in the UK Acid Waters Monitoring Network shows

133

Nitrogen processes in aquatic ecosystems

a number of sites with significant increasing trends in nitrate leaching discernible above the ‘noise’ of temporal variations at various timescales (Figure€7.8), despite static or declining levels of N deposition. While other drivers (climate, de-acidification) may explain some of these increases, the possibility of a nitrogen saturation signal cannot be ruled out (Butterbach-Bahl et€al., 2011, Chapter€6 this volume). As excess sulphate concentrations have fallen in response to declining S emissions, the importance of NO3− and NH4+ in acidification and preventing recovery has increased in relative terms (N vs. S) and at some sites may continue to increase in absolute terms (Curtis et€al., 2005; Rogora and Mosello, 2007).

7.5╇ Nitrogen processing in freshwater systems Figure€7.7 Nitrogen (NO3− + NH4+) concentration in runoff and total inorganic N concentrations in deposition for 54 European ICP Waters sites in 1990–93, 1996–99, and 2002–05. Deposition data are interpolated from adjacent EMEP stations. The sites include lake and stream sites from 8 European countries ranging from the Mediterranean to Scandinavia (1 mg N/l is equivalent to about 70 µeq/l) (from Kaste et€al., 2007).

Surface freshwaters, including streams, rivers, lakes and other standing waters receive nitrogen both as non-point sources from surface and groundwater runoff from their watershed, and from point sources from direct discharge of treated or untreated urban wastewater. As a rule, nitrate concentrations and fluxes

Figure€7.8 Annual mean concentrations of excess (non-marine) sulphate and nitrate in four sites from the UK Acid Waters Monitoring Network. Data for lakes (Lochnagar, Round Loch of Glen Head, Loch Chon) based on quarterly samples, stream (Dargal Lane) monthly samples.

134

Patrick Durand, Lutz Breuer and Penny J. Johnes

in rivers are significantly lower than in the soil sources in their catchments. A part of this abatement is explained by the role of riparian wetlands in transforming nitrate to gaseous forms and to organic N stored in wetland soil porewaters and biomass, en route from catchment to stream. Another part results from in-stream processes leading to transformation of the N load in transit from headwater streams to estuarine and coastal waters as well as from groundwater to surface water. The following section will discuss the dominant N-relevant processes of the different freshwater systems.

7.5.1╇ Groundwater Groundwater (saturated) and vadose (unsaturated) zones have four major roles in the N cycle: • infiltration and groundwater flow serve to transfer, often with significant time lags, nitrogen from soils to groundwater bodies (aquifers) and in turn to surface-water bodies via groundwater baseflow; • exchange of water between surface waters and groundwaters takes place through the biogeochemically active hyporheic zone where rapid cycling of Nr may take place; • low flow-rates and long travel distances allow these zones to store large quantities of nitrogen, principally as nitrate; • long residence times and specific biogeochemical conditions may allow nitrogen processing, especially through biotic reduction processes, particularly denitrification. Groundwater is hence both a receptor of pollution and a pathway to surface water receptors. Although elevated nitrate has been recognized in groundwater since the 1870s (Addiscott, 1996), the seriousness of this occurrence was not fully appreciated until a century later (Foster et€ al., 1986; Rivett et€ al., 2007).

Residence times and storage capacity Transfer of nitrogen to the subsurface zone (Figure€ 7.4) may occur via leaching of soils, point-source discharges, particularly via septic tank systems, or river seepage through the hypoÂ� rheic zone (Dages et€al., 2008). Although some discharges, e.g. point sources, may occur in the form of ammonium or DON, nitrification to nitrate is typical in aerated unsaturated zone and shallow groundwater. Nitrate migrates down through the unsaturated zone at a slow rate with nitrate fronts discernible in many systems (Kinniburgh et€ al., 1999). In many porous or dual-porosity porous-fissured aquifers such as sandstones or chalk-limestones, rates of downward nitrate movement are typically 0.2–1 m/yr (Rivett et€ al., 2007). Hence it may take years to decades for the main nitrate front to reach deep water tables as 10–50 m thick vadose zones are not uncommon. Even where units are fissured, nitrate diffusion to a porous rock matrix will cause limited flow rates. Below the water table, lateral groundwater flow causes nitrate to spread throughout the aquifer with some flow lines ultimately discharging to receiving surface water bodies and others migrating deeper to aquifer units potentially confined by overlying low permeability units.

Velocities in groundwater are governed by the hydraulic gradient, effective porosity and hydraulic conductivity of units that may vary over orders of magnitude (Martin et€al., 2006). Flow rates are typically in the range of 10–100 m/yr. However, in highly fissured karst systems, rates may reach kilometres per day, with significantly contrasting water ages occurring in the subsurface zone (Boehlke and Denver, 1995). Storage of nitrate in the subsurface-groundwater zone, thought to be relatively small at the global scale, can locally be a significant part of a catchment nitrogen budget, even in areas underlain with crystalline bedrocks (Durand, 2004; Basset-Mens et€al., 2006). Nitrate fronts slowly migrating down through unsaturated zones tens of metres thick and of high porosity (and hence large water storage capacity) is of particular concern. The principal impact on the underlying groundwater and in turn receiving surface waters may occur decades after N application at surface. Many aquifers used for water supply or replenishment of surface waters are relatively porous (10%–40%) and where such aquifer thicknesses are large (+50€ m) storage of nitrate may be substantial (Jackson et€ al., 2007). This is evident for example in aquifers of consolidated sedimentary sandstone or chalk/limestone deposits in some lowland parts of Europe. Water quality improvements arising from any protection zone initiatives taken to reduce nitrogen inputs at surface (Johnson et€ al., 2007) may hence become apparent years to decades after their implementation (Beaudoin et€al., 2005; Jackson et€al., 2008).

Transformation processes Microbial transformation of nitrate in the subsurface-groundwater zone is hence critical to the reduction of its impacts. Heterotrophic denitrification, with organic C as an electron donor ultimately forming nitrogen gas, is generally thought to be the dominant attenuation process amongst several other processes (Rivett et€ al., 2008). The latter include autotrophic denitrification by reduced iron or sulphur, dissimilatory nitrate reduction to ammonium, anoxic oxidation of ammonium (anammox) and abiotic nitrate reduction (Korom, 1992; Burgin and Hamilton, 2007). Biogeochemical conditions, for example redox and the presence and type of electron donors, are key to establishing their relative importance. Rivett et€al. (2008) conclude that as denitrifying bacteria are essentially ubiquitous in the subsurface zone, even to hundreds of metres depth (Neilsen et€ al., 2006), the critical limiting factors are electron donor concentration and availability. Variability of other environmental conditions (nitrate concentration, nutrient availability, microbial acclimation, pH, temperature, presence of toxins and other co-contaminants such as pesticides) appear to exert secondary influences on denitrification rates in groundwaters. Although the number of case studies assessing nitrate migration and attenuation steadily increases (Einsiedl et€al., 2005; Barkle et€al., 2007; Burgin and Hamilton, 2007; Ruckart et€al., 2007; Thayalakumaran et€al., 2007; Rivett et€al., 2007) the frequency of occurrence of transformation types and their rates remain largely unknown (Korom, 1992; Seitzinger et€al.,

135

Nitrogen processes in aquatic ecosystems

2006). These studies illustrate that it is technically challenging to investigate the various denitrification/transformation proÂ� cesses in the field with several lines of supporting evidence or methods typically required to characterize bacterial presence, their activity and specificity, nitrogen species present, electron donors, reaction. Often rates estimated are typically low; however, long residence times in the subsurface may result in even low rates of transformation proving important (Rivett et€al., 2007). There is hence a crucial need to further develop methods that reliably quantify aquifer denitrification capacities to allow improved prediction of nitrate fluxes (Green et€al., 2008). The ready availability of oxygen in the unsaturated zone and to a lesser extent in unconfined, outcrop aquifer units may cause denitrification to be effectively absent or present at only low rates. For example, studies of the dual-porosity Chalk of south-east England (summarized in Rivett et€ al., 2007) demonstrated only low rates of denitrification. This was ascribed to relatively good contact with the atmosphere via its fissure network and the low level of in-situ electron donors. The potential for aerobic denitrification was, however, recognized with the existence of anaerobic micro-sites occurring within generally aerobic environments. Although low oxygen conditions may be present in the chalk matrix, the fine pore matrix may exclude bacteria (~1 µm), causing bacterial activity to be largely restricted to fracture sites where it is difficult to establish anaerobic conditions necessary for denitrification without substantial labile C inputs, typically from a pollution source (Gooddy et€ al., 2002) or perhaps river infiltration.

Importance of DOC The presence of DOC in groundwater, often thought to be the primary electron donor in denitrification, is typically low. For example, a >11 000 sample dataset covering nine different aquifer types in England and Wales indicated aquifer mean DOC varying from 0.7 to 1.8 mg/l with DOC rarely exceeding 5 mg/l. Such a concentration would drive little denitrification. The stochiometry of heterotrophic denitrification indicates that 1 mg C/l of DOC may convert 0.93 mg N/l of nitrate to N2. The DOC is, however, preferentially oxidized by dissolved oxygen that requires 1 mg DOC-C/l to convert 2.7 mg O2/l. In air-saturated groundwater (10 mg O2/l), up to about 3.8 mg DOC-C/L must therefore be oxidized before denitrification can commence (Rivett et€al., 2008). For a fully oxygenated groundwater, this represents a DOC level below which anaerobic conditions may not develop and denitrification does not occur. The greatest concentrations of DOC occur near surface, either where groundwaters are being recharged or water is discharged through riparian zones, organic-rich wetlands or hyporheic zone riverbed sediments (Smith and Lerner, 2008). The relevance of DOC inputs into aquifers for sustaining heterotrophic denitrification has recently been challenged. Siemens (2003) and Green et€al. (2008) report DOC fluxes into a sandy aquifer in northern Germany and a set of four sandy

136

aquifers across the USA, and suggest that these are insufficient to account for oxygen and nitrate consumption. In addition, Green et€al. (2008) suggest that literature denitrification rates are biased towards high rates due to (1) a selection bias to aquifers with high denitrification rates, and (2) solely attributing temporal nitrate concentration gradients to denitrification and ignoring the history of N inputs. Regardless of the actual denitrification rate, the above studies indicate that denitrification in those aquifers relied mainly on the presence of fossil stocks of organic C and reduced sulphur (Kölle et€ al., 1985). Consumption of organic C, sulphur or ferrous minerals without replenishment increases the risk of a downward oxidation of electron donors and progression of nitrate-rich water further into aquifer units. In summary, although denitrification rates in groundwater may often be low and difficult to estimate, they are nevertheless important as they may lead to some mitigation of nitrogen pollution delivery to surface waters. The latest global assessment of groundwater denitrification (Seitzinger et€al., 2006) gives values ranging from 0 to 7020 kg N/km per yr, and suggests that European aquifers account for 30% of the global groundwater denitrification, which is estimated to be 44 Tg N/yr. This estimate is based on the assumption that heterotrophic denitrification is dominant, and since it is limited by DOC supply from recharge, it is presumed to occur principally in shallower groundwater systems. At present, however, there is insufficient evidence available to support this contention, and this remains an area requiring elucidation through further research.

7.5.2╇ Riparian wetlands Status, evolution and diversity of freshwater wetlands It is believed that at least 50% of the original area of wetlands has been lost in Europe due to drainage for agriculture and urbanization. This destruction is still in progress, since almost 4% of the remaining area has been lost between 1990 and 2000 (Figure€7.9). Freshwater wetlands comprise a range of very diverse ecosystems. Wetlands dominantly fed by rain water are highly sensitive to atmospheric N deposition. An associated loss of biodiversity occurs where the rate of N deposition increases over time, as is the case in many areas of Europe (Dise et al., 2011, Chapter€20 this volume). The other types can, under certain conditions, contribute to removal of inorganic nitrogen from water (see below). The conversion of riparian wetlands to urban or agricultural use has resulted not only in an increase in size of the potential source of nutrients subject to transfer to the wider environment, but also in the degradation and loss of ecosystems that are capable of reducing or ‘buffering’ the flux of inorganic N species from terrestrial to aquatic environments. This nitrate buffering potential of wetlands was first reported in detail in the early 1970s. Subsequently, there have been many publications produced describing the numerous benefits that can be gained from wetlands, often supported by data demonstrating

Patrick Durand, Lutz Breuer and Penny J. Johnes Figure€7.9 Changes in coverage of EUNIS 10 main habitat types from 1990 to 2000 (European Environment Agency, 2005).

their high efficiency in reducing nitrate flux to adjacent surface waters (see Haycock et€al., 1997, for a review). This section discusses the potential of freshwater wetlands for regulating water quality by storing and transforming nitrogen through the performance of a variety of physical, chemical and biological functions, and the role of hydrological residence time in controlling this behaviour.

Hydrological linkage between wetlands and the catchment There are three main aspects to the hydrology of a wetland that control its ability to transform and store nutrients:€(i) the hydrological linkage between the wetland and the wider catchment, (ii) the internal hydrological regime of the wetland and (iii) hydrological pathways within a wetland (McClain et€ al., 2003). The location of a wetland in the catchment often is the crucial factor in determining its effectiveness as a zone for nitrogen transformation (Johnston et€ al., 1990). Historically, the assessment of the impact of riparian wetlands on Nr flux at the catchment scale has focused on the total area of wetlands pre���sent. However, in wetlands where the main source of water is via diffuse shallow groundwater, it is often the wetland/terrestrial interface that is the active zone in terms of the reduction of nitrate concentrations in inflowing water. The location of nitrate removal at the upslope edge of a riparian buffer zone receiving water from the hillslopes is explained by the combination of high nitrate concentrations in the runoff, high carbon concentrations in the soil (essential as a respiratory substrate for the denitrifying bacteria), and anaerobic conditions resulting from elevated water table. This combination provides optimal conditions for denitrification to occur (Pinay et€ al., 1989). Nitrate concentrations are sometimes depleted over a few metres within this zone, and the rest of the wetland often shows very low denitrification rates (Sabater et€al., 2003). As a consequence, the length of interface between the wetland and upslope sources of nutrient rich runoff is often an important factor determining the impact of the wetland on nitrate reduction. While the hydrological connectivity between wetlands and sources of nutrients in the catchment controls the degree of contact between inflowing nutrients and wetland biogeochemical cycling processes, the effectiveness of that processing depends on the internal hydrological regime and hydrological

pathways of the wetland (Pinay et€al., 2002). Wetlands have a wide range of hydrological regimes, and these are influenced by many factors, including soil type, location in the catchment and geomorphology. In soils with low permeability, if the inflow is moderate (upstream locations, baseflow conditions), the residence time of water may be long enough for anoxic conditions to develop, and denitrification may be the dominant process influencing N concentrations in wetland soil porewaters. If the rate of inflow is higher (e.g. during storm events), this function may be bypassed through flushing of soil porewater stores and generation of preferential and/or overland flow to adjacent surface waters. In this circumstance, N retention rates and/or denitrification rates will be substantially reduced. In permeable soils, if the inflow is low, oxic conditions may dominate. This is also the case if the rate of inflow is very high because residence time is not long enough for anoxic conditions to develop. In both cases, denitrification is less significant, and plant uptake with subsequent microbial decomposition of plant material leads to the conversion of nitrate to PON and DON. In high flow events, when the water table rises in these systems, lateral flushing of DON-rich water can occur, so that these wetlands do not reduce total N flux to surface waters, they merely attenuate the rate and N form of the delivery (Prior and Johnes, 2002). It follows that, in general, the actual efficiency of riparian wetlands to denitrify inflow is much less than has been reported in studies where nitrate concentrations only are considered.

Side effects of nitrate removal by wetlands Although the beneficial effects of buffer zones on nitrate flux to surface waters have been widely documented, the complete environmental assessment of increased N loads in wetlands must include the potential limitations and adverse effects (Haag and Kaupenjohann, 2001). For example, increased N2O emission rates due to higher N availability and denitrification favouring conditions are likely. Factors such as acidic conditions and low temperatures further alter the N2O/N2 ratio in the denitrification process. Increased CO2 emissions are likely as well, due to increased organic matter oxidation in these systems. Another side effect is the release of DOC and DON into the streams, as mentioned above. Finally, increased N load can damage the ecological status of the wetland. This is obvious in the case of oligotrophic wetlands, but it is also the case for

137

Nitrogen processes in aquatic ecosystems

mesotrophic wetland ecosystems, leading to reduced biodiversity and altered structure and function in the wetland ecosystem. In addition, as the wetland nutrient store reaches capacity, the nitrogen storage and attenuation function of these wetlands will be compromised.

7.5.3╇ Running waters N cycling and transformation N cycling and transformations in river systems occur in the sediment and in the water column and are mainly controlled by the hydrological conditions (Figure€7.10). Organic matter decomposition and ammonium release by heterotrophic activity in rivers is mainly controlled by the input of organic material either from primary production or from allochthonous origins (litter fall, wastewater release and transfer from upstream). Typical values of bacterial production rate measured in the water column of European rivers ranges between 0.005–0.05 mg C/l per day for uncontaminated forest streams, between 0.01–0.5 mg C/l per day for large rivers, and can reach 2.5 mg C/l per day in rivers heavily contaminated with urban sewage (Billen et€al., 1995). Taking into account a C/N ratio of 6–7 for the organic material serving as substrate, and a growth yield of 0.5, these production rates correspond roughly to an in-stream release of ammonium of 0.0008–0.008 mg N/l per day, 0.0015–0.08 mg N/l per day and 0.4 mg N/l per day respectively. Nitrification occurs in rivers where large amounts of amÂ�monium are present, most often because of untreated urban wastewater release, and can lead to severe oxygen depletion (Chestérikoff et€al., 1992). Nitrification rates as high as 0.1 mg N/l per day have been recorded in European rivers (Brion and Billen, 2000). Seeding of river water by nitrifying bacteria

occurs through soil erosion but also by sewage release (Brion and Billen, 2000; Cebron et€ al., 2004; Cebron et€ al., 2005). Under low oxygen concentration (<2 mg O2/l), ammonium oxidizing bacteria are able to make use of nitrite as an alternative electron acceptor. The process, called nitrifying denitrification, leads to the production of large amounts of N2O (Tallec et€al., 2006; Garnier et€al., 2007). Denitrification occurs in the water column of river systems only when the oxygen concentration has dropped below a threshold of about 1 mg O2/l. There exist a number of well documented situations where this happened and resulted in the complete removal of the river nitrate load (Chestérikoff et€al., 1992). Fortunately, the progress of wastewater treatment which was primarily aimed at reducing the organic matter loading and improving the oxygen status of surface water has now resulted in the disappearance of these situations of severe oxygen depletion in most major European rivers. However, denitrification, which occurs in sediments, can be responsible for an important flux of nitrate from the water column (Figure€7.11). The role of periphytonic biofilms in denitrification has been highlighted in many studies, particularly in rivers without any significant deposit of fine sediments, and where significant hyporheic flow occurs through pebble, gravel or sand beds (Boulton et€al., 1998; Claret et€al., 1998; Mermillod-Blondin et€al., 2005; Revsbech et€al., 2005). River primary production and inorganic nitrogen uptake. In streams, effects of eutrophication on primary producers are less visible than in standing waters. Hydraulic flushing prevents algal blooms in these systems and hydraulic drag restricts vertical growth of macrophytes (Biggs, 1996; Hilton et€ al., 2006). Additionally, limitation by light can be important if the sediment load of a stream is high or if the stream is shaded by riparian forest. Nevertheless, there is much evidence Figure€7.10 The nitrogen spiralling in river systems:€a general scheme of the processes affecting nitrogen species during their downstream transfer in rivers.

138

Patrick Durand, Lutz Breuer and Penny J. Johnes

Figure€7.11 Benthic denitrification rates measured with belljars or benthic chambers in European rivers of different stream order (data from Billen et€al., 1985; Chestérikoff et€al., 1992; Garban et€al., 1995; Sanchez, 1996; Thouvenot et€al., 2007; Billen et€al., 2007; de Klein, 2008).

showing that both N and P affect periphyton and phyto�� benthos growth in streams (Elser et€ al., 2007) with dense development of filamentous and/or epiphytic algae characteristic of highly eutrophicated rapidly flowing rivers. Uptake by algae or macrophytes is the main mechanism by which Nr can be transformed into organic particulate (or dissolved) form. Low molecular weight DON can also be taken up directly by plants, and can represent an important pathway to support production in higher plant and algal communities in N limited waters from lakes to estuaries and coastal waters (Antia et€al., 1991; Seitzinger and Sanders, 1997).

In-stream nitrogen retention Empirical measurement of nitrogen in-stream retention capacity has been carried out in small experimental streams under steady discharge conditions. In brief, the experiments increase the background concentration of a nutrient in stream water and monitor its downstream spatial and temporal decay (Ensign and Doyle, 2006). The results show a high variability of retention capacity, the broad picture being that this is higher for ammonium than for nitrate and much higher in pristine sites than in N-enriched sites (Marti et€al., 2004). On the scale of catchments and individual waterbodies nitrogen removal can be estimated from balancing incoming (Nin) and outflowing (Nout) fluxes. The retention fraction (%) is thus defined as 100*(1 − Nout/Nin). For whole system N retention the average hydraulic residence time is a major controlling factor (Nixon et€al., 1996; Seitzinger et€al., 2002). Kelly et€ al. (1987) proposed a model of nitrogen removal through benthic denitrification in lakes and Howarth et al. (1996) extended this approach to river reaches. The model relates nitrogen loss through denitrification to water residence time (τ), mean water column depth (z), and an average mass transfer coefficient Sn (m/yr) which can be approximated by the ratio between mean areal denitrification rate and mean nitrate

Figure€7.12 Percent N removal (100*(1 − total Nout/Nin) vs. water column depth to residence time ratio in a number of European lakes and rivers stretches. Data include Italian lakes (◆); Danish lakes (● ) (Andersen, 1977), Seine Reservoirs (■) (Garnier et€al., 1999) and a number of river stretches (□) (Cooke and White, 1987; Christensen and Sorensen, 1988; Chestérikoff et€al., 1992; de Klein, 2008). Solid lines result from the application of Kelly’s model (with Sn respectively 100, 20 and 5 m/yr). The dashed line is the empirical formula proposed by Seitzinger et€al. (2002), the dotted line is the formula proposed by de Klein (2008).

concentration in the water column, if first-order denitrification kinetics is assumed %N removed =100* Sn/(z/τ + Sn).

(7.1) The ratio z/τ is referred to as water displacement (Seitzinger et€ al., 2002); Sn is typically in the range of 5–100 m/yr in Western European rivers. Figure€7.12 shows that this model fits reasonably well the empirical estimates obtained for a number of European lakes and river reaches. Seitzinger et€al. (2002) proposed an alternate empirical expression of relation (Eq.€(7.1)) which fits both river and lake observations: %N removed = 88.45 (z/τ )–0.37 .

(7.2)

De Klein (2008) obtained a very similar relationship based on the nitrogen budget of a number of lowland ditches and river systems: %N removed = 138 (z/τ )–0.44 .

(7.3)

They also stressed that, although the proportion of N inputs that is removed in an individual river stretch is generally low (<20%, see Figure€7.11), the cumulative effect of continued N removal along the entire flow path of a drainage network can result in much higher overall N removal.

7.5.4╇ Standing waters N processing in standing waters and major controlling factors In lakes, Nr can come from direct atmospheric inputs in the form of wet and dry deposition, and N2 fixed by cyanobacteria. These inputs may be more significant in lakes, compared to rivers, given the greater surface area for exchange to take place. In lakes, N is also derived from influx of both inorganic and organic N from catchment sources delivered by surface and subsurface flow pathways and in some geological environments,

139

Nitrogen processes in aquatic ecosystems

predominantly from groundwater stores. Internal loading derives from the microbial decomposition of N stored in aquatic plant and algal biomass and in lake sediment stores. Turnover of N in standing waters is controlled by the population dynamics of pelagic (open water), littoral (shoreline) and benthic (bottom dwelling) organisms, with the relative significance of these different groups varying among lake types. For example, in shallow lakes with a high littoral (shallow water) to profundal (deep water) ratio, N turnover is dominated by inorganic and organic N uptake by aquatic plants and organic N decomposition by the microbial community. In deeper lakes, with a low littoral to profundal ratio the free-floating planktonic community will be dominant and exert substantial control over N turnover in the lake. The balance of control is also influenced by water transparency (controlling the depth of light penetration), macro-nutrient availability (N, P and C, derived from external sources, internal stores and sediment-water column interactions), the vertical mixing regime (depending on currents, depth and climate), and the residence time for water entering the lake which determines the length of contact time between N entering the lake and the lake biota. The mixing regime in particular is a specific feature of standing waters as well as large, slow rivers:€while the typical pattern of temperate lakes is a dimictic regime with a winter stratification (cooler water at surface) and a summer one (cooler water at the bottom), some lakes are monomictic (stratified only in summer), polymictic (many mixes, e.g. shallow lakes and ponds), or meromictic (never fully mixed because of a stagnant bottom layer). These regimes affect the sensitivity of water bodies to eutrophication and the nitrogen cycling dynamics:€for example, in poorly mixed hypertrophic lakes, the hypolimnion (bottom layer including the sedimentwater interface) may become anoxic, so denitrification and ammonium accumulation are promoted.

Nitrogen loadings in European lakes and the question of N versus P limitation The N concentrations in lakes vary widely, depending on the relative rates of atmospheric N deposition, and the intensity of

human perturbation of land in the surrounding catchment; N relative to P loading is also highly variable, depending on the relative contribution of atmospheric and diffuse nutrient delivery from atmospheric and terrestrial sources (N rich) and point source (particularly sewage treatment works, P rich) (Johnes, 2007b). Data collated by the International Lake Environment Committee contains information on 40 European lakes (Figure€7.13), including records for varying time periods for 16 large European lakes for which total N and P loading rates are reported (Table€7.1). Annual loads to the catchment vary from <0.02 kg N/ha for Bodensee to 29 kg N/ha for Lake Paajarvi in Finland. Of these, 25% of lakes were reported as N limited or N and P co-limited, compared to over 75% P limited lakes. The question of N vs. P limitation in lakes has been largely debated in the literature, and what emerges from this debate is a more complex picture than the common idea that P is the cause of eutrophication in freshwaters. Although it is true that P limitation is most commonly reported, N and P co-limitation and N limitation are not rare, at least periodically, and may well occur in many European standing water bodies (Moss et€ al., 1996; Maberly et€al., 2002; James et€al., 2003). In cases of N and P enrichment, silica may be also limiting after diatom growth, giving way to possibly undesirable non-diatoms (Schelske and Stoermer, 1971). Note that N limitation or co-limitation can occur in different contexts. • Exceptionally P rich lakes, and lakes with very low N loads. In both cases, this may result in production by both the phytoplankton community in the open water areas, and the macrophyte community rooted in P rich sediment being limited by N rather than P availability. • Shallow, well mixed water bodies:€since the main store of P is in sediments, it is more readily available in shallow lakes and internal loading of P from these stores makes a substantial contribution to P availability in the open water areas. The diffusion to upper layers in deep, stratified lakes is much slower, typically resulting in P limitation of primarily phytoplankton production in the euphotic zone (where primary production is not limited by light availability). Figure€7.13 European lakes included in the World Lakes Database (International Committee Lake Foundation, 2010).

140

Patrick Durand, Lutz Breuer and Penny J. Johnes Table€7.1 Reported TN and TP loadings for large European lakes (International Committee Lake Foundation, 2010)

Lake

Country

Catchment area (km2)

TN loading (t/yr)

Lake Constanz

Germany

10 900

Lake Slapy

Czech

12 900

18 200

Lac d’Annecy

France

278

550

Lough Derg

Ireland

10 280

6245

215

Lac Leman

France

7975

1300

1350

Lake Balaton

Hungary

5181

3484

314

10.0

co-limited

Lago Maggiore

Italy

6387

11 000

550

20.0

P

422

21.0

Lago Trasimeno

Italy

396

433

Lake Tyrifjordan

Norway

9808

1600

Lake Mjosa

Norway

16 567

4430

Lake Hjalmaren

Sweden

3575

2750

Lake Malaren

Sweden

21 640

13 400

Lake Vanem

Sweden

41 182

17 380

Lake Vattern

Sweden

4530

3370

Lake Paajarvi

Finland

2550

Lake Paijanne

Finland

25 400

5475

Lake Pielinen

Finland

12 823

2600

While most of the studies have focused on limitation of phyto� plankton production, similar results have been found for periphyton (Maberly et€al., 2002) and for the macrophyte community (James et€al., 2005). A representative example of these findings is the recent paper by Phillips et€ al. (2008) who studied the relationship between growing season chlorophyll a (as a function of phyto� plankton production), total P and total N concentrations in over 1000 European lakes under the EU REBECCA programme (Figure€7.14). The database was dominated by lakes from north and central Europe, but a number of Mediterranean sites were also included. Phillips and colleagues concluded that growing season chlorophyll a concentrations were significantly related to both TN and TP and that although TP was the best descriptor of growing season chlorophyll a for the whole lake database, TN was a better predictor than TP for humic high alkalinity and very shallow polyhumic lakes, suggesting N limitation of phytoplankton productivity in the growing season in these lake types. High N loads have also been reported to be responsible for loss of biodiversity, especially for submerged plants. In a range of UK and Polish lakes the richest submerged plant

2.3 247 39.3

N:P ratio 9.13

N or P limited co-limited

73.7

P

14.0

P

29.1 0.96

P N

10.3

co-limited

22.9

co-limited

17.2

P

42.3

P

590

22.7

P

880

19.8

P

65.0

51.9

P

22.5

30.3

P

183

30.0

P

114

22.8

P

70.0 257 65.0

74.3

• Summer N limitation in eutrophic lakes where the earlier algal blooms have exhausted the dissolved N in the water column, and delivery from diffuse catchment sources is at its annual minimum. • Lakes in either Mediterranean or Nordic environments, where both N and P loadings (mainly atmospheric) are very low (Camacho et€al., 2003; Rekolainen et€al., 2004; Bergstroem and Jansson, 2006).

TP loading (t/yr)

communities were associated with winter nitrate concentrations not exceeding 2 mg N/l (James et€ al., 2005) and the authors propose this as an appropriate target concentration for enriched shallow European lakes to reach ‘good ecological status’ under the EU Water Framework Directive. Many shallow north temperate European lakes have concentrations well in excess of this limit. Van der Molen et€al. (1998) for example, reported that 65% of over 200 Dutch lakes had a summer mean total N concentration higher than 2.2 mg N/l. They argue that an appropriate target to protect against ecological deterioration and support recovery of eutrophic lakes is a summer mean concentration of 1.35 mg N/l. In only 30% of lakes studied by Moss et€al. (1996) was mean summer total N concentration below this target figure. It is clear that reduction of both N and P is required in shallow eutrophicated waters if both gross productivity is to be reduced and a stable, diverse submerged plant community is to be restored (James et€al., 2005). Jeppesen et€al. (2005) concluded that to improve the ecological status of shallow lakes it may be necessary to control both P and N loading rates. They reported that where external N loading has been reduced to the lakes in their study, the response time in terms of lake total N concentrations is relatively short (<5 years) compared to the response time for lake total P concentrations in relation to external P loading (10–15 years). Jeppesen and colleagues attributed these differences to the relative importance of the sedimentary P store to the lake P budget. These results also suggest that management of freshwaters needs to take a more holistic view of nutrient control than just relying on controlling the supply of P.

141

Nitrogen processes in aquatic ecosystems

as it applies to surface freshwaters, and the EU Groundwater Directive (2006/118/EC) designed to ensure protection of groundwater and groundwater-dependent wetlands against pollution and deterioration in terms of both quantity and chemical quality. The enrichment of European freshwater systems through increases in nitrogen export has compromised a wide range of ecosystem structures, functioning and services and will need to be brought under control through coordinated measures. Consideration will need to be given to a much wider range of sources, practices, and pathways for N delivery to waters and of the total N load delivered and transported to waters not merely the inorganic N fraction if the problem is to be fully addressed. Investigating all effects and managing the cascade of aquatic nitrogen on ecosystem services has not been conducted yet in a comprehensive way and there is a clear need for further research in this respect.

Acknowledgements This chapter was prepared with the support of the NinE Programme of the European Science Foundation, the NitroEurope IP (funded by the European Commission) and the COST Action 729.

Supplementary materials Supplementary materials (as referenced in the chapter) are available online through both Cambridge University Press:€ www.cambridge.org/ena and the Nitrogen in Europe website:€ www.nine-esf.org/ena.

References

Figure€7.14 Relationship between growing season chlorophyll a, total P and total N concentrations in over 1000 European lakes (Phillips et€al., 2008).

7.6╇ Conclusions Nitrogen enrichment of freshwaters has been taking place throughout Europe, from the advent of modern society, but particularly in the latter half of the twentieth century, where the increasing intensity of agricultural production together with increased industrial and traffic-associated emissions as well as changes in societal management of water and wastewater have led to increased inputs of all plant nutrients to freshwater systems. This enrichment has led to alteration of a wide range of aquatic ecosystem functions including the productivity of water bodies and their microbial metabolism, the microbial, plant and animal community species composition and their relative abundance, and the structure and balance of the aquatic food web. In terms of ecosystem function, there are relatively few European freshwaters where this has not been altered as a result of N enrichment, either from land-based or atmospheric sources, posing a problem in terms of achieving compliance with the EU Water Framework Directive (2000/60/EC)

142

Addiscott, T. M. (1996). Fertilizers and nitrate leaching. In Issues in Environmental Science and Technology 5, ed. R. E. Hester and R. M. Harrison. The Royal Society of Chemistry, Cambridge: pp. 1–26. Allott, T. E. H., Curtis, C. J., Hall, J., Harriman, R. and Battarbee, R. W. (1995). The impact of nitrogen deposition on upland surface waters in Great Britain:€a regional assessment of nitrate leaching. Water, Air and Soil Pollution, 85, 297–302. Andersen, J. M. (1977). Rates of denitrification of undisturbed sediment from six lakes as a function of nitrate concentration, oxygen and temperature. Archiv für Hydrobiologie, 80, 147–149. Anderson, M. R. and Kalff, J. (1986). Nutrient limitation of Myriophyllum spicatum growth in situ. Freshwater Biology, 16, 735–743. Antia, N. J., Harrison, P. J. and Oliveira, L. (1991). Phycological reviews 2:€The role of dissolved organic nitrogen in phytoplankton nutrition, cell biology and ecology. Phycologia, 30, 1–89. Barkle, G., Clough, T. and Stenger, R. (2007). Denitrification capacity in the vadose zone at three sites in the Lake Taupo catchment, New Zealand. Australian Journal of Soil Research, 45, 91–99. Barko, J. W., Gunnison, D. and Carpenter, S. R. (1991). Sediment interactions with submersed macrophyte growth and community dynamics. Aquatic Botany, 41, 41–65. Barko, J. W., Smart, R. M., McFarland, D. G. and Chen, R. L. (1988). Interrelationships between the growth of Hydrilla verticillata (L.€f.) Royle and sediment nutrient availability. Aquatic Botany, 32, 205–216. Basset-Mens, C., Anibar, L., Durand, P. and van der Werf, H. M. G. (2006). Spatialised fate factors for nitrate in catchments:€modelling

Patrick Durand, Lutz Breuer and Penny J. Johnes approach and implication for LCA results. Science of the Total Environment, 367, 367–382. Beaudoin, N., Saad, J. K., Van Laethem, C. et€al. (2005). Nitrate leaching in intensive agriculture in Northern France:€effect of farming practices, soils and crop rotations. Agriculture, Ecosystems and Environment, 111, 292–310. Beaulieu, J. J., Arango, C. P., Hamilton, S. K. and Tank, J. L. (2008). The production and emission of nitrous oxide from headwater streams in the Midwestern United States. Global Change Biology, 14, 878–894. Bergstroem, A. K. and Jansson, M. (2006). Atmospheric nitrogen deposition has caused nitrogen enrichment and eutrophication of lakes in the northern hemisphere. Global Change Biology, 12, 635–643. Bernal, S., Butturini, A. and Sabater, F. (2006). Inferring nitrate sources through end member mixing analysis in an intermittent Mediterranean stream. Biogeochemistry, 81, 269–289. Biggs, B. J. F. (1996). Hydraulic habitat of plants in streams. Regulated Rivers – Research and Management, 12, 131–144. Billen, G., Décamps, H., Garnier, J. et€al. (1995). Atlantic river systems of Europe. Ecosystems of the World, 22, 389–418. Billen, G., Somville, M., DeBecker, E. and Servais, P. (1985). Nitrogen budget of the Scheldt hydrographical basin. Netherlands Journal of Sea Research, 19, 223–230. Billen, G., Garnier, J., Némery, J. et€al. (2007). A long-term view of nutrient transfers through the Seine river continuum. Science of the Total Environment, 375, 80–97. Billen, G., Silvestre, M., Grizzetti, B. et€al. (2011). Nitrogen flows from European watersheds to coastal marine waters. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Blicher-Mathiesen, G. and Hoffmann, C. C. (1999). Denitrification as a sink for dissolved nitrous oxide in a freshwater riparian fen. Journal of Environmental Quality, 28, 257–262. Boehlke, J. K. and Denver, J. M. (1995). Combined use of groundwater dating, chemical, and isotopic analyses to resolve the history and fate of nitrate contamination in two agricultural watersheds, Atlantic Coastal Plain, Maryland. Water Resources Research, 31, 2319–2339. Boulton, A. J., Findlay, S., Marmonier, P., Stanley, E. H. and Valett, H.€M. (1998). The functional significance of the hyporheic zone in streams and rivers. Annual Reviews in Ecology and Systematics, 29, 59–81. Bouwman, A. F., Drecht, G. and Hoek, K. W. (2005). Surface N balances and reactive N loss to the environment from global intensive agricultural production systems for the period 1970–2030. Pedosphere, 15, 137–155. Brion, N. and Billen, G. (2000). Wastewater as a source of nitrifying bacteria in river systems:€the case of the River Seine downstream from Paris. Water Research, 34, 3213–3221. Brown, D. J. A. (1988). Effect of atmospheric N deposition on surface water chemistry and the implications for fisheries. Environmental Pollution, 54, 275–284. Burgin, A. J. and Hamilton, S. K. (2007). Have we overemphasized the role of denitrification in aquatic ecosystems? A review of nitrate removal pathways. Frontiers in Ecology and the Environment, 5, 89–96. Butterbach-Bahl, K., Gundersen, P., Ambus, P. et€al. (2011). Nitrogen processes in terrestrial ecosystems. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press.

Camacho, A., Wurtsbaugh, W. A., Miracle, M. R., Armengol, X. and Vicente, E. (2003). Nitrogen limitation of phytoplankton in a Spanish karst lake with a deep chlorophyll maximum:€a nutrient enrichment bioassay approach. Journal of Plankton Research, 25,€397. Cebron, A., Coci, M., Garnier, J. and Laanbroek, H. J. (2004). Denaturing gradient gel electrophoretic analysis of ammoniaoxidizing bacterial community structure in the lower Seine River:€impact of Paris wastewater effluents. Applied and Environmental Microbiology, 70, 6726. Cebron, A., Garnier, J. and Billen, G. (2005). Nitrous oxide production and nitrification kinetics by natural bacterial communities of the lower Seine river (France). Aquatic Microbial Ecology, 41, 25–38. Chapin, F. S., III, Moilanen, L. and Kielland, K. (1993). Preferential use of organic nitrogen for growth by a nonmycorrhizal arctic sedge. Nature, 361, 150–153. Chestérikoff, A., Garban, B., Billen, G. and Poulin, M. (1992). Inorganic nitrogen dynamics in the river Seine downstream from Paris (France). Biogeochemistry, 17, 147–164. Christensen, P. B. and Sorensen, J. (1988). Denitrification in sediment of lowland streams:€Regional and seasonal variation in Gelbaek and Rabis Baek, Denmark. FEMS Microbiology Letters, 53, 335–344. Claret, C., Marmonier, P. and Bravard, J. P. (1998). Seasonal dynamics of nutrient and biofilm in interstitial habitats of two contrasting riffles in a regulated large river. Aquatic Sciences – Research Across Boundaries, 60, 33–55. Clough, T. J., Bertram, J. E., Sherlock, R. R., Leonard, R. L. and Nowicki, B. L. (2006). Comparison of measured and EF5-r derived N2O fluxes from a spring-fed river. Global Change Biology, 12, 352–363. Cooke, J. G. and White, R. E. (1987). The effect of nitrate in stream water on the relationship between denitrification and nitrification in a stream-sediment microcosm. Freshwater Biology, 18, 213–226. Creed, I. F., Band, L. E., Foster, N. W. et€al. (1996). Regulation of nitrate-N release from temperate forests:€a test of the N flushing hypothesis. Water Resources Research, 32, 3337–3354. Curtis, C. J., Evans, C. D., Helliwell, R. C. and Monteith, D. T. (2005). Nitrate leaching as a confounding factor in chemical recovery from acidification in UK upland waters. Environmental Pollution, 137, 73–82. Curtis, C. J., Emmett, B. A., Reynolds, B. and Shilland, J. (2006). How important is N2O production in removing atmospherically deposited nitrogen from UK moorland catchments? Soil Biology and Biochemistry, 38, 2081–2091. Dages, C., Voltz, M. and Ackerer, P. (2008). Parameterization and evaluation of a three-dimensional modelling approach to water table recharge from seepage losses in a ditch. Journal of Hydrology, 348, 350–362. de Klein, J. (2008). From ditch to delta, nutrient retention in running waters. Ph.D. thesis, Wageningen University, Wageningen, The Netherlands. Dhondt, K., Boeckx, P., Hofman, G. and Van Cleemput, O. (2004). Temporal and spatial patterns of denitrification enzyme activity and nitrous oxide fluxes in three adjacent vegetated riparian buffer zones. Biology and Fertility of Soils, 40, 243–251. Dise, N. B., Ashmore, M., Belyazid, S. et€al. (2011). Nitrogen as a threat to European terrestrial biodiversity. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press.

143

Nitrogen processes in aquatic ecosystems Dong, L. F., Nedwell, D. B., Colbeck, I. and Finch, J. (2004). Nitrous oxide emission from some English and Welsh rivers and estuaries. Water, Air and Soil Pollution:€Focus, 4, 127–134. Dowdell, R. J., Burford, J. R. and Crees, R. (1979). Losses of nitrous oxide dissolved in drainage water from agricultural land. Nature, 278, 342–343. Duarte, C. M. (1992). Nutrient concentration of aquatic plants:€patterns across species. Limnology and Oceanography, 37, 882–889. Durand, P. (2004). Simulating nitrogen budgets in complex farming systems using INCA:€calibration and scenario analyses for the Kervidy catchment (W. France). Hydrology and Earth System Sciences, 8, 793–802. Durand, P. and Juan Torres, J. L. (1996). Solute transfer in agricultural catchments:€the interest and limits of mixing models. Journal of Hydrology, 181, 1–22. Durand, P., Cros-Cayot, S., Gascuel-Odoux, C . and Heddadj, D. (1999). Solute concentrations of overland flow water in a cultivated field:€spatial variations, intra- and inter-storm trends. Hydrological Processes, 13, 1465–1477. Einsiedl, F., Maloszewski, P. and Stichler, W. (2005). Estimation of denitrification potential in a karst aquifer using the 15N and 18O isotopes of NO3–. Biogeochemistry, 72, 67–86. Elser, J. J., Bracken, M. E. S., Cleland, E. E. et€al. (2007). Global analysis of nitrogen and phosphorus limitation of primary producers in freshwater, marine and terrestrial ecosystems. Ecology Letters, 10, 1135–1142. Ensign, S. H. and Doyle, M. W. (2006). Nutrient spiraling in streams and river networks. Journal of Geophysical Research, 111, G04009, doi:10.1029/2005JG000114. European Environment Agency (2005). The European Environment€– State and Outlook 2005. Copenhagen, http://www.eea.europa.eu/, Surface coverage data, June 2010. Evans, C. D., Monteith, D. T. and Cooper, D. M. (2005). Long-term increases in surface water dissolved organic carbon:€observations, possible causes and environmental impacts. Environmental Pollution, 137, 55–71. Fong, P., Fong, J. J. and Fong, C. R. (2004). Growth, nutrient storage, and release of dissolved organic nitrogen by Enteromorpha intestinalis in response to pulses of nitrogen and phosphorus. Aquatic Botany, 78, 83–95. Foster, S. S. D., Bridge, L. R., Geake, A. K., Lawrence, A. R. and Parker, G. R. (1986): The groundwater nitrate problem. In Hydrogeology Report No. 86/2, ed. Keyworth, British Geological Survey. Garban, B., Ollivon, D., Poulin, M., Gaultier, V. and Chesterikoff, A. (1995). Exchanges at the sediment-water interface in the river Seine, downstream from Paris. Water Research, 29, 473–481. Garnier, J., Billen, G. and Cebron, A. (2007). Modelling nitrogen transformations in the lower Seine river and estuary (France):€impact of wastewater release on oxygenation and N2O emission. Hydrobiologia, 588, 291–302. Garnier, J., Billen, G. and Palfner, L. (1999). Understanding the oxygen budget and related ecological processes in the river Mosel:€the RIVERSTRAHLER approach. Hydrobiologia, 410, 151–166. Garnier, J., Billen, G., Vilain, G. et€al. (2009). Nitrous oxide (N2O) in the Seine river and basin:€observations and budgets. Agriculture, Ecosystems and Environment, 133, 223–233. Gonzalez Sagrario, M. A., Jeppesen, E., Goma, J. et€al. (2005). Does high nitrogen loading prevent clear-water conditions in shallow lakes at moderately high phosphorus concentrations? Freshwater Biology, 50, 27–41.

144

Gooddy, D. C., Clay, J. W. and Bottrell, S. H. (2002). Redox-driven changes in porewater chemistry in the unsaturated zone of the chalk aquifer beneath unlined cattle slurry lagoons. Applied Geochemistry, 17, 903–921. Green, C. T., Puckett, L. J., Bohlke, J. K. et€al. (2008). Limited occurrence of denitrification in four shallow aquifers in agricultural areas of the United States. Journal of Environmental Quality, 37, 994–1009. Grennfelt, P. and Hultberg, H. (1986). Effects of nitrogen deposition on the acidification of terrestrial and aquatic ecosystems. Water, Air and Soil Pollution, 30, 945–963. Grizzetti, B., Bouraoui, F., Billen, G. et€al. (2011). Nitrogen as a threat to European water quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Haag, D. and Kaupenjohann, M. (2001). Landscape fate of nitrate fluxes and emissions in Central Europe. A critical review of concepts, data, and models for transport and retention. Agriculture, Ecosystems and Environment, 86, 1–21. Hasegawa, K., Hanaki, K., Matsuo, T. and Hidaka, S. (2000). Nitrous oxide from the agricultural water system contaminated with high nitrogen. Chemosphere€– Global Change Science, 2, 335–345. Haycock, N. E., Burt, T. P., Goulding, K. W. T. and Pinay, G. (1997). Buffer zones:€their processes and potential in water protection. In: Proceedings of the International Conference on Buffer Zones, September 1996. Harpenden, U.K: Quest Environmental. Hecky, R. E. and Kilham, P. (1988). Nutrient limitation of phytoplankton in freshwater and marine environments:€a review of recent evidence on the effects of enrichment. Limnology and Oceanography, 33, 796–822. Henriksen, A. and Brakke, D. F. (1988). Increasing contributions of nitrogen to the acidity of surface waters in Norway. Water, Air and Soil Pollution, 42, 183–201. Hilton, J., O’Hare, M., Bowes, M. J. and Jones, J. I. (2006). How green is my river? A new paradigm of eutrophication in rivers. Science of the Total Environment, 365, 66–83. Howarth, R. W., Billen, G., Swaney, D. et€al. (1996). Regional nitrogen budgets and riverine N and P fluxes for the drainages to the North Atlantic Ocean:€natural and human influences. Biogeochemistry, 35, 75–139. Huszar, V. L. M., Caraco, N. F., Roland, F. and Cole, J. (2006). Nutrient–chlorophyll relationships in tropical–subtropical lakes:€do temperate models fit? Biogeochemistry, 79, 239–250. Huttunen, J. T., Juutinen, S., Alm, J. et€al. (2003). Nitrous oxide flux to the atmosphere from the littoral zone of a boreal lake. Journal of Geophysical Research, 108, 4421, doi:10.1029/2002JD002989. International Committee Lake Foundation (2010). World Lakes Database. http://www.ilec.or.jp, June 2010. Jackson, B. M., Wheater, H. S., Wade, A. J. et€al. (2007). Catchmentscale modelling of flow and nutrient transport in the Chalk unsaturated zone. Ecological Modelling, 209, 41–52. Jackson, B. M., Browne, C. A., Butler, A. P. et€al. (2008). Nitrate transport in Chalk catchments:€monitoring, modelling and policy implications. Environmental Science and Policy, 11, 125–135. James, C., Fisher, J. and Moss, B. (2003). Nitrogen driven lakes:€The Shropshire and Cheshire Meres? Archiv für Hydrobiologie, 158, 249–266. James, C., Fisher, J., Russel, V., Collings, S. and Moss, B. (2005). Nitrate availability and hydrophyte species richness in shallow lakes. Freshwater Biology, 50, 1049–1063.

Patrick Durand, Lutz Breuer and Penny J. Johnes Jeppesen, E., Sondergaard, M., Jensen, J. P. et€al. (2005). Lake responses to reduced nutrient loading: an analysis of contemporary long-term data from 35 case studies. Freshwater Biology, 50, 1747–1771. Johansson, A. E., Klemedtsson, K., Klemedtsson, L. and Svensson,€B.€H. (2003). Nitrous oxide exchanges with the atmosphere of a constructed wetland treating wastewater. Tellus B, 55, 737–750. Johnes, P. J. (2007a). Meeting ecological restoration targets in European waters:€a challenge for animal agriculture. In:€Redesigning Animal Agriculture, ed. D. Swain, E. Charmley, J.€Steel and S. Coffey, Wallingford, UK:CAB International, Chapter€11, pp. 185–203. Johnes, P. J. (2007b). Uncertainties in annual riverine phosphorus load estimation:€Impact of load estimation methodology, sampling frequency, baseflow index and catchment population density. Journal of Hydrology, 332, 241–258. Johnes, P. J. and Burt, T. P. (1991). Water quality trends in the Windrush catchment:€nitrogen speciation and sediment interactions. IAHS Publication, 203, 349–357. Johnes, P. J. and Butterfield, D. (2002). Landscape, regional and global estimates of nitrogen flux from land to ocean:€errors and uncertainties. Biogeochemistry, 57/58, 429–476. Johnson, D., Fletcher, S. W. and Bray, K. (2007). Integrated assessment of nitrate concentrations for the designation of groundwater Nitrate Vulnerable Zones in England and Wales. Quarterly Journal of Engineering Geology and Hydrogeology, 40, 407. Johnston, C. A., Detenbeck, N. E. and Niemi, G. J. (1990). The cumulative effect of wetlands on stream water quality and quantity. A landscape approach. Biogeochemistry, 10, 105–141. Jones, D. L., Healey, J. R., Willett, V. B., Farrar, J. F. and Hodge, A. (2005). Dissolved organic nitrogen uptake by plants: an important N uptake pathway? Soil Biology and Biochemistry, 37, 413–423. Kaste, Ø., de Wit, H., Skjelkvåle, B. L. and Høgåsen, T. (2007). Nitrogen runoff at ICP Waters sites 1990–2005:€Increasing importance of confounding factors? In:€Trends in Surface Water Chemistry and Biota: The Importance of Confounding Factors, ICP Waters report 87/2007. NIVA serial.no. 5385, ed. H. de Wit and B. L. Skjelkvåle, pp. 29–38. Kelly, C. A., Rudd, J. W. M., Hesslein, R. H. et€al. (1987). Prediction of biological acid neutralization in acid-sensitive lakes. Biogeochemistry, 3, 129–140. Kemp, M. J. and Dodds, W. K. (2001). Spatial and temporal patterns of nitrogen concentrations in pristine and agriculturally-influenced prairie streams. Biogeochemistry, 53, 125–141. Kinniburgh, D. G., Gale, I. N., Gooddy, D. C. et€al. (1999). Denitrification in the Unsaturated Zones of the British Chalk and Sherwood Sandstone Aquifers, British Geological Survey Technical Report, WD/99/2. Kölle, W., Strebel, O. and Böttcher, J. (1985). Formation of sulphate by microbial denitrification in a reducing aquifer. Water Supply, 3,€35–40. Korom, S. F. (1992). Natural denitrification in the saturated zone:€a review. Water Resources Research, 28, 1657–1668. Kortelainen, P., Saukkonen, S. and Mattsson, T. (1997). Leaching of nitrogen from forested catchments in Finland. Global Biogeochemical Cycles, 11, 627–638. Kortelainen, P., Mattsson, T., Finer, L. et€al. (2006). Controls on the export of C, N, P and Fe from undisturbed boreal catchments, Finland. Aquatic Sciences – Research across Boundaries, 68, 453–468. Kroeze, C., Mosier, A. and Bouwman, L. (1999). Closing the global N2O budget:€a retrospective analysis 1500–1994. Global Biogeochemical Cycles, 13, 1–8.

Lipson, D. and Näsholm, T. (2001). The unexpected versatility of plants:€organic nitrogen use and availability in terrestrial ecosystems. Oecologia, 128, 305–316. Maberly, S. C., King, L., Dent, M. M., Jones, R. I. and Gibson, C. E. (2002). Nutrient limitation of phytoplankton and periphyton growth in upland lakes. Freshwater Biology, 47, 2136–2152. MacMahon, P. B. and Dennehy, K. F. (1999). N2O emissions from a nitrogen-enriched river. Environmental Science and Technology, 22, 21–25. Marti, E., Aumatell, J., Gode, L., Poch, M. and Sabater, F. (2004). Nutrient retention efficiency in streams receiving inputs from wastewater treatment plants. Journal of Environmental Quality, 33, 285–293. Martin, C., Molenat, J., Gascuel-Odoux, C. et€al. (2006). Modelling the effect of physical and chemical characteristics of shallow aquifers on water and nitrate transport in small agricultural catchments. Journal of Hydrology, 326, 25–42. Mattsson, T., Finér, L., Kortelainen, P. and Sallantaus, T. (2003). Brook water quality and background leaching from unmanaged forested catchments in Finland. Water, Air and Soil Pollution, 147, 275–297. Mattsson, T., Kortelainen, P., Laubel, A. et€al. (2009). Export of dissolved organic matter in relation to land use along a European climatic gradient. Science of the Total Environment, 407, 1967–1976. McClain, M. E., Boyer, E. W., Dent, C. L. et€al. (2003). Biogeochemical hot spots and hot moments at the interface of terrestrial and aquatic ecosystems. Ecosystems, 6, 301–312. Mengis, M., Gächter, R. and Wehrli, B. (1997). Sources and sinks of nitrous oxide (N2O) in deep lakes. Biogeochemistry, 38, 281–301. Mermillod-Blondin, F., Mauclaire, L. and Montuelle, B. (2005). Use of slow filtration columns to assess oxygen respiration, consumption of dissolved organic carbon, nitrogen transformations, and microbial parameters in hyporheic sediments. Water Research, 39, 1687–1698. Merrix, F. L., Lewis, B. R. and Ormerod, S. J. (2006). The effects of low pH and palliative liming on beech litter decomposition in acidsensitive streams. Hydrobiologia, 571, 373–381. Molenat, J., Durand, P., Gascuel-Odoux, C., Davy, P. and Gruau, G. (2002). Mechanisms of nitrate transfer from soil to stream in an agricultural watershed of French Brittany. Water, Air and Soil Pollution, 133, 161–183. Monteith, D. T., Evans, C. D. and Reynolds, B. (2000). Are temporal variations in the nitrate content of UK upland freshwaters linked to the North Atlantic Oscillation? Hydrological Processes, 14, 1745–1749. Moss, B., Johnes, P. J. and Phillips, G. L. (1996). The monitoring and classification of standing waters in temperate regions€– a discussion and proposal based on a worked scheme for British waters. Biological Reviews, 71, 310–339. Mulligan, D., Bauraoui, F., Grizzetti, B., Aloe, A. and Dusart, J. (2006). An atlas of pan-European data for investigating the fate of agrochemicals in terrestrial ecosystems. EUR22334. Joint Research Centre. Ispra. Murdoch, P. S. and Stoddard, J. L. (1992). The role of nitrate in the acidification of streams in the Catskill Mountains of New York. Water Resources Research, 28, 2707–2720. Neilsen, M. E., Fisk, M. R., Istok, J. D. and Pedersen, K. (2006). Microbial nitrate respiration of lactate at in situ conditions in ground water from a granitic aquifer situated 450 m underground. Geobiology, 4, 43–52. Nixon, S. W., Ammerman, J. W., Atkinson, L. P. et€al. (1996). The fate of nitrogen and phosphorus at the land sea margin of the North Atlantic Ocean. Biogeochemistry, 35, 141–180.

145

Nitrogen processes in aquatic ecosystems Oenema, O., Oudendag, D. and Velthof, G. L. (2007). Nutrient losses from manure management in the European Union. Livestock Science, 112, 261–272. Ormerod, S. J. and Durance, I. (2009). Restoration and recovery from acidification in upland Welsh streams over 25 years. Journal of Ecology, 46, 164–174. Phillips, G., Pietiläinen, O. P., Carvalho, L. et€al. (2008). Chlorophyllnutrient relationships of different lake types using a large European dataset. Aquatic Ecology, 42, 213–226. Pinay, G., Clement, J. C. and Naiman, R. J. (2002). Basic principles and ecological consequences of changing water regimes on nitrogen cycling in fluvial systems. Environmental Management, 30, 481–491. Pinay, G., Decamps, H., Arles, C. and Lacassin-Seres, M. (1989). Topographic influence on carbon and nitrogen dynamics in riverine woods. Archiv für Hydrobiologie, 114, 404–414. Prechtl, J., Kneip, C., Lockhart, P., Wenderoth, K. and Maier, U. G. (2004). Intracellular spheroid bodies of Rhopalodia gibba have nitrogen-fixing apparatus of cyanobacterial origin. Molecular Biology and Evolution, 21, 1477–1481. Prior, H. and Johnes, P. J. (2002). Regulation of surface water quality in a Cretaceous chalk catchment, UK:€an assessment of the relative importance of instream and wetland processes. Science of the Total Environment, 282, 159–174. Regina, K., Nykänen, H., Silvola, J. and Martikainen, P. J. (1996). Fluxes of nitrous oxide from boreal peatlands as affected by peatland type, water table level and nitrification capacity. Biogeochemistry, 35, 401–418. Rekolainen, S., Mitikka, S., Vuorenmaa, J. and Johansson, M. (2004). Rapid decline of dissolved nitrogen in Finnish lakes. Journal of Hydrology, 304, 94–102. Revsbech, N. P., Jacobsen, J. P. and Nielsen, L. P. (2005). Nitrogen transformations in microenvironments of river beds and riparian zones. Ecological Engineering, 24, 447–455. Reynolds, B. and Edwards, A. (1995). Factors influencing dissolved nitrogen concentrations and loadings in upland streams of the UK. Agricultural Water Management, 27, 181–202. Rivett, M. O., Smith, J. W. N., Buss, S. R. and Morgan, P. (2007). Nitrate occurrence and attenuation in the major aquifers of England and Wales. Quarterly Journal of Engineering Geology and Hydrogeology, 40, 335–352. Rivett, M. O., Buss, S. R., Morgan, P., Smith, J. W. N. and Bemment, C. D. (2008). Nitrate attenuation in groundwater:€a review of biogeochemical controlling processes. Water Research, 42, 4215–4232. Rogora, M. and Mosello, R. (2007). Climate as a confounding factor in the response of surface water to nitrogen deposition in an area south of the Alps. Applied Geochemistry, 22, 1122–1128. Ruckart, P. Z., Henderson, A. K., Black, M. L. and Flanders, W. D. (2007). Are nitrate levels in groundwater stable over time? Journal of Exposure Science and Environmental Epidemiology, 18, 129–133. Sabater, S., Butturini, A., Clement, J. C. et€al. (2003). Nitrogen removal by riparian buffers along a European climatic gradient:€patterns and factors of variation. Ecosystems, 6, 20–30. Sanchez, N. (1996). Le processus de dénitrification dans les sédiments du barrage-réservoir de la Marne:€Etude de sa cinétique et modélisation. Ph.D. thesis, University Pierre et Marie Curie, Paris. Schelske, C. L. and Stoermer, E. F. (1971). Eutrophication, silica depletion, and predicted changes in algal quality in Lake Michigan. Science, 173, 423–424.

146

Seitzinger, S., Harrison, J. A., Bohlke, J. K. et€al. (2006). Denitrification across landscapes and waterscapes:€a synthesis. Ecological Applications, 16, 2064–2090. Seitzinger, S. P. and Kroeze, C. (1998). Global distribution of nitrous oxide production and N inputs in freshwater and coastal marine ecosystems. Global Biogeochemical Cycles, 12, 93–113. Seitzinger, S. P. and Sanders, R. W. (1997). Contribution of dissolved organic nitrogen from rivers to estuarine eutrophication. Marine Ecology Progress Series, 159, 1–12. Seitzinger, S. P., Styles, R. V., Boyer, E. W. et€al. (2002). Nitrogen retention in rivers:€model development and application to watersheds in the northeastern USA. Biogeochemistry, 57, 199–237. Siemens, J. (2003). The European carbon budget:€a gap. Science, 302, 1681. Skoulikidis, N. and Amaxidis, Y. (2009). Origin and dynamics of dissolved and particulate nutrients in a minimally disturbed Mediterranean river with intermittent flow. Journal of Hydrology, 373, 218–229. Smith, J. W. N. and Lerner, D. N. (2008). Geomorphologic control on pollutant retardation at the groundwater–surface water interface. Hydrological Processes, 22, 4679–4694. Søvik, A. K., Augustin, J., Heikkinen, K. et€al. (2006). Emission of the greenhouse gases nitrous oxide and methane from constructed wetlands in Europe. Journal of Environmental Quality, 35, 2360–2373. Strahler, A. N. (1952). Hypsometric (area-altitude) analysis of erosional topography. Bulletin of the Geological Society of America, 63, 1117–1142. Tallec, G., Garnier, J., Billen, G. and Gousailles, M. (2006). Nitrous oxide emissions from secondary activated sludge in nitrifying conditions of urban wastewater treatment plants:€effect of oxygenation level. Water Research, 40, 2972–2980. Talling, J. F. (1979). Factor interactions and implications for the prediction of lake metabolism. Archiv für Hydrobiologie, 13, 96–109. Thayalakumaran, T., Bristow, K. L., Charlesworth, P. B. and Fass, T. (2007). Geochemical conditions in groundwater systems:€implications for the attenuation of agricultural nitrate. Agricultural Water Management, 95, 103–115. Thouvenot, M., Billen, G. and Garnier, J. (2007). Modelling nutrient exchange at the sediment–water interface of river systems. Journal of Hydrology, 341, 55–78. Van der Molen, D. T., Portielje, R., De Nobel, W. T. and Boers, P. C.€M. (1998). Nitrogen in Dutch freshwater lakes:€trends and targets. Environmental Pollution, 102, 553–557. Verhoeven, J. T. A., Keuter, A., Van Logtestijn, R., van Kerkhoven, M. B. and Wassen, M. (1996). Control of local nutrient dynamics in mires by regional and climatic factors:€A comparison of Dutch and Polish sites. Journal of Ecology, 84, 647–656. Wade, A. J., Butterfield, D. and Whitehead, P. G. (2006). Towards an improved understanding of the nitrate dynamics in lowland, permeable river-systems:€Applications of INCA-N. Journal of Hydrology, 330, 185–203. Willetts, V. B., Reynolds, B. A., Stevens, P. A., Ormerod, S. J. and Jones, D. L. (2004). Dissolved organic nitrogen regulation in freshwaters. Journal of Environmental Quality, 33, 201–209. Wright, R. F., Alewell, C., Cullen, J. M. et€al. (2001). Trends in nitrogen deposition and leaching in acid-sensitive streams in Europe. Hydrology and Earth System Sciences, 5, 299–31.

Chapter

8

Nitrogen processes in coastal and marine ecosystems Lead author: Maren Voss Contributing authors: Alex Baker, Hermann W. Bange, Daniel Conley, Sarah Cornell, Barbara Deutsch, Anja Engel, Raja Ganeshram, Josette Garnier, Ana-Stiina Heiskanen, Tim Jickells, Christiane Lancelot, Abigail McQuatters-Gollop, Jack Middelburg, Doris Schiedek, Caroline P. Slomp and Daniel P. Conley

Executive summary Nature of the problem • Nitrogen (N) inputs from human activities have led to ecological deteriorations in large parts of the coastal oceans along European coastlines, including harmful algae blooms and anoxia. • Riverine N-loads are the most pronounced nitrogen sources to coasts and estuaries. Other significant sources are nitrogen in atmospheric deposition and fixation.

Approaches • This chapter describes all major N-turnover processes which are important for the understanding of the complexity of marine nitrogen cycling, including information on biodiversity. • Linkages to other major elemental cycles like carbon, oxygen, phosphorus and silica are briefly described in this chapter. • A tentative budget of all major sources and sinks of nitrogen integrated for global coasts is presented, indicating uncertainties where present, especially the N-loss capacity of ocean shelf sediments. • Finally, specific nitrogen problems in the European Regional Seas, including the Baltic Sea, Black Sea, North Sea, and Mediterranean Sea are described.

Key findings/state of knowledge • Today, human activity delivers several times more nitrogen to the coasts compared to the natural background of nitrogen delivery. The source of this is the land drained by the rivers. Therefore, the major European estuaries (e.g. Rhine, Scheldt, Danube and the coastlines receiving the outflow), North Sea, Baltic Sea, and Black Sea as well as some parts of the Mediterranean coastlines are affected by excess nutrient inputs. • Biodiversity is reduced under high nutrient loadings and oxygen deficiency. This process has led to changes in the nutrient recycling in sediments, because mature communities of benthic animals are lacking in disturbed coastal sediments. The recovery of communities may not be possible if high productivity and anoxia persist for longer time periods.

Major uncertainties/challenges • The magnitude of nitrogen sources are not yet well constrained. Likewise the role of nutrient ratios (N:P:Si ratios) may be a critical variable in the understanding of the development of harmful algae blooms. • Whether only inorganic forms of nitrogen are important for productivity, or whether organic nitrogen is also important is not well understood and needs future attention.

Recommendations • For the future it will be necessary to develop an adaptive transboundary management strategy for nitrogen reduction. The starting point for such regulation is located in the catchments of rivers and along their way to the coastal seas. • An overall reduction of nitrogen inputs into the environment is urgently necessary, especially in the case of diffuse nitrogen inputs from agricultural activities.

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

147

Nitrogen processes in coastal and marine ecosystems

8.1╇ Introduction The marine and terrestrial nitrogen cycles are closely linked, although intellectual boundaries in disciplines often lead to separate treatment of both cycles (Gruber, 2004). Since humans have perturbed the nitrogen cycle considerably via the production of artificial fertilizers, fossil fuel combustion or animal husbandry, the man-made sources of reactive nitrogen (fixed nitrogen, Nr) are now larger than the amount produced by natural nitrogen fixation (Gruber and Galloway, 2008). Furthermore the transport of anthropogenically produced nitrogen to the ocean is accelerated relative to the previous century and close links between watersheds, airsheds and the marine system result in excess nutrient supply not only to the coastal zones (Rabalais, 2002) but also to the open ocean (Duce et€ al., 2008; Figure€ 8.1). Moreover Duce et€ al. (2008) argue that the coastal area is a sink not a source of nitrogen for the open ocean. Nitrogen may also strongly perturb natural fluxes and processes responsible for the production and release of trace gases which are relevant to climate such as nitrous oxide€(N2O). The global Nr budget has changed from one that was almost balanced in preindustrial times to one in present times with much higher inputs than losses (Vitousek et€ al., 1997) which impacts coastal systems significantly. Here, we summarize the current knowledge on N-pathways in marine waters, the anthropogenic impact and how the cycling, mass transfer and effects of this nitrogen have changed the European coastal waters (e.g. estuarine systems and enclosed seas).

8.2╇ Nitrogen-cycle processes in the open ocean and coastal systems The marine environment has unique characteristics that distinguish it from other aquatic systems. First of all the salinity varies from almost zero in inner estuaries to almost 40 in the Mediterranean (earlier salinity was given in grams of salts per litre, since 1978 no units are used because it refers to a

Figure€8.1 Schematic of the coupling of the marine and the terrestrial nitrogen cycles. The numbers are estimates of the natural plus anthropogenic N transports in Tg N yr−1 as taken from Gruber and Galloway (2008). The circles visualize the cycling of carbon (green), phosphorus (blue), and nitrogen (red). The industrial fixation is purely anthropogenic, and the atmospheric deposition on land (145) is dominated by 70% anthropogenic N. The graph is inspired by the same paper.

148

conductivity ratio of a seawater sample to a standard KCl solution). The salinity increases the density of the water and therefore strongly shapes the stratification of a water body so that a change of only 1 g l–1 equals a 5 °C difference in density. Salinity and temperature differences are responsible for the structure of a water body and a stable stratification is sustained by lighter waters on top of heavier ones. Only winds, tides and currents are able to break up the interfaces. Stratification prevents the exchange of dissolved substances between layers of water, so that nutrients may accumulate at a certain depth. Particulate material like phytoplankton aggregates, faecal pellets from zooplankton etc. sink through this interface and are degraded. The microbial degradation of this organic matter leads to the accumulation of nutrients and the depletion of oxygen below the interface. If the oxygen consumption is higher than the oxygen renewal of the deep waters, anoxia can develop with large scale die-offs of benthic animals. The abundance of such ‘dead zones’ is increasing in many coastal areas worldwide (Diaz and Rosenberg, 2008). Primary production in the oceans is to a large extent driven by the availability of inorganic and organic nitrogen compounds; mostly nitrate, ammonium, and dissolved organic nitrogen (DON). Coastal systems receive their nutrients from recycling of organic matter, river input, atmospheric deposition, onshore transport of nutrients from the open sea, and to a small extent from N2-fixation. For a long time, research on riverine N-inputs into coastal areas was mainly focused on inorganic N-species, especially nitrate and ammonium. In the last decade the importance of DON as a nutrient has received attention, especially after several studies showing that it can comprise up to 90% of the total nitrogen input (Seitzinger and Sanders, 1997), and that its bioavailability for phytoplankton and algae may be significant (Twomey et€al., 2005; Bronk et€al., 2007). In the past N was considered the major limiting nutrient of marine systems. However, since the delivery of P from watersheds has decreased (with improvement of P treatment in wastewater treatment plants), waters have become P limited in several locations (Cugier et€al., 2005; Lancelot et€al., 2007). However, the magnitude of delivery of both N and P has remained in excess relative to silica, which has often led to diatoms being replaced by harmful non-diatomaceous species (Billen and Garnier, 2007). In the open ocean regeneration of organic matter, plus convection (in temperate latitudes down to the seasonal thermocline), atmospheric deposition, and diazotroph N2-fixation provide most of the nutrients to the productive, sunlit surface waters. However, in the shallow coastal areas there is a tight coupling between the processes in the water-column and the sediments (Herbert, 1999). Groundwater efflux from sediments may also introduce nutrients to the water column (Slomp and Van Cappellen, 2004) in certain circumstances. Strong links exist between the N-, P- and C- cycles (Gruber and Galloway, 2008) as well as between trace metals and oxygen concentrations. The degradation and turnover processes of the various nitrogen compounds are mostly mediated by bacteria. A key process in the N-cycle is ammonification, carried out not only

Maren Voss

by bacteria but also by actinomycetes and fungi, which converts organic N (as found in proteins, amino-sugars, nucleic acids etc.) to inorganic ammonium (Herbert, 1999). Ammonium is taken up by phytoplankton and by bacteria, with bacterial uptake as high as 49% of total NH4+ uptake in the surface water and up to 72% at the bottom of the water column (Bradley et€al., 2010). Some NH4+ is oxidized during bacterial or archael nitrification via nitrite to nitrate. Nitrification is an obligatory aerobic two-step process; the first step is the oxidation of ammonia to nitrite by ammonia oxidizing bacteria and in the second step nitrite is oxidized to nitrate by nitrite oxidizers. The rate of nitrification is controlled by temperature unless there is an insufficient supply of oxygen and ammonium. Therefore numerous studies report seasonal cycles with increasing rates at higher temperatures (Tuominen et€al., 1998). At oxygen concentrations below 10 µmol l–1 the process additionally produces N2O as a by-product (Hynes and Knowles, 1984; Jorgensen et€ al., 1984) Nitrification occurs in the water column as well as in oxygenated sediment layers, and the generated nitrate is either assimilated by phytoplankton and algae or is reduced to N2 and N2O during bacterial denitrification. It was accepted for a long time, that denitrification is the only process that permanently removes reactive nitrogen from the aquatic and terrestrial environment. Under hypoxic conditions (<10 µmol l–1), the heterotrophic denitrifiers can use nitrate as a final electron acceptor during respiration and produce N2 gas as final end product and N2O as an intermediate. The extent of this process is strongly controlled by temperature, nitrate concentrations and the availability of organic carbon (Hulth et€al., 2004), which makes the process seasonally variable (Tuominen et€al., 1998). Autotrophic denitrification with CO2 as a carbon source may temporarily occur as the dominant process, at anoxic–oxic interfaces such as found in the Baltic Sea (Hannig et€al., 2007). In sediments below oxic water-Â�columns there is often a tight coupling between nitrification and denitrification, and the nitrate generated during nitrification in the oxic sediment layer diffuses downwards in sub-oxic sediment regions where it is rapidly denitrified (Herbert, 1999). Mulder et€al. (1995) was the first paper to describe an alternative pathway which generates N2 from reactive N. Although originally described from sewage-treatment plants anaerobic ammonia oxidation (anammox) is considered as important as denitrification in some upwelling regions (Kuypers et€al., 2006) but not in all (Ward et€al., 2009). During anammox, the strictly anaerobic chemoautotrophic bacteria of the group Planctomyces fix CO2 and use NH4+ to reduce NO2−, which results in the production of N2 (Schmidt et€ al., 2002). In sediments the process only occurs at oxygen concentrations below 1.1 µmol l–1 and can account for up to 80% of the total N2 production. Its importance seems to decrease with decreasing water depths (Thamdrup and Dalsgaard, 2002). There are several other N-transformation processes which most likely occur in the marine environment, but are not well understood:€ (i) sedimentary chemodenitrification, where manganese species react with nitrate or ammonium to produce N2 (Brandes et€al., 2007); (ii) dissimilatory nitrate reduction

to ammonia (DNRA), which is carried out by strictly anaerobic, fermentative bacteria (Herbert, 1999). High rates of DNRA are measured under highly reducing conditions in sediments (Christensen et€ al., 2000) and the water column (Lam et€ al., 2009); (iii) oxygen-limited autotrophic nitrification�denitrification (OLAND), which has been detected in sediments (Brandes et€al., 2007). Many N-transformation processes are strongly affected by humans (Figure€ 8.1). Together with increasing emissions of nitrogen and phosphorus (P) into the marine, terrestrial and atmospheric environment, these may lead to dramatic changes which can be summarized under the term of eutrophication (see Section 8.5). Coastal areas and estuaries suffer particularly from excess nutrient inputs, since they form the transition zone between the terrestrial and the marine environment. Anoxic coastal waters have been reported in many coastal areas worldwide such as the Gulf of Mexico, the Baltic Sea, or the Black Sea (Diaz, 2001). The increased input of nitrogen which is often accompanied by oxygen limitation has a strong negative effect on benthic metabolism and nitrogen mineralization (Karlson et€ al., 2007) (see Section 8.8). Therefore processes which remove reactive N and thus counteract eutrophication become of considerable interest. In this case denitrification is key, since anammox is usually not as important as denitrification in shallow waters (Thamdrup and Dalsgaard, 2002).

8.3╇ Dissolved gaseous nitrogen compounds (NO, N2O, N2, NH3)

Nitric oxide (NO), nitrous oxide (N2O), dinitrogen (N2) and ammonia (NH3) are constituents of the Earth’s atmosphere and play important roles in the chemistry and climate of the present-day Earth. Moreover, they are intermediates or byproducts of the marine nitrogen cycle (see above). An overview on current knowledge of the distribution and pathways of NO, N2O, N2 and NH3 in European marine ecosystems is given here. If not cited otherwise, further details and references can be found in Bange (2006, 2008).

8.3.1╇ Nitric oxide (NO) NO is chemically very reactive and thus it is a short-lived compound in aquatic systems. In nitrite-rich surface waters, NO is photochemically produced via the reduction of nitrite. The photochemically induced build-up of NO during the day is balanced by degradation during the night. Depth profiles of NO in the Pacific Ocean indicate that subsurface NO might be produced and consumed as an intermediate of nitrification and denitrification, respectively (Bange, 2008). Unfortunately, there are only a few measurements of oceanic NO available. In European marine ecosystems these are limited to one study of the NO flux from Wadden Sea sediments in northern Germany (Bodenbender and Papen, 1996). Because global oceanic NO emissions are only of marginal importance in comparison to emissions of N2O (see below), NO emissions from European coastal and marine ecosystems are most probably negligible as well.

149

Nitrogen processes in coastal and marine ecosystems

8.3.2╇ Nitrous oxide (N2O)

The world’s oceans (including their coastal zones) are sources of atmospheric N2O (a powerful greenhouse gas) and play a major role in the global budget of atmospheric N2O. N2O in oceanic environments is mainly formed as a by-product during nitrification (NH4+ → NO3–) and as an intermediate during denitrification (NO3– → N2). Based on the comprehensive compilation of N2O measurements from European coastal waters (Bange, 2006), three main conclusions can be drawn. (i) The highest concentrations of N2O occur in estuaries and fjords, whereas in open coastal waters (i.e. shelf waters not influenced by river plumes) N2O concentrations are close to the expected equilibrium with the atmosphere. This indicates that N2O is mainly formed in estuarine systems. (ii) It seems that sedimentary denitrification and water column nitrification are the major N2O formation processes. However, the yield of N2O from both processes strongly depends on the local O2 concentrations, thus O2 is the key factor in the regulation of N2O production (and its subsequent emissions to the atmosphere). Any pronounced changes of the O2 regimes in coastal waters (see Section 8.5.4 this chapter) may lead to conditions which are favourable for temporarily enhanced N2O production. In shallow waters this ‘excess’ N2O can be easily ventilated to the atmosphere, whereas in non-ventilated anoxic deep waters, e.g. in the central basin of the Baltic Sea, N2O can be consumed by water column denitrification. (iii) European coastal waters are a net source of N2O to the atmosphere. The major contribution comes from the estuarine/river systems rather than the open shelf areas. The mean overall N2O emissions calculated for the European coastal water area of 3.2 × 1012 m2 (Bange, 2006) are 0.5 Tg N yr–1 with a methodologically caused uncertainty of 0.3–0.7 Tg N yr–1. Therefore European coastal waters contribute significantly (about 9%) to the present global oceanic N2O emissions of 5.5 Tg N yr–1 (IPCC, 2007). Future N2O emissions from European coastal areas will be strongly influenced by nitrogen inputs to coastal waters and will most likely increase in the future. There might be two further, however, largely unknown N2O sources in European coastal areas.€First, small coastal upwelling sites along the European coasts from subsurface layers may be an additional, physically driven, sources of N2O to the atmosphere. In addition, a large coastal upwelling site is situated off the Atlantic coast of Portugal, but N2O emissions during upwelling events are not yet quantified. Second, organic matter release from fish farming activities could increase the dissimilatory nitrate reduction to ammonium (DNRA) in sediments (Christensen et€al., 2000). An additional source of N2O may be sedimentary DNRA and third, submarine groundwater discharge (Crusius et€al., 2008).

8.3.3╇ Dinitrogen (N2)

Despite the inert nature of N2, the atmospheric N2 pool is available for biological productivity via N2 fixation (N2 → NH4+). An example of which is the Baltic Sea, where fixation takes place in the surface layer during the summer. Estimates for N2 fixation in the Baltic Sea range from 0.03 to 0.9 Tg N yr–1 (Rahm et€al.,

150

2000; Schneider et€al., 2003). The release of dinitrogen (N2) from the fixed inorganic nitrogen pool (i.e. NO3–, NO2– and NH4+) is mediated by denitrification (NO3– → N2) and anaerobic ammonium oxidation (anammox, NO2– + NH4+ → N2). This N2 production has not yet been extrapolated to the European scale. Denitrification in sediments and in the water column of the Baltic Sea has been estimated to range from 0.5 to 0.6 Tg N yr−1 (Shaffer and Rönner, 1984; Eilola and Stigebrandt, 1999). In the Black Sea the annual water column production of N2 via anammox has been estimated to be about 0.4€ Tg€ N/â•›yr–1 (Kuypers et€al., 2003). Thus, we might roughly estimate a total N2 production in the Baltic and Black Seas of about 1 Tg N yr–1. This estimate should be regarded as conservative, because it does not include N2 production in the sediments of the Black€Sea.

8.3.4╇ Ammonium/ammonia (NH4+/ NH3)

The distribution of dissolved NH3 is tightly coupled to the distribution of dissolved NH4+ via the NH3/NH4+ equilibrium. Because the mean pH of today’s ocean surface layer is about 8.1, dissolved NH3 can exist in the seawater as a dissolved, nonprotonated gas and thus it is available for gas exchange across the ocean/atmosphere interface. At a pH of 8.1, a water temperature of 25 °C, and a salinity of 35, only about 6% of the sum of (NH3) and (NH4+) is available as dissolved NH3. NH4+ is the substrate or final product of major biological transformation processes of the oceanic nitrogen cycle such as bacterial and archaeal nitrification (NH4+ → NO3–), assimilation by phytoplankton (NH4+ → organic matter) and excretion by zooplankton. Moreover, NH4+ can be formed by photochemical decomposition of dissolved organic nitrogen. As the biological uptake of NH4+ is very rapid, concentrations of dissolved NH4+ and NH3 are generally low. In coastal areas which are heavily influenced by anthropogenic activity, such as the North and Baltic Seas, high atmospheric NH3 concentrations lead (despite high biological production and associated ammonia production) to a net flux of NH3 from the atmosphere to the ocean. Moreover, it was recently demonstrated that even at low atmospheric NH3 concentrations, the coastal North Sea is a sink for atmospheric NH3. Based on the limited amount of oceanic and atmospheric NH3 measurements available from European coastal areas, the NH3 exchange across the ocean/atmosphere interface is poorly known.

8.3.5╇ Outlook Despite the fact that our knowledge on the distribution of gaseous nitrogen compounds in European coastal waters is still associated with a high degree of uncertainty, a rough impact assessment of various parameters which might influence today’s emissions is given in Table€ 8.1. It is obvious that the ocean is a source of N2O and other gases under eutrophying conditions. Owing to the unknowns (Table€8.1) and in view of the ongoing environmental changes, we need integrated longterm measurement programmes in European coastal and marine ecosystems.

Maren Voss Table€8.1 Impact assessment of various parameters which might influence today’s emissions of gaseous nitrogen compounds from European coastal waters

Table€8.2 Atmospheric emissions of fixed nitrogen in 1993 (1012 mol N€yr−1) (based on Galloway et€al., 2004)

NO

N 2O

N2

NH3

Eutrophication/hypoxic events

?

++

++

++

Coastal upwelling

?

+

–

–

Fish farming

?

–

?

+

Ecosystem shifts due to climate change

?

?

?

?

Anthropogenic activity

NOx

NH3

0.5

0.6

Agricultural activity

0.2

2.8

Fossil fuel combustion

1.5

0.01

Industry

0.5

0.2

Soils, vegetation and animals

0.2

0.3

Lightning

0.4

—

0.06

0.06

0.04

0.04

Biomass burning

a

Natural

Classification scheme:€– = minor; + = moderate; ++ = high; ? = unknown. In essence; today’s N2O emissions are highly sensitive to eutrophication/ hypoxic events whereas the effect of fish farming activities on the present N2O emissions will be negligible.

Natural fires

b

Stratosphere exchange Ocean exchange

8.4╇ Atmospheric inputs

Total a

8.4.1╇ Atmospheric N emissions and transport to the oceans Atmospheric nitrogen emissions are predominantly in the form of oxidized (NO and NO2) and reduced nitrogen (NH3). Emissions of oxidized nitrogen are dominated by combustion processes and those of reduced nitrogen by agricultural activity. Although the emission estimates in Table€ 8.2 are rather uncertain, they do indicate that emissions of both forms of N are of broadly similar magnitude and that emissions are now several times greater than natural emissions. A recent global compilation not broken down by emission sector quoted total emissions in 2000 of between 3.7 and 4.6 × 1012 mol N yr−1 with 75% of this being anthropogenic (Duce et€al., 2008). Here we briefly summarize atmospheric N sources and transformations and their impacts, first on the open ocean and then on coastal€areas. There is an active inorganic N atmospheric chemistry and only the relevant parts are summarized here, based on Jickells (2006). A more detailed discussion can be found in Seinfeld and Pandis (1998). Upon emission to the atmosphere, NH3 gas will react with available acids and some of these reactions are reversible (e.g. the formation of ammonium nitrate) and some are essentially irreversible (e.g. the formation of ammonium sulphate). Hence both ammonia gas and aerosol ammonium species are often present in the atmosphere, with ammonium predominantly as fine mode aerosol (particles with diameters in the range 0.1–1 µm), which is the characteristic size distribution for aerosols formed by gas to particle reactions (Raes et€ al., 2000). NO and NO2 are usually referred to as NOx because these forms can interchange rapidly. They are oxidized on a timescale of hours to days to nitric acid which can react with bases such as ammonia in the atmosphere. In the marine boundary layer nitric acid will also react with sea salt by the reaction: HNO3(g) + NaCl(a) → NaNO3(a) + HCl(g).

(8.1)

The efficiency of this reaction means that there is little free nitric acid in the marine atmosphere and also that the

b

—

0.4

3.4

4.5

â•›Some of the tropical biomass burning included here could be natural. â•›Under ‘natural fires’ only high-latitude forest fires are included.

aerosol nitrate is primarily associated with the coarse mode (>1 µm) sea spray aerosol, which is mechanically generated (Raes et€al., 2000). Desert dust is also associated with this larger coarse mode and is alkaline, therefore in regions where dust supply is important such as in the tropical Atlantic, Indian and North West Pacific Oceans, reactions between dust aerosol and nitric acid can also be important (Usher et€al., 2003). In addition to these inorganic N forms, organic N is found in the atmosphere, particularly as aerosol, where soluble organic N represents a variable but significant fraction (often ~20%–30%) of total N (Cornell et€al., 2003). An insoluble organic N component may also exist (Russell et€al., 2003). The sources of this organic nitrogen are uncertain and likely to be many and varied, but recent evidence suggests it includes a significant anthropogenic component (Zhang et€ al., 2008). The bioavailability of this organic nitrogen is also uncertain, although Seitzinger and Sanders (1999) suggest that at least a part of it is readily bioavailable. Deposition of nitrogen to the oceans occurs by wet and dry deposition, and the rates of both processes depend in part on the aerosol size distribution, with large aerosols depositing more rapidly. Deposition rates for coarse mode aerosol are an order of magnitude or more greater than those of the fine mode (Duce et€ al., 1991), so the transformation of nitrate size distribution by the reaction with sea salt has a significant influence on the global flux distribution. Ammonia exchange at the ocean surface is a two way gaseous process. Increasing ammonia concentrations (as a result of human activity) have been suggested to have altered the direction of the flow of ammonia exchange in many regions (see Jickells, 2006, for a review). However, Johnson et€ al. (2008) emphasized the sensitivity of ammonia air–sea exchange to temperature and suggested that the flux will be predominantly into the oceans at low water temperatures and potentially out of the oceans at higher temperatures, although this is modulated by atmospheric ammonia concentrations.

151

Nitrogen processes in coastal and marine ecosystems

8.4.2╇ Effects of atmospheric N deposition on the€oceans Estimates of atmospheric nitrogen inputs (including organic N) to the global ocean (4.8 × 1012 mol N yr−1, ~80% of which is anthropogenic) now rival riverine (3.6–5.7 × 1012 mol N€yr−1) and natural biological N2 fixation (4.3–14.3 × 1012 mol N yr−1) rates (Duce et€ al., 2008). The distribution of these inputs is of course very different with fluvial inputs dominating in coastal waters and N2 fixation in tropical waters (Westberry and Siegel, 2006). There is considerable variability in N deposition fluxes to the oceans (Figure€8.2), reflecting global emission patterns and atmospheric transport pathways with most inputs falling into the North Pacific, Northern Indian and Atlantic Oceans. Duce et€ al. (2008) consider the impact of this increasing atmospheric input of fixed nitrogen on the open ocean. The dispersal of this flux over the vast areas of the ocean means that deposition at any one point is small and hence unlikely to trigger significant ecological impacts such as algal blooms or suppression of nitrogen fixation. However, large areas of the open oceans are believed to be nitrogen limited (Duce et€ al., 2008) and hence the deposition of nitrogen will allow somewhat higher total and ‘new’ primary production (in the terminology of Dugdale and Goering (1967)). The latter should on a long enough timescale be equivalent to the export of nitrogen to the deep sea within sinking organic matter€ – the ‘oceanic organic pump’€ – which can draw atmospheric CO2 into the oceans. Duce et€ al. (2008) estimate this enhanced drawdown due to ‘fertilization’ with atmospheric fixed N to be about 0.3 Pg C yr–1, which can be compared to a total oceanic uptake rate of CO2 equivalent to 2.2 ± 0.5 Pg C yr–1, emphasising the potential importance of this process. An alternative analysis by Krishnamurthy et€al. (2007) estimated a lower (0.16€Pg€C€yr–1),

but still significant, fertilization effect. Such an increase in productivity and carbon export to deep waters acts within a series of important feedbacks within the ocean atmosphere climate system. For instance, Duce et€al. (2008) note the potential for changes in oceanic N2O emissions, which could offset the CO2 storage benefits (in terms of greenhouse gas forcing) since N2O is a much stronger greenhouse gas than CO2. There are of course a wide variety of complex feedbacks between the ocean and atmosphere components of the Earth System and global change pressures such as nitrogen fluxes to the oceans do not exist in isolation. Changes in nitrogen inputs in coming decades will likely be accompanied by changes in temperature and ocean stratification which may act to enhance nutrient limitation. Atmospheric inputs of nitrogen also do not operate in isolation. A variety of nutrients and contaminants are transported together through the atmosphere and the full range of synergistic and antagonistic interactions between these is unknown. However, in terms of nutrients, we do know the transport and deposition of iron to the oceans reasonably well, although there are still many unknowns (Jickells et€al., 2005; Mahowald et€ al., 2005). Areas of high nitrogen deposition in the tropical North Atlantic, North Indian and Northwest Pacific Oceans are also regions of high dust and iron deposition. High dust/iron inputs can sustain nitrogen fixation in tropical waters (Mills et€al., 2004) and in some high latitude HNLC (high nutrient low chlorophyll) waters, allow more efficient phytoplankton growth and water column nitrogen and phosphorus uptake (Boyd et€al., 2007). It is also clear that the atmospheric supply of phosphorus in comparison to both nitrogen and iron is small when compared to the biological requirements and hence atmospheric supply of nutrients will tend to push the system toward P limitation (Baker et€al., 2003, 2007; Mahowald et€al., 2008).

Figure€8.2 Nitrogen deposition flux (mg N m–2 yr –1) to the Earth’s surface in 2000 for the S1 baseline scenario (after Dentener et€al., 2006).

152

Maren Voss Table€8.3 Atmospheric vs. riverine N deposition for some European Coastal Seas (103 tonnes yr−1)

Mediterranean

Atmospheric

Riverine

1084

1000

Baltic

185

830

North Sea

412

1073

Reference Guerzoni et€al. (1999) Voss et€al. (2005) Rendell et€al. (1993)

8.4.3╇ Effects of atmospheric N deposition on coastal waters The atmosphere makes a significant contribution to the total nitrogen input to many European coastal waters, as illustrated in Table€8.3. The exact fluxes will vary with time due to varying management regimes and the fluxes listed in Table€8.3 are for different years. They are also sensitive to slightly different assumptions about deposition processes and the role of different forms of nitrogen in both rivers and the atmosphere, but the table clearly illustrates that the atmospheric input is significant. Atmospheric inputs to coastal waters close to anthropogenic N sources may be substantially higher than to open ocean waters, while they may be very similar in coastal waters remote from anthropogenic sources. Meteorological conditions can also act to deliver atmospheric inputs in short intense bursts under certain conditions (Spokes and Jickells, 2005). Coastal areas receiving higher atmospheric inputs often also receive increased nutrient inputs from fluvial and groundwater sources. However, even in areas such as the North Sea (into which major river systems enriched in nutrients discharge), the atmosphere can still contribute a substantial part of the total land derived nutrient input (Spokes and Jickells, 2005). The consequences of these inputs vary greatly since the biogeochemistry of any particular coastal region is profoundly and closely linked to the physical environment, particularly the rates of exchange with open ocean waters (e.g. Jickells, 1998). This process can both remove land-derived nutrients and supply nutrients from deep ocean waters. Paerl and Withall (1999) considered the link between algal blooms and atmospheric deposition. However, Spokes and Jickells (2005) concluded that although atmospheric deposition may contribute significantly to overall terrestrially derived nutrient loadings, comparisons of nutrient supply to overall productivity in coastal waters, fuelled by terrestrial, offshore and internal recycling supplies of nutrients suggest the overall impact on productivity is modest. Hence atmospheric inputs are unlikely to directly trigger blooms. However, atmospheric inputs do contribute to overall nutrient loadings and hence to eutrophication pressure and under certain conditions may act to sustain blooms as they develop.

8.5╇ Linkages to other elemental cycles 8.5.1╇ C-cycling and ocean acidification The cycling of nitrogen is closely linked to other biogeochemical cycles, in particular to C and P. The tight coupling of these cycles was highlighted by the famous work of Alfred C. Redfield

(1890–1983). The ‘Redfield ratio’ describes the molar stoichiometric relationship between C, N and P in marine organic matter, which is 106:16:1 and is a cornerstone of marine biogeochemistry. However, the general applicability of the Redfield ratio is under debate and there are numerous examples which show its systematic deviation on the organism and species level, with the trophic status of the system, or over time and space (Banse, 1973; Geider and La Roche, 2002). Nevertheless, deviations of the C:N ratio in particulate organic matter from the Redfield ratio generally are within the range of 20%–30% (Sterner et€al., 2008), which is very narrow compared to terrestrial systems. A somewhat greater decoupling of C and N is observed for proÂ� cesses involving inorganic compounds (Banse, 1994). The uptake of more DIC (dissolved inorganic carbon) than that inferred from nitrate supply and Redfield stoichiometry is referred to as ‘carbon overconsumption’ (Toggweiler, 1993). Estimates of carbon overconsumption in the field vary between 17% and 300% (Sambrotto et€ al., 1993; Michaels et€ al., 1994; Marchal et€ al., 1996). Hypotheses that seek to explain carbon overconsumption are the preferential remineralization of organic nitrogen compounds (Thomas and Schneider, 1999), and the enhanced release of dissolved organic carbon (Engel et€al., 2002; Schartau et€al., 2007). The close coupling between N and C is of special relevance, because it constrains the biological draw down of CO2 in the ocean. In many oceanic domains as well as in coastal systems, the uptake of CO2 by primary production is limited through the bioavailability of nitrogenous nutrients. Biological nitrogen fixation is the major process to transform dinitrogen, N2, into combined forms, such as NH4+ and ultimately support the marine food web. Over long time scales the coupling between biological CO2 uptake and N2-fixation has therefore been proposed to affect natural climate cycles through indirect feedbacks to atmospheric CO2 (McElroy, 1983). However, primary production based on N2-fixation ultimately becomes limited by the availability of phosphorus (Tyrrell, 1999; Sanudo-Wilhelmy et€al., 2001) and in some regions by iron (Mills et€al., 2004). As a consequence, the role of the ‘biological pump’ in the uptake of anthropogenic CO2 is limited as long as nutrient concentrations in the world’s ocean or N2 fixation rates do not increase accordingly. There is still little known about the direct effects of anthropogenic perturbations, in particular the increase of CO2 concentrations and the associated acidification of seawater, on the coupling between N and C in marine systems. Recent studies have shown that rising CO2 concentration (to levels expected for the next century), stimulates growth and N2-fixation in Trichodesmium spp. (Barcelos e Ramos et€ al., 2007; Hutchins et€ al., 2007; Levitan et€ al., 2007), a tropical and subtropical

153

Nitrogen processes in coastal and marine ecosystems

cyanobacteria, which is responsible for over 50% of biological N2-fixation in the ocean (Capone et€al., 1997). Hutchins et€al. (2007) estimated that N2-fixation by Trichodesmium will increase by 35%–100% until the year 2100, which would substantially raise the total amount of pelagic CO2-fixation in the ocean. Determining the response of other diazotroph species (including unicellular cyanobacteria), to ocean acidification and the combined effects of nutrients and rising temperature, is a priority task for building an understanding of the future of the marine N-cycle. Another mechanism, which could potentially lead to a decoupling of C and N cycles in the future ocean, is the release and subsequent gel particle formation of non-utilized photosynthesis products by the cell. It has been demonstrated that increasing CO2 concentration can enhance photosynthesis in various phytoplankton species (Riebesell, 2004). In comparison to multicellular autotrophs, the spatial capabilities for storage of assimilates are limited in a phytoplankton cell. Excess carbohydrates are disposed to the surrounding seawater and often accumulate during vernal seasons. A fraction of these exudates comprises acidic polysaccharide, which aggregate into transparent exopolymer particles (TEP) and increase the C:N ratio of particulate matter (Engel, 2002). TEP production has been shown to increase with CO2 concentration in experimental studies (Engel, 2002; Engel et€al., 2004; Mari, 2008). Since TEP enhance particle aggregation and export, they may be of special relevance for the sustained or even enhanced decoupling of carbon from nitrogen in export fluxes in the future ocean (Schneider et€al., 2004; Arrigo, 2007).

8.5.2╇ P-cycling and eutrophication effects Phosphorus as a limiting nutrient and its availability in marine systems Along with N and iron (Fe), P is one of the key nutrient elements that can limit phytoplankton growth in marine environments. Phosphorus is assumed to be the ultimate limiting nutrient on geological time scales, based on the fact that at these time scales (> 1000 years), N requirements of phytoplankton

can always be met through N2 fixation from the atmosphere (Tyrrell, 1999). On shorter time scales, N availability typically controls phytoplankton growth in most coastal and marine systems (Howarth and Marino, 2006), with P being (co-) limiting in specific regions, such as the coastal zone of China (Harrison et€al., 1990), the Mediterranean Sea (Krom et€al., 1991b) and open ocean oligotrophic sites in the Atlantic and Pacific Ocean (Benitez-Nelson, 2000; Arrigo, 2005 and references therein). The availability of P in the oceans depends on the balance between the input of reactive P (i.e. biologically available P) from rivers, burial in sediments and the recycling in the marine system (Figure€8.3). In contrast to the N cycle, atmospheric inputs are generally unimportant, with the exception of the highly oligotrophic open ocean environments where dust inputs may alleviate both P and Fe co-limitation of N2 fixers (Mills et€al., 2004). Burial of P in sediments mostly takes place in the form of organic P, authigenically formed calcium-phosphate phases, such as carbonate fluor apatite (CFA), and as P bound to Fe (hydr)oxides. Fish debris (biogenic Ca-P) can be an important sink in lowÂ�oxygen settings (Schenau and de Lange, 2000). Although the major proportion of the total burial of P likely takes place in continental margin sediments (50%–90%; Follmi, 1996; Ruttenberg, 2003), the overall removal of reactive P to marine sediments is not well-quantified. As a consequence, current estimates of the oceanic residence time of P vary significantly, with estimated values ranging from 8000 to 40 000 years (Benitez-Nelson, 2000; Ruttenberg, 2003). This is considerably higher than the oceanic residence time for N (< 3000 years; Gruber, 2004). Given this relatively long oceanic residence time of P, distributions of dissolved inorganic phosphorus (DIP) in the water column of the open ocean are mainly determined by oceanic circulation patterns, temporal and spatial variability in biological activity and the rate of recycling (Louanchi and Najjar, 2000). In surface waters, the DIP has a rapid turnover time (<â•›days to weeks) suggesting that low DIP levels can support a high primary production (Benitez-Nelson, 2000). Turnover times for Dissolved Organic Phosphorus (DOP) are typically longer (> months). The DOP must first by hydrolyzed to DIP prior to uptake by phytoplankton but the rate of this regeneration from DOP is not well quantified (Ruttenberg, 2003; Paytan and McLaughlin, 2007).

Dust input

Figure€8.3 The marine phosphorus cycle (modified from Paytan and McLaughlin, 2007).

River input Photosynthesis Weathering & Human activities

Remineralization Loss to Coastal Zone Upwelling

Hydrothermal removal

154

Grazing Export of organic matter

Remineralization

Burial

Maren Voss

Fish debris

Organic P water

Figure€8.4 The sedimentary P cycle (modified from Slomp et€al., 1996). Fish debris refers to the bones and scales of fish; these consist of hydroxyl apatite.

sediment Fish debris

Organic P

Dissolved PO4

Fe-oxide P

Carbonate Fluorapatite (CFA)

In coastal environments, sediments play a critical role in the recycling and net removal of DIP. Fe-oxide bound P is most important as a temporary sink for P (Slomp et€al., 1998; Figure€ 8.4), with seasonal or more long-term variations in organic matter remineralization and sediment redox conditions playing a role in regulating sediment-water exchange fluxes of DIP. In the Baltic Sea, for example, water column DIP concentrations are correlated with the extent of the area of hypoxia, suggesting large scale periodic release of P from Fe-(hydr) oxide pools in the sediment (Conley et€ al., 2002). Authigenic Ca-P and organic P are the major burial sinks for P, both in near shore (Ruttenberg and Berner, 1993) and offshore (Slomp et€al., 1996) marine settings.

Increased P inputs to the coastal zone by humans Increased terrestrial inputs of N and P to the marine environment since 1950 are greatly modifying coastal nutrient cycling and are leading to problems with eutrophication and hypoxia worldwide. Given the enhanced release of P from sediments under low oxygen conditions, these changes are increasing the availability of P for primary producers in many coastal systems. The changes in the P cycle can be directly linked to human activities and are the result of the increased use of P fertilizers in agriculture and the discharge of P-containing wastewater to rivers and coastal waters (Mackenzie et€al., 2002). The results of a recent spatially explicit modelling study (Seitzinger et€ al., 2005 and references therein) suggest that anthropogenic activities account for 65% of the DIP exported to the coastal zone at the global scale, while the remainder is derived from natural weathering. Point sources (mainly human sewage) are by far the dominant source of anthropogenic DIP. Humans account for only 19% of total DOP export, with diffuse sources being dominant. Although both DIP and DOP export to the coastal zone is significant (at 1.09 and 0.67 Tg yr–1, respectively), total riverine P inputs to the coastal zone are dominated by particulate P (PP; 9.03 Tg yr–1). However, only a small part of this PP is likely to be bioavailable. In general, PP:DIP ratios are predicted to be lower in systems with more human activity (Seitzinger et€al., 2005).

Increased nitrogen loading is driving many large rivers to higher DIN/DIP ratios, affecting the phytoplankton community structure and the occurrence of harmful algal blooms. Increased submarine groundwater discharge of nutrients may further modulate nutrient ratios in coastal waters since DIN/ DIP ratios in fresh groundwater are typically far above the Redfield ratio (N:P = 16:1; Slomp and Van Cappellen, 2004). Apart from restricted basins, nutrient dynamics in more offshore areas are dominated by ocean inputs and are, as yet, not affected by anthropogenic P-inputs (Jickells, 1998).

8.5.3╇ Si-linkage and eutrophication In contrast to N and P riverine fluxes (which have been strongly modified in the past 50 years), silica fluxes (which originate essentially from the weathering of rocks) has remained rather constant or even decreased, due to eutrophication and/or trapping in reservoirs (Figure€ 8.5). Therefore silica has become a limiting factor for river diatoms in the main branch of the large rivers resulting in lower DSi/DIN and DSi/P ratios in estuaries and coastal regions. Whereas increased N, P deliveries to the coastal zone are recognized as a major threat to the ecological functioning of near shore coastal ecosystems, less attention has been paid to their imbalance in regard to silica (see Officer and Ryther, 1980; Conley et€al., 1993; Turner and Rabalais, 1994; Justic et€al., 1995a; Justic et€al., 1995b; Billen and Garnier, 1997; Turner et€ al., 1998; Conley, 1999; Humborg et€ al., 2000; Cugier et€al., 2005; Billen and Garnier, 2007; Humborg et€al., 2008). However, water column P/Si and N/Si ratios determine the phytoplankton community structure, especially the shift from diatoms to non-diatoms and these changes may have major impacts on water quality in the proximal, i.e. nearshore part of the coastal zone (Turner et€ al., 2003; Cugier et€al., 2005; Howarth and Marino, 2006).

8.5.4╇ Oxygen consumption and hypoxia Hypoxia in bottom waters, e.g. oxygen concentrations <2 ml l–1, is a growing problem worldwide (Diaz, 2001) and occurs when

155

Nitrogen processes in coastal and marine ecosystems

Figure€8.5 ‘Anthropocene’ silica transfers from land to sea. Routing of riverine silica, silica cycling, and retention along the aquatic continuum. Strahler ordination of the head water and diffuse Si sources; reservoir Si transformation and retention; large wetland Si retention; large city input of DSi and BSi; estuarine Si transformation and accumulation of BSi in the turbidity maximum; DSi limitation at the coastal zone and BSi recycling at sediment interface and redistribution within the water column (from Garnier et€al., 2006).

oxygen supply through physical processes does not meet the demand for the biological processes that consume it. Hypoxia is one of the common effects of eutrophication in coastal marine ecosystems and is not only harmful for organisms (see Section€8.8) (Diaz and Rosenberg, 1995, 2008), but also disrupts biogeochemical processes (Vahtera et€al., 2007). Essentially all of the nitrogen transforming processes are regulated by oxygen concentrations, thus making hypoxia a significant process for the evaluation of nitrogen turnover in coastal and marine ecosystems (Figure€8.6). As nitrogen removal (denitrification and anammox) is dependent on NO2– and NO3– produced in oxic conditions by nitrification, these processes are of key importance in enabling nitrogen removal from coastal and marine ecosystems. Two important factors allow for hypoxia to occur in marine environments. First, the bottom water must not be able to be mixed with the surface water so that the bottom layer is isolated from the surface layer. This happens when there is a large difference in salt content between the surface and the bottom water or a difference in temperature during the warm summer months. This causes the bottom water to be cut off from the normal resupply of oxygen from the surface. Second, there is organic matter that can be decomposed by bacteria utilizing the oxygen in the bottom waters. The source of the organic matter that decomposes and uses the oxygen in bottom water is mostly from the growth of phytoplankton. The growth is stimulated by nutrients, both N and P, delivered to the marine environment by rivers and from run-off from land. The increase in nutrient concentration in coastal waters has caused the amount of algae to increase and thus the amount of hypoxia to increase.

156

Significant alterations in nitrogen turnover processes occur with hypoxia and have been extensively studied in shallow coastal marine ecosystems. One of the most significant responses observed are large releases of NH4+ from sediments during hypoxia (Koop et€al., 1990) due to decreases in the sediment demand for NH4+, i.e. for nitrification and assimilation. The loss of nitrification is due to a number of different processes including the lack of oxygen, which is required for nitrification to occur, the lack of available NO3– substrate (Kemp et€al., 1992), and or the inhibition of nitrification by sulphide poisoning of the process (Joye and Hollibaugh, 1995). After reoxygenation of bottom water, nitrification can be rapidly re-established if the period of hypoxia has been short (Hietanen and Lukkari, 2007), however, it may not occur immediately with time needed for a nitrifying population to be re-established if sulphide poisoning occurs (Dalsgaard, 2003). Another important sink for remineralized NH4+ is the assimilation by the benthic microalgae community during primary production in shallow sediments where light can reach the bottom. Benthic primary production creates a demand for NH4+ at the sediment surface and benthic microalgae may act as a barrier for the release of NH4+ from the sediment to the water column (Sundbäck and McGlathery, 2005). The benthic primary producers die with extended hypoxia, thus removing this important sink. Furthermore, the pelagic primary production can increase during hypoxia. Significant increases in both pore water and bottom water NH4+ and PO43– concentrations are found during hypoxia (Souchu et€al., 1998). These nutrients can be mixed into surface waters creating algal blooms (Conley et€ al., 2007), and through sedimentation adding new organic matter to bottom waters that can keep the system hypoxic. The lack of oxygen during hypoxia can have significant effects on nitrogen removal and eutrophication in coastal and marine ecosystems (Smith and Hollibaugh, 1989). Denitrification and anammox are key processes removing NO2– and NO3–, yet the starting products, e.g. both nitrite and nitrate, are produced in oxic conditions. Nutrient enrichment from N and P input from watersheds increase the productivity of aquatic systems and potentially increase the occurrence of hypoxia, thus limiting the ability of the system to remove nitrogen and increasing the effects of eutrophication. In deep coastal and marine ecosystems and the enclosed seas of Europe it is rare that hypoxia occurs throughout the bottom waters and is often confined to the deepest waters. In these waters, it is possible to have denitrification occurring in the water and not only at the sediment-water interface. Recently, a significant negative correlation was found between the amount of dissolved inorganic nitrogen and the volume of hypoxic water in the Baltic Sea (Vahtera et€al., 2007), suggesting enhanced nitrogen removal during large-scale hypoxia. Water column denitrification has been found to occur at the interface between anoxic, stagnant deep water and overlying oxic water in the central Baltic Proper using acetylene blockage (Rönner and Sörensen, 1985; Brettar and Rheinheimer, 1992). However, a recent study using state-of-the-art stable isotope techniques failed to detect any denitrification potential in Baltic sub-oxic, sulphide-free water (Hannig et€al., 2007),

Maren Voss Figure€8.6 Schematic showing the interaction of N- and P-cycling depending on oxygen concentrations (modified from Boesch et€al., 2001).

however, bacteria capable of anammox were observed in the water column (Hannig et€al., 2007). In addition, potential for N2 production was found in the sulphide-containing, deeper layer, making it very likely a chemolithotrophic rather than heterotrophic process (Hannig et€ al., 2007). Chemolithtic N2 production occurs in the water column redox transition zone of the Cariaco Trench (Taylor et€al., 2001). The rates of water column denitrification can be substantial in sub-oxic deep waters (Deutsch et€al., 2007). Thus, in deep coastal and marine ecosystems the overall rates of denitrification and the loss of N may in fact be the same order of magnitude or even greater with hypoxia, although more research is need to clarify and estimate the rates.

8.6╇ Physical and biogeochemical processes in estuarine systems Estuaries are presented here briefly but most aspects on riverine loads and links to coastal eutrophication can be found in Billen et€ al. 2011 (Chapter 13 this volume). They mediate and transform land to ocean fluxes of nitrogen. An estuary is defined as part of the river confluence that experiences reversing landward flow due to tides. Estuaries come in all sizes and shapes ranging from large drowned river valleys, deep glacial excavated silled-fjords, tectonic depressions and much smaller and shallower longshore bar-build coastal lagoons. Whatever their size the estuaries are characterized by large salinity gradients resulting from the mixing of riverine freshwater from land with tidal inflow from the seawater. Apart from the plethora of sizes and shapes, the challenge in studying nitrogen transformation processes in estuaries stems from the large temporal and spatial variability inherent to these systems requiring a unique understanding of each estuarine system. Estuarine circulation patterns are highly variable and are influenced by their size and shape, tidal mixing and fresh water flux. The basic premise, however, is a downslope seaward flow

of lower-density fresher water at the surface and an inflow of saline water at depth (Figure€ 8.7). Entrainment of salt water into the outflowing upper layer is replaced by the underlying inflow from the sea. The distinct landward bottom flow of salt water that develops in some estuaries is called a salt wedge. Although only some estuaries develop a salt wedge all are characterized by a two-layer circulation which plays a pivotal role in N transformation in estuaries by trapping particles and nutrients. With the exception of very fine-grained material, sedimentary particles that are carried downstream to the estuary settle as the flow velocity decreases when the confined river flow discharges into the relatively open coastal zone. The landward-moving net bottom flow acts to carry particles upestuary, so that particles settling from the surface outflow can literally be carried back up the estuary at depths. In addition, when the particles are introduced to saltier water, a high ionic strength medium, they tend to aggregate to form larger and thus faster-settling particles. This process, termed flocculation, promotes sedimentation of riverine particulate organic matter in estuaries. Further, the salinity gradient has been shown to contribute to phosphorus desorption (Némery and Garnier, 2007) whereas silica dissolution could be ten times higher in the saline waters compared to the freshwater ones (Roubeix and Lancelot, 2008; Garnier et€ al., 2006). All these estuarine processes may affect the nutrient fluxes delivered to the marine waters. Finally, wetland systems bordering estuaries (tidal flats, reed beds, salt marshes, etc.) retards flow and retains particles particularly when the estuaries overflow into adjoining wetlands due to high tides and flood conditions. The net result is that estuaries effectively trap particulate organic matter delivered from rivers and produced within the estuary preventing it from reaching the open sea. One noteworthy feature of estuarine sedimentation is the turbidity maximum. This feature is maintained by the residual circulation, and is intrinsic to the idea that estuaries trap particles. The turbidity maximum consists of sediments settling

157

Nitrogen processes in coastal and marine ecosystems

typical zonation

low salinity

high turbidity

high Zooplankton high Chlorophylla

water flow in the estuary

net heterotroph

Figure€8.7 Idealized schematic of the zonation of an estuarine system showing major processes and currents along the gradient from fresh water (green) to full marine waters (blue).

net autotroph

from the river input plus resuspended bottom sediments brought up estuary by the saline bottom flow mixed with estuarine algal material and other anthropogenic discharges. The high concentrations of particles promote flocculation within the turbidity maximum. Some of the particles (characteristically those that are relatively fine-grained) are mixed upward into the outward-flowing surface layer, only to settle back later into the bottom inflow. Thus, a turbidity maximum should be viewed as a dynamic feature. In extreme cases, the turbidity maximum may stretch many kilometres and maybe so concentrated that it becomes a fluid mud lens. An example of this is seen in the Gironde Estuary in Southern France. Estuaries along European coasts are mostly net heterotrophic systems, where bacterial-mediated degradation of organic matter (heterotophy) exceeds primary production by estuarine phytoplankton (autotrophy). The surplus organic matter for heterotrophy is brought mostly from the land but also from the sea in salt wedge and partially-mixed estuaries. The turbidity maximum can host intense bacterial activity in the upper estuary due to high concentrations of particulate organic matter where fresh planktonic remains are mixed with old organic matter derived from land and resuspended from sediments. Thus the turbidity maximum can be viewed analogous to a bio-reactor, turning estuaries into hotspots for nitrogen processing through regeneration, nitrification and denitrification. In general, the high turbidity in the upper estuary inhibits phytoplankton growth by limiting light penetration. The net result is that many estuaries are net autotrophic only in the outer part towards the open ocean aided by nutrients carried from upper estuary (Figure€8.7). Anthropogenically and terrestrially-derived N enters estuaries through riverine transport or via atmospheric deposition, including both direct deposition to the surface of estuaries and indirect watershed runoff (Castro and Driscoll, 2002). In addition estuaries are historical population centres and today’s urban settlements release N-compounds also through direct discharge. Excess N input can lead to eutrophication, whereby overall increases in primary production are accompanied by major shifts in the dominant flora (Duarte, 1995) and fauna (Heip, 1995) if light penetration of the water column is adequate. The fact that estuaries trap particulate and dissolved

158

front

nitrogen makes them prone to severe oxygen depletion and in some extreme cases they can become anoxic. Such changes also alter N transformation processes and the capacity of estuaries to retain N (see Section 8.5.4) (Horrigan et€al., 1990). The extent to which DIN is processed in estuaries appears to be influenced by the size and fresh water residence time in European estuaries. For instance in large continental estuaries nitrate and ammonia showed removal during estuarine mixing (Middelburg and Nieuwenhuize, 2001). In contrast, in the smaller UK estuaries of the Tyne and Tweed rivers very little removal of nitrate and ammonia occurred which is attributed to their low fresh water residence time (Ahad et€al., 2006). Therefore alteration of estuarine circulation due to human activities can profoundly influence the estuarine capacity to process N. Intertidal and subtidal mud flats are areas of enhanced capacity to retain N through sedimentation resulting from retarding water flow and denitrification. For instance in the Humber estuary (UK), organic N burial is estimated to be 216 tonnes annually of the 57 400 tonnes of the annual total N load. The intertidal area of the Humber also enhances N processing by denitrification of about 997 tonnes of N. Increasing these intertidal areas by management for coastal defences could significantly increase this N retention within the estuary (Andrews et€al., 2006; Jickells, 2006). The extent of processing of Nr within estuaries therefore depends on a variety of factors such as oxygen status (particularly for processes involving redox change such as cycling of ammonia/ammonium and production of N2O and N2) and light climate for both benthic and water column photosynthesis. The other key issue is residence time which depends on estuarine volume and exchange rates controlled by river inflow and tidal exchanges with coastal waters. The residence times of waters in estuarine systems can vary from hours to months depending on the hydrographic setting. While the input of organic N and nitrate has declined in many urban estuaries due to sewage treatment, ammonia inputs through untreated sewage still occur. Elevated NH4+ inputs may lead to enhanced generation of N2O via nitrification in estuarine and coastal zones (Hashimoto et€al., 1999; de Wilde and de Bie, 2000). A comparative study of an urban and a rural estuary in UK showed evidence for intense nitrification in the urban estuary in the adjoining coastal zone (Ahad et€al.,

Maren Voss

2006). Therefore, increases in the atmospheric concentration of N2O related to human activity could have wide-reaching implications for global climate change (Galloway et€ al., 1995) by further enhancing already significant coastal emissions (Bange et€al., 1996; see Section 8.4). Currently global N inputs into the coastal zone are based on riverine fluxes and do not account for estuarine modification of these fluxes. Reliable global/ European estimates of N processing in estuaries are currently lacking due to the large variability between individual estuaries in processing N.

8.7╇ N-budgets in coastal systems€– linking inputs to transformations and losses The coastal ocean is a highly dynamic and spatially heterogenÂ� eous compartment at the land–ocean interface. It includes estuaries, salt marshes, seagrasses, mangroves, coral reefs and open continental shelf systems. Coastal systems receive relatively large amounts of nutrients and particles from land via rivers and groundwater and exchanges matter with the open ocean and the atmosphere. Coastal ecosystems are relatively shallow implying that sediment processes (including benthic primary production, nitrification, denitrification and anammox) play a major role. Moreover, most of the global (marine) carbon and nitrogen burial occurs in coastal sediments (Duarte et€al., 2005).

8.7.1╇ Input terms The delivery of land-derived nitrogen to the coastal ocean has been studied extensively and total riverine nitrogen input is about 66 Tg N yr−1. However, there is substantial spatial variability around the world in the magnitude and form of the N delivery (Seitzinger et€al., 2005). Moreover, present-day riverine N delivery to the ocean has already doubled relative to preindustrial delivery and it will probably further increase due to increased anthropogenic activity. Globally, about 40% of the total nitrogen is delivered as DIN, another 40% as PN and the remaining 20% as DON. We assume here that 50% of this N is natural and 50% anthropogenic, consistent with a doubling of total N delivery relative to pre-industrial inputs. Submarine groundwater discharge is increasingly recognized as a potential source of nutrients to coastal systems. The magnitude of groundwater nitrogen input is highly uncertain because of our limited knowledge of submarine groundwater input, the difficulty in separating net from gross water fluxes and few data on the nitrogen distribution and processing in aquifers. Present-day estimates of submarine groundwater discharge are on the order of 5%–10% of surface water inputs and if one assumes similar dissolved nitrogen concentrations as surface waters this relates to less than 4 Tg N yr−1. This input may be particularly important in certain regions, for instance limestone areas with few rivers, or certain regions of glacial sands. At the global scale, atmospheric N deposition to continental shelves is on the order of 8.4 Tg N yr−1 as of the mid 1990s compared to pre-industrial deposition of 1.4 Tg N yr−1

(Galloway et€al., 2004). Therefore, we assume 7 Tg N yr−1 are anthropogenic. Atmospheric N deposition rates are highly variable spatially, temporally, and in terms of composition. Deposition rates vary with distance from source and one would expect higher deposition rates in coastal systems than in the open ocean systems, but the relative importance of atmospheric versus other external N sources is likely lower in coastal systems. Biological N2 fixation represents a net addition of new nitrogen to oceanic systems and has been recognized as one of the key processes governing the marine nitrogen budget. During the last decade we have significantly advanced our knowledge on the identity and distribution of organisms involved in open ocean nitrogen fixation and we have seen a steady increase in published estimates of oceanic nitrogen fixation. However, few studies and consequently little progress have been made on nitrogen fixation in the coastal ocean. Capone (1988) summarized nitrogen fixation rates in the coastal zone and reported a total of about 15 Tg N yr−1, mainly due to the benthic compartment in shallow systems (12 Tg N yr−1). Nitrogen fixation estimates for the continental shelf and upwelling systems are scarce. Galloway et€al. (2004) reported a global shelf biological nitrogen fixation rate of 1.5 Tg N yr−1. Given the other terms in mass balance, in particular denitrification and exchange between shelf and open ocean waters, we will not further discuss biological nitrogen fixation although we do admit that this estimate may be highly conservative and may need upward revision.

8.7.2╇ Output terms Nitrogen loss terms include burial in coastal sediments, fish landings, and gaseous emission in the form of dinitrogen gas, nitrous oxide and other nitrogen containing gases. Nitrogen burial represents a small loss term and can be estimated from intensively studied carbon burial estimates for the coastal ocean assuming a C:N ratio of 10 for buried organic matter. We estimate coastal ocean (except bays and lagoons) burial to be about 4 Tg N yr−1. Global nitrous oxide emission from coastal systems is about 0.5–2.9 Tg N yr−1. Fish landings have been estimated at 3.7 Tg N yr−1 (Maranger et€al., 2008) and we attribute all of this to coastal systems. The conversion of fixed nitrogen into gaseous dinitrogen represents the major loss terms of nitrogen from the Earth system, the ocean and as well from the coastal systems. Most nitrogen gas production can be attributed to denitrification, the microbial conversion of nitrate via nitrite and nitrous oxide to dinitrogen, but some nitrogen gas is produced by the recently discovered process Anammox, the bacterially mediated process of anaerobic oxidation of ammonia with nitrite. For reconstruction of the coastal ocean nitrogen mass balance, it is not of first priority to partition nitrogen gas production to denitrification or Anammox or between the distal and proximal coastal ocean. Denitrification in estuaries amounts to about 8 Tg N yr−1 (Seitzinger et€al., 2006). Nitrogen gas production in continental shelf systems is primarily limited to sediments and governed

159

Nitrogen processes in coastal and marine ecosystems Table€8.4 Summary of global sources and sinks for the coastal seas in Tg N yr−1. Details in the text

Source

Tg N yr–1

Loss

River input (Seitzinger et€al., 2005)

66

Burial (Middelburg et€al., 1996)

4

Ground water

4

N2O emissions (Bange, 2006)

0.5–2.5

Atmospheric deposition (Galloway et€al., 2004)

8.4

Fish landing (Maranger et€al., 2008)

3.7

N-fixation (Capone, 1988)

Sum

15

93.4

Denitrification estuaries (Seitzinger et€al., 2006)

8

Denitrification shelf (Middelburg et€al., 1996; Seitzinger et€al., 2006)

107–250

Sum

123.2–268.6

primarily by organic nitrogen and carbon delivery to the sediments and only secondarily by nitrate concentration in the bottom water because most of the nitrate and or nitrite converted to dinitrogen are formed in the sediments (coupled nitrification-denitrification). Recent estimates of denitrification in continental shelf sediments vary from about 107 Tg N yr−1 (Middelburg et€ al., 1996, for continental shelf) to about 250 Tg N yr−1 (Seitzinger et€al., 2006). Although these authors have used different approaches, i.e. a more mechanistic model by Middelburg et€ al. and a more empirical model by Seitzinger et€al., both these estimates depend on the delivery of particulate organic carbon and nitrogen to shelf sediments. This flux of particulate organic matter to shelf sediments depends in turn on two main factors:€(1) the primary production of shelf ecosystems and (2) the water depth, because more organic matter is degraded in the water column with increasing depth so that less is available for degradation in the sediments (Middelburg and Soetaert, 2004).

8.7.3╇ Balance If we balance nitrogen inputs to the coastal systems with nitrogen losses (Table€8.4) due to burial and nitrous oxide emission (together 28.5 Tg N yr−1) and fish landings (3.7 Tg N yr−1), we can derive a minimum denitrification in the coastal zone of 61 Tgâ•›/â•›N yr−1, including 8 Tg N yr−1 denitrification in estuaries. This mass balance based number of 61 Tg N yr−1 is smaller than the sum of independent estimates for estuaries (8 Tg N yr−1) and global shelf sediment denitrification (107 and 250 Tg N yr−1). This implies that coastal systems import between 123 and 270 Tg N yr−1 from the open ocean, making coastal-open exchange a major input term. Although this nitrogen balance is not very accurate and will certainly be revisited in the coming years, it is clear that nitrogen losses due to sediment denitrification are primarily compensated by net import of nitrogen from the open ocean, consistent with previous nitrogen budget studies. We assume that all N input from the open ocean is natural. Walsh (1991) in his seminal paper on the importance of continental margins in marine biogeochemical cycling of carbon and nitrogen reported a transfer from the open ocean onto shelves of 560 Tg nitrate-N yr−1. This indicates that the net transfer of nitrogen onto continental shelves from the open ocean represents 9%–35% of total nitrogen flow. These calculations of the coastal zone as a net sink for oceanic N rather than

160

Tg N yr –1

a source are all based on model estimates, in part because estimating exchange rates between coastal waters and offshore are very difficult. Exchange rates between the continental shelves and open ocean are very variable temporally and spatially and difficult to quantify given the multiple processes involved. In a very thorough review Huthnance (1995) estimated that about 1 Sv (106 m3 s−1) is exchanged for each 1000 km of shelf edge. For a total shelf break length of 314 400 km, we then obtain a water exchange rate of 9915 × 1012 m3 yr−1. This then implies very small difference between inflowing and outflowing water (about 1 mmol N m−3) to balance the nitrogen budget. Such a very small difference in nitrogen concentration is difficult if not impossible to accurately constrain given the dynamic nature of cross-shelf edge exchange processes and natural spatial heterogeneity of nitrogen stock and cycling processes. In addition, at least during parts of the year, there are likely to be steep gradients in N concentrations with depth over the upper few hundred metres of the water column, and hence the depth from which offshore water is brought into coastal areas as well as the amount of such exchange is important.

8.8╇ Effects on coastal and marine biodiversity resulting from eutrophication Coastal and marine waters around Europe have to face a variety of anthropogenic impacts. Among those, nutrient enrichment has been seen as one major source for changes in pelagic and benthic communities and thus biodiversity. Similar changes and impacts have also been observed in other non-marine aquatic (Grizetti et€al., 2011, Chapter 17, this volume) or terrestrial (Dise et€al., 2011, Chapter 20, this volume) ecosystems. The biological responses to this nutrient enrichment are often grouped together under the term eutrophication. Generally, in coastal and marine waters, eutrophication can have both positive and negative effects on species’ occurrence and abundance. Biotic responses to organic enrichment and threshold levels show great variations in coastal and marine ecosystems, partly because eutrophication varies from region to region. How and to what extent the structure and functioning of marine ecosystems is affected by eutrophication also depends on other factors such as prevailing hydrographical conditions, climatic forcing, the amount and duration of exposure to increased nitrogen loads or other anthropogenic impacts, e.g. contaminants and habitat destruction. This may lead to synergistic effects which

Maren Voss

makes it more difficult to distinguish cause and effects and relate changes to a single environmental stressor. Nevertheless, there are some common features concerning the succession of reactions to increased organic matter and subsequent changes in diversity and ecosystem functioning.

8.8.1╇ Increase in biomass and change in species composition Increasing nutrient loads stimulate primary production as long as enough light is available, resulting in higher biomass and increased sedimentation of pelagic production. With an excess to N and P compared to Si phytoplankton biomass may increase, but spring diatoms may be replaced by unpalatable algae (e.g. Phaeocystis in the North Sea (Lancelot et€ al., 1987)) or toxic algae (e.g. Dinophysis in the Seine Bight (Cugier et€al., 2005)). Such shifts at the first trophic level can deeply modify the complex interactions within the food web (Pace et€al., 1999). Active suspension feeders (e.g. bivalves) can regulate pelagic primary production when the water is shallow, the residence time is long, and the suspension feeder biomass is high (Cloern, 1996). Filter-feeding and deposit-feeding macrobenthic species benefit from high production, but the relative dominance of major trophic groups changes as was shown for polychaetes. With increasing organic input, non-selective deposit feeders were favoured over suspension feeders and selected deposit feeders (Grall and Chauvaud, 2002). Furthermore, zoobenthos plays a major role in benthic nutrient regeneration, affecting primary production by supplying nutrients directly to the water and enhancing recycling rates. Experimental studies have shown that the impact of benthic fauna on benthic–pelagic coupling and nutrient release is considerable (Grall and Chauvaud, 2002). In the presence of macroÂ�fauna, degradation was faster and more efficient. Moreover, it became obvious that species’ function also play a role for particle degradation, transport and recycling. For instance, NH4+ and NO3– release were three times higher in sediments where the polychaete Nereis diversicolor€– a suspension feeder- was present, and were only 1.5 times higher in sediments where Nereis virens€– a deposit feeder€– was present. Moreover, the presence of both Nereis species increased both NO3– and silicate fluxes by two orders of magnitude (Grall and Chauvaud, 2002). Forster and Zettler (2004) demonstrate that the presence/absence of the bivalve Mya arenaria in the southern Baltic Sea impacts pore water-exchange and thus benthopelagic matter transport and fluxes.

8.8.2╇ Hypoxia Under ‘normal’ conditions the amount of primary production is in equilibrium with grazing and remineralization processes (Graneli et€ al., 1990). However, in case sedimentation rate exceeds the grazing capacity or food requirements of benthic organisms it will generate layers of biodeposits (Grall and Chauvaud, 2002), which leads to increased oxygen consumption in the near-bottom water and surface sediments causing hypoxia with severe consequences for certain biogeochemical

processes as discussed in Section 8.5.4. Moreover, it could have a strong impact on the local fauna and its diversity. The occurrence of hydrogen sulphide causes an additional stress factor for many benthic organisms (Diaz and Rosenberg, 1995; Gray et€al., 2002; Diaz and Rosenberg, 2008). Benthic species differ in their tolerance to hypoxia (Gray et€al., 2002), nevertheless, mass mortality of benthos and fish over large areas have been reported for many marine coastal areas, and sensitive species have been permanently or periodically removed (Wu, 2002). Prolonged presence of hypoxia affects the important role of benthic organisms for organic matter degradation, nutrient remineralization and particle fluxes. Hypoxia reduces growth and feeding of benthic organisms and eventually their general fitness. Prolonged hypoxia might eliminate sensitive species, and thereby cause major changes in species composition. Hypoxia decreases species diversity and species richness and changes abundances of functional groups (Wu, 2002). During hypoxia there is a general tendency for suspension feeders to be replaced by deposit feeders, demersal fish are replaced by pelagic fish and macrofauna by meiofauna (Wu, 2002). As mentioned before, tolerance of and sensitivity to low oxygen levels differ among species. However, the critical oxygen level for many benthic species is around 2.8 mg O2 l–1 while certain species can tolerate 0.5–1 mg O2 l–1 for several days or weeks (Wu, 2002; Vaquer, 2008). Polychaetes belong to a taxon most tolerant to stress due to organic loading and hypoxia. Therefore, frequent hypoxia may favour this taxonomic group. Moreover, seasonal hypoxic events often result in an increase in opportunistic species. For many Scandinavian and Baltic marine waters changes in dominant benthic species, their abundance and biomass have been documented in relation to oxygen deficiency caused by eutrophication (Karlson et€al., 2002; Berezina and Golubkov, 2008). Severe changes in species’ composition and trophic structure have also been observed in the Black Sea with dramatic consequences for the ecosystem. In€the Mediterranean, the analysis of long term time series from the lagoon of Venice documents how increased nutrients loads interact with changes in biotic compartments. Among others, a clear change in diversity is obvious as well as in species’ composition, turning a system with mixed feeders to one where detritus feeders predominate with a short life time (Pranovi et€al., 2008).

8.8.3╇ The role of macrophytes Dense assemblages of macrophytes covering sediments are also of great importance for nutrient cycling and they are likely to intercept nutrients from both the water column and the sediments. It has been argued that the water quality in macrophytes-dominated shallow-water systems is much better than in phytoplankton-dominated systems with similar nutrient loadings (Grall and Chauvaud, 2002). Presently, it has been observed that benthic eutrophication in estuaries and coastal lagoons can induce a shift from rooted plant communities, dominated by slow-growing species, like the eelgrass Zostera, towards free-floating (or partially free-floating), faster-�growing macroalgae, such as Enteromorpha or Ulva (Patricio et€ al.,

161

Nitrogen processes in coastal and marine ecosystems

2004). A large scale comparison of empirical relationships between distribution and abundance of marine vegetation documents that seagrasses and macroalgae generally respond to changes in eutrophication pressure by growing deeper, being more abundant and more widely distributed in clear waters with low nutrient concentrations as compared to more turbid and nutrient-rich environments (Krause-Jensen et€al., 2008). Eutrophication profoundly changes rocky shore communities. The abundance of perennial macroalgae and their associated communities have severely declined in the western Baltic and the Adriatic Seas where they have been replaced by few bloom-forming annual species (Vogt and Schramm, 1991; Munda, 1993). Similar shifts in rocky shore diversity have also been reported for the coasts of the North Sea (Thompson et€al., 2002) or the Baltic Proper (Kautsky, 1991; Berger et€al., 2003). These examples indicate that changes in species composition due to eutrophication do occur in different ecological compartments; phytoplankton being the first functional group where changes impact zooplankton and benthic communities but also fish. Based on studies performed in Danish waters a conceptual model has been developed (Figure€ 8.8) indicating the different interactions between trophic levels in response to increased nutrient loads (Ærtebjerg et€al., 2003).

8.8.4╇ Link to ecosystem functions and functioning Changes in species’ composition as described above are linked to losses of various functional types such as feeding guilds or plant growth forms or other functionally similar taxa. It is widely accepted that ecosystem functioning is dictated to a large degree by biodiversity and the community structure defined by

species richness and evenness and diversity. In turn community diversity characterizes an ecosystem (Gray, 1997). Changes in the diversity and therefore structure of the community may result in loss of specific functions with consequences for the functioning of the ecosystem (i.e. production, consumption, respiration, energy flow and cycling). As benthic communities are changed, biologically mediated geochemical cycles will be altered. The presence or absence of benthic fauna drastically alters the rates and pathways of organic matter mineralization, as well as overall sediment features. Species richness also affect production of ammonium in surface sediments and benthic oxygen and nutrient fluxes (Emmerson et€al., 2001; Raffaelli et€al., 2003; Waldbusser and Marinelli, 2006). In the Baltic proper, the deposition of organic matter has increased 1.7 times from the 1920s to the 1980s. During the same period the biomass of benthic fauna has increased although the number of benthic species has decreased (Karlson et€al., 2002). Bottom areas with laminated sediments, indicating lack of macrobenthic bioturbation have increased rapidly since 1960. While in the 1940s and 1950s less than 20â•›000 km² of Baltic Sea sediments were laminated sediments€– most of them naturally occurring, they covered about 70â•›000 km2 in 1990 (Jonsson et€al., 1990). Most of these areas correspond to the occurrence of severe hypoxia in bottom water. Hypoxia may also affect the functional diversity of microbial communities (Olenin, 1997; Mermillod-Blondin et€al., 2004). In contrast to fully marine ecosystems, the inner parts of Baltic Sea bays and fjords are often populated by one single species representing a functional guild (Olenin, 1997; Bonsdorff

Figure€8.8 Conceptual model illustrating the effects and consequences of nutrient enrichment and eutrophication in the marine environment (Ærtebjerg et€al.,€2003).

162

Maren Voss

and Pearson, 1999). This makes the systems even more sensitive to changes on bottom oxygen concentrations. Disappearance of such a key species would result in the loss of the functional group which, in turn, may change the biogeochemical cycling of the system essentially. A rough estimate has been made of the extent of bioturbation by a key-species, the polychaete Scoloplos armiger, and its role in biogeochemical cycling in one part of the Baltic Sea. Based on the particle reworking rate of Scoloplos spp. and the area occupied by this species it was calculated that it reworks 1.9 × 109 tons of sediments annually (Dippner et€al., 2008). High diversity seems to be important for maintaining ecosystem processes under changing conditions providing a ‘buffer’ against environmental fluctuations (Loreau et€al., 2001). Ecosystems with a higher degree of biodiversity and thus functional redundancy are assumed to be able to cope better with increased nutrient loads. On the other hand species which have successfully established in an ecosystem with strong natural gradients (e.g. Baltic Sea) might have higher tolerance levels to additional stress. Eutrophication processes do not occur in isolation but rather in parallel with other global change pressures including changing coastal management (Andrews et€al., 2006) and climate change, a process that has been shown for example to have caused major changes in plankton throughout the North Sea and Europe (Beaugrand et€al., 2002). When assessing the impact of eutrophication on organisms and species, climate change related changes and impacts on ecosystems have to be taken into account. Presently, we have only started to understand how an increase in temperature might affect primary production and its timing, and its likely consequences for other trophic levels and nutrient cycling. The adaptation capacity of organisms to higher temperature has to be considered and that the distribution range of certain species is likely to change due to climate change.

8.9╇ Examples from regional seas 8.9.1╇ The Mediterranean Sea example The Mediterranean is a large enclosed sea, connected through narrow sills to the Atlantic Ocean by the Strait of Gibraltar and to the Black Sea by the Strait of Dardanelles (Çanakkale). Its average depth is around 1500 m, with considerable spatial variability:€the Eastern Basin has several trenches exceeding 4000 m in depth, while the northern Adriatic, the shallowest part of the Mediterranean Sea, is less than 200 m deep. Unlike the other regional examples, the Mediterranean is oligotrophic (nitrate concentration in surface waters typically <0.2 μm) (EEA, 1999). There is a west-to-east nutrient gradient, with average concentrations in the waters of the Aegean about a factor of 12 less than in the Atlantic and eight times lower than in the Alboran Sea between Spain and Morocco (EEA, 1999). There is a limited supply of nutrients to the surface waters from the deep and from external sources (Cruzado, 1988), and the physical dynamics of the Mediterranean€– an outflow of highsalinity, high-nutrient deep waters, and inflow of nutrient-poor

surface Atlantic water€– and the rapid recycling of nutrients by the ecosystem (Krom et€al., 2005b; Lucea et€al., 2005) result in a tendency not to accumulate nutrients (Bryden and Stommel, 1984; Souvermezoglou, 1988; Caddy, 1993). Coste (1987) reports nitrate concentrations in the intermediate/deep waters of the Western Basin of 8.5 μm, while in the Eastern Basin, concentrations are around 6.0 μm N (Souvermezoglou et€al., 1992; Yilmaz and Tugrul, 1998). Unusually, primary production in the Mediterranean is phosphorus limited, most strongly in the Eastern Basin (Krom et€al., 2005a), while nitrogen and phosphorus co-limitation is evident elsewhere in the Mediterranean (Montserrat Sala et€al., 2002).

Nitrogen inputs and trends Although the main body of the Mediterranean is a low-nutrient system and assessments are that eutrophication is not a serious threat overall (EEA, 1999; Danovaro, 2003), nutrient enrichment is a concern in the near-coastal environment (Figure€8.1), where local and temporary occurrences of high biomass and productivity are seen especially in areas of restricted flow such as lagoon systems (Tett et€al., 2003). Winter mixing is a significant factor (Krom et€al., 1991a; Allen et€al., 2002), but the most concerns relate to nutrient enrichment in river plume areas, particularly in the shallow Adriatic (Degobbis and Gilmartin, 1990; Druon et€al., 2004). River inputs to the Mediterranean Sea are dominated by discharges from its northern side, so this is primarily a European concern. Although the Nile’s discharge is the greatest, at 89 km3 yr−1 at the level of the Aswan dam, its discharge is reduced to less than 5 km3 yr−1 as it reaches the Mediterranean (Skliris and Lascaratos, 2004), making the European inputs the major sources of nutrients and particulate matter (Bethoux, 1980; UNEP/FAO/WHO, 1996; UNEP-MAP, 2003). About a third of the annual water inflow comes from just two rivers:€ the Rhone (54 km3 yr−1) flowing into the Gulf of Lions, and the Po (46 km3 yr−1) flowing into the northern Adriatic (UNEP-MAP, 2003). Most of the Mediterranean rivers are affected by nitrate enrichment, albeit to a lesser degree than northern European rivers (UNEP-MAP (2003) reports average concentrations of 1.24 mg N l–1 for the 30 highest-output rivers flowing into the Mediterranean). However, nutrient enrichment is aggravated by the strong seasonality of river flows in this hot, dry region, which leads to episodic high concentrations in the river waters. The trends and the relative importance of riverine nitrogen input to the Mediterranean remain somewhat unclear. The EEA (1999) flags the ‘scarcity or unavailability of comparable and, in some cases, reliable data’ as a major concern, and more recently EEA (2005) reports that there is as yet no source apportionment for nutrient inputs into the Mediterranean (see also Kronvang et€al., 2004, 2005). Nevertheless, ‘in all documented cases’ (EEA, 1999), the overall riverine nitrogen input to the Mediterranean has increased (EEA, 1999; UNEP-MAP, 2003). Inputs almost doubled from ~330 kt N yr–1 prior to 1975 to ~600 kt N yr−1 in 1995, although in recent years, the increase is apparently slowing (UNEP-MAP, 2003). (Phosphate loads have dropped to about 1975 values as a result of the widely applied ban of phosphorus detergents and the general improvement

163

Nitrogen processes in coastal and marine ecosystems

of wastewater treatment facilities.) McGill (1969) determined that river run-off supplied 30% of the total input of nitrogen and phosphorus. Bethoux and Copin-Montegut (1986) used 1984 UNEP data to estimate terrestrial nitrogen discharges as 50%–70% of total input to the sea. Atmospheric deposition of nitrogen species is also significant (Bashkin et€al., 1997; Guerzoni et€al., 1999; Sandroni et€al., 2007), and has been associated with episodic phytoplankton blooms (Guerzoni et€ al., 1999). Loÿe-Pilot et€ al. (2004) reassessed the nitrogen budget for the Mediterranean Sea using 1974 UNEP regional data, and concluded that atmospheric inputs of dissolved inorganic nitrogen are of the same order as fluvial inputs in the western and northwestern zones of the Mediterranean (to the south of the more industrialised nations). The already high intensity of shipping in the Mediterranean is projected to increase (REMPEC, 2008), with implications for this component of the nitrogen budget. The Po and Rhone rivers, which dominate river inputs to the Mediterranean, also have the greatest potential impacts. They are both in catchments with major industrial and urban centres and agricultural activity, and both flow into shallower regions of the Mediterranean:€ the Po River flows into the northern Adriatic, and the Rhone into the Gulf of Lions (the continental shelf break is typically <10 km along the Western basin, but east of the Golfe de Fos it widens to over 50 km). The coastal and lagoon ecosystems influenced by these rivers and those in other intensely cultivated river basins of Italy, Spain (particularly the Ebro river) and Greece (the Axios, which also flows into the Adriatic) have been comparatively well-studied, for example in two recent EU integrated coastal/river catchment projects: • OAERRE (OAERRE 2010) and • EuroCat (EuroCat 2010) and • the Italian research network LaguNET (LaguNET 2010). Agriculture is a major source activity, with diffuse sources contributing ~55% of the nitrogen in the Po river (Crouzet et€al., 1999; Behrendt, 2004) and >80% in the Axios (Behrendt, 2004; Milovanovic, 2006). Sewage, waste-water and other point sources are regionally variable, accounting for 43% of total nitrogen in the Po (Crouzet et€ al., 1999), a somewhat higher proportion than for the other main rivers (Table€8.5).

Ecological status of the Mediterranean Sea Among all the European seas briefly presented here, the Mediterranean seems to be the most unaffected by human nutrient input. However, coastal areas especially along the northern shore are affected by enhanced productivity and algae blooms. The dam building activities e.g. in the Nile change the river flow and sediment transport considerably and seem to influence the fishery. The role of atmospheric deposition may therefore be more important than in other coastal areas with high riverine nutrient loads.

Future of the Mediterranean Sea

164

Across most of the Mediterranean, there is still limited nitrogen abatement (primarily waste-water treatment), including on the high-flow Po river. A recent UNEP-MAP Global Environment Facility (GEF) activity (UNEP-MAP/RAC/CP, 2004) prepared

Table€8.5 Relative importance of the different nitrogen species in the Po, Rhone and Ebro rivers (NH3 contributes ~5% of DIN in all rivers)

TN

DIN

DON

PN

%

%

%

3.1

71.0

17.0

12.0

Rhone

1.7

84.8

8.4

6.9

Ebro

2.6

80.8

11.5

7.7

mg l Po

1 2

3

−1

Sources:€(1) Tartari et€al. (1991), (2) Pont (1996), (3) Munoz and Prat (1989). Cited in UNEP-MAP (UNEP-MAP, 2003).

regional guidelines on good nutrient management practices that will be implemented more widely in coming years to ensure compliance with European legislation. Nevertheless, with the continuing intensification of agriculture, urbanization and shipping, particularly in the eastern end of the Mediterranean, and in a context where a comparatively small proportion of the coastline is protected under legislation and monitored, threats to ecosystems from anthropogenic nitrogen remain.

8.9.2╇ The Baltic Sea example The Baltic Sea is a shallow brackish water sea average depth 52.3m with limited water exchange through the Danish sounds, resulting in an average residence time of the water of 30 years. Surrounded by heavily industrialized countries and a population of c. 85 million people the Baltic Sea is strongly impacted by nutrient loading, consequently eutrophication is the major problem of the Baltic Sea (HELCOM, 2004). The impacts of eutrophication are manifested as various symptoms such as increased nutrient concentrations and phytoplankton biomass, and oxygen deficiency and elimination of benthic fauna (review by Lundberg, 2005).

Nutrient input and trends The early signs of eutrophication were already present in early 1900 close to the larger municipalities that discharged their waste waters directly with minimal treatment to the coastal waters (Laakkonen and Laurila, 2007). Since then the nutrient loading from rivers has increased approximately seven times for nitrogen and four times for phosphorus (HELCOM, 2004) and the pools of nutrients have also increased considerably in the whole Baltic Sea (Larsson et€al., 1985; Nehring and Matthäus, 1991). Recently decreasing trends of nutrient levels have been reported in different parts of the Baltic (FlemingLehtinen et€al., 2008). However, many regions, particularly the Gulf of Finland, Baltic Proper, and the Southern Baltic, show clear signs of eutrophication (HELCOM, 2002; Rönnberg and Bonsdorff, 2004; Fleming-Lehtinen et€al., 2008). The total riverine nitrogen load to the Baltic Sea varies between 600–800 ktons per year being closely related to the river run-off. About half of the total riverine nitrogen load is discharged via three large rivers:€Vistula, Oder and Nemunas to the Southern Baltic Sea (HELCOM, 2002). Diffuse loading from agriculture and forestry is the major source of nitrogen (70%–95%) to the Baltic catchments (HELCOM, 2004). Other major nitrogen sources of the Baltic Sea are atmospheric deposition, estimated to be 185 ktons N in 1997 (HELCOM,

Maren Voss

1997), and nitrogen fixation, estimated to vary between 370 ktons N (Wasmund et€ al., 2001) and 926 ktons N per year (Schneider et€al., 2003). The effects of riverine nitrogen loading are more pronounced in the coastal areas than in the open Baltic Proper, and the coastal sediments appear to be very efficient in removing nitrogen by denitrification (Voss et€al., 2005; Vahtera et€al., 2007).

Ecological status of the Baltic Sea The anoxic conditions in the Baltic are promoting release of inorganic phosphorus from the sediments (Pitkänen et€ al., 2001; Conley et€al., 2002) and may increase denitrification as the area covered by oxic and anoxic interfaces increase (Vahtera et€al., 2007). The supply of phosphorus favours the blooms of cyanobacteria that fix their nitrogen from the dissolved N2. This has been described as a vicious circle resembling the situation in many lakes where internal sources of nutrients maintain eutrophication (Tamminen and Andersen, 2007; Vahtera et€al., 2007). The internal processes and feedbacks that maintain and enforce the eutrophied status of the Baltic Sea, suggest that there has been a regime shift from an earlier more oligotrophic to more eutrophic status that may be difficult to reverse (Österblom et€al., 2007).

The future of the Baltic Sea The EU Water Framework Directive calls for restoration of the ‘good ecological status’ of coastal waters by 2015. This is a key piece of legislation for the protection of the Baltic Sea as it requires holistic management of pressures and impacts on the catchment scale including the coastal zone. The Baltic Sea Action Plan (HELCOM, 2007) sets the ecological objectives and country allocations for nutrient reductions based on best scientific knowledge and information available. The Baltic Sea Action Plan is a pilot for implementing the new EU Marine Strategy Directive that calls for good marine environmental status to be reached in 2020, and adds to the requirements of the EU Water Framework Directive in the coastal zone. Nutrient loading reductions that are required to reach the ‘good status’ of the Baltic are anticipated to be costly. Therefore more drastic management actions have been suggested to improve the status of the open Baltic Sea. Large scale engineering with artificial ventilation of the Baltic Proper deep waters with concurrent fertilization with nitrogen has been suggested to halt the ‘vicious circle’ of internal phosphorus loading and cyanobacterial blooms in the Baltic Proper (Stigebrandt and Gustaffson, 2007). Artificial ventilation could keep the patient alive, but not necessarily cure the cause of the disease. Halting the biological pump producing organic material that is suffocating the ‘patient’ would require tuning down the supply of both nitrogen and phosphorus at the source (Tamminen and Andersen, 2007; Vahtera et€al., 2007). Treating the central regions of the Baltic Sea separately would not necessarily improve the coastal areas where the majority of people meet the sea, and which provide a number of ecosystem services to humans. The future climate change scenarios include a potential increase in the temperatures and precipitation in Northern Europe. Concurrent potential intensification of agricultural

practices, and increased precipitation and transport of nutrients to the warmer Baltic Sea is a scary scenario that calls for urgent planning to counteract the risk of magnified eutrophication.

8.9.3╇ The Black Sea example The Black Sea is almost cut off from the rest of the world’s oceans and is up to 2200 m deep. The only connection to the open sea is through the Bosporus Strait which has a depth of only 40 m. The Black Sea drains an extensive catchment consisting of 23 countries, covering a land area of 2 400 000 km2, and receiving waste water from more than 190 million people (daNUbs, 2005). The freshwater input to the Black Sea amounts to 350€km³, and comes from the largest European Rivers the Danube, Dnieper and Dniester. The Black Sea is naturally stratified below approximately 70 m, and is the world’s largest permanently anoxic water body. The mean residence time of the water is approximately 1000 years. Because it receives considerable riverine input, the Black Sea in general and the shallow northwestern shelf in particular are susceptible to the effects of nutrient loading. During the 1960s the Soviet agricultural revolution resulted in intensive fertilizer use and livestock production in the Black Sea catchment (Mee et€al., 2005). This led to severe eutrophication of the Black Sea’s northwestern shelf during the 1970s and 1980s. During these years, the shelf experienced an increase in the magnitude, extent and frequency of algal blooms (Bodeanu, 1993) as well as the occurrence of a number of harmful algal bloom (or red tide) events (Moncheva et€al., 2001). The phytoplankton composition changed drastically between 1960 and 1990 (Bodeanu, 2002), much of the shelf area was hypoxic (Zaitsev and Mamaev, 1997), and the decline and in some cases loss of benthic floral and faunal communities was observed (Zaitsev and Mamaev, 1997). Alterations to the phytoplankton community precipitated changes in the zooplankton, including a decrease in non-gelatinous zooplankton species and an increase in gelatinous species such as Noctiluca scintillans and Mnemiopsis leidyi, an invasive ctenophore (Shiganova and Bulgakova, 2000). Species diversity and abundance of several valuable fish species (bonito, Sarda sarda; bluefish, Pomatomus saltatrix; flounder, Platichthys flesus; turbot, Psetta maxima; sole, Solea lascaris) decreased (Shiganova and Bulgakova, 2000). These disruptions to the food web, along with unregulated fishing efforts, damaged area fisheries (Daskalov, 2002).

Nutrient input and trends From the 1950s to the early 1980s, the annual discharge of nitrates to the Black Sea increased from 155 000 to 340 000 tons (Zaitsev and Mamaev, 1997). Overall, shelf waters of the Black Sea are phosphorus-limited and open waters are nitrogenÂ�limited (Cociasu et€al., 1998; daNUbs, 2005). Riverine input is the primary source of land-based nitrogen to the Black Sea, contributing approximately 63% of the Black Sea’s anthropogenic nitrogen load, 52% of which comes from the Danube River alone (Figure€8.9a, b) (Black Sea Environmental, 1996; Commission, 2002). Although the amount of fertilizer

165

Nitrogen processes in coastal and marine ecosystems Seaof Bulgaria Dniester Don Georgia Romania 1% Azov 7% 12% 4% 0% 7% Dneiper 2% Russia 2% Turkey 6% Danube 52% Ukraine 7%

Nitrogen

(b)

3500 3000 Fertilizer consumption (kt)

(a)

600

Atmospheric deposition

400

Ktyr–1

2000 1500 1000 500 1960

500

1970

1980

1990

2000

Figure€8.10 Consumption of fertilizer in the Danube catchment (after Mee,€2006).

Domestic

300

24 Industrial

200

22

Riverine

100

20 18

TN

µM

0

Figure€8.9 Nitrogen inputs to the Black Sea (a) by source (Black Sea Environmental, 1996) and (b) by anthropogenic sector. Atmospheric contribution applies to northwestern shelf exclusively (data from Commission, 2002; Langmead et€al., 2007).

16 14 12 10 8

1980

1984

1988

1992

1996

2000

2004

Figure€8.11 Time-series of total nitrogen concentrations measured at Constanta, Romania (from Cociasu and Popa, 2004).

900 800 700 600 Kt Year –1

consumed in the Danube catchment has decreased significantly (Mee, 2006) (Figure€8.10), the amount of nitrogen in Danube waters has decreased only slightly due to retention of the nutrient in catchment soils and groundwater (daNUbs, 2005); this is reflected in the lack of clear trend in nitrogen concentrations in shelf waters (Figure€ 8.10). The decreased nitrogen load since 1990 (Figure€8.11) may be attributed to implementation of the Nitrates Directive in EU countries (Union, 1991; Commission, 2002), reduced fertilizer usage after the collapse of the communist bloc (Commission, 2002), and greater waste water treatment (Commission, 2002). As a result of decreasing nitrogen loads in the Danube, the concentration of nitrogen in the Black Sea shelf has recently begun to decline (Figure€8.12) (Cociasu and Popa, 2004). Industry is the second greatest contributor of land-based nitrogen (30%) directly to the Black Sea (Commission, 2002). Most industrial nitrogen entering the Black Sea comes from Ukraine (49%), Russia (30%) and Bulgaria (18%) (Figure€8.13a) (Black Sea Environmental, 1996). Domestic waste water is a minor contributor of nitrogen (<5%) to the Black Sea, as is atmospheric deposition (10%) (Figures€ 8.9b, 8.13b). The majority of industries in all Black Sea coastal states are connected to municipal wastewater treatment systems. Therefore, with implementation of the Urban Waste Water Treatment Directive in the accession states and recovering economies in other Black Sea nations enabling more efficient sewage treatment, the amount of untreated domestic and industrial

166

2500

500 400 300 200 100 0

1960

1990

2000

Agriculture Background Urban sources (diffuse and point) Other diffuse sources Figure€8.12 Sources of nitrogen to the Danube (data from daNUbs, 2005; Mee et€al., 2005).

Maren Voss

(b) Domestic sources

(a) Industrial sources

Romania Turkey Georgia Bulgaria Russia Ukraine

1 7 578

Romania Russia Turkey Ukraine Georgia Bulgaria

TN (tons/year)

30, 954 44, 394

TN (tons/year)

414 877

1, 577 2, 482 5, 430 9, 514

71, 000

Figure€8.13 (a) Industrial and (b) domestic sources of nitrogen to the Black Sea (from Black Sea Environmental, 1996).

Figure€8.14 (a) The Greater North Sea and its major contributors; (b) SeaWiFS-derived chlorophyll a in the Greater North Sea for March–May 2003. Generated by NASA’s Giovanni (Giovann.gfsc.nasa.gov)

waste entering the Black Sea directly is expected to decrease (Commission, 2002).

Ecological status Although a recovery of the Black Sea ecosystem was observed in the past decades the system still seems to be fragile and needs attention and further nutrient reduction to become a stable oceanic environment. Nutrients are partly stored in sediments and deeper waters and can easily be mobilized (see below). The same may happen when nutrient loads from the rivers increase (Langmead et€ al., 2007) e.g. through land use changes in the drainage basins.

The future of the Black Sea Due to economic decline in the 1990s of all Black Sea coastal countries except Turkey, and the consequent lack of support for centrally administered agriculture, the marine ecosystem has recently shown signs of recovery (Mee et€ al., 2005). The

late 1990s saw fewer phytoplankton blooms than previous decades and the community structure has returned to a diatomdominated state (Bodeanu, 2002). No hypoxic events were recorded on the shelf between 1993 and 2000 (Mee, 2006). However, although possible, recovery of the Black Sea is far from certain. During the exceptionally warm years of 2001 and 2002, 15 monospecific blooms occurred in Romanian waters and the phytoplankton community was again dominated by non-diatoms (Bodeanu et€ al., 2004). Additionally, the climatic conditions of 2001 triggered a large scale hypoxic event which resulted in fish mortalities in shelf waters (Mee et€ al., 2005). These signs indicate that nutrient concentrations in shelf waters remain sufficiently elevated to support eutrophic events, particularly in combination with anomalous climatic conditions. Additionally, overfishing continues, with further depletion of remaining stocks likely (Langmead et€al., 2007). Despite recent ecosystem improvements, the Black Sea’s future remains uncertain. Because of expected growth of farming in

167

Nitrogen processes in coastal and marine ecosystems

Eastern Europe and additional pressures from shipping, poorly regulated fisheries, sewage treatment without nutrient removal, and changes in climate, the Black Sea may potentially revert back to its eutrophic state (Langmead et€al., 2007). Alternatively, this period of newfound regional prosperity, particularly with the expansion of the EU to include Bulgaria, Romania, and possibly Turkey, may be used as an opportunity for remediation of the Black Sea ecosystem. Even if eutrophication and overfishing are mitigated, there is no guarantee that the ecosystem will return to its pre-eutrophic state (Mee et€al.,€2005).

8.9.4╇ The North Sea example The Greater North Sea is a semi-enclosed, epi-continental large marine ecosystem off North-Western Europe (Figure€ 8.14a). For the purpose of nutrient loading, the English Channel, the Kattegat and margins such as the Wadden Sea are considered as an integral part of the larger North Sea ecosystem (NSTF et€al., 1994). Depths greater than 100m prevail in the seasonallystratified Northern North Sea, a rather oceanic system dominated by the influence of Atlantic water.

Nutrient input and trends The southern part is a shallow (<50 m deep and <20 m near the coast) and continuously mixed region that receives the majority of the riverine nutrient loads to the North Sea (Seine, Thames, Scheldt, Rhine, Ems, Weser, Elbe; Figure€8.14a). Owing to the general water mass circulation, nutrient loads cumulate along a SW–NE gradient in the region (Lancelot et€al., 1987). In the plume area of these large rivers draining watersheds among the most industrialized, cultivated and densely populated (ca 185 million inhabitants) over the world, the increased nutrient loading has resulted in severe eutrophication problems visible as increased winter nutrient concentrations (Lancelot et€al., 1987; Anonymous, 1993; Radach and Pätsch, 1997) and spring phytoplankton biomass (Figure 8.14b; Radach and Pätsch, 1997), foam accumulation (Lancelot, 1995), oxygen deficiency (Radach, 1992) and toxin release (Maestrini and Graneli, 1991).

Ecological status of the North Sea Eutrophication in the North Sea manifests itself as undesirable blooms of essentially two Haptophytes:€ Phaeocystis globosa forming colony blooms every spring (April–May) in coastal waters off France, Belgium, the Netherlands and Germany. There are associated ‘foam events’ (Lancelot, 1995) and, the toxin-producer Chrysochromulina spp. blooming between April and August in the Kattegat and Skagerrak off the coasts of Denmark, Sweden and Norway (Dahl et€ al., 2005). The toxicity of Chrysochromulina varies however between species and within the same species, under control of environmental conditions (Johansson and Graneli, 1999). Extensive blooms of harmful C. polylepis were up to now only recorded in May–June 1988 when they decimated farmed fish (Dahl et€al., 2005). In contrast, foam deposits on the beaches along the French, Belgian, Dutch and German beaches are well-known recurrent spring phenomena that correspond to the explosive development of large mucilaginous colonies of Phaeocystis. These

168

colonies enlarge when growing and reach sizes unmanageable for the current copepods (Weisse et€ al., 1994). These blooms are not recent, having been reported already at the end of the nineteenth century (Cadée and Hegeman, 1991). However a model reconstruction of pristine time for the Seine and Scheldt watersheds (Lancelot et€al., 2009) suggest that grazable forms of Phaeocystis colonies were blooming under natural conditions in the Southern North Sea and efficiently transferred their production to higher trophic levels. Eco-physiological studies show that both these Haptophytes follow an early-spring diatom bloom (Rousseau et€ al., 2002; Dahl et€al., 2005) and their growth is sustained by excess nitrate under low phosphate conditions (Lancelot et€ al., 1998; Dahl et€al., 2005; Breton et€al., 2006). Both species have indeed shown their ability to use organic forms of phosphorus (Veldhuis and Admiraal, 1987; Estep and MacIntyre, 1989; Veldhuis et€ al., 1991; Nygaard and Tobiesen, 1993). Resistance to grazers was found as essential for forming exceptional blooms, Phaeocystis colonies by enlarging their size, Chrysochromulina by releasing toxins under phosphate depletion (Edvardsen et€al., 1990; Johansson and Graneli, 1999).

Linking harmful algal blooms with land-based nutrients It is accepted that coastal eutrophication problems over the world are caused not only by increased nutrient loads but rather by the unbalance in the delivery of nitrogen and phosphorus in excess with respect to silica (Officer and Ryther, 1980; Billen et€ al., 1991; Conley et€ al., 1993; Turner et€ al., 1998). Hence undesirable coastal eutrophication results with the development of non-siliceous algae responding to new sources of N and P. Land-based nutrients are discharged in the Greater North Sea through three pathways:€direct tributaries, river inputs and atmospheric deposition. Annual nitrogen and phosphorus inputs to the North Sea (Table€8.6.) estimated for the 1989–1990 period show the large contribution of rivers (60% N, 77% P) some 70% of which is brought by rivers Seine, Rhine and Elbe, and point to an excess of N with respect to P delivery (molar N:P > 16; Table€ 8.5). This N excess with respect to P propagates in the coastal area, possibly explaining the successful development of both Phaeocystis and Chrysochromulina (Lancelot et€al., 1987; Dahl et€al., 2005). Supporting this, a positive relationship has been found between Phaeocystis abundance and nitrates (Lancelot et€ al., 1998; Breton et€ al., 2006; Lancelot et€ al., 2007) in the Southern Bight of the North Sea while Chrysochromulina abundance has been shown to be positively correlated with spring and summer N:P (Dahl et€al., 2005).

The future of the North Sea Historical trends of river nutrient inputs into the North Sea and the related coastal eutrophication problems have been appraised based on compilation of available data of nutrients and phytoplankton biomass (Radach and Patsch, 1997), and model simulations constrained by reconstructed river nutrient loads based on either available data (Radach and Pätsch, 1997) or river simulations (Billen and Garnier, 2007; Lancelot et€al., 2007). All results point towards an accelerated increase

Maren Voss Table€8.6 Land-based sources of N and P (1989–1990) to the North Sea

N kt yr –1

P kt yr –1

Molar N:P

Rivers

720

46

34

Direct

88

14

14

412

—

>>>

1,220

60

45

Atmosphere Total

Sources:€Rendell et€al., 1993; OSPAR et€al., 2005.

of N and P river inputs by a factor of ~5 between 1950 and 1985 while Si loads were unchanged. This N and P increase was linked to the combination of three factors:€(i) the generalization of modern intensive agriculture and the use of synthetic fertilizers, (ii) the presence of polyphosphates in washing powders and (iii) the increased urban waste water collection and lack of efficient nutrient reduction treatment (Billen and Garnier, 2007). As a result spring phytoplankton biomass increased by a factor of ~3 in the river plume areas. This nutrient enrichment was beneficial to diatoms up to 1965 after which non-siliceous algae became highly dominant as for instances reported for the Helgoland time series (Radach and Pätsch, 1997) due to the imbalanced N:P:Si nutrient enrichment. Model simulations (Radach and Pätsch, 2007) suggest that the penetration of landbased nutrients extends to the area bordered by the line from the English coast south of 54 °N to northern Denmark, sustaining the high phytoplankton biomass recorded in that region by satellite-derived Chl a maps (Figure€8.14b). North of this area, the impacts of land based nutrient supply is much less evident with an increasing importance for offshore nutrient sources and little evidence of changes in plankton communities resulting from terrestrially sourced nutrients but important climatically driven changes (Beaugrand et€al., 2002). In the late 1980s, however, the perception of coastal eutrophication problems was reaching a maximum in countries bordering the greater North Sea and recommendations for reducing nutrient inputs to the North Sea were made (North Sea conference, 1987). Instruments and procedures were implemented such as the OSPAR Strategy to combat eutrophication (OSPAR et€al., 2005) and the Water Framework Directive of the European Union (2000/60/EC). As a result of improved wastewater treatment P loads decreased considerably for all rivers (~ 65% in 2004) while the decrease of N loads was less efficient (~â•›30% in 2004) due to the weak implementation of agro-environmental measures (Billen and Garnier, 2007). As a consequence, present-day measured and modelled river loads show severe imbalanced molar N:P ratiosâ•›>>â•›16 (range:€22–60 for the different rivers) which propagate directly in the continental coastal strip (winter N:P> 25; Rousseau et€al., 2004) and favour the development of harmful Phaeocystis and Chrysochromulina blooms in the North Sea. In agreement, model reconstruction of diatom and Phaeocystis blooms in the Southern Bight of the North show a positive feedback of decreased nutrient loads after 1990 to both diatoms and Phaeocystis with however a larger impact on diatoms (Lancelot et€al., 2007). We conclude that future management of nutrient emission reduction aiming at decreasing harmful algal blooms in the southern North Sea without impacting diatom blooms need

to target a decrease of N loads through the implementation of specific agro-environmental measures.

References Ærtebjerg, G., Andersen, J. H. and Hansen, O. S. (2003). Nutrients and Eutrophication in Danish Marine Waters: A Challenge for Science and Management. National Environmental Research Institute. Ahad, J. M. E., Ganeshram, R. S., Spencer, R. G. M. et€al. (2006). Evaluating the sources and fate of anthropogenic dissolved inorganic nitrogen (DIN) in two contrasting North Sea estuaries. Science of the Total Environment, 372, 317–333. Allen, J. I., Somerfield, P. J. and Siddorn, J. (2002). Primary and bacterial production in the Mediterranean Sea:€a modelling study. Journal of Marine Systems, 33, 473–495. Andrews, J. E., Burgess, D., Cave, R. R. et€al. (2006). Biogeochemical value of managed realignment, Humber estuary, UK. Science of the Total Environment, 371, 19–30. Anonymous (1993). North Sea Quality Status Report 1993, North Sea Task Force. ICES/OSPAR Commissions, London. Arrigo, K. R. (2005). Marine microorganisms and global nutrient cycles. Nature, 437, 349–355. Arrigo, K. R. (2007). Marine manipulations. Nature, 450, 491–492. Baker, A. R. et€al. (2007). Dry and wet deposition of nutrients from the tropical Atlantic atmosphere:€links to primary productivity and nitrogen fixation. Deep-Sea Research, 54, 1704–1720. Baker, A. R., Kelly, S. D., Biswas, K. F., Witt, M. and Jickells, T. D. (2003). Atmospheric deposition of nutrients to the Atlantic Ocean. Geophysical Research Letters, pp. 30. Bange, H. W. (2006). Nitrous oxide and methane in European coastal waters. Estuarine, Coastal and Shelf Science, 70, 361–374. Bange, H. W. (2008). Gaseous nitrogen compounds (NO, N2O, N2, NH3) in the ocean. In:€Nitrogen in the Marine Environment, ed. D. G. Capone et€al. Elsevier, Amsterdam, pp. 51–94. Bange, H. W., Rapsomanikis, S. and Andreae, M. O. (1996). Nitrous oxide in coastal waters. Global Biogeochem Cycles, 10, 197–207. Banse, K. (1973). On the interpretation of data for the carbon-tonitrogen ratio of phytoplankton. Limnology Oceanography, 695–699. Banse, K. (1994). Uptake of inorganic carbon and nitrate by marine plankton and the Redfield ratio. Global Biogeochem Cycles, 8, 81–84. Barcelos e Ramos, J., Biswas, H., Schulz, K. G., LaRoche, J. and Riebesell, U. (2007). Effect of rising atmospheric carbon dioxide on the marine nitrogen fixer Trichodesmium. Global Biogeochemical Cycles, 21. Bashkin, V. N., Erdman, L. K., Abramychev, A. Y. et€al. (1997). The Input of Anthropogenic Airborne Nitrogen to the Mediterranean Sea through its Watershed, MAP Technical Reports Series. UNEPMAP/MedPol/WMO, Athens. Beaugrand, G., Reid, P. C., Ibañez, F., Lindley, J. A. and Edwards, M. (2002). Reorganization of North Atlantic marine copepod biodiversity and climate. Science of the Total Environment, 296,€1692–1694. Behrendt, H. (2004). Past, Present and Future Changes in Catchment Fluxes. European Commission. Benitez-Nelson, C. R. (2000). The biogeochemical cycling of phosphorus in marine systems. Earth Science Reviews, 51, 109–135. Berezina, N. and Golubkov, S. M. (2008). Effect of drifting macroalgae Cladophora glomerata on benthic community dynamics in the easternmost Baltic Sea. Journal of Marine Systems, 74, 80–85. Berger, R., Henriksson, E., Kautsky, L. and Malm, T. (2003). Effects of filamentous algae and deposited matter on the survival of

169

Nitrogen processes in coastal and marine ecosystems Fucus vesiculosus L. germlings in the Baltic Sea. Aquatic Ecology, 37,€1–11. Bethoux, J. P. (1980). Mean water fluxes across sections in the Mediterranean Sea evaluated on the basis of water and salt budgets and of observed salinities. Oceanologica Acta, 3, 79–88. Bethoux, J. P. and Copin-Montegut, G. (1986). Biological fixation of atmospheric nitrogen in the Mediterranean Sea. Limnology and Oceanography, 31, 1353–1358. Billen, G. and Garnier, J. (1997). The Phison River Plume:€coastal eutrophication in response to changes in land use and water management in the watershed. Aquatic Microbial Ecology, 13, 3–17. Billen, G. and Garnier, J. (2007). River basin nutrient delivery to the coastal sea:€assessing its potential to sustain new production of non-siliceous algae. Marine Chemistry, 106, 148–160. Billen, G., Lancelot, C. and Meybeck, M. (1991). N, P and Si retention along the Aquatic Continuum from Land to Ocean. In:€Ocean Margin€Processes in Global Change, ed. R. F. C. Mantoura, J.-M.€Martin, and R. Wollast. John Wiley & Sons, New York, pp. 19–44. Billen, G., Silvestre, M., Grizzetti, B. et€al. (2011). Nitrogen flows from European watersheds to coastal marine waters. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Black Sea Environmental (1996). Black Sea Transboundary Diagnostic Analysis. Global Environmental Facility. Bodeanu, N. (1993). Microalgal blooms in the Romanian area of the Black Sea and contemporary eutrophication conditions. In:€Toxic Phytoplankton Blooms in the Sea, ed. T. J. Smayda and Y. Shimizu. Elsevier, Amsterdam, pp. 203–209. Bodeanu, N. (2002). Algal blooms in Romanian Black Sea waters in the last two decades of the 20th century. Cercatari Marine, 34, 7–22. Bodeanu, N., Andrei, C., Boicenco, L., Popa, L. and Sburlea, A. (2004). A new trend of the phytoplankton structure and dynamics in Romanian marine waters. Cercatari Marine, 35, 77–86. Bodenbender, J. and Papen, H. (1996). Bedeutung gasförmiger Komponenten an den Grenzflächen Sediment/Atmosphäre und Wasser/Atmosphäre. Sylter 20 Wattenmeer Austauschprozesse (SWAP), Projektsynthese, pp. 252–278. Boesch, D. F., Brinsfield, R. B. and Magnien, R. E. (2001). Chesapeake Bay eutrophication:€scientific understanding, ecosystem restoration, and challenges for agriculture. Journal of Environmental Quality, 30, 303–320. Bonsdorff, E. and Pearson, T. H. (1999). Variation in the sublittoral macrozoobenthos of the Baltic Sea along environmental gradients:€a functional-group approach. Australian Journal of Ecology, 312–326. Boyd, P. W. et€al. (2007). Mesocale iron enrichment experiments 1993–2005:€synthesis and future directions. Science, 315, 612–617. Bradley, P. B., Sanderson, M. P., Frischer, M. E. et€al. (2010). Inorganic and organic nitrogen uptake by phytoplankton and heterotrophic bacteria in the stratified Mid-Atlantic Bight. Estuarine, Coastal and Shelf Science, 88, 429–441. Brandes, J. A., Devol, A. H. and Deutsch, C. (2007). New developments in the marine nitrogen cycle. Chemical Reviews, 107, 577–589. Breton, E., Rousseau, V., Parent, J. Y., Ozer, J. and Lancelot, C. (2006). Hydroclimatic modulation of diatom/Phaeocystis blooms in the nutrient-enriched Belgian coastal waters (North Sea). Limnology and Oceanography, 51, 1–14. Brettar, I. and Rheinheimer, G. (1992). Influence of carbon availability on denitrification in the central Baltic Sea. Limnology and Oceanography, 37, 1146–1163.

170

Bronk, D. A., See, J. H., Bradley, P. and Killberg, L. (2007). DON as a source of bioavailable nitrogen for phytoplankton. Biogeosciences, 283–296. Bryden, H. L. and Stommel, H. M. (1984). Limiting processes that determine basic features of the circulation in the Mediterranean Sea. Oceanologica Acta, 7, 289–296. Caddy, J. F. (1993). Contrast between recent fishery trends and evidence from nutrient enrichment in two large marine ecosystems:€the Mediterranean and the Black Seas. In:€Large Marine Ecosystems:€Stress, Mitigation, and Sustainability, ed. K. Sherman et€al. American Association for the Advancement of Science, Washington DC, pp. 137–147. Cadée, G. C. and Hegeman, J. (1991). Historical phytoplankton data of the Marsdiep. Hydrobiological Bulletin, 24, 111–118. Capone, D. G. (1988). Benthic nitrogen fixation. In:€Nitrogen Cycling in Coastal Marine Environments, ed. T. H. Blackburn, John Wiley & Sons, NewYork, pp. 85–123. Capone, D. G., Zehr, J. P., Paerl, H. W., Bergman, B. and Carpenter, E. J. (1997). Trichodesmium, a globally significant marine cyanobacterium. Science, 276, 1221–1229. Castro, M. S. and Driscoll, C. T. (2002). Atmospheric nitrogen deposition to estuaries in the mid-Atlantic and northeastern United States. Environmental Science and Technology, 36, 3242–3249. Christensen, P. B., Rysgaard, S., Sloth, N. P., Dalsgaard, T. and Schwaerter, S. (2000). Sediment mineralization, nutrient fluxes, denitrification and dissimilatory nitrate reduction to ammonium in an estuarine fjord with sea cage trout farms. Aquatic Microbial Ecology, 27, 73–81. Cloern, J. E. (1996). Phytoplankton blooms dynamics in coastal ecosystems:€a review with some general lessons from sustained investigation of San Francisco Bay, California. Review of Geophysics, 34, 127–168. Cociasu, A. and Popa, L. (2004). Significant changes in Danube nutrient loads and their impact on the Romanian Black Sea coastal waters. Cercatari Marine, 35, 25–37. Cociasu, A., Popa, L. and Buga, L. (1998). Long-term evolution of the nutrient concentrations on the north-western shelf of the Black Sea. Cercatari Marine, 13, 29. Commission, Black Sea (2002). State of the Environment of the Black Sea:€Pressures and Trends:€1996–2000. Black Sea Commission. Conley, D. J. (1999). Biogeochemical nutrient cycles and nutrient management strategies. Hydrobiologia, 410, 87–96. Conley, D. J., Carstensen, J., Ærtebjerg, G. et€al. (2007). Longterm changes and impacts of hypoxia in Danish coastal waters. Ecological Applications, 17, 165–184. Conley, D. J., Humborg, C., Rahm, L., Savchuk, O. P. and Wulff, F. (2002). Hypoxia in the Baltic Sea and basin scale changes in phosphorus biogeochemistry. Environmental Science and Technology, 36, 5315–5320. Conley, D. J., Schelske, C. I. and Stoermer, E. F. (1993). Modification of the biogeochemical cycle of silica with eutrophication. Marine Ecology Progress Series, 81, 121–128. Cornell, S. E., Jickells, T. D., Cape, J. N. et€al. (2003). Organic nitrogen deposition on land and coastal environments:€a review of methods and data. Atmospheric Environment, 37, 2173–2191. Coste, B. (1987). Les sels nutritifs dans le basin occidental de la Méditerranée. Rapport Commission International Mer Méditerranée, 30,€399–410. Crouzet, P., Leonard, J., Nixon, S. et€al. (1999). Nutrients in European Ecosystems. EEA Environmental Assessment Report, Copenhagen.

Maren Voss Crusius, J., Kroeger, K., Bratton, J. et€al. (2008). N2O fluxes from coastal waters due to submarine groundwater discharge. Geochimica et Cosmochimica Acta, 72, A191–A191, Suppl. 191. Cruzado, A. (1988). Eutrophication in the pelagic environment and its assessment. In:€Eutrophication in the Mediterranean Sea:€Receiving Capacity and Monitoring of Long-term Effects, UNESCO Reports in Marine Science, Paris, France, pp.€57–66. Cugier, P., Billen, G., Guillaud, J. F., Garnier, J. and Ménesguen, A. (2005). Modelling the eutrophication of the Seine Bight (France) under historical, present and future riverine nutrient loading. Journal of Hydrology, 304, 381–396. Dahl, E., Bagoien, E., Edvardsen, B. and Stenseth, N. C. (2005). The dynamics of Chrysochromulina species in the Skagerrak in relation to environmental conditions. Journal of Sea Research, 54,€15–24. Dalsgaard, T. (2003). Benthic primary production and nutrient cycling in sediments with benthic microalgae and transient accumulation of macroalgae. Limnology and Oceanography, 48, 2138–2150. Danovaro, R. (2003). Pollution threats in the Mediterranean Sea:€an overview. Chemistry and Ecology, 19, 15–32. daNUbs (2005). DaNUbs:€Nutrient Management of the Danube Basin and its Impact on the Black Sea. DaNUbs. Daskalov, G. M. (2002). Overfishing drives a trophic cascade in the Black Sea. Marine Ecology Progress Series, 225, 53–63. de Wilde, H. P. J. and de Bie, M. J. M. (2000). Nitrous oxide in the Schelde estuary:€production by nitrification and emission to the atmosphere. Marine Chemistry, 69, 203–216. Degobbis, D. and Gilmartin, M. (1990). Nitrogen, phosphorus and biogenic silicon budgets for the Northern Adriatic Sea. Oceanologica Acta, 13, 31–45. Dentener, F. et€al. (2006). Nitrogen and sulfur deposition on regional and global scales:€a multimodel evaluation. Global Biogeochem. Cycles, 20. Deutsch, C., Sarmiento, J. L., Sigman, D. M., Gruber, N. and Dunne, J. P. (2007). Spatial coupling of nitrogen inputs and losses in the ocean. Nature, 445, 163–167. Diaz, J. D. (2001). Overview of hypoxia around the world. Journal of Environmental Quality, 30, 275–281. Diaz, R. J. and Rosenberg, R. (1995). Marine benthic hypoxia:€a review of its ecological effects and the behavioural responses of benthic macrofauna. Oceanography and Marine Biology, 33, 245–303. Diaz, R. J. and Rosenberg, R. (2008). Spreading dead zones and consequences for marine ecosystems. Science, 321, 926–929. Dippner, J. W., Vuorinen, I., Daunys, D. et€al. (2008). Climate-related marine ecosystem change. In:€Assessment of Climate Change for the Baltic Sea Basin, ed. T. B. A. Team, Springer, New York, pp.€309–377. Dise, N. B., Ashmore, M., Belyazid, S. et€al. (2011). Nitrogen as a threat to European terrestrial biodiversity. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Druon, J.-N., Schrimpf, W., Dobricic, S. and Stips, A. (2004). Comparative assessment of large-scale marine eutrophication:€North Sea area and Adriatic Sea as case studies. Marine Ecology Progress Series, 272, 1–23. Duarte, C. (1995). Submerged aquatic vegetation in relation to different nutrient regimes. Ophelia, 41, 87–112. Duarte, C. M., Middelburg, J. and Caraco, N. (2005). Major role of marine vegetation on the oceanic carbon cycle. Biogeosciences, 2,€1–8.

Duce, R. A. et€al. (1991). The atmospheric input of trace species to the world ocean. Global Biogeochemical Cycles, 5, 193–259. Duce, R. A. et€al. (2008). Impacts of atmospheric anthropogenic nitrogen on the open ocean. Science, 320, 893–897. Dugdale, R. C. and Goering, J. J. (1967). Uptake of new and regenerated forms of nitrogen in primary productivity. Limnology and Oceanography, 12, 196–206. Edvardsen, B., Moy, F. and Paasche, E. (1990). Hemolytic activity in extracts of Chrysochromulina polylepis grown at different levels of selinite and phosphate. In:€Physiological Ecology of Harmful Algal Blooms, ed. E. Graneli, B. Sundstrom and L. Edler, Springer, Berlin, pp. 190–208. EEA (1999). State and Pressures of the Marine and Coastal Mediterranean Environment. Office for Official Publications of the European Communities, Luxembourg, Downloadable from reports.eea.europa.eu/medsea/en/medsea_en.pdf EEA (2005). Source Apportionment of Nitrogen and Phosphorus Inputs into the Aquatic Environment. European Environment Agency, Copenhagen. Eilola, K. and Stigebrandt, A. (1999). On the seasonal nitrogen dynamics of Baltic proper biogeochemical reactor. Journal of Marine Research, 57, 693–713. Emmerson, M. C., Solan, M., Paterson, D. M. and Raffaelli, D. (2001). Consistent patterns and the idiosyncratic effects of biodiversity in marine ecosystems. Nature, 411, 73–77. Engel, A. (2002). Direct relationship between CO2 uptake and transparent exopolymer particles production in natural phytoplankton. Journal of Plankton Research, 24, 49–53. Engel, A., Goldthwait, S., Passow, U. and Alldredge, A. (2002). Temporal decoupling of carbon and nitrogen dynamics in a mesocosm diatom bloom. Limnology and Oceanography, 47, 753–761. Engel, A., Delille, B., Jacquet, S. et€al. (2004). Transparent exopolymer particles and dissolved organic carbon production by Emiliania huxleyi exposed to different CO2 concentrations:€a mesocosm experiment. Aquatic Microbial Ecology, 34, 93–104. Estep, K. and MacIntyre, F. (1989). Taxonomy, life cycle, distribution and dasmotrophy of Chrysochromulina:€a theory accounting for scales, haptonema, muciferous bodies and toxicity. Marine Ecology Progress Series, 57, 11–21. EuroCat (2010). www.cs.iia.cnr.it/EUROCAT/project.htm Fleming-Lehtinen, V., Laamanen, M., Kuosa, H., Haahti, H. and Olsonen, R. (2008). Long-term development of inorganic nutrients and chlorophyll a in the open northern Baltic Sea. Ambio, 37, 86–92. Follmi, K. B. (1996). The phosphorus cycle, phosphogenesis and marine phosphate-rich deposits. Earth Sciences Reviews, 40,€55–124. Forster, S. and Zettler, M. L. (2004). The capacity of the filter-feeding bivalve Mya arenaria L. to affect water transport in sandy beds. Marine Biology, 144, 1183–1189. Galloway, J. N. et€al. (2004). Nitrogen cycles:€past, present, and future. Biogeochemistry, 70, 153–226. Galloway, J. N., Schlesinger, W. H., Levy, H., Michaels, A. and Schnoor, J. L. (1995). Nitrogen-fixation€– anthropogenic enhancement€– environmental response. Global Biogeochemical Cycles, 9, 235–252. Garnier, J., Sferratore, A., Meybeck, M., Billen, G. and Dürr, H. (2006). Modelling silica transfer processes in river catchments. In: The Silicon Cycle:€Human Perturbations and Impacts on Aquatic Systems, ed. V. Ittekkot et€al., Island Press, Washington DC, p. 296. Geider, R. J. and La Roche, J. (2002). Redfield revisited:€variability of C:N:€P in marine microalgae and its biochemical basis. European Journal of Phycology, 37, 1–17.

171

Nitrogen processes in coastal and marine ecosystems Grall, J. and Chauvaud, L. (2002). Marine eutrophication and benthos:€the need for new approaches and concepts. Global Change Biology, 8, 813–830. Graneli, E., Wallström, K., Arsson, U., Granelli, W. and Elmgren, R. (1990). Nutrient limitation of primary production in the Baltic Sea area. Ambio, 19, 142–151. Gray, J. (1997). Marine Biodiversity:€patterns, threats and conservation needs. Biodiversity and Conservation, 6, 153–175. Gray, J., Shiu-sun Wu, R. and Or, Y. Y. (2002). Effects of hypoxia and organic enrichment on the coastal environment. Marine Ecology Progress Series, 238, 249–279. Grizzetti, B., Bouraoui, F., Billen, G. et€al. (2011). Nitrogen as a threat to European water quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Gruber, N. (2004). The dynamics of the marine nitrogen cycle and its influence on atmospheric CO2 variations in carbon–climate interactions. In:€Carbon-Climate Interactions, ed. M. Follows and T. Oguz, John Wiley & Sons, New York, pp. 1–47. Gruber, N. and Galloway, J. N. (2008). An Earth-system perspective of the global nitrogen cycle. Nature, 451, 293–296. Guerzoni, S. et€al. (1999). The role of atmospheric deposition in the biogeochemistry of the Mediterranean Sea. Progress in Oceanography, 44, 147–190. Hannig, M., Lavik, G., Kuypers, M. M. M. et€al. (2007). Shift from denitrification to anammox after inflow events in the central Baltic. Limnology and Oceanography, 53, 1336–1345. Harrison, P. J., Hu, M. J., Yang, Y. P. and Lu, X. (1990). Phosphate limitation in estuarine and coastal waters of China. Journal of Experimental Marine Biology and Ecology, 140, 79–87. Hashimoto, S., Gojo, K., Hikota, S., Sendai, N. and Otsuki, A. (1999). Nitrous oxide emissions from coastal waters in Tokyo Bay. Marine Environmental Research, 47, 213–223. Heip, C. (1995). Eutrophication and zoobenthos dynamics. Ophelia, 41, 113–136. HELCOM (1997). Airborne Pollution Load to the Baltic Sea 1991–1995. HELCOM (2002). Baltic Sea Environment Proceedings, Helsinki Commission.€Environment of the Baltic Sea area 1994–1998. Baltic Sea Environment Proceedings, Helsinki Commission. HELCOM (2004). The 4th Baltic Sea Pollution Load Compilation. Baltic Sea Environment Proceedings, Helsinki Commission. HELCOM (2007). Helcom Baltic Sea Action Plan. http://www.helcom. fi/BSAP/en_GB/intro/. Herbert, R. A. (1999). Nitrogen cycling in coastal marine ecosystems. Microbiology Reviews, 23, 563–590. Hietanen, S. and Lukkari, K. (2007). Effects of short-term anoxia on benthic denitrification, nutrient fluxes and phosphorus forms in coastal Baltic sediment. Aquatic Microbial Ecology, 49, 293–302. Horrigan, S. G., Montoya, J. P., Nevins, J. L. and McCarthy, J. J. (1990). Natural isotopic composition of dissolved inorganic nitrogen in the Chesapeake Bay. Estuarine, Coastal and Shelf Science, 30, 393–410. Howarth, R. W. and Marino, R. (2006). Nitrogen as the limiting nutrient for eutrophication in coastal marine ecosystems:€evolving views over three decades. Limnology and Oceanography, 51, 364–376. Hulth, S., Allerb, R. C., Canfield, D. E. et€al. (2004). Nitrogen removal in marine environments:€recent findings and future research challenges. Marine Chemistry, 94, 125–145. Humborg, C., Conley, D. J., Rahm, L. et€al. (2000). Silicon retention in river basins:€far-reaching effects on biogeochemistry and aquatic food webs in coastal marine environments. Ambio, 29, 45–51.

172

Humborg, C., Smedberg, E., RodriguezMedina, M. and Mörth, C.-M. (2008). Changes in dissolved silicate loads to the Baltic Sea€– The effects of lakes and reservoirs. Journal of Marine Systems, 73,€223–235. Hutchins, D. A., Fu, F.-X., Zhang, Y. et€al. (2007). CO2 control of Trichodesmium N2 fixation, photosynthesis, growth rates, and elemental ratios:€implications for past, present and future ocean biogeochemistry. Limnology and Oceanography, 52, 1293–1304. Huthnance, J. M. (1995). Circulation and water masses at the ocean margin:€the role of physical processes at the shelf edge. Progress in Oceanography, 35, 353. Hynes, R. K. and Knowles, R. (1984). Production of nitrous oxide by Nitrosomonas europaea:€effects of acetylene, pH and oxygen. Canadian Journal of Microbiology, 30, 1397–1404. IPCC (2007). Climate Change 2007:€The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel 25 on Climate Change. Cambridge University Press. Jickells, T. D. (1998). Nutrient biogeochemistry of the coastal zone. Science, 281, 217–222. Jickells, T. D. (2006). The role of air–sea exchange in the marine nitrogen cycle. Biogeosciences, 3, 271–280. Jickells, T. D. et€al. (2005). Global iron connections between desert dust, ocean biogeochemistry and climate. Science, 308, 67–71. Johansson, N. and Graneli, E. (1999). Cell density, chemical composition and toxicity of Chrysochromulina polylepis (Haptophyta) in relation to different N:P supply ratios. Marine Biology, 135, 209–217. Johnson, M. T. et€al. (2008). Field observations of the oceanatmosphere exchange of ammonia:€fundamental importance of temperature as revealed by a comparison of high and low latitudes. Global Biogeochemical Cycles, 22. Jonsson, P., Carman, R. and Wulff, F. (1990). Laminated sediments in the Baltic: a tool for evaluating nutrient mass balances. Ambio, 19, 152–158. Jorgensen, S. K., Jensen, H. B. and Sorensen, J. (1984). Nitrous oxide production from nitrification and denitrification in marine sediments at low oxygen concentrations. Canadian Journal of Microbiology, 30, 1073–1078. Joye, S. B. and Hollibaugh, J. T. (1995). Influence of sulfide inhibition of nitrification on nitrogen regeneration in sediments. Science, 270, 623–625. Justic, D., Rabalais, N. N., Turner, R. E. and Dortch, Q. (1995a). Changes in nutrient structure of river-dominated coastal waters:€stoichiometric nutrient balance and its consequences. Estuarine, Coastal and Shelf Science, 40, 339–356. Justic, D., Rabalais, N. N. and Turner, R. E. (1995b). Stoichiometric nutrient balance and origin of coastal eutrophication. Marine Pollution Bulletin, 30, 41–46. Karlson, K., Bonsdorff, E. and Rosenberg, R. (2007). The impact of benthic macrofauna for nutrient fluxes from Baltic Sea sediments. Ambio, 36, 161–167. Karlson, K., Rosenberg, R. and Bonsdorff, E. (2002). Temporal and spatial large-scale effects of eutrophication and oxygen deficiency on benthic fauna in Scandinavian and Baltic waters: a review. Oceanography and Marine Biology, 40,€427–489. Kautsky, H. (1991). Influence of Eutrophication on the distribution of phytobenthic plant and animal communities. Internationale Revue der gesamten Hydrobiologie, 76, 423–432. Kemp, W. M., Sampou, P. A., Garber, J., Tuttle, J. and Boynton, W.€R. (1992). Seasonal depletion of oxygen from bottom waters

Maren Voss of Chesapeake Bay: roles of benthic and planktonic respiration and physical exchange processes. Marine Ecology Progress Series, 85,€137–152. Koop, K., Boynton, W. R., Wulff, F. and Carman, R. (1990). Sedimentwater oxygen and nutrient exchanges along a depth gradient in the Baltic Sea. Marine Ecology Progress Series, 63, 65–77. Krause-Jensen, D., Sagert, S. and Schubert, H. C. B. (2008). Empirical relationships linking distribution and abundance of marine vegetation to eutrophication. Ecological Indicators, 8, 515–529. Krishnamurthy, A., Moore, J. K., Zender, C. S. and Luo, C. (2007). Effects of atmospheric inorganic nitrogen deposition on ocean biogeochemistry. Journal of Geophysical Research, 112. Krom, M. D., Brenner, S., Israilov, L. and Krumgalz, B. (1991a). Dissolved nutrients, preformed nutrients and calculated elemental ratios in the south-east Mediterranean Sea. Oceanologica Acta, 14, 189–194. Krom, M. D., Kress, N., Brenner, S. and Gordon, L. (1991b). Phosphorus limitation of primary production in the eastern Mediterranean. Limnology and Oceanography, 36, 424–432. Krom, M. D., Thingstad, T. F., Brenner, S. et€al. (2005a). Summary and overview of the CYCLOPS P addition Lagrangian experiment in the Eastern Mediterranean. Deep-Sea Research€– Part II, 52, 3090–3108. Krom, M. D., Woodward, E. M. S., Herut, B. et€al. (2005b). Nutrient cycling in the south east Levantine basin of the eastern Mediterranean:€Results from a phosphorus starved system. Deepsea Research€– Part II, 52, 2879–2896. Kronvang, B., Larsen, S. E., Jensen, J. P., Andersen, H. E. and Lo Porto, A. (2004). Catchment Report:€Enza, Italy – Trend Analysis, Retention and Source Apportionment, EUROHARP report 4–2004, NIVA report SNO 4787–2004. Oslo, Norway. Kronvang, B., Larsen, S. E., Jensen, J. P., Andersen, H. E. and Reisser, H. (2005). Catchment Report:€Vilaine, France. Trend Analysis, Retention and Source Apportionment. EUROHARP report 15–2005, NIVA report SNO 5081–2005. Oslo, Norway. Kuypers, M. M. M. et€al. (2003). Anaerobic ammonium oxidation by anammox bacteria in the Black Sea. Nature, 422, 608–611. Kuypers, M. M. M., Lavik, G. and Thamdrup, B. (2006). Anaerobic ammonium oxidation in the marine environment. In:€Past and Present Water Column Anoxia, ed. L. N. Neretin, Springer, New York, pp.€311–335. Laakkonen, S. and Laurila, S. (2007). Changing environments or shifting paradigms? Strategic decision making toward water protection in Helsinki 1850–2000. Ambio, 36, 212–219. LaguNET (2010). www.dsa.unipr.it/lagunet/english/index.htm Lam, P., Lavik, G., Jensen, M. M. et€al. (2009). Revising the nitrogen cycle in the Peruvian oxygen minimum zone. Proceedings of the National Academy of Sciences of the USA, 106, 4752–4757. Lancelot, C. (1995). The mucilage phenomenon in the continental coastal waters of the North Sea. Science of the Total Environment, 165, 83–112. Lancelot, C., Billen, G., Sournia, A. et€al. (1987). Phaeocystis blooms and nutrient enrichment in the continental coastal zones of the North Sea. Ambio, 16, 38–46. Lancelot, C., Keller, M., Rousseau, V., Smith Jr, W. O. and Mathot, S. (1998). Autoecology of the Marine Haptophyte Phaeocystis sp. In:€Series G. Ecological Science€– NATO Advanced Workshop on the Physiological Ecology of Harmful Algal Blooms, ed. D. A. Anderson, A. M. Cembella and G. Hallegraeff, John Wiley & Sons, New york, pp. 209–224. Lancelot, C., Gypens, N., Billen, G., Garnier, J. and Roubeix, V. (2007). Testing an integrated river-ocean mathematical tool for linking marine eutrophication to land use:€the Phaeocystis-dominated

Belgian coastal zone (Southern North Sea) over the past 50 years. Journal of Marine Systems, 64, 216–228. Lancelot, C., Rousseau, V. and Gypens, N. (2009). Ecologically based indicators for Phaeocystis disturbance in eutrophied Belgian coastal waters (Southern North Sea) based on field observations and ecological modeling. Journal of Sea Research, 61,€44–49. Langmead, O., McQuatters-Gollop, A. and Mee, L. D. (2007). European Lifestyles and Marine Ecosystems:€Exploring Challenges for Managing Europe’s Seas. University of Plymouth Marine Institute, Plymouth, UK. Larsson, U., Elmgren, R. and Wulff, F. (1985). Eutrophication and the Baltic Sea. Ambio, 14, 9–14. Levitan, O., Rosenberg, G., Setlik, I. et€al. (2007). Elevated CO2 enhances nitrogen fixation and growth in the marine cyanobacterium Trichodesmium. Global Change Biology, 13, 531–538. Loÿe-Pilot, M. D., Martin, J. M. and Morelli, J. (2004). Atmospheric input of inorganic nitrogen to the Western Mediterranean. Biogeochemistry, 9, 117–134. Loreau, M., Naem, S., Inchausti, P. et€al. (2001). Biodiversity and ecosystem functioning:€current knowledge and future challenge. Science, 294, 804–808. Louanchi, F. and Najjar, R. G. (2000). A global monthly climatology of phosphate, nitrate, and silicate in the upper ocean:€spring– summer export production and shallow remineralization. Global Biogeochemical Cycles, 14, 957–977. Lucea, A., Duarte, C. M., Agusti, S. and Kennedy, H. (2005). Nutrient dynamics and ecosystem metabolism in the Bay of Blanes (NW Mediterranean). Biogeochemistry, 73, 303–323. Lundberg, C. (2005). Conceptualizing the Baltic Sea ecosystem:€an interdisciplinary tool for environmental decision making. Ambio, 34, 433–439. Mackenzie, F. T., Ver, L. M. and Lerman, A. (2002). Century-scale nitrogen and phosphorus controls of the carbon cycle. Chemical Geology, 190, 13–32. Maestrini, S. and Graneli, E. (1991). Environmental conditions and ecophysiological mechanisms which lead to the Chrysochromulina bloom:€a hypothesis. Oceanologica Acta, 14, 397–413. Mahowald, N. et€al. (2008). The global distribution of atmospheric phosphorus deposition and anthropogenic impacts. Global Biogeochemical Cycles, 19. Mahowald, N. M. et€al. (2005). The atmospheric global dust cycle and iron inputs to the ocean. Global Biogeochemical Cycles, 19. Maranger, R., Caraco, N., Duhamel, J. and Amyot, M. (2008). Nitrogen transfer from sea to land via commercial fisheries. Nature Geoscience, 2, 111–113. Marchal, O., Monfray, P. and Bates, N. R. (1996). Spring–summer imbalance of dissolved inorganic carbon in the mixed layer of the northwestern Sargasso Sea. Tellus, 48B, 115–134. Mari, X. (2008). Does ocean acidification induce an upward flux of marine aggregates? Biogeosciences Discussion, 5, 1631–1654. McElroy, M. B. (1983). Marine biological controls on atmospheric CO2 and climate. Nature, 302, 328–329. McGill, D. A. (1969). A budget for dissolved nutrient salts in the Mediterranean Sea. Cahiers Océanographiques, 21, 543–554. Mee, L. D. (2006). Reviving dead zones. Scientific American, 295,€54–61. Mee, L. D., Friedrich, J. and Gomoiu, M.-T. (2005). Restoring the Black Sea in times of uncertainty. Oceanography, 18, 32–43. Mermillod-Blondin, F., Rosenberg, R., Francois-Carcaillet, F., Norling, K. and Mauclaire, L. (2004). Influence of bioturbation by three benthic infaunal species on microbial communities and

173

Nitrogen processes in coastal and marine ecosystems biogeochemical processes in marine sediments. Aquatic Microbial Ecology, 36, 271–284. Michaels, A. F., Bates, N. R., Buesseler, K. O., Carlson, C. A. and Knap, A. H. (1994). Carbon-cycle imbalance in the Sargasso Sea. Nature, 372, 537–540. Middelburg, J. J. and Nieuwenhuize, J. (2001). Nitrogen isotope tracing of dissolved inorganic nitrogen behaviour in tidal estuaries. Estuarine, Coastal and Shelf Science, 53, 385–391. Middelburg, J. J. and Soetaert, K. (2004). The role of sediments in shelf ecosystem dynamics. In:€The Sea, ed. A. R. Robinson and K. H. Brink, Harvard University Press, Cambridge, MA, pp.€353–374. Middelburg, J. J., Soetaert, K., Herman, P. M. J. and Heip, C. H. R. (1996). Marine sedimentary denitrification:€a model study. Global Biogeochemical Cycles, 10, 661–673. Mills, M. M., Ridame, C., Davey, M., La Roche, J. and Geider, R. J. (2004). Iron and phosphorus co-limit nitrogen fixation in the eastern tropical North Atlantic. Nature, 429, 292–294. Milovanovic, M. (2006). Water quality assessment and determination of pollution sources along the Axios/Vardar River, Southeastern Europe. Desalination, 213, 159–173. Moncheva, S., Doncheva, V. and Kamburska, L. (2001). On the long-term response of harmful algal blooms to the evolution of eutrophication off the Bulgarian Black Sea coast:€are the recent changes a sign of recovery of the ecosystem€– the uncertainties. In:€Harmful Algal Blooms 2000, ed. G. M. Hallegraeff, S.€I.€Blackburn, C. J. Bolch and R. J. Lewis, UNESCO, Paris, pp.€177–181. Montserrat Sala, M., Peters, F., Gasol, J. M. et€al. (2002). Seasonal and spatial variations in the nutrient limitation of bacterioplankton growth in the northwestern Mediterranean. Aquatic Microbial Ecology, 27, 47–56. Mulder, A., Van de Graaf, A. A., Robertsonm, L. A. and Kuenen, J. G. (1995). Anaerobic ammonium oxidation discovered in a denitrifying fluidized bed reactor. FEMS Microbiology Ecology, 16, 177–184. Munda, I. A. (1993). Changes and degradation of seaweed stands in the Northern Adriatic. Hydrobiologia, 260/261, 239–253. Munoz, I. and Prat, N. (1989). Effects of river regulation on the lower Ebro River, Northeast Spain. Regulated Rivers:€Research and Management, 3, 345–354. Nehring, D. and Matthäus, W. (1991). Current trends in hydrographic and chemical parameters and eutrophication in the Baltic Sea. Internationale Revue der gesamten Hydrobiologie, 76, 297–316. Némery, J. and Garnier, J. (2007). Dynamics of Particulate Phosphorus in the Seine estuary (France). Hydrobiologia, 588, 271–290. NSTF, Ducrotoy, J. P., Pawlak, J. and Wilson, S. (1994). The 1993 Quality Status Report of the North Sea. International Council for the Exploration of the Sea, Oslo and Paris. Nygaard, K. and Tobiesen, A. (1993). Bactivory in algae:€a survival strategy during nutrient limitation. Limnology and Oceanography, 38, 273–279. OAERRE (2010). www.lifesciences.napier.ac.uk/OAERRE/index.htm Officer, C. B. and Ryther, J. H. (1980). The possible importance of silicon in marine eutrophication. Marine Ecology Progress Series, 3, 383–391. Olenin, S. (1997). Benthic zonation of the Eastern Gotland Basin. Netherlands Journal of Aquatic Ecology, 30, 265–282. OSPAR, Pastch, J. and Radach, G. (2005). Long-term simulation of the eutrophication of the North Sea:€temporal development of

174

nutrients, chlorophyll and primary production in comparison to observations. Journal of Sea Research, 38, 275–310. Österblom, H. O., Hansson, S., Larsson, U. et€al. (2007). Humaninduced trophic cascades and ecological regime shifts in the Baltic Sea. Ecosystems, 10, 877–889. Pace, M. L., Cole, J. J., Carpenter, S. R. and Kitchell, J. F. (1999). Trophic cascades revealed indiverse ecosystems. Trends in Ecology and Evolution, 14, 483–488. Paerl, H. W. and Whitall, D. R. (1999). Anthropogenically-derived atmospheric nitrogen deposition, marine eutrophication and harmful algal bloom expansion:€is there a link? Ambio, 28, 307–311. Patricio, J., Ulanowicz, R., Pardal, M. A. and Marques, J. C. (2004). Ascendency as an ecological indicator:€a case study of estuarine pulse eutrophication. Estuarine, Coastal and Shelf Science, 60, 23–35. Paytan, A. and McLaughlin, K. (2007). The oceanic phosphorus cycle. Chemical Reviews, 107, 563–576. Pitkänen, H., Lehtoranta, J. and Räike, A. (2001). Internal nutrient fluxes counteract decreases in external load:€the case of the estuarial eastern Gulf of Finland, Baltic Sea. Ambio, 30, 195–201. Pont, D. (1996). Evaluation des charges polluantes du Rhône à la Méditerranée: Synthèse et Recommendations. Agence de l’Eau Rhône-Méditerranée-Corse, Lyon, France. Pranovi, F., Da Ponte, F. and Torricelli, P. (2008). Historical changes in the structure and functioning of the benthic community in the lagoon of Venice. Estuarine, Coastal and Shelf Science, 76, 753–764. Rabalais, N. (2002). Nitrogen in aquatic ecosystems. Ambio, 31, 102–112. Radach, G. (1992). Ecosystem functioning in the German Bight under continental nutrient inputs by rivers. Estuaries, 15, 477–496. Radach, G. and Pätsch, J. (1997). Climatological annual cycles of nutrients and chlorophyll in the North Sea. Journal of Sea Research, 38, 231–248. Radach, G. and Pätsch, J. (2007). Variability of continental riverine freshwater and nutrient inputs into the North Sea for the years 1977–2000 and its consequences for the assessment of eutrophication. Estuaries and Coasts, 30, 66–81, 10.1007/ BF02782968. Raes, F. et€al. (2000). Formation and cycling of aerosols in the global troposphere. Atmospheric Environment, 34, 4215–4240. Raffaelli, D. G., Emmerson, M., Solan, M., Biles, C. and Paterson,€D. (2003). Biodiversity and ecosystem processes in shallow coastal waters:€an experimental approach. Journal of Sea Research, 49, 133–141. Rahm, L. et€al. (2000). Nitrogen fixation in the Baltic proper:€an empirical study. Journal of Marine Systems, 25, 239–248. REMPEC (2008). Study of maritime traffic flows in the Mediterranean Sea. Rendell, A. R., Ottley, C. J., Jickells, T. D. and Harrison, R. M. (1993). The atmospheric input of nitrogen species to the North Sea. Tellus, 45B, 53–63. Riebesell, U. (2004). Effects of CO2 enrichment on marine phytoplanktons. Journal of Oceanography, 60, 719–729. Rönner, U. and Sörensen, F. (1985). Nitrogen transformation in the Baltic proper:€denitrification counteracts eutrophication. Ambio, 14, 134–138. Roubeix, V. and Lancelot, C. (2008). Effect of salinity on growth, cell size and silicification of an euryhaline freshwater diatom, Cyclotella meneghiniana, Kütz. Transitional Water Bulletin, 1, 31–38. Rousseau, V., Leynaert, A., Daoud, N. and Lancelot, C. (2002). Diatom succession, silicification and silicic acid availability in Belgian

Maren Voss coastal waters (Southern North Sea). Marine Ecology Progress Series, 236, 61–73. Rousseau, V., Breton, E., De Wachter, B. et€al. (2004). Identification of Belgian maritime zones affected by eutrophication (IZEUT). Towards the establishment of ecological criteria for the implementation of the OSPAR Common Procedure to combat eutrophication. Belgian Science Policy Publications, 77. Russell, K. M., Keene, W. C., Maben, J. R., Galloway, J. N. and Moody, J. L. (2003). Phase partitioning and dry deposition of atmospheric nitrogen at the mid-Atlantic U.S. coast. Journal of Geophysical Research, 108, 1–1–1–16, doi:10.1029/2003JD003736. Ruttenberg, K. C. (2003). The Global Phosphorus Cycle. Elsevier, New€York. Ruttenberg, K. C. and Berner, R. A. (1993). Authigenic apatite formation and burial in sediments from non-upwelling continental margins. Geochimica et Cosmochimica Acta, 57, 991–1007. Sambrotto, R. N., Savidge, G., Robinson, C. et€al. (1993). Elevated consumption of carbon relative to nitrogen in the surface ocean. Nature, 363, 248–250. Sandroni, V., Raimbault, P., Migon, C. and Garcia, N. E. G. (2007). Dry atmospheric deposition and diazotrophy as sources of new nitrogen to northwestern Mediterranean oligotrophic surface waters. Deep-Sea Research – Part I, 54, 1859–1870. Sanudo-Wilhelmy, S. A., Kusta, A. B., Gobler, C. J. et€al. (2001). Phosphorus limitation of nitrogen fixation by Trichodesmium in the central Atlantic ocean. Nature, 411, 55–59. Schartau, M., Engel, A., Schröter, J. et€al. (2007). Modelling carbon overconsumption and the formation of extracellular particulate organic carbon. Biogeosciences, 4, 433–454. Schenau, S. J. and de Lange, G. J. (2000). A novel chemical method to quantify fish debris in marine sediments. Limnology and Oceanography, 45, 963–971. Schmidt, I., Sliekers, A. O., Schmid, M. et€al. (2002). Aerobic and anaerobic ammonia oxidizing bacteria: competitors or natural partners? FEMS Microbiology Ecology, 39, 175–181. Schneider, B., Engel, A. and Schlitzer, R. (2004). Effects of depth- and CO2-dependent C:N ratios of particulate organic matter (POM) on the marine carbon cycle. Global Biogeochemical Cycles, 18, 1–13. Schneider, B. et€al. (2003). The surface water CO2 budget for the Baltic Proper:€a new way to determine nitrogen fixation. Journal of Marine Systems, 42, 53–64. Seinfeld, J. H. and Pandis, S. N. (1998). Atmospheric Chemistry and Physics:€from Air Pollution to Climate Change. Wiley-Interscience, New York. Seitzinger, S. P. and Sanders, R. W. (1997). Contribution of dissolved organic nitrogen from rivers to estuarine eutrophication. Marine Ecology Progress Series, 159, 1–12. Seitzinger, S. P. and Sanders, R. W. (1999). Atmospheric inputs of dissolved organic nitrogen stimulate estuarine bacteria and phytoplankton. Limnology and Oceanography, 44, 721–730. Seitzinger, S. P., Harrison, J. A., Dumont, E., Beusen, A. H. W. and Bouwman, A. F. (2005). Sources and delivery of carbon, nitrogen, and phosphorus to the coastal zone:€an overview of Global Nutrient Export from Watersheds (NEWS) models and their application. Global Biogeochemical Cycles, 19, 1–11. Seitzinger, S., Harrison, J. A., Böhlke, J. K. et€al. (2006). Denitrification across landscapes and waterscapes:€a synthesis. Ecological Applications, 16, 2064–2090. Shaffer, G. and Rönner, U. (1984). Denitrification in the Baltic Proper deep water. Deep-Sea Research – Part I, 31, 197–220.

Shiganova, T. A. and Bulgakova, Y. V. (2000). Effects of gelatinous plankton on Black Sea and Sea of Azov fish and their food resources. ICES Journal of Marine Science, 57, 641–648. Skliris, N. and Lascaratos, A. (2004). Impacts of the Nile River damming on the thermohaline circulation and water mass characteristics of the Mediterranean Sea. Journal of Marine Systems, 52, 121–143. Slomp, C. P., Epping, E. H. G., Helder, W. and van Raaphorst, W. (1996). A key role for iron-bound phosphorus in authigenic apatite formation in North Atlantic continental platform sediments. Journal of Marine Research, 54, 1179–1205. Slomp, C. P., Malschaert, J. F. P. and van Raaphorst, W. (1998). The role of sorption in sediment-water exchange of phosphate in North Sea continental margin sediments. Limnology and Oceanography, 43, 832–846. Slomp, C. P. and Van Cappellen, P. (2004). Nutrient inputs to the coastal ocean through submarine groundwater discharge:€controls and potential impact. Journal of Hydrology, 295, 64–86. Smith, S. V. and Hollibaugh, J. T. (1989). Carbon controlled nitrogen cycling in a marine “macrocosm”:€an ecosystem scale model for managing coastal eutrophication. Marine Ecology Progress Series, 52, 103–109. Souchu, P., Gasc, A., Collos, Y. et€al. (1998). Biogeochemical aspects of bottom anoxia in a Mediterranean lagoon (Thau, France). Marine Ecology Progress Series, 164, 135–146. Souvermezoglou, E. (1988). Comparaison de la distribution et du bilan d’échanges des sels nutritifs et du carbone inorganique en Méditerannée et en Mer Rouge. Oceanologica Acta, SI9 103–109. Souvermezoglou, E., Hatzigeorgiou, E., Pampidis, I. and Siapsali, K. (1992). Distribution and seasonal variability of nutrients and dissolved oxygen in the northeastern Ionian Sea. Oceanologica Acta, 15, 585–594. Spokes, L. J. and Jickells, T. D. (2005). Is the atmosphere really an important source of reactive nitrogen to coastal waters? Continental Shelf Research, 25, 2022–2035. Sterner, R. W., Andersen, T., Elser, J. J. et€al. (2008). Scale-dependent carbon :€nitrogen :€phosphorus seston stoichiometry in marine and freshwaters. Limnology and Oceanography, 53, 1169–1180. Stigebrandt, A. and Gustaffson, B. G. (2007). Improvement of Baltic Proper water quality using large-scale ecological engineering. Ambio, 36, 280–286. Sundbäck, K. and McGlathery, K. (2005). Interactions between benthic macroalgal and microalgal mats. In:€Interactions between Macro- and Microorganisms in Marine Sediments, ed. E.€Kristensen, R. R. Haese and J. E. Kostka, American Geophysical Union, Washington DC, pp. 7–29. Tamminen, T. and Andersen, T. (2007). Seasonal phytoplankton nutrient limitation patterns as revealed by bioassays over Baltic Sea gradients of salinity and eutrophication. Marine Ecology Progress Series, 340, 121–138. Tartari, G., Milan, C. and Felli, M. (1991). Idrochimica dei nutrienti. Quaderni Instituto di Ricerca Sulle Acque, 92, 1–29. Taylor, G. T., Iabichella, M., Ho, T. -Y. et€al. (2001). Chemoautotrophy in the redox transition zone of the Cariaco Basin:€a significant mid-water source of organic carbon production. Limnology and Oceanography, 46, 148–163. Tett, P., Gilpin, L., Svendsen, H. et€al. (2003). Eutrophication and some European waters of restricted exchange. Continental Shelf Research, 23, 1635–1671. Thamdrup, B. and Dalsgaard, T. (2002). Production of N2 through anaerobic ammonium oxidation coupled to nitrate reduction

175

Nitrogen processes in coastal and marine ecosystems in marine sediments. Applied and Environmental Microbiology, 68,€1312–1318. Thomas, H. and Schneider, B. (1999). The seasonal cycle of carbon dioxide in Baltic Sea surface waters. Journal of Marine Systems, 22,€53–67. Thompson, R. C., Crowe, T. P. and Hawkins, S. (2002). Rocky intertidal communities:€past environmental changes, present status and predictions for the next 25 years. Environmental Conservation, 29, 168–191. Toggweiler, J. R. (1993). Carbon overconsumption. Nature, 363, 210–211. Tuominen, L., Heinänen, A., Kuparinen, J. and Nielsen, L. P. (1998). Spatial and temporal variability of denitrification in the sediments of the northern Baltic Proper. Marine Ecology Progress Series, 172, 13–24. Turner, R. E., Qureshi, N. A., Rabalais, N. N. et€al. (1998). Fluctuating silicate:nitrate ratios and coastal plankton food webs. Proceedings of the National Academy of Sciences of the USA, 95, 13048–13050. Turner, R. E. and Rabalais, N. N. (1994). Evidence for coastal eutrophication near the Mississippi River Delta. Nature, 368, 619–621. Turner, R. E., Rabalais, N. N., Justic, D. and Dortch, Q. (2003). Global patterns of dissolved N, P and Si in large rivers. Biogeochemistry, 64, 297–317. Twomey, L. J., Piehler, M. F. and Paerl, H. W. (2005). Phytoplankton uptake of ammonium, nitrate and urea in the Neuse River estuary, NC, USA. Hydrobiologia, 533, 123–134. Tyrrell, T. (1999). The relative influences of nitrogen and phosphorus on oceanic primary production. Nature, 400, 525–531. UNEP-MAP (2003). Riverine Transport of Water, Sediments and Pollutants to the Mediterranean Sea, MAP Technical Series no. 141. UNEP-MAP/RAC/CP (2004). Guidelines for the Application of Best Environmental Practices (BEPs) for the Rational Use of Fertilizers and the Reduction of Nutrient Loss from Agriculture for the Mediterranean Region, MAP Technical Series no. 143. UNEP/FAO/WHO (1996). Assessment of the state of eutrophication in the Mediterranean Sea, MAP Technical Series no. 106. European Union (1991). Nitrates Directive. Directive 91/676/EEC. Usher, C. R., Michel, A. E. and Grassian, V. H. (2003). Reactions on mineral dust. Chemical Reviews, 103, 4883–4939. Vahtera, E., Conley, D. J., Gustafsson, B. G. et€al. (2007). Internal ecosystem feedbacks enhance nitrogen-fixing cyanobacteria blooms and complicate management in the Baltic Sea. Ambio, 36,€186–194. Vaquer, R. C. D. (2008). Thresholds of hypoxia for marine biodiversity. Proceedings of the National Academy of Sciences of the USA, 105, 15452–15457.

176

Veldhuis, M. J. W. and Admiraal, W. (1987). Influence of phosphate depletion on the growth and colony formation of Phaeocystis pouchetii. Marine Biology, 95, 47–54. Veldhuis, M. J. W., Colijn, F. and Admiraal, W. (1991). Phosphate utilization in Phaeocystis pouchetii (Haptophyceae). Marine Ecology Progress Series, 12, 53–62. Vitousek, P. M., Mooney, H. A., Lubchenko, J. and Melillo, J. M. (1997). Human domination of Earth’s ecosystems. Science, 277, 494–499. Vogt, H. and Schramm, W. (1991). Conspicuous decline of fucus in Kiel Bay (Western Baltic):€what are the causes? Marine Ecology Progress Series, 69, 1105–1118. Voss, M., Emeis, K. C. Hille, S., Neumann, T. and Dippner, J.€W. (2005). Nitrogen cycle of the Baltic Sea from an isotopic perspective. Global Biogeochemical Cycles, 19, 1–16. Waldbusser, G. G. and Marinelli, R. L. (2006). Macrofaunal modification of porewater advection:€role of species function, species interaction, and kinetics. Marine Ecology Progress Series, 311, 217–231. Walsh, J. J. (1991). Importance of continental margins in the marine biogeochemical cycling of carbon and nitrogen. Nature, 350, 53–55. Ward, B. B., Devol, A. H., Rich, J. J. et€al. (2009). Denitrification as the dominant nitrogen loss process in the Arabian Sea, Nature, 461, 78–82. Wasmund, N., Voss, M. and Lochte, K. (2001). Evidence of nitrogen fixation by non-heterocystous cyanobacteria in the Baltic Sea and re-calculation of a budget of nitrogen fixation. Marine Ecology Progress Series, 214, 1–14. Weisse, T., Tande, K., Verity, P., Hansen, F. and Gieskes, W. W. C. (1994). The trophic significance of Phaeocystis blooms. Journal of Marine Systems, 5, 67–79. Westberry, T. K. and Siegel, D. A. (2006). Spatial and temporal distribution of Trichodesmium blooms in the world’s oceans. Global BiogeochemIcal Cycles, 20. Wu, R. S. S. (2002). Hypoxia:€from molecular responses to ecosystem responses. Marine Pollution Bulletin, 45, 35–45. Yilmaz, A. and Tugrul, S. (1998). The effect of cold- and warm-core eddies on the distribution and stoichiometry of dissolved nutrients in the northeastern Mediterranean. Journal of Marine Systems, 16, 253–268. Zaitsev, Y. and Mamaev, V. O. (1997). Biological diversity in the Black Sea:€a study of change and decline. Black Sea Environmental Series, 3, 208. Zhang, Y., Zheng, L., Liu, X. et€al. (2008). Evidence for organic N deposition and its anthropogenic sources in China. Atmospheric Environment, 42, 1035–1041.

Chapter

9

Nitrogen processes in the atmosphere Lead author: Ole Hertel Contributing authors: Stefan Reis, Carsten Ambelas Skjøth, Albert Bleeker, Roy Harrison, John Neil Cape, David Fowler, Ute Skiba, David Simpson, Tim Jickells, Alex Baker, Markku Kulmala, Steen Gyldenkærne, Lise Lotte Sørensen and Jan Willem Erisman

Executive summary Nature of the problem • The two main groups of atmospheric reactive nitrogen compounds (reduced and oxidized nitrogen) have different fates due to differences in governing processes. • Abatement strategies need to take into account these differences when assessing the impact on the sensitive ecosystems.

Approaches • The chapter outlines the governing physical and chemical processes for the two main groups of reactive nitrogen compounds. • The chapter is divided into sections concerning:€emissions, transformation, aerosol processes, dry deposition and wet deposition.

Key findings/state of knowledge • Reactive nitrogen compounds consist of reduced nitrogen (ammonia and its reaction product ammonium), oxidized nitrogen (nitrogen oxides) and organic nitrogen compounds. • Nitrogen oxides have little impact close to the sources since they are emitted as nitrogen monoxide and nitrogen dioxide with low dry deposition rates. These compounds need to be converted into nitric acid (about 5% per hour) before deposition is efficient. • Ammonia has a high impact near the sources due to high dry deposition rates. Ammonia may therefore have significant impact on ecosystems in areas with intense agricultural activity leading to high emissions of ammonia. • Both ammonia and gaseous nitrogen oxides lead to formation of aerosol phase compounds (ammonium and nitrate) which are transported over long distances (up to more than 1000 km). • Very little is known either quantitatively or qualitatively about organic nitrogen compounds, other than that they can contribute a significant fraction of wet-deposited N, and are present in gaseous and particulate forms in the atmosphere.

Major uncertainties/challenges • Ambient air concentrations of reactive nitrogen compounds are fairly well (often within +/− 20%–30%) reproduced by state-of-the-art models, but estimates of deposition are most more uncertain (often more than +/− 50%). • Bidirectional fluxes of reactive nitrogen are still not well understood. • Sources and forms of organic nitrogen deposition are largely unknown.

Recommendations • There is a significant need for studies of fluxes of reactive nitrogen compounds over sensitive ecosystems. These studies need to include detailed field studies, parameterization, application and testing of chemistry-transport models.

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

177

Nitrogen processes in the atmosphere

9.1╇ Introduction

9.2.1╇ Ammonia emission handling

The Nr compounds take part in a series of chemical and physical processes in the atmosphere. Figure€9.1 outlines the main pathways of Nr compounds in the atmosphere. In this context N2O is disregarded, as it is mainly of importance in the stratosphere, and this compound is therefore only scarcely treated in the following. The left side of the figure illustrates path ways of NHx whereas path ways of NOy are illustrated to the right. However, the full picture is more complex than shown in the simple sketch. The aim of this chapter is thus to outline state-of-the-art of what is known to govern turnover processes for Nr compounds in the troposphere. The chapter will therefore describe anthropogenic and biogenic N emissions (Section€9.2), transformations of gas phase N compounds in the atmosphere (Section€ 9.3), aerosol processes involving N compounds (Section€9.4), as well as dry (Section€9.5) and wet (Section€ 9.6) deposition to both terrestrial and marine ecosystems of both gas phase and aerosol phase N compounds. Atmospheric transport and dispersion of Nr compounds is handled in Simpson et€ al. 2011 (Chapter 14, this volume), which also deals with the assessment of atmospheric N deposition.

9.2╇ Emissions of N compounds to the atmosphere

Agricultural activities are the dominating sources of NH3 emissions (Bouwman et€al., 1997). In western countries these activities contribute 85%–100% of the atmospheric releases (Anderson et€al., 2003; Sutton et€al., 2000), and e.g. in Denmark 98% of the anthropogenic NH3 emissions are related to agricultural production with the remaining 2% being emitted from petrol cars with catalytic converters (Gyldenkærne et€al., 2005; Mikkelsen et€al., 2005). NH3 emission from animal waste is a physical process Â�taking place from wet surfaces (Elzing and Monteny, 1997). The process is highly temperature dependent and varies significantly over day and season (Gyldenkærne et€ al., 2005; Skjøth et€ al., 2004). The regional variation is a function of production methods and agricultural practice, whereas temporal variations depend mainly on meteorology (but equally on agricultural practice). Organic bound N in the manure is not a direct source of NH3. The emission strength is therefore mainly related to the manure or fertilizers content of TAN (Total Ammoniacal N = NH3 + NH4+), pH, temperature, and wind speed. The agricultural sources of NH3 may be grouped as: • point sources, i.e. animal houses and manure storages, • application of manure and mineral fertilizer to the fields, • grazing animals, • other sources including plants.

The Nr released to the atmosphere consists of three parts:€NH3, NOx, and N2O, respectively, with smaller (and poorly quantified) contributions from organic compounds such as amines.

These categories are in the following described in detail with focus on a parameterization of spatial and temporal variation developed for the Danish area (Gyldenkærne et€al., 2005), and

Reaction acid gases & aerosols

Uptake aerosols & reaction NH3 Atmospheric aerosol NO3–

Atmospheric aerosol NH4+

Transport & Dispersion Atmospheric gaseous NHO3

Transport & Dispersion

NO2 reaction OH radical Atmospheric gaseous NH3

Atmospheric gaseous NOx (NO & NO2) NO2 on vegetation

Transport & Dispersion Emission NH3 mainly agricultural

Transport & Dispersion

Dry

Wet Ground

Dry

Transport & Dispersion Emission NOx (NO & NO2) traffic, industry & power production

Figure 9.1 Illustration of the path ways of reactive nitrogen (Nr) in the atmosphere. The left side of the figure illustrates the atmospheric path ways of NHx (gas phase ammonia (NH3) and aerosol phase ammonium (NH4+)) compounds, and the right side of the figure illustrates path ways of the NOy (NOx and reaction products) compounds. NH3 is emitted mainly from agricultural sources. In the atmosphere it is subject to transport and dispersion, but also scavenging by dry deposition and by transformation to aerosol bound NH4+ in reactions with acid gases and aerosols. Aerosol bound NH4+ has generally a long lifetime in the atmosphere and may therefore be transported over long distances (>1000 km). The NH4+ containing aerosols are mainly removed by wet deposition. Nitrogen oxides (NOy) are emitted to the atmosphere as NOx (the sum of nitrogen monoxide (NO) and nitrogen dioxide (NO2)), where also these compounds are subject to transport and dispersion. NO2 may be dry deposited to the vegetation, but it is mainly scavenged from the atmosphere by reaction with OH radical in the formation of nitric acid (HNO3). HNO3 has a very short lifetime in the atmosphere, since it is quickly scavenged by uptake in aerosols, reaction with NH3 or by dry deposition (it sticks to any surface€– aerosol as well as ground). Uptake in aerosols or formation of new aerosols by the reaction with NH3 leads to aerosol bound nitrate (NO3−). Also NO3− containing aerosols (in some cases the same aerosols that contain NH4+) are mainly scavenged by wet deposition. Aerosol phase NH4+ and NO3– may under certain circumstances depending on humidity and temperature€– be released back to gas phase NH3 and HNO3. N2O is excluded in the figure, as it does not play an important role in the atmospheric N deposition. For simplicity, reservoir compounds like HONO, HO2NO2, PAN and PAN-like species have also been disregarded from the sketch. Source:€Hertel et€al. (2006).

178

Ole Hertel

(a)

Animal houses and manure storages

Application of manure and mineral fertilizer The emission of NH3 from field application of manure and mineral fertilizer takes place at distinct times of year with

25 20 15

–15

–5

5

15

25

35

25

35

Temperature, outdoor [°C] 35 Temperature, stables [°C]

(b)

30 25 20 15 10 5 0 –25

–15

–5

5

15

Temperature, outdoor [°C] (c)

0.45

Ventilation, stable [m/s]

0.89 0. 26 (9.1) ╅╇ E (t ) = C × T (t ) xV (t ) , where C is a constant which can be related to the amount of N or TAN in the manure at a given time and location, T(t) is the temperature as function of time, and V is the wind speed or the rate of ventilation. This formula may help to distribute a known annual emission into smaller time step ν. Concerning the differences between animal types, pig and poultry have a critical temperature that is relatively high compared with cattle. This means that in Northern Europe stables with pigs and poultry are heated during winter. The magnitude of the overall annual emission from a given manure storage reflects the type of storage, whereas the temporal variation in the emission reflects variations in ambient air temperature (Gyldenkærne et€ al., 2005). In warm areas and during warm periods of time, the emissions from all buildings will reflect outdoor temperatures. Buildings containing pigs and poultry thus have significant emissions also during cold periods, when cattle barns and storages have low emission rates. Applying the parameterization in equation (9.1) and the simplified functions shown in Figure€ 9.2, the temporal variation in NH3 emissions may be simulated. Figure€9.3 shows as an example of calculated temporal variation in NH3 emission from a pig stable, cattle barn and manure storage.

30

10 –25

0.4 0.35 0.3 0.25 0.2 0.15 0.1 –30

–20

–10

0

10

20

30

40

Temperature, outdoor [°C] (d)

0.23 0.225

Ventilation [m/s]

Significant variations in NH3 emissions are found for different types of animals and housings (Koerkamp et€ al., 1998). The variations are related to amount of TAN in the manure, Â�stable temperature, and ventilation rate. Highly complex surface models describe the NH3 emissions from agricultural buildings (Muck and Steenhuis, 1982; Olesen and Sommer, 1993; Oudendag and Luesink, 1998; Zhang et€al., 1994), but are not applicable for large scale modelling (Gyldenkærne et€al., 2005; Pinder et€ al., 2004). A simplified parameterization based on wind speed and ambient air temperature is in this case a more practical approach. Most inventories are using total N or TAN content in the manure together with fixed average emission factors (EF) for different types of animals, housing systems, and manure applications. The EF should be location specific and account for local conditions, and such data are currently being processed for the entire European area. At large scales, it has been shown that the variation in emissions E(t) from stables and storages may be described by a simple parameterization (Gyldenkærne et€al., 2005; Skjøth et€al., 2004):

35 Temperature, stable [°C]

currently considered as the best and most advanced (Pinder et€al., 2007). The parameterization has been implemented into the atmospheric transport models:€the ACDEP model (Skjøth et€al., 2004), the Danish Eulerian Hemispheric Model (DEHM), the Unified EMEP model (Fagerli et€ al., 2007), and the local scale model OML-DEP (Sommer et€al., 2009).

0.22 0.215 0.21 0.205 0.2 0.195

0

100

200

300

Day Figure€9.2 The indoor temperature as function of outdoor ambient air temperature for (a) isolated stables, (b) open barns. Ventilation rate for (c) isolated stables as function of ambient air temperatures, and (d) open barns as function of day of year.

relatively short duration compared with point source emissions. The application method is crucial for the magnitude of the emission. Very high emissions are related to the application

179

Nitrogen processes in the atmosphere

(a)

Temperature

Pig houses

Cattle barns

Storage

30

Temperature

25 20 15 10 5 0 –5 –10 1

27

53 79 105 131 157 183 209 235 261 287 313 339 365 Day

Emission, g N/dag*ha

(b)

Cows

Pigs

Storage

18 16 14 12 10 8 6 4 2 0 1

31

61

91

121 151 181 211 241 271 301 331 361 Day

Figure 9.3 Simulations for open and isolated stables and manure storages. The top figure shows the calculated daily mean temperature, and the lower figure shows the calculated daily emission at Tange, Denmark in the year 2000. Source:€Ellermann et€al. (2004).

of broad spread methods, whereas soil injection methods lead to very low emissions. National regulations may in some cases play a significant role with regards to the seasonal variation. In some countries manure application is constrained with almost no regulation, whereas in many North European countries manure application is abandoned during winter time. To overcome a shortage in manure storage capacity farmers often empty their slurry tanks in the autumn. This may give high emissions in the autumn. The timing of mineral fertilizers and the related NH3 differ also between regions giving different temporal emission patterns, where farmers in the Southern parts of Europe initiate application of fertilizer earlier than in Northern Europe. Figure€9.4 shows the temporal variation in NH3 emission having four typical application times during the year 2000 at Tange in Denmark.

Emission of NH3 from grazing animals The emission from grazing animals depends on time spent in the field and N content in the grass. The latter leads to excretion of large amounts of excessive N as TAN compared with more extensive grassland. However, the urine is easily entering the soil, which will lower the emission potential compared to surface applied slurry. In Southern Europe, animals are in the field most of the year. Sheep may remain in the field most of the time, whereas dairy cattle in many countries are inside

180

Figure 9.4 The emission strength and temporal variation in NH3 emission in the Tange area in Denmark related to application of manure during spring, summer and autumn (top) and mineral fertilizer during spring and summer (bottom). Source:€Ellermann et€al. (2004).

the stables approximately half of the year (see e.g. Regional Air Pollution INformation and Simulation model (RAINS) database (RAINS, 2010)). The number of grazing animals will in general follow the availability of grass, or the season of growth. There is little knowledge on the effect of temperature on the NH3 emission.

Emission of NH3 from other sources Legumes and plants taking up excess fertilizer are emitting NH3 (Larsson et€ al., 1998). The emission depends on the enrichment of the apoplast with NH4+, and the compensation point (Farquhar et€al., 1980) (see Section€9.5), which is a function of the plant status with respect to growth and stress, etc. The emission is still not well described with respect to magnitude, as well as temporal and spatial variation. Emissions from non-agricultural sources are in general not well described but include sweat from humans, exhaust from gasoline cars with catalytic converters, excreta from pets and wild animals, and evaporation from waste deposits (Sutton et€ al., 2000). In Europe, the largest national non-agricultural NH3 emission has been reported for the UK with a fraction of about 15% (Sutton et€al., 2000), compared with about 2% reported for Denmark (it has been disputed whether the Danish inventory accounts for natural NH3 emissions).

Ole Hertel

Figure€9.5 Annual NH3 emissions in Europe based on:€(top left) Edgar data from 1995 (1° × 1°) in kg NH3/grid cell, (top right) EMEP data from 2000 (50 km × 50 km) in Mg NH3/grid cell, and (bottom) NERI data based on EMEP and GENEMIS data (16.67 km × 16.67 km) in tonnes NH3/grid cell.

181

Nitrogen processes in the atmosphere Figure€9.6 The relative distribution between different NH3 sources in national emissions. Data derived for the year 2000 (Hertel et€al., 2006).

Spatial distribution in NH3 emissions At the European level, EMEP (Co-operative Programme for Monitoring and Evaluation of the Long-range Transmission of Air Pollutants in Europe (EMEP, 2010)) and CORINAIR (CORe INventory AIR emissions) compile inventories of the annual mean emissions on a grid with a spatial resolution of 50 km × 50 km. The EDGAR (Emissions Database for Global Atmospheric Research; EDGAR, 2010) and GEIA (Global Emissions Inventory Activity; GEIA, 2010) databases are available on 1°Â€ ×€ 1° resolutions, and the EUROTRAC (EUREKA project on the transport and chemical transformation of trace constituents in the troposphere over Europe) GENEMIS (Generation and Evaluation of Emission Data) project (Eurotrac, 2010) compiled inventories with a grid resolution of 16.67 km × 16.67 km. The GENEMIS data was for the year 1994, but has in some studies been used to redistribute EMEP emission inventories for following years, assuming unchanged relative distribution over the years (Hertel et€al., 2002; Spokes et€al., 2006). NH3 emission has evidently a non-uniform distribution throughout Europe (Figure€9.5). Figure€9.6 shows the relative national distribution in NH3 emission between the typical agricultural emission source categories.

Site specific long term trends in NH3 emissions In Denmark before 1989, the NH3 emission was relatively low during winter time as a result of low activity and low temperatures. The accumulated manure during winter was applied to crops in the fields during spring, but also to grass fields during summer. Finally the manure storages were emptied in autumn (Figure€9.7). This pattern is typical in Northern Europe with moderate to large agricultural activity and limited legislative control. In the 1990s Denmark implemented the until now strongest European regulation on local agricultural activities in order to reduce emissions to air, soil and water (Grant and Blicher-Mathiesen, 2004; Skjøth et€ al., 2008). This included improving the entire production chain with respect to

182

Figure€9.7 Temporal variation in daily NH3 emission [g N/ha/day] from different sources in Tange, Denmark for the years 1989 (top) and 2000 (bottom). Source: Skjøth et€al. (2008).

reducing NH3 emissions. The farmers had to increase the fraction of manure applied during growth of the crops in spring, and similarly decrease the fraction applied in summer and autumn. This is seen in Figure€9.8, where the emission peak in spring is more pronounced and emissions in summer and autumn reduced when comparing 1989 to 2000. The overall Danish release of NH3 was decreased significantly (Table€9.1) despite an increase in animal production over this period (Skjøth et€al., 2008).

Long term trends in NH3 emissions on European scale The NH3 emissions have been reduced in countries like Denmark, Germany and the Netherlands, whereas for

Ole Hertel Figure€9.8 Comparison of the trends in computed monthly mean NH3 emissions (pink) and observed monthly mean ambient air NH3 concentrations (blue) for the Danish monitoring station Tange during the time period 1989 to 2003 (Skjøth et€al., 2008). All values are relative to the annual mean in 1989.

Table€9.1 Annual emission of ammonia (ktonnes N) from selected European countries during the period 1985 to 2000

1985

1990

1995

2000

Denmark

113

109

92

83

Germany

706

630

523

514

Netherlands

204

187

153

126

France

642

628

624

649

Sweden

44

42

50

46

Norway

19

19

21

21

Finland

35

31

29

27

5500 5000

Gg NH3

4500 4000 3500 3000 2500 2000

1990

1995

2000

2005

Figure 9.9 Trend in NH3 emissions on European scale for the period 1990 to 2005, and further projected to 2010. Source:€EMEP (2010).

countries like France, Sweden and Norway only very small changes have occurred over the past 15 years (Table€ 9.1). Over the EMEP domain, emissions have decreased since 1990, and projections point at a further decrease in 2010 (Figure€9.9).

9.2.2╇ Nitrogen oxide emission handling Nitrogen oxides (NOx is the sum of nitrogen monoxide (NO) and nitrogen dioxide (NO2)) are generated at high temperatures

in combustion processes mainly from oxidation of free atmospheric nitrogen (N2).

European emission inventories EMEP€ – For Europe various inventories on different scales and compiled with different objectives are available. National inventories are aimed at fulfilling the obligations due to international agreements and conventions, e.g. the European Commissions’ National Emission Ceilings Directive (NECD) or the UNECE Convention on Long-range Transboundary Air Pollution (CLRTAP). NOx emissions are subject to the NECD and the NOx Protocol as well as the Gothenburg Protocol under CLRTAP. While the NECD covers only EU member states, the CLRTAP protocols cover the whole UNECE region (see Figure€9.10). However, not all countries have ratified all protocols and thus may not be obliged to report. These inventories focus on anthropogenic emissions and sources, but biogenic and natural emissions are reported by some countries. The officially reported emissions are often incomplete and subject to gaps, and a gap-filling procedures have thus to be applied when used, i.e. in modelling (see example in Figure€9.10). EDGAR€– In contrast to the officially reported compilations of national emissions, the EDGAR (EDGAR, 2010) is an example of an independent bottom-up inventory with global coverage for all main pollutants and greenhouse gases and compiled from expert estimates and available emission factors and activity data.

National inventories European inventories provide spatial resolution of e.g. 50€kmÂ€× 50 km (Figure€9.10) or 0.5° × 0.5°. For many European countries, however, data are available at 10 km × 10 km, 5 km × 5 km or even 1 km × 1 km (see Figure€9.11). Vehicle emissions can be computed using an emission module in the Danish Operational Street Pollution Model (OSPM, 2010). This module uses COPERT IV (see COPERT, 2010) emission factors and a road and traffic database. A GIS script compiles kilometres travelled as function of road type and vehicle category (Jensen et€al., 2008) on a 1 km × 1 km grid, and another programme (UrbEmi) calculates the road traffic emissions.

183

Nitrogen processes in the atmosphere Figure 9.10 European NOx emissions on the 50€km × 50 km EMEP grid for the year 2000 (EMEP, 2010).

Table€9.2 NOx sectoral contribution based on EDGAR data for Europe in the year 2000, including 37 countries (source:€F T32, EDGAR 2010)

Sector

Emissions in Gg

Share

Power generation

6721.75

25.2%

Industrial production

4092.07

15.4%

Residential & commercial combustion

1603.53

6.0%

Road transport

6085.19

22.9%

Other mobile sources

3646.38

13.7%

International shipping

3012.02

11.3%

613.80

2.3%

17.23

0.1%

829.81

3.1%

Air transport Waste incineration Biomass & agricultural burning Total

26 621.79

Data from:€EU27, Croatia, FYR of Macedonia, Turkey, Switzerland, Norway, Albania, Russian Federation, Bosnia & Herzegovina, Ukraine, Federal Republic of Yugoslavia

Key source analysis The source sectors are summarized in Table€ 9.2. The bulk of NOx emissions in Europe stem from both stationary and mobile combustion sources. Road transport and public power generation are by far the largest contributors to NOx emissions, followed by residential and commercial combustion and offroad sources. The NatAir project (NatAir, 2010) compiled emissions of NO from forest and agricultural soils at about 98.22 Gg for

184

the EU15 (Friedrich, 2002). Up-scaled to the entire European area, this is pointing at forest and agricultural NO emissions in the order of 200 to 300 Gg NO yr−1. Emissions from lightning activities calculated within NatAir show 325 Gg yr−1 for the EU15 region, which according to the EDGAR inventory is in the same order as biomass and agricultural burning emissions.

Past and future trends Trends for NOx emissions in the period 1990 to 2005 have steadily pointed downwards (Figure€ 9.12) due to legislation to reduce emissions from stationary sources (e.g. the EC Large Combustion Plants Directive) and mobile sources (mainly the EURO emission standards for vehicles), amounting to approx. 34% reduction of EU27 emissions. Further reductions are anticipated with legislation for road transport sources over the next few years. A review on European NOx emissions with emphasis on road transport is provided in Vestreng et€al. (2008). Commitments for phasing out nuclear energy may, however, lead to a replacement of nuclear by biomass or coal burning plants and thereby increase emissions from this sector. International shipping emissions are rising steadily and this increase is projected to continue, although the reductions from regulating quality of marine fuels and other abatement strategies are not accounted for in these projections. NOx to NO2 ratio€ – Ambient NO2 concentrations do not decrease at the same rate as NOx in European hotspots (Carslaw et€al., 2007; Lambrecht, 2007), mainly due to increasing NOx/NO2 ratio in late diesel technology vehicles. Hourly NO2 limit values are mandatory in Europe from 2010 (EC Daughter Directive 99/30/EC, and primary NO2 in vehicle

Ole Hertel

Figure 9.12 Sector trends in European NOx emissions 1880–2005 (unit TgNO2) (Vestreng et€al., 2008).

exhausts may thus need to be addressed in future emission inventories. Secondary emission control options of especially large stationary combustion units include selective catalytic reduction (SCR). A common problem with SCR systems is the release of unreacted NH3 referred to as ammonia slip. Slip can occur when catalyst temperatures are not in the optimal range for the reaction or when too much NH3 is injected. SCR systems are increasingly used also for mobile sources, specifically heavy duty trucks, where ammonia slip is an issue more difficult to optimize than for stationary combustion units.

Ecosystems and soil NO and N2O emissions

Figure 9.11 National high resolution maps of NOx emissions, displaying (a) UK National Atmospheric Emission Inventory data (Source: UK NAEI: http:// www.naei.org.uk/mapping/mapping-2008.php; Crown copyright) and (b) spatial distribution of NOx road emissions in Denmark in 2004 on a 1â•›×â•›1 km2 grid.

Globally, agricultural soils represent the largest soil source of N2O (54%), followed by tropical wet forests (21%) and temperate forests, savannahs and natural grasslands contribution at almost equal parts to the remaining 25%. The impact of N fertilizer on N2O emissions in temperate climates is well studied, whereas less is known about agricultural systems in tropical countries or tropical wet forest systems. Ongoing measurements in tropical forest and agricultural systems in Borneo suggest much larger emissions in wet tropical countries than in temperate regions (Nemitz and Skiba, personal communication). Of the agricultural soils intensively managed grasslands in the UK, Netherlands and other temperate climates are hotspots of N2O emissions, because (a) grasslands receive larger rates of mineral fertilizer and manure, (b) grasslands occur in high rainfall regions, and (c) grazed grasslands have compacted soils. Such conditions all favour NO production. NOx emission from soils may represent >40% of the emissions at the global scale (Davidson and Kingerlee, 1997; IGAC, 2000), and >10% for some European countries (Butterbach-Bahl et€al., 2004; Skiba et€al., 1997; Stohl et€al., 1996). Emissions resulting from fertilizer use could represent 40% of soil emissions at global scale (IGAC, 2000) and up to 65% for the USA (Hall et€al., 1996). Rural agricultural areas receiving N fertilizers in countries with long dry periods are likely the largest sources of soil NO. The NitroEurope Integrated Project (NEU, 2010) and the NOFRETETE (Nitrogen oxides emissions from European Forest Ecosystems) project point at Europe forests as large sources of NO (Pilegaard et€ al., 2006). The coniferous forest at Höglwald,

185

Nitrogen processes in the atmosphere

DE, receiving high atmospheric N deposition is a large source of NO, whereas the boreal forest at Hyytiälä, FI, and moorland sites in FI and UK have very small emissions (Skiba et€al., 1997). In semi-natural/Â�natural ecosystems that do not receive N from fertilization or grazing, atmospheric N deposition significantly affect NO and N2O emissions. The NOFRETETE project showed correlation between N deposition rates and NO emissions from coniferous forest soils (Pilegaard et€ al., 2006). Along a wet deposition gradient in Cumbria (17–40 kg N ha−1 yr−1), a linear relationship was observed between wet deposition of N, KCl extractable NH4+ and NO3− and NO and N2O emissions from semi-natural grassland on peat (Skiba et€al., 2007).

9.3╇ Transformation of N compounds in the€atmosphere The following section provides a description of atmospheric transformation processes of Nr and highlights where these processes play a significant role.

9.3.1╇ Reactions between NH3 and acid gases and€aerosols In the reactions between gas phase NH3 and gas phase acids, new aerosol particles are formed. However, NH3 may also condense onto existing atmospheric particles. Gaseous NH3 will practically always react with sulphuric acid (H2SO4) in gas or aerosol phase, if H2SO4 is present. H2SO4 is formed from gas phase oxidation of SO2 by hydroxyl (OH) radical or from aerosol phase conversion by hydrogen peroxide (H2O2) and ozone (O3). The later process is pH dependent, and may in fact be catalysed by NH3, since uptake of NH3 increases the pH of aerosols (Apsimon et€al., 1994; Junge and Ryan, 1958). The reaction between NH3 and H2SO4 is usually considered irreversible. In traditional CTM, it is thus common to describe the reaction as irreversible and taking place in two steps forming ammonium bisulphate (NH4)HSO4 and ammonium sulphate (NH4)2SO4 (Hov et€al., 1994), respectively:

NH3 + H2 SO4 → NH4 HSO 4

(9.2)



NH3 + NH4 HSO4 → (NH4 )2 SO4 .

(9.3)

The rate of reaction between NH3 and H2SO4 has been analysed in detail in laboratory studies (Baldwin and Golden, 1979; Gupta et€al., 1995; Huntzicker et€al., 1980; McMurry et€al., 1983). At high RH, the limiting factor for the transformation is the molecular diffusion of NH3 to the acid particles, whereas at low RH only between 10% and 40% of the collisions between NH3 gas molecules and H2SO4-containing particles lead to reaction (Huntzicker et€ al., 1980; McMurry et€ al., 1983). For small particles, the relatively large surface area makes the diffusion process more efficient. Organic material on the surface of the particles may, however, limit the uptake of NH3 (Daumer et€al., 1992).

186

Whereas the NH3 reaction with H2SO4 may be considered irreversible, this is not the case for the reactions with other acid gaseous compounds. Presence of nitric acid (HNO3) and/or hydrochloric acid (HCl) together with NH3 lead to equilibrium between these gases and their aerosol phase reaction products€– the ammonium salts:€ammonium nitrate (NH4NO3) and ammonium chloride (NH4Cl). For the reaction with HNO3 this may be expressed as:

NH3 + HNO3 ↔ NH4 NO3 .

(9.4)

Experimental studies show that to a good approximation an equilibrium product, keq = [NH3][HNO3], of the gas phase concentrations of NH3 and HNO3 at saturation of the air, may be expressed by a function depending solely on temperature and humidity (Stelson et€al., 1979; Stelson and Seinfeld, 1982). The RH at the point of deliquescence RHd [%] = 856.23/T + 1.2306, and ln( K ) = ln(K eq ) = 0.78 −

ln( K ) = ln( K eq ) −

24, 220 ln(T ) − 6.1 RH < RH d T 298

(9.5)

20, 75 + ln(K eq )  RH − RH d    RH ≥ RH d 101 − RH  100 − RH d  (9.6)

where T is in K. Besides the reactions with H2SO4 and HNO3, NH3 may also take part with HCl and form NH4Cl (Pio and Harrison, 1987a):

HNO3 + NaCl → NaNO3 + HCl

(9.7)



NH3 + HCl ↔ NH4 Cl.

(9.8)

Whereas HCl is a primary pollutant emitted by coal burning and incineration, HNO3 is the main secondary pollutant from oxidation of NOx emissions (see Section€ 9.3.3). New measurement data indicate that in NW Europe, HCl concentrations are similar to those of HNO3 in summertime, in terms of mixing ratio. However, NH4Cl concentrations are much lower than NH4NO3 concentrations. HCl is emitted from anthropogenic sources, but it is also released in displacement reaction in sea spray particles when these take up HNO3 (Wall et€al., 1988): ╇╇╇ NaCl + HNO3 → NaNO3 + HCl.



(9.9)

In the first EMEP model this reaction was accounted for by a first order decay of HNO3 of 10–5 s−1, and a reverse reaction rate coefficient of half this size (Hov et€al., 1994). Measurements in California showed that NO3− in the coarseparticle mode is Â�primarily associated with high Na+ levels in marine air (Wall et€ al., 1988), and that NO3− in course fraction particles has a peak at 3 μm diameter, where the product of the Na and mass distribution also peaks. This displacement reaction is thus the most likely explanation for

Ole Hertel

HCl concentrations of up to 250€ pptv (Harris et€ al., 1992) observed in the marine boundary layer. Experimental studies have determined an equilibrium product at saturation of the air with the two gases NH3 and HCl (Pio and Harrison, 1987b). See also Figure€9.13. NH4NO3 and NH4Cl are semi-volatile and the salts are deliquescent under most tropospheric conditions in northern Europe and may dissolve in pre-existing aerosol droplets or absorb onto the surface of any pre-existing aerosol particles. Thereby NO3−, Cl− and NH4+ are incorporated in suspended particulate matter in the particle size range, mainly in the submicron accumulation mode size-range, and therefore contribute to PM2.5 and PM10, the metrics used for human health assessment (Moldanova et€ al., 2011, Chapter€ 18 this volume). It may be reasonable to assume equilibrium of NH3 and HNO3, and NH3 and HCl. However, observations of particle size distribution of inorganic N, S, and Cl species in maritime air over the North Sea show products of partial pressures of [NH3][HNO3] and [NH3][HCl] that often fall below the theoretical lines of equilibrium (Ottley and Harrison, 1992), and one should therefore be careful when applying the assumption of equilibrium. This is due to sources and sinks, but mainly because the theoretical lines are for pure salts, while the co-existence of sulphate in the aerosol can dramatically decrease the equilibrium vapour concentration product.

9.3.2╇ Changed NH3 to NH4+ conversion rate due to changes in S emissions Early experiments carried out by Mckay (1971) showed that 50% of the available NH3 is converted into ammonium sulphate in about 35 minutes, based on concentrations present in the atmosphere at that time (20 µg m−3 SO2 and 2.7 µg m−3 NH3). Models like the EMEP Unified Model assume an instantaneous and irreversible formation of ammonium sulphate, only limited by the availability of the least abundant of NH3 and SO42−. Any excess NH3 may then react with HNO3, forming NH4NO3. Over the last decades, a dramatic decrease in SO2 emissions occurred (Figure€9.14). Especially in the East European countries, the SO2 emissions dropped by approximately 60% in the late 1980s/early 1990s. This drop in SO2 emissions and resulting ambient concentrations has affected the formation of (NH4)2SO4. Measurements show that in the Netherlands in the early 1980s the NH3/NH4+ conversion rate was 28.8% per hour, while at present it is about 5% per hour (Van Jaarsveld, 2004). Trends in observations and EMEP model results for wet deposited N are in compliance with trends in emissions (Fagerli and Aas, 2008). For air concentrations less information is available, since most of EMEP sites did not start measuring TIA (Total Inorganic NHx = gas phase NH3 + aerosol phase NH4+) and TIN (Total Inorganic NO3− = HNO3 + aerosol phase NO3−) until the end of the 1980s and only a few

2007 Figure 9.13 HCl concentrations 2006 (left) and 2007 (right) based on 30 sites with monthly monitoring. Source:€Pollutant deposition (CEH, 2010).

187

Nitrogen processes in the atmosphere

sites (~20) have reported continuously. Moreover, the gas and particulate phases have very different chemical (e.g. their role in the NH4+–NH3–HNO3–NO3–SO42− equilibriums) and physical properties (e.g. the aerosols have a much longer residence time and are transported over longer distances), and the trend in the gas and particulate phase may thus be different. In some Eastern European countries NH3 emissions have declined by 30%–60%, but NHx concentrations decreased only by 20%–30% (EMEP, 2010). In Germany, however, NHx concentrations declined by 20%–30%, whilst emission reductions are reported to have been 10%–20%. The explanation is a combination of a less efficient formation of NH4+ aerosol (due to decreasing SOx) and less efficient dry deposition of NH3 due to less acidic surfaces; both effects leading to a shift towards gaseous NH3 relative to particulate NH4+.

9.3.3╇ NOy chemistry in the troposphere

The emission of NOx takes place mainly in the form of nitrogen monoxide (NO) and to a lesser extent (usually 5%–20%) as nitrogen dioxide (NO2). The fraction of directly emitted NO2 from road traffic in western countries has increased in recent years as a result of the use of catalytic converters. However, in the tropospheric boundary layer the distribution between NO and NO2 is governed to a large degree by O3 that reacts very fast with NO to form NO2. In sunlight NO2 photo dissociates (wavelengths 200–420 nm) to form NO and the very shortlived oxygen (O(3P)) radical. The latter will in most cases again form O3 in a reaction with free oxygen (O2).

NO + O3 → NO2 + O2



(9.10)



NO2 + hv → NO + O( 3 P)

(9.11)

O( 3 P) + O2 + M → O3 + M



In the above reactions M is a third body (either an N2 or O2 molecule) that absorbs excess vibrational energy and thereby stabilizes the formed O3 molecule. These reactions all have time scales of seconds to minutes. The reaction rate of reaction (9.9) is temperature dependent, but has a typical value about 4 × 10–4 ppbv−1 s−1. Under typical atmospheric boundary layer conditions, reaction (9.9) will either lead to the complete conversion of all the O3 to NO2, or to the conversion of all NO to NO2 (Clapp and Jenkin, 2001). In highly polluted atmosphere (e.g. an urban area) or close to pollution sources, the former behaviour is usually observed because although O3 is widely distributed in the lower atmosphere, its concentration is not usually high compared with NO in the highly polluted atmosphere, and hence O3 concentrations become rapidly depleted. During daylight, the main fate of NO2 is to undergo photolysis (9.10), reforming O3 (9.11) and NO (Dickerson et€al., 1982). This reaction has a typical rate coefficient under summer Â�conditions in the mid afternoon at mid altitudes of about 7 × 10–3 s−1. Reaction (9.12) is the only production path for O3 in the atmosphere. Figure€9.15 illustrates the NO–NO2–O3 chemistry in urban streets using a highly simplified module (Palmgren et€al., 1996) in the Operational Street Pollution Model (OSPM) (Berkowicz, 2000). This module includes the reactions (9.10) and (9.11) and an assumption of reaction (9.12) being instantaneous. In adÂ�dition the model includes a ventilation rate between the street canyon and the surrounding air, and a distribution of the direct emission of NO and NO2 from street traffic (Palmgren et€al., 1996).

80-03 80-98

–40

–20

0

20

Change in emission (in %)

188

40

(9.12)

60

80

Figure 9.14 EMEP emission changes for different European countries for two periods:€1980–1998 and 1980–2003. Source:€EMEP (2010).

Ole Hertel

Whenever NO is present, the most important atmospheric reaction of the hydroperoxy radical (HO2) is the conversion of NO to NO2:

NO + HO2 → NO2 + OH.

(9.16)

The hydroperoxy radical is one of many peroxy radicals that take part in the conversion of NO to NO2. Organic peroxy radicals (RO2) play likewise an important role and are mainly formed by the attack of the OH radical on the organic compounds ubiquitously present in the polluted atmosphere. These reactions follow a similar path as the CO oxidation, and may in a simplified form be presented as:

OH + RXH → R + H2 O

(9.17)



R + O 2 + M → RO2 + M.

(9.18)

RXH represents the organic compound, whereas R is an organic radical such as the alkyl radical and RO2 an alkyl peroxy radical. The only important atmospheric pathway of the alkyl radical is reaction (9.18) with O2 to form alkyl peroxy radicals (Finlayson-Pitts and Pitts, 1986). The RO2 radical may subsequently covert NO to NO2 in the same way as the HO2 radical (reaction (9.16)). During combustion processes at high temperatures, e.g. inside the motor of a petrol or diesel-driven vehicle, NO is formed from ambient N2. However, in the very NO rich air inside the exhaust pipe of vehicles and inside emitting chimneys, another oxidation path than (9.11) takes place:

Figure 9.15 The chemistry of NOx in urban streets. (Top) the observed relationship between NO2 and NOx. For NOx concentrations below about 20 ppb, all NOx is in the form of NO2, since the air contains sufficient O3 for converting all NO to NO2. For higher NOx concentrations, only the direct emission of NO2 contributes to further increase in NO2 concentrations. (Bottom) Comparison between observed and calculated hourly mean concentrations of NO2. All data are from Jagtvej in 2003, and calculations performed with OSPM. Only working days during daytime (800–1600) are included. Correlation coefficient (R2) = 0.7 (Hertel and Brandt, 2009).

The OH radical initiates the oxidation of a wide range of compounds in the atmospheric boundary layer. OH interacts with peroxy radicals that are responsible for the formation of excess concentrations of photo oxidants like O3. In the background troposphere, carbon monoxide (CO) plays a role in this system:

CO + OH → CO2 + H

(9.13)



H + O2 + M → HO2 + M.

(9.14)

This conversion is so rapid that for all practical purposes reaction (9.13) may actually be written as:

2 CO + OH + M O → CO2 + HO2 + M.

(9.15)

NO + NO + O2 → 2NO2 .

(9.19)

This reaction is a third order reaction with a second order dependence of the NO concentration, and has a reaction rate coefficient of 2.3 × 10–38 cm6 molecules−2 s−1 (Hampson and Gavin, 1978). The further transformation of NO2 to HNO3 takes place with a typical rate of about 5% per hour in the troposphere:

NO2 + OH → HNO3.

(9.20)

The hydroxyl radical (OH) is formed in the daytime in the presence of sunlight (Finlayson-Pitts and Pitts, 1986). The photo dissociation of O3 leads to the formation of both O(3P) and O(1D) radicals, a fraction of the latter reacts with water vapour to form two OH radicals:

O3 + hv → O2 + O( 3 P)

(9.21)



O3 + hv → O2 + O(1 D)

(9.22)



O( 1 D) + H2 O → 2OH.

(9.23)

Reaction (9.17) takes place in competition with O(1D)’s reaction with third body O2 or N2 molecules to form O(3P), that in turn reform O3 in reaction (9.12). The OH radicals initiate

189

Nitrogen processes in the atmosphere

most of the degradation of hydrocarbons in the atmosphere, a chain of reactions that e.g. lead to the formation of high O3 concentrations during summer. During night-time the nitrate (NO3) radical has a less important but still somewhat similar role for the degradation of hydrocarbons in the atmosphere as the OH radical in daytime. Despite the considerably lower reactivity compared with OH, its higher peak concentrations in the night-time troposphere allow the NO3 radical to play a major role in transformation of organic compounds. The NO3 radical is formed during nighttime in reaction with NO2. The dinitrogen pentoxide (N2O5) is a reservoir compound for the NO3 radical at low temperatures, but it is broken down to its precursors NO2 and NO3 at higher temperatures:

NO2 + O3 → NO3 + O2

(9.24)



NO3 + NO2 + M → N 2 O5 + M

(9.25)



N2 O5 + M → NO2 + NO3 + M.

(9.26)

The typical night-time NO3 radical concentrations in the atmospheric boundary layer are in the order 107 to 108 molÂ� ecules m−3 (which is the pptv range). During daytime both NO3 and N2O5 photo dissociate so fast that the concentrations of these compounds are insignificant. In the tropospheric boundary layer the photolysis of NO3 radical (with a typical noon lifetime of about 5 s) follow two different paths (λ represents here the wavelength):

NO3 + hv (ν < 700nm ) → NO + O2

(9.27)



NO3 + hv (ν < 580nm ) → O2 + O(3 P).

(9.28)

Close to pollution sources from combustion processes, e.g. road traffic or power plants, the NO3 radical is quickly removed by reaction with NO:

NO3 + NO → 2NO2 .

(9.29)

During night-time the heterogeneous conversion of N2O5 to HNO3 is an important process:

N 2 O5 + H2 O → 2HNO3 .

(9.30)

The lifetime of N2O5 with respect to this removal is on the order of minutes in the tropospheric boundary layer. This production of HNO3 may in winter be equally important as daytime conversion of NO2 by OH radical. As already described, particulate nitrate (NO3−) is formed when HNO3 reacts with NH3 and form new aerosol particles, and when it sticks to existing particles in the atmosphere. In addition organic NO3− may be formed from gaseous NO2 on the surfaces of aerosols in other heterogeneous reactions (see Section€9.3.4). The NO3 radical attacks alkanes by hydrogen abstraction in a similar way as the reactions of the OH radical:

190

RH + NO3 → R + HNO3 .



(9.31)

Followed by formation of a peroxy radical (RO2) that may oxidize an NO molecule to NO2. Also for alkenes, the attack of the NO3 radical is similar to the reactions of the OH radical; the NO3 radical adds to the double bond. This reaction is followed by rapid O2 addition which leads to the production of a peroxy radical. Reaction of soil emissions of NO with atmospheric OH has been suggested to provide an in-canopy source of HNO3 (Farmer et€al., 2006). HNO3 concentrations usually peak during the day, regulated by the emissions of NOx, photochemical activity and the gas/aerosol equilibrium of NH4NO3 shifting towards the gas phase with increasing temperature and decreasing relative humidity. A reaction similar to reaction (9.22), but less important, is the reaction between NO and OH radical forming nitrous acid (HONO):

NO + OH + M → HONO + M.

(9.32)

During daytime HONO photo dissociate (λ < 400 nm) rapidly back to the reactants:

HONO + hv → NO + OH.

(9.33)

Therefore, HONO formed in the late evening may serve as a night-time reservoir of OH and NO, which are subsequently liberated again the following morning, when sunlight starts up reaction (9.23) again. Studies in the highly polluted Po Valley in Northern Italy have shown an interrelation between simultaneous peaks in NOx concentrations and aerosol surfaces and peak HONO concentrations during foggy periods (Notholt et€al., 1992). This was taken as evidence for heterogeneous �conversion on aerosol surfaces through either of the reactions:

NO + NO2 + H2 O → 2HONO

(9.34)



2NO2 + H2 O → HNO3 + HONO.

(9.35)

Probably this type of conversion plays an important role also in many urban areas over Europe, but so far only few studies have been carried out.

9.3.4╇ Organic N compounds in the atmosphere The presence of atmospheric organic N compounds has been evident for years (Cornell et€ al., 2003; Neff et€ al., 2002), but direct measurements of individual species are rather sparse. The evidence of organic N comes from analysis of precipitation samples for total N and comparison with inorganic N content, to give ‘dissolved organic N’ (DON) by difference. This approach has been prone to several analytical artefacts (Cape et€al., 2001), but DON may in fact contribute up to half of total water-soluble N in precipitation. The fraction depends highly on location, and on whether air masses are of marine or terrestrial origin. DON has been ignored in estimating environmental consequences of N deposition, although there is reason to believe that many, if not all, components of DON are biologically available (Krab et€ al., 2008; Lipson and Nasholm,

Ole Hertel

2001; Paerl and Whitall, 1999; Qualls and Haines, 1992). There appears to be a DON contribution from marine air masses (Cornell et€al., 1995, 2001), and DON proportions are consistently high in unpolluted air, especially in the tropics. For continental/terrestrial samples, annual average concentrations of DON in precipitation appear to correlate better spatially with NH4+ than with NO3−, suggesting an agricultural source, but the seasonal variation is not correlated with NH4+ concentrations, implying that different sources are involved (Cape et€al., 2004). There is limited evidence available from sampling of air diÂ�rectly using a nebulizing mist sampler that both gas phase and particle phase components contribute to water-soluble organic N (WSON) in the atmosphere, which leads to the occurrence of DON in precipitation. Organic N is measured in fog (Zhang and Anastasio, 2001) and cloud water (Hill et€ al., 2007), but there is some concern that most analyses for DON are made on bulk rainfall samples (i.e. collected using an open funnel) and that a significant fraction of the measured DON might have been dry-deposited on the funnel surface. This presents problems of interpretation, but does not remove the problem of identifying the source, composition and fate of organic N compounds (see also Figure€9.16).

nitrates that serve as important reservoirs of NO2. The most abundant of these nitrates is the peroxy acetyl nitrate (PAN):

CH3 CHO + hv → CH 3C(O)

(9.36)



CH3 CHO + OH → CH 3 C(O) + H2 O

(9.37)



CH3 C(O) + O2 → CH3 C(O)OO

(9.38)

CH3 C(O)OO + NO2 + M → CH3 C(O)OONO2 + M (9.39) Reaction (9.38) is very fast, and reactions (9.38) and (9.39) may for many practical purposes be regarded as taking place in one step. PAN is thermally unstable and equilibrium between peroxy acetyl radical and NO2 on one side and PAN on the other side is established in the boundary layer. High PAN and O3 concentrations are often observed together during photo chemical smog episodes. The thermal degradation gives PAN a lifetime of ~ 1.7 h at 273 K and 50 h at 263 K. The PAN formation is competing with NO degradation of peroxy acetyl radical: CH3 C(O)OO + NO → CO2 + NO2 + CH3O2



Evidence for reduced organic N in the atmosphere Some measurements of individual components of reduced organic N in gas, particulate and aqueous phases have been reported and indicate potential sources and fates of these compounds (Table€9.3), but in most cases rather small concentrations are measured and these cannot account for the rather high proportions of DON in precipitation.

Formation of organic oxidized N compounds When aldehydes are photo dissociated or react with OH, an alkyl radical is formed, which in turn may form peroxy alkyl

This reaction is totally dominating at ppbv levels of NO meaning that PAN and other peroxy alkyl nitrates are formed only in the background atmosphere, and i.e. not inside urban areas. However, substantial PAN concentrations may still be observed in urban areas, especially at relatively low temperatures. The peroxy alkyl nitrates include compounds �produced in a similar way as PAN, but generated from biogenic isoprene emissions that may be of importance in southern Europe, and have similar thermal degradation pathways as PAN.

Figure 9.16 Illustration of the interaction between the various nitrogen oxide (NOy) compounds in the tropospheric boundary layer. The symbol Δ represents energy leading to thermal degradation, hν solar radiation leading to photo dissociation and RH a hydrocarbon reacting with the specie in question. PPN is a notation for peroxy propionyl nitrate, but also other peroxy nitrates than PPN and PAN play a role in this context (Derwent and Hertel, 1998).

HONO OH, HONO, NO2

H, H

O

NO

hυ

O

NO

hυ

hυ

,N

O

2

O3, HO2, RO2, NO3

ROO2

NO2

∆ HO2

∆ , hυ

HO2NO2

∆,

hυ

NO

NO

3

NO3

hυ, O3, NO

N2O5

2

∆,

hυ

RH

PAN PPN

(9.40)

OH

HNO3

191

Nitrogen processes in the atmosphere Table€9.3 Reported data on reduced organic N compounds in the atmosphere

Species

Sources

Atmospheric reactions and fate

Other direct measurements

Amines and Urea

Direct emissions from some industrial processes and from agricultural activity such as slurry spreading (Kallinger and Niessner, 1999). Some data exist for the latter process relative to ammonia emissions, which may be helpful in modelling emissions, but release rates are c. 1% of NH3. Ocean water may be source or sink (Hatton and Gibb, 1999).

Water soluble (Calderon et€al., 2007; Gibb et€al., 1999). Likely acid–base reactions and form particles (cf. NH4+) (Angelino et€al., 2001; Blando and Turpin, 2000). Photo degradation possible (McGregor and Anastasio, 2001). O3 reaction (Tuazon et€al., 1994). Likely high dry deposition rates (low surface resistance€– ‘sticky’ molecules).

PTR-MS data on trimethylamine from slurry (Twigg, 2006). Published data (Beddows et€al., 2004; Gorzelska et€al., 1992; Gorzelska et€al., 1994; Gronberg et€al., 1992; Gundel et€al., 1993; Palmiotto et€al., 2001; Schade and Crutzen, 1995) (Cornell et€al., 1998; Mace et€al., 2003b; Mace et€al., 2003c; Mace et€al., 2003a; Mace and Duce, 2002).

Amino acids, proteins and peptides

Ocean surface (Milne and Zika, 1993), grassland (soil) (Scheller, 2001).

Photodestruction (Anastasio and McGregor, 2000; McGregor and Anastasio, 2001; Milne and Zika, 1993; Saxena and Hildemann, 1996).

Measurement data (Gorzelska et€al., 1992; Kieber et€al., 2005; Mace et€al., 2003b; Mace et€al., 2003c; Mace et€al., 2003a; Matsumoto and Uematsu, 2005; Scheller, 2001; Zhang and Anastasio, 2001; Zhang and Anastasio, 2003).

Hydrogen cyanide and methyl cyanide (Acetonitrile)

Biogenic emissions (terrestrial) (Shim et€al., 2007) and biomass burning (Bertschi et€al., 2003; Cicerone and Zellner, 1983; Holzinger et€al., 1999; Li and Tan, 2000); anthropogenic (Holzinger et€al., 2001).

(Bange and Williams, 2000; Cicerone and Zellner, 1983; Karl et€al., 2004). Other direct measurements:€(Sprung et€al., 2001; Warneke et€al., 2001).

9.4╇ Dry deposition and bi-directional fluxes of N compounds Nr compounds are being monitored in many regional networks across the world, such as the European EMEP programme (EMEP, 2010), the NitroEurope Integrated Project (NEU, 2010) the US National Atmospheric Deposition Network (NADP, 2010), the Acid Deposition Monitoring Network in East Asia (EANET, 2010) and several others. However, these networks measure air concentrations rather than fluxes, and deposition is estimated using inferential modelling approaches, which are underpinned by often sparse databases of campaign based process studies with limited geographical coverage. This is partly due to the fact that instrumentation to measure fluxes of sticky compounds such as NH3, HNO3 or HONO are expensive and labour intensive to operate. The measurement of each individual Nr compound is technically more challenging than that of CO2 fluxes, for example. Robust low cost flux measurement approaches are lacking, although recent developments of a Conditional Time-Averaged Gradient (COTAG) method (Famulari et€al., 2010) show promise for wide-scale deployment over long periods for short vegetation. A first regional flux measurement network for Nr compounds is established within the European NitroEurope IP. This network takes a three-tier approach, where selected Nr compounds

192

are measured at a network of 13 supersites, using advanced micrometeorological flux measurement techniques. At a further nine regional sites the novel COTAG systems are being deployed, while deposition is derived at a further 50+ sites from concentration measurements, using inferential techniques (Tang et€al., 2009). Spatial coverage of Nr deposition can only be achieved through numerical modelling. The gaseous Nr compounds most commonly considered for dry deposition are NH3, HNO3 and NO2. Their relative contributions to N deposition depend on the pollution climate. In agricultural areas NH3 may dominate the atmospheric N loading, while in more industrial and urban areas HNO3 and NO2 may be more important. In addition, NH3 deposition depends on the N status of the receiving surface, with fertilized vegetation and vegetation receiving high atmospheric N deposition inputs acting as a less efficient sink or even net source of NH3. In dry regions, stomatal deposition may make a larger relative contribution to net exchange than in wet regions, where leaf cuticles provide a very efficient sink for water soluble gases (NH3 and HNO3). In the UK, dry deposition of NH3, HNO3 and NO2 is estimated to have contributed 48 (14.5%), 57 (17.3%) and 9 (0.03%) Gg, respectively, to the total N deposition of 330 Gg N in 2004, with the rest originating from wet and cloud deposition (211 Gg, 63.9%) and aerosol deposition (16 Gg N, 0.05%) (Fowler

Ole Hertel

et€ al., 2009). The UK deposition model uses detailed knowlÂ� edge of land use to estimate the vegetation-dependent deposition velocities and fluxes as a function of land use in each 5â•›km × 5â•›km grid square of the country, combined with longterm measurements of air concentrations which are unique in Europe in terms of spatial coverage.

9.4.1╇ The dry deposition process Dry deposition is the direct deposition of gases or aerosols at terrestrial or marine surfaces. The dry deposition of gases and particles is a continuous process and governed by their air concentrations, turbulent transport processes in the boundary layer, the chemical and physical nature of the depositing species, and the capability of the surface to capture or absorb the species. In relation to deposition transport, the boundary layer may be considered to consist of two layers:€the fully turbulent layer and the quasi-laminar layer. The quasi-laminar layer is introduced to quantify the way in which pollutant transfer differs from momentum transfer in the immediate vicinity of the surface (Hicks et€al., 1987). In this layer, the transport is dominated by molecular diffusion. Once at the surface, the chemical, biological and physical nature of the surface determines the capture or absorption of the gases and particles. Deposition to water surfaces (oceans or fresh waters) may thus be very different from deposition to vegetated surfaces on land. The deposition process may be considered as a series of resistances, by analogy with an electrical circuit (Monteith and Unsworth, 2008). The resistances refer to the transport proÂ� cesses through the various ‘layers’ defined above:€ turbulent transfer (usually denoted Ra), quasi-laminar (Rb) and surface (Rc). For a complex surface with several potential absorption sinks (e.g. vegetation) the resistance Rc may be viewed as a network of parallel resistances, representing transfer to the external leaf surface, through stomata, to water on the surface, or through the canopy to the underlying soil surface. The total resistance (RT) is the sum of all the series and parallel resistances (Ra€+€Rb€+€Rc), and is usually expressed in units of s€m−1. The inverse of the total resistance (1/RT) is known as the deposition velocity (vd) and has units of m s−1. The turbulent transfer resistance (Ra) depends upon the height at which the deposition flux is measured, so the total resistance (RT) and deposition velocity (vd) also vary with height above the surface. The transfer flux (F) is defined as the product of the air concentration of a gas or particles at height z, multiplied by the deposition velocity at height z, and (in the absence of competing chemical reactions (Sorensen et€ al., 2005)) does not vary with height, provided that the air concentration is horizontally uniform. This formulation assumes that the surface concentration of the gas is zero€– where this is not the case (see below) the effect can be described either as a decreased driving force for deposition (concentration difference between height z and the surface) or as an increased surface resistance. The deposition velocity (vd) is often reported as a constant even though it depends on a set of variables, e.g. wind

speed, surface roughness and atmospheric stratification. Joffre (1988) has suggested a parameterization which depends on the meteorological conditions, roughness length and the molecular diffusion coefficient for the compound of interest. The various components of the total transport resistance can be estimated from meteorological data if several assumptions are made concerning spatial and temporal homogeneity. The atmospheric turbulent resistance (Ra) can be calculated from:

╇╇╇

Ra ( z ) =

1  z  z z0    ln −ψ  ,   , κ u *  z0  L L 



(9.41)

where z is the reference height, z0 is the roughness length, u* is the friction velocity, κ is the von Karman constant (≈ 0.4) and L is the Monin–Obukhov length. For the diabatic surface layers (Businger, 1982) a stability function ϕ is introduced (Businger et€al., 1971). For neutral conditions ϕ = 1 and ϕ is greater/less than unity for stable/unstable stratifications. In the above equation, ψ is the integrated stability function. The resistance of the underlying thin molecular lamÂ� inar sub-layer is given by (Kramm, 1989; Kramm et€al., 1991; Kramm and Dlugi, 1994): Rb = ∫

z

zs

Di + K

=

  uz + Bi   u u 



(9.42)

╇╇╇ where uz0 is a characteristic velocity for the layer zs < z < z0, Bi is the sub-layer Stanton number, which is a function of the roughness Reynolds number Re* = u* z/ν and the Schmidt number, Sci = ν/Di. The Stanton number can be estimated as (Kramm and Dlugi, 1994): Bi−1 = aSc ib Re *c + ε , (9.43) where the following values are suggested for smooth surfaces a = 13.6, b = 2/3, c = 0 and ε = −15.5 and for rough surfaces, a = 7.3, b = 0.5, c = 0.25 and ε = −5. The surface resistance term depends on the physical and chemical nature of the absorbing surface, and parameterizations should be adapted to the surface concerned. The value of vd is often expressed as annual or seasonal averages, for the purpose of calculating deposition fluxes as the product of air concentrations and deposition velocities. Deposition velocities and concentrations should refer to the same height€– usually the height at which the concentrations are measured. Tall vegetation causes increased atmospheric turbulence, so Ra values are smaller, and deposition velocities are larger, than for short vegetation. Consequently, estimating deposition of different components to the countryside requires knowledge about land use as well as the spatial pattern of air concentrations. The air–sea gas exchange of the very soluble gases HNO3 and NH3 is rate limited by the vertical transport in the boundary layer, because the uptake at the water surface is very fast relative to other commonly studied gases. Of the two very soluble N-gases, HNO3 exchange rates are larger than NH3 due

193

Nitrogen processes in the atmosphere

to the higher solubility. The less soluble NO2 and NO gases, deposit much slower to the marine surface. The surface resistance is the most important resistance for slightly soluble gases and relates to the transfer velocity Kc, which is also used for air-sea exchange of other gases like CO2, DMS and CH4. The surface resistance is a key parameter for the deposition of a gas to a water surface, and may be expressed as:



F 1 = Kc = c . ∆c w Rc

(9.44)

Here Fc is the flux across the surface and Δcw is the concentration gradient across the laminar sub-layer in the water. The resistance across the water surface is controlled by the Henry’s law coefficient (H), which describes the solubility of different gases, and is a strong function of temperature. The effective overall surface resistance is therefore:

Rc ,e ff = Rc H *,

(9.45)

where H* is the dimensionless Henry’s law coefficient (Table€9.4). The process of dry deposition of particles differs from that of gases in two respects. • Deposition depends on particle size, since transfer to the surface involves Brownian diffusion, inertial impaction/ interception and sedimentation (all of which are a strong function of particle size) (Slinn, 1982). • It is assumed that the surface resistance for particles less than 10 μm diameter (Hicks and Garland, 1983) is negligibly small to all surfaces. For submicron particles, the transport through the boundary layer is more or less the same as for gases. However, transport of particles through the quasi-laminar layer can differ. For particles with a diameter <0.1 μm, deposition is controlled by diffusion, whereas deposition of particles with a diameter >10 μm is more controlled by sedimentation. Deposition of particles with a diameter between 0.1 μm and 1 μm is determined by the rates of impaction and interception and depends heavily on the turbulence intensity. Transfer through the quasi-laminar layer close to the surface presents a considerable restriction on the deposition of 0.1–1.0 µm diameter particles. Uptake of particles by surfaces is thus largely controlled by micro-structures and turbulence intensity. Most of the theory and measurements of particle fluxes have focused on sulphate particles (SO42−), which mostly occur in the submicron size range as (NH4)2SO4. Other submicron aerosol particles are expected to behave similarly, although semi-volatile particles may form or evaporate depending on the local equilibrium with the constituent gases (e.g. NH4NO3 and NH3/HNO3). The most widely used model is an empirical parameterization (Wesely et€ al., 1985), which is based upon flux measurements of SO42− over grass. In this model, vd is represented as a function of the friction velocity u* and the Monin–Obukhov length L. Then for SO42− particles and low vegetation, vd can be calculated by using (Erisman and Draaijers, 1995):

194

vd = vd =



2 /3 u*   300   ⋅ 1 +    500   − L  

u* 500

L<0 L>0



(9.46)

Ruijgrok proposed another parameterization derived from measurements over coniferous forest (Ruijgrok et€ al., 1997). In this approach, which is simplified from the Slinn model (Slinn, 1982), vd is not only a function of u*, but also of relative humidity (RH) and surface wetness. Inclusion of RH allows for particle growth under humid conditions and for reduced particle bounce when the canopy is wet. Dry deposition velocity is expressed as: 1 1 = Ra + , vd vds



(9.47)

where Ra is the aerodynamic resistance, which is the same as for gaseous species, and vds is the surface deposition velocity. For tall canopies vds is parameterized by (Ruijgrok et€ al., 1997) as vds = E ⋅



u 2* , uh

(9.48)

where uh is the wind speed at the top of the canopy, which is obtained by extrapolating the logarithmic wind profile from ZR to the canopy height h. Now uh can be expressed as:

uh =

u* k

  10 ⋅ z0 − d   z0    10 ⋅ z0 − d   ln   −ψ h  h  + ψh    . (9.49) z L  L     0   

Note that E is the total efficiency for canopy capture of particles, parameterized for dry and wet surface separately (Erisman et€al., 1997). For dry surfaces, for SO42− particles (Brook et€al., 1999): 0.005 u 0.28 *  E = 0.28 0.005 u * 

RH − 80   ⋅ 1 + 0. 18 ⋅ exp 20  

RH ≤ 80%

(9.50)

RH > 80% .



For wet surfaces, for SO42− particles (Brook et€al., 1999): 0.08 u*0.45  E = RH − 80  0.45  0.08 u* ⋅ 1 + 0.37 ⋅ exp 20    

RH ≤ 80%

RH > 80% ,

where RH is taken at the reference height.



(9.51)

Ole Hertel

Erisman and Draaijers used the following general form for the calculation of vd (Erisman and Draaijers, 1995):

vd =

1 Ra +

1 vds

+ vs ,

Compound

(9.52)



where vs is the deposition velocity due to sedimentation, to represent deposition of large particles, and vds can be estimated from (9.48). Relations for E for different components and conditions may be derived from model calculations and multiple regression analysis (Erisman and Draaijers, 1995). For larger supermicron particles (Na+, Ca2+ and Mg2+), and therefore for some NO3− particles, and for low vegetation (for all particles), the sedimentation velocity has to be added:

vs = 0.0067 m ⋅ s −1 vs = 0.0067 ⋅ e

0.0066⋅ RH 1.058 − RH

RH ≤ 80 m ⋅ s −1

RH > 80%.

Table€9.4 Summary of Henry’s law coefficients of various gaseous nitrogen compounds

(9.53)

For sulphate deposition velocity, observations suggest that there is a distinct upper limit which depends on land use type. As a result, it is required that

vds ≤ vm , (9.54) where vm is the observed maximum deposition velocity (Walcek et€al., 1986).

9.4.2╇ Bi-directional fluxes of N-containing gases Plant fixation of N2 provides the single largest atmospheric N input to the biosphere worldwide. However, since it is not associated with acidifying effects, controlled by the plants themselves and its rate is not altered through human activity (other than through land-use change), it is not usually considered in atmospheric N deposition budgets. Direct measurement approaches of N2 fixation are lacking as the flux to the biosphere is very small compared with ambient N2 concentrations. Instead N2 fixation is measured in laboratory, e.g. with isotope techniques. Nitrous oxide (N2O), an important greenhouse gas with a lifetime of 114 years, is usually assumed to be emitted by terrestrial surfaces (see Jarvis et€al., 2011, Chapter€10, this volume). Although reports of transient N2O deposition fluxes in the literature are increasing in number, see, for example, Flechard et€al. (2007), the magnitude of N2O uptake is small and negligible compared with the main contributors to atmospheric N deposition. For the other N containing gases there are several parallel pathways of pollutant exchange with vegetation, which include adsorption to the leaf cuticles, exchange through the stomata with the sub-stomatal cavity and exchange with the soil. All these processes are potentially bi-directional, depending on the relative magnitude of the air concentration and the gaseous concentrations in chemical equilibrium with the leaf surface, the apoplastic fluid and the soil solution, respectively. The likelihood for uptake increases with the water solubility

Henry’s law coefficient at 25 °C in water [mol kg−1 bar−1]

NH3

61

HNO3

2.6 × 106

HONO

49

NO

0.0019

NO2

0.012

N2O

0.025

PAN

4.1

and Henry’s law coefficient of the gas, which vary over Â�several orders of magnitude (Table€ 9.4). A database of Henry’s law coefficients is available (Mainz, 2010).

Nitric acid Because of its high deposition rate, HNO3 makes a significant contribution to Nr deposition in areas exposed to air containing emitted NOx. HNO3 is highly water soluble and commonly assumed to deposit at the maximum rate permitted by turbulence, i.e. surface resistance is negligible. While this is probably a reasonable approximation for most situations, several authors have observed emission gradients or reduced uptake rates of HNO3, probably owing to non-zero HNO3 surface concentrations in equilibrium with NH4NO3 aerosol deposited to leaf surfaces (Neftel et€al., 1996; Nemitz et€al., 2004; Zhang et€al., 1995). In the case of trace gases with negligible surface resistance, the deposition velocity is very sensitive to the atmospheric resistances (Ra and Rb), which over aerodynamically rough surfaces are small (5–10 s m−1). In such conditions, even a very small surface resistance for HNO3 would strongly influence deposition rates. Currently there are insufficient field data to show whether HNO3 deposition is subject to a surface resistance, and this remains a research priority.

Ammonia NH3 dominates atmospheric N deposition to semi-natural vegetation in agricultural areas, especially in Northern Europe where NH3 deposition is favoured at high humidity and cold temperatures, although, these conditions also favour conversion to ammonium aerosol. NH3 is less water soluble than HNO3. Thus, NH3 previously absorbed to wet leaf surfaces may more readily be desorbed (re-emitted) as leaf water layers dry out again (Flechard et€al., 1999). Another complication is that plants under certain conditions may release NH3. Generally plants contain inorganic N in the form of NH4+ and NO3−. These nutrients are mainly present in the liquid part (apoplast) between the cells of the plant. NH4+ is an important by-product of plant biochemical pathways resulting in non-zero NH4+ concentrations in the leaf apoplast, which results in non-zero gas-phase concentrations (stomatal compensation points, χs) in equilibrium with this NH4+apo

195

Nitrogen processes in the atmosphere

concentration at the apoplastic pH, for example. Current evidence suggests that NH4+apo increases with increasing N supply to the plant, either through fertilization or high atmospheric N inputs. The compensation point χs is the product of a temperature function describing the Henry’s Law equilibrium and the ratio of Γs = [NH4+apo]/[H+apo]. Values of Γs range from <100 for semi-natural vegetation in clean, remote environments over values around 1500 for semi-natural vegetation in environments with high N deposition to >10 000 after fertilization. At 10 ºC, this equates to values of χs of < 0.15, 2.3 and > 15 µg m−3, respectively. Emission potentials of fertilized soils can be even larger. This large range illustrates that the direction of NH3 exchange is often difficult to estimate a priori. Several papers have recently reviewed the literature on bidirectional NH3 exchange and compiled extensive database on compensation points (Massad et€ al., 2010b; Zhang et€ al., 2010) in order to provide the necessary input for application in atmospheric transport models. The compensation points increase with N input as it is the main driver of apoplast and bulk leaf NH4+ concentrations (Massad et€al., 2010a), but the compensation point also vary between different plant species and with growth stage and season (Riedo et€ al., 2002). The decomposition of litter has been found to play a dominant role (Zhang et€al., 2010). The stomatal pathway for NH3 exchange is only available when stomata are open during daytime, and thus deposition to (often wet) leaf surfaces is the dominant pathway during the night, unless soil surfaces provide a major source and are well exposed to the atmosphere. Deposition fields of NH3 are particularly uncertain, due to (i) uncertainties in the overall magnitude as well as spatial and temporal patterns of agricultural NH3 emissions and (ii) the large variability of NH3 deposition rates to different surfaces. Specific dry deposition sub-models for the surface resistance that include the description of a canopy compensation point for NH3 have been derived and implemented in connection with the analysis of different plant surfaces, e.g. for beans (Farquhar et€ al., 1980), oilseed rape plants (Husted et€al., 2000), and heather (Calluna vulgaris) (Schjørring et€al., 1998). It is common to apply a two-pathway process description (Fowler et€al., 2009; Loubet et€al., 2001):€(a) a stomatal pathway, which is bi-directional and modelled using a stomatal compensation point, and (b) a plant surface pathway, which denotes exchange with water surfaces or waxes on the plant surface. The stomatal compensation point may be calculated from knowledge of the aqueous phase chemistry. The equilibrium NH3 ambient air concentration for the stomatal compensation point has been expressed as (Sorteberg and Hov, 1996):  NH4+  NH 3 ( g )  = χ cp = 10 (1.6035 − 4207.62 /T )  +  ,  H  (9.55) where χcp is the compensation point concentration of NH3, and [NH4+] and [H+] are the concentrations of ammonium and hydrogen ion in stomatal cavity, respectively.

196

The leaf surface may work as a capacitance for NH3 and SO2 uptake, and this capacitance increases with humidity (Van hove et€ al., 1989). This transport is independent of solar radiation and contrary to the uptake through stomata, it also takes place during night. Sutton et€al. (1998) defined the canopy compensation point as: ′ (9.56) where χz09 is the canopy compensation point, χ is the NH3 ambient air concentration, z is the height above ground, d is the displacement height, and Fg is the vertical flux. The vertical flux Fg may be divided into a flux towards the leaf surface Fw and a flux through stomata Fs:

Fg = Fw + Fs .



(9.57)

And these fluxes may be written as: ′

′

where Rw and Rs are leaf surface and stomatal resistances, and expressions for these may be found in the work by Sutton et€al. (1993, 1998). From the above two equations the total flux Fg may be expressed as: cp

Fg =

−

′ z0

−

′ z0

.

(9.58)



Similarly the total flux may be derived from the expression of the canopy compensation point: s

w

− ( z) . Ra ( z ) + Rb ′

(9.59) Combining these two equations and eliminating the total flux Fg provides a general expression for the canopy compensation point: Fg =

′

z0



z0

 ( z )  + cp    Ra ( z ) + Rb  Rs = . −1 ( Ra ( z ) + Rb ) + Rs −1 + Rw −1

(9.60)

Several generalized parameterizations of bi-directional NH3 exchange have recently been developed for inclusion in regional CTMs (Gore et€al., 2009; Massad et€al., 2010b; Zhang et€al., 2010), but these have not yet been tested in the spatial modelling environments. In an earlier study, Sorteberg and Hov implemented a simpler parameterization of bi-directional fluxes of NH3 into a Lagrangian long-range transport model, assuming pH to be a constant value of 6.8 (Sorteberg and Hov, 1996). Concerning the concentration of NH4+(aq), they assumed this to be 150 and 50 μmol l−1 for crop and grassland, respectively. The model with these relatively crude assumptions was applied for the European area for the year 1993, and compared with basic scenario without bi-directional flux

Ole Hertel

parameterization. The results indicated a reduction of 0%–20% in total sulphur deposition and a 0%–25% increase in NH3 deposition compared with a simple flux model. The emission through stomata was found to account approximately 0.1% of the total NH3 emission. Loubet et€al. applied a 2D local scale model with the above parameterization of bi-directional fluxes of NH3 based on the canopy compensation point approach to a moorland area (Loubet et€al., 2001). With the FIDES (Flux Interpretation by Dispersion and Exchange over Short Range) model they simulated transport and dispersion to a moorland placed 260 m downwind from a pasture grazed with sheep. Experimental studies have shown that over the sea the atmospheric fluxes of NH3 may also be upward or downward (Lee et€ al., 1998; Quinn et€ al., 1988; Sørensen et€ al., 2003) depending on the meteorological conditions and the relationship between the pH and contents of NH4+ in the upper surface waters on the one side, and the NH3 concentrations in ambient air just above the water surface on the other side. The bidirectional NH3 flux over sea is expressed as an exchange with the water surface: ╇

F = Ve ( Ceq − C air ) ,



(9.61)

where Ve is the exchange velocity between air and sea (that equals 1/(Ra + Rb)), Ceq is the NH3 concentration in the air at equilibrium with the NHx in the water, and Cair is the actual ambient air concentration of NH3. F is the flux of NH3; the flux is positive when the sea emits NH3 and negative when deposition takes place. The ambient air NH3 concentration at equilibrium is expressed as (Asman et€al., 1994): M NH3 [ NH xs ]

Ceq =

R ×T × H NH3

,  1 10 − pHs  +    γ NH3 γ NH4 × K NH4 





 1 56 EX P 4092  T 

1  , 298 .15  

 1  1 K NH4 = 5.67 ×10 −10 EX P −6286  − .  T 298 .15   

Nitric oxide NO is rather water-insoluble and there is no efficient mechanism for NO to react on the surface or inside leaves, so its deposition rate is rather slow. By contrast, soils commonly act as a source for NO. Some of these soil emissions of NO are oxidized to NO2 (and possibly HNO3) within plant canopies, and taken up more efficiently than NO and thus the behaviour of NO still needs to be taken into account in surface–atmosphere exchange. Nitrogen dioxide Plant uptake of NO2 is slower than that of the more water soluble gases (HNO3, NH3), but it is a significant contributor to N deposition. The NO2 deposition to vegetation is primarily regulated by stomata, and for most plants the internal resistance is negligible, and NO2 deposition velocities may thus be computed from a knowledge of stomatal resistance or conductance (Thoene et€ al., 1991). Studies indicate a small effective stomatal compensation point for NO2 for some plant species, in the range of > 0 to 2 ppb; e.g. an American experimental study found a value of 1.5 nmol mol−1 for the canopy compensation point for NO2 over deciduous forest (Horii et€al., 2004). However, the underlying process is not currently understood, and some laboratory work has failed to reproduce the field observations. Because of its low water solubility, deposition to (and reaction with) surface water, including sea water, is also slow (Cape et€al., 1993).

Nitrous acid (9.62)

where Ceq is in [μg m−3], [NHxs] is the NHx concentration in the sea [μm], MNH3 is the molecular mass of NH3 [g mol−1], γNH3 is the activity coefficient of NH3×H2O, γNH4 is the activity coefficient of NH4+ in sea water, R is the gas constant (8.2075×10–5 atm. m3 mol−1 K−1) and HNH3 is the Henry’s law coefficient for NH3 [m atm−1], pHs is the pH in sea water, which is a measure of the activity of H+ in sea water, and KNH4 is the dissociation constant for NH4+ [M]. The values for HNH3 and KNH4 are expressed as: H NH3

The above formulation was developed for computing the impact of bi-directional fluxes over the North Sea and applied to measured data (Asman et€al., 1994). The formulation has since been applied in the Lagrangian ACDEP model (Sørensen et€ al., 2003), where the results showed a redistribution of N deposition in the coastal region off the coast of the Netherlands.

The biosphere/atmosphere exchange of HONO is generally bi-directional, and daytime concentrations of HONO are low, as it is rapidly photolysed in sunlight. With solubility similar to NH3, HONO is deposited to vegetation under most conditions. Observations of HONO emission have been attributed to production of HONO at surfaces, e.g. through the reaction of NO2 with NO on wet surfaces (Harrison and Kitto, 1994) or NO2 reduction on humic acid (Stemmler et€ al., 2006). In connection with an experimental study, a parameterization of bi-directional fluxes of both NH3 and HONO was applied for estimating dry deposition of N compounds to the Amazon Basin from measured ambient air concentrations (Trebs et€al., 2006).

(9.63)

Organic nitrogen compounds

(9.64)

Organic N compounds account for approximately 20%–30% of the total N deposition in precipitation (Cape et€ al., 2001; Cornell et€al., 2003; Holland et€al., 1999) although this is often not included in N deposition estimates. Much of this organic contribution is presumably due to scavenging of organic N

197

Nitrogen processes in the atmosphere Figure€9.17 The processes of capture of pollutants by cloud and rain.

cloud droplet

AB

A CDEF

SO2

CDEF

NO3–

SO42–

HNO3 NO2 NO

precipitation A - dissolution E - impaction

B - oxidation C - diffusiophoresis D - Brownian diffusion F - cloud condensation nuclei pathway

compounds in the aerosol phase and cloud water. However, the contribution of gaseous organic N compounds to N deposition is even less studied. PAN is considered an important N reservoir species, responsible for much of the N transport in remote regions. PAN is thought to deposit slowly and remains stable at cold temperatures. At warmer tropospheric temperatures PAN decomposes quickly. Newly developed instruments have resulted in new measurements indicating deposition rates of PAN (and other PAN-like compounds) that are significantly larger than classical predictions (Turnipseed et€al., 2006; Wolfe et€al., 2009), especially to wet vegetation. Thus the lifetime of PAN with respect to deposition may be shorter than previously thought. In addition, PAN is water insoluble and the comparably large deposition fluxes to wet surfaces indicate that the current mechanistic understanding of the deposition process is incomplete. There are parallels to the deposition of O3, which also appears to exhibit larger deposition rates to wet surfaces than can be explained by its solubility (Fowler et€al., 2001). The importance of alkyl nitrates has recently been demonstrated for Blodgett Forest, Sierra Nevada, USA (Farmer et€al., 2006), although it appears that the pollution climate of their site is quite unique. Nevertheless, information is lacking to form a robust picture of the importance of these compounds across the full range of European conditions. Although amines have been measured as emitted from agricultural activities (Schade and Crutzen, 1995), there is currently no information on their dry deposition.

198

9.4.3╇ Deposition of N containing aerosols Deposition of particles containing SO42−, NO3−, Cl− and NH4+ contributes to the potential acidification and eutrophication (N components) of ecosystems. Compared to gaseous deposition of acidifying compounds onto low vegetation, particle deposition fluxes are usually found to be small. However, in difference from wet deposition it takes place all the time and furthermore it is believed that the fluxes of small particles are currently underestimated for very rough surfaces like forests. Erisman et€al. (1997) found that deposition of aerosols to the Speulder forest contributed 20% and 40% to the total dry deposition of S and N, respectively. Parameterizations of aerosol dry deposition velocities to forests differ greatly between models (Tang et€al., 2009).

9.5╇ Wet scavenging of N compounds from the atmosphere Wet deposition or scavenging is defined as the removal of gases and aerosol from the atmosphere by precipitation snow, rain. Unlike dry deposition, the wet deposition processes are indirect; rain, hail and snow are the vectors for transport of the pollutant to the surface. The apparent simplicity of the measurement approach for wet deposition, a simple precipitation collector placed on the ground contrasts appreciably with the underlying physical and chemical pathways of solutes into the collected precipitation

Ole Hertel

sample. There is also significant uncertainty in the relative magnitudes of dry deposition of trace chemical species as gases and aerosols onto the collecting equipment. The incorporation of pollutants in clouds and precipitation include many different processes, which will be considered in turn. The Nr compounds are present in aerosols or as gases. Regarding aerosols, the N is mainly present as NH4+ or NO3− (although some organic N is also present). The bulk of the aerosol mass is present in the size range 0.1–1.0 μm (diameter). These aerosols are removed through interception by falling rain or snow, a process known as washout or by incorporation of the aerosol into cloud droplets within clouds, a process known as rainout (Figure€9.17). Washout is responsible for 10%–20% of the N in wet deposition on average, but depends naturally on the relative amounts of N present in cloud water and in the air through which the precipitation falls. The aerosol scavenging within cloud occurs through a number of physical and chemical pathways (Figure€9.17) as C, D, E and F while the gases are incorporated through solution and oxidation processes (A and B). The phoretic process includes diffusiophoresis, in which aerosol particles are transported in the direction of a mean flux of vapour molecules. In the case of a cloud droplet growing by vapour

Figure 9.18 Orographic enhancement of precipitation in the UK; an East–West transect.

Figure 9.19 The seeder–feeder process of orographic enhancement of precipitation. Source: Fowler and Battarbee, 2005.

diffusion of water molecules towards the droplet surface, aerosols would move along the vapour flux towards the growing droplet. Additional phoretic mechanisms are presented by electrical and thermal gradients (electrophoresis and thermophoresis respectively). The phoretic processes contribute relatively small amounts of the solute in cloud water (Goldsmith et€al., 1963). Aerosols may also be captured by cloud droplets following Brownian diffusion (D) to the droplet surface and rates of Brownian diffusion vary strongly with particle size, being significant for particles smaller than 100 nm in diameter. However, diffusion rates are very small relative to molecular diffusion and diffusional mechanisms make only minor contributions to the wet removal pathway. The remaining minor process leading to capture of aerosols by cloud droplets is impaction and interception (E). As implied in the name these processes lead to the capture of aerosols by droplets when one is unable to follow the streamlines of airflow around the other and the aerosol and droplet collide. The bulk of the aerosol N in cloud water is incorporated through the activation of aerosols containing NO3− or NH4+ into cloud droplets. The N containing aerosols are effective cloud condensation nuclei and are readily incorporated into cloud droplets through the nucleation scavenging pathway. Thus the main route is nucleation scavenging for aerosol NO3−, and NH4+ (Pruppacher and Jaenicke, 1995). The pathway for below wet scavenging of the gaseous N compounds depends on the solubility and reactivity of the specific gas. In the case of NH3 and HNO3, which are highly soluble, clouds and rain remove these gases effectively from the air. The contribution of NO and NO2 to dissolved N in precipitation is very small as these gases are not very soluble (relative to NH3 or HNO3). Wet deposition is monitored by simple methods (precipitation collectors) analysed for major anthropogenic ions SO42−, NO3−, NH4+, H+ and marine ions Cl−, Na+, Mg2+. The networks of collectors for precipitation chemistry are much less dense than precipitation collectors for the national meteorological services, mainly because of the costs of chemical analysis. Furthermore, precipitation chemistry collectors are located a height above ground to reduce contamination from ground based sources, and the practice of locating collectors above the ground reduces the capture of small droplets due to aerodynamic screening by the collector. The relative contributions to deposition from dry and wet deposition change with distance from source as primary pollutant concentrations decline and oxidation from gas to particle remove gas phase species which dry deposit quickly. Thus the areas more than a few hundred km from sources receive most of their N deposition in precipitation. In regions in which the amounts of precipitation are large, wet deposition dominates the N loads, as in most of the uplands of Europe. However, it is not simply the precipitation amount that needs to be considered in assessing the relative contributions of wet and dry deposition. The processes leading to orographic enhancement of rainfall amount have a profound effect on the overall scavenging of pollutants from the atmosphere. The meteorological process which enhances precipitation in much of maritime northern Europe is the seeder–feeder

199

Nitrogen processes in the atmosphere Figure 9.20 The incorporation of pollutant aerosols into orographic cloud. Source: Fowler et€al., (1991).

mechanism, in which orographic cloud, formed over hills and mountains is washed out by precipitation falling from higher levels in the troposphere, as shown in Figure€ 9.18 and first described by Bergeron (1965). The process occurs widely and is responsible for most of the enhancement of precipitation over uplands in the UK and Scandinavia. The process has been extensively studied in the UK, where, especially in the West of the country, annual rainfall is in the range 1000–3000 mm with the amounts in excess of 100 mm being mainly generated through seeder-feeder scavenging. The mountains are very effective in increasing rainfall (Figure€9.18) and wet deposition by the seeder–feeder process (Figure€9.19) in which low level hill cloud droplets are washed out by falling precipitation from higher level. The hill cloud is more polluted than higher level cloud because boundary layer aerosols are effectively activated into cloud droplets as they are forced to rise and cool over the hills and mountains. The seeder–feeder effects on precipitation amount have been simulated in process-based models and are able to simulate observed spatial patterns in precipitation (Carruthers and Choularton, 1983). Models have also been used to simulate the wet deposition of pollutants over mountains (Dore et€al., 1990) and compared with detailed campaign measurements in an upland area. Extending the modelling of orographic enhancement of wet deposition to the country scale has enabled detailed spatially resolved wet deposition maps to be generated (Dore et€al., 1990). As orographic enhancement of wet deposition has been shown to be a major contributor to the total deposition in upland Britain the explicit inclusion of the process in deposition maps has been regarded as a routine component of wet deposition mapping (NEGTAP, 2001). The resulting wet, and total N deposition maps show a strong influence of altitude and requires a grid resolution on the same scale as the complex topography to reproduce (<10 km). Thus deposition modelling and mapping at a 50 km × 50 km scale fails to capture the spatial structure in wet

200

deposition, as described in Simpson et€al. (2011, Chapter€14, this volume). In principle the models are able to simulate the process as the literature shows, but the grid resolution of both the underpinning meteorological model and the model applied for deposition calculations need to be able to capture the topographic scale of the variability.

Cloud droplet deposition Unlike aerosols in the size range 0.1–1.0 µm, which are not deposited efficiently on vegetation, the hill cloud droplets are large enough (3–10 µm in diameter), to impact efficiently on vegetation (Fowler et€ al., 1990); this deposition pathway is termed cloud deposition or occult deposition. For the UK it provides a very small contribution to the total but it is important for hills which are frequently shrouded in cloud (see Figure€9.20). As the concentrations of major ions in hill cloud are enhanced, this deposition pathway leads to the exposure of vegetation to very large concentrations (SO42−, NH4+, NO3−, 1000 μeq 1–1–2000 μeq 1–1) (Fowler et€al., 1990). The orographic enhancement of wet deposition is not included in the assessments of wet deposition in all countries, and for regions with only small areas of upland, this will not lead to significant underestimates in wet deposition. However, for areas of Europe in which seeder–feeder scavenging of pollutant represents a substantial contribution to total deposition, it is important to simulate the process in mapping regional wet deposition, to avoid underestimating wet deposition and exceedances of critical loads.

9.6╇ Summary of emerging issues and unanswered questions There are many uncertainties and unanswered questions regarding the fate of Nr species in the atmosphere. These uncertainties and unanswered questions relate to various aspects of the governing atmospheric processes:€ emission, transformation,

Ole Hertel

transport (see Simpson et€al., 2011, Chapter€14, this volume) and deposition. Most of the N studies are concentrated on few mainly northern European countries. This applies for studies on agricultural NH3 emissions. There is therefore a strong need for more field and monitoring studies of atmospheric N compounds in other parts of Europe, i.e. southern European countries.

9.6.1╇ Emissions There is a need for emission inventories of higher spatial and temporal resolution than what is currently applied in atmospheric transport models, i.e. going from the traditional 50€km horizontal grid resolution towards inventories of 1–5 km and applying source specific seasonal and diurnal variations that account for actual meteorological conditions, local praxis, etc. The need for higher resolution inventories applies especially for NH3 for which the near-source deposition plays a significant role, and current deposition mappings therefore face significant uncertainties due to insufficient resolution in the inventories. Detailed information about agricultural practice is necessary in order to derive these inventories that need to account for meteorological conditions and how these affect the emissions from different agricultural source categories.

9.6.2╇ Transformation In relation to obtaining a better understanding of the transformation processes, there is a strong need for speciated field measurements and source apportionment studies of gas phase and particulate Nr compounds in the atmosphere. This applies especially to the organic N compounds for which the processes are not fully explored and the impact on the overall atmospheric N budget therefore still is fairly unknown. Aerosol processes are in general subject for further development in atmospheric transport models as they still cannot fully explain observed PM2.5 and PM10 mass, and this applies naturally also for the Nr compounds.

9.6.3╇ Deposition New parameterizations are needed to describe bi-directional NH3 exchange in atmospheric transport models, which currently tend to overestimate dry deposition to fertilized vegetation or semi-natural vegetation subject to large atmospheric N inputs. To put these parameterizations onto robust footings it is necessary to (i) compile databases of stomatal compensation points for the major biomes, (ii) derive operational, mechanistic parameterizations for cuticular uptake resistances (e.g. in response to local pollution climates), and (iii)€ compile European agricultural management profiles of fertilizer application. Litter decomposition is emerging as a potentially important source of atmospheric NH3, which is not well quantified, understood or parameterized. Improved mechanisms are required to deal with sub-grid variability in NH3 deposition. Due to limitation of the spatial

resolution, operational atmospheric transport models are usually not able to predict deposition to patch-work landscapes correctly. Because they average over rather large areas, the models underestimate the hotspots of dry deposition, which commonly results in the under prediction of critical loads exceedances. The large spatial variability in NH3 concentrations across the typical European landscape also makes it hard to validate the models. Direct micrometeorological measurement of dry deposition of reactive N compounds is limited to small-scale measurement campaigns, due to the costs of maintaining expensive instrumentation. Therefore dry deposition estimates usually rely on inferential modelling where deposition is derived from air concentrations which may either be measured or modelled, and (often uncertain) estimates of deposition velocities. Lowcost approaches are urgently required to provide robust direct measurements of Nr deposition across an extensive flux measurement network. The contribution of organic N compounds to total gaseous and particulate dry deposition is largely unknown. Evidence from the western USA suggests that alkyl nitrates can make a significant contribution to the total N flux (Farmer et€al., 2006), but for European conditions, this question remains largely unanswered. No fluxes have been measured for gaseous amines. New evidence suggests that PAN dry deposition rates may be much larger than assumed in classical modelling approaches (Turnipseed et€ al., 2006). The contribution of PAN to dry deposition of N may therefore have been underestimated. This also highlights the fact that our current understanding of the dry deposition mechanism is incomplete as PAN appears to deposit faster than can be explained by its solubility alone.

Acknowledgements This chapter was prepared with the support of the NinE Programme of the European Science Foundation, the NitroEurope IP (funded by the European Commission) and the COST Action 729.

References Anastasio, C. and McGregor, K. G. (2000). Photodestruction of dissolved organic nitrogen species in fog waters. Aerosol Science and Technology, 32, 106–119. Anderson, N., Strader, R. and Davidson, C. (2003). Airborne reduced nitrogen:€ammonia emissions from agriculture and other sources. Environment International, 29, 277–286. Angelino, S., Suess, D. T. and Prather, K. A. (2001). Formation of aerosol particles from reactions of secondary and tertiary alkylamines:€characterization by aerosol time-of-flight mass spectrometry. Environmental Science and Technology, 35, 3130–3138. Apsimon, H. M., Barker, B. M. and Kayin, S. (1994). Modeling studies of the atmospheric release and transport of ammonia in anticyclonic episodes. Atmospheric Environment, 28, 665–678. Asman, W. A. H., Harrison, R. M. and Ottley, C. J. (1994). Estimation of the net air–sea flux of ammonia over the Southern Bight of the North-Sea. Atmospheric Environment, 28, 3647–3654. Baldwin, A. C. and Golden, D. M. (1979). Heterogeneous atmospheric reactions: sulfuric-acid aerosols as tropospheric sinks. Science, 206, 562–563.

201

Nitrogen processes in the atmosphere Bange, H. W. and Williams, J. (2000). New directions:€acetonitrile in atmospheric and biogeochemical cycles. Atmospheric Environment, 34, 4959–4960. Beddows, D. C. S., Donovan, R. J., Harrison, R. M. et€al. (2004). Correlations in the chemical composition of rural background atmospheric aerosol in the UK determined in real time using timeof-flight mass spectrometry. Journal of Environmental Monitoring, 6, 124–133. Bergeron, X. (1965). On the low-level redistribution of atmospheric water caused by orography. In: Proceedings of the International Conference on Cloud Physics, Tokyo and Sapporo, 24 May–1 June 1965, pp. 96–100. Berkowicz, R. (2000). OSPM: a parameterised street pollution model. Environmental Monitoring and Assessment, 65, 323–331. Bertschi, I., Yokelson, R. J., Ward, D. E. et€al. (2003). Trace gas and particle emissions from fires in large diameter and belowground biomass fuels. Journal of Geophysical Research – Atmospheres, 108. Blando, J. D. and Turpin, B. J. (2000). Secondary organic aerosol formation in cloud and fog droplets:€a literature evaluation of plausibility. Atmospheric Environment, 34, 1623–1632. Bouwman, A. F., Lee, D. S., Asman, W. A. H. et al. (1997). A global high-resolution emission inventory for ammonia. Global Biogeochemical Cycles, 11, 561–587. Brook, J. R., Zhang, L. M., Di Giovanni, F. and Padro, J. (1999). Description and evaluation of a model of deposition velocities for routine estimates of air pollutant dry deposition over North America. Part I:€Model development. Atmospheric Environment, 33, 5037–5051. Businger, J. A. (1982). The fluxes of specific enthalpy, sensible heat and latent-heat near the Earth’s surface. Journal of the Atmospheric Sciences, 39, 1889–1892. Businger, J. A., Wyngaard, J. C., Izumi, Y. and Bradley, E. F. (1971). Flux-profile relationships in atmospheric surface layer. Journal of the Atmospheric Sciences, 28, 181–190. Butterbach-Bahl, K., Kesik, M., Miehle, P., Papen, H. and Li, C. (2004). Quantifying the regional source strength of N-trace gases across agricultural and forest ecosystems with process based models. Plant and Soil, 260, 311–329. Calderon, S. M., Poor, N. D. and Campbell, S. W. (2007). Estimation of the particle and gas scavenging contributions to wet deposition of organic nitrogen. Atmospheric Environment, 41, 4281–4290. Cape, J. N., Storetonwest, R. L., Devine, S. F., Beatty, R. N. and Murdoch, A. (1993). The reaction of nitrogen-dioxide at low concentrations with natural-waters. Atmospheric Environment Part A – General Topics, 27, 2613–2621. Cape, J. N., Kirika, A., Rowland, A. P. et€al. (2001). Organic nitrogen in precipitation:€real problem or sampling artifact? The Scientific World, 1, 230–237. Cape, J. N., Tang, Y. S., van Dijk, N. et al. (2004). Concentrations of ammonia and nitrogen dioxide at roadside verges, and their contribution to nitrogen deposition. Environmental Pollution, 132, 469–478. Carruthers, D. J. and Choularton, T. W. (1983). A model of the feeder seeder mechanism of orographic rain including stratification and wind-drift effects. Quarterly Journal of the Royal Meteorological Society, 109, 575–588. Carslaw, D. C., Beevers, S. D. and Bell, M. C. (2007). Risks of exceeding the hourly EU limit value for nitrogen dioxide resulting from increased road transport emissions of primary nitrogen dioxide. Atmospheric Environment, 41, 2073–2082.

202

Cicerone, R. J. and Zellner, R. (1983). The atmospheric chemistry of hydrogen-cyanide (HCN). Journal of Geophysical Research – Oceans and Atmospheres, 88, 689–696. Clapp, L. J. and Jenkin, M. E. (2001). Analysis of the relationship between ambient levels of O3, NO2 and NO as a function of NOx in the UK. Atmospheric Environment, 35, 6391–6405. COPERT (2010). http://lat.eng.auth.gr/copert/ Cornell, S., Rendell, A. and Jickells, T. (1995). Atmospheric inputs of dissolved organic nitrogen to the oceans. Nature, 376, 243–246. Cornell, S. E., Jickells, T. D. and Thornton, C. A. (1998). Urea in rainwater and atmospheric aerosol. Atmospheric Environment, 32, 1903–1910. Cornell, S., Mace, K., Coeppicus, S. et al. (2001). Organic nitrogen in Hawaiian rain and aerosol. Journal of Geophysical Research – Atmospheres, 106, 7973–7983. Cornell, S. E., Jickells, T. D., Cape, J. N., Rowland, A. P. and Duce, R. A. (2003). Organic nitrogen deposition on land and coastal environments:€a review of methods and data. Atmospheric Environment, 37, 2173–2191. Daumer, B., Niessner, R. and Klockow, D. (1992). Laboratory studies of the influence of thin organic films on the neutralization reaction of H2SO4 aerosol with ammonia. Journal of Aerosol Science, 23, 315–325. Davidson, E. A. and Kingerlee, W. (1997). A global inventory of nitric oxide emissions from soils. Nutrient Cycling in Agroecosystems, 48, 37–50. Derwent, R. and Hertel, O. (1998). Transformation of air pollutants. In:€Urban Air Pollution: European Aspects, ed. J. Fenger, O. Hertel and F. Palmster, Kluwer Academic Publishers, Dordrecht. Dickerson, R. R., Stedman, D. H. and Delany, A. C. (1982). Direct measurements of ozone and nitrogen dioxide photolysis rates in the troposphere. Journal of Geophysical Research – Atmospheres, 78, 4933–4946. Dore, A. J., Choularton, T. W., Fowler, D. and Stortonwest, R. (1990). Field-measurements of wet deposition in an extended region of complex topography. Quarterly Journal of the Royal Meteorological Society, 116, 1193–1212. EANET (2010). www.eanet.cc EDGAR (2010). http://edgar.jrc.ec.europa.eu Ellermann, T., Hertel, O., Monies, C., Kemp, K. and Skjøth, C. A. (2004). Atmospheric Deposition 2003, NOVA 2003. (In Danish:€Amosfærisk deposition 2003, NOVA 2003.) Danish Ministry of the Environment, National Environmental Research Institute, Roskilde, Denmark. Elzing, A. and Monteny, G. J. (1997). Ammonia emission in a scale model of a dairy-cow house. Transactions of the ASAE, 40, 713–720. EMEP (2010). http://www.emep.int Erisman, J. W. and Draaijers, G. (1995). Atmospheric Deposition in Relation to Acidification and Eutrophication, Elsevier, Amsterdam. Erisman, J. W., Draaijers, G., Duyzer, J. et al. (1997). Particle deposition to forests: summary of results and application. Atmospheric Environment, 31, 321–332. Eurotrac (2010). http://www.gsf.de/eurotrac Fagerli, H. and Aas, W. (2008). Trends of nitrogen in air and precipitation:€model results and observations at EMEP site in€Europe, 1980–2003. Environmental Pollution, 154(3), 448–461. Fagerli, H., Hertel, O., Skjøth, C. A., Gyldenkærne, S. and Hutchings, N. (2007). Implementation of a dynamical ammonia emission

Ole Hertel parameterization in the regional Unified EMEP chemical transport model Costa do Sauípe, Bahia, in Northeast Brazil. Famulari, D., Fowler, D., Nemitz, E. et al. (2010). Development of a low-cost system for measuring conditional time-averaged gradients of SO2 and NH3. Environmental Monitoring and Assessment, 161, 11–27. Farmer, D. K., Wooldridge, P. J. and Cohen, R. C. (2006). Application of thermal-dissociation laser induced fluorescence (TD-LIF) to measurement of HNO3, Sigma alkyl nitrates, Sigma peroxy nitrates, and NO2 fluxes using eddy covariance. Atmospheric Chemistry and Physics, 6, 3471–3486. Farquhar, G. D., Firth, P. M., Wetselaar, R. and Weir, B. (1980). On the gaseous exchange of ammonia between leaves and the environment: determination of the ammonia compensation point. Plant Physiology, 66, 710–714. Finlayson-Pitts, B. J. and Pitts, J. N., Jr. (1986). Atmospheric Chemistry:€Fundamentals and Experimental Techniques. Wiley & Sons, New York. Flechard, C. R., Fowler, D., Sutton, M. A. and Cape, J. N. (1999). A€dynamic chemical model of bi-directional ammonia exchange€between semi-natural vegetation and the atmosphere. Quarterly Journal of the Royal Meteorological Society, 125, 2611–2641. Flechard, C. R., Ambus, P., Skiba, U. et€al. (2007). Effects of climate and management intensity on nitrous oxide emissions in grassland systems across Europe. Agriculture Ecosystems and Environment, 121, 135–152. Fowler, D., Duyzer, J. H. and Baldocchi, D. D. (1991). Inputs of trace gases, particles and cloud droplets to terrestrial surfaces. Proceedings of the Royal Society of Edinburgh Section B – Biological Sciences, 97, 35–59. Fowler, D., Flechard, C., Cape, J. N., Storeton-West, R. L. and Coyle, M. (2001). Measurements of ozone deposition to vegetation quantifying the flux, the stomatal and non-stomatal components. Water, Air and Soil Pollution, 130, 63–74. Fowler, D. and Battarbee, R. , (2005). Climate change and pollution in the mountains: The nature of change. In: Mountains of Northern Europe: Conservation Management, People and Nature, ed. Thompson, D. E. B. A., Price, M. F. and Galbraith, C. A. Edinburgh, UK: Scottish Natural Heritage. Fowler, D., Pilegaard, K., Sutton, M. A. et€al. (2009). Atmospheric composition change: Ecosystems-Atmosphere Interactions. Atmospheric Environment, 43, 5193–5267. Friedrich, R. (2002). Anthropogenic Emissions, ed. P. Midgley and M.€Reuther, Backhuys Publishers, Leiden, The Netherlands. GEIA (2010). http://www.geiacenter.org Gibb, S. W., Mantoura, R. F. C. and Liss, P. S. (1999). Oceanatmosphere exchange and atmospheric speciation of ammonia and methylamines in the region of the NW Arabian Sea. Global Biogeochemical Cycles, 13, 161–177. Goldsmith, P., Delafield, H. J. and Cox, L. C. (1963). The role of diffusiophoresis in the scavenging of radioactive particles from the atmosphere. Quarterly Journal of the Royal Meteorological Society, 89, 43–50. Gore, M. L., Aneja, V., Cooter, E. J., Dennis, R. and Pleim, J. E. (2009). Ammonia Emissions in the U.S.:€Assessing the Role of Bi-directional Ammonia Transport using the Community Multi-scale Air Quality€(CMAQ) Model. Centre for Ecology and Hydrology, Edinburgh, UK. Gorzelska, K., Galloway, J. N., Watterson, K. and Keene, W. C. (1992). Water-soluble primary amine compounds in rural continental

precipitation. Atmospheric Environment Part A – General Topics, 26, 1005–1018. Gorzelska, K., Talbot, R. W., Klemm, K. et€al. (1994). Chemicalcomposition of the atmospheric aerosol in the troposphere over the Hudson-Bay Lowlands and Quebec-Labrador regions of Canada. Journal of Geophysical Research – Atmospheres, 99, 1763–1779. Grant, R. and Blicher-Mathiesen, G. (2004). Danish policy measures to reduce diffuse nitrogen emissions from agriculture to the aquatic environment. Water Science and Technology, 49, 91–99. Gronberg, L., Lovkvist, P. and Jonsson, J. A. (1992). Measurement of aliphatic-amines in ambient air and rainwater. Chemosphere, 24, 1533–1540. Gundel, L. A., Daisey, J. M., Decarvalho, L. R. F., Kado, N. Y. and Schuetzle, D. (1993). Polar organic-matter in airborne particles: chemical characterization and mutagenic activity. Environmental Science and Technology, 27, 2112–2119. Gupta, A., Tang, D. and McMurry, P. H. (1995). Growth of monodisperse, submicron aerosol-particles exposed to SO2, H2O2, and NH3. Journal of Atmospheric Chemistry, 20, 117–139. Gyldenkærne, S., Skjøth, C. A., Hertel, O. and Ellermann, T. (2005). A dynamical ammonia emission parameterization for use in air pollution models. Journal of Geophysical Research – Atmospheres, 110. Hall, S. J., Matson, P. A. and Roth, P. M. (1996). NOx emissions from soil:€implications for air quality modeling in agricultural regions. Annual Review of Energy and the Environment, 21, 311–346. Hampson, R. F. and Gavin, D. (1978). Reaction Rate and Photochemical Data for Atmospheric Chemistry€– 1977. National Bureau of Standards, US Government, Washington DC. Harris, G. W., Klemp, D. and Zenker, T. (1992). An upper limit on the HCl near-surface mixing-ratio over the Atlantic measured using Tdlas. Journal of Atmospheric Chemistry, 15, 327–332. Harrison, R. M. and Kitto, A. M. N. (1994). Evidence for a surface source of atmospheric nitrous-acid. Atmospheric Environment, 28, 1089–1094. Hatton, A. D. and Gibb, S. W. (1999). A technique for the determination of trimethylamine-N-oxide in natural waters and biological media. Analytical Chemistry, 71, 4886–4891. Hertel, O. and Brandt, J. (2009). Air pollution modelling. In: Air Pollution: From Local to Global Importance, ed. J. Fenger and J. C. Tjell, RSC Publishing and Polyteknisk Forlag, Copenhagen, pp. 293–326. Hertel, O., Skjøth, C. A., Frohn, L. M. et€al. (2002). Assessment of the atmospheric nitrogen and sulphur inputs into the North Sea using a Lagrangian model. Physics and Chemistry of the Earth, 27, 1507–1515. Hertel, O., Skjøth, C. A., Løfstrøm, P. et€al. (2006). Modelling nitrogen deposition on a local scale: a review of the current state of the art. Environmental Chemistry, 3, 317–337. Hicks, B. B. and Garland, J. A. (1983). Overview and suggestions for further research on dry deposition. In:€Precipitation Scavenging, Dry Deposition, and Resuspension, ed. H. R. Pruppacher, R.€G.€Semonin and W. G. N. Slinn, Elsevier, New York. Hicks, B. B., Baldocchi, D. D., Meyers, T. P., Hosker, R. P. and Matt, D. R. (1987). A preliminary multiple resistance routine for deriving dry deposition velocities from measured quantities. Water, Air and Soil Pollution, 36, 311–330. Hill, K. A., Shepson, P. B., Galbavy, E. S. et€al. ( 2007). Processing of atmospheric nitrogen by clouds above a forest environment. Journal of Geophysical Research – Atmospheres, 112.

203

Nitrogen processes in the atmosphere Holland, E. A., Dentener, F. J., Braswell, B. H. and Sulzman, J. M. (1999). Contemporary and pre-industrial global reactive nitrogen budgets. Biogeochemistry, 46, 7–43. Holzinger, R., Warneke, C., Hansel, A. et€al. (1999). Biomass burning as a source of formaldehyde, acetaldehyde, methanol, acetone, acetonitrile, and hydrogen cyanide. Geophysical Research Letters, 26, 1161–1164. Holzinger, R., Jordan, A., Hansel, A. and Lindinger, W. (2001). Automobile emissions of acetonitrile:€assessment of its contribution to the global source. Journal of Atmospheric Chemistry, 38, 187–193. Horii, C. V., Munger, J. W., Wofsy, S. C. et€al. (2004). Fluxes of nitrogen oxides over a temperate deciduous forest. Journal of Geophysical Research – Atmospheres, 109. Hov, Ø., Hjøllo, B. A. and Eliassen, A. (1994). Transport distance of ammonia and ammonium in Northern Europe. 1. Model description. Journal of Geophysical Research – Atmospheres, 99, 18735–18748. Huntzicker, J. J., Cary, R. A. and Ling, C. S. (1980). Neutralization of sulfuric-acid aerosol by ammonia. Environmental Science and Technology, 14, 819–824. Husted, S., Schjørring, J. K., Nielsen, K. H., Nemitz, E. and Sutton, M. A. (2000). Stomatal compensation points for ammonia in oilseed rape plants under field conditions. Agricultural and Forest Meteorology, 105, 371–383. IGAC (2000). IGAC activities newsletter no. 22, December 2000. Jarvis, S., Hutchings, N., Brentrup, F., Olesen, J. and van der Hoek, K. (2011). Nitrogen flows in farming systems across Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman, et€al., Cambridge University Press. Jensen, S. S., Hvidberg, M., Petersen, J. et€al. (2008). GIS-based National Road and Traffic Database 1960–2005. (In Danish:€GISbaseret national vej- og trafikdatabase 1960–2005.) National Environmental Research Institute, University of Aarhus, Roskilde, Denmark. Joffre, S. (1988). Modelling dry deposition of soluble gases to the sea surface. Atmospheric Environment, 22, 1137–1146. Junge, E. and Ryan, T. G. (1958). Study of the SO2 oxidation in solution and its role in atmospheric chemistry. Quarterly Journal of the Royal Meteorological Society, 84, 46–55. Kallinger, G. and Niessner, R. (1999). Laboratory investigation of annular denuders as sampling system for the determination of aliphatic primary and secondary amines in stack gas. Mikrochimica Acta, 130, 309–316. Karl, T., Potosnak, M., Guenther, A. et€al. (2004). Exchange processes of volatile organic compounds above a tropical rain forest:€implications for modeling tropospheric chemistry above dense vegetation. Journal of Geophysical Research – Atmospheres, 109. Kieber, R. J., Long, M. S. and Willey, J. D. (2005). Factors influencing nitrogen speciation in coastal rainwater. Journal of Atmospheric Chemistry, 52, 81–99. Koerkamp, P. W. G. G., Metz, J. H. M., Uenk, G. H. et€al. (1998). Concentrations and emissions of ammonia in livestock buildings in Northern Europe. Journal of Agricultural Engineering Research, 70, 79–95. Krab, E. J., Cornelissen, J. H. C., Lang, S. I. and van Logtestijn, R. S. P. (2008). Amino acid uptake among wide-ranging moss species may contribute to their strong position in higher-latitude ecosystems. Plant and Soil, 304, 199–208.

204

Kramm, G. (1989). A numerical-method for determining the dry deposition of atmospheric trace gases. Boundary-Layer Meteorology, 48, 157–175. Kramm, G. and Dlugi, R. (1994). Modeling of the vertical fluxes of nitric-acid, ammonia, and ammonium-nitrate. Journal of Atmospheric Chemistry, 18, 319–357. Kramm, G., Muller, H., Fowler, D. et€al. (1991). A modified profile method for determining the vertical fluxes of NO, NO2, ozone, and HNO3 in the atmospheric surface-layer. Journal of Atmospheric Chemistry, 13, 265–288. Lambrecht, U. (2007). Legislation and future requirements for NOx reduction. Frankfurt, Germany. Larsson, L., Ferm, M., Kasimir-Klemedtsson, A. and Klemedtsson, L. (1998). Ammonia and nitrous oxide emissions from grass and alfalfa mulches. Nutrient Cycling in Agroecosystems, 51, 41–46. Lee, D. S., Halliwell, C., Garland, J. A., Dollard, G. J. and Kingdon, R. D. (1998). Exchange of ammonia at the sea surface: a preliminary study. Atmospheric Environment, 32, 431–439. Li, C. Z. and Tan, L. L. (2000). Formation of NOx and SOx â•›precursors during the pyrolysis of coal and biomass. Part III: Further discussion on the formation of HCN and NH3 during pyrolysis. Fuel, 79, 1899–1906. Lipson, D. and Nasholm, T. (2001). The unexpected versatility of plants:€organic nitrogen use and availability in terrestrial ecosystems. Oecologia, 128, 305–316. Loubet, B., Milford, C., Sutton, M. A. and Cellier, P. (2001). Investigation of the interaction between sources and sinks of atmospheric ammonia in an upland landscape using a simplified dispersion-exchange model. Journal of Geophysical Research – Atmospheres, 106, 24183–24195. Mace, K. A. and Duce, R. A. (2002). Determination of urea in atmospheric aerosols and natural waters: a cation exchange method. International Journal of Environmental Analytical Chemistry, 82, 341–352. Mace, K. A., Artaxo, P. and Duce, R. A. (2003a). Water-soluble organic nitrogen in Amazon Basin aerosols during the dry (biomass burning) and wet seasons. Journal of Geophysical Research – Atmospheres, 108. Mace, K. A., Duce, R. A. and Tindale, N. W. (2003b). Organic nitrogen in rain and aerosol at Cape Grim, Tasmania, Australia. Journal of Geophysical Research – Atmospheres, 108. Mace, K. A., Kubilay, N. and Duce, R. A. (2003c). Organic nitrogen in rain and aerosol in the eastern Mediterranean atmosphere:€an association with atmospheric dust. Journal of Geophysical Research€– Atmospheres, 108. Mainz (2010). www.mpch-mainz.mpg.de/~sander/res/henry.html Massad, R. S., Nemitz, E. and Sutton, M. A. (2010a). Review and parameterisation of bi-directional ammonia exchange between vegetation and the atmosphere. Atmospheric Chemistry and Physics Discussions, 10, 10335–10398. Massad, R. S., Tuzet, A., Loubet, B., Perrier, A. and Cellier, P. (2010b). Model of stomatal ammonia compensation point (STAMP) in relation to the plant nitrogen and carbon metabolisms and environmental conditions. Ecological Modelling, 221, 479–494. Matsumoto, K. and Uematsu, M. (2005). Free amino acids in marine aerosols over the western North Pacific Ocean. Atmospheric Environment, 39, 2163–2170. McGregor, K. G. and Anastasio, C. (2001). Chemistry of fog waters in California’s Central Valley Part II: Photochemical transformations of amino acids and alkyl amines. Atmospheric Environment, 35, 1091–1104.

Ole Hertel Mckay, H. A. C. (1971). Atmospheric oxidation of sulphur dioxide in water droplets in presence of ammonia. Atmospheric Environment, 5, 7. McMurry, P. H., Takano, H. and Anderson, G. R. (1983). Study of the ammonia (gas) sulfuric-acid (aerosol) reaction-rate. Environmental Science and Technology, 17, 347–352. Mikkelsen, M. H., Gyldenkærne, S., Poulsen, S. D., Olsen, J. E. and Sommer, S. G. (2005). Inventory and Calculation Method for Agricultural Emissions of Ammonia and Greenhouse Gases€in the Time Period 1985–2002. (In Danish:€Opgørelse og beregningsmetode for landbrugets emissioner af ammoniak og drivhusgasser€– i perioden 1985–2002.) National Environmental Research Institute, University of Aarhus, Denmark, Roskilde, Denmark. Milne, P. J. and Zika, R. G. (1993). Amino-acid nitrogen in atmospheric aerosols: occurrence, sources and photochemical modification. Journal of Atmospheric Chemistry, 16, 361–398. Moldanová, J., Grennfelt, P., Jonsson, Å . et€al. (2011). Nitrogen as a threat to European air quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Monteith, J. and Unsworth, M. (2008). Principles of Environmental Physics, 3rd edn, Elsevier, New York. Muck, R. E. and Steenhuis, T. S. (1982). Nitrogen losses from manure storages. Agricultural Wastes, 4, 41–54. NADP (2010). http://nadp.sws.uiuc.edu/ NatAir (2010). http://natair.ier.uni-stuttgart.de/ Neff, J. C., Holland, E. A., Dentener, F. J., McDowell, W. H. and Russell, K. M. (2002). The origin, composition and rates of organic nitrogen deposition:€a missing piece of the nitrogen cycle? Biogeochemistry, 57, 99–136. Neftel, A., Blatter, A., Hesterberg, R. and Staffelbach, T. (1996). Measurements of concentration gradients of HNO2 and HNO3 over a semi-natural ecosystem. Atmospheric Environment, 30, 3017–3025. Nemitz, E., Sutton, M. A., Wyers, G. P. and Jongejan, P. A. C. (2004). Gas-particle interactions above a Dutch heathland:€I. Surface exchange fluxes of NH3, SO2, HNO3 and HCl. Atmospheric Chemistry and Physics, 4, 989–1005. NEU (2010). http://www.nitroeurope.eu Notholt, J., Hjorth, J. and Raes, F. (1992). Formation of HNO2 on aerosol surfaces during foggy periods in the presence of NO and NO2. Atmospheric Environment, 26A, 2111–2117. Olesen, J. E. and Sommer, S. G. (1993). Modeling effects of windspeed and surface cover on ammonia volatilization from stored pig slurry. Atmospheric Environment Part A – General Topics, 27, 2567–2574. OSPM (2010). ospm.dmu.dk Ottley, C. J. and Harrison, R. M. (1992). The spatial-distribution and particle-size of some inorganic nitrogen, sulfur and chlorine species over the North-Sea. Atmospheric Environment Part A€– General Topics, 26, 1689–1699. Oudendag, D. A. and Luesink, H. H. (1998). The manure model: manure, minerals (N, P and K), ammonia emission, heavy metals and the use of fertiliser in Dutch agriculture. Environmental Pollution, 102, 241–246. Paerl, H. W. and Whitall, D. R. (1999). Anthropogenically-derived atmospheric nitrogen deposition, marine eutrophication and harmful algal bloom expansion:€is there a link? Ambio, 28, 307–311.

Palmgren, F., Berkowicz, R., Hertel, O. and Vignati, E. (1996). Effects of reduction of NOx on the NO2 levels in urban streets. Science of the Total Environment, 190, 409–415. Palmiotto, G., Pieraccini, G., Moneti, G. and Dolara, P. (2001). Determination of the levels of aromatic amines in indoor and outdoor air in Italy. Chemosphere, 43, 355–361. Pilegaard, K., Skiba, U., Ambus, P. et€al. (2006). Nitrogen load and forest type determine the soil emission of nitrogen oxides (NO and N2O). Biogeosciences, 3, 651–661. Pinder, R. W., Strader, R., Davidson, C. I. and Adams, P. J. (2004). A temporally and spatially resolved ammonia emission inventory for dairy cows in the United States. Atmospheric Environment, 38, 3747–3756. Pinder, R. W., Adams, P. J. and Pandis, S. N. (2007). Ammonia emission controls as a cost-effective strategy for reducing atmospheric particulate matter in the eastern United States. Environmental Science and Technology, 41, 380–386. Pio, C. A. and Harrison, R. M. (1987a). The equilibrium of ammonium-chloride aerosol with gaseous hydrochloric-acid and ammonia under tropospheric conditions. Atmospheric Environment, 21, 1243–1246. Pio, C. A. and Harrison, R. M. (1987b), Vapor-pressure of ammonium-chloride aerosol: effect of temperature and humidity. Atmospheric Environment, 21, 2711–2715. Pruppacher, H. R. and Jaenicke, R. (1995). The processing of watervapor and aerosols by atmospheric clouds, a global estimate. Atmospheric Research, 38, 283–295. Qualls, R. and Haines, B. L. (1992). Measuring adsorption-isotherms using continuous, unsaturated flow through intact soil cores. Soil Science Society of America Journal, 56, 456–460. Quinn, P. K., Charlson, R. J. and Bates, T. S. (1988). Simultaneous observations of ammonia in the atmosphere and ocean. Nature, 335, 336–338. RAINS (2010). http://www.iiasa.ac.at/rains/index.html Riedo, M., Milford, C., Schmid, M. and Sutton, M. A. (2002). Coupling soil–plant–atmosphere exchange of ammonia with ecosystem functioning in grasslands. Ecological Modelling, 158, 83–110. Ruijgrok, W., Tieben, H. and Eisinga, P. (1997). The dry deposition of particles to a forest canopy:€a comparison of model and experimental results. Atmospheric Environment, 31, 399–415. Saxena, P. and Hildemann, L. M. (1996). Water-soluble organics in atmospheric particles:€a critical review of the literature and application of thermodynamics to identify candidate compounds. Journal of Atmospheric Chemistry, 24, 57–109. Schade, G. W. and Crutzen, P. J. (1995). Emission of aliphatic-amines from animal husbandry and their reactions: potential source of N2O and HCN. Journal of Atmospheric Chemistry, 22, 319–346. Scheller, E. (2001). Amino acids in dew: origin and seasonal variation. Atmospheric Environment, 35, 2179–2192. Schjørring, J. K., Husted, S. and Poulsen, M. M. (1998). Soil–plant– atmosphere ammonia exchange associated with Calluna vulgaris and Deschampsia flexuosa. Atmospheric Environment, 32, 507–512. Shim, C. S., Wang, Y. H., Singh, H. B., Blake, D. R. and Guenther, A. B. (2007). Source characteristics of oxygenated volatile organic compounds and hydrogen cyanide. Journal of Geophysical Research–Atmospheres, 112. Simpson, D., Aas, W., Bartnicki, J. et€al. (2011). Atmospheric transport and deposition of nitrogen in Europe. In:€The European Nitrogen

205

Nitrogen processes in the atmosphere Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Skiba, U., Fowler, D. and Smith, K. A. (1997). Nitric oxide emissions from agricultural soils in temperate and tropical climates:€sources, controls and mitigation options. Nutrient Cycling in Agroecosystems, 48, 139–153. Skiba, U., Nemitz, E. and Flechard, C. (2007). Land atmosphere exchange of reactive nitrogen. In: Proceedings of the ACCENT Workshop, Urbino, Italy, 23–27 July 2007. Skjøth, C. A., Hertel, O., Gyldenkærne, S. and Ellermann, T. (2004). Implementing a dynamical ammonia emission parameterization in the large-scale air pollution model ACDEP. Journal of Geophysical Research€– Atmospheres, 109. Skjøth, C. A., Gyldenkærne, S., Ellermann, T., Mikkelsen, M. H. and Hertel, O. ( 2008). Footprints on ammonia concentrations from emission regulation. Journal of Air and Waster Management, 58, 1158–1165. Slinn, W. G. N. (1982). Predictions for particle deposition to vegetative canopies. Atmospheric Environment, 16, 1785–1794. Sommer, S. G., Østergård, H. S., Løfstrøm, P., Andersen, H. V. and Jensen, L. S. (2009). Validation of model calculation of ammonia deposition in the neighbourhood of a poultry farm using measured NH3 concentrations and N deposition. Atmospheric Environment, 43, 915–920. Sørensen, L. L., Hertel, O., Skjøth, C. A., Lund, M. and Pedersen, B. (2003). Fluxes of ammonia in the coastal marine boundary layer. Atmospheric Environment, 37, S167–S177. Sørensen, L. L., Pryor, S. C., de Leeuw, G. and Schulz, M. (2005). Flux divergence of nitric acid in the marine atmospheric surface layer. Journal of Geophysical Research – Atmospheres, 110. Sorteberg, A. and Hov, Ø . (1996). Two parametrizations of the dry deposition exchange for SO2 and NH3 in a numerical model. Atmospheric Environment, 30, 1823–1840. Spokes, L., Jickells, T., Weston, K. et€al. (2006). MEAD:€An interdisciplinary study of the marine effects of atmospheric deposition in the Kattegat. Environmental Pollution, 140, 453–462. Sprung, D., Jost, C., Reiner, T., Hansel, A. and Wisthaler, A. (2001). Acetone and acetonitrile in the tropical Indian Ocean boundary layer and free troposphere:€aircraft-based intercomparison of AP-CIMS and PTR-MS measurements. Journal of Geophysical Research€– Atmospheres, 106, 28511–28527. Stelson, A. W., Friedlander, S. K. and Seinfeld, J. H. (1979). Note on the equilibrium relationship between ammonia and nitric-acid and particulate ammonium-nitrate. Atmospheric Environment, 13, 369–371. Stelson, A. W. and Seinfeld, J. H. (1982). Thermodynamic prediction of the water activity, NH4NO3 dissociation-constant, density and refractive-index for the NH4NO3–(NH4)2SO4–H2O system at 25-Degrees-C. Atmospheric Environment, 16, 2507–2514. Stemmler, K., Ammann, M., Donders, C., Kleffmann, J. and George, C. (2006). Photosensitized reduction of nitrogen dioxide on humic acid as a source of nitrous acid. Nature, 440, 195–198. Stohl, A., Williams, E., Wotawa, G. and KrompKolb, H. (1996). A European inventory of soil nitric oxide emissions and the effect of these emissions on the photochemical formation of ozone. Atmospheric Environment, 30, 3741–3755. Sutton, M. A., Burkhardt, J. K., Guerin, D., Nemitz, E. and Fowler, D. (1998). Development of resistance models to describe

206

measurements of bi-directional ammonia surface-atmosphere exchange. Atmospheric Environment, 32, 473–480. Sutton, M. A., Fowler, D. and Moncrieff, J. B. (1993). The exchange of atmospheric ammonia with vegetated surfaces. 1. Unfertilized vegetation. Quarterly Journal of the Royal Meteorological Society, 119, 1023–1045. Sutton, M. A., Dragosits, U., Tang, Y. S. and Fowler, D. (2000). Ammonia emissions from non-agricultural sources in the UK. Atmospheric Environment, 34, 855–869. Tang, Y. S., Simmons, I., van Dijk, N. et€al. (2009). European scale application of atmospheric reactive nitrogen measurements in a low-cost approach to infer dry deposition fluxes. Agriculture, Ecosystems and Environment, 133, 183–195. Thoene, B., Schroder, P., Papen, H., Egger, A. and Rennenberg, H. (1991). Absorption of Atmospheric NO2 by Spruce (Picea-abies€l Karst) Trees. Part I: NO2 influx and its correlation with nitrate reduction. New Phytologist, 117, 575–585. Trebs, I., Lara, L. L., Zeri, L. M. M. et€al. (2006). Dry and wet deposition of inorganic nitrogen compounds to a tropical pasture site (Rondonia, Brazil). Atmospheric Chemistry and Physics, 6, 447–469. Tuazon, E. C., Atkinson, R., Aschmann, S. M. and Arey, J. (1994). Kinetics and products of the gas-phase reactions of O3 with amines and related-compounds. Research on Chemical Intermediates, 20, 303–320. Turnipseed, A. A., Huey, L. G., Nemitz, E. et€al. (2006). Eddy covariance fluxes of peroxyacetyl nitrates (PANs) and NOy to a coniferous forest. Journal of Geophysical Research – Atmospheres, 111. Twigg, M. (2006). Application of tunable diode laser absorption spectroscopy for the investigation of surface/atmosphere exchange of ammonia. Ph.D. thesis, University of Ulster. Van Hove, L. W. A., Adema, E. H., Vredenberg, W. J. and Pieters, G. A. (1989). A study of the adsorption of NH3 and SO2 on leaf surfaces. Atmospheric Environment, 23, 1479–1486. Van Jaarsveld, H. (2004). The Operational Priority Substances Model:€Description and Validation of OPS-Pro 4.1. National Institute for Public Health and Environment, Bilthoven, The Netherlands. Vestreng, V., Nziachristos, L., Semb, A., Isaksen, I. S. A. and Tarrason, L. (2008). Evolution of NOx emissions in Europe with focus on road transport control measures. Atmospheric Chemistry and Physics Discussions, 8, 10697–10747. Walcek, C. J., Brost, R. A., Chang, J. S. and Wesely, M. L. (1986). SO2, sulfate and HNO3 deposition velocities computed using regional land-use and meteorological data. Atmospheric Environment, 20, 949–964. Wall, S. M., John, W. and Ondo, J. L. (1988). Measurements of aerosol size distributions for nitrate and major ionic speices. Atmospheric Environment, 22, 1649–1656. Warneke, C., van der Veen, C., Luxembourg, S., de Gouw, J. A. and Kok, A. (2001). Measurements of benzene and toluene in ambient air using proton-transfer-reaction mass spectrometry:€calibration, humidity dependence, and field intercomparison. International Journal of Mass Spectrometry, 207, 167–182. Wesely, M. L., Cook, D. R., Hart, R. L. and Speer, R. E. (1985). Measurements and parameterization of particulate sulfur dry deposition over grass. Journal of Geophysical Research – Atmospheres, 90, 2131–2143. Wolfe, G. M., Thornton, J. A., Yatavelli, R. L. N. et€al. (2009). Eddy covariance fluxes of acyl peroxy nitrates (PAN, PPN and MPAN) above a Ponderosa pine forest. Atmospheric Chemistry and Physics, 9, 615–634.

Ole Hertel Zhang, L., Wright, L. P. and Asman, W. A. H. (2010). Bi-directional air-surface exchange of atmospheric ammonia€– a review of measurements and a development of a big-leaf model for applications in regional-scale air-quality models. Journal of Geophysical Research – Atmospheres. Zhang, Q. and Anastasio, C. (2001). Chemistry of fog waters in California’s Central Valley€– Part 3:€concentrations and speciation of organic and inorganic nitrogen. Atmospheric Environment, 35, 5629–5643. Zhang, Q. and Anestasio, C. (2003). Free and combined amino compounds in atmospheric fine particles (PM2.5) and fog

waters from Northern California. Atmospheric Environment, 37, 2247–2258. Zhang, R. H., Day, D. L., Christianson, L. L. and Jepson, W. P. (1994). A computer-model for predicting ammonia release rates from swine manure pits. Journal of Agricultural Engineering Research, 58, 223–229. Zhang, Y. H., ten Brink, H., Slanina, J. and Wyers, G. P. (1995). The influence of ammonium nitrate equilibrium on the measurement of exchange fluxes of ammonia and nitric acid. In:€Acid Rain Research:€Do We Have Enough Answers?, ed. G. J. Heij and J.€W.€Erisman, Elsevier Science, Amsterdam.

207

Part

III

Nitrogen flows and fate at multiple spatial scales

Chapter

10

Nitrogen flows in farming systems across Europe Lead author: Steve Jarvis Contributing authors: Nick Hutchings, Frank Brentrup, Jorgen Eivind Olesen and Klaas W. van de Hoek

Executive summary Nature of the issue • Farms represent operational units which determine N-use efficiency and incorporation into products and, collectively, at the wider scale, determine the extent of environmental losses from agriculture. • The basic principles and objectives of using N, from whatever source, pertain to different systems across the wide range of farming types across Europe. • In addition to managing external inputs (fertilisers), there is much opportunity to improve N transfers within the farm. Mineral fertilisers are added to balance supply/demand for crops. Some systems rely on legume-N which, once incorporated into farm cycles, behaves in the same way as other N forms.

Approaches • Farm N cycles, their constituent parts and controlling influences are described and generalised principles identified. • Farm budgets for a range of systems, focussing on typical practice in NW Europe are shown which illustrate some general, important differences between farming systems.

Key findings/state of knowledge • Benefits of using N effectively are far reaching with immediate impact in promoting production. Use of N also provides an effective and flexible management tool for farmers. • Crop N requirements are determined from response curves and economic optima. Advice is supplied to farmers from various sources but the extent to which it is taken depends on many factors. New technologies are available to improve N-use efficiency. The basis of good N management is to optimise efficiency of added and soil N by increasing the temporal and spatial coincidence between availability and uptake of N. • Current management drivers often cause farms to be ‘open’ with N losses. By changing focus from productivity-only to balances between productivity, product quality and environmental impact, managements can be redesigned to increase N use efficiency. • Livestock farming presents particular problems with large potential N losses. Previously, animal manure was considered as a waste product rather than a nutrient source. • Farm-based budgets are a simple way of representing gross flows of N into and from farms and provide important insights into N behaviour. Illustrative budgets show important differences between typical farming systems including conventional arable and livestock (pig, beef and diary) and organic dairy systems in NW Europe in their emissions and the ratio of emissions per unit of N in products.

Major uncertainties/challenges • Nitrogen is mobile and potentially leaky:€it is readily available for farmers to use (at cost) and easy to apply to crops, but requires skilled management. • Technologies to improve efficiency are available, but need continued revision:€farmer knowledge about the requirements for N use from both production and environmental perspectives is increasing, but there is much opportunity to extend this. A major challenge of modern agriculture has been to change perceptions about manure and to demonstrate the value and more efficient use of N from this source.

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

211

Nitrogen flows in farming systems

Recommendations • Continued development of farm-scale models (including simple cycles and budgets) is required for policy but also for farm practice. Improved knowledge of farm budgets, including those from other farm types and regions is required. Continued programmes of providing advice to farmers are required so that new, available technologies are taken up. • Research programmes are required to ensure a sound base on which to develop alternative managements and options to meet future economic and environmental (viz. climate change) challenges.

10.1╇ Background Until about the middle of the twentieth century, the main N input to European farms was via fixation by legumes. This N was made available to non-leguminous crops after the decomposition of plant residues and by the recycling of animal manure. Thereafter, there was a period of 30–40 years during which the importance of manure-N to the farmer was greatly diminished, because of the availability of cheap and reliable synthetic N fertilisers and the increased demand for N created by agronomic practices that increased potential yield. This led to a tendency for manure to be considered as a waste product of animal husbandry, rather than a valuable source of nutrients. Over the past 20 years, the role of manure nitrogen has regained some of its former status and the use of synthetic N fertilisers has been decreasing. There have been three main forces driving this latest development:€the increasing price of synthetic N fertilisers, the introduction of nutrient management legislation (e.g., the EU Nitrates Directive) and a greater awareness amongst farmers and consumers of the nutrient value of manures. Farms represent the operational units at which decisions are made, which have impact on the efficiency of N use and incorporation into products, and collectively at the wider scale, determine the extent of transmissions of excess N into waters or the atmosphere. There is an enormous range across Europe Operational drivers for nutrient use Soil management Crop/forage management Production Animal management

in the ways that farms operate; this is dependent upon the type of production, farm location, soil type, climate and individual farm operational decisions amongst other determining factors. Nevertheless, the basic principles and objectives of using N, from whatever source, pertain to all of these different systems and are shown diagrammatically in Figure€10.1. The decision-making processes that determine N use are complex and highly interactive with both internal and external factors playing a role. This chapter attempts to describe some of the important factors which determine the flows and use of N at the farm scale. The losses of N from natural ecosystems tend to be small, either because the supply of nutrients from natural sources, such as that through biological fixation, is relatively small and/ or because labile forms of N are rapidly captured by the plants present. The purpose of farming is to produce food and fibre, and this is achieved by increasing both the inputs of nutrients and their mobility within the plant/soil ecosystem. Losses of nutrients from agricultural ecosystems will therefore nearly always exceed those from natural ecosystems. However, the extent to which food and fibre production (and income) is ‘Â�traded-off ’ against environmental pollution is much determined by the skill of farmers, and their aims and objectives, as these are influenced by many factors (Figure 10.1) in Figure 10.1 Influences and controls over N flows at the farm scale.

Biological Chemical Physical processes

Influences on management • Legislation • Consumer demand • Climate weather • Farm structure • Farming policies

Housing system

P Profit

Manure management Farm system

F Flexibility S Sustainability

Nutrient balance surplus

Loss

212

Steve Jarvis Animal products

1 Livestock Feed

7

2

5

Livestock

8

Animal housing

12

17

Manure Storage

13

15

4

11 3

6

Fields

Crops 18

19

Manure

14

16 Fertiliser Manure Fixation Seed Atm. dep.

Figure 10.2 Schematic diagram of annual nitrogen flows on a farm. Atm. dep. = deposition from the atmosphere, DON = dissolved organic nitrogen. The numbers refer to the flow or transformation processes described in the text.

NH3, N2, N2O, NO

Bedding

Change in soil N

9

Crop products

NH3 N2 N2O NO

10

NO3– NH4+ DON

controlling nutrients on their farms. However, some loss is inevitable because many forms of N are, or have potential to be, very mobile and the farm is an open system which operates with some unpredictable factors such as rainfall. The farmscale N cycle therefore controls the availability of excess N and contributes to a lesser or greater degree to effects at the larger, landscape or catchment scales within which it is a constituent part. The farm is therefore an important and convenient scale at which to consider effects and impacts, and is the operational scale at which any necessary controls to reduce flows of N to the wider environment are applied practically. The totality of N flows for a farm system is shown in Figure€10.2 in which the numbers refer to the flows or transformations described in the following text. In many situations, there is an import of N from outside livestock farms both in imported animals and imported animal feed (1) and bedding (2). For both livestock and tilled cropping systems, N can be added to the fields from outside the farm in imported manure, mineral fertiliser, and seed and from the atmosphere via atmospheric deposition or nitrogen fixation in legumes (3). There is an export from the farm of crop products such as cereals and straw (4). Nitrogen is also exported from the farm in animal products such as livestock sold or milk (5). Some manure may be exported from the farm (6). The other N removals from the farm are losses as gases to the atmosphere from the components of livestock production (7, 8) and cropped or grazed fields as NH3, N2, N2O or NO (9), or in run-off or leaching as NO3−, NH4+ or dissolved organic N (DON) (10). There may also be direct losses of these forms from animal houses, yards and manure storage areas. The farm N cycle also involves much internal transfer and transformation. Thus in livestock systems, N not incorporated into animal protein or into milk is excreted in dung and urine either on the fields during grazing (11) or in animal housing, animal holding areas and feedlots (12). From there, it is either

applied directly to land or enters the manure management system (13, 14, 15). The other important internal N transfers are the uptake into the crop either to be consumed directly by livestock (16, 17) or into tillage-crop production (18). There are also many internal transfers and transformations in the soil (19) which result in either sequestration into relatively immobile forms or release and transformation into forms that are either available for uptake by plants or further transfer into losses. That which remains in the soil is therefore the net effect of additions made to the soil, and a balance of the net effects of mineralisation, immobilisation, nitrification, denitrification, ammonia volatilisation and plant uptake. The concept of a farm N cycle is inter-connected with that of budgets:€the system shown in Figure 10.2 can, if all the various components can be quantified, be used to provide a systems balance or budget. The detail that a systems balance provides is most relevant to those involved in research and in developing complex models of the N flows. For more practical purposes, soil, livestock and farm gate balances are those which are used. The soil balance represents the net effect of inputs and removals from a specified area, usually a field, and thus enables a prediction of the potential supply of N for future crops and is used for estimating additional supplies required from fertiliser inputs. Both the soil and livestock balances can be used for calculating the N use efficiency of these components, as well as the need for nutrient inputs. The ‘farm gate’ balance is a simplification of the full system balance and simply calculates external inputs from all the sources (flows 1+2+3 in Figure 10.2) and the removal in products (flows 4+5+6 in Figure 10.2). From this information, the surplus or deficiency in the farm system can be determined. This surplus has been used to demonstrate the efficiency of N use in the system and any potential for leakage to the wider environment, i.e. an indicator of pollution potential. The flows of N within a farm scale are controlled by the same processes and transformations as discussed elsewhere

213

Nitrogen flows in farming systems

(de Vries et╯al., 2011, Chapter 15, this volume), but the local, intense concentration of various forms at particular stages of the cycle, means that the equilibrium at these points is pushed towards a faster rate of cycling resulting in a greater probability of loss than in natural ecosystems. This effect is at least in part dependent on the efficiency of utilisation of N in the various components of any farming system (see elsewhere and later in this chapter). However, the livestock component, whether integrated into mixed systems or operating as a separate enterprise, has major effects on environmental consequences. This is largely because their excretal products will almost entirely be returned to land either on the same or another enterprise. The major immediate environmental impact is through NH3 volatilisation and, N2O and NO volatilisation, either directly from urine patches in the field, from excreta in sheds, yards and hard standings, from stored solid and liquid manures, or from their application to land. The NH3 emitted can be deposited back to land either within the same holding or after being transported (sometimes many hundreds of kilometres) away from the farm unit. If the farmer treats the manure as a waste rather than resource, there is a danger that the crops will be over-supplied with N, leading to increases in NO3− leaching. If urea fertilisers are used there will also be volatilisation especially when applied to the surface of the soil without rapid incorporation, in grasslands for example. This may also occur with NH4+-based fertilisers in alkaline soils.

10.2╇ Controlling on-farm N supplies 10.2.1╇ On-farm sources In addition to relying on external inputs from fertilisers etc., there is much opportunity to use the N that is transferred between various internal sources. Some farming systems utilise, and organic systems rely on, N input from legumes, which is acquired through biological fixation. This fixed N, once it has become part of the crop, whether it be grain or forage, then enters the farm cycle in exactly the same way as other forms of N entering the farm. Because this input is usually relatively small, especially when compared with intensive managements with large levels of fertiliser, the losses from legume based systems are often viewed as being smaller than those from fertiliser-based managements. However, this comparison needs to be made on a like-for-like basis when the whole system is considered:€where N inputs of fertiliser and from legume are similar, then losses are also comparable (Hutchings and Kristensen, 1995). Grazed monoculture white clover achieved losses as NO3− and NH3 which were comparable with those from a sward receiving 400 kg N/ha/year (Jarvis et╯al., 1989). However, there is no direct loss, as may sometimes occur when fertilisers are applied and, other than a biological cost to the plant, there are no other costs associated with biologically-fixed N production. Nitrogen fixed in this way is, however, more difficult to control and manage compared with fertiliser N when trying to target N supplies for cropping. The role of the animal and its inefficiency have already been noted. The N contained within manures and slurries is an

214

important resource. However, the composition of these materials is notoriously variable; the N that they contain is there partially in mobile forms, mostly as NH4+ ions, but large proportions are also present in organic forms. The amount of the NH4+ lost as NH3 can be very variable and the rate of decomposition of organic N to plant-available mineral N is also difficult to determine. This makes it much more difficult to use than fertiliser-N in supplying N at the required rates and times to growing crops. The reliability and cheapness of fertiliser-N combined with the growing need to fill the gap between crop needs and other available supplies has led to an enormous increase in its use over recent decades, sometimes completely replacing animal manure, which once was the main fertilizer resource available to farmers. Nevertheless, the realisation that manure-N is an important resource has recently grown again, and farmers have increasingly incorporated a consideration of supplies from this source into their nutrient planning. Some farms have remained almost totally reliant on manures to supply the N requirements of their cropping systems. The other important on-farm N resource is that contained within the soils. With the exception of legumes, crops are dependent upon N present in the mineral N pools as NO3− and /or NH4+ ions. The other major pool within the soil is the organic pool, in fact comprising a number of smaller pools containing materials of different ages and each with a different potential to supply N. Nitrogen can be added (immobilised) to, or be released (mineralised) from, these organic pools:€this depends upon the action of the soil microbial biomass and the resistance of the organic matter to microbial attack. Supplies from soil organic matter are therefore very dependent upon the local environmental conditions (water and temperature especially), the soil texture and the nature of the organic materials that have been added (plant residues, manures and other organic supplements) in the recent or long-term past. Again, supplies are difficult to predict other than in general terms and it is therefore much more difficult to provide an effective index system to define the supply capability for N than it is for P and K:€ this may be even more difficult when manures have been applied over an extended period. However, supply from the soil is very important and, if the contribution it makes is not understood, then it cannot effectively be incorporated in nutrient planning. This introduces a further degree of inefficiency.

10.2.2╇ Balanced N fertiliser supplies One of the keys to successful crop growing is the supply of the correct amounts of nutrients at the correct time in relation to peaks and troughs of crop growth. Where this cannot be achieved, there can, on the one hand, be a deficit of N in relation to demand, and on the other, a surplus. Where the former occurs, growth potential is restricted and when the latter occurs there is potential for loss. Because of the issues noted above, it is more difficult to achieve the correct balance of N supply from all sources than for the other major nutrients. To do so effectively for N requires knowledge of the supplies from all sources to be able to capitalise on these. In the past, in many circumstances and whilst fertiliser prices were relatively

Steve Jarvis

cheap in relation to the gain required, many farmers tended to use more rather than less N than may actually be required. However, perspectives have changed and N use efficiency has improved for many reasons. Crop requirements have, in the main, been determined on the basis of response curves and a defined economic optimum, the point at which the returns in yield/income per unit of fertiliser applied are considered not to be viable. The response curve follows the principle of diminishing returns at this stage. Advice is supplied to farmers by independent advisors, government agencies and through various interactions with other farmers. The extent to which advice is followed depends upon the background (social, educational, peer influence) of the particular farmer, the cost and the income foreseen, and the way that current economic, legislative and other pressures influencing decision making (Figure 10.1). The way that advice is used or is supplied varies greatly from region to region, but is usually available as specific fertiliser recommendations, Codes of Good Agricultural Practice, in booklets or as computer based systems. As well as straightforward N effects, there are many interactions with other factors which influence the efficiency of N use. Thus a shortage of water will restrict growth and hence uptake, as will a shortage of other nutrients. In the latter case, Liebig’s Law of the minimum will be followed and use of the non-limiting nutrients through growth will be dependent on the availability of the limiting nutrient. Thus a sulphur shortage has been shown to reduce N uptake by grass swards and creates a surplus of N in the soil, enhancing the potential for its loss (Brown et╯al., 2000), and conversely, as shown in Figure 10.3, N application has enhanced the use of potassium and phosphorus.

farmers decide on the acceptable level of risk associated with each their operations to determine nutrient application/management regimes. Farmers have multiple roles:€they are custodians of important resources, the farm and the soils on which it resides, and they are managers and risk takers. And their skills determine the level of risk they are prepared to take to achieve financial gain and/or environmental benefit. However, the majority of farmers are businessmen and women, and many are entrepreneurs, whose primary aim is to optimise their production system to the benefit of themselves and perhaps of society as well. Further improvement of efficiency of N use within farming systems is dependent on the effective uptake of new knowledge and approaches as they become available from new research. Figure 10.4 shows the difference between what was technically achievable and that being achieved by the best Dutch farmers in 2000. In all components of the system shown, there was a significant shortfall between what was technically possible and that achieved in practice. A similar examination of farmers with

Hanninghof long-term trial (since 1958) Average P and K uptake in 17 years with oats (grain and straw) 70.00 60.00 50.00 40.00

–N +N

30.00 20.00 10.00 0.00

10.2.3╇ The role of the farmer

P uptake (kg P/ha)

The importance of individual farmer decisions has been noted:€much depends upon the skill and precision with which

Figure 10.3. Effects of nitrogen application on phosphorus and potassium uptake (kg/ha) (F. Brentrup, personal communication).

100

Utilisation efficiency (%)

K uptake (kg K/ha)

90

Technically possible

80

Realised in practice

Figure 10.4 Potential for change for increased N efficiency (%) and that achieved in practice by skilled farmers (from Jarvis and Aarts, 2000).

70 60 50 40 30 20 10 0 Soil to crop

Crop to intake

Intake to– product

Excretato soil

Whole farm Input to output

215

Nitrogen flows in farming systems

less skill than those indicated in Figure 10.4 would show much larger discrepancies. Future research will extend the technically achievable efficiency of use. Skilled farmers will keep up with this and create the need to ensure that other farmers also increase their efficiency of N use.

10.3╇ Mechanisms affecting N use and loss in farming systems The fundamental doctrine of N management is to optimise efficiency of both introduced and native soil N by increasing the temporal and spatial coincidence (synchronisation and ‘synlocation’) between availability and root uptake of mineral N (Christensen, 2004; Crews and Peoples, 2005). The management needed to assure the vitality of highly productive crops most often causes agro-ecosystems to be relatively ‘open’ with respect to losses. By moving the focus from productivity-only drivers to a balance between yield, product quality and environmental impact, farm management and associated agroecosystems can be re-designed to increase N use efficiency. Important management measures to improve N efficiency on farms include improved feeding efficiency of animals, reducing NH3 losses and improving N retention in the crop–soil system as well as timing, rate, source/material and method of supply. The latter implies crop sequences that incorporate cover (or catch) crops, judicious use of soil tillage, improved timing and Input

Rooting zone

Mineral fertiliser

Animal manure

In the soil, N undergoes a variety of largely microbial mediated transformations, which are associated with organic matter (OM) turnover. Agricultural soils contain a large pool of organically bound N:€ Figure 10.5 shows an expanded view of the soil compartment shown earlier in Figure 10.2. The soil layers exploited by plant roots typically contain 5000–15â•›000 kg N/ha.

Harvest

Denitrification Leaching

NO3–

Volatilisation

NH4+

Decomposer biomass Labile organic matter

N2 fixation

Stabilised organic matter

216

10.3.1╇ Nitrogen turnover in agricultural soils

Output

Plant Biomass

Atmospheric deposition

use of animal manures, crop residues and mineral fertilisers and a suitable balance between the plant production potential and animal stocking density. Livestock production systems (in particular ruminants) have a considerably lower N use efficiency than those based on cash crops. When plant biomass is utilised by the livestock, up to 80–90% of the plant N is recycled on the farm. The handling and subsequent use of this N will unavoidably lead to losses with all the entailing environmental impacts already described. The main challenge for N management in farming systems is tightening the cycle, in particular those of livestock production systems. The total amount of N excreted by livestock in the EU-27 peaked at about 11 Tg in the late 1980s, which was very similar to the 12 Tg used as fertiliser (Oenema et╯al., 2007). This illustrates the importance of considering both fertiliser and manure N inputs as well as the inputs from biological N fixation.

Figure 10.5 The nitrogen flows in a typical arable soil showing major inputs, outputs and pools of nitrogen (from Christensen, 2004).

Steve Jarvis

However, only 1%–2% of this large pool may in any given year become available to crop uptake within the growing period. The pools of N that dominate the short-term N cycle are the decomposer-biomass and labile OM pools. These N pools are relatively dynamic and respond readily to inputs of plant residues and animal manure, changes in moisture and temperature and to soil disturbances, as caused for example by tillage. Nitrogen occurs in the soil in chemical forms with widely different characteristics in terms of availability to plants and susceptibility to losses. Ammonia located at the soil surface can easily be lost by volatilisation, particularly at alkaline pH (Sommer et╯al., 2003). NO3− is very mobile in the soil solution and thus susceptible to leaching, whereas NH4+ is retained in the soil through sorption to soil colloids or fixation in clay minerals. Nitrification and, in the main, denitrification are the sources of gaseous losses as NO, N2O and N2. By far the largest fraction of organically bound N is retained in the soil, but under some circumstances soluble organic N losses may be significant (Murphy et╯al., 2000). Through the mineralisation–immobilisation turnover proÂ� cesses, mineral N becomes available from soil organic matter (SOM) for plant uptake, but also for losses to the environment. The rate and the seasonal and spatial distribution of these proÂ� cesses influence the composition and productivity of the vegetation. Natural ecosystems (including many grasslands) exhibit a large degree of synchrony and synlocation between release and uptake potentials and losses are generally small. In contrast, most arable cropping systems are relatively open, primarily because annual crops with a large N demand during the vegetative growth phase are used. Management also introduces massive physical disturbance of the soil structure through tillage and affects hydrology through drainage and irrigation. Management of these systems therefore causes the dynamics of the processes to differ substantially from those of natural ecosystems, in particular by reducing the synlocation and synchrony in the N turnover and a reduced return of organic matter to the soil, which collectively reduce N-use efficiency.

efficiency (REA), which is the increase in N yield (or total biomass) divided by the amount of N applied and (ii) the direct recovery efficiency (RED), which is the amount of labelled N that is taken up in a crop (usually only in above-ground material) following application of addition of 15N labelled fertiliser. The RED of mineral fertiliser N applied in autumn has been measured at 11–42%. For spring-time applications this increases to 42–78%, illustrating the effect of improved timing of the application for improving synchrony with crop uptake (Christensen, 2004). Typical REA values in research plots are c. 40–50% for small-grain cereals, when defined on grain N yield, and this is increased to 60–70% when based on total aboveground N uptake (Balasubramanian et╯al., 2004; Olesen et╯al., 2009). The RED values are generally smaller than REA values because some of the applied N is incorporated into microbial biomass and possibly into SOM. Experiments with 15N-labelled fertilisers applied to wheat have shown larger RED values in humid than in dry environments, illustrating the importance of environment for N-use efficiency. However, the retention of residual 15N in the soil increased with increasing dryness. Postharvest NO3− losses of residual fertiliser N is usually less than 5%, indicating that NO3− that is susceptible to leaching during autumn and winter in humid environments mainly originates from mineralisation of organic N. Fertiliser applications should be calculated to provide the smallest rate necessary to obtain the optimum crop yield achievable at the specific site, and to ensure the quality of the crop. European farmers usually meet the needs of the crop by using several separate N applications to prevent deficiency in periods of peak demand as well as to ensure no over-supply. Properly applied N application allows farmers to manage the development of the crop and to influence yield by the promotion or indirect inhibition of individual yield components, and directly improve yield quality. Such fertiliser strategies are widely used to contribute to the yield and quality management of cereals, which occupy more than 50% of EU arable land.

10.3.2╇ Mineral fertiliser

When increasing amounts of N are applied to different plots on the same field, in the same year, and on the same crop, the yields obtained from the plots usually form a typical response curve (Figure 10.6). The economic optimum for the farmer is usually defined as when the cost of the last unit of N applied is still covered by the value of the additional yield it produces. Establishing the correct fertiliser rate for a crop is a complex process, which involves many different factors such as crop type, expected yield and quality, nutrients available in the soil and changes in available nutrients during the growth period. In many countries, soil analysis is used to estimate the mineral N content in the rooting zone at the start of plant growth in spring to improve the decision for the first application. However, during the following growing season, the available N will be subject to the conversion processes noted elsewhere and which vary both in space and time. Furthermore, the need of the growing crop is also influenced by favourable or unfavourable growing conditions (Figure 10.7). As a result, the economic optimum fertiliser rate for a specific crop changes from

The use of mineral fertiliser as one source of plant nutrients is an essential component of current agricultural practice. Mineral fertilisers are applied in order to balance the gap between the nutrients required for economically optimal crop development and the nutrients supplied by the soil and by available organic sources. This gap results from a permanent export of nutrients from the field with the agricultural products. Today, the N gap is closed by an annual application of 97 Mt mineral fertiliser N at the global scale in 2006 (IFA, 2006). Mineral fertiliser N should, in principle, be applied at the time and location that is optimal for crop uptake and thus lead to potentially high N-use efficiencies. However, in practice many factors may reduce the actual efficiency obtained and although some farmers are more effective than others, this is one of the areas where there is still potential to make improvements (Figure 10.4). There are many ways to define and measure N use efficiency. Here, we will apply two different approaches:€(i) the apparent recovery

Response curves and farmer choice of optimal rates

217

Nitrogen flows in farming systems

Yield (t/ha) 10 9 8 7 6 5 4 3 2 1 0 0

Average of 172 annual trials (Yara, Germany, 1996–2008)

Long-term trial (Broadbalk Experiment, 1st wheat, 1996–2000)

50

100

150 200 250 N fertiliser rate (kg N/ha)

300

350

Figure 10.6 Average yield response to increasing annual N application rates in winter wheat (172 annual field trials in Germany, 1996–2008) and yield response in a long-term field trial with winter wheat (Broadbalk Experiment, average yields 1996–2000) (F.Brentrup, personal communication). 250 N uptake by the crop

kg N/ha

200

10.3.3╇ Manure handling and N use efficiencies N fertiliser demand

150

100

50

N supply from the soil

0 Feb

March April

May

June

July

Aug

Figure 10.7 The amount of nitrogen taken up by a crop depends on the growing conditions of the particular field and varies according to the growing conditions of the year (between the green lines). Mineralisation also varies from year to year (between the red lines). Therefore, the ‘correct’ application rate for the same crop in the same field (red arrows) will differ from year to year and may need adjustment during the growing season (F. Brentrup, personal communication).

year to year and from field to field and, for N, methods based only on soil analysis to determine that available for the crop have limited reliability and can serve to estimate rates needed at the start of growth, but require supporting decisions during the season. Split application strategies based on the N status of the plant can assist growers to adjust the available supply several times during the growth period. In this way the problem of a varying demand in different fields and in different years can be better managed. Thus, during the past 20 years, scientists and farmers have focused on methods based on direct plant analysis in the field to determine the optimum rate. Different methods have been developed for practical use (such as the NO3– sap test and, making, chlorophyll meter) to assist the decision. The above-mentioned methods are based on representative samples and thus provide a single average recommendation for the field, which makes them appropriate for smaller fields. Soil properties, nutrient availability, crop growth and final yield can vary widely within single fields (Figure 10.8). As a consequence, optimum fertiliser rates also vary and since the early 1990s,

218

variable rate technology has been developed to improve efficiency of inputs and lead to economic and environmental benefits. This technology, or more generally ‘Precision Agriculture’, aims to manage crop variability by tailoring inputs to specific needs in any particular part of the field. Variable application of N is of particular interest because N has the largest immediate effect on crop growth, yield and quality. The most promising systems for measuring within-field variation in crop growth are based on imaging crops by remote sensing and spectral indices derived from the reflectance spectra have been shown to be indirectly related to their N status. Using this information, spatially variable fertiliser application plans can be made to meet the optimum in each part of the field, which can be illustrated as a ‘map’. Experiments and practical experience indicate several potential economic and environmental benefits, including increased N efficiency, more uniform crop stands, ripening, and quality, easier harvesting and greater yields.

For at least one millennium, manure of livestock fed on seminatural grassland or with legume fodder crops was the only available fertiliser resource for croplands. During the last part of the 20th century, the abundance of manufactured fertiliser with predictable yield effects caused farmers to consider animal manure as a waste product of animal husbandry rather than a valuable nutrient source. One of the main challenges of modern agriculture has been to change this perception and to document and demonstrate a more efficient use of this N. Some 70–80% of livestock excreta are collected in housing systems in EU-27, with a tendency for this to increase (Oenema et╯al., 2007). The remaining 30–20% of livestock excreta is dropped at grazing, where it is difficult to manage but contributes to the N economy of the system. More than half the manure collected in housing systems is managed in the form of slurry or liquid, while the remainder is managed in a solid form and often includes larger quantities of bedding material (e.g., deep litter or farm-yard manure). There is a huge regional variation in manure management systems in Europe (Menzi, 2002). Slurrybased systems are dominant in the Netherlands and Denmark (>90%), while separate collection of slurries and solids dominate in UK, France and Central/Eastern Europe (<50% slurry/ liquid). Most of the slurries are stored in tanks with or without covers, but some is stored in unsealed pits in Central Europe. The EU Nitrates Directive obliges Member States to properly store (for up to nine months) and manage manure. However, in practice, implementation in some countries has been slow. There are many loss pathways for N after excretion in the animal house, manure storage or after application in the field. However, the dominant loss is through NH3 volatilisation. The urea in the excreted urine is rapidly hydrolysed to NH3 which can be volatilised, especially if it is placed on open surfaces, if pH is alkaline and if temperatures are high (Sommer et╯al., 2006). Modelling studies indicate that in 2000 almost 30% of the N excreted in animal housing systems in EU-27 was lost during storage; approximately 19% via NH3 emissions, 7% via nitrification and denitrification (NO, N2O and N2) and 4% via

Steve Jarvis Nitrogen application map based on crop scanning by tractor mounted sensor Spectral Index High

Sensor Map

0m

Figure 10.8 Nitrogen application ‘map’ based on crop scanning by a tractor-mounted N-SensorTM (winter barley, 25 May 1999, F.€Brentrup, personal communication).

100 m 200 m 300 m 400 m

Low

N-Application Map

N (kg N/ha) 100 90

Winter barley, N-Sensor-measurement and N-application on the 25thof May 1999

80 0m

100 m 200 m 300 m 400 m

70 60 50 40

N leaching and runoff (Oenema et╯al., 2007). Another 17% of the N excreted in the housing was lost via NH3 volatilisation following application. Thus in total, 48% of the N excreted in animal housing was lost during storage and immediately after application. Because of the significance of this, we pay special attention to this loss route.

Ammonia emission from animal production During the last decades much research on NH3 emission reduction has focused on constructional measures for animal houses and on low emission application techniques for manures. The guiding principle is minimising the contact surface and contact time between animal manure and the surrounding air. Examples are decreasing the evaporating area of the manure in the storage pit and frequently removing the manure to an outside storage and using sod injection for slurries on grassland and direct ploughing after land spreading on arable land (Starmans and Van der Hoek, 2007). Measures with respect to animal feeding have effects on excretion and on NH3 emission. Dutch research on cattle feeding showed a linear relationship between milk urea content and the emissions from housing. An increase in the milk urea concentration from 20 to 40 mg/100 gram milk resulted in increasing emissions from 5 to 9 kg NH3 per cow in a cubicle cow during the 190-day winter season (Van Duinkerken et╯al., 2005). The resulting decrease of the NH4-N content of the manure will, however, reduce volatilisation losses during manure storage and manure application. Dutch research on pig feeding showed reductions in emissions of up to 70% as a combined effect of reducing the N content of the feed, additives affecting the pH of the manure and resulting in a change of N in urine into faecal protein (Aarnink and Verstegen, 2007). Emission from solid poultry manure is governed by manure characteristics such as pH, temperature and water

content. Microbial breakdown of uric acid and undigested proteins into NH3 is dependent on moisture content. The positive effect of drying poultry manure on lowering emissions was demonstrated in pilot studies and on practical farms (Groot Koerkamp, 1994; Groot Koerkamp et╯al., 1998; Starmans and Van der Hoek, 2007). Slurry injection into bare soil and trailing hose application and injection (Nyord et╯al., 2008) to growing arable crops (Sommer et╯al., 1997) reduce NH3 emissions substantially. Sod injection on grassland or ploughing directly after manure spreading on arable land is very effective in reducing emissions (Huijsmans et╯al., 2001; 2003). In theory, choosing the right meteorological conditions for spreading can help to reduce emissions from land spreading of manure. However, farmers may have limited choice about the timing of manure applications, because of operational constraints such as availability of contractors or regulatory considerations (such as those imposed by the restrictions of the Nitrates Directive). The efficacy of this approach has yet to be proved in practice.

Improving N use efficiency from manures Many different technologies for reducing housing and storage emissions and improving manure quality have been tested and are increasingly being implemented. These include reducing fouled surface areas in animal houses, covering manure stores, acidification of slurry to reduce pH, slurry separation, biogas digestion, incineration of solid manures, etc. Some of these treatments not only reduce N losses but may have other advantages such as providing energy or increasing the total fertiliser value of the manure. Fewer measures are available for reducing gaseous N losses from solid manures than for slurry, partly because much of the scientific research and technical development has been in areas

219

Nitrogen flows in farming systems

of Europe where slurry is the dominant form and because NH3 losses from slurry are much greater than from solid manures. Significant losses of NH3 can occur from stored solid manure, if there is composting or self-heating (Sommer, 2001; Dämmgen and Hutchings, 2008). When implementing measures to reduce N emissions from the manure management system, it is important to take a whole system approach. Reducing NH3 emissions from animal housing may result in greater emissions from subsequent stages storage, field application unless additional measures are taken (Hutchings et╯al., 1996). For animal manure to be a reliable source of plant-available N, NH3 losses need to be small, and the distribution across the field should be as uniform as for mineral fertiliser. This often requires use of expensive equipment supplemented with measurements of NH4+-N in the manure just before application to precisely target crop needs. Crop recovery of N in manure varies widely. Compared with the 42–78% RED values cited above for mineral fertilisers, the RED values for manure applied in spring is only 18–31% for faeces, while it is 61–88% for urine and poultry excreta (Christensen, 2004). The recovery of N in manure in the second year is comparatively small with RED values of 2–6%, and is almost independent of source of N. Storage facilities must allow manure to be kept over winter without significant loss of€N. Application in spring is a prerequisite for maximising the N use from this source.

10.3.4╇ Crop residues (and rhizodeposition) Crop residues returned to the soil represent a significant input, not only of N, but also of easily decomposable carbon, to the microbial decomposers in the soil. Application of residues with a wide C:N ratio (e.g., straw) can therefore lead to immobilisation of soil mineral N. Application of manure and plant material (as in a green manure crop) with a smaller C:N ratio will lead to more rapid decomposition and release of mineral N. Mature crop residues applied in autumn typically result in RED values of 8–13% and a grass incorporated in spring one of 70% (Christensen, 2004). Crop residues from legume crops provide an important source of N in many farming systems (in particular in organic farming). Studies using labelled N have typically shown RED values of 10–30% for N incorporated in legume N residues, which is considerably less than for fertiliser N (Crews and Peoples, 2005). These values are, however, misleading since a considerably larger proportion of the legume N is incorporated into the soil microbial N pool, and some of the microbial N is released as a consequence. In many cases, almost equally good REA values have been found for N in fresh legume residues as compared with that in mineral fertilisers. Rhizodeposition of N during plant growth, which is the release of labile organic N into the soil from plant roots and from the nodules of legumes, may be an important source of N, and could represent from 25% to 43% of total N recovered in the crop (Russell and Fillery, 1996). Mayer et╯al. (2003) clearly indicate that N rhizodeposition of grain legumes (beans, peas and lupins) represents a significant contribution to the balance and N dynamics in crop rotations.

220

10.3.5╇ Management of N in agro-ecosystems There are many management factors that affect N use efficiency in crop production. They can roughly be grouped into strategies that try to either (i) increase plant demand, (ii) manipulate supply or (iii) capture excess inorganic N before it is lost. The most obvious strategy is to adjust the N inputs as closely as possible in time and space to the requirements of the crop. However, to ensure an efficient N uptake requires a healthy crop, which is one of the most important, but often overlooked factors by which plant N demand can be increased. Break crops in temperate wheat production not only improve yield by improving N supply, but also by ensuring a healthier root system that enables the crop to better utilise soil N (Kirkegaard et╯al., 2008). Similarly, leaf diseases in cereals have been found to reduce N use efficiency (Olesen et╯al., 2003). Soil tillage leads to disturbance of soil structure and this influences N turnover in the soil by modifying aeration and soil moisture that affects plant roots and soil organisms. Tillage also leads to a better mixing of soil and N-containing substrates that will favour decomposition and subsequently lead to release of mineral N. Soil tillage in autumn therefore often leads to enhanced losses of NO3− through leaching, whereas tillage in spring can lead to enhanced uptake of N by the crop, but also to greater N2O emissions (Chatskikh and Olesen, 2007). Water availability is one of the key factors controlling N processes (nitrification, denitrification, mineralisation, N leaching, etc.) and crop yield. This factor can be controlled more effectively in irrigated than in rain-fed systems. The intensification of irrigation to obtain economical benefit has grown in many areas, especially in arid and semi-arid regions, where a large amount of N fertiliser is often combined with a high volume of applied water. This has had an adverse effect on leaching losses contributing to groundwater pollution in important areas of these countries (Diez et╯al., 1994), but good agricultural practices have emerged in response to the need to provide the right amount of water for each crop, avoiding therefore, as much as possible, over-watering (A. Vallejo, personal communication). It is pertinent to make a special note at this point on the debate and the issue of sustainability of organic and conventional farming. Because organic farms do not use synthetic fertilisers, they have in general a lower yield per hectare than conventional farms. Comparing the emissions per unit of production provides more insight in both systems. Probably the best comparison will be made with an equal N intensity per hectare on organic and conventional farms, as has been discussed by Goulding et╯al. (2008) and Olesen et╯al. (2006).

10.3.6╇ Crop rotations Crop rotations affect N use and losses in several ways. First, the rotation defines the sequence of crops of various N demands and of various amounts of N in residues returned to the soil. Second, the crop sequence defines the time of break between the different crops, time of tillage operations and possibilities (and needs) for growing cover crops. Third, the crop sequence

Steve Jarvis

affects crop growth by modifying soil properties and preventing (or propagating) weeds and diseases. The challenge is to design crop rotations that both maximises crop production (usually driven by the need to generate acceptable incomes) and, at the same time, reduces N losses. For grassland systems this may include avoiding late season grazing, high protein supplementary feeds and autumn ploughing of grasslands. For arable cropping, this could mean reduced autumn tillage, use of cover crops and spring application of manure with application technologies to reduce NH3 volatilisation. Intercropping with mixtures of two or more species is also effective in reducing risk of N losses, e.g., for grass-clovers compared with ryegrass monocultures (Eriksen et╯al., 2004). Crop rotations that include grass or grass/clover leys require special care; there is an accumulation of organic N in the soil under the ley N management of the crop following the grassland needs to take account of the substantial net mineralisation that results when the grassland is ploughed (Eriksen et╯al., 1999; Hutchings et╯al., 2007). Under a sustainable agriculture management therefore, crop rotation is an important tool to improve nitrogen use efficiency. There is a greater degree of synchronisation between crop N-uptake and N dynamics in rotations than in monocultures (Pierce and Rice, 1988). Cover crops (or catch crops) are crops that are grown in breaks between main crops, often to capture excess soil mineral N or to capture N by biological fixation (green manure crops). Both types of cover crops reduce N leaching when growing during the wet part of the year, when no main crop is present. The efficiency with which excess N is taken up depends on time of establishment and final root depth of the cover crop. Cover crops when used in the spring do not always lead to positive effects on crop N nutrition, since those with a large C:N ratio may lead to microbial immobilisation of N, and soils without a cover crop often have a higher initial mineral N content than where a cover crop has been growing (Thorup-Kristensen, 1994). In systems dominated by spring-sown cereals, cover crops can be established as an under-sown crop or sown just before harvest of the main crop. Such cover crops are therefore grown during autumn and winter and ploughed in spring, enhancing the synchrony of N uptake and supply, thereby reducing N leaching losses (Figure 10.9). Cropping systems for ruminant husbandry provide plant biomass for grazing and for feed for housed animals. Animal grazing has traditionally relied on perennial leys of grass mixtures and grass-clover. However, with more intensive systems such leys may be included in the crop rotation, which further enhances N cycling and also the risks of losses. There is typically a non-linear relationship between input and output and, conversely, between input and losses (Figure 10.10). Above a certain N input, there is no further increase in productivity and N losses increase. In intensive systems, grazing management is a key factor for N use efficiency on dairy farms. Restricted grazing contributes to increasing nutrient use efficiency at farm level through better utilisation of animal excreta, because these are collected in manure stored instead of being deposited in the meadow. However, in general, increasing the actual grazing period leads

Winter

Spring

Summer

Autumn

Percolation

Nitrogen availability

Crop uptake potential

Spring barley

D

J

F

M

A

M

J

Nitrate catch crop

J

A

S

O

N

Figure 10.9 The seasonal dynamics of potentials for percolation (NO3− leaching loss), availability of mineral nitrogen (mineralisation + external inputs) and crop uptake under typical northwest European climatic conditions. The spring barley crop is under-sown with ryegrass acting as a catch crop. The vertical grey zones indicate periods with increased susceptibility to elevated NO3− leaching losses (from Christensen, 2004).

Figure 10.10 Relationship between total annual N input and N output as products (milk, meat and crops€– shown as triangles) and as losses (volatilisation, denitrification, run-off, leaching and transfer to unproductive areas€– shown as circles) in dairy and beef production systems that involve some grazing (from Rotz et╯al., 2005).

to smaller emissions of NH3 and to greater leaching of NO3− and greater consumption of manufactured fertiliser. Increasing the grazing period also means fewer possibilities for adjusting the protein content of the animal feed. Finally, grazing is cheaper than housing and benefits animal welfare. For a comprehensive comparison of grazing-based and confinementbased cattle production we refer to the literature (Arsenault et╯al., 2009). In extensive farming systems, grazing management has multiple goals including sustainability in terms of animal feeding resources and ecological and sociological functions (Hadjigeorgiou et╯al., 2005).

221

Nitrogen flows in farming systems

10.3.7╇ Use of modelling Process-based knowledge of N and C cycling has, in many instances, been integrated into mechanistic and dynamic simulation models, many of them operating at farm and agro-ecosystems scales. Models are often used for decision making at the macro policy level, but not necessarily for on-farm management. They therefore frequently influence bigger scale policy decisions, but not those concerned with production on farms on a widespread basis. However, such models offer the potential to analyse the contribution of individual components to the total system N cycling and losses. This is typically done through sensitivity analyses and inter-model comparisons, which may be used to identify gaps in current process understanding. Modelling can also serve as a tool for interpreting experimental results and extrapolating to new environmental and management conditions (Smith et╯al., 1997). The available models often have different strengths in scale or loss pathways. Most models function at the plot or field scale (Li et╯al., 1992), whereas a few models integrate interactions also at the farm scale (Berntsen et╯al., 2003; Brown et╯al., 2005; Rotz et╯al., 2005). Most of the models simulate NO3− leaching, some simulate denitrification and N2O emissions (Li et╯al., 1992), whereas few models simulate NH3 volatilisation (Sommer et╯al., 2003). Models are often applied for estimating losses at larger spatial or temporal scales. However, for feasibility this often involves simplifying model inputs or model structure. These models are rarely applied to practical decision support since they have large requirements on the accuracy of input data that cannot be met in practice. Instead more simple and empirical tools are used that rely on N balance sheets, supported by bookkeeping of easily measured and estimated field or farm-scale inputs and outputs of N (see Sections 10.1 and 10.4). This is often supplemented with use of N response curves from field experiments and various approaches that apply N use efficiency considerations. These systems are, however, far from perfect and far from widespread, and there is particular need to improve the estimation of how much mineral N is released from mineralisation of soil OM and incorporated crop residues and catch crops.

10.4╇ Example farming systems In this section, we consider the farm budgets of a range of farming systems. We focus on typical farming practice in �north-western

Europe and use information from various sources and expert opinions to model and construct generalised budgets (further information is provided in supplementary material for this chapter). The examples were chosen to illustrate some general differences between farming systems; it is therefore necessary to emphasise here that, in practice, a wide range of budgets can be found within each system type. The numbers in each Figure refer to annual flows of N per ha within the system.

10.4.1╇ Arable farms The simplest budget is that of conventional arable farms (Figure 10.11). Production is largely driven by the use of imported mineral N fertiliser, although in areas containing livestock farms, imported manure may be an additional N source. The input of N via fixation is usually small unless it is an organically based system, since the crop rotation is usually dominated by nonfixing crops. Unless manure is imported or urea is the choice of fertiliser, NH3 emissions will be relatively low. The fate of the remaining mineral N is divided between crop uptake, loss to the atmosphere as N2, N2O and NO via denitrification, loss in water as NO3− and DON in leachate and runoff or accumulation in SOM. The soil organic matter is also a source of N via mineralisation. Over longer periods (annually or more), the balance between mineralisation and immobilisation in SOM is dependent upon the extent to which soil processes have reached an equilibrium with soil management and climatic conditions. The factors encouraging net immobilisation include the presence of crops or use of management practices that add larger amounts of plant residues (e.g., the presence of grass or the incorporation of straw), increased waterlogging (e.g., from changes in climate or drainage) and acidification (because of base-ion leaching). The proportion in crop uptake tends to increase as the growth potential of the crop increases (e.g., more productive varieties) and the proportion of the year in which a crop is established increases (e.g., through the use of crops with a long growing season or via the planting of catch crops). The partitioning of N between NO3– leaching, runoff and denitrification is dependent on a wide range of factors, including the type and timing of fertiliser addition, the soil type and drainage and the climate. In general, it appears that on freely-drained soils, NO3− leaching predominates whereas on poorly-drained soils and those with a high water table, denitrification predominates. Figure 10.11 Annual nitrogen flows (kg/ha) in a conventional arable farming system.

222

Steve Jarvis

Figure 10.12 Annual nitrogen flows (kg/ha) in a pig farming system.

Recovery of N in crops and the fate of the un-recovered N should be estimated by measurement where it is technically and economically possible and modelling where it is not. In practice, a combination of both is the norm. It is important that estimates are not only based on experimental results, where conditions are usually close to optimal, but also from commercial farms where they may not be. Figure 10.11 shows values for Danish agriculture, so are typical for North West Europe and balanced N addition designed for yields that are about 10% below the economic optimum. Inputs and outputs were measured whereas the partitioning of the farm N surplus is based on assumptions that are reasonable for Danish agriculture. Note that leaching is estimated at the base of the root zone; the subsequent fate of the leached NO3− and DON is not determined in this example. The partitioning is particularly sensitive to the climatic and soil conditions. For example, in situations where the upper limit of the groundwater is within the root zone for part of the year, leaching would be much smaller and gaseous emissions from denitrification much greater. In the absence of balanced fertiliser inputs, the removal of N in crop would be marginally greater, but the losses to the environment would probably be substantially greater.

10.4.2╇ Livestock:€non-ruminants On farms with pigs, poultry and other non-ruminants, there is a substantial additional input of N in the feed imported for the livestock which supplies most or all of their requirements. Figure 10.12 shows an example of this for pig farms. As for arable farms, the input of N via fixation is usually limited. The large quantities of manure produced can sometimes be difficult to utilise effectively on the farm, so manure may have to be exported from the farm. If manure is exported, not all of the losses of N associated with the livestock production will be reflected in the farm N budget, since a proportion will occur

on the farm receiving the manure. The values in Figure 10.12 are typical for the Danish situation in terms of climate, soils and balanced N fertiliser management. As a consequence, the import of N in mineral fertiliser is less than for the arable farm. As for the arable example, the farm inputs and outputs are based on measurements whereas the partitioning of the farm N surplus is based on assumptions that are reasonable for the Danish situation.

10.4.3╇ Livestock:€ruminant animals A proportion of the diet of ruminant animals is normally in the form of roughage (high-fibre plant material), because such feeds are required for good rumen function and because it is often relatively cheap compared with the alternatives. Roughage feed is usually bulky and therefore expensive to transport, so is usually produced on the farm rather than imported. On specialist ruminant livestock farms, much or all the farm’s land will be dedicated to growing roughage feed, so a large proportion of the diet is produced on the farm (Figure 10.13). If the livestock manure is applied to this land, this creates a feedback mechanism in which the amount of N in the feed affects the amount in the manure, which affects the N supplied to the roughage and therefore the N content of the roughage fed to the animals. Ruminant livestock farms also differ from non-ruminant farms in three other aspects. The first is that a proportion of the roughage feed is often harvested by grazing, so the N deposited in the accompanying dung and urine bypasses the manure management system. The second is that the feed conversion efficiency for the ruminants is generally poorer than for the non-ruminants. This in part reflects the inherently lower efficiency of the ruminant digestion and in part the tendency for the inclusion of roughage in the diet to lead to an oversupply of protein. The third difference is that it is not uncommon to include plants such as clover in the roughage crops, so the input of N from fixation may be greater than on other types of farms.

223

Nitrogen flows in farming systems

Figure 10.13 Annual nitrogen flows (kg/ha) in a beef farming system.

Figure 10.14 Annual nitrogen flows (kg/ha) in a dairy farming system.

Figure 10.13 illustrates the N flow on a beef farm where grazed herbage is a major component of the animals’ diet. As a consequence, there are significant flows of N from the pasture to the cattle via grazing and from the cattle to pasture via the associated deposition of dung and urine. Because the animals spend a large proportion of the year at pasture (here assumed to be 0.6), the amount of N managed as manure is small, relative to that typical for pig farms. As a consequence, the gaseous emissions of N are also smaller but the leaching losses are larger. On intensive dairy farms (Figure 10.14), the energy demand of highly productive livestock cannot be satisfied by roughage alone, so feed with a high energy concentration (e.g., grain)

224

will be imported onto the farm. The proportion of the animal feed demand satisfied from the farm’s resource is therefore smaller than for more extensive ruminant systems, such as the beef system (Figure 10.13). Although the young replacement cattle will often be grazed for much of the year, the dairy cows will spend a significant proportion of time in animal housing, even during the growing season. The weighted average proportion of time spent indoors is assumed here to be 0.4. As for the pig farms, large amounts of manure are produced in the animal housing, leading to substantial losses of gaseous N. Despite the differences between the example beef and dairy situations in the magnitude of individual inputs and flows of N, the total amount of N entering the soil are similar in both

Steve Jarvis

Figure 10.15 Annual nitrogen flows (kg/ha) in an organic dairy farming system.

cases. As a consequence, losses by leaching and denitrification are similar. The final example shown here is of organic dairy farming (Figure 10.15). The main distinguishing management features of the organic dairy farm as far as N flows are concerned, are the lower intensity of production and the absence of mineral fertiliser inputs. The reduced intensity leads to a greater contribution of N fixation to the N input to the crops and a greater efficiency of recovery in the offtake from the farm in animal products and crops from all external inputs to the soil (about 30% compared with 25% for the conventional dairy farm). There is also a greater accumulation of N in the soil than found for the other farming types. The consequent reduction of losses of N to the environment is therefore the result of a combination of a lower stock density and a greater efficiency of N use in the field. It can be noted that although the typical total losses of N to the environment in the organic system (75 kg/ha) are much smaller than the conventional dairy farming system (Figure 10.14) (143 kg/ha), there are also substantial differences in the total products produced. Thus the total animal and crop products of the organic system in Figure 10.15 contain 39 kg/ha N, while those of the conventional system in Figure 10.14 contain 56 kg/ha N. Expressed as losses of N to the environment per unit N in products, the losses are about 30% greater in the organic example (ratio of 2.5) compared with the conventional dairy (ratio 1.9) (Table 10.1). The comparison above is with an organically based system, but it is likely that similar differences between dairy managements would be obtained when other low-input systems are compared with those based on greater levels of input. Table 10.1 provides a summary of the differences in annual losses from the farm budgets described in this chapter and indicates some other important differences. For example, whilst there are some substantial losses from the pig farm, the ratio of loss to that in the product is relatively much smaller than that in the beef system.

However, caution is needed in interpreting these ratios. For example, pig production has a smaller ratio than beef but this is because these farms import large quantities of animal feed and the emissions associated with the feed production occur in another farm. In contrast, the cattle farms produce a larger proportion of the animal feed themselves.

10.4.3╇ Other farming systems The previous estimates have, of necessity, centred on NW European examples for which there is a growing bank of information on the various components comprising a farm system. This is not the case for systems in many other parts of Europe, especially those in the south and also for many of those based on low input managements. As an example, in southern European farms, the N balance and flows are very dependent on soil moisture conditions and whether the crops are irrigated or rain-fed. The effects of the warmer temperatures and different moisture status on N flows can result in very different effects to those described earlier. For example, in areas close to intensive livestock production, such as pig farms where slurry is often used to supply N instead of mineral N, effects of moisture and temperature can be accentuated. Thus, although the N balance is quantitatively very similar when mineral N is applied, depending on the slurry management (rate, timing, method of application, etc.), the relative importance of pathways of N flows and losses may be very different, particularly those of NH3 volatilisation (A. Vallejo, personal communication). Other information for systems in other parts of Europe is given in De Clercq et╯al. (2001), which provides simple �farm-gate balances for typical farming systems in each of the EU-16 countries. Whilst such balances provide useful insights into the N surpluses for a wide range of farming systems and their efficiency of N use, they are based on limited

225

Nitrogen flows in farming systems Table 10.1 Summary of annual N in products and losses (kg/ha) derived from the typical farm nitrogen budgets (Figures 10.11–10.15), with losses also expressed per unit N in products

Nitrogen in crop and animal products Farm management Arable

Nitrogen losses N losses per unit N in products (as ratio)

kg/ha/year N 99

84

0.85

Pig

159

131

0.82

Beef

40

108

2.7

Dairy (conventional)

56

143

2.55

Dairy (organic)

39

75

1.92

data and do not, for example, illustrate the internal transfers that occur and can be seen in the earlier examples. There is therefore much scope to gain wider understanding for these systems and thus identify opportunities to improve N management and reduce flows to the wider environment under other circumstances.

10.5╇ Conclusions • Nitrogen is an essential component of the requirements for producing food, fibre and energy. There is no opportunity of avoiding inputs from some source or other without reducing production potential. As a consequence, there will always be some losses:€natural ecosystems also leak N into the wider environment. It therefore takes much skill and effort and is difficult, but by no means impossible, to reconcile the dual aims of sustaining or increasing production with an ever-increasing demand for reduced losses to the environment when there is an increasing demand for food. • Farming systems within the EU are diverse, occupying wide ranges of climate, soil type, topography and managements. Our examples have centred on current farming practice in parts of NW Europe. Those operating in Southern Europe will have very different objectives, operational structures and, although the same mechanisms and pathways for N are involved, different flows of N will be encountered as demonstrated in the examples of farm-type budgets. • The role of the individual farmer is crucial in optimising the flows of N to meet the dual targets of maximising production and minimising environmental cost. The aim must be to optimise supplies to the crops and animals by more effectively matching supply and demand, in time and space. • The farm N-cycle, which we have described in fairly simplistic terms, is actually rather complex:€each component of the cycle can be divided into other internal cycles, which may be more or less complex. Equally, the farm cycle is a smaller component of the larger-scale cycling that occurs and which is generally considered at catchment or river-basin scales. Superimposed on all of these scales is societal influence, which can be multidirectional in its effects, requirements and impacts. The

226

farm scale is, however, the operational scale at which many of these interactive effects are demonstrated or at which implementation policies have to be deployed, either for production or environmental requirements. • Livestock farming presents particular issues and problems and, particularly where it is separated from tillage land, can result in accumulations of N with the potential of overloading the system and generating much leakage of excess N. Again technologies are available and are being increasingly employed to reduce the impact of, for example, NH3 volatilisation. • Agriculture and the technologies that it employs have reacted positively to the changes required to meet the demands placed upon agriculture. Decision-making in the use of N has become more precise, but there is opportunity to do more. However, there are limits to the increases in efficiency that can be achieved (see Figure 10.4). Important in achieving the potential to increase the efficiency of N use is the maintenance of a high skill base and awareness of environmental impacts amongst the farmer community. The benefits of using N effectively on farms are important and large. The strengths of using N are that it has immediate impact in promoting growth and production, that we know a great deal about its controls and flows, and that it is also a very effective management tool for farmers to provide flexibility and other requirements. The weaknesses of using N are that it is a very mobile and leaky nutrient, it is readily available (although at cost) and easy to use, but does require skilled management. There are increasing opportunities to employ ever-developing technologies to maximise efficiencies (precision application of fertilisers and manures, improved animal diets, improved breeding to optimise supplies, etc.) and an ever-increasing knowledge base amongst land managers about the requirements for its use from both production and environmental perspectives. There are, however, continued threats to the use of nitrogen in agriculture, which may involve complex interactions and feedback mechanisms with climate change, while possible future revised environmental standards may put increased demands on farmers to reduce losses further. Finally, changÂ� ing price structures for inputs (including N fertilisers) and the

Steve Jarvis

goods produced, coupled with other public pressures, may make the challenges for using N even more demanding.

Acknowledgements This chapter was prepared with the support of the NinE Programme of the European Science Foundation, the NitroEurope IP (funded by the European Commission), and the COST Action 729.

Supplementary materials Supplementary materials (as referenced in the chapter) are� available online through both Cambridge University Press:€ www.cambridge.org/ena and the Nitrogen in Europe website: www.nine-esf.org/ena.

References Aarnink, A. J. A. and Verstegen, M. W. A. (2007). Nutrition, key factor to reduce environmental load from pig production. Livestock Science, 109, 194–203. Arsenault, N., Tyedmers, P. and Fredeen, A. (2009). Comparing the environmental impacts of pasture-based and confinementbased dairy systems in Nova Scotia (Canada) using life cycle assessment. International Journal of Agricultural Sustainability, 7, 19–41. Balasubramanian, V., Alves, B., Aulakh, M. et╯al. (2004). Crop, environmental, and management factors affecting nitrogen use efficiency. In: Agriculture and the Nitrogen Cycle:€Assessing the Impacts of Fertilizer Use on Food Production and the Environment, ed. A. R. Mosier, J. K. Syers and J. R. Freney, SCOPE 65, Island Press, Washington DC, pp. 19–33. Berntsen, J., Jacobsen, B. H., Olesen, J. E., Petersen, B. M. and Hutchings, N. J. (2003). Evaluating nitrogen taxation scenarios using the dynamic whole farm simulation model FASSET. Agricultural Systems, 76, 817–839. Brown, L., Scholefield, D., Jewkes, E. C. et╯al. (2000). The effect of sulphur application on the efficiency of nitrogen use in two contrasting grassland soils. Journal of Agricultural Science, 135, 131–138. Brown, L., Scholefield, D., Jewkes, E. C., Lockyer, D. R. and Del Prado, A. (2005). NGAUGE :€A decision support system to optimise N fertilisation of British grassland for economic and environmental goals. Agriculture, Ecosystems & Environment, 109, 20–39. Chatskikh, D. and Olesen, J. E. (2007). Soil tillage enhanced CO2 and N2O emissions from loamy sand soil under spring barley. Soil & Tillage Research, 97, 5–18. Christensen, B. T. (2004). Tightening the nitrogen cycle. In:€Schjønning, P., Elmholt, S. and Christensen, B. T. (eds.) Managing Soil Quality:€Challenges in Modern Agriculture, CAB International, Wallingford, UK, pp. 47–67. Crews, T. E. and Peoples, M. B. (2005). Can the synchrony of nitrogen supply and crop demand be improved in legume and fertilizer-based agroecosystems? A review. Nutrient Cycling in Agroecosystems, 72, 101–129. Dämmgen, U. and Hutchings, N. J. (2008). Emissions of gaseous nitrogen species from manure management:€a new approach. Environmental Pollution, 154, 488–497. de Clercq, P., Gertis, A. C., Hofman, G. et╯al. (2001). Nutrient Management Legislation In European Countries. Wageningen Academic Publishers, Wageningen, The Netherlands.

de Vries, W., Liep, A., Reinds, G. J. et╯al. (2011). Geographic variation in terrestrial nitrogen budgets across Europe. In: The European Nitrogen Assessment, ed. Sutton, M. A., Howard, C. M., Erisman, J. W. et╯al., Cambridge University Press. Díez, J. A., Roman, R., Cartagena, M. C. et╯al. (1994). Controlling nitrate pollution of aquifers by using different nitrogenous controlled release fertilizers in maize crop. Agriculture, Ecosystems & Environment, 48, 49–56. Eriksen, J. Askegaard, M. and Kristensen, K. (1999). NO3-leaching in an organic dairy/crop rotation as affected by organic manure type, livestock density and crop. Soil Use & Management, 15, 176–182. Eriksen, J., Vinther, F. P. and Søegaard, K. (2004). NO3-leaching and N2 fixation in grasslands of different composition, age and management. Journal of Agricultural Science, 142, 141–151. Goulding, K., Jarvis, S. and Whitmore, A. (2008). Optimizing nutrient management for farm systems. Philosophical Transactions of the Royal Society, B, 363, 667–680. Groot Koerkamp, P. W. G. (1994). Review on emissions of ammonia from housing systems for laying hens in relation to sources, processes, building design and manure handling. Journal of Agricultural Engineering Research, 59, 73–87. Groot Koerkamp, P. W. G., Speelman, L. and Metz, J. H. M. (1998). Litter composition and ammonia emission in aviary houses for laying hens. Part 1:€Performance of a litter drying system. Journal of Agricultural Engineering Research, 70, 375–382. Hadjigeorgiou, I., Osoro, Fragoso de Almeida, K. P. and Molle,€G., (2005). Southern European grazing lands:€Production, environmental and landscape management aspects. Livestock Production Science, 96, 51–59. Hutchings, N. J. and Kristensen, I. S. (1995). Modelling mineral nitrogen accumulation in grazed pasture:€will more nitrogen leach from fertilized grass than unfertilized grass/clover? Grass & Forage Science, 50, 300–313. Hutchings, N. J., Olesen, J. E., Petersen, B. M. and Berntsen, J. (2007). Modelling spatial heterogeneity in grazed grassland and its effects on nitrogen cycling and greenhouse gas emissions. Agriculture Ecosystems & Environment, 121, 153–163. Hutchings, N. J., Sommer, S. G. and Jarvis, S. C. (1996). A model of ammonia volatilization from a grazing livestock farm. Atmospheric Environment, 30, 589–599. Huijsmans, J. F. M., Hol, J. M. G. and Hendriks, M. M. W. B. (2001). Effect of application technique, manure characteristics, weather and field conditions on ammonia volatilization from manure applied to grassland. Netherlands Journal of Agricultural Science, 49, 323–342. Huijsmans, J. F. M., Hol, J. M. G. and Vermeulen, G. D. (2003). Effect of application method, manure characteristics, weather and field conditions on ammonia volatilization from manure applied to arable land. Atmospheric Environment, 37, 3669–3680. IFA (2006). International Fertilizer Manufacturers Association, www.fertilizer.org (site accessed 24 September 2010). Jarvis, S. C. and Aarts, H. P. M. (2000). Nutrient management from a farming systems perspective. In:€Soegaard, K., Ohlsson, C., Hutchings, N. J., Kristensen, T. and Sehested, J. (eds). Grassland Farming Balancing Environmental and Economic Demands, Proceedings of the 18th General Meeting of the European Grassland Federation, Aalborg, Denmark, 22–25 May 2000 (Grassland Science in Europe, Vol. 5), pp. 363–373. Jarvis, S. C., Hatch, D. J. and Roberts, D. H. (1989). The effects of grassland management on nitrogen losses from grazed swards

227

Nitrogen flows in farming systems through ammonia volatilization; the relationship to excretal N returns from cattle. Journal of Agricultural Science, 112, 205–216. Kirkegaard, J., Christen, O., Krupinsky, J. and Layzell, D. (2008). Break crop benefits in temperate wheat production. Field Crops Research, 107, 185–195. Li, C., Frolking, S. and Frolking, T. A. (1992). A model of nitrous oxide evolution from soil driven by rainfall events:€1. Model structure and sensitivity. Journal of Geophysical Research, 97, 9759–9776. Mayer, J., Buegger, F., Jensen, E. S., Schloter, M. and Heβ, J. (2003). Estimating N rhizodeposition of grain legumes using a 15N in situ stem labelling method. Soil Biology & Biochemistry, 35, 21–28. Menzi, H. (2002). Manure management in Europe:€results of a recent survey. In:€Venglovsky, H. and Greserova, G. (eds.) RAMIRAN 2002. 10th International Conference on Hygiene & Safety, Strbske Pleso, High Tatras, Slovak Republic, pp. 93–102. Murphy, D. V., MacDonald, A. J., Stockdale, E. A. et╯al. (2000). Soluble organic nitrogen in agricultural soils. Biology & Fertility of Soils, 30, 374–382. Nyord, T., Sogaard, H. T., Hansen, M. N. and Jensen, L. S. (2008). Injection methods to reduce ammonia emission from volatile liquid fertilisers applied to growing crops. Biosystems Engineering, 100, 235–244. Oenema, O., Oudendag, D. and Velthof, G. L. (2007). Nutrient losses from manure management in the European Union. Livestock Science, 112, 261–272. Olesen, J. E., Askegaard, M. and Rasmussen, I. A. (2009). Winter cereal yields as affected by animal manure and green manure in organic arable farming. European Journal of Agronomy, 30, 119–128. Olesen, J. E., Jørgensen, L. N., Petersen, J. and Mortensen, J. V. (2003). Effects of rates and timing of nitrogen fertiliser on disease control by fungicides in winter wheat. 1. Crop yield and nitrogen uptake. Journal of Agricultural Science, 140, 1–13. Olesen, J. E., Schelde, K., Weiske, A. et╯al., (2006). Modelling greenhouse gas emissions from European conventional and organic dairy farms. Agriculture, Ecosystems & Environment, 112, 207–220.

228

Pierce, F. J. and Rice, C. W. (1988). Crop rotation and its impact of efficiency of water and nitrogen use. In:€Hargrove, W. L. (ed.) Cropping Strategies for Efficient Use of Water and Nitrogen, American Society of Agronomists, Madison, WI, pp.€101–113. Rotz, C. A., Taube, F., Russele, M. P. et╯al. (2005). Whole-farm perspectives of nutrient flows in grassland agriculture. Crop Science, 45, 2139–2159. Russell, C. A. and Fillery, I. R. P. (1996). Estimates of lupin belowground biomass nitrogen, dry matter, and nitrogen turnover to wheat. Australian Journal of Agricultural Research, 47, 1047–1059. Smith, P., Smith, J. U., Powlson, D. S. et╯al. (1997). A comparison of the performance of nine soil organic matter models using datasets from seven long-term experiments. Geoderma, 81, 153–225. Sommer, S. G. (2001). Effect of composting on nutrient loss and nitrogen availability of cattle deep litter. European Journal of Agronomy, 14, 123–133. Sommer, S. G., Friis, E., Bach, A. and Schjorring, J. K. (1997). Ammonia volatilization from pig slurry applied with trail hoses or broadspread to winter wheat:€effects of crop developmental stage, microclimate, and leaf ammonia absorption. Journal of Environmental Quality, 26, 1153–1160. Sommer, S. G., Géneremont, S., Cellier, P. et╯al. (2003). Processes controlling ammonia emission from livestock slurry in the field. European Journal of Agronomy, 19, 465–486. Sommer, S. G., Oenema, O., Chadwick, D. et╯al. (2006). Algorithms determining ammonia emission from buildings housing cattle and pigs and from manure stores. Agronomy Journal, 89, 264–335. Starmans, D. A. J. and Van der Hoek, K. W. (eds.) (2007). Ammonia: The Case of the Netherlands, Wageningen Academic Publishers, Wageningen, The Netherlands. Thorup-Kristensen, K. (1994). The effect of nitrogen catch crop species on the nitrogen nutrition of succeeding crops. Fertilizer Research, 37, 227–234. Van Duinkerken, G., André, G., Smits, M. C. J., Monteny, G. J. and Šebek, L. B. J. (2005). Effect of rumen-degradable protein balance and forage type on bulk milk urea concentration and emission of ammonia from dairy cow houses. Journal of Dairy Science, 88, 1099–1112.

Chapter

11

Nitrogen flows and fate in rural landscapes Lead author: Pierre Cellier Contributing authors: Patrick Durand, Nick Hutchings, Ulli Dragosits, Mark Theobald, Jean-Louis Drouet, Oene Oenema, Albert Bleeker, Lutz Breuer, Tommy Dalgaard, Sylvia Duretz, Johannes Kros, Benjamin Loubet, Joergen Eivind Olesen, Philippe Mérot, Valérie Viaud, Wim de Vries and Mark A. Sutton

Executive summary Nature of the problem • The transfer of nitrogen by either farm management activities or natural processes (through the atmosphere and the hydrological network) can feed into the N cascade and lead to indirect and unexpected reactive nitrogen emissions. • This transfer can lead to large N deposition rates and impacts to sensitive ecosystems. It can also promote further N2O emission in areas where conditions are more favourable for denitrification. • In rural landscapes, the relevant scale is the scale where N is managed by farm activities and where environmental measures are applied.

Approaches • Mitigating nitrogen at landscape scale requires consideration of the interactions between natural and anthropogenic (i.e. farm management) processes. • Owing to the complex nature and spatial extent of rural landscapes, experimental assessments of reactive N flows at this scale are difficult and often incomplete. It should include measurement of N flows in the different compartments of the environment and comprehensive datasets on the environment (soils, hydrology, land use, etc.) and on farm management. • Modelling is the preferred tool to investigate the complex relationships between anthropogenic and natural processes at landscape scale although verification by measurements is required. Up to now, no model includes all the components of landscape scale N flows:€farm functioning, short range atmospheric transfer, hydrology and ecosystem modelling.

Key findings/state of knowledge • The way N is managed, as well as the location of farming activities, can have a strong influence on N flows at landscape scale. Consequently, environmental measures can be more or less effective according to the landscape and farming system, and the interactions between them. • The magnitude of nitrate transfers and subsequent impacts is linked to the hydrology of the area (e.g. subsurface versus deep hydrological flows). • Source–sink relationships for atmospheric transfer are linked to land use (e.g. patchiness, hedgerows) and distance between sources and sensitive areas. • A verified integrated landscape model would be useful for investigating the N flows in rural landscapes, as well as evaluating different N management strategies and environmental measures at the landscape scale.

Major uncertainties/challenges • The multiple pathways of N transfer, the interactions between natural and anthropogenic processes and the risk of pollution swapping require complex high resolution modelling. Linkage of the different model components and the verification and uncertainty assessment of the integrated model are major challenges. • A network of European landscapes, including different climatic conditions, hydrology and farming systems, should be established as case studies to assess the influence of landscape processes on N budgets. • When designing and implementing new environmental measures, greater attention should be given to the landscape scale in order to take into account processes (such as N deposition to sensitive areas or indirect N2O emissions) that maximize the efficiency of the measures.

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

229

Nitrogen flows and fate in rural landscapes

Recommendations • The implementation of environmental measures should consider the variety of landscape types and allow adaptation to local conditions since their effectiveness might vary according to landscape features and farming systems. • Environmental measures applied to different landscapes and farming systems should be established and evaluated by modelling and verified, if possible, by monitoring once the measures are in place.

11.1╇ Introduction Rural landscapes, especially in Europe where there is a long history of agriculture and forestry, were mostly shaped by man in the past decades and centuries. For a long time, the concept of landscape was mainly related to its aesthetic quality as a portion of the earth surface captured by human eye. In landscape ecology, landscape is often described using three concepts:€ patch, corridor, and matrix (Forman and Godron, 1986). Patch is a ‘nonlinear surface area differing in appearance from its surroundings’. The mosaic of patches evolves and changes according to changes in land use (e.g. de(re)forestation, urban/road construction) and succession (e.g. crop rotation, grassland–cropland succession). Corridors are ‘narrow strips of land which differ from the matrix on either side’. Roads and water streams represent landscape corridors, as well as hedgerows, ditches and grassed strips. Roads are ‘disturbance corridors’, whereas rivers are ‘environmental resource corridors’. Matrix is the ‘most extensive and most connected landscape element type, and therefore plays the dominant role in the landscape’ functioning. Landscape is thus understood as a spatially heterogeneous mosaic (Forman and Godron, 1981) with interactions between ecological and anthropogenic processes (e.g. farm management, rural development, land conversion). The study of these interactions provides a practical dimension to landscape, because it is at this scale that planning, management, conservation, and land use change occur (Rapport et╯al., 1998). This approach of the landscape, which was initially designed for biodiversity issues (Forman, 1997), is relevant to describe the structure of a landscape for other purposes, and can, to some extent, be applied to nitrogen (N) issues (Liu and Taylor, 2002). In European rural landscapes with high inputs of N, cropland/grassland constitutes the matrix in most cases, although in e.g. Northern Europe the matrix might be the forest. The most relevant patches would be hot-spots in emissions or deposition. For instance, the livestock buildings of farmsteads are large point sources of atmospheric ammonia (NH3), forests are patches with potentially large atmospheric deposition rates and wetlands could be patches consisting of sinks for nitrate (NO3−), but sources for nitrous oxide (N2O). The corridors also have a role in N transfer and transformation, which will make them of specific interest. The rivers and ditches play a key role in NO3– transfer and denitrification in riparian zones, as well as water and N retention time. The roads and tracks network influences fertilizer transfer by the farmer and cattle displacements. However, for N issues, the cropland/grassland matrix cannot be considered as a homogeneous medium. It is itself a mosaic of sources and sinks of N according to the

230

crop type, management practices and proximity of reactive N (Nr) sources. To this extent, the distinction between matrix and patches is not straightforward. Moreover, N transfers mostly occur through the atmosphere, the hydrological network and farm management, which means that the connectivity between landscape elements for N transfer is different to that for biodiversity issues. At large spatial scales, either global (Turner et╯al., 1994) or European scales (Bouma et╯al., 1998; Meeus, 1993), a variety of regionally differentiated landscapes is observed. This is mainly due to the ecological adaptation to different constraints (such as geology and climate), and due to integration of agriculture production into regional socio-economic context (food industries, transport pathways (e.g. roads, rivers, valleys and urban areas)). The individual patch areas and the spatial density of linear structures may vary over several orders of magnitude depending on the region and landscape in question. In agricultural areas, one of the main driving forces of landscape design processes is the farming system which is often linked to a regional differentiation of agriculture (Westergaard, 2005). As an example, Figure 11.1 shows the difference in land use distribution for two dairy farming systems in Brittany (France) with different levels of intensification. In this chapter, we first highlight why the landscape scale is relevant for N issues, from the point of view of process analysis, flux estimation and agro-environmental policies. In the two next sections, we analyse how a landscape can be described and what are the processes which are the most relevant at landscape scale compared to plot scale. This leads on to the next section, which examines to what extent modelling is able to simulate landscape scale processes leading to practical application and scenario analysis. In the last section, we discuss the opportunities to integrate the landscape perspective into N assessment and management and conclude with the future challenges at landscape scale.

11.2╇ Why consider the landscape for N issues? 11.2.1╇ The N cascade in rural landscapes In rural landscapes, Nr mainly comes from fertilizers and livestock production. Plants absorb mineral N and mainly transform it into organic forms. Animals transform organic N from pasture or feed coming from either within or outside the farm into other forms. Hence, most of this N is managed by man. The amount of N that is manipulated, the methods and the timing

Pierre Cellier

Figure 11.1 Two landscapes composed of dairy farms in the “Zone-atelier Pleine-Fougéres” in Brittany (western France). In landscape (a), farm areas are large, field patterns are clustered around the farmstead (shown in red) and enable an intensive use of space (large field) with specialized patches of cash crop, forage and pastures. In landscape (b), farms are smaller than in (a); field patterns are fragmented, scattered and dispersed; crops, forage and pastures are very mixed in the landscape giving a heterogeneous crop mosaic.

of production, storage and application depend a lot on farming system and on the production intensity (see Jarvis et╯al., 2011, Chapter 10, this volume), which are also related to climate and to the links between agriculture and agro-industry. At field or farmstead scales, processes of N transformation and transfer have been extensively studied, and have given a fair insight into the fate of N at small space and time scales. When going beyond the field or farmstead boundaries (i.e. the landscape, watershed, regional scales), N can be transferred in significant amounts from Nr sources (e.g. farmsteads, field after slurry/fertilizer application, etc.) to the recipient ecosystems by a variety of pathways. For example, atmospheric NH3 emitted from animal housing or a field can be re-deposited to the foliage of nearby ecosystems in amounts that increase the closer the source is horizontally to the recipient ecosystem and vertically to the soil surface (Fowler et╯al., 1998; Loubet et╯al., 2006). Similarly, wetlands or crops/grasslands at the bottom of slopes can recapture NO3− in the groundwater that originates from N applied further up the slope. In both cases, this can lead to large inputs of N to the receptor ecosystem that may have potential impacts on the ecosystem (Pitcairn et╯al., 2003) and the biogeochemical cycles, possibly leading to enhanced N2O and NO emission (Beaujouan et╯al., 2001; Skiba et╯al., 2004) and further feeding the N cascade (Galloway et╯al., 2003) (Figure 11.2). These N2O emissions resulting from N transfer in receptor ecosystem are usually called indirect emissions and may represent a significant fraction of total N2O emissions, although how much remains uncertain (Mosier et╯al., 1998). The importance of uncultivated or marginal areas that are outside or peripheral to the agricultural systems for flows and budgets of energy and matter, including N, emphasizes the need to adopt a landscape perspective.

11.2.2╇ Consequences of heterogeneity on N flows and budgets When going from the plot scale to the landscape scale, one major new feature that appears is the heterogeneity in land use, in natural features and in farming activities (location of fields/grasslands/forests/ditches, hedgerows, livestock holdings, N application). A range of processes linked to the spatial heterogeneity, either natural or anthropogenic (mainly farm scale), has to be considered, such as non-random application of N at farm scale and the interaction between the farmstead and landscape features (e.g. soil, topography), NH3 transfer and deposition to vegetation, especially forest and hedgerows, N2O emissions from wetlands and streams, preferential pathways for N through the ditches and tree belts networks. As a whole, the fluxes of deposition (atmosphere) or recapture (groundwater) are most important when N flows from one system to another with different characteristics (see e.g., Beaujouan et╯al., 2001; Loubet et╯al., 2009). For example, the deposition of NH3 is especially large at forest edges because the abrupt change in canopy type/height increases the turbulent exchange and the surface area of vegetation in contact with the plume from a nearby source (Fowler et╯al., 1998; Weathers et╯al., 2001). This leads to hot-spots in deposition (Dragosits et╯al., 2002). For example, Loubet et╯al. (2009) estimated that a forest belt can capture more than 15% of the emission from an animal house. For hydrological transfer, Beaujouan et╯al. (2001) estimated€– by modelling the water and N flow through the hydrological network, including possible N removal by soil/plants in wetlands and denitrification€ – that most of N2O emission is expected to occur in wetlands, i.e. in places where N has not been directly applied but has been transported by hydrological transfer (Oehler et╯al., 2009; Figure 11.3). Beaujouan et╯al.

231

Nitrogen flows and fate in rural landscapes Figure 11.2 The Nitrogen Cascade in rural landscapes (adapted from Sutton et al., 2011, Chapter 1, this volume).

Figure 11.3 Modelling of the spatio-temporal extension of soil saturation due to the rise of groundwater table (left) and subsequent denitrification (right) using the TNT2 model in a rural catchment (Britanny, France). After Oehler et€al., 2009 (modified) and unpublished data.

(2001) also suggested that the recapture of aqueous NO3− was greatest when landscape fragmentation was largest and sources and sinks more intimately mixed.

11.2.3╇ Landscape as a scale to mitigate adverse effects of N Processes of either recapture or transformation of N can be used for mitigating fluxes at larger scale than the scale at which the Nr is applied or produced. An example can be given for the case of NH3. Agricultural sources of NH3 are, by their nature, quite localized, e.g. a fertilized field or an animal house. This

232

means that the exposure of a receptor and the related deposition is largely determined by the spatial relationships between the receptor and the nearby sources. Hence, by deliberately locating ‘sink’ vegetation downwind of a source, the local recapture of atmospheric NH3 and the enhancement of turbulent mixing in the low atmosphere can be used to mitigate the impacts of atmospheric NH3 further downwind by reducing the concentration within the plume near the surface (Theobald et╯al., 2001; Dragosits et╯al., 2006; Loubet et╯al., 2009). Possible strategies include planting of tree belts (to enhance local deposition and dispersion) in order to reduce the NH3 deposition to sensitive receptors further downwind (Sutton et╯al., 2004).

Pierre Cellier

Similarly, restoration, management or even construction of wetlands (riparian strips, slow flowing meanders, ponds, etc.) along the course of small rural streams has often been proposed as an efficient measure to mitigate surface water contamination by NO3− leaching from agricultural land (Haycock et╯al., 1997; Woltemade, 2000; Tanner et╯al., 2003). Indeed, it has been estimated in a number of regional catchments that denitrification in riparian wetlands stops 20%–60% of N coming from diffuse sources from entering the drainage network (Billen and Garnier, 2000; see Billen et╯al., 2011, Chapter 13, this volume). At the landscape scale, the efficiency of riparian wetlands depends strongly on the hydrological setting (Haag and Kaupenjohann, 2001). However, it has to be borne in mind that such measure could give rise to pollution swapping (Butterbach-Bahl et╯al., 2011, Chapter 6, this volume) due to possible enhanced N2O emissions. Similarly, the introduction of extensive farm management such as set-aside grassland instead of intensive cropland in designated environmentally sensitive areas may be an efficient way to protect groundwater quality. This was successfully studied in Denmark by modelling (Dalgaard, 2009) and in France by practical application (Vittel mineral water protection area; Deffontaines et╯al., 1994; Gras and Benoît, 1998).

11.2.4╇ Synthesis:€relevance of the landscape scale for environmental and policy issues As illustrated above, the principal issue concerning N at the landscape scale is the question of N transfer at short distance (101–103 m) by atmospheric or hydrological processes or by farm management transfer. The magnitude of this transfer is linked to the magnitude of the sources, the relative positions between sources and sinks, the heterogeneity of the landscape and the type of Nr (e.g. NH3 is generally deposited nearer the source than NOx). Secondly, the landscape is composed of a range of ecosystems and anthropogenic systems (farmstead, roads, etc.) within which N cycling can be very different and which can have a large effect on the potential for N transformation and consequently lead to different Nr budgets. Thirdly, in rural landscapes, farm management is a key component, as farming systems are by far the main source of Nr. Moreover, it is expected that the consequences of agricultural practices and the choice of farming systems may be very different according to the environmental conditions (climate, topography, soil types, proximity of sensitive areas). These relationships determine the landscape function and highlight a dynamic view of what a landscape is (Leibowitz et╯al., 2000). Any change in landscape structure will change the dynamics of N flows at field and landscape scales. This gives rise to questions such as:€what are the consequences of N transfer on the production of Nr or vice versa? What is the influence of landscape features on N transfer? And to what extent can farm management be adapted to landscape conditions to help mitigate the emissions of Nr? This gives an insight into the possibility to adapt environmental policies to regional conditions. To some extent, the

landscape scale is a very practical scale for researching solutions to N related problems as it considers both farming systems and environmental features. One practical application that is especially relevant at landscape scale is the protection of sensitive areas. Assessing the threat from nearby activities to these areas requires an estimation of the N flow from the source to the receptor, which itself requires knowledge of the patchwork of sources and sinks in and around the sensitive area, the intensity of agricultural activities, the features of nearby ecosystems and the conditions for atmospheric dispersion or N flow in the soil and aquatic systems.

11.3╇ Landscape description and functioning for N issues 11.3.1╇ Landscape characterization Landscape scale In all parts of the world, but more especially in Western Europe where the anthropogenic influence on the environment is large, a great variety of landscapes can be observed. This gives rise to the question of what distinguishes one landscape from another. Forman defined the landscape as follows in 1997:€‘A landscape is a mosaic where the mix of local ecosystems or land uses is repeated in similar form over a kilometre-wide area. Within a landscape several attributes tend to be similar and repeated across the whole area, including geologic land forms, soil types, vegetation types, local faunas, natural disturbance regimes, land uses and human aggregation patterns. Thus a repeated cluster of spatial elements characterises a landscape’. This highlights that, despite large small scale variability, there is a scale where some degree of homogeneity can be observed. This implies a definition of the landscape scale, i.e. an area of several square kilometres. It could certainly be much larger in regions that are more uniform than in Europe. Defining the landscape scale for N issues, i.e. the transfer processes and farm management, a landscape consists of a repeated cluster of small catchments (typically several hectares to square kilometres each) for hydrological transfer, a repeated mosaic of ecosystems (including farmsteads) for atmospheric transfer, and several farms. Considering the spatial scale for atmospheric and hydrological processes (typically 101– 103â•›m) and the average size of farms in Europe (101–102 ha), the landscape scale can be considered to represent domains ranging from 1 km² to 100 km², with an interest in the spatial interactions within this domain, such as may occur on scales of a few metres to several kilometres. At this scale, it must be noted that all the landscapes elements are under the influence of the same climate and they may share a similar geomorphology.

Characterization of landscape elements A first indicator characterizing the landscape is the relative area of the different land cover types related to the area of interest (Willems et╯al., 2000). Sometimes this indicator is related to some specific landscape structure, e.g. the percentage of forest

233

Nitrogen flows and fate in rural landscapes

within a wetland (Vogt et╯al., 2004) or the valley bottom. For linear structures, a similar rough approach is based on relative abundance measured as a density (length of a specific linear structure related to a specific area). Such indexes are well suited for large scales and allow us to describe the major trends at that scale. Nevertheless it does not account for the landscape heterogeneity and the spatial arrangement. The predictive power of such statistical approach fails in small catchments (less than 1–10 km²), suggesting that the spatial arrangement of landscape patches may become critical at these small scales (Strayer et╯al., 2003). Beyond the main mosaic of sources and sinks that characterize the landscape, several elements have a specific importance and can be described according to landscape functioning. They are represented in Figure 11.4 (Haag and Kaupenjohann 2001). Ecotones and corridors These refer to the relation between ecosystems inside the landscape. The corridors determine the connectivity of the landscape. For N issues, they are areas within which there is a high rate of N transport, relative to the rate of change in the N transformations. The ecotones are ecological transition areas between two ecosystems (e.g. riparian zones between cropland and the river). There is a high rate of change in space in the nature of the N transformations (e.g. denitrification) among these areas. Hedgerows This is one of the archetypal landscape elements that may influence the biophysical and ecological functioning of the whole landscape. A large diversity of hedgerow structures and densities can be observed in European countries. The recent introduction of field margin ecology in Europe (Marshall,

2002) highlighted the biophysical and ecological interest of hedgerows (Baudry et╯al., 2000; McCollins, 2000) and showed their interest to consider the lateral transfer of N in the landscape. Hot spots and buffer zones In hydrological systems, hot spots occur where flow paths meet substrates or other flow paths containing complementary or missing reactants. It might also occur in ecosystems where surface or subsurface conditions are different from the surroundings, promoting transformation processes (e.g. denitrification in grassed strips along rivers). They can also be related to deposition to sensitive ecosystems where critical loads can be exceeded due to the proximity of a strong source (Dragosits et╯al., 2002; Loubet et╯al., 2009) or changes in surface conditions (e.g. forest edge). When N retention or transformation is observed, these structures may also be called ‘buffer zones’. These processes give them a specific function in the N cascade and potential for mitigation and they have become an important subject of research and a management tool for reducing pollution.

Importance of the location of landscape elements Because landscape elements clearly interact with the other structures of the environment, we need to take into account the precise location of these landscape elements relative to the sources and sinks of Nr. This specifically applies on the pathway of N pollutants along the transfer lines. Both hydrological and atmospheric transfer are concerned, the effectiveness of which might depend on transfer conditions, e.g. on groundwater depth or meteorological conditions. For example, a landscape structure such as a hedge will not have the same impact on N fluxes in a calcareous environment with deep groundwater level than the same hedge located on a soil with a shallow Figure 11.4 Scheme of corridors and retention compartments. The sequence of compartments depends upon the specific hydrological setting and is spatio-temporally variable (redrawn from Haag and Kaupenjohann, 2001) with permission from Elsevier.

234

Pierre Cellier

groundwater level. In the second case the ground flow goes through the superficial soil and the rooting zone, making it possible for ecosystem to capture NO3− and for denitrification and subsequent N2O emissions to occur. Similarly, the effect of landscape structures like riparian buffer zones on N flows and transformation depends on their place within the catchment, up- or downstream (Mourier et╯al., 2008). The same principles apply to atmospheric transport at a landscape scale. For example, the potential benefit of a wooded buffer zone for NH3 dispersion will depend on the nature of the source and the location, dimensions and structure of woodland, as well as the location of the receptor area to be protected. All this clearly indicates that the spatial location of all landscape elements must be explicitly accounted for in landscape description and modelling.

11.3.2╇ Interactions between farming systems and landscape structure Spatial organization of crop mosaic and farm practices Rural landscapes result from the aggregation of multiple farms and their relationship with other land uses. The location of the farmsteads (more or less dispersed) play a significant role in the design of landscape patterns. Within a single agricultural region, landscape pattern results to a large extent from decisions made at the farm level and how the farming systems integrate at landscape level (Deffontaines et╯al., 1995). Firstly, farming systems control the composition of the landscape mosaic in terms of surface area used for agricultural production (arable fields, grasslands). Secondly, for a given farming system, crop allocation to the fields is controlled by the combination of agronomic constraints for crop succession (Colbach et╯al., 1997), environment constraints, soil quality (Stockle et╯al., 2003), and specific constraints of the farm field pattern, including accessibility, field size, distance to farmstead (Thenail and Baudry, 2004; Rounsevell et╯al., 2003), and market forces (e.g. quotas, market prices). The relative balance and hierarchy between the constraints mentioned above, differ among the farming systems. Thenail (2002) has emphasized the strong spatial pattern of the crop mosaic of dairy farms in north-eastern Brittany (western France), which is to a large extent determined by distance to farmstead. Land use is organized into approximate concentric circles around the farmstead (Figure 11.5):€pastures grazed by dairy cows are located as close to the farmstead as possible, because dairy cows move daily from the farmstead into the fields. A second circle consists of fields used for crops and forage. The outer circle consists of permanent grasslands grazed by heifers, extensive lands or woodlands, which require little management. This applies to many locations in north-western Europe. By contrast, crop allocation in crop farming systems or intensive breeding farming systems are expected to be less controlled by the distance to farmstead. More generally, the degree of spatial specialization varies according to the farming systems, the diversity of crop rotations and the specific constraints of the farm.

Figure 11.5 Theoretical organization of the crop mosaic according to the distance to the farmstead in dairy farms in Brittany (western France) (redrafted from Thenail and Bandry, 2002). With permission from Elsevier.

Interactions between farming systems and the semi-natural landscape elements Strong interactions exist between farming systems, the crop/ grassland mosaic, and the pattern of semi-natural, often perennial, landscape elements such as those located on the field margins (grass strips, hedges, ditches, woodland plot, wetlands, etc.). Thenail and Baudry (2004) found different degrees of landuse allocation in farms depending on the hedgerow density:€the presence of hedgerows distorts to a certain extent the concentric pattern of crop allocation described above in dairy farms. In Jutland (Denmark), Kristensen et╯al. (2001) showed that the management of hedgerows, woodlands, and permanent grasslands varies according to the type of farming systems. Thenail and Baudry (2005) have focused on the management of small riparian wetlands and their interaction with farming systems. In the example of dairy farms in Brittany (western France, see, for example, Figure 11.5) where riparian wetlands form a large part of the farm area and are located close to the farmstead, they are intensively used for animal grazing or fodder. On the contrary, where they represent a small part of the farm area, or where they are far away from the farmstead, they tend to be abandoned by the farmers.

11.4╇ N transfer and transformation processes from the plot to the landscape scale Detailed descriptions of N processing in terrestrial, freshwater and atmospheric systems are presented in Butterbach-Bahl

235

Nitrogen flows and fate in rural landscapes

et╯al., 2011; Durand et╯al., 2011; Hertel et╯al., 2011 (Chapters 6, 7 and 9 this volume). N transfers at the farm scale are described in Jarvis et╯al. (2011, Chapter 10, this volume). In this section we specifically highlight the ecosystem, hydrological and atmospheric processes that contribute most to interactions and modify fluxes and budgets at the landscape scale.

11.4.1╇ N processes at ecosystem scale The landscape scale is first characterized by a variety of interlinked ecosystems of varying sizes. In cropland and some grassland, the large inputs of mineral fertilizer create a much larger pool of Nr than in natural ecosystems. A key question at landscape scale is that of the spatial transfer of N from eutrophic to oligotrophic ecosystems, leading to N impacts in the latter. Consequently, there should be a special focus on the understanding of environmental conditions and C–N turnover and transformation in these ecosystems with low direct N input but significant indirect N input through lateral transfers (e.g. wetlands, forests, grass strips). This means that the N turnover in the soil and vegetation (litter, dead leaves) have higher relevance than in agro-ecosystems with high direct N input (see Butterbach-Bahl et╯al., 2011, Chapter 6 this volume). It is also necessary to better understand and quantify the processes of N capture in these oligotrophic ecosystems, from both the atmosphere (dry and wet deposition) and the soil water (uptake from the groundwater to the biogeochemically active upper soil layers). In the former, NH3 and NO2 absorption by stomata (including compensation point modelling) and further recycling in the plant metabolism must be considered, as well as N capture by the leaf surface and subsequent transfer to the soil surface by rain washing (Hertel et╯al., 2011, Chapter€ 9, this volume). For this, a minimum description of the canopy structure (height, leaf area density, etc.) is essential. In the latter, an understanding is needed of the rooting depth, groundwater depth and the water transfer in the soil and nutrient absorption by roots under conditions close to saturation and anoxia (Beaujouan et╯al., 2001; Durand et╯al., 2011, Chapter 7 this volume).

11.4.2╇ Vertical and lateral transport processes Surface and deep hydrology Water transport in the natural environment can be roughly separated into vertical flow (e.g. water infiltration from the surface to groundwater) and lateral flow (surface runoff, subsurface and deep lateral flow, base flow) (Cirmo and McDonnell, 1997). For landscape analysis, two types of catchment are generally considered:€shallow groundwater catchments, in which the vertical flow feeds into the subsurface lateral flow, and deep groundwater watersheds (e.g. karst situations) with deep vertical flow. In the first case, surface hydrology can lead to significant redistribution of N that feeds into the N cascade and modifies Nr fluxes and budgets. In the second case, hydrology generally does not create significant interactions at the landscape scale. In the majority of situations, the relative importance

236

of shallow and deep water pathways varies rapidly in space and time. These pathways have very different time scales (minutes to hours/days for surface flow, months to years/decades for deep flow), resulting in complex patterns of residence times and seasonal variations. A consequence is that the fate of the N applied in a catchment and its impact on stream water quality, which is always a mixture of waters with contrasting histories (Boehlke and Denver, 1995; Durand and Juan Torres, 1996), will depend on the location of its application within the catchment and on the landscape structure. Some specific events can occur at some places in the landscape. In areas with an impermeable layer close to the soil base, a significant lateral flow can occur beneath or in the soil, with consequences for N transport (Molenat et╯al., 2008). In some places such as wetlands, the surface water–groundwater interaction (Dahm et╯al., 1998) might be very important for N capture by vegetation and subsequent possible denitrification. This might be important for N2O and NO3− budgets at landscape scale and it is therefore an issue in landscape modelling (Beaujouan et╯al., 2001). As a whole, due mainly to differences in local water balance and pathways, NO3– leaching is larger at points further down the slope, and denitrification and plant recapture is more important downhill. Consequently the interactions between ecosystems could require more attention in the valleys, e.g. at the wetlands–arable land/grassland interface.

Atmospheric transfer In rural landscapes, emissions of Nr into the atmosphere are predominantly the result of agricultural activities. In these landscapes, mainly NH3, but also N2O and N oxides are emitted from livestock housing, the storage of manures and slurries and the application of organic and inorganic fertilizers to fields. Some of these gases are also emitted, to a much lesser extent, from mobile agricultural sources (trucks, tractors, etc.). Although N2O emissions can be strongly affected by landscape structure, once emitted, N2O does not interact significantly with the landscape. Similarly, although atmospheric NOx concentrations can vary substantially across landscape, the dry deposition velocities of NO2 are small, so that it generally has only a small influence on local spatial patterns of total Nr deposition. By contrast, NH3 dispersion and deposition is very important for processes at the landscape scale because it is both subject to relatively large local emission variations (farmstead, field) and high dry deposition velocities. Therefore high atmospheric concentrations and large deposition rates can occur close to the source (Fowler et╯al., 1998; Van Pul et╯al., 2008; Dragosits et╯al., 2002). Uncertainty analysis shows that most of the uncertainty in predicting the fate of atmospheric NH3 is due to the uncertainty in deposition processes (Loubet et╯al., 2009), including compensation points.

11.4.3╇ Transfer linked to farm activity As mentioned above, farm activity is by far the main source of Nr either in mineral or organic form in European rural landscapes. Large amounts of organic matter, and hence

Pierre Cellier

N, are manipulated by farm operations (manure, harvest). Moreover, crop fertilization and animal feed often result in a large import of N to the farm, which is distributed within the landscape according to farm management. In contrast, crop harvest and animal production lead to exports from the farm and landscape. The magnitude of these N transfers and their organization depend on the farming system and farmer’s decision making. They are described in more detail in Jarvis et╯al. (2011, Chapter 10 this volume). The location of the crops and grasslands relative to the other ecosystems, as well as animal displacement, also depend to a large extent on the farming system.

11.4.4╇ Anthropogenic modifications of transport processes Anthropogenic structures and activities such as urban areas, transport pathways (roads, tracks, canals) or agriculture modify the natural pathway of water in most places in Europe. In agricultural land, some specific modifications can influence N flows at different scales. • Ploughing (plough layer) and soil compaction by large machinery (Lipiec and Pniewski, 1995) has a large effect on water infiltration and might increase surface lateral flow and modify the local hydrology and related N flows. • High density ditch drainage systems have traditionally existed in most European landscapes for centuries and tile drainage has also been set up in recent decades. Both have a strong impact on lateral flows of water and dissolved N and hence modify the transfer time of water as well as spatial relationships between ecosystems. They have significant effects on NO3− leaching (Dinnes et╯al., 2002) and N2O emissions (Reay et╯al., 2003). • As already mentioned, the presence of woodland and hedgerows in a landscape increase the surface roughness and hence the atmospheric dispersion. Moreover, patchiness increases the number of hot-spots in local deposition, as deposition is the largest at transition zones such as forest edges (Loubet et╯al., 2009). • Constructed wetlands are implemented in many countries because they have the potential for reducing NO3− contamination from agricultural areas (Spieles and Mitsch, 2000, Woltemade, 2000; Tanner et╯al., 2003). However, there is suspicion that they may emit significant amounts of N2O and CH4 (Sovik et╯al., 2006). It is thus of utmost importance to study these features in close relation with studies that are performed on natural wetlands.

11.4.5╇ Conclusion on landscape N transfer and transformation processes It is necessary to combine knowledge of the spatial distribution and extent of N sources with knowledge of the pathways connecting those sources with adjacent aquatic or terrestrial ecosystems, in order to understand the relationship between anthropogenic sources and natural receptors. This is also

necessary to both predict the impact of changing land use and management on N fluxes and budgets at plot (as a source) and landscape scale. This requires an improvement of our understanding of some anthropogenic drivers (e.g. water/N flows from farmsteads, transfer through ditch networks) and of some specific processes in natural ecosystems (e.g. recapture and transformation of N€– coming from upslope€– by soil and vegetation in wetlands). As mentioned before, all the relevant processes at landscape scale have similar space scales, between several metres and several kilometres, but the frame is different according to the type of transfer. For hydrological transfer, the catchment is obviously the relevant scale, with the watershed limits€ – where a nil flux condition can generally be applied€– giving the boundaries of the domain. Such limits and conditions do not exist for atmospheric transfers. The simulation domain is often a square and it is necessary to prescribe boundary conditions from measurements or from a higher scale model. For farm activities, the domain consists of the farmstead and the fields and naturals areas depending on the farm. It is a discontinuous domain with some of the fields possibly outside the studied landscape. Conversely some parts of the landscape can be attached to farmsteads outside the landscape. This mismatch between the domains for the different types of transfer makes it difficult to have a unified approach in investigating landscape from an N perspective. Nevertheless the consistency between space scales is a facilitating factor. Considering the time scales, the range is much larger. The farm scale processes are event-based and proceed along the crop cycle, the animal breeding cycle or the year. The natural processes can be very short (seconds to minutes) like NH3 deposition close to a farm building or surface run-off during a heavy rain, or very long. The latter is the case of hydrological transfer which can last several years or decades between the rainfall and the exit at the catchment outlet. As a whole, at landscape scale, atmospheric processes have short time scales (seconds to day) and hydrological transfer have much longer time scales, from days to years. Consequently, it is difficult to establish relationships between N inputs and outputs from a given landscape and to assess budgets at landscape scale both from experimental and modelling points of view, unless working on the long term.

11.5╇ Landscape modelling of N Although measurements have been made of N flows between individual landscape elements such as the transfer of atmospheric NH3 from a source to downwind vegetation (see Loubet et╯al., 2009; Theobald et╯al., 2001) or N flows along drainage or stream networks (see Boehlke and Denver, 1995; Dahm et╯al., 1998; Molenat et╯al., 2008). N flows within and across entire landscapes are still beyond the capabilities of current technology or research budgets. For example, remote sensing can be used to study the spatial distribution of atmospheric NH3 (Clarisse et╯al., 2009) but the current horizontal and vertical resolution of this technique is not adequate for the landscape scale. The simultaneous measurement of the flows and

237

Nitrogen flows and fate in rural landscapes

interactions of multiple Nr species through multiple media (atmospheric, hydrological and plant and soil systems) would require a complex network of sensors and a large amount of researcher time. Moreover, interpreting these measurements requires knowledge of the agricultural practices at field and farms scale all over the studied landscape and its surroundings. This might apply for several years, due to the time scales of Nr transfer. This type of experimental approach is currently beyond the capabilities of most research projects. This is one of the main reasons why the use of process modelling approaches seems to be the way forward to study landscape N flows and budgets. However, field data are still required to verify model predictions although measurements at a lower resolution (both spatially and temporally) can be used for this. Landscape modelling of N extends the process modelling approach of single component fluxes (e.g. emission/deposition/leaching) to follow N from source through the different compartments of a study landscape with multiple routes in the N cascade (including hydrology, atmosphere and farm management). By contrast to plot-scale modelling, landscape modelling focuses on spatial transport and transformation, both within and between the compartments of the landscape. Comprehensive landscape models must therefore take into account the nature, location and size of the emission sources, the distribution of land cover within the landscape, the hydrology, the meteorological conditions, etc. It should also account for transformation of N within the different components of the landscape. This approach requires a clear understanding of all landscape compartments, as well as the boundaries between them, and how Nr transfers across these boundaries. It also requires that attention be given to hot-spots of N emissions, which are the drivers of a large proportion of N transfer, as well as to more diffuse sources. Recycled fluxes in the cascade must also be considered, such as the fate of atmospherically deposited N, which is an important driver of impacts and further feeds the N cascade. Defining compartment boundaries between model domains is not trivial, because of the contrasting needs of the different model components. For example, the boundaries relevant for hydrological transfer (watershed) have no or little relevance for atmospheric transfer or anthropogenic farm transfer. Landscape models need to be appropriately calibrated and verified, which may use both spatial and temporal datasets. This section focuses on detailed comprehensive models and their ability to identify the main issues and investigate the relevance of some policy measures. However, these models can also provide the basis for setting up simpler models, i.e. dealing with a limited number of processes and/or considering simplified formulations of transfer and transformation processes. These models can be also be used to develop and support landscape management decisions.

elements, such as individual fields, livestock buildings, patches of woodland, hedgerows, streams, etc. These input data include properties such as soil type, building height, building ventilation rates as well as management activities related to farming, such as the application of mineral fertilizer or livestock manures, planting/sowing/harvesting of crops, and grazing/ housing of livestock. While average conditions and activity data may be adequate for regional scale modelling, real world farm management data are required to understand the flows of N in a specific landscape, as well as diurnal/seasonal/inter-annual variability. The input data should also have sufficient spatial resolution to consider small spatial elements (e.g. hedgerows, grassed strips) that are relevant for landscape processes. Environmental variables, such as temperature, precipitation, wind speed/direction, solar radiation, etc., are also required. However, the different components of a landscape model do not need to be as detailed as a single-compartment model (i.e. ecosystem, atmospheric, hydrological, farm model), where processes are investigated in greater detail. Moreover, while process models tend to focus on particular compounds (NH3, N2O, NO3−, etc.) and particular aspects of the N cycle (e.g. atmospheric transport modelling; catchment modelling, crop or grassland modelling), a landscape model needs to bind all these components together. One challenge in landscape modelling is thus to achieve the right balance between describing the details of the individual processes and ensuring consistency between the different models. An essential requirement for landscape modelling is that all elements and activities are assigned to a spatial location, i.e. a map location is recorded, and a spatial database developed, which provides input data for the model. It should be noted that, despite the assumed importance of corridors in transfers at landscape scale, these processes are generally not accounted for in landscape models. This is because it would generally require a very high spatial resolution (e.g. for a ditch network) and the processes are not yet well quantified. For N transfers in rural landscapes, such corridors are mainly relevant for farm transport of fertilizers, feed and products, and for riverine fluxes, which can be specified by the models. For practical application of landscape processes in regional models, it is not realistic to collect and use detailed field and farm input data. For this reason, landscape models are tested, for example landscapes where detailed datasets are collected. This means that for upscaling the findings of landscape models to the regional level, there is a need to generalize the processes and consequences, as well as to determine landscape typologies, based on global indicators for landscape structure and farming systems. Typologies could be derived from e.g. remote sensing data, vegetation and topography maps and regional agricultural censuses. At present, such indicators are still to be defined and their relevance assessed.

11.5.1╇ Key issues for comprehensive landscape modelling of N

11.5.2╇ Examples of landscape scale models for N

To be able to model the main interactions within a landscape, detailed input data are needed for a large number of landscape

238

Over the past few years, a number of modelling assessments have been carried out at the landscape scale. From a historical point of view, landscape modelling of N may be seen as a

Pierre Cellier

logical extension from the separate fields of interest in stream and groundwater flows of NO3− and in local scale atmospheric transport and deposition modelling of NH3 and NOx. In integrating these different elements, the approach must also account for ecosystem processes and be placed in the context of farm management. Current models have been developed with a focus on environmental science disciplines (hydrology, atmospheric sciences, farming system) or environmental issues (impacts on water quality, air pollution, sensitive ecosystems).

Modelling flux heterogeneity at the landscape scale The model Initiator2 (Integrated Nutrient ImpacT Assessment Tool On a Regional scale; De Vries et╯al., 2005) simulates (i) emissions of NH3 and greenhouse gases (CO2, CH4 and N2O) from animal housing systems and agricultural soils and (ii) leaching and runoff of nutrients (specifically N and phosphorus) from agricultural soils to groundwater and surface water. In this approach, the modelled NH3 emissions from fields and housing systems form the input to an atmospheric transport model (OPS; Van Jaarsveld, 1995), which is used to assess the N deposition to agricultural and non-agricultural systems using a grid resolution of 250 m. Initiator2 was used to make an integrated assessment of the present environmental status (year 2004) of the Noardlike Fryske Wâlden area (NFW) in the north of the Netherlands and of impacts of management measures that are being applied in the area (Figure 11.6). The input database contained animal numbers, agricultural practices and land management, such as manure application techniques for each farm in the NFW area, based on results from questionnaires from the Dutch Central Bureau on Statistics. This database was linked to detailed topographic data, spatially explicit soil data (soil map 1:€50.000) and hydrology (Kros et╯al., 2007). The results of the analysis for the NFW give an insight into the high spatial heterogeneity in NH3 and N2O emissions as well as in the NO3− concentration in upper groundwater. In almost 6% of the area the EC NO3− groundwater limit of 50€ mg€ l−1 was exceeded, even though for the NFW area as a whole, the average NO3− concentration was only 10 mg NO3 l−1. The map of N2O fluxes (Figure 11.6b) shows larger emissions for wet/ peaty locations, in contrast to the map of NO3− concentrations (Figure 11.6c), where the values are largest for dry/sandy locations. However, this type of model cannot capture the lateral spatial interactions at the landscape scale, as it does not simulate small scale (i.e. 10–100 m) atmospheric and hydrological processes.

Modelling N interactions at the landscape scale A number of modelling studies have recently been carried out to address spatial N interactions at the landscape scale, including farm management. In the following paragraphs, examples of different approaches for landscape modelling are briefly reviewed. The focus is not on the technical aspects of the individual models, but on how the flows and transformations of N are represented, and the insights gained from working at this scale. The UK LANAS integrated model (Landscape Analysis of Nitrogen and Abatement Strategies) is centred around

atmosphere, ecosystem and soil interactions, with leaching only included as an end point, i.e. vertical or horizontal underground/in-stream flows of N are not represented (Theobald et╯al., 2004; Dragosits et╯al., 2005). The LANAS model consists of established process-based models for the main components of the landscape, NGauge (used at field scale for grassland systems, Scholefield et╯al., 1991), SUNDIAL (crop systems, Smith et╯al., 1996), and LADD (atmosphere, Hill 1998; Dragosits et╯al., 2002). These models and a simple farmstead model, FYNE (Theobald et╯al., 2004), were coupled via a ‘wrapper’ programme to control the data exchanges through a spatial database which stores, sends and receives input and output to/from the component models during simulation. Vertical and horizontal flows are only fully represented in the atmosphere component of the landscape, with the field models acting as plot models for each of the grass/crop fields in the landscape. Output from the LANAS model at the landscape scale includes NH 3 and nitrous oxide emissions, dry deposition of NH3 and leaching of NO3− out of the bottom of the ecosystem models. An example of LANAS output is given in Figure 11.9. The Danish ARLAS project (Dalgaard et╯al., 2002; Hutchings et╯al., 2004) focussed on farms, ecosystems and effects of N management on drinking water boreholes in an area of central Jutland (Denmark). The model developed under the ARLAS project did not include an atmospheric component model, so that emissions of NH3 to the atmosphere were not dispersed or deposited during the model simulation. The groundwater and hydrological component was of central interest in the project, which analysed how the water quality could be improved by restricting N losses from agricultural sources. The main aim of the model application was scenario testing, including the estimation of N-surpluses from organic farming (Dalgaard et╯al., 2002). Similarly, the introduction of extensive farm management such as set-aside grassland instead of intensive cropland in designated environmentally sensitive areas was shown to be an efficient way to protect groundwater quality (see Figure 11.7; Dalgaard, 2009). This illustrates the practical nature of landscape scale modelling with its emphasis on local sources, sinks and flows of N. The French EcoSpace project (Beaujouan et╯al., 2001) coupled a hydrological model (TNT, based on TOPMODEL, Beven and Kirkby, 1979; Beven, 1997) with an existing generic crop model (STICS, Brisson et╯al., 1998). The two models were coupled with the ‘soil store’ for N and water being controlled by STICS, and the ‘drainage store’ being controlled by TNT. An atmospheric model was developed to account for deposition to ecosystems close to farmsteads or slurry application, but it was not fully coupled to the STICS/TNT model. An example of the model output is given in Figure 11.3. The main aim of the study was to investigate mitigation options for the improvement of stream and surface water quality, and in particular to minimize N pollution. Beaujouan et╯al. (2001) used the model to evaluate the influence of the spatial distribution and size of patches of crops in six theoretical agricultural catchments of different types and shapes (convergent/parallel/intermediate catchments with either concave or convex

239

Nitrogen flows and fate in rural landscapes

(a)

(b)

NH3 emission application (kg NH3-N/ha) 0 - 10 10 - 20 20 - 30 30 - 40 >40

N2O emission (kg N2O-N/ha) 0-5 5 - 10 10 - 15 15 - 25 >25

0

2

4

0

6 Kilometers

2

4

6 kilometers

(c) NO3 groundwater (mg NO3/I) 0-5 5 - 10 10 - 25 25 - 50 >50

0

2

4

6 Kilometers

Figure 11.6 Maps calculated using Initiator2 for the NFW region in the Netherlands for 2004:€(a) annual NH3 emissions from manure application; (b) total annual N2O emissions; (c) nitrate concentrations in the upper groundwater (from Kros et al., 2007).

slopes). In the scenarios, source areas were placed upstream or downstream of sink areas, as well as spread in checkerboard patterns throughout the catchments. When applied to real cases, the model output compared favourably with real catchments in Brittany (NW France). More recently, a consortium of European research groups has started an ambitious project on landscape analysis of N interactions, as a component of the EU NitroEurope Integrated Project (see also www.nitroeurope.eu; Sutton et╯al., 2007). One of the aims of the landscape component of the project is the

240

joint development of an integrated landscape scale model, NitroScape, to simulate the flows of N between all components of rural landscapes. The NitroScape model is a framework coupling suitable existing component models for the atmosphere, ecosystems and hydrological components, as well as farm scale processes, with a spatial database (Cellier et╯al., 2006; S. Duretz, personal communication, 2010) (Figure 11.8). The approach is similar to the one used in the LANAS project described above, but with a more sophisticated model coupler, which allows interaction between the component models

Pierre Cellier

Figure 11.7 Modelled example of landscape scale mitigation of nitrate leaching via the introduction of non-N-fertilized set aside grassland in a drinking water borehole catchment (boundaries in blue line), situated in the Tyrebæk stream watershed, Central Jutland, Denmark. The ‘before’ and ‘after’ maps show results from crop rotation, manure, farm and hydro-geological models, before and after introducing extensive farming systems in the borehole catchment (after Hutchings et al., 2004; Dalgaard, 2009) with permission.

during run-time and minimum adaptation of existing models. NitroScape will consider the majority of the components of N transfer at landscape scale. It will be tested and verified over a range of rural landscapes under different climatic conditions, with different farming systems including livestock. Each landscape has a specific topic and includes natural areas where impacts of N can be predicted.

11.5.3╇ Conclusion on landscape modelling Progress has been made by a number of recent and current studies exploring landscape scale modelling from different starting points and for different purposes, whether to

investigate strategies for the provision of clean drinking water or to protect sensitive semi-natural areas from excess atmospheric N deposition. This required the consideration of both natural and anthropogenic processes and modelling them with sufficient levels of detail in a spatial context. A clear challenge emerges of how to implement the interaction between different component models, using the right tools. These models have also improved the understanding of the relative importance of transfer and transformation processes in rural landscapes. However, there is still much to learn about the interactions of the different elements in the landscape and the development of new models can help with this.

241

Nitrogen flows and fate in rural landscapes

Figure 11.8 Schematic of the NitroScape modelling framework to provide a fully integrated treatment of N exchange fluxes at the landscape scale. The landscape is envisaged as integrating farms, fields, seminatural land and non-agricultural sources, with lateral and vertical dispersion fluxes through the atmosphere and hydrosphere (from Sutton et al., 2007).

11.6╇ The importance of integrating the landscape perspective into N assessment and management 11.6.1╇ N mitigation at the landscape scale The following examples of land use management, often referred to as ‘spatial abatement’ or ‘spatial planning’ (Bleeker and Erisman, 1998; Lekkerkerk, 1998; Theobald et╯al., 2001; Sutton et╯al., 2004; Dragosits et╯al., 2006; Schou et╯al., 2006) highlight the relevance of the landscape scale for mitigating N impacts on the environment. • Establishing tree belts around NH3 sources (e.g. animal housing) or sensitive areas has been suggested as an efficient tool to diminish deposition to sensitive ecosystems and could be used as a tool for their protection (Sutton et╯al., 2004; Dragosits et╯al., 2006). • Constructing wetlands or more generally restoration and management of wetlands (Haycock et╯al., 1997; Woltemade, 2000; Tanner et╯al., 2003; Viaud et╯al., 2004) have proved to significantly decrease the NO3− concentration in surface waters and thus are an efficient buffering element, protecting the river course from the impact of N (Haycock et╯al., 1997; Viaud et╯al., 2004). This has led eco-engineers to the implementation of constructed wetlands for water quality objectives. It is typically a landscape issue because their efficacy and their management depend on the catchment (including hydrological functioning, hedgerow network and grassed strips) that contains the wetlands and on the farming systems (Haycock et╯al., 1997). • On-farm spatial planning provides means to help protect sensitive areas by locating certain activities in the most suitable location. This can include locating farmsteads, crops

242

and grasslands, as well as high emission activities, such as manure spreading to locations that reduce emissions and/or impacts of the emissions. Such strategies can also help protect fresh water by decreasing NO3− leaching and groundwater contamination (see Figure 11.7; Dalgaard, 2009; Deffontaines et╯al., 1994), as well as help protect sensitive ecosystems such as Natura 2000 sites from NH3 deposition. For example, Dragosits et╯al. (2005) modelled the effect of burning poultry manure for power generation (instead of spreading it on fields) or moving poultry houses away from a nature reserve on NH3 and N2O emission, N deposition (Figure 11.9) and NO3− leaching. These measures can exploit spatial relationships to reduce emissions (e.g. arranging activities to reduce N2O emissions) as well as use the source–sink relationship to decrease local impacts of NH3 on sensitive ecosystems (Loubet et╯al., 2009). These approaches can be considered as extending the vision of ‘precision farming’ from the field to the landscape scale. In all cases, practitioners are faced by the complexity of the landscape because it involves not only the studied system (wetlands, tree belt, etc.), but also the surrounding landscape. Modifying crop spatial allocation needs to consider the whole farming system for consistency and its interactions with the landscape. All these measures, therefore, must be placed in a landscape perspective and consider long-term interactions.

11.6.2╇ Using landscape-scale interactions to improve regional models Air pollution or climate models at regional or national scale often use a grid size of between 5 × 5 km2 and 50 × 50 km2, limiting simulations of atmospheric concentration or deposition

Pierre Cellier Figure 11.9 Difference in N deposition (NH3 dry deposition) due to moving of poultry from two sets of buildings in the immediate vicinity of a nature reserve (hatched area) to a more distant location (approx. 1.5 km east/right) (from Dragosits et al., 2005). With permission from Elsevier.

to this resolution. In reality, atmospheric deposition of N, especially NH3 dry deposition, can vary by several orders of magnitude within a grid square of a national or regional model (Dragosits et╯al., 2002). This variability is mainly due to the localized nature of NH3 emission sources and the high dry deposition velocity for NH3 for semi-natural vegetation (sinks). Using data from a regional model could, therefore, significantly underestimate (or overestimate) the environmental impacts since the actual deposition at a particular location could be much higher (or lower) than the model simulation. Landscape-scale atmospheric models can take into account the sub-grid short-scale interactions between sources and sinks and should therefore be used to better assess the uncertainty of national or regional models by estimating the statistical distribution of deposition values within the grid square. This would help to assess local deposition and impacts on conservation areas at a regional scale (see e.g. Loubet et╯al., 2009; Hertel et╯al., 2009). Similarly, in regional scale water quality models, diffuse sources of nutrients from agricultural areas are most often estimated either from empirically determined export coefficients or from an additive approach based on the output of separately run plant/soil/water models at the plot scale. In the best case, they use an arbitrary reduction coefficient accounting for ‘landscape’ or ‘riparian’ retention (see e.g. Billen et╯al., 2009). None of these approaches are able to simulate the effect of changes in the spatial structure or functioning at the landscape scale. Landscape-scale transfer models can help draw a more complete picture by quantifying the storage/release of N pools in soils and groundwater, which are, per se, an important issue for N management, and by describing the intra-annual dynamics of the N delivery to the streams. These models are also better suited to complex scenario analyses, especially to quantify the effects of management practices on N losses. Such results could be aggregated as input to larger scale models, based on the catchment/subcatchment aggregation.

11.6.3╇ Role of the landscape scale in environmental N policy measures A number of policies and measures in the EU and various Member States (see Oenema et╯al., 2011, Chapter 4 this volume) address the importance of landscape structure and functions in relation to N. The potential for considering the landscape scale in these policies depends in part on the level of detail that can be used by Member States to implement them. • Water related policies (Water Framework Directive, Nitrates Directive, Urban Waste Water Directive and Groundwater Directive):€the Water Framework Directive applies a river basin and a catchment approach, while the Nitrates Directive distinguishes Nitrate Vulnerable Zones and various areas at farm level (near water courses, sloping areas, wet soils, etc.). In the case of the Groundwater Directive, groundwater bodies or aquifers are distinguished. Member States have some degree of freedom to interpret the spatial variability within landscapes according to these directives. • Air related policies (Air Quality Directive, NEC and IPPC Directives):€the main environmental targets relate to emission ceilings at national level, concentration levels in the air and the implementation of best available techniques at farm, car, machine and company level. Its spatial component depends on envisaged measures and the target. The landscape scale is to some extent addressed in the case of the protection of sensitive areas (e.g. permit for farm extension close to a Special Area of Conservation). However, there is potential for greater consideration of landscape planning approaches as a means to maximize the environmental benefit for any given national emissions ceiling. • Nature protection policies (Habitats Directive, Birds Directive):€these policies have a strong spatial component through the identification of high nature value areas

243

Nitrogen flows and fate in rural landscapes

(Special Areas of Conservation and Special Protection Areas, making up the Natura 2000 network). It is left to the Member States to identify and to prescribe conditions and measures applicable to these areas, and also around these areas. For example, some Member States have restrictions on farming activities, especially on the intensification of farming activities within and near Natura 2000 areas (see e.g. Hertel et╯al., 2009). Up to now, the assessments generally only consider the location of point sources, e.g. animal housing. Diffuse sources, such as fertilizer and organic manure application, are rarely considered, but have significant local impacts. There is potential for further use of buffer zones in source areas for both atmospheric and water based nitrogen inputs. • Rural Development Regulations and Agri-Environmental Regulations have a strong spatial component. Farmers in less favourable areas and/or near high nature value areas may be supported in exchange for landscape maintenance and forbearance of intensification of farming activities. Farmers may also receive support for introducing low-NH3 emissions techniques for manure storage and application. The landscape perspective also provides the means to link EU agri-environment support more effectively through ‘cross compliance’ with other Directives. For example, where farm management plans associated with support payments are considered as ‘plans or projects’ under the Habitats Directive, landscape analysis provides the means to optimize spatial Nr management. In summary, there are a large number of opportunities provided by EU Directives and Regulations to address the landscape scale. These are needed to better account for local conditions in relation to the wide variety of farming systems and environmental conditions. As yet, there is a huge difference in the interpretation of the EU Directives and Regulations between Member States, and this also is also the case for addressing the landscape scale. This is notably the case with the Nitrates Directive (Smith et╯al., 2007) and protection of the Natura 2000 areas (COST 729, 2009). Our analysis suggests that there are ample possibilities to address the landscape scale, with so far only limited use being made of this scale. Up to now, the Â�policy-maker is faced with a lack of practical tools for supporting this type of analysis, such as user-friendly landscape models. Moreover, there is a need for case studies and improved databases for analysis at this scale.

11.6.4╇ The importance of detailed and simple tools for landscape assessment All the cases described above have highlighted the relevance of the landscape scale for N assessment and management. However, no simple rule exists of how to make an assessment of an environmental measure or abatement technique at the landscape scale. Depending on the level of detail to be applied, this may need to consider a large number of N sources and sinks, with complex and changing relationships between them. Hence it is not straightforward to identify similarities between

244

situations and thus to extrapolate a conclusion for one location directly to another location/situation or to derive simple rules that are generally applicable at the landscape scale. It is clear that comprehensive modelling will be the privileged approach to investigate potential strategies and make an assessment of measures and scenarios at the landscape scale. This requires detailed modelling of processes, as described in Section 11.4. Application of such models to multiple cases and/or regional or larger scales would need detailed landscape databases and the development of landscape typologies. Nevertheless, it is clear that simple practical tools are also needed. While detailed approaches are needed to understand and quantify the interactions, the outcomes of such models also need to be generalized. In this respect, the development of publicly accessible screening tools provides an important step forward. These simpler models can be based on simplifying assumption allowing analytical relationships to be derived or on simpler numerical schemes. This makes it possible to investigate with reasonable accuracy the flows (including input to sensitive ecosystems) and concentration fields of N species. For example instead of using complicated atmospheric transport models Rihm and Kurz (2001) used a function of deposition vs. distance that was developed for the Netherlands (10-year average, averaged over all wind directions) and applied it to Switzerland. It was coupled to a spatially detailed NH3 emission inventories (200 × 200 m2 or less), that formed the input for the calculations of NH3 concentration fields. Although this should not be done in principle as the Swiss climate differs from the Dutch climate, a good correlation was obtained between modelled and measured values for 17 sites. Later Thöni et╯al. (2004) refined the method adjusting the function deposition vs. distance, so that an optimum correlation was obtained for the Swiss situation. Similar examples can be found in other countries (e.g. the SCAIL model in the UK; Theobald et╯al., 2009) and for hydrological modelling (e.g. Durand and Torres, 1996) or ecosystem models (e.g. Strayer et╯al., 2003).

11.7╇ Future challenges The examples above have shown that analysing the N cascade at the landscape scale make it possible to integrate the major processes that modify the N flows and balance. To this extent, the landscape scale also appears to be a very practical scale for implementing and assessing environmental measures. However, it is also highlighted that analysing and modelling landscape interaction for N is a complex task and that no approach has yet been found to be completely satisfactory for the complete analysis. At the same time, there is a parallel need for the development of simple practical tools that can support landscape level decision making in the rural environment. The major questions faced for the coming years include the following. How do we best account for the interactions between farming systems and landscape? Spatial heterogeneity, as well as interactions with farm management, is shown to have strong effects on N flows and transformation at landscape scale. As exemplified in Figure 11.1 farm activity may determine the

Pierre Cellier

spatial arrangement of fields, roads and hedgerows. Moreover farm activity and hence N application to land is not only organized according to the distance from the farm (Figure 11.5) but also to the topography (e.g. grasslands are often located in wetter and less productive areas). These interactions are complex and dependent on local conditions. Hence, there is a need for more study and analysis on the interrelationship between farming systems and landscape features. How can we develop a landscape typology to describe landscape variety in modelling at European scale? European rural landscapes present a wide range of variability, due to climate, physical environment (e.g. topography, soils) and history. Moreover, experiences in landscape modelling have shown that it requires detailed local data, including spatial data on activities/environmental variables, etc. National average data are usually not sufficient to represent local spatial and temporal (diurnal, seasonal or inter-annual) variability. Consequently, there is a need to develop methods to derive a landscape typology giving a limited number of landscape classes based on landscape features and farming systems. These could be based on either real landscape description using aerial photography or remote sensing, or on a farming system approach (see e.g. Figure 11.7) or both. Such a landscape typology would allow landscape processes to be treated more effectively in largerscale operation models. Is it feasible to derive scenarios of future landscapes at 2030 or 2100 horizon? Due to different drivers such as climate change, population increase, extension of urban areas or changes in agricultural and environmental policies, European rural landscapes are expected to change significantly in the next few decades. This could have a significant effect on N flows and efficiency of policy measures. There is a need to examine potential scenarios for future landscape structure and dynamics in order to account for this in climate change and land use change scenarios. How do we develop and test monitoring approaches to assess N flows and budgets at the landscape scale? While modelling at the landscape scale is now becoming firmly established, as illustrated by the studies described above, monitoring approaches for landscape level assessment also need to be developed further, at least to enable the validity of the landscape modelling to be tested. This monitoring should integrate measurement of the spatial and temporal variability of NH3, N2O, NOx and NO3− including the role of hot-spots. Further testing and verification of bioindicators of N responses could be integrated with the physicochemical monitoring activities. How do we best account for landscape issues in environmental N policies? Landscape scale models should be adapted for practical use by landscape planners, farm advisers or policy-Â�makers. This effort will also need databases based on case studies which could be used as a basis for analysis. The use of a landscape typology (see above) would make it possible to integrate and make assessments at a larger scale. There is an ongoing need for simple tools to support the implementation of landscape scale N policies, complementing the detailed models.

How do we assess pollution swapping? In the frame of environmental policies, the risk of pollution swapping (within or beyond the landscape) is increasingly important and must be further explored. The landscape scale is especially relevant, as N transformations often occur in locations different from where N has been applied. Landscape scale modelling can help to understand the origin and magnitude of these transformations by linking together the processes between landscape elements, allowing the synergies and trade-offs to be better quantified.

Acknowledgements We are grateful for support for this work through the NitroEurope Integrated Project of the European Commission, and for coordination funds through the COST 729 and ESF Nitrogen in Europe (NinE) programmes.

References Baudry, J., Bunce, R. G. H. and Burel, F. (2000). Hedgerows:€an international perspective on their origin, function and management. Journal of Environmental Management, 60, 7–22. Beaujouan, V., Durand, P. and Ruiz, L. (2001). Modelling the effect of the spatial distribution of agricultural practices on nitrogen fluxes in rural catchments. Ecological Modelling, 137, 93–105. Beven, K. J. (ed.) (1997). Distributed Modelling in Hydrology:€Applications of the TOPMODEL Concept. Wiley, Chichester, UK. Beven, K. J. and Kirkby, M. J. (1979). A physically based variable contributing areas model of basin hydrology. Hydrological Sciences Bulletin, 24, 43–69. Billen, G. and Garnier, J. (2000). Nitrogen transfers through the Seine drainage network:€a budget based on the application of the Riverstrahler model. Hydrobiologia, 410, 139–150. Billen, G., Thieu, V., Garnier, J. and Silvestre, M. (2009). Modelling the N cascade in regional watersheds:€the case study of the Seine, Somme and Scheldt rivers. Agriculture, Ecosystems and Environment, 133, 234–246. Billen, G., Silvestre, M., Grizzetti, B. et╯al. (2011). Nitrogen flows from European watersheds to coastal marine waters. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J.€W.€Erisman€et╯al. Cambridge University Press. Bleeker, A. and Erisman, J. W. ( 1998). Spatial planning as a tool for decreasing nitrogen loads in nature areas. Environmental Pollution, 102, 649–655. Boehlke, J. K. and Denver, J. M. (1995). Combined use of groundwater dating, chemical, and isotopic analyses to resolve the history and fate of nitrate contamination in two agricultural watersheds, Atlantic Coastal Plain, Maryland. Water Resource Research, 31, 2319–2339. Bouma, J., Varallyay, G. and Batjes, N. H. (1998). Principal land use changes anticipated in Europe. Agriculture Ecosystems and Environment, 67, 103–119. Brisson N., Mary B., Ripoche D. et╯al. (1998). STICS:€a generic model for the simulation of crops and their water and nitrogen balances. I. Theory and parameterization applied to wheat and corn. Agronomie, 18, 311–346. Butterbach-Bahl, K., Gundersen, P., Ambus, P. et╯al. (2011). Nitrogen processes in terrestrial ecosystems. In:€The European Nitrogen

245

Nitrogen flows and fate in rural landscapes Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Cellier, P., Bleeker, A., Bog, E. et al. (2006). Landscape scale assessment of nitrogen interactions. In:€The GHG Cycle in the Northern Hemisphere, Open Science Conference, European Commission, Sissi-Lassithi, Crete (GRC), Nov. 14–18/11/2006, p. 149 (oral communication, abstract). Cirmo, C. P. and McDonnell, J. J. et╯al. (1997). Linking the hydrologic and biogeochemical controls of nitrogen transport in near-stream zones of temperate-forested catchments:€a review. Journal of Hydrology, 199, 88–120. Clarisse, L., Clerbaux, C., Dentener, F., Hurtmans, D. and Coheur, P. F. (2009). Global ammonia distribution derived from infrared satellite observations. Nature Geoscience, 2, 479–483. Colbach, N., Lucas, P. and Meynard, J. M. (1997). Influence of crop management on take-all development and disease cycles on winter wheat. Phytopathology, 87, 26–32. COST729 (2009). Nitrogen deposition and Natura 2000 workshop. See http://cost729.ceh.ac.uk/n2kworkshop Dahm, C. N., Grimm, N. B., Marmonier P., Valett H. M. and Vervier, P. (1998). Nutrient dynamics at the interface between surface waters and groundwaters. Freshwater Biology, 40, 427–451. Dalgaard, T. (2009). Landscape agroecology:€managing interactions between agriculture, nature and socio-economy. In €Proceedings from the Multi-level Processes of Integration and Disintegration, ed. Nagel et╯al. Green Week Scientific Conference, International Congress Centre, Berlin, Jan. 14–15 2009. Margraf Publishers. Dalgaard, T., Heidman, T. and Mogensen, L. (2002). Potential N-losses in three scenarios for conversion to organic farming in a local area of Denmark. European Journal of Agronomy, 16, 207–221. Deffontaines J.-P., Brossier, J., Benoît, M. et╯al. (1994). Agricultural practices and water quality: a research development project. In:€Brossier,€J., de Bonneval, Laurence, Landais, E. (eds.), Coll. Science Update, pp. 31–61. Deffontaines, J. P., Thenail, C. and Baudry, J. (1995). Agricultural systems and landscape patterns€– how can we build the relationship? Landscape and Urban Planning, 31, 3–10. De Vries, W., Kros, J. and Velthof, G. (2005). Integrated evaluation of agricultural management on environmental quality with a decision support system. In 3rd International Nitrogen Conference, Oct.€12–16 2004, Nanjing. China, pp. 859–870. Dinnes, D. L., Karlen D. L., Jaynes D. B. et╯al. (2002). Nitrogen management strategies to reduce nitrate leaching in tile-drained midwestern soils. Agronomy Journal, 94, 153–171. Dragosits, U., Theobald, M. R., Place, C. J. et╯al. (2002). Ammonia emission, deposition and impact assessment at the field scale:€a case study of sub-grid spatial variability. Environmental Pollution, 117, 147–158. Dragosits U., Theobald M. R., Place C. J. et╯al. (2005). Interaction of nitrogen pollutants at the landscape level and abatement strategies. In 3rd International Nitrogen Conference, Nanjing, China, Oct. 12–16 2004, pp. 30–34. Dragosits, U., Theobald, M. R., Place, C. J., ApSimon, H. M. and Sutton, M. A. (2006). The potential for spatial planning at the€landscape level to mitigate the effects of atmospheric ammonia deposition. Environmental Science and Policy, 9, 626–638. Durand, P. and Juan Torres, J. L. (1996). Solute transfer in agricultural catchments:€the interest and limits of mixing models. Journal of Hydrology, 181, 1–22.

246

Durand, P., Breuer, L., Johnes, P. et╯al. (2011). Nitrogen processes in aquatic ecosystems. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Forman, R. T. (1997). Land Mosaics: The Ecology of Landscapes and Regions. Cambridge University Press. Forman, R.T. and Godron, M. (1981). Patches and structural components for a landscape ecology. BioScience, 31, 733–740. Forman, R. T. and Godron, M. (1986). Landscape Ecology. Wiley, New€York. Fowler, D., Pitcairn, C. E. R., Sutton, M. A. et al. (1998). The mass budget of atmospheric ammonia in woodland within 1 km of livestock buildings. Environmental Pollution, 102, 343–348. Galloway J. N., Aber, J. D., Erisman, J. W. et╯al. (2003). The Nitrogen Cascade. BioScience, 53, 341–356. Gras, F. and Benoît, M. (1998). Influence des systèmes de culture et des pratiques agricoles sur la qualité de l’eau minérale de Vittel. Le programme de recherches AGREV. Comptes Rendus de l’Académie d’Agriculture de France (FRA), 5, 166–168. Haag, D. and Kaupenjohann, M. (2001). Landscape fate of nitrate fluxes and emissions in Central Europe:€a critical review of concepts, data, and models for transport and retention. Agriculture, Ecosystems and Environment, 86, 1–21. Haycock, N. E., Burt, T. P., Goulding, K. W. T. and Pinay, G. (1997). Buffer Zones:€Their Processes and Potential in Water Protection. Quest Environmental, Harpenden, UK. Hertel, O., Reis, S. and Ambelas Skjøth, C. (2011). Nitrogen processes in the atmosphere. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al. Cambridge University Press. Hill, J. (1998). Application of computational modelling to ammonia dispersion from agricultural sources. PhD thesis, University of London. Hutchings, N., Dalgaard, T., Rasmussen, B. M. et╯al. (2004). Watershed nitrogen modelling. In:€Controlling Nitrogen Flows and Losses, ed. D. J. Hatch et╯al.,Wageningen Academic Publishers, Wageningen, The Netherlands, pp. 47–53. Jarvis, S., Hutchings, N., Brentrup, F., Olesen, J. and van der Hoek, K. (2011). Nitrogen flows in farming systems across Europe. In:€The European Nitrogen Assessment. ed. M. A. Sutton, C. M. Howard, J.€W. Erisman et╯al. Cambridge University Press. Kristensen, S. P., Thenail, C. and Kristensen, L. (2001). Farmers’ involvement in landscape activities:€An analysis of the relationship between farm location, farm characteristics and landscape changes in two study areas in Jutland, Denmark. Journal of Environmental Management, 61, 301–318. Kros, J., de Vries, W., Voogd, J. C. H., Gies, T. J. A. and Roelsma,€J. (2007). Meervoudige milieumonitoring Noordelijke Friese Wouden: gebiedsstatus van emissie en depositie van ammoniak in relatie tot gebiedsdoelstellingen. Alterra, Wageningen, The Netherlands. Leibowitz, L. G., Loehle, C., Li, B. L. and Preston, E. M. (2000). Modeling landscape functions and effects:€a network approach. Ecological Modelling, 132, 77–94. Lekkerkerk, L. (1998). Implications of Dutch ammonia policy on the livestock sector. Atmospheric Environment, 32, 581–587. Lipiec, J. and Pniewski, W. (1995). Effects of soil compaction and tillage systems on uptake and losses of nutrients. Soil and Tillage Research, 35, 37–52. Liu, J. and Taylor, W. W. (2002). Integrating Landscape Ecology into Natural Resource Management. Cambridge University Press.

Pierre Cellier Loubet, B., Cellier, P., Milford, C. and Sutton, M. A. (2006). A coupled dispersion and exchange model for short-range dry deposition of atmospheric ammonia. Quarterly Journal of the Royal Meteorological Society, 132, 1733–1763. Loubet, B., Asman, W. A., Theobald, M. R. et╯al. (2009). Ammonia deposition near hot spots:€processes, models and monitoring methods. In:€Atmospheric Ammonia: Detecting Emission Changes and Environmental Impacts€– Results of an Expert Workshop under the Convention on Long-range Transboundary Air Pollution, ed. M.€A. Sutton, S. Reis and S. M. H. Baker, Springer, New York. Marshall, E. (2002). Editorial:€introducing field margin ecology in Europe. Agriculture, Ecosystems and Environment, 89, 1–4. McCollin, D. (2000). Editorial:€hedgerow policy and protection€– changing paradigms and the conservation ethic. Journal of Environmental Management, 60, 3–6. Meeus, J. H. A. (1993). The transformation of agricultural landscapes in Western Europe. Science of the Total Environment, 129, 171–190. Molenat, J., Gascuel-Odoux, C., Ruiz, L. and Gruau, G. (2008). Role of water table dynamics on stream nitrate export and concentration. in agricultural headwater catchment (France). Journal of Hydrology, 348, 363–378. Mosier, A., Kroeze, C., Nevison, C. et╯al. (1998). Closing the global N2O budget:€nitrous oxide emissions through the agricultural nitrogen cycle. Nutrient Cycling in Agroecosystems, 52, 225–248. Mourier, B., Walter, C. and Merot, P. (2008). Soil distribution in valleys according to stream order. CATENA, 72, 395–404. Oehler, F., Durand, P., Bordenave, P., Saadi, Z. and SalmonMonviola,€J. (2009). Modelling denitrification at the catchment scale. Science of the Total Environment, 405, 1726–1737. Oenema, O., Bleeker, A., Braathen, N. A. et╯al. (2011). Nitrogen in current European policies. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al. Cambridge University Press. Pitcairn, C. E. R., Fowler, D., Leith, I. D. et╯al. (2003). Bioindicators of enhanced nitrogen deposition. Environmental Pollution, 126, 353–361. Rapport, D. J., Gaudet, C., Karr, J. R. et╯al. (1998). Evaluating landscape health:€integrating societal goals and biophysical process. Journal of Environmental Management, 53, 1–15. Reay, D. S., Smith, K. A. and Edwards, A. C. (2003). Nitrous oxide emission from agricultural drainage waters. Global Change Biology, 9, 195–203. Rihm, B. and Kurz, D. (2001). Deposition and critical loads of nitrogen in Switzerland. Water, Air and Soil Pollution, 130, 1223–1228. Rounsevell, M. D. A., Annetts, J. E., Audsley, E., Mayr, T. and Reginster,€I. (2003). Modelling the spatial distribution of agricultural land use at the regional scale. Agriculture Ecosystems and Environment, 95, 465–479. Scholefield, D., Lockyer, D. R. Whitehead, D. C. V. and Tyson, K. C. (1991). A model to predict transformation and losses of N in UK pastures grazed by beef cattle. Plant and Soil, 131, 165–178. Schou, J. S., Tybirk, K., Løfstrøm, P. and Hertel, O. (2006). Economic and environmental analysis of buffer zones as an instrument to reduce ammonia loads to nature areas. Land Use Policy, 23, 533–541. Skiba, U., Pitcairn, C., Sheppard, L. et╯al. (2004). The influence of atmospheric N deposition on nitrous oxide and nitric oxide fluxes and soil ammonium and nitrate concentrations. Water, Air and Soil Pollution, Focus, 4, 37–43.

Smith, J. U., Bradbury, N. J. and AdT, M. (1996). SUNDIAL:€a PC-based system for simulating nitrogen dynamics in arable land. Agronomy Journal, 88, 38–43. Smith, R. J. N., Glenn, G. A., Parkinson, R. and Richards, J. P. (2007). Evaluating the implementation of the Nitrates Directive in Denmark and England using an actor-orientated approach. European Environment, 17, 124–144. Sovik, A. K., Augustin, J., Heikkinen, K. et╯al. (2006). Emission of the greenhouse gases nitrous oxide and methane from constructed wetlands in Europe. Journal of Environmental Quality, 35, 2360–2373. Spieles, D. J. and Mitsch, W. J. (2000). The effects of season and hydrologic and chemical loading on nitrate retention in constructed wetlands:€a comparison of low- and high-nutrient riverine systems. Ecological Engineering, 14, 77–91. Stockle, C. O., Donatelli, M. and Nelson, R. (2003). CropSyst, a cropping systems simulation model. European Journal of Agronomy, 18, 289–307. Strayer, D., Beighley R. E., Thompson, L. C. et╯al. (2003). Effects of land cover on stream ecosystems:€roles of empirical models and scaling issues. Ecosystems, 6, 407–423. Sutton, M. A., Dragosits, U., Theobald, M. R. et╯al. (2004). The role of trees in landscape planning to reduce the impacts of atmospheric ammonia deposition. In:€Landscape Ecology of Trees and Forests, ed. R. Smithers, IALE (UK)/ Woodland Trust, Grantham, UK, pp.€143–150. Sutton, M. A., Nemitz, E., Erisman, J. W. et╯al. (2007). Challenges in quantifying biosphere-atmosphere exchange of nitrogen species. Environmental Pollution, 150, 125–139. Sutton, M. A., Howard, C. M., Erismar, J. W. et╯al. (2011). The challenge to integrate nitrogen science and policies. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J.€W. Erisman et╯al. Cambridge University Press. Tanner, C. C., Nguyen, M. L. and Sukias, J. P. S. (2003). Using constructed wetlands to treat subsurface drainage from intensively grazed dairy pastures in New Zealand. Water Science and Technology, 48, 207–213. Thenail, C. (2002). Relationships between farm characteristics and the variation of the density of hedgerows at the level of a micro-region of bocage landscape: study case in Brittany, France. Agricultural Systems, 71, 207–230. Thenail, C. and Baudry, J. (2004). Variation of farm spatial land use pattern according to the structure of the hedgerow network (bocage) landscape:€a case study in northeast Brittany. Agriculture Ecosystems and Environment, 101(1), 53–72. Thenail, C. and Baudry, J. (2005). Farm riparian land use and management:€driving factors and tensions between technical and ecological functions. Environmental Management, 36, 640–653. Theobald, M. R., Milford, C., Hargreaves, K. J. et╯al. (2001). Potential for ammonia recapture by farm woodlands:€design and application of a new experimental facility. The ScientificWorld, 1(S2), 791–801. Theobald, M. R., Dragosits, U., Place, C. J. et╯al. (2004). Modelling nitrogen fluxes at the landscape scale. Water, Air and Soil Pollution, Focus, 4, 135–142. Theobald, M. R., Bealey, W. J., Tang, Y. S., Vallejo, A. and Sutton, M. A. (2009). A simple model for screening the local impacts of atmospheric ammonia. Science of The Total Environment, 407, 6024–6033. Thöni, L., Brang, P., Braun, S., Seitler, E. and Rihm, B. (2004). Ammonia monitoring in Switzerland with passive samplers:€patterns, determinants and comparison with modelled concentrations. Environmental Monitoring Assessment, 98, 93–107.

247

Nitrogen flows and fate in rural landscapes Turner, B. L., Meyer, W. B. and Skole, D. L. (1994). Global land-use land-cover change – towards an integrated study. Ambio, 23, 91–95. Van Jaarsveld, J. A. (1995). Modelling the long-term atmospheric behaviour of pollutants on various spatial scales. PhD Thesis, Universiteit Utrecht, Utrecht. Van Pul, W.A.J., van Jaarsveld, J.A., Vellinga, O.S., van den Broek, M. and Smits, M.C.J. (2008). The VELD experiment:€An evaluation of the ammonia emissions and concentrations in an agricultural area. Atmospheric Environment, 42, 8086–8095 Viaud, V., Merot, P. and Baudry, J. (2004). Hydrochemical buffer assessment in agricultural landscapes:€from local to catchment scale. Environmental Management, 34, 559–573. Vogt, J., Puumalainen, J., Kennedy, P. and Folving, S. (2004). Integrating information on river networks, catchments and major forest types:€towards the characterisation and analysis of European landscapes. Landscape Urban Planning, 67, 27–41.

248

Weathers, K., Cadenasso, M. and Pickett, S. (2001). Forest edges as nutrient and pollutant concentrators:€Potential synergisms between fragmentation, forest canopies, and the atmosphere. Conservation Biology, 15, 1506–1514. Westergaard, K. (2006). The landscape composition of organic and conventional, dairy and crop farms in two different geological regions in Denmark. Agriculture Ecosystems and Environment, 117, 63–70. Willems, E., Vandevoort, C., Willekens, A. and Buffaria, B. (2000). Landscape and land cover diversity index. In:€From Land Cover to Landscape Diversity in the European Union, European Commission:€DG AGRI, EUROSTAT and the Joint Research Centre (Ispra). Woltemade, C. J. (2000). Ability of restored wetlands to reduce nitrogen and phosphorus concentrations in agricultural drainage water. Journal of Soil and Water Conservation, 55, 303–312.

Chapter

12

Nitrogen flows and fate in urban landscapes Lead authors: Anastasia Svirejeva-Hopkins and Stefan Reis Contributing authors: Jakob Magid, Gabriela B. Nardoto, Sabine Barles, Alexander F. Bouwman, Ipek Erzi, Marina Kousoulidou, Clare M. Howard and Mark A. Sutton

Executive summary Nature of the problem • Although cities take only 1.5%–2% of the Earth’s land surface, due to their dense population, settlement structure, transportation networks, energy use and altered surface characteristics, they dramatically change the regional and global nitrogen cycle. Cities import and concentrate Nr in the form of food and fuel, and then disperse it as air and water pollution to other ecosystems covering much larger areas.

Approaches • A mass-balance approach was used in order to quantify the fluxes of reactive nitrogen (Nr) in and out of cities. • Cities can be characterised either as a source of Nr (i.e. emitting large amounts as liquid or solid household waste, automobile exhaust, air pollution from power plants) or a sink of Nr (through importing more food, fossil fuels, etc., and having fewer emissions to the air and water). • Paris metropolitan area is used as a case study, which represents an evolving European capital with much available data.

Key findings/state of knowledge • The Paris Metropolitan Area changed from being a sink in the eighteenth and nineteenth centuries to a source of Nr today. Major changes in the city functioning occurred before 1950, but especially recent decades have been characterised by an unprecedented amplification of those changes. • The major part of Nr output is attributed to the combustion of fossil fuels for transport and energy, which converts both atmospheric N2 and fossil Nr to reactive NOx. The second largest Nr contribution comes from incineration of solid waste, and third highest emissions come from sewage water treatment plants. • Urban wastewater discharge into rivers largely contributes to N contamination of the aquatic environment, although sophisticated and expensive tertiary treatment techniques are now available to drastically reduce Nr emissions. • Denitrification in urban landscapes is controlled by the presence of water bodies and green areas. These areas have lower biomass and decomposition rates than natural ecosystems.

Major uncertainties/challenges • A better understanding is needed regarding the following uncertainties:€(a) the mechanisms of dry-deposition in urban systems with patchy vegetation; (b) the complex patterns of air flow in densely built-up areas; (c) the fate of Nr in urban soils with altered water regimes and impermeable surfaces.

Recommendations • To achieve sustainability of urban systems in relation to the N cycle, road transport of goods and passengers has to be reduced, household waste generation minimised, and wastewater treatment improved. • More attention should be given to future sewage processing systems that process Nr (and other nutrients) for reuse as a fertiliser rather than losing the Nr resource by denitrifying it back to N2. • Such measures could eventually turn urban areas from sources of Nr to N-neutral or even N-sink areas. Regional adaptation measures should be specifically tailored to the individual urban ecosystems of Europe.

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

249

Nitrogen flows and fate in urban landscapes

12.1╇ Introduction 12.1.1╇ Problem setting and approach In this chapter we ask and try to answer the following questions:€ what are the specific features of urban landscapes in Europe and how does the local urban N cycle differ from that for natural ecosystems? What are the important issues concerning N management in cities in Europe as a region? As a case study we chose Paris for a more in-depth quantification of terrestrial and atmospheric fluxes, input and output to the system-city. But first of all, why is a chapter dedicated to nitrogen fluxes in urban systems required for the European Nitrogen Assessment (ENA)? Although cities take only 1.5%–2% of the Earth’s land surface, due to their dense population, settlement structure, transportation networks, energy use and altered surface characteristics, they change the N cycle substantially. Firstly, they fix substantial amounts of atmospheric N2 to Nr as NOx through the high temperature combustion of fossil fuels. Secondly, they drive the industrial fixation of Nr to fertilisers, importing the Nr produced in food for burgeoning urban populations, subsequently dispersing it in air and water pollution to other ecosystems over much larger areas than the cities themselves. In other words, they act as concentrators, transformers and dispersers of N, representing new entities of the Earth System. We use an ecosystem approach in this chapter. Any ecosystem is an open system, whose functioning is supported by inand out-fluxes of matter and energy. These fluxes constitute the system itself and determine its boundaries (Odum, 1973).

12.1.2╇ The city as an ecosystem Any city, and especially an industrial one, represents an incomplete or ‘heterotrophic’ system, receiving energy, food, materials, water and other substances from outside of its boundaries. The city could be considered as a specific heterotrophic ecosystem that differs very much from a natural heterotrophic ecosystem. In fact, a city has a more intensive metabolism per area unit, requiring a significant inflow of artificial energy. Its consumption per urban area unit may be three to four orders of magnitude higher than for a same-sized rural non-agricultural area. One hectare of the city area may use 1000 times more energy than the same area of the rural territory. During the process of its own metabolism, a city consumes large amounts of various materials:€food, water, wood, metals, etc., all that we call ‘grey energy’. Products of city’s metabolism are large volumes of substances that are more toxic than those produced by natural ecosystems. Most cities have wide green belts (consisting of trees, bushes, lawns, as well as ponds and lakes), so it could be said that some autotrophic component is also present. Wakernagel and Rees (1997) introduced the concept of the ecological footprint, which provides an account of the total area of productive surfaces required to produce, under prevailing technology, the resources consumed by a country or a city. Feeding modern cities is associated with one-third to half of the global ecological footprint, ranging world-wide from 0.8 to

250

3.8 global-hectares per inhabitant (WWF, 2002). This simple ‘footprint’ indicator, however, neither fully describes the complexity of the relationships that a city establishes with its rural hinterland for the supply of food, nor how, over time, these relationships impact upon the development of both the city and the countryside. Billen et╯al. (2009) introduced the so-called ‘food-print’ indicator that takes into account the area required for producÂ� ing agricultural goods, being expressed in terms of the effective surface of the surrounding territory needed to support the food requirements of a city. Billen et╯al. (2009) found that despite a 50-fold increase in the population of Paris since the fifteenth century, the food-print of the city barely increased twofold. In contrast, the further doubling of the population in the twentieth century was paradoxically accompanied by a fivefold decrease in the food-print, because of the intensification of agricultural production. While emphasising the scale of the changes for Paris, this example also illustrates that the ‘food-print’ is only a partial indicator of the impact of a city. For example, it does not include a comprehensive account of all resources required, such as energy, fertilisers and waste disposal, which might be converted into additive ‘global hectare equivalent’ units. From an ecological point of view, a modern city could be defined as a ‘parasite’ of its rural surroundings. As things are at present, a city does not produce significant quantities of food or other organic substances, it does not purify its air and it does not return water or inorganic substances to the pristine state of their respective biogeochemical cycles. However, theoretically, one could consider a city from a different perspective, for example looking at it as being in a symbiotic relationship with its environment, because a city produces and exports goods and services, money and cultural values, receiving in return goods and services as well.

12.1.3╇ The urbanisation processes Urbanisation is considered to be one of the most powerful and characteristic anthropogenic forces on Earth in the twenty-first century. Although, as noted above, cities occupy only 1.5%–2% of the Earth’s land surface, they are home to over 50% of the world’s population. The number of city dwellers has increased from a mere 14% of world population in 1900 and will further increase to an estimated 60% by 2030 (UNCHS, 2003). The total number of people living in major European cities in the 1400s was only 1.1 million, increasing to 3.5 million in the 1700s, and 3.8 million in the 1800s. However, during the twentieth century, an exponential growth of urban dwellers took place, accompanied by a large increase in the number of large cities (Hohenberg and Hollen, 1985). At present, urbanisation proÂ�cesses in Western Europe have reached the so-called ‘stagnation period’, giving an example of saturated growth. This is, however, accompanied by the expansion of medium-sized cities (~1–2 million). The dynamics of European urbanisation, the differences between urban patterns in Western and Eastern Europe and the implications for the terrestrial, water and atmospheric parts of the N cycle are discussed in greater detail in the following sections.

Anastasia Svirejeva-Hopkins and Stefan Reis

The N cycle, one of the main cycles in ‘Biosphera Machina’, is closely interconnected with the carbon, water and oxygen cycles. Urbanisation-related disturbances in the main driving cycles, caused by urban human activities, lead to global, regional and local environmental problems, such as global warming, photochemical smog, soil acidification, and eutrophication pollution of surface, ground and coastal waters. Even though in some cities urban population might stabilise or even slightly decrease, actual urban areas as such are expected to continue to expand in the twenty-first century, accompanied by growing energy production, increased food demand, and expanding transportation and industrialisation. The demands of high production to feed the urban population alter land cover, biodiversity, and hydrology, both locally and regionally. Similarly, urban waste discharge affects local to global biogeochemical cycles and climate. Although agricultural production is by far the largest influence which has caused the amount of Nr entering the biospheric cycle to double compared to pre-industrial conditions (Smil, 1999), today more than half the crops produced in rural areas are consumed in urban zones. Transportation and industry are concentrated in urban centres, making them point sources of N containing greenhouse and other trace gases such as NOx, N2O and many organic Nr forms (Pataki et╯al., 2006). Air and in particular water pollution influence nutrient cycling and primary production in adjacent ecosystems, especially as many major European cities are located along rivers and coastlines. The Nr in solid wastes generated in cities also eventually enters the air and water, affecting biogeochemical cycles, while the extent of influence depends on the vectors by which materials are carried (e.g. rubbish disposed of by landfill or incineration, dispersal of emitted NOx and NH3, etc.). In addition, it is important to study urban areas in the context of the ENA, since two of the five key threats identified in ENA (Sutton et╯al., 2011, Chapter 5, this volume), namely water quality and air quality, are important aspects of the functioning of urban landscapes. Moreover, pollution generation by cities is a matter of increasing concern, especially when urbanisation outpaces societal capacity to implement pollution-control measures. Therefore, it is important to assess the current situation and to forecast the dynamics of N biogeochemical functioning of urban landscapes. This chapter discusses the past and current situation and touches upon the future trends of the development of Europe’s urban systems that are related to the N cycle.

12.2╇ Urban geography 12.2.1╇ Regional physiography Europe’s long historical development has led to well established trade and communication with the rest of the world. Almost nowhere in Europe is far from water, and water routes have historically facilitated contact between peoples and cultures resulting in the circulation of goods and ideas. The earliest European cities were thus built and developed

adjacent to water routes. The hundreds of miles of navigable waterways, straits and channels between many islands, peninsulas and the mainland, the accessible Mediterranean, North and Baltic Seas, all provided the routes for the exchange of merchandise. Later, the oceans became the means of longdistance spatial interaction. This historical advantage of moderate distances applies to the mainland as well. No place in Europe is very far from any other place on the continent, although nearby places are often different from each other. Short distances and large differences enable much interaction, which is typical of European geography over the past 1000 years. The climate in Europe is mild and temperate. Europe’s biomes (Bazilevich, 1979) are on average high in biomass density; for example, storage of Nr in the deciduous forest that covers large parts of central Europe can be as high as 1100 kg N/ha (for oak-dominated forest including roots). On average, the Nr content in biomass of the deciduous forest biome of Central Europe is 310 kg /ha, while in the mixed forest surrounding Moscow it is estimated at 496 kg N/ha (Bazilevich and Tytlianova, 2007). European soils are generally fertile. For example, there are many alluvial soils, formed in the river valleys and deltas, used to grow crops for centuries, while the Russian and Ukrainian black-soil (‘chernozems’) are extremely fertile. The North European Lowland topographic region, extending from Southern Britain to Western Russia, is the most densely populated physical region of Europe and a route of contact between Europeans and their neighbours to the East. The United Nations (UN) regional classification divides European countries according to their developmental stage as either Highly Industrialised (Western Europe) or Economies in Transition (Eastern Block and former USSR). The dynamics of city development and the process of urbanisation in general, differ somewhat according to region, as discussed in the next sub-section.

12.2.2╇ Urban demography and projections The total population of Europe is around 702 million, distributed over a land area of 9.8 million km2. Following the demographic explosion which started in the middle of the last century, European rates of population growth and of Nr produced by anthropogenic activities broadly matched the global trends, with similar increases during the period 1980 to 1990 (Erisman et╯al., 2008). However, these growth rates have declined from an all time peak (2.1% per year between 1965 and 1970) to about 1.6% per year, while human influence has increased faster than population growth (Cohen, 1995). The UN forecasts that virtually all population growth from now until 2030 will be concentrated in the urban areas of the world. In Europe, the percentage of the population living in urban areas is expected to rise from 77% in 2000 to 84% in 2030, with some countries reaching 90%. However, the proportion of people living in very large urban agglomerations is still small:€in 2000, 48% of the population in developed countries lived in cities of less than 1 million inhabitants and by 2015 that proportion is expected to rise to 49% (UN, 2008).

251

Nitrogen flows and fate in urban landscapes Table 12.1 The largest urban areas in Europe, ranked from 200 world largest (base year 2005)

Rank (World)

Country

Urban Area

Population (106)

Land area (103 km2)

Density (103 person/km2)

16

Russia

Moscow

13.3

4.5

2.9

22

France

Paris

10.4

3.1

3.4

29

UK

London

8.3

1.6

5.1

38

Germany

Essen-Dusseldorf

7.3

2.6

2.8

52

Spain

Madrid

5.0

0.9

5.2

63

Russia

St Petersburg

4.6

1.9

3.9

68

Italy

Milan

4.2

2.4

4.6

79

Greece

Athens

3.7

0.7

5.4

111

Italy

Naples

3.0

0.8

3.9

124

Italy

Rome

2.8

9

3.2

134

Ukraine

Kiyev

2.5

0.5

4.6

143

Portugal

Lisbon

2.3

0.9

2.5

154

Germany

Frankfurt

2.7

0.7

3.4

155

UK

Birmingham

2.3

0.6

3.8

176

Netherlands

Rotterdam-Hague

2.1

0.8

2.5

181

Hungary

Budapest

2.1

0.9

2.4

183

Germany

Cologne-Bonn

2.0

0.9

2.1

186

Poland

Warsaw

2.0

0.5

3.7

Source:€Demographia, 2009.

Not all large urban agglomerations experience fast population growth. In fact, some of the fastest growing cities have small populations and, as population size increases, the growth rate tends to decline. Some European cities are actually shrinking, with Rome having a negative urban population growth rate of −0.04%; for example, Budapest −0.01% and St.€ Petersburg −0.09% (Demographia, 2009). Some East German cities are also experiencing a reduction of population due to the changing labour markets. However, in general for Europe, even cities with shrinking populations tend to be sprawling in area. In other words, even where there is little or no population pressure, a variety of factors are still driving ‘sprawl’ (i.e. increase in city area). These changes are rooted in the desire of people to realise new lifestyles in suburban areas, outside the inner city. The factors reflect micro and macro socio-economic trends, e.g. transport quality, land prices, planning policies, cultural traditions and the attractiveness of respective urban areas. Thus, four-fifths of European citizens now live in towns and cities. As cities start to increase in size, the social infrastructure grows faster than population, i.e. the surface areas of streets, electricity network length, etc., lag behind the city’s population growth, while income and certain measures of innovation outpace it (Bettencourt, et╯al., 2007). According to the same author, individual human needs (housing, employment, household electrical consumption etc.) are scaled linearly for cities of different sizes. This observation provides us with the grounds for using per capita-based calculations later in this chapter.

252

On a global scale, cities with a population of 0.5 million and smaller are anticipated to grow the most rapidly in the course of the next 50 years. In Europe, already-existing cities of between 0.5 and 2 million inhabitants are projected to expand the most rapidly in the course of the next 40 years, which calls for expansion of the cities considered in the present assessment. The model of Svirejeva-Hopkins and Schellnhuber (2008) estimates that in Western Europe, the total urban area (including cities smaller than 1 million inhabitants) will increase from 1645 * 103 km2 in the year 2010 to 1661 * 103 km2 in 2030 and subsequently decrease again slightly below the 2010 value to 1641 * 103 km2, owing to the saturation in the urban population growth rate. The dynamics are different for Eastern Europe, where urban areas are projected to grow from 133 to 134 * 103 km2 by 2030, then decrease in 2040, and then increase again to 134 * 103 km2 by 2050. There is also a substantial difference in the regional relative urban areas that are currently 40% for Western Europe (of the total land) and only around 2% for Eastern Europe. This indicates the different patterns of urban development€– sprawl versus density increase.

City sizes and types With a population of more than 13 million, Moscow is the largest European urban area today (see Table 12.1). Paris is the second largest, followed by London, Essen-Dusseldorf, Madrid and St. Petersburg. There are, in total, 58 urban areas in Europe with populations of one million and more. An urban area (urbanised area agglomeration or urban centre) is defined

Anastasia Svirejeva-Hopkins and Stefan Reis

as a continuously built up landmass of urban development (Demographia, 2009). It generally defines the ‘urban footprint’, or the lighted area that can be observed from an airplane at night. This chapter confines urban areas to a single metropolitan area or labour market area. What constitutes a particular metropolitan area is a matter of professional judgment. However, there is a necessity to ‘draw a line’, especially where adjacent urban areas have ‘grown together’, but remain fairly distinct labour markets. Having considered the list of all European cities, we could categorise city types as: • small cities with a population of under half a million, • medium-sized cities ranging from 0.5 to 2 million, • large cities with a population of 3 million people or more. The so-called combined urban areas (‘mega-cities’) are not characteristic of Europe; this is a typical American, Chinese, and to some extent Indian phenomenon, while there are no mega-cities in Europe to date. While Moscow and London have populations of 13 and 8 million respectively, they neither have catastrophic urban densities nor occupy vast amount of land, as typical mega-cities do. However, some twin cities are already emerging in Europe; for example Essen-Dusseldorf (Germany), Marseille–Aix-en-Provence (France), and Rotterdam-Hague (Netherlands). In this chapter we focus on the settlements of one million and more, since in terms of N fluxes, they play the role of indirect consumers (N for fertiliser, fixed to ‘feed’ them) and sources of Nr at the regional scale. Special consideration, in view of future urban trends, should be given to the type of settlements known as ‘new Â�cities’€ – planned communities, also called ‘new towns’, since they are still relatively small. Carefully planned from the start and typically constructed in a previously undeveloped area, they contrast with settlements that evolve in an ‘old-fashioned’ way. There are only a few of them in Europe so far, their populations are small (a few thousand) and some of them are not developing as planned. One successful example is Louvainla-Neuve, the French-speaking university town in Belgium built in 1972 with originally only 600 permanent residents in it, which has experienced rapid growth, reaching 10 477 inhabitants in 1981. This town is also an example of the ‘New Pedestrianism’ movement, e.g. where roads are in many cases directed under the city. Another representative new city is Tapiola (population of 16 000), constructed in the 1950s and 1960s by the Finnish apartment foundation and designed as a garden city. An interesting example is the city of Slobomir in Eastern Europe, which aims to become one of the major cities of post-war Bosnia–Herzegovina and Serbia, which is still under construction.

12.2.3╇ Urban density as an integral indicator The western part of Europe is characterised by the most uniform demographic processes. France, Europe’s second largest country, has the lowest number of large-city dwellers, at only 10.4%. By contrast, Russia has one of the highest figures€– 42%

of inhabitants reside in large towns and cities. Countries that formed part of the former Soviet Union are similar in that respect. In the Ukraine, 37% of the population live in cities with more than 150 000 residents, in Belarus 40%. The historical development of Germany and Italy led to the formation of a large number of important but smaller cities. In Germany, 26% of the population live in Großstädten (‘large cities’), while only 21% of Italians reside in cities with populations of more than 150 000 people. Poland is the country where the share of the rural population is the highest:€only 24% of people live in large towns and cities. With more than 8 million residents, London alone accounts for almost 12% of the UK population. According to statistics, about 51% of Britons live in towns and cities with more than 150 000 inhabitants. However, this figure could be inaccurate, since some smaller towns have been administratively merged with their surrounding rural districts. Suburban structure varies throughout Europe depending on the individual city’s location, the urban spatial typology, social status and functioning. Today, suburbs are a mosaic of mainly isolated fragments of different housing types, enriched by infrastructure facilities like retail stores and offices subÂ� divided by transport networks. Central Western Europe and especially France and Great Britain have experienced a tough social exclusion of non-privileged classes that have had to settle in compact suburbs consisting of social housing facilities. In Central and Eastern Europe, housing construction was limited to huge, industrialised mass-housing estates. Western-style suburbia never developed, which is one of the core differences between Eastern and Western urban development. Currently, this difference is rapidly disappearing, and luxurious housing estates multiply in the suburban areas of former socialist cities. The urban core is losing its inhabitants, while the suburbs grow as a whole. This is particularly true for Eastern Germany. Figure 12.1 shows the European part of Tobler’s updated world population density map (Tobler, 1995); urban densities are notably redder and large urban areas are detectable. In broad terms, this map of urban density provides an indicator of NOx emissions intensity, as can be seen by comparison with mapped tropospheric NO2 concentrations (Beirle et╯al., 2004; Simpson et╯al., 2011, Chapter 14, this volume). In Figure 12.2, one can clearly see that different types of cities form different clouds in the plot. For example, such major cities as Moscow, Paris and London clearly stand out; however their densities are not especially high compared to some unsustainable Eastern European and Russian cities. Samara and Ekaterinburg€– although occupying relatively small areas and being only medium-sized€ – have the highest population densities in Europe with 8.4 and 9 thousand inhabitants per km2 respectively. Bucharest, the capital city of Romania, is of medium size (2 million) and is also quite dense. If we look at Table 12.2 showing the time taken to travel to work in a sample of cities, we see that Bucharest is characterised by the highest time of almost 80 minutes. Since the area of city is not too large, this may indicate a high level of vehicle congestion, that the public transport networks are not efficient or that the city has developed without one core financial commercial district,

253

Nitrogen flows and fate in urban landscapes

Figure 12.1 European population density and the location of major cities.

but with many small business areas. At the same time, it may reflect that people mainly live in dense suburbs and travel to work in a rather small downtown area. All these factors would directly influence the air quality in this city and consequently pattern of NOx emissions and atmospheric concentrations. Figure 12.3 illustrates another set of urban indicators. This shows, for example, that only 0.01% of wastewater was treated in Bucharest in 1999! Hopefully the situation has changed since then, because with 20% waste incinerated, the rest was deposited in open dumps. Obviously, the emissions of Nr into the water would be very high in this case. Based on Figure 12.2, some medium-sized German and French cities of the lower left corner of the diagram such as Marseille and Lyon appear to be more sustainably managed

254

than other cities shown. Lyon has a medium-sized population, low urban density and 100% of its wastewater is treated, while most of the solid waste is incinerated and only 4% properly land-filled, with some recycling taking place as well. While these indicators do not specify the efficiency of the water treatment and incineration plants, it is notable that Lyon also has a well developed public transport system combining buses, metro, funiculars and tram lines. The picture is of a city with a carefully organised infrastructure having the potential to reduce Nr emissions to air and water. When dealing with urban transportation, not only the average urban density and the geographical expanse of urban areas are important, but also the differences in internal population density, i.e. density gradient. The average urban density data could mask significant variations within urban areas. For

Anastasia Svirejeva-Hopkins and Stefan Reis Table 12.2 Mean travel time to work as an indicator, reflecting internal urban density gradients for selected European cities and areas (year 1999)

Mean travel time to work (in minutes)

example, London and Athens have similar population densities; however the core (central business district) densities in Athens are considerably higher than in London. The Athens suburbs, however, are among the least dense in the world. Similarly, the Essen-Dusseldorf and Milan urban areas have almost identical densities, yet core densities are considerably higher in Milan. This is because with the geographical expanse of nearly all modern, high-income urban areas, automobiles provide by far the greatest coverage, with considerably shorter travel times than public transport. For example, automobiles account for 88% of travel in the Essen-Dusseldorf urban area and somewhat more than 77% in Milan, with its steeper density gradient. These gradients also play a central role when considering wastewater treatment, which is addressed in detail in the following section. Urban density or structure has an impact on air quality and in turn on the health of urban residents. Results of one study (Ferreira et╯al., 2008) indicate that although compact cities provide better air quality compared with dispersed cities, the former have greater exposures and thus a higher health risk, due to high population density. Urban density could clearly serve as some integral indicator that reflects the quality of life, including Nr pollution levels, in cities of different types. As already mentioned, many cities of lower middle ‘cloud’ in Figure 12.2 are expected to expand at a high rate in the next 50 years. Therefore, the direction they will move on a plot has important implications for the anthropogenic urban Nr emissions.

City

Country

Amsterdam

Netherlands

22

Athens

Greece

53

Copenhagen

Denmark

22

Glasgow

United Kingdom

32

Hertfordshire

United Kingdom

27

Koeln

Germany

32

Lyon

France

32

Paris

France

35

Stockholm

Sweden

35

Donetsk

Ukraine

51

Minsk

Belarus

51

Moscow

Russian Federation

62

Nizhny Novgorod

Russian Federation

35

Tbilisi

Georgia

70

Yerevan

Armenia

52

Belgrade

Serbia

35

Bucharest

Romania

78

Budapest

Hungary

40

Prague

Czech Republic

57

Riga

Latvia

27

Comparing European sub-regions

Sofia

Bulgaria

35

Warsaw

Poland

34

Zagreb

Croatia

26

One important difference highlighted by the comparison of Bucharest and Lyon, is that they are situated in Eastern and Western Europe, respectively. The UN sub-divided the world according to the economic developmental stage and, as expected, the difference between Economies in Transition (ET) and Highly Industrialised (HI) countries within Europe is clearly reflected in the urban densities. When comparing total urban areas and populations in cities of the two regions, we can see that ET exceed the values of HI region by two to six times, and that European cities have in general lower density than cities of Russia and Eastern Europe. For cities of half a million or more, Western Europe has an average urban density of 3150 inhabitants per km2; Western Europe outside the UK€ – 3000; UK€ – 4100; Europe except Russia€ – 4200; Russia€– 4900 (Demographia, 2009).

Source:€UNCHS Global Urban Indicators Database (2003).

12.3╇ N-fluxes and city sub-systems, including a case study of the Paris Metropolitan Area

Figure 12.2 Population size versus density for the European cities of 1€million people or more.

Cities show symptoms of the biogeochemical imbalances that they help to create. In urban systems, additional N inputs occur primarily via the importation of foodstuffs for humans, as well as by inadvertent ‘fertilisation’ through the production and subsequent deposition of NOx derived from the combustion of fossil fuels. Cities also experience high acid deposition.

255

Nitrogen flows and fate in urban landscapes

100% other

90% 80%

recycling

70% 60%

open dump

50% incineration

40% 30%

land-fill

20% 10%

Nitrogen transfers in human-dominated ecosystems are inherently inefficient; there is leakage of N at each point of the food chain from fertilisation through human excretion. These leaks could lead to increased storage in soil and groundwater pools and losses to rivers. Air pollutants are transported over both short and long distances (as far as a few thousand kilometres) before being deposited on surface water, vegetation or soil (Bobbink, 1998). In this way, vegetation over a large area or in remote regions can be influenced by airborne pollutants (see Fowler et╯al., 1998; Asman et╯al., 1998). Elemental mass balances can frame this problem, because they identify potential excesses of inputs over outputs and likely sinks within the urban landscape (Baker et╯al., 2001). Usually cities are hotspots of accumulation of N, P, and metals and, consequently, harbour a pool of material resources. By constructing mass balances at scales from the household to the city, human choice can be linked directly to biogeochemical cycling (Kaye et╯al., 2005). The following sections describe the general functioning of city sub-systems and development of the urban N budget for the Paris Metropolitan Area (PAM). They provide a view of the region’s history, current status and projected impacts of N accumulation in adjacent areas, which generally has caused negative impacts. The N budget serves as a planning tool that is based on the estimation of gross N contributions from different N sources/components of the N cycle entering the system, as well as the amount of N leaving it. Such an analysis illustrates the spatial heterogeneity in both Nr creation and distribution of N from a local to a regional scale. The relatively simple N budget for Paris provides an assessment of the relative contributions of sources and the potential benefit of changes of management practices in the PAM. The subsequent sections (Sections 12.3.3 to 12.3.6) discuss urban N fluxes in a more regional context, supporting the statements of the case study for the PAM.

Zagreb (CRO)

Sofia (BG)

Warsaw (PL)

Riga (LAT)

Prague (CZR)

Budapest (HUN)

Belgrade (SRB)

Bucharest (ROM)

Tbilisi (GRG)

Yerevan (ARM)

Nizhny Novgorod (RUS)

Minsk (BLR)

Moscow (RUS)

Donetsk (UKR)

Stockholm (SE)

Lyon (F)

Paris (F)

Koeln (DE)

Glasgow (UK)

Hertfordshire (UK)

Athens (GR)

Copenhagen (DK)

Amsterdam (NL)

0%

256

Figure 12.3 Liquid and solid household waste indicators for selected European cities for the year 1999. A substantial share of wastewater is treated in the majority of the cities listed (exceptions are Zagreb, Belgrade and Bucharest) and the level of land-filling is quite diverse among Western European cities, whereas open dumps are frequently used in Eastern and South-Eastern Europe.

share of wastewater treated

12.3.1╇ A case study and its historic development The Paris Metropolitan Area occupies the Île-de-France, the geographic region constituting the lowland area around the city. This area, which forms the heartland of France, is drained largely by the Seine River and its major tributaries converging on Paris. The natural vegetation of the basin, broad-leafed deciduous forest biome, has been almost entirely lost to civilisation, except for a few relict forests. In order to emphasise what is happening in urban areas concerning N it is relevant to go back as far as the end of the eighteenth century. Paris, an old European capital, is a good illustration, since some major historical changes occurred before 1950, while recent decades have been characterised by an unprecedented amplification of those changes. The population of Paris has dramatically increased since the beginning of the nineteenth century (see Figure 12.4). Paris at the end of the eighteenth century and beginning of the nineteenth century already represented a substantial hub of Nr flows. The concentration of humans and animals (especially horses) in the city is estimated to have required an input of Nr of 24 g per head per day to meet the combined dietary needs. Material from cesspools and other organic matter placed in the streets came from households, animals, butchers, slaughterhouses and other industrial activities. As a result of waste infiltration through the surfaces, the content of Nr in soils and underground water was high, for example the nitrate content in water from Paris wells was up to 2.2 g/L (Boussingault, 1858). Household water supply was nearly non-existent, but the river Seine was more or less preserved from human excreta discharge. This was however not the case for small industrial rivers (like the Bièvre in Paris) where water quality was poor. Much of the N was recovered from the city waste, with urine and ‘night soil’ being collected in carriages and transferred to

Anastasia Svirejeva-Hopkins and Stefan Reis

2500

2000

Firewood (*1,000m3) Charcoal (*1,000 m3) Coal (*1,000 tons)

1500

1000

500

0 1855

Figure 12.4 The human population of Paris, of the Seine catchment, of the urban area of the Seine and of the whole of the Île-de-France region, 1801–1999 (millions of people).

the city refuse depots to obtain the Nr rich urine fraction and a phosphate rich fertiliser powder ‘poudrette’. The latter also contained significant amounts of Nr, though much less than the liquid fraction, especially as part of the Nr was volatilised as ammonia during its preparation (Barles and Lestel, 2007). Wood and, to lesser extent, coal combustion were responsible for emissions to the air€– to these emissions of N2O (from local denitrification) would have been added, although the latter can be considered as having a much smaller scale. Overall, the main urban impact on the N cycle at that time was in the form of underground accumulation, riverine discharge and emissions to the air. The next stage of Paris’s evolution was from the mid nineteenth century to the beginning of the twentieth century. The rates of human and animal concentration in cities kept increasing with more people moving in. Food production therefore became a central issue and so did the greater needs for fertilisers. From the 1820s onwards, cities came to be recognised as sources of fertilisers and the main concern was N recovery. There were many discussions between Boussingault and Liebig about this issue. As Jean Baptiste Dumas said:€‘one of the most beautiful problems in agriculture lies in the art of obtaining nitrogen at low cost’ (Dumas, 1844, as translated by Barles and Lestel, 2007). In addition to the production of poudrette, from the 1830s, ammonium sulphate was manufactured in Paris using urine, and many patents involving the use of human manure and dry fertilisers were developed all around Europe. The processing of Nr rich waste was particularly well developed in Paris (Barles and Lestel, 2007). Thus by the late nineteenth century around 50% of the city’s excreta was collected and industrially processed. The excreta were settled and the eau vanne distilled industrially to produce ammonia (Vincent, 1901). Using the Margueritte process, the yield was estimated at 2.5–3 kg NHx-N per m3 of eau vanne (Vincent, 1901, p 6 ff, p 19). Based on these estimates, this would have amounted to around 800 000 tonnes of excreta processed industrially every year, from which around 2000 to 2400 tonnes of ammonium N were produced, mainly as ammonium sulphate. Combined with the processing of excretal solids to produce poudrette and other fertilisers, N recovery rates increased, but not enough to

1865

1875

1885

1895

1905

1915

1925

1935

1945

Figure 12.5 Firewood, charcoal, and coal consumption, Paris, 1855–1943.

counterbalance the effect of urban population increase in the inner city and suburbs. Water supply to households had been very much improved, but the question still remained of what to do with the water once used. Untreated wastewater provided a major source of pollution to the River Seine, with both high organic matter and ammonium content. Energy consumption (heating systems, gas production, industrial development) continued to increase. Coal progressively replaced firewood and was used either directly or turned into gas, causing the increase in related emissions (see Figure 12.5). By the start of the twentieth century a major change occurred with the rapid growth of the household water supply, the introduction of British-style flushing toilets and the development of the piped sewage system, to which 10% of the population was connected in 1895 and 70% by 1914 (Barles, 2007). Flushing toilets produced much more dilute sewage streams, which were supplied as a liquid fertiliser to surrounding agricultural land. However, industrial processing of dilute sewage was much less cost-effective, and this was therefore a major factor contributing to the obsolescence of the system of recycling the Nr containing wastes (Barles, 2007; Barles and Lestel, 2007). Thus, by 1913, the production of ammonium sulphate from sewage was already down to 600 tonnes of N (Barles, 2007), substantially less than that estimated above for the turn of the century. Between the beginning of the twentieth century and the 1970s, Paris grew as a source of N emissions to the water and air. The human population continued to increase in Paris, while the city sprawled over an even bigger area. However, the animal population decreased substantially. As horses were moved out of the city, total per capita Nr inputs decreased, yet consumption patterns and higher living standards still caused an increase in human Nr inputs, since a larger fraction of food was not eaten and contributed to waste. Table 12.3 shows these changes. New sources of industrial Nr fertilisers were discovered, such as the Haber–Bosch process (Smil, 1999; Erisman et╯al., 2008). During the first half of the twentieth century, the cheaper costs associated with this process removed the immediate need to use sewage Nr as a fertiliser. In time, Nr fertiliser manufacture from the excreta of Paris became completely uneconomic, so that by the 1920s, the industry was effectively at an end;

257

Nitrogen flows and fate in urban landscapes Table 12.3 Main characteristics of dietary nitrogen balance, Paris, 1817, 1869, 1913, 1931 (Barles, 2007)

1817

1869

1913

1931

Human population

716 000

1 840 000

2 893 000

2 885 000

Horses population

16 500

50 000

55 000

10 000

Food inflows (Gg N)

6.0

17.6

23.5

19.7

Urban fertiliser produced Street sludge (Gg N)

0.5

1.3

2.1

0.7

Horse manure (Gg N)

0.6

1.8

1.8

0.4

Human manure (Gg N)

0.1

1.1

1.2

0.1

Wastewater to sewage farms (Gg N)

0

±0

4.0

4.0b

Total outflows to agriculture (Gg N)

1.2

4.2

9.1

5.2

% of food inflows

20

24

40

26

Direct discharge to Seine (Gg N)

?

?

3.1

7.0

% of food inflows

?

?

13

36

a

b

 substantial fraction of the sewage was processed industrially for ammonium sulphate production A (Vincent, 1901). b╇ This concerns only the dietary nitrogen. The total amount of nitrogen in wastewaters is more important. a╇

Table 12.4 Sewage treatment capacity increase for the growing population (Billen et╯al., 2009)

Year

Rate of connection to sewage collection system (%)

Installed domestic wastewater treatment capacity (inhabitant equivalent)

1954

9

300 000

1962

18

500 000

1971

23

2 000 000

1976

31

4 900 000

1980

38

7 200 000

1985

50

9 600 000

1991

60

9 700 000

1996

70

11 300 000

urban Nr, which had been a previously valued product, became a waste product for disposal. Wastewater treatment plants were thus constructed, focusing on removing nitrogen through denitrification, which gradually increased their treatment capacities (see Table 12.4). Surface water contamination continued to increase during this period, since the environmental issue alone (the agricultural pressure disappeared) was not important enough at that time to provoke water treatment enhancement. As heavy industry was moved out of Paris, emissions from industry (industrial N and other pollutants) became distinct from those generated by other urban processes. This, however, did not mean that industrial N emissions decreased, but rather that urban N emissions to the air became impacted by energy transitions. On the one hand, they decreased because of the increased share of electricity in the energy system (which was earlier dominated by gas produced from coal or coal itself). However, urban N

258

emissions decreased substantially only with the construction of nuclear power plants or even since coal power plants were taken outside the city. At the same time, emissions to the air increased as fuel-powered transport replaced horse transport and overall traffic increased due to urban sprawl. From the 1950s the use of Nr fertilisers increased substantially. The mean application rate was 13 kg N per ha/yr on agricultural land in France, which thus increased to 114 kg N per ha/yr in 1996. While ammonium in the River Seine was practically undetectable before the middle of the 1960s, the maximum contamination was reached during the 1970s, owing both to increased urban population (mostly in the downstream part of the sub-basins, as a result of the expansion of the Paris agglomeration) and to increased rates of sewage collection, often released into surface water without treatment. However, later on progress in wastewater treatment led to a considerable decrease in ammonium contamination during the 1990s (Barles, 2007).

12.3.2╇ Nitrogen budget for the Paris Metropolitan Area The urban nitrogen budget (balance) can be considered as a subset of national and regional N budgets. It incorporates imports of Nr-containing products into the city, their conversion within the city boundaries and exports outside of the urban sphere. The urban sphere incorporates the three dimensional space surrounding the urban habitat and spans all environmental media, water, air and soil. We make the first step by creating a detailed Nr mass balance for Paris and its urbanised surroundings in order to estimate the magnitude of major fluxes across the urban landscape and to see how N cycling varies among urban system components. This will help to determine which budget terms are most open to management efforts to reduce N pollution to recipient systems. The budget is shown in Figure 12.6.

Anastasia Svirejeva-Hopkins and Stefan Reis

Figure 12.6 Nitrogen flows quantified for the Paris Metropolitan Area for the year 2006 (PAM, numbers in Gg N per year). The quantified fluxes displayed reflect major N flows through the PAM originating from food import and fossil fuel use, as well as N2 out-flux from wastewater treatment. Notes: The calculation of fossil fuel input has been based on total gasoline consumption for France, weighted by the urban population of the Paris Metropolitan area. Rough assumptions had to be made regarding gasoline N content, fuel N conversion rate in combustion and that the average per capita consumption for France was a suitable indicator for Paris. In reality, it is likely that the urban population will rely more on public transport and this figure may be an overestimation, however, it indicates that the bulk of emissions from fossil fuel combustion stem from N2 conversion of air N content in the combustion process, whereas the fuel N contribution is comparatively small. It has to be stated as well, that this figure only covers gasoline consumption, which is assumed to be the largest contributor of fossil fuel Nr import into the city. â•… The dry and wet deposition of Nr on the soils and surface waters of the PAM were not quantified due to a lack of modelling results for this specific spatial domain. Considering a total emission to air of approx. 50 Gg N per year, dry and wet deposition may lead to a substantial contribution to nutrient input into urban soils and waters. â•… If solid waste has been thermally treated, the remaining Nr content of the waste deposited in landfills should be negligible. Because of the relatively low temperatures at which municipal waste furnaces operate, 70%–80% of NOx formed in municipal waste furnaces is associated with nitrogen in the waste and is emitted to air, alongside small amounts of NO2 and N2O (in particular from emission control equipment for flue-gas treatment of incineration plants). â•… Emissions of nitrogen containing species (mainly NH3) from untreated municipal waste in landfills due to rotting and chemical conversion processes are difficult to quantify, as they depend on the content of the waste stored, moisture and other parameters. In most cases, only CH4 emissions from landfills are monitored and used for power generation. Overall, based on Sutton et╯al. (2000), volatilisation emissions of NH3 in the PAM may be of the order of 1–2 Gg Nr per year.

The estimates in this figure are derived from the mass balances of urban food consumption and nutrient flows (Faerge et╯al., 2001; Magid et╯al., 2006), data on N2O emissions from wastewater treatment plants (Thomsen and Lyck, 2005) with the focus on the urban sub-systems. Obviously, the urban consumption of resources produced elsewhere (notably food) gives rise to substantial leakages of Nr, and should be included in an ecological footprint analysis, as in Rees (1997) and Wackernagel and Rees (1997). Based on the indicative calculation illustrated in Figure 12.6, the Paris Metropolitan Area is a source of Nr, emitting in total the amount of 50 Gg per year to the atmosphere, the major part being attributed to the emissions from transport and energy. Although much smaller, emissions of Nr to air from the incineration of solid waste are also substantial, contributing 2 Gg per year. The amount emitted to the aquatic environment, at about 12€Gg N/yr, greatly depends on the type of wastewater treatment adopted. Disposal of solid wastes and incineration residues in landfills or of sewage sludge on agricultural and

non-agricultural soils (potentially leaking to the ground water over time), together amount to 17 Gg N/yr. Regarding the transformations between N2 and Nr, the largest of these occurs outside this budget, in the production of fertiliser Nr to provide food. Overall, the food Nr import of 63 Gg per year is of a similar order of magnitude to the inadvertent fixation of N2 to Nr through combustion processes. However, the fate of the Nr produced by the two processes is very different. In the case of Nr in food, most is transferred to waste waters, with over half of this being denitrified to N2 in wastewater treatment, i.e. 32 Gg, with only around 0.2 Gg per year being emitted as N2O (Tallec et al., 2007). Although NH3 volatilisation from wastewater treatment is unquantified, based on UK estimates (Sutton et╯al., 2000), it is expected to be similar at around 0.2€Gg per year. As a result of the major loss by denitrification, this leaves only around 12 Gg per year which is returned to agricultural and non-agricultural land, with 12 Gg per year of Nr lost to the environment in receiving waters. The small fraction of the food import Nr being reused on agricultural and

259

Nitrogen flows and fate in urban landscapes

non-agricultural land of 20%, compares with a value of nearly 40% achieved in 1913 (Table 12.3). In the case of Nr from combustion processes, effectively all of this is exported from Paris as Nr. Thus four times as much Nr is released to the environment of Paris from combustion proÂ� cesses than from the Nr originating in food imported to Paris. If this highlights the problem of Nr emissions from combustion sources for this city, it should not be forgotten that the Nr denitrified in wastewater treatment represents the loss of a valuable resource. Without commenting here on the economic viability of recycling wastewater Nr, it may simply be noted that, at an indicative value of €1 per kg fertiliser Nr, the denitrification of wastewater Nr in Paris represents an annual resource loss of €32 million per year.

12.3.3╇ Human food sub-system An important criterion for the assessment of a city’s functioning in relation to the N cycle could be set as the relation between urban population growth and rural productivity (nitrogencontaining food products). In other words, is there enough land surrounding a city to feed its population and how does its rate of growth compare to the growth rate of the urban population? The concept of the so-called food print has been suggested by Billen et╯al. (2009) and discussed in the previous sections. However, this is not a simple relation, since the amount of land could remain the same while its productivity increases due to the introduction of new technologies based on artificial Nr fertilisation, etc. The food could also be imported. This is precisely what caused the food print of Paris to shrink during the second half of the twentieth century, while its population dramatically increased, eventually reaching 10 million inhabitants. Inputs to the human food system include imported and internally produced food, while outputs consist of discharges of unutilised food to landfills and excretion to wastewaters. The size of the area that feeds Paris remained more or less the same and corresponds to the Seine watershed being around 60 000 km2. However, if we look at the ecology of a city, we also need to be sure that water and air pollution do not impair the hinterland surrounding it, thus making the lands less productive. The average intake of protein by the population of the EU is 105 g per person per day, while for France this value is 118 g per person per day, one of the highest in the world (FAO Nutritional Studies). Based on this figure, for France the total direct human consumption of nitrogen is 420 Gg Nr per year, corresponding to about 14% of total EU human consumption of protein. Consequently, for the Paris Metropolitan Area it should be 79.9 Gg N per year. Most food protein is contained in animal products, the common protein sources for Europeans being meat from pigs, cattle and poultry and eggs (FAO Yearbook, 2005/2006). Normal well-fed adults exhibit a nitrogen balance where Nr ingested equals Nr excreted (Voegt and Voegt, 2003), therefore 100% of the outputs from the human food sub-system enter directly into the wastewater subsystem. In other words, human direct consumption should be about the same as human direct excrement to the wastewater subsystem. However, for the mass balance calculations presented

260

earlier in the text, we modified the figure to 63.3 Gg N per year based on Magid et╯al. (2006), who estimated the annual production of nitrogen per person to be 6 kg, of which 0.37 kg is from detergents and the like used in washing water. The rest is related to food intake, either directly or to the food waste going to the bin or kitchen sink.

12.3.4╇ Sewage system:€N in liquid and solid fractions of urban waste The main point sources of nitrogen in discharges are human or industrial sewage treatment plants, larger agricultural units (husbandry) and, of course, untreated wastewater from urban areas (Figure 12.7). In Western Europe, population increased from 466 to 519 million inhabitants (+11%) between 1970 and 2000, which is a slower growth than in North America, for example. Human Nr in sewage increased from 4.8 to 5.7 kg per person per year in the period 1970–2000. Similarly, human P emissions increased from 0.8 to 1.0 kg per person per year, while detergent P emissions decreased from 0.3 to 0.2 kg per person per year, with a peak in the 1980s (Van Drecht et╯al., 2009) . In 1970 about 64% of the population of Europe was connected to sewage systems, increasing to 79% in 2000; in the same period the amount of N removed in wastewater treatment (as denitrification to N2) increased from 10% to 50% of the Nr and P removal from 11% to 59% (Van Drecht et╯al., 2009). The sum of all these changes was a slight decrease of the Nr discharge to surface water (after treatment) from 1326 to 1192 Gg/yr (Figure 12.8); P discharge to surface water decreased from 333 to 216 Gg/yr. Hence, despite the enormous investments in the construction of sewage systems and wastewater treatment facilities, the Nr flow from households to surface waters is still considerable. With higher P removal rates the flows of P are reduced more effectively. For Europe, nitrogen discharge has become an important issue, when in the mid 1970s massive recurrent blooms of gelatinous Phaeocystis flagellata (producing toxins that kill marine animals) colonies and cells were observed each spring in the coastal areas of the North Sea. Practically all investigators came to one conclusion:€the change of dominance from diatoms to Phaeocystis was a consequence of the nitrate enrichment of coastal waters in response to the cumulative nutrient discharges by the major North-West European Rivers, and especially of the decrease of Redfield ratio due to the abundance of phosphorus in the input flow of biogenic pollutants. A detailed description of the precise spatio-temporal interactions between human activities and the functioning of river basin ecosystems and estuaries is presented in Billen et╯al., 2011 (Chapter 13, this volume). The urban sewage treatment process is primarily designed to reduce the level of pollution of watercourses by organic matter, which results in the oxidation of nutrients to inorganic forms. Technologies of nutrient removal are relatively new:€ ‘tertiary treatment’ normally follows the normal twostage treatment of sedimentation, followed in turn by biological treatment. These technologies allow most of the Nr and phosphorus to be removed. But the costs of removal grow very fast according to the degree of cleaning required.

Anastasia Svirejeva-Hopkins and Stefan Reis Figure 12.7 N effluent from sewage systems after wastewater treatment for 2000 for Europe (based on data from Van Drecht et╯al., 2009).

Figure 12.8 Trends in human N emission by type of sanitation, sewage and N removal for 1970, 1990 and 2000 for Europe. Improved sanitation indicates N from households with connection to public sewerage, but also to other systems such as septic systems, simple pit latrines, pour-flush, and ventilated improved pit latrines. It is assumed that the N from households with no improved sanitation or with no sewage connection does not end in surface water. Therefore, the N from households and small industries that enters the surface water is the N from sewage systems with no treatment or after treatment (the red parts of the bars). Figure based on data from Van Drecht et╯al. (2009).

Traditional wastewater treatment can remove 85%–95% Nr and 90%–95% phosphorus. The operational cost is around 1€ per kg nitrogen removed (i.e. denitrified to N2) and 1.5€ per kg phosphorus removed. Construction cost is at a similar level. In principle, up to 100% of N and 90% of P can be removed from wastewaters, but these are very expensive technologies (Henze et╯al., 2008). The nitrogen in wastewater is, in general, in the form of ammonia, also some nitrite is present in low concentrations, but both are toxic for fish. Most technologies use a two-step process:€nitrification (ammonia → nitrate) and denitrification (nitrate → gaseous nitrogen). Usually, combined treatment plants are used, where nitrification and denitrification occur in different zones controlled by oxidation. For instance, the ‘Carousel’ system allows up to 50%–70% removal of the total nitrogen. Certainly, physical and chemical processes exist for ammonia removal, although they are more expensive than the microbiological ones. Chemical stripping by the addition of lime (so that the pH of the sludge rises above 11), and further passage of it through an aeration tower can raise the degree of removal up to 90% (Hammer and McKichan, 1981). In some areas septic systems (installed in allotments and suburban

261

Nitrogen flows and fate in urban landscapes

residences) are suspected of causing an increased level of ground-water contamination. Intermittent loading or recycling of nitrified effluent are suggested as methods of improving denitrification in septic systems. This becomes especially relevant in view of the growing suburbanisation of Europe. These days, European cities usually direct all their wastewaters to treatment plants, however in the 1980s and 1990s the situation was very different, especially for the Eastern European countries (see Figure 12.3):€in Bucharest, for example, only 0.01% of wastewater was treated, in Belgrade 12% and in Warsaw 36%. Since then, the Urban Wastewater Treatment (UWT) Directive was issued in 1991, which regulates wastewater treatment in Europe. It requires the collection and treatment of wastewater in all agglomerations of >2000 population equivalents, p.e., (where 1 p.e. is the organic biodegradable load having a five-day biochemical oxygen demand (BOD5) of 60 g of oxygen per day); secondary treatment of all discharges from agglomerations of >2000 p.e., and more advanced treatment for agglomerations >10 000 p.e. in designated sensitive areas and their catchments. The UWT Directive requires the pre-authorisation of all discharges of urban wastewater, of discharges from the food-processing industry and of industrial discharges into urban wastewater collection systems; it moreover monitors the performance of treatment plants and receiving waters and controls the sewage sludge disposal and reuse, and treated wastewater reuse whenever it is appropriate. Sewage water reuse is one of the adaptation strategies listed in the IPPC 4th Assessment Report for the water sector (IPCC, 2007). Segregated wastewater collection enables efficiency in reuse of water and the nutrients found in wastewater. Consequently, water resources are conserved and nutrients are returned back to the soil. In this system, greywater and blackwater are collected separately from urban households. Rainwater is also harvested before it reaches wastewater collection systems. The UNDP Environment and Energy Program defines Ecological Sanitation (ECOSAN) as ‘an approach to human excreta disposal that aims at recycling nutrients back into the environment and productive systems’ (see further discussion by Oenema et╯al., 2011, Chapter 23, this volume). Ideally, a community using the ECOSAN approach disposes no raw or treated wastewater into the water bodies, limiting the disposal of xenobiotics, including endocrine disrupting chemicals (EDCs), pharmaceuticals and personal care products (PPCPs) along the way. It should be noted that this approach cannot be applied to urban areas with established centralised wastewater collection and treatment systems. However, this is easily adoptable in newly developing urban settlements. Strict legislation is lacking, however, the World Health Organization (WHO, 2006) has issued ‘Guidelines for the Safe use of Wastewater, Excreta and Greywater’. Greywater is rich in terms of phosphorus but the nitrogen content is limited (Atasoy, 2007). Urine contains approximately 80% of the Nr and 55% of the phosphorus found in domestic wastewater (Leeming and Stenstrom, 2002). As was already mentioned, in conventional treatment systems, nitrogen and phosphorus are removed in tertiary treatment. Sludge, containing some of the remaining nutrients, is then disposed of most commonly either by landfill or incineration. With segregated

262

Table 12.5 Nitrogen emitted from wastewater treatment for all European cities of over 1 million, contrasting a scenario of current water treatment (80% treatment, with denitrification based approaches) with a system of latrine water recycling; based on per capita recalculations (Magid et╯al., 2006).

Gg N per year New system of latrine water recycling

Receiving media

N Gg per year 80% treatment

Water (Nr)

157

26.2

Sludge (Nr)

157

52.4

Air (denitrified to N2)

418

0

0

629

732

708

Recycled (as fertiliser Nr) Total

water collection, on the other hand, water is reused and nutrients are returned back to the soil as fertiliser. One interesting historical example of the latrine sewage recycling, implemented in the middle of the nineteenth century in Copenhagen, is described in the Box 12.1. It also describes the situation in London at that time. Box 12.2 describes the beginning stage of centralised urban water management in Russia. It is indeed possible to reduce the amount of Nr entering the surface waters substantially and to entirely eliminate Nr emissions to the atmosphere from wastewater treatment plants (Magid et╯al., 2006). Figure 12.14 shows the scheme suggested for the recycling of sewage waters. It is relevant to estimate the total amounts of Nr in different fluxes for all major European cities (> 1 million population) using the traditional cleaning method and the suggested utilisation. Table 12.5 shows the calculated values, which clearly highlight the advantages of the proposed utilisation scheme. Overall the production of fertiliser in recycling Nr for major European cities would have the theoretical potential to produce over 600 Gg Nr per year, equivalent to around 600 million € per year, at the same time as reducing polluting losses to the environment. Box 12.1╇ Urban waste management in the nineteenth century: London and Copenhagen

London In 1840 Thomas Cubitt wrote ‘… Fifty years ago nearly all London€had every house cleaned into a large cesspool …. Now sewers having been very much improved, scarcely any person thinks of making a cesspool, but it is carried off at once into the river. The Thames is now made a great cesspool instead of each person having one of his own …’ . By then London had reached over 2€million inhabitants, and was the largest city in the world. Cholera outbreaks had begun some years earlier, but the cause for this was not understood. The main reason for public debate was caused by the stink of the Thames. This fired a debate on how to manage waste. At this time Justus von Liebig wrote a letter to the Prime Minister of the UK Sir Robert Peel. ..The cause of the exhaustion of the soil is sought in the customs and habits of the towns people, i.e., in the construction of water closets, which do not admit of a collection and preservation

Anastasia Svirejeva-Hopkins and Stefan Reis

of the liquid and solid excrement. They do not return in Britain to the fields, but are carried by the rivers into the sea. The equilibrium in the fertility of the soil is destroyed by this incessant removal of phosphates and can only be restored by an equivalent supply. …If it was possible to bring back to the fields of Scotland and England all those phosphates which have been carried to the sea in the last 50 years, the crops would increase to double the quantity of former years…’. In his book on Agricultural Chemistry (1862) von Liebig later stated that ‘The introduction of water-closets into most parts of England results in the loss annually of the materials capable of producing food for three and a half million people; the greater part of the enormous quantity of manure imported into England being regularly conveyed to the sea by the rivers …like a vampire it hangs upon the breast of Europe, and even the world; sucking its life-blood. Although von Liebig focused his argument on phosphorus, it is clear that they applied just as much to Nr. When London’s authorities decided to construct a sewage disposal rather than a recycling system suggested by Liebig, he increased his effort to find ways to replace the fertility removed by cities from farmland by artificial means. He focused in particular on developing artificial fertilisers to keep the agricultural land productive in order to feed the cities.

Copenhagen At around the same time as these developments Copenhagen was bankrupt. Similar problems with waste arose, although on a smaller scale and cholera outbreaks eventually€ visited

Figure 12.9

Figure 12.10

Copenhagen in 1853. The future waste management system was hotly debated, but in the end the state prohibited sewers in 1858 due to insurmountable costs, and the city negotiated contracts with farmers for collection of latrine waste. Eventually this system was developed into an elaborate service industry that ensured timely collection and daily transport of latrine contents to the eastern and western outskirts of the city. Figure 12.9 shows show night soil workers empty stainless steel drums into large wooden barrels, and subsequently wash and steam rinse the drums. Furthermore they show farmers collecting night soil from the latrine wagons that were commonly known as‘The Royal Train’and‘The Chocolate Express’(see Figure 12.10). Cholera subsided during these years, and the hinterland farming community gained access to fertiliser as well as a growing market for perishable foods, resulting in better welfare. This system persisted until after the Second World War, but gradually gave way to sewers and water closets. Peri-urban farmers were strong stakeholders, protesting vociferously against the decline of the system and the resulting negative effects on their farmland productivity.

Box 12.2╇ The history of centralized water supply and canalisation in Moscow There were no centralised systems of water supply for cities in Russia before the end of the nineteenth century. The water was taken from the streams and wells. Cities were supplied with water in barrels (fig. 12.11).

Figure 12.11 Water, brought to the city in barrel (Miksashevsky and Korolkova, 2000).

The domestic waste discharges were dumped in the nearby water body or just on the streets. Therefore the water bodies were polluted and were the sources of infectious diseases. The canalisation systems were constructed earlier in Europe:€first in London, then in Paris and Berlin. The positive results were immediate, like in Berlin in the course of one year after the centralised sewage system was built, the water quality was greatly improved and the number of people who contacted cholera dropped by half, and soon the disease was entirely eliminated. With the growth of large cities and rapid increase of their inhabitants, the need for the centralised water supply developed. The late development of centralised water supply in Moscow, contrasted with the fact that smaller-scale water

263

Nitrogen flows and fate in urban landscapes

Box 12.2 (cont.) supply systems existed earlier in Eastern than in Western Europe. For example, the archeological findings suggest their presence on the territory of Caucusus, (Russia) Great Novgorod and Ukraine (see Figure 12.12).

Figure 12.12 An example of small scale water supply: Kremlyn palace, Moscow (17th century).

The centralised large water supply system started to operate in Moscow in 1892, when the two main pumping towers were constructed (see Figure 12.13).

Figure 12.13 “Kretsletz” main water pumping towers.

In 1874, engineer M. A. Popov brought to the attention of the Russian government that sewage channels needed to be built in Moscow and suggested that sewage waters be removed from the city and purified using special irrigation fields (with later usage as fertiliser). Popov used his own funds for collection of topographic and soil data and made sewage application capacity calculations, based on fundamental population growth projections. He developed the entire project of combined sewage system and estimated the construction costs as well as costs of using sewage residue. Unfortunately, the implementation of an actual plan was delayed, due to disagreement with the external evaluator from Berlin, Gobrecht, who supported the plan at first, but then found some flaws and offered to take it over. In€1890, a segregated sewage system project, developed by the engineer Kastilsky, was implemented. By 1898, 262 km of pipelines had been laid and the main pumping station was built. By August 1899, the system began to function to distribute sewage waters to agricultural irrigation fields.

12.3.5╇ Urban N fluxes due to the combustion of fossil fuels in stationary and mobile sources 264

The main contribution to urban air quality problems is made by the combustion of fossil fuels. Emissions come from both

stationary (residential and commercial combustion for heating and process water purposes, combined heat and power plants) and mobile sources (road and off-road transport and machinery). The general mechanisms leading to the formation of the most relevant pollutants (NO2, NOx, NH3, ozone and secondary aerosols/particular matter are illustrated by Hertel et╯al. (2011, Chapter 9, this volume), which gives a detailed account of the processes leading from emissions to ambient concentrations. Here, only specific aspects of urban air quality will be discussed. Kousoulidou et╯al. (2008) analysed the projections of road transport emissions until 2020 and state that while significant reductions are to be expected for relative emissions per vehicle and kilometre driven, NO2 concentrations in urban areas are not expected to fall as dramatically. This is mainly due to the change in the NO2/NOx emission ratio of new technologies, aiming for instance to reduce PM emissions from vehicle exhausts (see also Keuken et╯al., 2010). This trend will most likely have implications for the attainment of ambient air quality standards for NO2 concentrations in all large European cities. Beevers and Carslaw’s (2005) earlier work concluded this for central London. In addition to the technology changes in vehicles and control equipment, an increase in annual average mileage driven in urban areas may arise from a continuing urbanisation towards the development of urban sprawls, as discussed by De Ridder et╯al. (2008). Stationary sources of emissions in urban areas are residential and commercial combustion plants on the one hand (household heating and process water, open fireplaces, etc.) and€– with the deregulation of the energy markets and increasing fuel prices€– decentralised small power plant units (in most cases combined heat and power, CHP, plants based on natural gas or renewable fuels, e.g. biomass) on the other hand. While solid fossil fuels have been banned for use in private stoves in some countries and regions/cities, they contribute a significant share of household heating especially in Scandinavian countries and Central and Eastern Europe. The major contributing sources obviously vary from city to city, however a few patterns can be identified by looking at the information available from individual large urban areas in Europe, such as Greater London (Table 12.6) or Berlin (Table€12.7). Table 12.7 illustrates the situation in Berlin, showing a bulk of NOx emissions stemming from road transport sources. In contrast to Greater London, however, industrial emissions contribute a significantly larger share in Berlin with about 32% of facilities requiring a permit to operate, and thus being subject to regulation. Domestic fuel combustion makes up only about 11% of NOx emissions in Berlin. The above tables illustrate the relative contribution of major activities to urban air quality problems, namely high ambient concentrations of NO2, ozone and particular matter. Figure 12.15 displays the percentage of the urban population in Europe experiencing pollutant concentrations above the respective target/limit values. A clear downward trend can be observed for SO2 together with a less pronounced one for NO2. Concentrations of NO2 in general and population exposure to very high (>40μg/m3) concentrations have declined in the 10 years between 1997 and 2006 (see Figure€12.16).

Anastasia Svirejeva-Hopkins and Stefan Reis Table 12.6 Share of emissions within Greater London and on a national scale for the year 1999. This table illustrates the relevance of road transport sources for urban air quality and indicates a significantly larger proportion of nitrogen (58.2%) being contributed by urban road transport. Industrial sources, in contrast, play a minor role (8.9%) and only about 33% of urban NO x emissions can be attributed to other sources

Nitrogen oxides (NOx) Fine particles (PM10) Sulphur dioxide (SO2) Carbon monoxide (CO)

Total emissions in Gg per year

percentage of emissions in Greater London

percentage of national emissions

All sources

Road transport

Road transport

Industry

Industry

68.13

58.2

8.9

44

37

2.75

67.9

22.3

20

44

3.55

38.3

39.1

1

89

173.38

93.7

1.4

69

16

Source:€Mayor of London’s Air Quality Strategy 2010).

8g to air

Present day distribution of Nitrogen 14 g mostly form urine and faeces delivered via waste water to sewage treatment

3.0 g to the sea

1.7 g in solid waste (organic waste from household and garden)

3g Sludge

N distribution after increased utilization or urine and faeces 12 g utilised as fertiliser for agricultural land 0 g to air

1.5 g delivered via waste water to sewage treatment

0.5 g to the sea

2.2 g in solid waste (organic waste from household and garden) to be recycled upon treatment (composting or bio-gas production)

1g Sludge

Figure 12.14 Current versus increased utilisation method of N distribution (g per capita) (Magid et╯al., 2006).

At the same time as Figures 12.15 and 12.16 show the decline in exposure to high NO2 concentrations in urban areas, a more frequent occurrence of exposures to high ambient levels of ground level ozone above 120 μh/m3 (8 h mean) are observed (Figure 12.17). It is difficult to assess to what extent this increase of exposure of the urban population to high ambient levels of ground level ozone is caused by urban emissions

(resp. the reduction of urban NOx emissions and the resulting decrease of the titration effect in NOx-rich environments) and to what extent by the slowly increasing concentrations of global background ozone levels. In particular, exposure to high ambient concentrations of ozone and PM lead to adverse effects on human health, which are discussed in detail in Moldanová et╯al., 2011 (Chapter 18, this volume).

265

Nitrogen flows and fate in urban landscapes Table 12.7 Emissions of NOx in Berlin according to emitting groups (Gg per year)

Data in Gg per year Nitrogen oxide

1989

1994

2000

2002

Trend 2005

Germany

2862

2226

1815

1640

1447

Berlin

70.0 (2.4%)

42.4

26.1

22.1

19.8

17.5

Emittent approved facilities

41.8

16.2

6.0

5.8

8.3 (31.9%)

6.5

Trend 2010

Domestic fuel

2.7

3.1

2.9 (10.9%)

2.9

2.7

2.6

Small trade

1.2

0.7

0.2 (0.7%)

0.2

0.2

0.1

21.4

19.0

12.4 (47.5%)

10.5

8.9

7.0

Traffic (other)

1.4

1.3

1.1 (4.3%)

1.1

1.1

1.1

Other sources

1.5

2.1

1.2 (4.6%)

1.0

1.0

0.9

Traffic (motor vehicles only)

(Source:€Senate Berlin, 2010).

100

PM10

NO2

O3

Figure 12.15 Percentage of urban population resident in areas where different air pollutant concentrations are higher than selected limit/target values, EEA member countries, 1997–2006. (Source:€EEA, 2010.)

SO2

80 60 40 20 0 1997

1998

1999

0-26 µg/m3

100

2000

2001

2002

26-32 µg/m3

2003

2004

32-40 µg/m3

2005

2006

2007

Figure 12.16 Percentage of population resident in urban areas potentially exposed to NO2 concentration levels exceeding the annual limit value in EEA member countries for the period 1997–2006. (Source:€EEA, 2010.)

>40 µg/m3

75

50

25

0 1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

% of urban population

12.3.6╇ Urban green areas and urban soils Urban open space and green areas In European cities the area occupied by open space, consisting of parks and recreation areas, is on average 30%. Organic matter production does not play a significant role in the operating mechanisms within a city. However, while the green belts play

266

purely recreational and aesthetic roles, they are very important also because they even out the air temperature fluctuations within a city, reduce noise and other pollution, and serve as a habitat for small animals and birds. It is not cost-free, however, to support their functioning, and the labour and fuel spent on irrigation, lawn management, tree planting and care, etc., increases the energy and monetary expenses of a city.

Anastasia Svirejeva-Hopkins and Stefan Reis

12.4╇ Conclusions, uncertainties and the future development of European cities

Figure 12.17 Exposure of urban population in EEA member countries to maximum ozone concentration above the 8 h-daily mean target value of 120€µg/m3. (Source:€EEA, 2010.)

Trees can nevertheless play an important role in Nr related issues in cities, such as by helping to reduce the effects of particulate matter (PM) and NOx pollution. The amounts of gaseous pollutants and particulates and the interception of aerosols are greater in woodlands than in shorter vegetation (Fowler et╯al., 1998), since they have broader leafs and create turbulent mixing of air. Therefore, urban woodlands and the presence of trees in the urban environment can improve air quality quite significantly. McPherson (1998) estimated that in Chicago trees removed 234 tonnes of PM10 in 1991 and improved average hourly air quality by 0.4% (2.1% in the heavily wooded areas), while Nowak et╯al. (1997) calculated that trees in Philadelphia improved air quality by 72% by removal of PM10.

Urban soils Urban soils, which play an important role as sources, sinks and transformers in the nitrogen cycle, represent an area of great concern as regards food supply and supply of sustainable drinking water, and are important in terms of aesthetics and recreation. Soils are designated as urban soils if they are located in watersheds that provide drinking water, food, waste utilisation, and natural resources to cities. Urban soils include also all soils located within cities in park areas, recreation areas, community gardens, green belts, lawns, septic absorption fields, sediment basins or other open or sealed soils inside the city. The N status and dynamic of urban soils is determined by external factors like N deposition, temperature, rainfall, groundwater N content and groundwater level and internal factors like the geogenic parent material, technogenic substrates, water and air holding capacity, dry bulk density, microbial activity, etc. Technogenic substrates (rubble, construction material, sewage sludge, refuse, and dust) play a key role in the genesis of urban soils. Furthermore they are relatively comparable among cities. Examining the N status of the most important technogenic substrates makes it possible to assess the potential behaviours of urban soils.

The present day situation in our case study, the PAM, reflects to a large extent the metropolisation processes in the area. The urban population is continuing to increase due to urban sprawl, while the density does not increase greatly, which is typical for the Western European region. The further development of the nitrification–denitrification process in wastewater treatment plants has reduced surface water pollution from cities over the last century. However, such wastewater treatment plants operate by denitrifying Nr back to N2 which can be considered as a substantial waste of an expensive resource. For Paris alone, this loss equates to a potential fertiliser value of around € 30 million per year. Urban sprawl is responsible for an increase of car use in the urban context (public transport is still not adapted to low density areas) and there is an increase of trip length. The globalisation of trade leads to an increase in transport-related N emissions. However, both the trend towards moving industries from Europe to other parts of the world and the strict regulation of industrial pollutants lead to a decrease in industrial N emissions. The renaissance of the city is a hot topic in Europe. Generally the term addresses the renaissance of the inner city and is applied to the city centre only. However, while suburbanisation is increasing, more and more European cities are expected to turn into urban regions. A partial renaissance of the inner city would probably take place, as well as partial growth of suburbia. Both will be accompanied by either partial decay of suburbia or partial decay of the inner city. There are already vast and increasing differences among cities. Since the breakdown of communism, the development is very different in different regions, but most European cities are currently exposed to drastic economic and social changes. They face tremendous new challenges such as globalisation, ageing societies, shrinking population figures, shrinking household sizes, increasing social divisions, decreasing resources of public authorities, etc. This change from a relatively stable industrial society towards a post-industrial society will shape the development of cities in Europe over the coming decades. The rejuvenation of an attractive city centre can offer the best service locations, plus it can tie a highly mobile urban middle class to a city in the long run. Creation of an efficient public transportation network connecting the suburbs with the city core is an essential aspect. The growth of European suburbia is a dynamic process, yet in most European cities, urban planning efforts concentrate on the city centre, such as in London and Berlin. If we aim to create a ‘neutral’ Nr state for cities in Europe, we have to increase recycling of food and water, minimise household waste either through reusing sewage waters or technologically improving treatment plants, and reduce Nr emissions to the atmosphere limiting travel by car as much as possible. The importance of these sources is clearly illustrated by the nitrogen budget of Paris (Figure 12.6). In particular, reducing road traffic NOx emissions has the largest single potential to decrease Nr emissions from a city such as Paris, while the use of new re-use based sewage systems, have the potential to avoid

267

Nitrogen flows and fate in urban landscapes

the waste of Nr inherent in denitrification-based water treatment systems. Such measures could eventually turn urban areas from being a source of nitrogen to becoming nitrogen neutral. These adaptation measures have to be carefully planned and individually tailored. The following uncertainties regarding nitrogen cycling in an urban system need a better understanding:€the mechanisms of dry-deposition processes in urban systems with patchy vegetation; high NOx emissions and the complex patterns of air flow in densely built-up areas. The N dynamics of urban soil are very uncertain, and while soil represents a major sink of N in natural ecosystems, what happens in urban soils due to, for example, impervious surfaces (roads, etc.) has been little studied. Factors that control denitrification in urban landscapes are related to the presence of green areas within city, but those areas differ from natural landscapes. They have lower densities of biomass and altered decomposition rates. Interactions between increasing temperatures, especially in built-up areas, and photochemical smog (NO2 and ground level ozone) are complex and difficult to quantify. Yet, it can be expected that increasing global ambient temperatures may contribute to more frequent occurrences of Nr-related adverse health effects in cities. There is still some uncertainty regarding the fate of Nr in the septic tanks in low-density suburban residential areas. For example, many of these are fairly old and may not function properly, causing leaking to the groundwater. Also storm events often cause septic tanks to overflow, in which case the untreated sewage is transported directly to the surface waters. The most immediate task to bring a city to a neutral state in relation to the nitrogen cycle, is to control transport emissions in cities. There are already some examples of sustainable transport policies in cities, showing that public transport can be attractively organised for a densely built-up city, as well as for a large metropolitan area. In the city of Basel, the traffic policy aims to calm traffic and to promote the use of the bicycle. In the 1980s and the early 1990s the Basel traffic policy implemented a variety of environmentally compatible measures in different areas of transport. This multi-level policy could serve as a model for urban development in other cities. Local measures include implementation of a traffic policy with an effective combination of green transport modes; the successful testing of traffic calming measures in residential districts; the safeguarding of high standards for bicycle use; the diversification of modernised bus, tram, and rail systems; the introduction of a customer-friendly pricing policy in public transport systems; the passing of legal regulations in favour of green modes of transport. Restructuring the labour market (which is the second most important driver of urbanisation after population growth) plays an important role in creating of sustainable transportation network. The city of Copenhagen, which in 1993 introduced a Municipal Plan aiming to design a compact urban structure based on public transport, provides a good example. This required a long-term restructuring of working places according to public transport stations, enhancing and transforming the growth of the city in the harbour area,

268

strengthening the ‘green’ aspects of the city and restoring and maintaining the historical quality of specific city districts and their diversity. An increasing interest in deploying the tram-trains concept is growing across Europe in order to fight congestion whilst also cutting carbon and nitrogen emissions. This approach, using proven technologies, combines heavy rail routes with tramways to allow passengers to access key destinations in city centres from the suburbs without making a change, with the aim of attracting people who previously used cars. Germany pioneered the utilisation of combining heavy rail and street running fixed link systems but in the last few years there has also been an upsurge of interest in France and a trial is under way in the UK (connecting Sheffield, Huddersfield and Rotherham in Yorkshire). Generally speaking, the most effective N management strategies are those that are specifically tailored to individual cities and the ecosystems surrounding them. To develop such schemes will require the construction of detailed, ecosystemlevel Nr balances, to help with a deeper understanding of the interplay of inputs, geographical and climatic factors, nonspecific management practices, and deliberate Nr management practices that control the fate of Nr in urban landscapes. Nitrogen budgets can be used as a tool to provide a context for the evaluation of the extent to which human intervention in the N cycle has changed Nr distribution from local to global scales. To gain first insight into the spatial heterogeneity of Nr creation and distribution in urban landscapes, we examined an urban N budget. This is important as it illustrates the differences in Nr creation and distribution as a function of the level of urban development and geographic location.

Acknowledgements The authors gratefully acknowledge support from the Nitrogen in Europe (NinE) programme of the European Science Foundation, from the NitroEurope IP funded by the European Commission and from the COST Action 729.

References Asman, W. A. H., Sutton, M. A. and Schjørring, J. K. (1998). Ammonia:€emission, atmospheric transport and deposition. New€Phytologist, 139, 27–48. Atasoy, E., Murat, S., Baban A. and Tiris, M. (2007). Membrane bioreactor treatment of segregated household wastewater for reuse. Clean, 35, 465–472. Baker, L., Hoper, D., Xu, Y., Edmonds, J. and Lauver, L. (2001). Nitrogen balance for the Central Arizona–Phoenix (CAP) ecosystem. Ecosystems, 4, 582–602. Barles, S. (2007). Feeding the city:€food consumption and flow of nitrogen, Paris, 1801–1914. Science of the Total Environment, 375, 48–58. Barles, S. and Lestel, L. (2007). The nitrogen question:€urbanization, industrialization, and river quality in Paris 1830–1939. Journal of Urban History, 33, 794–812. Bazilevich, N. I. (1979). Biogeochemisty of the Earth and functional models of exchange processes in natural ecosystems. In:€Modern Concepts and Problem of Biogeochemistry

Anastasia Svirejeva-Hopkins and Stefan Reis Proceedings of the Biogeochemical Laboratory, vol. 17, Nauka, Moscow, pp. 56–73. Bazilevich, N. I. and Tytlanova, A. A. (2007). Biological Cycling on Five Continents:€Nitrogen and Elements in Natural Ecosystems. Novosibinsk, Nauka (in Russian). Beirle, S., Platt, U. and Wagner, T. (2004). Global Mean Tropospheric Nitrogen Dioxide (NO2) Vertical Column Density (VCD) for Europe: Map. Institute for Environmental Physics, University of Heidelberg. Beevers, S. D. and Carslaw, D. C. (2005). The impact of congestion charging on vehicle emissions in London. Atmospheric Environment, 39, 1–5. Bettencourt, L. M. A., Lobo, J., Helbing, Y., Kuchnert, C. and West,€G.€B. (2007). “Growth, innovations scale and the pace of life in cities”, Vol. 104, No. 17 pp. 7301–7306. Billen, G., Barles, S., Garnier, J. et╯al. (2009). The food-print of Paris:€long-term reconstruction of the nitrogen flows imported into the city from its rural hinterland. Regional Environmental Change, 9, 1436–1498. Billen, G., Silvestre, M., Grizzetti, B. et╯al. (2011). Nitrogen flows from European watersheds to coastal marine waters. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al. Cambridge University Press. Bobbink, R. (1998). Impacts of tropospheric ozone and airborne nitrogenous pollutants on natural and seminatural ecosystems:€a commentary. New Phytologist, 139, 161–168. Boussingault, J. B. (1858). Note relative à l’emploi industriel et domestique des eaux de puits de Paris. Gazette hebdomadaire de médecine et de chirurgie, 5, 507–508. Cohen, J. E. (1995). Population growth and Earth’s human carrying capacity. Science, 269, 341–346. Demographia (2009). World Urban Areas & Population Projections:€5th Comprehensive Edition (Revised April 2009). http://www.demographia.com (site accessed 30/07/2010). Dumas, J. B. (1844). Essai de statique chimique des êtres organisés, 3rd edn. Paris. De Ridder, K., Lefebre, F., Adriaensen, S. et╯al. (2008). Simulating the impact of urban sprawl on air quality and population exposure in the German Ruhr area. II. Development and evaluation of an urban growth scenario. Atmospheric Environment, 42, 7070–7077. EEA (2010). http://www.eea.europa.eu/data-and-maps/figures (site accessed 30/07/2010). Erisman, J.-W., Sutton, M. A., Galloway, J. and Klimont, Z. (2008). How a century of ammonia synthesis changed the world. Nature Geoscience, 1, 636–639. Faerge, J., Magid, J. and de Vries, F. W. T. P. (2001). Urban nutrient balance for Bangkok. Ecological Modelling, 139, 63–74. FAO Nutritional Studies no.24. http://www.fao.org/ (site accessed 30/07/2010). FAO Yearbook (2005/2006). http://www.fao.org/ (site accessed 30/07/2010). Ferreira, J., Martins, H., Miranda, A. I. and Borrego, C. (2008). Population Exposure to Atmospheric Pollutants:€The Influence of Urban Structure. Wessex Institute of Technology Publications. Fowler, D., Flechard, C., Skiba, U., Coyle, M. and Cape, N. (1998). The atmospheric budget of oxidized nitrogen and its role in ozone formation and deposition. New Phytologist, 139, 11–23. Hammer, M. J., McKichan, K. A. (1981). Hydrology and Quality of Water Resources. John Wiley, New York.

Henze, M., van Loosdrecht, M. C. M., Ekama, G. A. and Brdjanovic,€D. (eds.) (2008). Biological Wastewater Treatment: Principles, Modelling and Design. Hertel, O., Reis, S. and Ambelas Skjøth, C. et╯al. (2011). Nitrogen processes in the atmosphere. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman, et╯al. Cambridge University Press. Hohenberg, P. M. and Hollen, L. (1985). The Making of Urban Europe, 1000–1950. Harvard University Press, Cambridge, MA. IPCC (2007). Climate Change 2007:€Synthesis Report. Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. IPCC, Geneva, Switzerland. Kaye, J. P., McCulley, R. and Burke, I. C. et╯al. (2005). Carbon fluxes, nitrogen cycling and soil micro-organisms in adjacent urban, native and agricultural ecosystems. Global Change Biology, 11, 575–587. Keuken, M. P., Jonkers, S., Wilmink, I. R. and Wesseling, J. (2010). Reduced NO2 and PM10 emissions on urban motorways in The Netherlands by 80 km/h speed management. Science of the Total Environment, 408, 2517–2526. Kousoulidou, J., Ntziachrios, L., Mellios, G. and Samaras, Z. (2008). Road transport emissions projections in European urban environments. Atmospheric Environment, 42, 7465–7475. Leeming, C. R. and Stenstrom, T. A. (2002). Faecal contamination of source-separated human urine based on the content of faecal sterols. Water Research, 36, 1965–1972. Magid, J., Eilersen, A. M., Wrisberg, S. and Henze, M. (2006). Possibilities and barriers for recirculation of nutrients and organic matter from urban to rural areas:€a technical theoretical framework applied to the medium-sized town Hillerod, Denmark. Ecological Engineering, 28, 44–54. Mayor of London (2010). http://www.london.gov.uk/mayor/ strategies/air_quality/docs/chap_2.pdf McPherson, E. G. (1998). Structure and sustainability of Sacramento’s urban forest. Journal of Arboriculture, 24, 174–190. Miksashevsky, N. V. and Korolkova, S. V. (2000). Clean Water Treatment Systems and Fitters. BHV–Petersburg, (in Russ.) Moldanová, J., Grennfelt, P., Jonsson, Å . et╯al. (2011). Nitrogen as a threat to European air quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al. Cambridge University Press. Nowak, D. J. et╯al. (1997). Environmental and economic benefits€of preserving forests within urban areas:€air and water€quality. In:€de Brun, Constance T. F. (ed.), The Economic Benefits of Lland Conservation. The Trust for Public Land, pp.€28–47. Odum, H. T. (1973). Energy, ecology and economics. Ambio, 2, 220–227. Oenema, O., Salomez, J., Branquinho, C. et╯al. (2011). Integrated approaches to nitrogen management. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al. Cambridge University Press. Pataki, D. E., Alig, R. J., Fung, A. S. et╯al. (2006). Urban ecosystems and the North American carbon cycle. Global Change Biology, 12, 2092–2102. Rees, W. E. (1992). Ecological footprints and appropriated carrying capacity:€what urban economics leaves out. Environment and Urbanisation, 4 (2), 121–130. Senate Berlin (2010). http://www.berlin.de/sen/umwelt/luftqualitaet/ de/emissionen/index.shtml Simpson, D., Aas, W., Bartnicki, J. et╯al. (2011). Atmospheric transport and deposition of nitrogen in Europe. In:€The European Nitrogen

269

Nitrogen flows and fate in urban landscapes Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al. Cambridge University Press. Smil, V. (1999). How many billions to go? Nature, 401, 429. Sutton, M. A., Dragosits, U., Tang, Y. S. and Fowler, D. (2000). Ammonia emissions from non-agricultural sources in the UK. Atmospheric Environment, 34, 855–869. Sutton, M. A., Howard, C. M., Erisman J. W. et╯al. (2011). The need to integrate nitrogen science and policies. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al. Cambridge University Press. Svirejeva-Hopkins, A. and Schellnhuber, H. J. (2008). Urban expansion and its contribution to the regional carbon emissions:€using the model based on the population density distribution. Ecological Modelling, 216. Tallec, G., Garnier, G., Billen, G. and Gousailles, M. (2007). Nitrous oxide emissions from denitrifying activated sludge of urban wastewater treatment plants, under anoxia and low oxygenation. Bioresource Technology, 99 2200–2209. Thomsen, M and Lyck, E. (2005). Emission of CH4 and N2O from Wastewater Treatment Plants (6B). No. 208, 1–46. 2005, Research notes from National Environmental Research Institute, Roskilde, Denmark.

270

Tobler, W., Deichmann, V., Gottsegen, J. and Maloy, K. (1995). The Global Demography Project, Technical Report TR-95–6. National Center for Geographic Information Analysis, University of Santa Barbara, CA. UNCHS (2003). Global Urban Indicators Database. http://ww2. unhabitat.org/programmes/guo/urban_indicators.asp 2003 Van Drecht, G., Bouwman, A. F., Harrison, J. and Knoop, J. M. (2009). Global nitrogen and phosphate in urban waste water for the period 1970–2050. Global Biogeochemical Cycles, 23, GB0A03, doi:10.1029/2009GB003458. Vincent, C. (1901). Ammonia and its Compounds: Their Manufacture and Uses, translated by M. J. Salter. Scott, Greenwood & Co., London. Voegt, D. and Voegt, J. G. (2003). Biochemistry, 2nd edn. John Wiley and Sons, New York. Wackernagel, M. and Rees, W. E. (1997). Perceptual and structural barriers to investing in natural capital:€Economics from an ecological footprint perspective. Ecological Economics, 20, 3–24. WHO (2006). Guidelines for the Safe Use of Wastewater, Excreta and Greywater, Vol. 1, Policy and Regulatory Aspects. World Health Organization, Geneva. WWF (2002). Living Planet Report. http://www.panda.org/ about_our_earth/.

Chapter

13

Nitrogen flows from European regional watersheds to coastal marine waters Lead author: Gilles Billen Contributing authors: Marie Silvestre, Bruna Grizzetti, Adrian Leip, Josette Garnier, Maren Voss, Robert Howarth, Fayçal Bouraoui, Ahti Lepistö, Pirkko Kortelainen, Penny Johnes, Chris Curtis, Christoph Humborg, Erik Smedberg, Øyvind Kaste, Raja Ganeshram, Arthur Beusen and Christiane Lancelot

Executive summary Nature of the problem • Most regional watersheds in Europe constitute managed human territories importing large amounts of new reactive nitrogen. • As a consequence, groundwater, surface freshwater and coastal seawater are undergoing severe nitrogen contamination and/or eutrophication problems.

Approaches • A comprehensive evaluation of net anthropogenic inputs of reactive nitrogen (NANI) through atmospheric deposition, crop N fixation, fertiliser use and import of food and feed has been carried out for all European watersheds. A database on N, P and Si fluxes delivered at the basin outlets has been assembled. • A number of modelling approaches based on either statistical regression analysis or mechanistic description of the processes involved in nitrogen transfer and transformations have been developed for relating N inputs to watersheds to outputs into coastal marine ecosystems.

Key findings/state of knowledge • Throughout Europe, NANI represents 3700 kgN/km²/yr (range, 0–8400 depending on the watershed), i.e. five times the background rate of natural N2 fixation. • A mean of approximately 78% of NANI does not reach the basin outlet, but instead is stored (in soils, sediments or ground water) or eliminated to the atmosphere as reactive N forms or as N2. • N delivery to the European marine coastal zone totals 810 kgN/km²/yr (range, 200–4000 depending on the watershed), about four times the natural background. In areas of limited availability of silica, these inputs cause harmful algal blooms.

Major uncertainties/challenges • The exact dimension of anthropogenic N inputs to watersheds is still imperfectly known and requires pursuing monitoring programmes and data integration at the international level. • The exact nature of ‘retention’ processes, which potentially represent a major management lever for reducing N contamination of water resources, is still poorly understood. • Coastal marine eutrophication depends to a large degree on local morphological and hydrographic conditions as well as on estuarine processes, which are also imperfectly known.

Recommendations • Better control and management of the nitrogen cascade at the watershed scale is required to reduce N contamination of ground- and surface water, as well as coastal eutrophication. • In spite of the potential of these management measures, there is no choice at the European scale but to reduce the primary inputs of reactive nitrogen to watersheds, through changes in agriculture, human diet and other N flows related to human activity.

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

271

Nitrogen flows from European regional watersheds

13.1╇ Introduction A regional territory comprises a number of natural, seminatural and artificial landscapes, themselves composed of a mosaic of interacting ecosystems. The preceding chapters in this volume have emphasised that the complexity of landscape interactions, often occurring at the interface between ecosystems, prevents a simple additive approach to the functioning of large systems and their nitrogen budget; this is particularly true for regional territories. A regional watershed can be defined as a territory structured by a drainage network. Defining the limits of a territory in accordance with the limits of a watershed simplifies budget� ing approaches, as it allows a direct estimate of export through the hydrosystem, which is one of the major output terms in the nitrogen budget. However, this simple matter of budgeting convenience is not the sole reason to focus a discussion of the nitrogen cascade on the scale of regional watersheds. Indeed, drainage networks historically played a major role in structuring the European geographical space, often determining city settlement locations, the commercial

exchanges between them and the surrounding rural areas, hence the development of agriculture. Regional watersheds are therefore pertinent spatial units for studying the interactions between humans and the environment. Moreover, the coastal marine systems located at the outlet of regional watersheds are strongly influenced by the fluxes of water and nutrients delivered by the river, so that the whole continuum of ecosystems, including the catchments’ terrestrial systems, the drainage network, the estuarine zone and the coastal sea, should all be viewed and managed as a single integrated system. This is the point of view adopted in the present chapter. The major European watersheds are shown in Figure 13.1, grouped according to the marine coastal zones where they discharge. The full database of European watersheds used for this study is available as on-line supplementary material (see Supplementary materials, Section 13.7). It includes 5872 Â�individual catchments, most of them very small rivers. The major ones, with an area larger than 10 000 km², account for 67% of the total European coastal watershed area.

Figure 13.1 Major regional watersheds in Europe and their receiving coastal marine systems.

272

Gilles Billen

Figure 13.2 Schematic representation of the flows of reactive nitrogen through a regional watershed.

In this chapter, the nitrogen cascade will be examined at the scale of the major watersheds in Europe. The fate of reactive nitrogen brought into these regional territories through atmospheric deposition, synthetic fertiliser application, crop nitrogen fixation and commercial import of food and feed will be discussed at the regional basin scale. Particular emphasis will be placed on the riverine transfer of nitrogen from the terrestrial watershed to ground, surface and marine coastal waters, and on the consequences for the health of the marine systems. The analysis will be guided by the conceptual representation of nitrogen transfers through the different components of a regional watershed illustrated in Figure 13.2. Although this figure does not show all the complex interactions between the different components of the system, it clearly indicates the major sources of reactive nitrogen, the three major types of nitrogen output (to the atmosphere in its gaseous form, to other territories as food and feed, to the coastal seas as river loading) and the major pools with a long residence time (soils and aquifers) where nitrogen can be stored (and possibly remobilised) within the system. We will first establish the reactive nitrogen input–output budget of European watersheds and discuss the difference, often improperly termed ‘retention’, between total inputs and riverine outputs to the coastal zone. We then will take stock of the various modelling approaches used at the regional scale for relating nitrogen inputs to riverine outputs. Using both model results and observed fluxes, we will then examine the longterm trends of nitrogen riverine delivery, and its relations with phosphorus and silica, which is the key to understanding their potential for coastal marine eutrophication. The role of estuaries, acting as the last filter before delivery to the sea, will be briefly examined, prior to discussing the state of eutrophication of European coastal zones.

13.2╇ Input–output nitrogen budgets of regional watersheds 13.2.1╇ Inputs to watersheds As depicted in Figure 13.2, reactive nitrogen is brought into watersheds from atmospheric deposition, crop N2 fixation and synthetic fertiliser use, as well as by net commercial import of food and feed. All these terms are estimated at a rather fine geographical resolution scale for the whole of Europe, as discussed by Leip et╯al. (2011, Chapter 16, this volume). We present here only a short summary of these data. Data on atmospheric deposition of nitrogen as nitrogen oxides and as ammonium are available from the calculation of the EMEP project. Owing to the much shorter residence time of NH3 than nitrogen oxides in the atmosphere, a large part of deposited reduced nitrogen is short-distance re-deposition of emitted ammonia. Therefore, for the purposes of estimating net input of N to large watersheds, local emissions of ammonium by agricultural sources should be subtracted from deposition figures, or, as often done, only the figures for oxidised nitrogen deposition should be considered. Synthetic fertiliser application rates are calculated from the CAPRI database, which is fed by the national fertiliser application rate by crop, communicated by EFMA (European Fertilizers Manufacturers Association). Crop N2 fixation is evaluated from the data on legume crop and grassland distribution considering their respective rates of N2 fixation. As also discussed by Leip et╯al. (2011, Chapter 16, this volume), net commercial input/output of nitrogen as agricultural goods can be deduced from a budget of food and feed production by agriculture (autotrophy) versus local consumption by human and domestic animals (heterotrophy), both fluxes being expressed in terms of nitrogen (Billen et╯al., 2007, 2009a,

273

Nitrogen flows from European regional watersheds Figure 13.3 Distribution of the balance between autotrophy and heterotrophy of the main watersheds across Europe (EU-27) (as calculated from the CAPRI-DNDC database, Leip et╯al., 2011, Chapter 16, this volume). Green watersheds have an autotrophic status, while orange or red areas represent systems with heterotrophic status; yellow watersheds are balanced.

274

15000 Somme Severn autotrophy, kgN.km–2.yr –1

2010). Urban areas, where food is consumed but not produced, are obviously of dominant heterotrophic status. Rural regions specialised in crop farming and exporting their production to distant markets have an autotrophic status, while those which orient their agricultural activities towards intensive animal farming based on the import of feed from other regions have a heterotrophic status (Figure 13.3). When applied to European watersheds, this approach shows basins such as the Scheldt or the Po which have a strongly heterotrophic status, while basins such as the Seine or the Somme are highly autotrophic systems (Figure 13.4). Based on agricultural data from the second half of the twentieth century, Figure 13.4 shows the opposite trends in the historical trajectory of two exemplary basins, the Seine and the Scheldt, during the past 50 years:€ the former, turning towards exclusive cereal farming, become more and more autotrophic, while the latter, specialised in intensive animal husbandry, increased its heterotrophic status. These trends are important from the perspective of the nitrogen cascade, since the dominant source of nitrogen in autotrophic watersheds consists of inorganic fertilisers, while organic forms of nitrogen dominate in the nitrogen inputs to watersheds with heterotrophic status, which modifies the subsequent cascading nitrogen pathways. Summing up all the atmospheric deposition, fertiliser application and N2 fixation data, as well as the above-calculated net commercial imports (or exports) of nitrogen as food and feed, the net anthropogenic nitrogen input (NANI, Howarth et╯al., 1996) to each European watershed can be calculated. Expressed per square kilometre, it represents the intensity of anthropogenic disturbance of the N cycle by introduction of reactive nitrogen into the biosphere at the regional basin scale (Figure 13.5). Values range from a few tens of kgN/km²/yr in Scandinavian watersheds, where atmospheric deposition dominates the total, to over 10 000 kgN/km²/yr in watersheds

10000 1990

Ems

2000 Seine 5000

0

1970

Scheldt 1950 1980 2005 Po

1955 19th c

Thames

Shannon Elbe

Tevere

pre-medieval medieval Kalix 0 5000

10000 –2

heterotrophy, kgN km

yr

15000 –1

Figure 13.4 Autotrophy/heterotrophy diagram showing the historical trajectory of the Seine (●  ) and Scheldt (■) watersheds (Billen et╯al., 2009b). Autotrophy represents the agricultural production of food and feed, while heterotrophy represents local human and livestock consumption. The position of a number of other European basins (◯) is also shown.

bordering the North Sea, where either fertilisers or commercial imports of feed dominate, depending on their autotrophic or heterotrophic status. Net total nitrogen inputs (NTNI) to watersheds can also be defined; this differs from NANI by the natural rate of atmospheric nitrogen fixation both by lightning and by biological N2 fixation in the soils of natural ecosystems. In Europe this rate has been evaluated at 1.5–2.5 kgN/ha/yr in Scandinavian forests, 5–25 kgN/ha/yr in temperate forests and 10–35 kgN/ha/yr

Gilles Billen Figure 13.5 Basin averaged net anthropogenic nitrogen inputs to European watersheds, based on CAPRI-DNDC data (Leip€et╯al., 2011, Chapter 16, this volume).

in Mediterranean shrubland, based on the compilation by Cleveland et╯al. (1999).

13.2.2╇ Observed riverine nitrogen fluxes at the catchment outlet A database of nutrient fluxes delivered from large European watersheds at their outlet into estuarine/coastal waters has been assembled as part of NinE activities. The database (available as on-line supplementary material, see Section 13.7) includes recent data on total N, P and Si fluxes. Typically, average values of annual fluxes observed between 1995 and 2005 are recorded. When only inorganic nitrogen data where available, total nitrogen was estimated using the relationship between TN and DIN discussed by Durand et╯al. (2011, Chapter 6, this volume). Figure 13.6 summarises the available data. Documented watersheds in the NinE database cover 69% of the total European watershed area. Nitrogen delivery rates range from less than 200 to more than 4000€kgN/km²/yr. Total nitrogen fluxes exported from small watersheds may be even higher, with values over 10â•›000 kgN/km²/yr. Figure 13.7 shows that values higher than 2000 kgN/km²/yr are always associated with the presence of agriculture as a major share of land use in the catchment, although the exported nitrogen fluxes from watersheds with the same percentage of agricultural land may vary greatly, reflecting differences in agricultural practices (including the proportion of low-intensity grazing) as well as climatic and hydrological conditions.

inputs of nitrogen to watersheds is actually exported by the river to the coastal zone. Although misleading, the term ‘retention’ is used extensively in the literature to designate all the processes preventing nitrogen load (i.e. NTNI) to a watershed being transferred to the outlet of the drainage network (Dillon et╯al., 1990; Howarth et╯al., 1996; Windolf et╯al., 1996; Arheimer, 1998; Lepistö et╯al., 2001). It accounts for the net effect of various biogeochemical processes responsible for temporary or permanent N removal from the water phase (such as biological uptake and biomass production, sedimentation and denitrification) or N removal from the land phase (such as gaseous losses by denitrification and nitrification, volatilisation and N storage in permanent vegetation, soils and groundwater). Howarth et╯al. (1996, 2006), Boyer et╯al. (2002) and Alexander et╯al. (2002) showed that the flux of nitrogen exported by North American and Western European watersheds, over a background export of 107â•›kgN/km²/yr, accounts for a mean 26% of net anthropogenic nitrogen inputs (NANI), implying that 74% of the anthropogenically introduced nitrogen is retained or eliminated in the watershed. The data gathered in the NinE database allow testing this empirical relationship for the European watersheds for which an estimate of N delivery at the outlet is available (Figure 13.8a). Apparent retention, expressed as the fraction of NANI, varies from 90% to 50%, with a mean value of 82%. The regression of N delivery versus NANI is, however, highly significant:

13.2.3╇ Overall ‘retention’ within regional watersheds

Nflx (kgN/km²/yr) = 0.18 * NANI (kgN/km²/yr) + 228 (r²= 0.58)  (13.1)

Comparing the data in Figures 13.5 and 13.6 immediately shows that only a limited fraction of the total anthropogenic

The modelled background nitrogen export of 228 kgN/ km²/yr, although burdened with substantial uncertainty

275

Nitrogen flows from European regional watersheds

Riverine TN flux, kg/km²/yr

Figure 13.6 Available observed data on N delivery by European watersheds. Data from Humborg et╯al. (2003, 2006, 2008); Radach and Pätsch (2007); Lancelot et╯al. (1991); Billen et╯al. (2009a, b); Neal and Davies (2003); Meybeck et╯al. (1988); Ludwig et╯al. (2009); Cociasu et╯al.(1996); Johnes and Butterfield (2002); OSPAR (2002); REGINE (2010). See online supplementary material (Section 13.7) for original data.

14000 12000 10000 8000 6000 4000 2000 0 0

20

40

60

80

100

% Agriculture

Figure 13.7 Exported total nitrogen fluxes at the outlet of small to mediumsized watersheds from different areas of Europe as a function of the share of agricultural land (arable land and managed grassland) in total land cover. (data from Baltic countries (○), Germany and Czech Rep (□), United Kingdom (■), France (● ) and the Netherlands (◆)).

(± 100 kgN/km²/yr), agrees well with the experimental data from the boreal zone (Mattsson et╯al., 2003; Kortelainen et╯al., 2006; see below). Looking further to the variability of the retention factor, Howarth et╯al. (2006) found that it could be correlated with a simple climate variable such as precipitation or discharge (Q), with retention decreasing from 95% to 40% of total NANI when the specific discharge increases from 100 to 800 mm/yr in North American watersheds. The even more pronounced effect of specific discharge on N retention was also underlined by Behrendt and Opitz (2000) and Billen et╯al. (2009b). A sigmoid relationship of runoff has been proposed by Billen et╯al. (2010) for world watersheds. From the data presented in Figure 13.8a, no clear relationship of retention with specific runoff emerges, which could explain the variability observed around relation (13.1), although a

276

trend around a sigmoid relationship of discharge is observed when the data are grouped into Scandinavian, temperate or Mediterranean river systems (Figure 13.8b). Other factors such as variability in temperature- and soil moisture-induced biogeochemical processes in watershed soils are likely to play a role as well. The presence of lakes and their location in the watershed with respect to the outlet are also important factors contributing to overall retention (Arheimer and Brandt, 1998; Lepistö et╯al., 2006). On the basis of relationship (13.1), it is possible to use the distributed data on NANI (Figure 13.5) to calculate the most likely value of riverine-specific nitrogen delivery by undocumented European watersheds, thus interpolating the observed data of Figure 13.6. The overall nitrogen riverine delivery and retention for the watersheds of the major costal areas of Europe calculated on this basis are summarised in Table 13.1. According to this analysis, the total flux of nitrogen discharged by rivers into European coastal waters is 4760 ktonN/yr, accounting for 22% of the total amount of new nitrogen brought by anthropogenic processes to the corresponding watersheds (21 550 kton/yr). Although sizeable uncertainties affect these estimations, they clearly show the extreme perturbation of the N cycle in most European watersheds. They also stress that water resources contamination is only one of many important pathways of the cascade followed by anthropogenic nitrogen introduced into regional watersheds. Understanding and predicting the relation between N-related human activities in a watershed and the amount of N transferred by the hydrosystem is therefore a key scientific question. It is also a major management issue, as many measures can potentially act on retention processes.

Gilles Billen 5000 50% ret y = 0.18x + 228 r 2 = 0.54

FlxN, kgN/km2/yr

4000

3000

2000

90%ret

1000

0 0

5000

10000

15000

20000

2

NANI, kgN/km /yr

a.

1.25 nordic

fraction of NTNI exported

1.00

temperate

0.75 southern

0.50

13.3.2╇ A typology of models for regional watershed N transfers

0.25

0.00 0 b.

aquatic processes of nitrogen retention and makes no difference between the pathways through which nitrogen inputs are introduced to the hydrosystem, either as diffuse processes on the terrestrial watershed or as discrete point injection directly into the drainage network (Figure 13.9). This distinction between diffuse and point sources of nitrogen is important because different retention processes act on each of them:€landscape processes including storage in soil organic matter or biomass pools, soil denitrification or ammonia volatilisation, storage in deep aquifers, etc., act on diffuse sources of nitrogen. In-stream processes, including river bed denitrification or sediment storage, act on point sources and on diffuse sources after the action of landscape processes. Identifying and quantifying the various pathways of nitrogen contamination of and transformation in hydrosystems is therefore of prime importance for understanding the nitrogen cascade at the regional scale. Various models have been developed for this purpose and will be briefly described in the following section. To implement them, detailed data on point and non-point sources of nutrients to the hydrosystem are required. As far as point sources of nitrogen from urban wastewater are concerned, a detailed inventory is available at the European scale, thanks to the efforts of the EC within the implementation of the Water Framework Directive (Bouraoui et╯al., 2009). The geographical pattern shown (Figure 13.10) closely follows the distribution of large cities across Europe, with the transversal dorsal of rich cities extending from Birmingham to Milan (‘The Blue Banana’, Brunet, 2002). The definition of diffuse sources depends considerably on the particular model used, as will be shown below.

500

1000

1500

2000

runoff, mm/yr

Figure 13.8 (a) Specific N flux delivered at the outlet of European watersheds as a function of the net anthropogenic nitrogen inputs (NANI). The heavy line represents the regression across all points. The lighter lines represent N retention of 50% and 90% of NANI. (b) Fraction of NTNI delivered at the outlet as a function of runoff for Scandinavian (◊), temperate (■) and Mediterranean (▲) watersheds. The line represents the best fit of the sigmoid relationship proposed by Billen et╯al. (2010):€fraction NTNI exported = exp(−(Q−Qm)²)/Qs²), where Q is the mean specific runoff (mm/yr) of each watershed and Qm and Qs are climate-specific parameters.

13.3╇ Modelling N fluxes through watersheds 13.3.1╇ Point and diffuse sources of nitrogen to the hydrosystem The above NANI approach is based on a pure black-box input– output budget of the watershed as a whole. It quantifies the overall retention without identifying the processes responsible for it. In particular, it does not distinguish between terrestrial and

A large number of models have been used for quantifying nutrient transport and retention at the regional river-basin scale, which relies on a wide range of different assumptions and different methods for the description of nutrient sources, catchment characteristics and the physical and biogeochemical processes involved. Table 13.2 lists a number of such models, with their general basic equations and principles, their required input data and a list of watersheds where they have been applied. All models have been validated in diverse catchment areas and all are capable of satisfactorily predicting nutrient export from land use and point inputs, but they differ in the system boundaries, their spatial resolution, the complexity of their representation of the processes and their temporal resolution. A first difference lies in the definition of the system modeled:€some models work with the drainage network only; others encompass the entire watershed, including part or all of the soil and groundwater landscape components as well. This difference implies different ways of defining the diffuse sources of nitrogen and of dealing with the nitrogen dynamics in the soil–plant system above the root zone (plant growth, mineralisation, immobilisation, denitrification, etc.). For instance, many models, such as N-exret and RivR-N, are typically

277

Nitrogen flows from European regional watersheds Table 13.1 Overall nitrogen input, riverine delivery and percentage retention for the watershed of the major European coastal areas. (The documented area represents the percentage total watershed area for which observed data of nitrogen delivery are available; relationship (13.1) with NANI is used for the undocumented areas, then the overall specific delivery is calculated for the whole coastal zone watershed area.)

Basin area km²

Documented %

Specific delivery (weighted average) kgN/km²/yr

Net input (NANI) ktonN/yr

Total delivery ktonN/yr

Arctic Ocean

Retention of NANI % —

Norway & Finland Arctic coast

122 929

28

192

10

24

—

NW Norway coast

132 905

38

225

11

30

—

White Sea

281 927

0

244

—

69

—

83 974

66

309

116

26

78

Bothian Sea

499 488

88

211

181

105

42

Gulf of Finland

421 306

90

271

872

114

87

Gulf of Riga

137 846

90

634

712

87

88

S Baltic and Belt area

506 045

86

699

2720

354

87

47 135

27

339

—

16

—

Skagerak & Kattegat

196 115

80

509

302

100

67

Wadden Seas & W Danish coast

447 931

83

2017

3392

904

73

Seine Bight€– Belgian coast fringe

129 219

78

2024

888

262

70

E English & Scottish coast

115 361

48

1502

582

173

70

34 356

99

2667

289

92

68

Biscay Bay

267 545

92

1175

1217

314

74

Portuguese & W Spanish coasts

370 252

61

623

1202

231

81

W Scottish, Irish & Welsh coasts

Baltic Sea SE Sweden coast

North Sea W Norway coast

Atlantic Ocean Brittany

134 597

58

1386

473

186

61

S Welsh & English Coast

18 569

14

1374

103

26

75

Irish Sea

42 638

24

1758

303

75

75

S Spanish coast

42 440

0

988

179

42

76

305 052

74

811

1135

247

78

Tyrrhenian Sea

74 650

24

1122

359

84

77

Central Mediterranean Sea

13 433

0

995

57

13

77

Mediterranean Sea Lyon Gulf

Ionian Sea

61 775

9

767

193

47

75

Adriatic Sea

237 711

40

1191

1148

283

75

Aegean Sea

155 260

65

530

789

82

90

Black Sea W Black Sea coast Total

991 235

81

782

4317

775

82

5 868 102

69

811

21 550

4761

78

drainage network models which define diffuse inputs of nutrients on the basis of an export coefficient approach, i.e. by associating an empirically determined mean annual nitrogen flux of nutrients to the watershed’s different land use classes, without taking into account specific internal soil processes. The Seneque/Riverstrahler model uses a similar approach, except that the mean annual concentrations of surface and base flow

278

runoff are associated with land use/lithologic classes, these concentrations being either empirically defined or from offline runs of plant–soil models at the plot or landscape scales, as described by Cellier et╯al. (2011, Chapter 11, this volume). Full watershed models, such as Swat, Inca, Green or EveNFlow, fully integrate a description of the processes occurring in the top soil-plant system and define diffuse sources as the inputs

Gilles Billen Figure 13.9 Comparison of the input– output view of the watershed behind the black-box NANI approach (a), with a view of the watershed distinguishing between point sources from urban wastewater and diffuse sources from agricultural soils, and considering landscape and in-stream retention processes separately (b). Refer to Figure 13.2 for a more detailed view.

to soil as atmospheric deposition, biological nitrogen fixation, inorganic fertiliser and manure application. The complexity in the description of soil nutrient dynamics varies considerably, however, between these models, from simple regression relationships (e.g. Green) to detailed process modelling (e.g. Swat). Other models, such as Sparrow, Polflow and Moneris, define diffuse sources as the soil nitrogen surplus, i.e. the difference between total inputs to soil and outputs as crop yield, which is assumed to be directly transferred to the hydrosystem. The spatial resolution also differs widely between the models. Lumped approaches do not consider any spatial distribution of sources and sinks within the watershed; fully distributed models, on the other hand, account for the spatial variability of processes, nutrient inputs and watershed characteristics. The latter obviously require high-resolution spatial referencing of all constraint data, which might be difficult to obtain for large regional applications. Semi-distributed models illustrate the

intermediate case where sub-basins are divided into uniform units (e.g. in terms of land use or vegetation zones) as in the Green or Swat models, or a drainage network into a regular scheme of confluence of tributaries with mean characteristics by stream-order as in the Riverstrahler approach. Most often, depending on data availability and spatial resolution, models may compromise between fully distributed and totally lumped methods, also adapting to the level of process-description details. Indeed, another major difference between models lies in the complexity of their representation of the processes affecting the nitrogen transfer and transformations, ranging from a very simplified to an extremely complex description. Statistical regression models, such as Green, Sparrow and Polflow, consist in simple correlations of stream monitoring data with watershed sources and landscape properties and provide empirical estimates of nutrient stream export, based on a few explanatory

279

Nitrogen flows from European regional watersheds Figure 13.10 N point source emissions (urban communities and industries) on a 1â•›km² grid over Europe (from Bouraoui et al., 2009).

variables or predictors. The low data and time requirements of such methods explain their popularity for modelling nutrient fate in large river basins. Nevertheless, as soon as forecasting, i.e. explaining and describing the evolution over time of nutrient export and its dependence on the several controlling factors, is needed, mathematical models should be used for mechanistically describing the physical and biogeochemical underlying processes, as in the Inca, Swat and Seneque/ Riverstrahler approaches. Owing to their very detailed deterministic description of the processes, these models reduce site specificity to generically explain nutrient transfers with no or minimal need for calibration, hence providing a true understanding of the mechanisms involved. However, the intensive data requirement of such approaches leads to hybrid methods based on empirical relations for quantifying processes whose interactions are mechanistically expressed. While most regression models only predict mean annual nutrient fluxes, mechanistic models such as Swat, Inca and Seneque/Riverstrahler necessarily take into account seasonal patterns and provide inter-annually and seasonally variable flux and concentration results, which might be of prime importance for assessing the effect on receiving marine ecosystems. Finally, the models may differ in terms of the variables described, either total N or the different chemical (organic and mineral, dissolved or particulate) forms, and possibly other nutrients such as P and Si.

13.3.3╇ Comparison of different modelling approaches (EUROHARP) As illustrated above, a large range of modelling approaches of the nitrogen cascade in large watersheds are available, each of them corresponding to a specific objective and perspective.

280

The EUROHARP project (Kronvang et╯al., 2009) aimed at providing a broad range of end-users with unbiased guidance for an appropriate choice of model to satisfy existing European requirements on harmonisation, reliability and transparency for quantifying diffuse nutrient losses. It focused on diffuse nutrient losses, nitrogen and phosphorus in particular, from agricultural land to surface freshwater systems and coastal waters, to help end-users implement the Water Framework Directive and the Nitrate Directive. Nine different models were applied to 17 catchments in Europe covering a broad range of climatic, pedologic and farming practices gradients. The models were selected by each participant in the project as one of the official models being used in assessing nutrient losses to surface waters. The models selected included Moneris (Behrendt et╯al., 2002), Swat (Neitsch et╯al., 2001) and TRK/ HBV-NP (Brandt and Ejhed, 2002) (see Table 13.2), among others. The models are fully described in Schouman et╯al. (2003) and vary from simple loading functions to complex fully distributed mechanistic models. All models were applied to three core catchments, located in the UK, Italy and Norway, in order to fully investigate the similarities and differences in the various approaches not only in estimating the losses, but also in assessing the contribution of different pathways of losses, nutrient turnover, etc. At least four of the models were applied to each of the 14 remaining catchments in order to test their applicability. Overall it was concluded that no single model appeared consistently superior in terms of its performance across all three core catchments. Indeed, according to the output variable considered, depending on the goodness of fit of the test used, the models ranked differently on the three core catchments. The largest variations between model predictions (largest standard deviations) were found for the three Mediterranean catchments mostly due to the limited data availability when

281

Distributed drainage network model (1×1â•›km grid cells)

Lumped full watershed models

Semi-distributed (lumped by subbasins)

Semi-distributed drainage network (HBV-NP) or full watershed (TRK) model (sub-basins divided into elevation zones)

Lumped watershed model

Global-NEWS models Dumont et╯al., 2005 Seitzinger et╯al., 2009 Mayorga et╯al., 2010

Green Grizzetti et al., 2005, 2008 Bouraoui et╯al., 2009

(TRK)/HBV-NP Arheimer, 1998 Arheimer et╯al., 1998 Bergström et╯al., 1987 Brandt, 1990 Petersson et al., 2001

Moneris Behrendt, 2002

Geographical resolution

N-Exper is over Lepistö et al., 2001, 2006

Model and authors

Land use including tile drained areas Runoff divided into several pathways Point emissions (PS) Diffuse sources (DF) defined as soil N surplus

Daily meteorological data, point inputs (P), atmospheric deposition, land-use specific soil-leaching concentrations, potentially produced by a soil model (TRK)

Land use, rainfall, drainage network morphology (length, lake area), point emissions (PS), diffuse sources (DS) (fertiliser and manure application, atm. deposition, biological N2 fixation)

Land use, runoff, lake and reservoirs (dam) area, point emissions (PS) diffuse sources (DS) (fertiliser and manure application, atm. deposition, biological N2 fixation, minus N in crops and grass removed from land)

Point emissions, distributed land use, drainage network morphology,

Required input data

Nutrient outlet load = Σ (for different pathways) α.f(residence time).[DS.+ PS] where αs are a priori calibrated retention parameters

Mechanistic model of N soil dynamics (TRK) Simple calibrated first order, temperature-dependent, in-stream retention parameter (HBV-NP)

Nutrient outlet load = α.f(L,Area).[DS.β.f(R) + (PS+UL)] where α and β are calibrated in-stream and watershed retention parameters L is the total river length and Area the lake area in the watershed, R is rainfall, UL is the upstream load.

Nutrient outlet load = α.f(WA, RA, IR).[DS.β.f(Roff ) + PS)] where α and β are calibrated in-stream and watershed retention parameters WA is the watershed area RA is the reservoir (dammed) area IR is the water removed for irrigation Roff is runoff

N export coefficient by land use class; in-stream and riparian retention parameterised

Basic equation(s)/principle(s) for nutrient transfer representation

Table 13.2 A summarised description of a sample of models for nutrient transport and retention at the scale of regional watersheds

Mean annual

Daily

Annual mean

Annual mean

Annual mean

temporal resolution

Total N total P

DIN, orgN, totP

Total N total P

DIN, DON, PON

Total N

variables

German river basins

Entire Baltic Sea drainage basin

All European basins

All world watersheds

All Finnish watersheds

Watersheds

282

Distributed full watershed model (regular grid cells)

Semi-distributed full watershed model (sub-basin structure based on monitored reaches)

Distributed full watershed model

Semi-distributed full watershed model

Semi-distributed (Riverstrahler) or fully distributed (Sénèque) drainage network model

Sparrow Smith et╯al., 1997; Preston and Brakebill, 1999; Alexander et╯al., 2000, 2001

Inca Whitehead et╯al., 1998 Wade et╯al., 2002a, b

Swat Arnold et╯al., 1998, 1999 Neitsch et╯al., 2001, 2005

Riverstrahler/Sénèque Billen et╯al., 1994 Garnier et╯al., 1995, 2002; Ruelland et╯al., 2007; Thieu et╯al., 2009

Geographical resolution

Polflow De Wit, 2001

Model and authors

Table 13.2 (cont.)

Meteorological data, drainage network morphology, point emissions, land use, nutrient concentrations of superficial and base flow to streams for each land use class.

Meteorological data, topographic slope, soils, land use, nutrient emissions, agricultural management strategies

Daily meteorological and hydrological series, basin characteristics, point inputs, land use, growing seasons of crops, diffuse emissions (included fertilisers and livestock)

Basin characteristics (air temperature, precipitation, land-surface slope, soil permeability, stream density, and wetland area) and drainage network characteristics (discharge, time of travel), discharge data Point sources (PS) Diffuse sources (DF) defined as fertilisers and manure application, nonagricultural runoff and atm. deposition

Runoff and aquifer residence times, basin topography, soil & aquifer types, Point sources (PS) Diffuse sources (DF) defined as soil N surplus for each grid cell

Required input data

Kinetic formulation for each process describing the in-stream dynamics of nutrients, phytoplankton, zooplankton, bacteria (Rive model)

Mechanistic description of water, nutrient and pesticide routing and transformation in the watershed; mixing equations and simple parametric relationships for drainage network processes

Detailed mechanistic approach: * differential equations for describing losses in plant/ soil system and instream processes (nitrification, denitrification, sediment dynamics, biological uptake); * reaction rates are calibrated.

Nutrient load in each reach x = αf(cz, tt). [DS(x).β + (PS(x)+ Σ UL(x))] where α and β are calibrated in-stream and watershed retention parameters In-stream N retention depends on channel size (cz) and time of travel (tt) (first-order kinetics) Watershed retention depends on basin characteristics. UL is the upstream load.

Nutrient load at grid cell x= αf(slope).[DS(x).βf(s,r) + (PS(x)+UL(x))] where α and β are calibrated in-stream and watershed retention parameters In-stream N retention depends on slope and runoff Watershed retention depends on soil type and residence time in aquifers Denitrification in groundwater:€regression on residence times and infiltration. UL is the upstream load.

Basic equation(s)/principle(s) for nutrient transfer representation

10-Day periods

Daily

Daily

Mean annual

Mean annual

temporal resolution variables

NO3, NH4, diss&partorg N, P, Si, orgC, phyto/ zooplankton, bacteria

NO3, NH4, diss&partorgN

NO3, NH4, total P

Total N total P

Total N total P

Watersheds

Seine, Somme, Scheldt, Mosel, Danube, Kalix, Lule, Red rivers

Many watersheds in Europe and America

A wide range of catchments across Europe

Major US watersheds

Rhine, Elbe, Norrström

Gilles Billen Table 13.3 Budget of nitrogen to the basins of a number of rivers as calculated by the MONERIS Model for the 2001–2005 period (Behrendt et╯al., unpublished data)

kgN/km²/yr (%)

Danube

Rhine

Weser

Elbe

Odra

Input to land (before landscape retention)

2080

4400

5390

3810

2840

â•… from fertilisers and manure

930

2660

3500

2300

1430

â•… from atmospheric deposition

1150

1740

1890

1510

1410

Landscape retention

1490

3010

4030

2740

2100

â•… diffuse sources (after landscape retention)

590 (70)

1390 (76)

1360 (87)

1070 (78)

740 (80)

â•… background

60

80

70

50

40

â•…â•… from fertilisers and manure

210

670

720

530

360

â•…â•… from atmospheric deposition

330

740

570

490

340

Point sources

250 (30)

450 (24)

210 (13)

310 (22)

190 (20)

Total inputs to hydrosystem

840 (100)

1840 (100)

1570 (100)

1380 (100)

930 (100)

Delivery at outlet

560 (67)

1490 (81)

1300 (83)

920 (67)

400 (43)

In-stream retention

280 (33)

350 (19)

270 (17)

460 (33)

530 (57)

Table 13.4 . Nitrogen budget for three large European watersheds under wet and dry hydrological conditions, calculated by the Riverstrahler model (Trifu-Raducu, 2002; Thieu et╯al., 2009)

Danube

Seine

Scheldt

kgN/km²/yr (%)

1993 (dry)

1996 (wet)

1996 (dry)

2001 (wet)

1996 (dry)

2001 (wet)

Diffuse sources

966

1079

2012

3905

1227

2837

Point sources

281

281

553

553

1013

1013

Total inputs

1247 (100)

1360 (100)

2566 (100)

4459 (100)

2240 (100)

3850 (100)

Delivery at outlet

474 (38)

667 (49)

1378 (54)

2311 (52)

1213 (54)

2084 (54)

Groundwater storage

—

—

356 (14)

707 (16)

193 (9)

392 (10)

Riparian retention

273 (22)

312 (23)

524 (20)

1161 (26)

607 (27)

1167 (30)

In-stream retention

492 (39)

367 (27)

185 (7)

132 (3)

225 (10)

207 (5)

Reservoir retention

8 (0.6)

14 (1)

13 (0.5)

40 (1)

—

—

compared to the other catchments. Another critical factor affecting model results in these catchments was the model formulation, since in general most models were not developed to cover the Mediterranean regions typically characterised by nonpermanent flow, high rainfall intensity, etc. A similar limitation was found in the Norwegian catchments where none of the models considered frozen soils. The EUROHARP project highlighted that one of the major sources of discrepancy between the models is the quantification of the retention process. As most models are based on a mass balance approach, in order to accurately quantify the export of nitrogen at the catchment outlet, the models tended to adjust river retention accordingly, resulting in differences in retention estimates larger than one order of magnitude. It is important to note that even if most models did reproduce water and nutrient losses at the outlet reasonably well, the pathways of losses differed considerably between the models. To increase the reliability of the prediction of diffuse losses, it is suggested to scrutinise the internal processes and pathways simulated by the models whenever possible. The overall conclusion was that

the selection of the best model for N loss estimation should be made on a case-by-case basis depending on the catchment type, the purpose of the application, data availability, model limitations, expertise, etc. The parallel use of several models should always be recommended.

13.3.4╇ Environmental controls of N retention processes As explained in the above discussion, different models provide different visions of the nature and quantitative importance of retention processes. It is possible, however, to draw a number of general conclusions from the results of various models. Of particular interest in this respect are the results of those models which provide a detailed estimation of different pathways of nitrogen transfer through the watershed and the drainage network, and the corresponding retention. Examples of such results are presented in Tables 13.3 and 13.4, respectively from the Moneris model (a lumped, annual, calibrated model) and the Sénèque model (a distributed, seasonal, mechanistic model).

283

Nitrogen flows from European regional watersheds 80

The effect of nitrogen delivery on the coastal zone is highly dependent on the accompanying fluxes of the other nutrients required for the development of marine phytoplankton, particularly phosphorus and silica. For this reason, P and Si delivery rates were also gathered in the database established in the scope of the ENA (Figure 13.12). Table 13.5 summarises the data grouped according to the main coastal marine receiving areas. These are inter-annual average values, and it must be stated again that annual fluxes at the outlet of a regional basin can vary within a factor of two between a dry and a wet hydrological year. Moreover, since not all basins in each coastal zone watershed are documented, missing information has been obtained by extrapolation from nearby documented areas. The figures in Table 13.5 should therefore be considered rough estimates.

70 60

% N retention

50 40 30 20 10 0 –10 0

5

10

15

20

% lake area in the watershed Figure 13.11 In-stream N retention (%) as a function of the percentage lake area in the watershed for 30 Finnish river systems (data from Lepistö et╯al., 2006).

These models show the significance of processes occurring in the watershed’s upper soil layers, in the unsaturated zone and in the riparian wetlands for eliminating or storing nitrogen surplus from agricultural soil. Conversely, tile drainage, which affects agricultural soils in large areas in Europe (including Great Britain, France, the Netherlands, Denmark, Norway, the Baltic countries, Poland and Germany), accelerates nitrogen transfer to surface water. In-stream retention largely depends on the residence time of water masses through the drainage network and is therefore dependent on both the specific runoff and the presence of lakes and ponds (Figure 13.11).

13.4╇ N, P and Si delivery from watersheds 13.4.1╇ The present situation The level of nitrogen surface water contamination as revealed by the N delivery at the outlet of the major regional watersheds of Europe is depicted in Figure 13.6 above. Delivery rates are at least twice the background value in most of Europe except in the Scandinavian areas, and rates more than ten times the background are not unusual. This reflects a severe level of surface and groundwater contamination, which is described and discussed in Grizzetti et╯al., 2011 (Chapter 17 this volume). The total flux of nitrogen delivered to the sea along the EU27 coasts can be estimated at 4.8 TgN/yr of which 4.3 TgN/ yr comes from EU27 (4 171 851 km² watershed) and 0.5 TgN/ yr from outside EU27 (449 085 km²). This rate of N delivery is nearly five times the estimated natural background (0.98€TgN/yr).

284

13.4.2╇ The potential for coastal eutrophication (ICEP) It is now well recognised that the basic cause of coastal eutrophication is related not only to the general nutrient enrichment of the marine system, but also to the imbalance in the delivery of nitrogen (and phosphorus) with respect to silica. Indeed, many authors (Officer and Ryther, 1980; Conley et╯al., 1993; Conley, 1999; Turner and Rabalais, 1994; Justic et╯al., 1995; Billen and Garnier, 1997, 2007; Turner et╯al., 1998; Cugier et╯al., 2005) have shown that coastal eutrophication is the consequence of excess nitrogen and phosphorus delivery with respect to silica, in relation to the requirements of diatom growth. They underlined that coastal enrichment with nutrients brought in proportion of the Redfield ratios (Redfield et╯al., 1963), characterising the requirement of diatom growth, seldom causes problems, but, on the contrary, stimulates a healthy and productive food web, as is the case in upwelling areas where new planktonic primary production is mostly ensured by diatoms, while non-siliceous algae are restricted to regenerated production. By contrast, coastal eutrophication problems are the manifestation of new production of non-siliceous algae sustained by external inputs of nitrogen and phosphorus brought in excess over silica, thus in conditions where diatom growth is limited. Based on this view of coastal eutrophication, Billen and Garnier (2007) developed an indicator of coastal eutrophication potential (ICEP) of riverine nutrient inputs. This represents the carbon biomass potentially produced in the receiving coastal water body through new production sustained by the flux of nitrogen or phosphorus (depending on which one is limiting with respect to the other) delivered in excess over silica. For the purposes of a river-to-river comparison, it is expressed by unit of watershed area, in kgCkm2/day. It can be calculated by the following relationships (based on the Redfield molar C:N:P:Si ratios 106:16:1:20): N-ICEP = [NFlx / (14*16)€– SiFlx / (28*20) ] * 106 * 12 if N/P < 16 (N limiting) P-ICEP = [PFlx / 31€– SiFlx / (28*20) ] * 106 * 12 if N/P > 16 (P limiting)

Gilles Billen Figure 13.12 Available observed data on total P (a) and Si (b) delivery by European watersheds (see Figure 13.6 and supplementary material for references).

where PFlx, NFlx and SiFlx are, respectively, the mean specific fluxes of total phosphorus, total nitrogen and dissolved silica delivered at the outlet of the river basin, expressed in kgP/km²/ day, in kgN/km²/day and in kgSi/km²/day. A negative ICEP value indicates that silica is present in excess over the limiting nutrient (among nitrogen and phosphorus) and thus characterises the absence of eutrophication problems. Positive values indicate an excess of nitrogen or phosphorus over the potential for diatom growth, thus a condition for harmful non-siliceous algae development. As defined,

the ICEP does not take into account the particular conditions determining the response of the coastal zone into which the river is discharging, but simply represents the potential impact of the riverine fluxes. According to the N/P ratio of nutrient loading, N or P is the potential limiting nutrient. The ICEP should theoretically be calculated with respect to this nutrient. However, even in the case where P is limiting, a large excess of nitrogen with respect to silica is probably a risk for coastal eutrophication. This is because P is rapidly recycled in the marine environment, so

285

Nitrogen flows from European regional watersheds Table 13.5 Average specific fluxes of N, P and Si delivered by rivers into the different European coastal areas (P and Si flux values in italics and in brackets are educated guesses for undocumented areas). Corresponding Indicator of Coastal Eutrophication Potential (Billen and Garnier, 2007), calculated as the C€equivalent of either N (N-ICEP) or P (P-ICEP) brought in excess of Si with respect to the requirements of diatom growth

Weighted average river loading kgN/km²/yr

kgP/km²/yr

ICEP (N)

kgSi/km²/yr

ICEP (P)

Limitation

mgC/km²/day

Arctic Ocean Norway & Finland Arctic coast

192

3.1

865

−2.4

−5.0

P

NW Norway coast

225

6.0

992

−2.7

−5.5

P

White Sea

244

(4.5)

(900)

−1.8

−5.0

P

309

7.5

268

3.1

−0.8

P

−4.2

Baltic Sea SE Sweden Coast Bothian Sea

210

10

860

−2.1

Gulf of Finland

271

12

145

3.3

Gulf of Riga

634

15

434

7.2

−1.0

P

S Baltic and Belt area

698

33

497

7.8

0.6

P

W Norway coast

339

17

1427

−3.6

−7.0

P

Skagerak & Kattegat

509

644

3.9

−3.2

P

8.1

P

5.0

P

0.45

P P

North Sea

Wadden seas & W Danish coast

2017

Seine Bight€– Belgian coast fringe

2024

E English & Scottish coast

1502

Brittany

2667

Biscay Bay

7.4 136

1158

24

93

881

26

265

1074

17

23

N

21

858

36

−2.9

P

Atlantic ocean 1175

62

1429

9.4

−1.9

P

Portuguese & W Spanish coast

623

32

485

6.7

0.6

P

W Scottish, Irish & Welsh coast

1386

73

(1250)

14

0.5

P

S Welsh and English coast

1374

148

(1250)

14

8.9

P

Irish Sea

1758

216

1054

21

988

(30)

(485)

12

811

44

815

S Spanish coast

18

P

0.35

P

−0.11

P

Mediterranean Sea Lyon Gulf Tyrrhenian Sea

7.5

1122

30

800

12

−1.6

P

Central Mediterranean Sea

995

(25)

(800)

10.5

−2.1

P

Ionian Sea

767

21

800

6.9

−-2.6

P

Adriatic Sea

1191

92

2130

5.3

−2.9

P

Aegean Sea

530

246

1360

−0.2

19

N

782

31

267

11.5

Black Sea W Black Sea coast

that P limitation might not be effective as long as high nitrogen concentrations are available. Moreover, there is evidence that toxin production by non-siliceous as well as siliceous algae is enhanced in high nitrogen concentrations (Murata et╯al., 2006). An additional reason for considering the N-ICEP even in situations where the N/P ratio is above the Redfield ratio is that excess N not used in a coastal zone is likely to be exported to adjacent areas where it might cause eutrophication problems.

286

1.8

P

For the European rivers for which N, P and Si loading are documented the (N- and P-) ICEPs have been calculated (Figure€13.13). The mean values extrapolated to all European coastal areas are summarised in Table 13.5. Figure 13.13 clearly shows that excess N or P delivery with respect to silica is widespread in Europe, with the exception of northern Scandinavia. In most of Europe, nitrogen excess is much more pronounced than phosphorus excess; the reverse is true only in the southern part of the Balkan peninsula,

Gilles Billen Figure 13.13 Calculated values of N-ICEP (upper panel) and P-ICEP (lower panel) at the outlet of European watersheds.

where specific phosphorus delivery is still very high (see Figure 13.12).

13.4.3╇ Historical trends The land- and waterscape of Europe is the heritage of millennia of a complex human history which modified the land cover as well as the river morphology and hydrology. For a number of watersheds, retrospective studies have reconstructed past trends of nutrient inputs, transfer and delivery to the

coastal sea in response to changes in the constraints imposed by human society, using both historical records and modelling approaches (Andersson and Arheimer, 2003, for Swedish rivers; Behrendt et╯al., 2002, for the Odra and Danube; Billen et╯al., 2005, for the Scheldt basin; Billen et╯al., 2007, for the Seine basin; Stalnacke et╯al., 2003, for Latvian rivers). Such historical studies allow assessing the present degree of perturbation of European watersheds with respect to either pristine or historical situations. They are also particularly useful

287

Nitrogen flows from European regional watersheds

to examine the time lag involved in the response of compartments of the system with a very long life time, such as large aquifers and urban structures. We present here a summary of the general findings of these studies at the European scale.

Reconstituted pristine situations The pristine level of nitrogen inputs to river systems corresponds to background nitrogen concentration in runoff water from unperturbed forested areas, plus the input of litter from riparian trees. For a hypothetical pristine, entirely forested Seine watershed, Billen et╯al. (2007) estimated this to be 120– 300 kgN/km²/yr. The corresponding delivery at the outlet of the drainage network was in the range 60–150 kgN/km²/yr according to hydrological conditions. Similarly, Thieu et╯al. (2010) calculated values in the range 50–250 kgN/km2/yr for the pristine state of the Seine, Somme and Scheldt rivers. These figures are consistent with the value of 228 kgN/km²/yr found above for the y intercept of delivery vs NANI (Figure 13.8, relation (13.1)). They are also close to the values reported for present delivery rates of Swedish and Finnish rivers (see Table 13.1). Spatially representative long-term databases from 42 unmanaged headwater catchments covered by peatlands and forests showed average long-term N export around 130 kgN/km2/yr (site specific range, 29–230â•›kgN/km2/yr; Kortelainen et╯al., 2006) and 140 kgN/km2/yr (site specific range, 77–230â•›kgN/ km2/yr; Mattsson et╯al., 2003). Corresponding modelled pristine figures for phosphorus and silica delivery from the Seine, Somme and Scheldt basins are in the range of 8–30 kgP/km²/yr and 350–1500 kgSi/km²/yr (Thieu et╯al., 2010a, b). Observed phosphorus delivery in boreal Finnish rivers is 2–5 kgP/km²/yr. The average long-term P export from unmanaged Finnish catchments was 5.0 kgP/km²/yr (range, 1.7–15 kgP/km²/yr; Kortelainen et╯al., 2006) and 5.4€kgP/km²/yr (range, 2.1–18 kgP/km²/yr; Mattsson et╯al., 2003).

Preindustrial agricultural systems Traditional agricultural practices involved rotation alternating a fallow period and one or two cereal crops, and using manure fertilisation. Estimated nitrogen delivery from landscapes characterised by such agrarian systems varies between 300 and 800€kgN/Â�km²/yr in the Seine basin based on a few available measurements dating back to the nineteenth century (Billen et╯al., 2007). In an attempt to evaluate the nitrogen delivery from the Seine, Somme and Scheldt basins under a hypothetical scenario with generalised organic farming over their whole agricultural areas, Thieu et╯al. (2010a, b) obtained figures ranging from 430 to 950 kgN/km²/yr depending on the basin and the hydrology. Phosphorus release from agricultural soils increased significantly with respect to pristine levels because of higher erosion losses. Direct release of phosphorus from point sources from even small cities also leads to increased phosphorus contamination of surface water. For the Seine watershed, the estimated delivery in the periods preceding the twentieth century was in the range of 15–50 kgP/km²/yr. The question of the role played by agriculture in increasing silica delivery is still under debate. It was generally assumed

288

that dissolved silica concentration in runoff water, because it originates from rock weathering, only depends on the watershed’s lithology. However, some authors stressed the role of vegetation in a terrestrial silica cycle involving active uptake of silica from soil by plants and release of biogenic silica under the form of phytoliths, the dissolution or erosion of which contributes to the inputs of silica from soils to the surface water. Agriculture could therefore have influenced the diffuse sources of silica to river systems (Conley, 2002; Humborg et╯al., 2004). Rantakari and Kortelainen (2008) demonstrated that in a randomly selected Finnish lake database, SiO2 had highest correlation coefficient with TIC and CO2 in lakes surrounded by peatlands, the relation between SiO2 and inorganic carbon was less close in lakes surrounded by forests or agricultural land, supporting the important role played by biogenic Si cycling. The decomposition of organic matter produces organic acids and carbon dioxide, both of which enhance weathering and thus SiO2 concentrations.

From 1950 to 1985 In most regions of Europe, the second half of the twentieth century was characterised by both increased urbanisation, often with few wastewater treatment infrastructures, and generalisation of modern agricultural practices with increased use of synthetic fertilisers. In the Seine basin, N delivery peaked in the 1980s at 1500–3000â•›kgN/km²/yr (according to hydrological conditions). In the Odra river (where low specific discharge is responsible for high retention), Behrendt et╯al. (2005a) calculated an increase from 270 kgN/km²/yr in 1960 to 595 kgN/km²/yr in 1980. As far as phosphorus is concerned, this period is also characterised by the substitution of traditional soap products by P-containing washing powders, which led to a fourfold increase in the per capita P release rate in urban wastewater. In the Seine basin, delivery rates as high as 350 kgP/km²/yr were reached at the end of 1980. In the Odra River the increase during the 1960–1980 period was from 20 to 50€kgP/km²/yr. Silica release from domestic wastewater, although not insignificant (Sferratore et╯al., 2006), is relatively low with respect to the N and P content of urban wastewater. Moreover, eutrophication of surface water, owing to N and P contamination, often resulted in increased retention of dissolved silica related to a more intense diatom development in rivers. Impoundments of large reservoirs also led to increased silica retention either by algae development and trapping of dissolved silica or biogenic particulate silica produced upstream, or by reducing rock weathering in flooded, former wetlands (Humborg et╯al., 2004, 2006). As a result, the period of industrialisation and urbanisation of the second half of the twentieth century was characterised by a significant decrease in silica delivery, while N and P fluxes increased tremendously.

The economic transition of Eastern countries The period following the collapse of the former USSR was characterised by major changes in agricultural and industrial activity in all countries of Eastern Europe, resulting in a considerable decrease in fertiliser application and industrial wastewater discharge.

Gilles Billen 10

per capita N flux, kgN/cap/yr

In the Baltic countries (Estonia, Latvia, and Lithuania), formerly specialised in cattle farming, import of mineral fertilisers and feedstuff decreased by a factor of 15 between 1987 and 1996, and the livestock was reduced fourfold, decreasing the use of manure. Yet, this dramatic reduction of the intensity of agriculture led to only a slow and limited response in Latvian rivers’ N load, due to the inertia of the soils and aquifer compartments (Stalnacke et╯al., 2003), while the response in terms of P load was more visible. Similar observations are reported by Behrendt et╯al. (2005a) for the Odra River. For the Danube basin, although some disagreement exists on its amplitude, a clear decrease of N and P delivery to the Black Sea was reported in the early 1990s (by 9–23% for N, by 25–35% for P), as a result of decreased diffuse sources and point sources (Behrendt et╯al., 2005b).

13.5╇ The estuarine filter 13.5.1╇ Typology of European estuaries Before reaching the sea, the flux of nutrients delivered at river outlets has to cross their estuarine zones, which are often biogeochemically very active systems. The different coastal systems of Europe, however, offer a wide range of estuarine types, differing in their filtering effect for riverine nutrients. Meybeck and Dürr (2009) have proposed a typology of estuaries, based on coastal morphology, tidal influence and freshwater discharge, distinguishing (i) fjords and fjärds (deep glacial valleys filled with marine water, but where snow melt leads to rapid transit of surface freshwater to the coastal zone), (ii) rias (drowned river valleys, dominated by seawater dynamics), (iii) macrotidal estuaries (where the tidal circulation gives rise to the development of a turbidity maximum zone), (iv) deltas (prograding wedges of sediment at the river mouth with restricted entrance of seawater) and (v) lagoons (littoral shallow brackish

6

Seine Oder Danube Scheldt

4

2

0 1900

The recent trends

2 per capita P flux, kgP/cap/yr

During the past 10–15 years, considerable efforts have been devoted to improving surface water quality in most European countries. This resulted in a spectacular decrease of point inputs of nutrients through wastewater discharge. The effect is particularly striking for phosphorus, as improved wasteÂ� water treatment was accompanied by the substitution of polyphosphate as a sequestering agent in washing powders, which resulted in a three- to four-fold reduction of the per capita rate of phosphorus release in domestic wastewater. Tertiary treatment of nitrogen in wastewater purification plants is also in progress, particularly in northern European countries. Diffuse inputs of nutrients by agriculture, however, are still at a high level, in spite of the agro-environmental measures advocated by most European Water Authorities. The inertia of soil and aquifer reservoirs, mentioned above, is here added to the conservatism of many components of the socio-economic agricultural sphere. As a result, phosphorus delivery is rapidly decreasing at the outlet of most European rivers, while nitrogen delivery is still increasing or at best levelling off (Figure 13.14).

8

1920

1940

1960

1980

2000

1960

1980

2000

Seine Oder Danube Scheldt

1

0 1900

1920

1940

Figure 13.14 Trends of N and P delivery for different European rivers during the twentieth century, normalised to the total population of the watershed. Data are a combination of observations and a model reconstruction from Billen et╯al., 2005, 2007 (the Scheldt and the Seine), Behrendt et╯al., 2002, 2005a,b (the Oder and the Danube). For the Seine and Scheldt, the error bars show the values for wet and dry hydrological conditions, while the other values refer to mean hydrological conditions.

ponds with permanent or temporary sea water exchange). Karstic areas are characterised by direct inputs of groundwater to the sea. Figure 13.15 shows the dominant types of estuaries along the coasts of Europe.

13.5.2╇ Estuarine nutrient retention Estuarine nutrient processing is highly varied and too few studies are available to make any generalising quantitative statement on the filtering effect of estuaries on riverine nutrient fluxes. Figure 13.16 summarises a number of European case studies where the retention of the land-based nitrogen loading during the transit through the estuarine zones has been evaluated. These studies highlight the effect of residence time on overall retention. Nixon et╯al. (1996) proposed a relationship similar to that proposed by Kelly et╯al. (1987) for lakes, relating N retention during estuarine transit to depth and residence time; this relationship fits generally well with the data assembled for European estuaries (Figure 13.16).

289

Nitrogen flows from European regional watersheds Figure 13.15 Dominant types of estuaries along Europe’s coastlines (Meybeck and Dürr, 2009).

100

13.6╇ Nitrogen delivery and coastal eutrophication

% N retention

80 Arcachon lagoon

60

13.6.1╇ Coastal eutrophication in European coastal waters

Scheldt

40

Oder

20 Seine 0 1

10

Norsmind fjord Tweed 100

Danube Pô Tyne 1000

10000

depth/residence time, m/yr Figure 13.16 Observed N retention during transit through some European estuaries, plotted versus the depth/residence time ratio. The line represents the relationship found by Nixon et╯al. (1996) for a number of North Atlantic American estuaries. European case studies include deltas [the Danube (TrifuRaducu, 2002); the Po (De Wit and Bendoricchio, 2001), the Rhone (Pettine et╯al., 1998; El-Habr and Golterman, 1987), Oder (Pastuszak et╯al., 2005)], macrotidal estuaries (Seine (Garnier et╯al., 2010), the Scheldt (Billen et╯al., 1985), the Tyne and Tweed (Ahad et╯al., 2006)), a fjord (Norsminde fjord (Nielsen et╯al., 1995) and a lagoon Arcachon lagoon, DeWit et╯al., 2005).

Those estuarine systems where substantial nitrogen processing is occurring, including nitrification and denitrification, are often characterised by N2O concentrations far above saturation, indicating that they act as a source for this greenhouse gas. This has been observed in the Tamar (Law et╯al., 1992), the Humber (Barnes and Owens, 1998), the Scheldt (de Wilde and de Bie, 2000), and the Seine (Garnier et╯al., 2006) estuaries, where emission rates ranged from 0.4 to 5 gN-N2O/m²/yr. In the case of the Tyne estuary where high ammonium release occurs in the rapidly flushed estuarine zone, an area of high N2O emission is observed in the adjacent coastal zone (Ahad et╯al., 2006).

290

Riverine delivery considerably affects (positively or negatively) the ecological functioning of coastal marine ecosystems, as it most often represents the major source of new nutrients for primary production. Satellite determination of coastal marine algal biomass have been available since the early 2000s, based on the radiometric observation of changes of seawater colour from blue to green as the chlorophyll concentration increases. A composite image of chlorophyll distribution in European coastal zones is shown as an example in Figure 13.17. The conversion of the optical signal to in situ pigment concentration relies on the calibration of algorithms which are highly dependent on the presence of various organic and inorganic constituents of seawater and can lead to severe overestimation of actual biomass (Darecki and Stramski, 2004). Qualitatively, however, Figure 13.17 clearly shows the effect of riverine nutrient discharge on algal biomass distribution in European coastal zones. In and of itself, this enhancement of primary producer biomass would not be a problem, if it were not often accompanied by profound changes in the structure of the food webs and a decline of zooplankton grazing and commercial fish production (Vasas et╯al., 2007), as well as by diverse harmful manifestations such as organic matter accumulation, toxin production, anoxia, etc. A detailed account of the problems related to coastal eutrophication is provided by Voss et╯al. (2011, Chapter€8, this volume).

Gilles Billen

Figure 13.17 Composite satellite image of mean chlorophyll concentration along the European coasts in 2007 (from MODIS-Aqua satellite data, source:€JRC, http://marine.jrc.ec.europa.eu/). Annual mean and maximum values in brackets of direct measurements at selected stations are also shown to provide an absolute reference (Lancelot et╯al., 2005; Solidoro et╯al., 2009; M. Voss, personal communication).

13.6.2╇ Comparing indicators and observation of coastal eutrophication Ignoring the role played by the estuarine filter on riverine nutrient delivery, the data gathered in Table 13.5 can be compared with the available observations of coastal eutrophication along European coastlines (Figure 13.18). On the basis of the N/P ratio of nutrient river loading calculated in Table 13.5, phosphorus presently appears as the potentially limiting nutrient in most coastal areas, except in the Aegean Sea, where phosphorus loading is still extremely high. This situation is recent and contrasts with that of the 1980s when, owing to much higher P loading, nitrogen was likely to be the limiting nutrient in most European coastal areas (see Section 13.4.3 and Figure 13.14). Note that the Gulf of Finland, as well as many other areas of the Baltic Sea, are still regarded as N-limited for most of the growth period (Graneli et╯al., 1990; Tamminen and Andersen, 2007). P limitation increases towards the north in the Gulf of Bothnia (Tamminen and Andersen, 2007). Admittedly,

the N/P calculation based on riverine deliveries does not take into account the effect of the biogeochemical processes in receiving coastal waters (sedimentation, denitrification, sediment release). Thus, in the Baltic Sea these processes tend to shift the ratios from estuarine P towards N limitation in the open sea (Pitkänen and Tamminen, 1995; Tamminen and Andersen, 2007). The ICEP shows negative values (whether it is estimated on the basis of N or P) in the Arctic coastal zones as well as in the northern Baltic. Positive ICEP values are reached in the southern Baltic, as well as in the North Sea and along most Atlantic coasts. In the Mediterranean, positive ICEP values are observed in rivers flowing into the Adriatic and Tyrrhenian seas. The western Black Sea coast is also characterised by high ICEP values. The distribution of European coastal areas designated as subject to the risk of eutrophication in the discussion above fits well with the observations of eutrophication problems, although the manifestations of eutrophication might be quite different depending on the local physiographical and hydrological

291

Nitrogen flows from European regional watersheds

Figure 13.18 Calculated Indicator of Coastal Eutrophication Potential (ICEP) by European coastal region, based on the data from Table 13.5. Identification of the major coastal areas where eutrophication problems are recorded.

conditions:€blooms of toxic algae, as in the Seine Bight (Cugier et╯al., 2005) and in the Baltic (Hansson, 2008; HELCOM, 2009), massive development of mucilaginous, unpalatable, algal species in the North Sea (Lancelot et╯al., 1987, 2005, 2007), the Black Sea (Cociasu et╯al., 1996) and the Adriatic Sea (Marchetti, 1991), deposition of increasing amounts of organic material resulting in anoxic bottom waters as in the northern Adriatic (Justic, 1991), Danish coastal waters (Babenerd, 1990) and Baltic coastal zones (HELCOM, 2009). In Brittany and on the other Atlantic coasts, the very rapid dilution of fresh water masses due to tidal currents often prevents the development of dense planktonic blooms. Eutrophication is mainly apparent from the development of benthic macro algae close to the coast, although development of toxic dinoflagellate blooms might also be a problem during summer when the water column is stratified. The continental coastal zone of the English Channel and the North Sea, from Normandy to the Danish coast, is one of the more severely eutrophicated areas in the world, with the occurrence of heavy blooms of Phaeocystis globosa colonies every spring, responsible for the accumulation of mucus foam on the beaches (Lancelot, 1995). Marine ecological model simulations constrained by river nutrient load simulations suggest that the maximum biomass reached by Phaeocystis increased threefold from 1950 to 1990 and has now decreased by about 20% (Lancelot et╯al., 2007). The Baltic Sea is a nearly enclosed brackish-water area, with seawater renewal occurring through the narrow Danish Straits and Sound areas linking the Baltic to the North Sea. Major inflows of seawater have only occurred rarely in recent decades, leaving the water in the deeper basins without a renewal of oxygen. Salinity stratification, small water volume and long residence time are the main physical reasons for the

292

sensitivity of the Baltic Sea to eutrophication (Leppäranta and Myrberg, 2009). The sea is heavily impacted by nutrient loading and anoxic conditions promoting release of inorganic phosphorus from the sediments (Pitkänen et al, 2001). The impacts of eutrophication are manifested as various symptoms such as increased nutrient concentrations and phytoplankton biomass, oxygen deficiency and elimination of benthic fauna, as well as frequent blooms of filamentous cyanobacteria (Lundberg, 2005). The northwestern Adriatic Sea, subject to the inputs of the Po River, also suffers from the development of non-siliceous algae leading to the production of mucilaginous substances. Detection of organic-walled dinoflagellate on sediment cores revealed a clear shift to eutrophication conditions from 1930 onwards, reaching a peak in the 1960–1980 period. Subsequently, eutrophication levels decreased, although dinocyst diversity suggests that the ecosystem has not completely recovered (Sangiorgi and Donders, 2004). The western coast of the Black Sea has been experiencing a severe process of degradation since the early 1960s. From a diverse ecosystem with rich ecological resources, it evolved into a low biodiversity zone where jellyfish and ctenophores replaced zooplankton–fish food chains. An almost total collapse of fisheries occurred in the late 1980s (Mee, 1992; Lancelot et╯al., 2002). The considerably reduced Danube nutrient discharge over the past 15 years, following the collapse of industry and agriculture in the former Soviet countries of the Danube catchment area, however, induced a trend towards restoration of the marine ecosystem. The species diversity of macrozooÂ� benthos has increased since 1996 in front of the Danube delta (Horstmann et╯al., 2003), but ctenophores and medusa still dominate the zooplankton, preventing the full regeneration of fish populations.

Gilles Billen

13.7╇ Conclusions Urbanisation and the spread of industrial fertilisation techniques in agriculture in most European territories have led to an unprecedented opening of the nitrogen cycle which resulted in increased inputs of reactive nitrogen to watersheds. Compared with the background pristine inputs of N through natural biological fixation and atmospheric deposition (460– 1800 ktonN/yr, based on the figures compiled by Cleveland, 1999), net anthropogenic inputs of reactive nitrogen to EU27 (21 540 ktonN/yr) are two to ten times higher. A fraction of only about 20% of these inputs ultimately reaches the outlet of the hydrographic network of large river systems, while both landscape and aquatic processes contribute to retention of the remaining 80% of anthropogenic inputs. Landscape processes include storage of nitrogen in the soil organic matter pool and in the groundwater. This is temporary storage, which simply confers a great inertia of the response of riverine delivery to changes in diffuse inputs:€ depending on the residence time of nitrogen in these reservoirs, the reaction to any change in land use and agricultural practices in terms of nitrogen flux at the basin outlet can be delayed by several decades. Soil and riparian zone denitrification are other processes contributing substantially to landscape retention; the elimination of nitrate by this pathway unfortunately is accompanied by harmful emissions of N2O. In-stream nitrogen retention processes are dominated by benthic denitrification both in the river bed and in small water storage structures such as ponds and shallow reservoirs. This process also leads to N2O emissions, however. The relative role played by lakes in terms of N retention within watersheds is important:€ two major processes involved are denitrification and sedimentation. Wastewater treatment must be considered a retention process when it involves specific processes for N elimination, most often through denitrification, accompanied, once again, by N2O emissions. In spite of these effective retention mechanisms, many of which can still be improved by suitable management, nitrogen delivery to coastal systems at the European scale, now totalling 4750 ktonN/yr, increased more than fourfold with respect to the pristine state, and approximately threefold with respect to the pre-1950 situation. At the same time, phosphorus delivery increased, but is now decreasing again close to preindustrial levels, owing to effective P abatement measures in urban wastewater purification implemented in most European countries. Silica delivery, on the other hand, is decreasing due to both reduced rock weathering and enhanced retention in watersheds, mostly linked to dam construction (Humborg et╯al., 2008). The consequence of these changes is that the riverine input of nutrients to the coastal zones, which used to be a major factor contributing to the richness of these areas providing most of the fish catch, is now largely imbalanced, resulting in severe eutrophication problems. Particularly affected are the south-eastern continental coast of the North Sea, the Baltic Sea (except the Gulf of Bothnia), the coasts of Brittany, the Adriatic Sea and the western Black Sea coastal area.

Better knowledge and understanding of the processes leading to retention and elimination of reactive nitrogen once introduced within watersheds would certainly allow better management of land- and waterscapes with the objective of reducing the N fluxes transferred to the sea and to the atmosphere as reactive species. However, whatever the potential of such management measures, there will be no other choice for durably improving the situation than reducing the anthropogenic nitrogen load, through changes in agriculture, human diet and other nitrogen flows related to modern human activity.

Acknowledgements This chapter was prepared with the support of the NinE Programme of the European Science Foundation, the NitroEurope IP (funded by the European Commission) and the COST Action 729. It grew from the discussions held in two workshops held in Paris in January 2007 (with the support of both NinE and LOICZ) and in Dourdan in November 2008 (with the support of NinE and COST729). It also benefited from the participation in other collaborative networks including TIMOTHY Interuniversity Attracting Pole of the Belgian Science Policy and the AWARE EC-FP7 programme. We acknowledge Amelie Danacq for establishing Table 13.2. We also acknowledge Hast Behrendt for his active participation in this workshops mentioned above; he has since passed away. We dedicate this chapter to his memory.

Supplementary materials Supplementary materials (as referenced in the chapter) are available online through both Cambridge University Press:€ www.cambridge.org/ena and the Nitrogen in Europe website:€www.nine-esf.org/ena.

References Ahad, J. M. E., Ganeshram R. S., Spencer, R. G. M. et╯al. (2006). Evaluating the sources and fate of anthropogenic dissolved inorganic nitrogen (DIN) in two contrasting North Sea estuaries. Science of the Total Environment, 372, 317–333. Alexander, R. B., Smith, R. A. and Swartz, G. E. (2000). Effect of stream channel size on the delivery of nitrogen to the Gulf of Mexico. Nature 403, 758–761. Alexander, R. B., Smith, R. A., Schwarz, G. E. et╯al. (2001). Atmospheric nitrogen flux from the watersheds of major estuaries of the United States:€an application of the SPARROW watershed model. In: Nitrogen Loading in Coastal Water Bodies:€An Atmospheric Perspective, ed. R.€Valigura,R . Alexander,M . Castro, et╯al. American Geophysical Union, Madison, WI, pp.€119–170. Alexander, R. B., Johnes, P. J., Boyer, E. W. and Smith, R. A. (2002). A comparison of models for estimating the riverine export of nitrogen from large watersheds. Biogeochemistry, 2002, 295–339. Andersson, L. and Arheimer, B. (2003). Modelling of human and climatic impact on nitrogen load in a Swedish river 1885–1994. Hydrobiologia, 497, 63–77. Arheimer, B. (1998). Riverine nitrogen€– analysis and modelling under nordic Conditions. Ph.D. Thesis. Linköping University, Sweden. Arheimer, B. and Brandt, M. (1998). Modelling nitrogen transport and retention in the catchments of southern Sweden. Ambio, 27, 471–480.

293

Nitrogen flows from European regional watersheds Arnold, J. G., Srinivasan, R., Muttiah, R. S. and Williams, J. R. (1998). Large area hydrologic modeling and assessment. Part I:€Model development. Journal of the American Water Resources Association, 34, 73–89. Arnold, J. G., Srinivasan, R., Muttiah, R. S., Allen, P. M. and Walker,€C. (1999). Continental scale simulation of the hydrologic balance. Journal of the American Water Resources Association, 35, 1037–1052. Babenerd, B. (1990). Eutrophication and algal blooms. Water Pollution Research Reports, 12,€215–223. Barnes, J. and Owens, N. J. P. (1998). Denitrification and nitrous oxide concentrations in the Humber Estuary, UK, and adjacent coastal zones. Marine Pollution Bulletin, 37, 247–260. Behrendt, H. (2002) The Model System MONERIS. EUROCAT WD. Behrendt, H. and Opitz, D. (2000). Retention of nutrients in river systems:€dependence on specific runoff and hydraulic load. Hydrobiologia, 410, 111–112. Behrendt, H., Huber, P., Kornmilch, M. et al. (2002). Estimation of the nutrient inputs into river basins: experiences from German rivers, Regional Environmental Changes, 3, 107–117. Behrendt, H., Opitz, D., Korol, R. and Stronszka, M. (2005a). Changes of the nutrient loads of the Odra River during the last century: their causes and consequences. In: Proceedings of the ICID 21st European Regional Conference ‘Integrated Land and Water Management:€Towards Sustainable Rural Development’, Frankfurt/ Oder, Germany, April 2005. Behrendt, H., van Gils, J., Schreiber, H. and Zessner, M. (2005b). Point and diffuse nutrient emissions and loads in the transboundary Danube River Basin. Part II: Long term changes. Archives of Hydrobiology (Suppl.), 158, 221–247. Bergström, S., Brandt, M. and Gustafson, A. (1987). Simulation of runoff and nitrogen leaching from two fields in southern Sweden. Hydrological Science Journal, 32, 191–205. Billen, G . and Garnier, J. (1997). The Phison River Plume:€coastal eutrophication in response to changes in land use and water management in the watershed. Aquatic Microbiology and Ecology, 13, 3–17. Billen, G. and Garnier, J. (2007). River basin nutrient delivery to the coastal sea:€assessing its potential to sustain new production of non siliceous algae. Marine Chemistry, 106, 148–160. Billen, G., Somville, M., De Becker, E. and Servais, P. (1985). A nitrogen budget of the Scheldt hydrographical basin. Netherlands Journal of Sea Research, 19, 223–230. Billen, G., Garnier, J. and Hanset, P. (1994). Modelling phytoplankton development in whole drainage networks:€the RIVERSTRAHLER model applied to the Seine river system. Hydrobiologia, 289, 119–137. Billen, G., Garnier, J. and Rousseau, V. (2005). Nutrient fluxes and water quality in the drainage network of the Scheldt basin over the last 50 years. Hydrobiologia, 540, 47–67. Billen, G., Garnier, J., Némery, J. et╯al. (2007). Nutrient transfers through the Seine river continuum:€mechanisms and long term trends. Science of the Total Environment, 375, 80–97. Billen, G., Barles, S., Garnier, J., Rouillard, J. and Benoit, P. (2009a). The food-print of Paris:€long term reconstruction of the nitrogen flows imported to the city from its rural hinterland. Regional Environmental Change, 9, 13–24. Billen, G., Thieu, V., Garnier, J. and Silvestre, M. (2009b). Modelling the N cascade in regional watersheds:€the case study of the Seine, Somme and Scheldt rivers. Agriculture, Ecosystems and Environment, 133, 234–246. Billen, G., Beusen, A., Bouwman, L. and Garnier J. (2010). Anthropogenic nitrogen autotrophy and heterotrophy of the

294

world’s watersheds:€past, present, and future trends. Global Biogeochemical Cycles, 24, GB0A11, doi:10.1029/2009GB003702. Boyer, E. W., Goodale, C. L., Jaworski, N. A. and Howarth, R. W. (2002). Anthropogenic nitrogen sources and relationships to riverine nitrogen export in the northeastern USA. Biogeochemistry, 57/58, 137–169. Brandt, M. (1990). Simulation of runoff and nitrogen transport from mixed basins in Sweden. Nordic Hydrology, 21, 13–34. Brandt,M. and Ejhed, H. (2002). TRK, Transport-RetentionKällfördelning, Belastning Pahavet, Swedish Enviroment Protection Agency, Report 5247, Stockholm, Sweden, 2002, (in Swedish). Brunet, R. (2002). Lignes de force de l’espace Européen. Mappemonde, 66,€14–19. Cellier, P., Durand, P., Hutchings, N. et╯al. (2011). Nitrogen flows and fate in rural landscapes. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Cleveland, C. C., Townsend, A. R., Schimel, D. S. et╯al. (1999). Global patterns of terrestrial biological nitrogen fixation in natural ecosystems. Global Biogeochemical Cycles, 13, 623–645. Cociasu, A., Dorogan, L., Humborg, C. and Popa, L. (1996). Longterm ecological changes in Romanian coastal waters of the Black Sea. Marine Pollution Bulletin, 32, 32–38. Conley, D. J. (1999). Biogeochemical nutrient cycles and nutrient management strategies. Hydrobiologia, 289, 87–96. Conley, D. J. (2002). Terrestrial ecosystems and the global biogeochemical silica cycle. Global Biogeochemical Cycles, 16, 1121, doi: 10.1029/2002GB001894. Conley, D. J., Schelske, C. L. and Stoermer, E. F. (1993). Modification of the biogeochemical cycle of silica with eutrophication. Marine Ecology Progress Series, 81, 121–128. Cugier, P., Billen, G., Guillaud, J. F., Garnier, J. and Ménesguen,€A. (2005). Modelling eutrophication of the Seine Bight under present, historical and future Seine river nutrient loads. Journal of Hydrology, 304, 381–396. Darecki, M. and Stramski, D. (2004). An evaluation of MODIS and SeaWiFS bio-optical algorithms in the Baltic Sea. Remote Sensing of the Environment, 89, 326–350. de Wilde, H. P. J. and de Bie, M. J. M. (2000). Nitrous oxide in the Schelde estuary:€production by nitrification and emission to the atmosphere. Marine Chemistry, 69, 203–216. De Wit, M. (2001). Nutrient fluxes at the river basins scale. Part I:€The PolFlow model. Hydrological Processes, 15, 743–759. De Wit, M. and Bendoricchio, G. (2001). Nutrient fluxes in the Po basin. Science of the Total Environment, 273, 147–161. De Wit, R., Leibreich, J., Vernier F. et╯al. (2005). Relationship between land-use in the agro-forestry system of les Landes, nitrogen loading to and risk of macro-algal blooming in the Bassin d’Arcachon coastal lagoon (SW France). Estuarine, Coastal and Shelf Science, 62, 453–465. Dillon, P. J., Evans, R. D. and Molot, L. A. (1990). Retention and resuspension of phosphorus, nitrogen, and iron in a central Ontario lake. Canadian Journal of Aquatic Science, 47, 1269–1274. Dumont, E., Harrison, A., Kroeze, C., Bakker, E. J. and Seitzinger, S.€P. (2005). Global distribution and sources of dissolved inorganic nitrogen export to the coastal zone:€result from a spatially explicit, global model. Global Biogeochemical Cycles 19, GB4S02, doi:10.1029/2005 GB002488 Durand, P., Breuer, L. and Johnes, P. J. (2011). Nitrogen turnover processes and effects in aquatic ecosystems. In:€The European Nitrogen Assessment, ed. M., Sutton, C., Howard, J.W. Erisman, et╯al. Cambridge University Press.

Gilles Billen El-Habr, H. and Golterman, H. L. (1987). Input of nutrients and suspended matter into the Golfe du Lion by the river Rhone? Chemosphere, 16, 417–418. Garnier, J., Billen, G. and Coste, M. (1995). Seasonal succession of diatoms and chlorophyecae in the drainage network of the River Seine: observations and modelling. Limnology and Oceanography, 40,€750–765. Garnier, J., Billen, G., Hannon, E. et╯al. (2002). Modeling transfer and retention of nutrients in the drainage network of the Danube River. Estuarine and Coastal Shelf Science, 54, 285–308. Garnier, J., Cébron, A., Tallec, G. et╯al. (2006). Nitrogen behaviour and nitrous oxide emission in the tidal Seine River estuary (France) as influenced by human activities in the upstream watershed. Biogeochemistry, 77, 305–326. Garnier, J., Billen, G., Némery, J. and Sebilo, M. (2010). Transformations of nutrients (N, P, Si) in the turbidity maximum zone of the Seine estuary and export to the sea. Estuarine and Coastal Shelf Science (in press). Graneli, E., Wallström, K., Larsson, U., Graneli, W. and Elmgren, R. (1990). Nutrient limitation of primary production in the Baltic Sea area. Ambio, 19, 142–151. Grizzetti, B. and Bouraoui, F. (2006). Assessment of Nitrogen and Phosphorus Environmental Pressure at European Scale. European Commission, Joint Research Centre, Luxembourg. Grizzetti, B., Bouraoui, F., de Marsily, G. and Bidoglio, G. (2005). A statistical approach to estimate nitrogen sectorial contribution to total load. Water Science and Technology, 51, 83–90. Grizzetti, B., Bouraoui F. and De Marsily, G. (2008). Assessing nitrogen pressures on European surface water, Global Biogeochemical Cycles, doi:10.1029/2007GB003085. Grizzetti, B., Bouraoui, F., Billen, G. et╯al. (2011). Nitrogen as a threat to European water quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Hansson, M. (2008). Cyanobacterial blooms in the Baltic Sea. HELCOM Indicator Fact Sheets 2008. Available online:€http:// www.helcom.fi/environment2/ifs/ifs2008/ en_GB/ cyanobacteriaBlooms/ [Viewed 24 March 2009] HELCOM (2009). Eutrophication in the Baltic Sea: An Integrated Thematic Assessment of the Effects of Nutrient Enrichment in the Baltic Sea Region. HELCOM, Helsinki. Horstmann, U., Davidov, A., Cociasu, A. and Velikova, V. (2003). Der Einfluss verringerter Nährstofffrachten der Donau auf das Schwarze Meer. Österreichische Wasser- und Abfallwirtschaft, 55, 205–212. Howarth, R. W., Billen, G., Swaney, D. et╯al. (1996). Regional nitrogen budgets and riverine N and P fluxes for the drainages to the North Atlantic ocean:€natural and human influences. Biogeochemistry, 35, 75–139. Howarth, R. W., Boyer, E. W., Marino, R. et╯al. (2006). The influence of climate on average nitrogen export from large watersheds in the northeastern United States. Biogeochemistry, 79, 163–186. Humborg, C., Danielsson, A., Sjoberg, B. and Green, M. (2003). Nutrient land-sea fluxes in oligothrophic and pristine estuaries of the Gulf of Bothnia, Baltic Sea. Estuarine, Coastal and Shelf Science, 56, 781–793. Humborg, C., Smedberg, E., Blomqvist, S. et╯al. (2004). Nutrient variations in boreal and subarctic Swedish rivers:€landscape control of land-sea fluxes. Limnology and Oceanography, 49, 1871–1883. Humborg, C., Pastuszak, M., Aigars, J. et╯al. (2006). Decreased silica land-sea fluxes through damming in the Baltic Sea catchment: significance of particle trapping and hydrological alterations. Biogeochemistry, 77, 265–281.

Humborg, C., Smedberg, E., Rodriguez Medina, M., Mörth, C.-M. (2008). Changes in dissolved silicate loads to the Baltic Sea€– The effects of lakes and reservoirs Journal Marine Systems. 73:€223–235 Johnes, P. J. and Butterfield, D. (2002). Landscape, regional and global estimates of nitrogen flux from land to sea:€errors and uncertainties. Biogeochemistry, 57/58,€429–476. Justic, D. (1991). Hypoxic conditions in the northern Adriatic sea: historical development and ecological significance. Geological Society of London, Special Publications, 58, 95–105 Justic, D., Rabalais, N. N., Turner, R. E. and Dortch, Q. (1995). Changes in nutrient structure of river-dominated coastal waters, stoichiometric nutrient balance and its consequences. Estuarine and Coastal Shelf Science, 40, 339–345. Kelly, C. A., Rudd, J. M, Hesslein, R. H. et╯al. (1987). Prediction of biological acid neutralization in acid-sensitive lakes. Biogeochemistry, 3, 129–140. Kortelainen, P., Mattsson, T., Finer, L. et╯al. (2006). Controls on the export of C, N, P and Fe from undisturbed boreal catchments, Finland. Aquatic Sciences, 68, 453–468. Kronvang, B., Borgvang, S. A. and Barkved, L. J. (2009). Towards European harmonised procedures for quantification of nutrient losses from diffuse sources:€the EUROHARP project. Journal of Environmental Monitoring, 11, 503–505. Lancelot, C. (1995). The mucilage phenomenon in the continental coastal waters of the North Sea. Science of the Total Environment, 165, 83–102. Lancelot, C., Billen, G., Sournia, A. et╯al. (1987). Phaeocystis blooms and nutrient enrichment in the continental coastal zones of the North Sea. Ambio, 16, 38–46. Lancelot, C., Billen, G. and Barth, H. (1991). The Dynamics of Phaeocystis Blooms in Nutrient Enriched Coastal Zones, Water Pollution Research Reports Series No. 23, Environmental RD Programme , CEC, Brussels. Lancelot, C., Martin, J. M., Panin, N. and Zaitsev, Y. (2002). The North-Western Black-Sea:€a pilot site to understand the complex interaction between human activities and the coastal environment. Estuarine, Coastal and Shelf Science, 54, 279–283. Lancelot, C., Spitz, Y. and Gypens, N. et╯al. (2005). Modelling diatomPhaeocystis blooms and nutrient cycles in the Southern Bight of the North Sea:€the MIRO model. Marine Ecology Progress Series, 289, 63–78. Lancelot, C., Gypens, N., Billen, G., Garnier, J. and Roubeix, V. (2007). Testing an integrated river–ocean mathematical tool for linking marine eutrophication to land use:€The Phaeocystis-dominated Belgian coastal zone (Southern North Sea) over the past 50 years. Journal of Marine Systems, 64, 216–228. Law, C. S. Rees, A. P. and Owens N. J. P. (1992). Nitrous oxide:€estuarine sources and atmospheric flux. Estuarine, Coastal and Shelf Science, 35, 301–314. Leip, A. Achermann, B., Biller, G. et╯al. (2011). Integrating nitrogen fluxes at the European scale. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al.,€Cambridge University Press. Lepistö, A., Kenttämies, K. and Rekolainen, S. (2001) Modeling combined effects of forestry, agriculture and deposition on nitrogen export in a northern river basin in Finland. Ambio, 30, 338–348. Lepistö, A., Granlund, K., Kortelainen, P. and Räike, A. (2006). Nitrogen in river basins:€Sources, retention in the surface waters and peatlands, and fluxes to estuaries in Finland. Science of theTotal Environment, 365, 238–259. Leppäranta, M. and Myrberg, K. (2009). Physical Oceanography of the Baltic Sea. Springer, New York.

295

Nitrogen flows from European regional watersheds Ludwig, W., Dumont, E., Meybeck, M. and Heussner, S. (2009). River discharges of water and nutrients to the Mediterranean and Black Sea:€major drivers for ecosystem changes during past and future decades? Progress in Oceanography, 80, 199–217. Lundberg, C. (2005). Conceptualizing the Baltic Sea ecosystem an interdisciplinary tool for environmental decision making. Ambio, 34, 433–439. Marchetti, R. (1990). Algal blooms and gel production in the Adriatic sea. In: Barth, H., Fegan, L. (eds.) Eutrophication Related Phenomena in the Adriatic Sea and in other Mediterranean Coastal Zones, Water Pollution Research Reports Series No. 16, Enviromental RD Programme, CEC, Brussels, pp. 21–42. Mattsson, T., Finer, L., Kortelainen, P. and Sallantaus, T. (2003). Brook water quality and background leaching from unmanaged forested catchments in Finland. Water, Air and Soil Pollution, 147,€275–297. Mayorga, E., Seitzinger, S. P., Harrison, J. A. et al. (2010). Global Nutrient Export from WaterSheds 2 (NEWS 2):€model development and implementation. Environmental Modeling and Software, 25, 837–853. Mee, L. D. (1992). The Black Sea in crisis:€the need for concerted international action. Ambio, 21, 278–286. Meybeck, M., Cauwet, G., Dessery, S. et╯al. (1988). Nutrients (organic C, P, N, Si) in the eutrophic River Loire (France) and its estuary. Estuarine Coastal and Shelf Science, 27, 595–624. Meybeck, M. and Dürr, H. H. (2009). Cascading filters of river material from headwaters to regional seas:€the European example. In SCOPE 70, Chapter 7. Murata, A., Sandric Chee Yew Leonga, Yuji Nagashima, Y. and Taguchi, S. (2006). Nitrogen: phosphorus supply ratio may control the protein and total toxin of dinoflagellate Alexandrium tamarense. Toxicon, 48, 683–689. Neal, C. and Davies, H. (2003). Water quality fluxes for eastern UK rivers entering the North Sea:€a summary of information from the Land Ocean Interaction Study (LOIS). Science of the Total Environment, 314/316, 821–882. Neitsch, S. L., Arnold, J. G., Kiniry, J. R. and Williams, J. R. (2001). Soil and Water Assessment Tool-use Manual Version 2000. Blackland Research Center, Agricultural Research Service, Texas, USA. Neitsch, S. L., Arnold, J. G., Kiniry, J. R. and Williams, J. R. (2005). SWAT Theoretical Documentation. www:bcr.tamus.edu/swat/ Nielsen, K., Nielsen, L. P. and Rasmussen, P. (1995). Estuarine nitrogen retention independently estimated by the denitrification and mass balance methods:€a study of Norsminde Fjord, Denmark. Marine Ecology Progress Series, 119, 275–283. Nixon, S. W., Ammerman, J. and Atkinson, L. (1996). The fate of nitrogen and phosphorus at the land–sea margin of the North Atlantic Ocean. Biogeochemistry, 35, 141–180. Officer, C. B. and Ryther, J. H. (1980). The possible importance of silicon in marine eutrophication. Marine Ecology Progress. Series., 3, 383–391. OSPAR (2002). Data Report on the Comprehensive Study of Riverine Inputs and Direct Discharges (RID) in 2000. OSPAR Commission, London. Pastuszak, M., Witek, Z., Nagel, K., Wielgat, M. and Grelowski, A. (2005). Role of the Oder estuary (southern Baltic) in transformation of the riverine nutrient loads. Journal of Marine Systems, 57, 30–54. Peterson, B. J., Wollheim, W. M., Mulholland, P. J. et╯al. (2001). Control of nitrogen export from watersheds by headwater streams. Science, 292, 86–90. Pettine, M., Patrolecco, L., Camusso, P. and Crescenzio, S. (1998). Transport of carbon and nitrogen to the northern Adriatic Sea by the Po River. Estuarine, Coastal and Shelf Science, 46, 127–142.

296

Pitkänen, H. and Tamminen, T. (1995). Nitrogen and phosphorus as production limiting factors in the estuarine waters of the eastern Gulf of Finland. Marine Ecology Progress Series, 129, 283–294. Pitkänen, H., Lehtoranta, J. and Räike, A. (2001). Internal nutrient fluxes counteract decreases in external load:€the case of the estuarial eastern Gulf of Finland, Baltic Sea. Ambio, 30, 195–201. Preston, D. P. and Brakebill, J. W. (1999). Application of Spatially Referenced Regression Modeling for the Evaluation of Total Nitrogen Loading in the Chesapeake Bay Watershed, (WRI Report 99–4054. USGS Water Resources Division MD-DE-DC, Maryland). Radach, G. and Pätsch, J. (2007). Variability of continental riverine freshwater and nutrient inputs into the North Sea for the years 1977–2000 and its consequences for the assessment of eutrophication. Estuaries and Coasts, 30, 66–81. Rantakari, M. and Kortelainen, P. (2008). Controls of organic and inorganic carbon in randomly selected Boreal lakes in varied catchments. Biogeochemistry, 91, 151–162. Redfield, A. C., Ketchum, B. H., Richards, F. A. (1963). The influence of organisms on the composition of sea-water. In: The Sea, ed. M.€N.€Hill, John Wiley & Sons, New York, pp. 12–37. REGINE (2010). www.nve.no Ruelland, D., Billen, G., Brunstein, D. and Garnier, J. (2007). SENEQUE 3:€a GIS interface to the RIVERSTRAHLER model of the biogeochemical functioning of river systems. Science of the Total Environment, 375, 257–273. Sangiorgi, F. and Donders, T. H. (2004). Reconstructing 150 years of eutrophication in the north-western Adriatic Sea (Italy) using dinoflagellate cysts, pollen and spores. Estuarine, Coastal and Shelf Science, 60, 69–79. Schouwman, O. F., Silgram, M. P., Groenendijk, P. et╯al.(2003). Description of nine nutrient loss models:€capabilities and suitability based on their characteristics. Journal of Environmental Monitoring, 11, 506–514. Seitzinger, S. P., Mayorga, E., Kroeze, C. et╯al. (2009). Global river nutrient export trajectories 1970–2050:€a millennium ecosystem assessment scenario analysis. Global Biogeochemical Cycles, doi:10.1029/2009GB003587. Sferratore, A., Garnier, J., Billen, G., Conley, D. and Pinault, S. (2006). Silica diffuse and point sources in the Seine watershed. Environmental Science and Technology, 40, 6630–6635. Smith, R. A., Schwarz, G. E. and Alexander, R. B. (1997). Regional interpretation of water-quality monitoring data. Water Resources Research, 33, 2781–2798. Solidoro, C., Bastianini, M., Bandelj, V. et╯al.(2009). Current state, scales of variability, and trends of biogeochemical properties in the northern Adriatic Sea. Journal of Geophysical Research, 114, doi:10.1029/2008JC004838. Stalnacke, P., Grimvall, A., Libiseller, C., Laznik, M. and Kokorite, I. (2003). Trends in nutrient concentrations in Latvian rivers and the response to the dramatic change in agriculture. Journal of Hydrology, 283, 184–205. Tamminen, T. and Andersen, T. (2007). Seasonal phytoplankton nutrient limitation patterns as revealed by bioassays over Baltic Sea gradients of salinity and eutrophication. Marine Ecology Progress Series, 340, 121–138. Thieu, V., Billen, G. and Garnier, J. (2009). Nutrient transfer in 3 contrasting N-W European watersheds:€the Seine, Somme and Scheldt rivers. A comparative application of the Seneque/ Riverstrahler model. Water Resources Research, 43, 1740–1748. Thieu, V., Garnier, J. and Billen, G. (2010a). Assessing the effect of nutrient mitigation measures in the watersheds of the Southern Bight of the North Sea. Science of the Total Environment (in€press).

Gilles Billen Thieu, V., Billen, G., Garnier, J. and Benoît, M. (2010b). Nutrient cycling in a hypothetical scenario of generalized organic agriculture in the Seine, Somme and Scheldt watersheds. Regional Environmental Changes (in press). Trifu-Raducu, M.-C. (2002) Transfert des nutriments dans le bassin du Danube et apports à la Mer Noire:€modélisation et bilans. Thèse, Université P & M Curie. Paris Turner, R. E. and Rabalais, N. N. (1994). Evidence for coastal eutrophication near the Mississippi River Delta. Nature, 368, 619–621. Turner, R. E., Qureshi, N. A., Rabalais, N. N. et╯al. (1998). Fluctuating silicate:nitrate ratios and coastal plankton food webs. Proceedings of the National Academy of Sciences of the USA, 95, 13048–13051. Vasas, V., Lancelot, C., Rousseau, V. and Jordan, F. (2007). Eutrophication and overfishing in temperate nearshore pelagic food webs:€a network perspective. Marine Ecology Progress Series, 336, 1–14. Voss, M., Baker, A. and Bange, H. W. (2011). Nitrogen processes in coastal and marine systems. In:€The European Nitrogen Assessment,

ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Wade, A. J., Durand, P., Beaujouan, V. et╯al. (2002a). A nitrogen model for European catchments:€INCA, new model structure and equations. Hydrology and Earth System Sciences, 6, 559–682. Wade, A. J., Whitehead, P. G. and Butterfield, D. (2002b). The Integrated Catchments model of phosphorus dynamics (INCA-P), a new approach for multiple source assessment in heterogeneous river systems:€model structure and equations. Hydrology and Earth System Sciences, 6, 583–606. Whitehead, P. G., Wilson, E. J. and Butterfield, D. (1998). A semidistributed Nitrogen Model for Multiple Source Assessments in Catchments (INCA):€Part 1€– Model Structure and Process Equations. Science of the Total Environment, 210/211, 547–558. Windolf, J., Jeppesen, E., Jenseen, J. P. and Kristensen, P. (1996). Modelling of seasonal variation in nitrogen retention and in-lake concentration:€a four-year mass balance study in 16 shallow Danish lakes. Biogeochemistry, 33, 25–44.

297

Chapter

14

Atmospheric transport and deposition of reactive nitrogen in Europe Lead author: David Simpson Contributing authors: Wenche Aas, Jerzy Bartnicki, Haldis Berge, Albert Bleeker, Kees Cuvelier, Frank Dentener, Tony Dore, Jan Willem Erisman, Hilde Fagerli, Chris Flechard, Ole Hertel, Hans van Jaarsveld, Mike Jenkin, Martijn Schaap, Valiyaveetil Shamsudheen Semeena, Philippe Thunis, Robert Vautard and Massimo Vieno

Executive summary Nature of the problem • Observations of atmospheric reactive nitrogen (Nr) deposition are severely restricted in spatial extent and type. The chain of processes leading to atmospheric deposition emissions, atmospheric dispersion, chemical transformation and eventual loss from the atmosphere is extremely complex and therefore currently, observations can only address part of this chain.

Approaches • Modelling provides a way of estimating atmospheric transport and deposition of Nr at the European scale. A description of the different model types is provided. • Current deposition estimates from models are compared with observations from European air chemistry monitoring networks. • The main focus of the chapter is at the European scale; however, both local variability and and intercontinental Nr transfers are also addressed.

Key findings/state of knowledge • Atmospheric deposition is a major input of Nr for European terrestrial and freshwater ecosystems as well as coastal sea areas. • Models are key tools to integrate our understanding of atmospheric chemistry and transport, and are essential for estimating the spatial distribution of deposition, and to support the formulation of air pollution control strategies. • Our knowledge of the reliability of models for deposition estimates is, however, limited, since we have so few observational constraints on many key parameters. • Total Nr deposition estimates cannot be directly assessed because of a lack of measurements, especially of the Nr dry deposition component. Differences among European regional models can be significant, however, e.g. 30% in some areas, and substantially more than this for specific locations.

Major uncertainties/challenges • There are very few measurements of many of the key compounds (e.g. gaseous HNO3, coarse-nitrate, NH3), which are needed to enable comprehensive model evaluation. Data on all compounds should be available at the same site if the mass-balance of Nr is to be assessed, pointing to the need for integrated site measurements in air monitoring networks. • The main needs for oxidised Nr compounds are to evaluate how well the models capture the partitioning between gaseous HNO3 and either fine or coarse nitrate aerosol. For reduced Nr compounds, better estimates of NH3 emissions are needed, and how these are affected by meteorological factors as well as agricultural practices, coupled with an understanding of biosphere–atmosphere exchange. • Dry deposition of particles, sub-grid fluxes of NHx compounds, and effects of topography on wet deposition are especially difficult to parameterise properly.

Recommendations • There is a significant need for studies to constrain uncertain model parameters. This includes measurements of both the gas and particle phases of Nr compounds, and of atmosphere–biosphere fluxes of Nr compounds over sensitive ecosystems. • A balanced programme of observations and models is needed and is critical to future understanding of atmospheric transport and deposition of Nr containing pollutants at local to global scales. The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

298

David Simpson

14.1╇ Introduction This chapter attempts to answer the overriding question:€what are the atmospheric inputs of reactive nitrogen (Nr) in Europe, and how well can we estimate these? This issue is of particular concern for semi-natural ecosystems and sea areas, where the atmospheric supply of nitrogen can form an appreciable part of the total nitrogen load. As discussed elsewhere in this report, nitrogen measurements are of course essential for understanding the state of the atmosphere, and hence to help answer the first part of this question. However, as outlined by Hertel et€al., 2011 (Chapter 9, this volume), the chain of processes linking emissions, atmospheric dispersion, chemical transformation and loss from the atmosphere of Nr compounds is extremely complex. Observations can typically address only a small portion of this chain. In particular for this chapter, observations of atmospheric deposition are severely restricted in spatial extent and type. Typically only the ‘wet’ deposition of atmospheric nitrogen can be observed, and even this issue is fraught with uncertainty when we try to measure deposition to canopies. Nitrogen in the form of both ammonium and nitrate, together with most other plant nutrients, is strongly affected by canopy exchange (mainly uptake on the surface of the foliage), which affects throughfall composition. Unlike for S-species, N-species can be retained by the forest canopy, and throughfall is not a reliable indicator of total deposition. Another emerging and difficult field is that of organic nitrogen and its contribution to especially wet deposition (Cape et€ al., 2001; Gonzalez Benitez et€ al., 2009, 2010). (This issue is discussed by Hertel et€al., 2011 (Chapter€9, this volume), but not in this chapter as the sources of much of the measured organic nitrogen are still unclear. Further, the models presented in this chapter only consider oxidised organics, such as PAN, rather than reduced compounds.) The situation for dry deposition is even worse, with no routine method of measuring dry deposition. Some flux data are available from a limited number of sites employing microÂ�meteorological methods (Fowler et€al., 2001, 2009), but estimates of particle deposition rates are still very uncertain (Pryor et€al., 2008a,b,c). Measurements of the dry deposition of gaseous nitrogen species usually rely on the measurement of concentrations and estimates of deposition velocity (Zhang et€al., 2009). Given the lack of an observed deposition field over Europe, models are thus an essential tool for our understanding of the nitrogen cycle. A wide variety of models is available, but most aim to provide some or all of the following benefits. • To allow for spatially comprehensive estimates of pollutant concentrations, and for mapping of deposition patterns over large areas. • To integrate our understanding of atmospheric chemistry and transport. Models address emissions, dispersion and transport over multiple scales, chemical transformation, and dry and wet removal of pollutants. • To allow an exploration of the relative importance of different physical/chemical processes, in order to test hypotheses, and to focus attention on the most important mechanisms.

• To predict future pollution levels, including ‘what-if ’ scenarios in which different policy options are explored. • A comparison of model predictions against observed values is essential if scientists and policy makers are to have confidence that we understand the nitrogen cycle. Models are of course necessarily approximations to the real world, and they have to be evaluated thoroughly against Â�measurements if we are to have any confidence in their Â�abilities. For these reasons this chapter will focus mostly on models (in particular chemical transport models, CTMs) and their results, although with strong coupling to measurements. Section 14.2 will briefly present the types of models Â�typically used to assess atmospheric deposition, Section 14.3 will present deposition estimates from global to local scales. Model evaluation will be discussed in Section 14.4, and Section 14.5 will discuss the remaining uncertainty and challenges. The main focus of the chapter is the European regional scale, and deposition issues, but we will also present results covering scales from global to local (~1 km scale) in order to place the results in context.

14.2╇ Types of models A bewildering variety of models is available for air pollution studies, with applications ranging from near-source dispersion or process studies to global scale. For example, the European Topic Centre on Air and Climate Change model documentation system lists 123 different models, developed around the world (EIONET, 2010). The United States Environmental Protection Agency maintains a similar list (EPA, 2010). Recent reviews of different types of models and their applications can be found in Bleeker et€al. (2009), Hertel et€al. (2006), Holmes and Morawska (2006), Seinfeld and Pandis (1998), Sportisse (2007) and van Pul et€al. (2009), for example. The number of models partly reflects the difficulties of the task at hand, with models limited by basic theoretical principles as well as by practical problems. Difficulties arise from our limited understanding of many biological, meteorological and chemical processes, the difficulty of specifying many of the important inputs for modelling nitrogen exchange (e.g. NH+4 levels in vegetation, soil water, atmospheric emissions, surface properties), and the still-real problems of computer processing power. Thus, all models are compromises in which all aspects of the problem are simplified to some extent. The goal, and art, of modelling is to capture the most important processes for the problem at hand, so that the model is useful for its purpose, and can be relied upon to a reasonable extent. In this chapter we will discuss the main types of models typically used for problems related to oxidised or reduced nitrogen in relation to air quality issues. The main focus is Europe, but we also discuss applications from local scale to global scale. This section discusses the main types of models which are typically used to calculate nitrogen inputs to ecosystems or water surfaces, namely plume, Lagrangian or Eulerian models. For a discussion of other types of models (e.g.

299

Atmospheric transport and deposition of reactive nitrogen

computational fluid dynamics models, canyon models) see Hertel et€al. (2006).

Plume models Plume models are widely used in operational local-scale modelling of releases from industry and power plants. Examples of such models are the American AERMOD (Cimorelli et€al., 2005), the UKADMS (Carruthers et€al., 1994), the Dutch OPS (Duyzer et€al., 2001; van Pul et€al., 2004) and the Danish OML (Olesen et€ al., 1992). An inter-comparison study of plume models has shown a reasonably good agreement for most conditions (Olesen, 1995) and this type of model is in general suited for application in local-scale air pollution regulation, or for emissions verification. An example of a plume model applied specifically for NH3 deposition modelling is OMLDEP (Hertel et€ al., 2006). Further discussion of these and other models can be found in Hertel et€al. (2006) and Holmes and Morawska (2006).

Lagrangian models In Lagrangian models, an air parcel is tracked along a trajectory computed from wind speed and wind direction. Lagrangian models may use just one, or many, vertical layers. Where more than one layer is used, the approximation is usually made that all layers are transported with the same velocity, e.g. for ACDEP (Hertel et€ al., 1995) or FRAME (Singles et€ al., 1998; Fournier et€ al., 2005). Although global-scale models usually use the Eulerian framework, the UK STOCHEM model uses a Lagrangian formulation in which very many independent air parcels are followed and allowed to exchange material with each other (Collins et€al., 1997). Lagrangian models are typically computationally fast since they are usually applied to a restricted number of receptor points or air parcels, and with simplified treatment of meteorology and dispersion. In some models, this allows for a more advanced treatment of other aspects, e.g. of Â�chemistry€ – the UK Photochemical Trajectory Model makes use of the Master Chemical Mechanism (MCM), with MCM v3–1 treating about 13 500 reactions between 5900 species (Johnson et€ al., 2006; Jenkin et€al., 2003), or of detailed aerosol dynamics, e.g. UHMA (Korhonen et€al., 2004). Figure 14.1 provides a relevant example of the type of detailed atmospheric processing which can be analysed with MCM.

Eulerian models In Eulerian models, calculations are performed simultaneously for a grid of cells. For each of these grid-cells, advection, tur� bulent exchange, chemistry, and dry and wet deposition are computed. Examples of such models are the EMEP model (Berge and Jakobsen, 1998; Simpson et€al., 2003), CHIMERE (Bessagnet et€al., 2004), LOTOS (Schaap et€al., 2008), MATCH (Robertson et€al., 1999), RADM (Chang et€al., 1987), CMAQ (Binkowski and Roselle, 2003), STEM (Carmichael et€al., 1991), and DEHM-REGINA (Frohn et€al., 2001). Eulerian models are generally more computer-resource demanding than plume and Lagrangian models, especially when a high geographical resolution is desired. However, such

300

Figure 14.1 Example of the use of a Lagrangian box model for chemical simulation. These calculations used a pseudo-Lagrangian boundary layer box model to represent initial passage of an air mass over an urban area (three hours duration), during which time the box received enhanced emissions. Subsequently, the box received background emissions, which were based on the UK average. Discussed in more detail in Jenkin et€al. (2006), these calculations demonstrated particularly important contributions from organic nitrates and PANs, a conclusion that is in broad agreement with observations.

models are generally recognised to provide the most comprehensive framework for chemical transport models (Sportisse, 2007; Seinfeld and Pandis, 1998). Unlike plume or Lagrangian models, Eulerian modelling involves calculations for the full spatial domain, and the structure allows for straightforward inclusion of complex meteorology and multiple, chemically interacting sources. Such models have been developed with simple one-way nesting (Kessler et€al., 2001; Vieno et€al., 2009, 2010) and more accurate and advanced models with two-way nesting also exist (Frohn et€al., 2001).

Inferential models The last type of model mentioned here, inferential modelling, is very different, in that no chemical transport modelling is done at all. Where flux measurements are not available (i.e. at the vast majority of sites), an estimate of dry deposition fluxes may still be made by combining measured concentration data with relevant (micro)meteorological data and estimated deposition velocities. Inferential modelling is important for its ability to provide a deposition estimate which is heavily observation-based, and as a framework for evaluating differences between model formulations for deposition velocity. This method has been applied within e.g. the US-CASTNET network (Clarke et€ al., 1997) and the EU-NitroEurope network (Sutton et€al., 2007; Flechard et€al., 2010). See also Table 14.1.

14.3╇ Atmospheric deposition of reactive nitrogen This section gives an overview of modelling results concern�ing the deposition of reactive nitrogen to land and sea areas. Scales ranging from global to local are covered, but most emphasis is given to the European scale. Unfortunately, observations cannot provide maps of total Nr deposition, as typically only wet deposition can be measured, and then at spatially heterogeneous

David Simpson Table 14.1 Advantages and disadvantages of common types of chemical transport models (adapted from Hertel et€al., 2006)

Advantages

Disadvantages

Scale T

Highly simplified formulation. Difficulties with complex meteorology, chemistry. Cannot account for interactions between sources.

0–20 km

Short-falls in the description of transport and dispersion. The uncertainty increases with distance along the trajectory. Forward trajectory models can only handle simplified chemistry. Computationally demanding for a large number of receptor points.

1–500 km

Generally computationally demanding€– especially for three dimensional models with high resolution, e.g. including nesting techniques. Difficulties in handling plumes.

10 km–global

Plume Fast, analytic solutions, easy to apply.

Lagrangian Fast for carrying out multiple model runs that concern a limited number of receptor points. Generally easy to apply for most purposes. Eulerian Allows comprehensive description of combined transport, dispersion and chemical modelling. Enables high-resolution three-dimensional simulations, treatment of complex terrain, and one or two-way nesting.

Notes:€There are examples of all model types at essentially all scales, but we give here the main domain of application of the different types.

networks, so this section focuses on model results. However, Section 14.4 will present further data on observed wet deposition in the context of model evaluation studies, and discuss some of the uncertainties surrounding these estimates.

14.3.1╇ Atmospheric deposition:€global scale Global emissions of NO, NH3 and SO2 may have increased by more than a factor of three since the pre-industrial era. Regionally, these increases have been even more substantial, and emissions from large portions of North America, Europe and Asia increased by more than a factor of ten during the past century (van Aardenne et€al., 2001). Recent studies (Galloway et€al., 2004) indicate substantial further increases of emissions and deposition toward 2050. Other scenario studies suggest that increasing air pollution control will stabilise or reduce emissions by 2030 (Cofala et€ al., 2007). The need to understand and predict such changes has led to a flurry of activity on global-scale modelling in recent years, further promoted by the establishment of the UNECE Task Force on Hemispheric Transport of Air Pollution (HTAP, 2010). An extensive recent study of global N-deposition is that of Dentener et€al. (2006). This study focused on global and regional deposition fluxes of both oxidised and reduced nitrogen compounds for the present day and near future (2030), using an ensemble of 23 models. This study showed reasonable agreement with observations in Europe and North America, where 60%–70% of the model-calculated wet deposition rates agree to within ±50% with quality-controlled measurements (Dentener et€al., 2006). The same models systematically overestimate NHx deposition in South Asia, and underestimate NOy deposition in East Asia (Figure 14.2). These questions were addressed in several multi-model studies of nitrogen deposition. Recently the UNECE Task Force on Hemispheric Transport of Air Pollution evaluated the hemispheric transport of ozone, aerosol and precursors between

four world regions. As discussed in Sanderson et€al. (2008) (see also Erisman et€al., 2011, Chapter 2, Figure 2.10, this volume), Europe substantially impacts parts of Asia and North America, and, vice versa, Europe is mostly influenced by emissions from North America. A few percent of NOy emissions from North America reach Europe. The TF HTAP interim report states that on average 75% of the NOx emissions in Europe are deposited within Europe, with small fractions falling on North America€ (1%), South Asia (2%), East Asia (2.5%), and the remainder deposited in the oceans, and Russia.

14.3.2╇ Atmospheric deposition over Europe In this section we focus on modelling results for the so-called regional scale models, those which are designed to run over large areas of Europe, with grid sizes of typically 30–50 km. One important model in this context is the EMEP model (Berge and Jakobsen, 1998; Simpson et€al., 2003, 2006a), as it is widely employed within the European air pollution abatement strategy and legislation work (Sliggers and Kakebeeke, 2004). This model is typically run with a 50 × 50 km2 grid size, although first results for 10 × 10 km2 are now available (Fagerli et€al., 2008). Other models typically applied at this scale include CHIMERE (Bessagnet et€ al., 2004), LOTOS (Schaap et€ al., 2008), MATCH (Robertson et€al., 1999), and DEHM-REGINA (Frohn et€ al., 2001). Intercomparison of some or all of these models was presented in van Loon et€al. (2004, 2007), Vautard et€al. (2007) and Stern et€al. (2008). We will begin by illustrating the results from a so-called ensemble of chemistry transport models, which includes most of those mentioned above. Using an ensemble of models rather than a single model to simulate air quality for assessment or emission scenario evaluation purposes provides two new pieces of information. Firstly, the average (or the median) over this ensemble is a new result by itself, which is expected to have a smaller error because individual model errors cancel each

301

Atmospheric transport and deposition of reactive nitrogen

Figure 14.2 Annual wet deposition of (A) NO –3 (HNO3 and aerosol nitrate), (B) NHx (NH3 and aerosol ammonium), for current-year (~€year€2000) simulation along with measurements grouped in 5° latitude and 10° longitude. The numbers within the circles indicate the number of stations in this latitude/longitude band. Units mg(N) m−2 (100€mg(N)€m−2€= 1 kg(N) ha−2). From Dentener et€al. (2006).

other to a certain extent. Secondly, the spread of the ensemble can be a measure of the uncertainty in model simulations. In the EURODELTA study (van Loon et€al., 2007; Schaap et€al., 2010; Vautard et€al., 2008) seven modelling teams simulated the air quality over the European domain for the full year of 2001 using a harmonised emission database. Figure 14.3 �illustrates the total Nitrogen deposition obtained from the ensemble-mean, along with the standard deviation of these results. Firstly, these results demonstrate the strong spatial variation in nitrogen deposition, with clear maxima over the Benelux area and Po Valley region of Northern Italy. The standard deviation of model results is large, however, e.g. representing about 30% of the mean value over the Netherlands. Maps of just the wet-deposition component of one model, EMEP, will be presented and compared with observations in Section 14.4.3.

Deposition to ecosystems Figure 14.3 presented total N deposition to model grids, but for assessing the vulnerability of ecosystems to deposition one

302

needs to know the deposition rates to each type of land-cover within the grids. Importantly, deposition loads to forests are typically greater than to other ecosystems, enhanced by their greater aerodynamic roughness, and their ability to capture fine-particles (Ruijgrok et€al., 1997; Pryor et€al., 2008b). Here we present some examples calculated with the EMEP model, as this model utilises a so-called mosaic approach, in which deposition rates are calculated for up to 18 different landcover types per grid (Simpson et€al., 2001, 2003). Calculated deposition to two important ecosystems are illustrated in Figure 14.4, with deposition given per unit area of ecosystem. This figure clearly illustrates the large gradients of N-deposition across Europe, and that areas in north-west Europe receive the highest loadings of N-deposition. Deposition to forests is significantly higher than to semi-�natural areas. Similar calculations for croplands show even lower deposition rates than to semi-natural, partly due to the fact that crop lands are only vegetated for part of the year. In order to further illustrate the sources of this nitrogen deposition, Figure 14.5 shows the calculated relative

David Simpson

Figure 14.3 Calculated nitrogen deposition from an ensemble of seven models for 2001, together with the standard deviation (right) of the model estimates. Units:€mg(N)m−2 (100mg(N)m−2 = 1â•›kg(N)ha−2) (EURODELTA study, see text). (a)

Sum CF

(b)

100

Sum CF

100 3000

3000 2000

80

2000

80

1500

1500 1100

60

1100

60

900

900 700

40

700

40

500

500 300

20

300

20

100

100

40

60

80

100

120

40

60

80

100

120

Figure 14.4 Calculated N-deposition densities (mg(N) per m2 of ecosystem per year) to different ecosystems (year 2000):€(a) coniferous forest, (b) semi-natural. Source:€EMEP MSC-W.

contributions of dry and wet deposition, for oxidised and reduced nitrogen, to the total reactive nitrogen deposition to forests. In the Nordic countries, dry and wet deposition of oxidised nitrogen dominate, although wet deposition of reduced nitrogen accounts for around 20%–30% of the total in forest ecosystems, and somewhat more for non-forest ecosystems. Dry deposition of reduced-N is, however, the most significant contributor in many areas of central Europe, including parts of France, UK, Ireland and the Netherlands. Over southern Europe dry deposition tends to dominate over wet, as should be expected given the lower precipitation rates.

Deposition to European seas The atmospheric input of Nr to sea is significant. It has been estimated that approximately one quarter of the total nitrogen

input to the Baltic Sea comes from airborne nitrogen deposited directly into the sea (HELCOM, 2005) and around 30% for the North Sea (Rendell et al., 1993). A number of modelling studies have examined deposition to sea areas, with the Baltic and North Sea receiving most attention (Bartnicki and Fagerli, 2008; Hertel et€al., 2002, 2003; Langner et€ al., 2009; de Leeuw et€ al., 2001, 2003; Schlunzen and Meyer, 2007). For example, Hertel et€al. (2002) estimated around 40% of the nitrogen deposition over the North Sea to originate from agriculture activities and around 60% from emissions from combustion sources. As seen in Figure 14.6, wet deposition dominates over the dry deposition of nitrogen for three of the four sea areas. The dominance of wet deposition was also found by de Leeuw et€al. (2003) and Hertel et€al. (2002) for the North Sea (more than 80%). This dominance is expected

303

Atmospheric transport and deposition of reactive nitrogen (a)

(b)

100

50

30

80

100

50

30

80

20 60

15

10

40

20 60

15

10

40

5 20

5 20

80

100

120

140

80

(c)

100

120

140

(d)

100

50 30

80

100

50 30

80

20

20 60

15 10

40

60 15 10

40

5 20

5 20

80

100

120

140

80

100

120

140

Figure 14.5 Calculated percentage contributions to total nitrogen deposition over coniferous forest:€(a) dry deposition of oxidised N; (b) dry deposition of reduced N; (c) wet deposition of oxidised N; (d) wet deposition of reduced N. Calculations for the year 2000, from Simpson et€al. (2006a).

since compounds such as NO2 and PAN have low deposition rates to water surfaces. Further, the sea surface is usually aerodynamically smooth compared to land (especially forest), and so dry deposition of even soluble compounds is relatively less important over sea than land. In general, nitrogen deposition originating from emissions on land have a strong gradient towards the sea. Ammonia is efficiently dry deposited close to the source areas and most of the reduced nitrogen that reaches the open sea comes in the form of ammonium particles which are efficiently wet deposited. NOx deposition has a somewhat weaker gradient, reflecting a longer residence time in the atmosphere (NO and NO2 do not deposit efficiently, but are transformed to HNO3 which is efficiently dry deposited or forms nitrate aerosols.) Furthermore,

304

slower deposition processes of aerosols over water surfaces are assumed in the model. The Mediterranean Sea is also different with respect to the share of agriculture related nitrogen deposition and deposition originating from emissions from combustion sources. Whilst the other seas have similar contributions of oxidised and reduced nitrogen deposition, the share of oxidised nitrogen deposition is more than 70% for the Mediterranean Sea, owing largely to the large contribution from ship traffic emissions. In fact, for three of the four European seas discussed here, ship traffic emissions are among the most important contributor to oxidised nitrogen deposition to the sea area (Table 14.2). The exception is the Black Sea where emissions from Russia,

David Simpson (a)

Table 14.2 Contribution from international ship traffic emissions to oxidised nitrogen deposition in European seas (%). From Bartnicki and Fagerli (2008)

Receptor

Ship contribution

Baltic Sea

22%

North Sea

17%

Mediterrenean Sea

34%

Black Sea

(b)

7%

Turkey and Ukraine contribute around 20% each, with contribution from international ship traffic of around 7%. The main contributors to reduced nitrogen deposition are in general countries along the coast lines.

14.3.3╇ The local scale and scaling issues

(c)

(d)

Figure 14.6 Time series of annual atmospheric load (Gg N/yr) of nitrogen to the European seas in the period 1995–2005. Oxidised and reduced dry and wet deposition and total nitrogen deposition are shown. From Bartnicki and Fagerli (2008).

As noted above, the grid resolution of models varies from typically less than 1 km in the most detailed local modelling to around 1° (c. 100 km) or larger for global scale models. Grid resolution affects not only the detail of model outputs, but has profound effects on the treatment of non-linear processes. Affected processes include for example the rate of oxidation of NOx in plumes, and subsequent partitioning of NOy into either rapidly depositing HNO3 or longer-lived aerosol nitrate particles, or the bi-directional exchange of both oxidised and reduced nitrogen, where a mosaic of regions with high and low concentrations may well have a different net exchange to that found in a calculation where all concentrations are smeared out over a grid square. Many of the scaling problems associated with especially NH3 modelling have been addressed in a series of recent reviews and so are not covered in detail here€ – the reader is referred to Bleeker et€al. (2009), Hertel et€al. (2006), Holmes and Morawska (2006), Loubet et€al. (2009) and van Pul et€al. (2009). Here we will concentrate on the comparability of regional and local scale models€– and on the issues associated with bridging these scales. It was noted in Loubet et€al. (2009) (and refs cited therein) that the combination of hot-spot sources and effective deposition processes lead to sources and sinks of NHx being spatially heterogeneous at a scale of a square kilometer or less. Direct measurement of NHx deposition near hot spots is challenging due to intense local advection, and indirect estimates using mass balance, 15N labelling, SF6 to NH3 ratio methods, as well as modelling studies, have estimated that the fraction recaptured within 2 km downwind from the source of NH3 emitted ranges between 2% and 60% (Asman, 1998; Loubet and Cellier, 2001; Sommer and Jensen, 1991; Theobald et€al., 2001; Loubet et€al., 2006). As another example, field studies in the Netherlands (Asman et€ al., 1988) and the UK (Fowler et€ al., 1998) show that individual sources lead to a large downwind gradient in concentration and deposition. Figure 14.7 shows an example of a farm scale emission and deposition gradient from 28 to 2â•›g m−3 and 40 to 5 kgNâ•›ha−1â•›yr−1, respectively, within a distance of about 300 m.

305

Atmospheric transport and deposition of reactive nitrogen

4800 kg NH3-N y –1

Distance in m

Poultry Unit

15 50 42 16 28.9 9.5

76 12

126 8

270

5

Deposition kg ha–1

6.5 3.9 1.6

Concentration NH3 µgm–3 Sum of deposition within 270m of farm woodland is: 155 kg N y –1 (3.2% of emissions)

Figure 14.7 Farm scale NH3 emission and deposition, illustrating the rapid fall-off in deposition levels with distance from source (adapted from Fowler et€al., 1998).

Figure 14.8 Annual mean cumulative deposition of NHx species as a function of downwind distance, calculated with the OPS model.

Figure 14.8 illustrates the cumulative deposition of ammonia and ammonium plotted against the distance downwind of a source (e.g. an animal house) as calculated by the OPS model (van Jaarsveld, 2004). Owing to the high dry deposition velocity of ammonia and the relatively low release height the loss

of material is substantial in the first kilometres. Almost 20% is already deposited after 1 km transport and 50% after 50 km. Indeed, an important aspect of ammonia is that local deposition is almost fully determined by dry deposition of NH3. After approx. 50 km wet deposition of NH+4 becomes the dominant deposition form. Modelling the transport and deposition of ammonia, therefore, requires relatively high resolutions, both in the horizontal and vertical dimension. As of today, no single model is capable of reproducing a sufficiently wide range of length and time scales. Practical solutions include the (dynamical) nesting of small scale models into large-scale models or the use of output from large scale models to provide the boundary conditions for small scale models. The development of Eulerian models with flexible �resolution allows a systematic assessment of the effects of scale on model predictions. For example, Figure 14.9 illustrates the effect of increasing resolution on modelled deposition of reduced nitrogen using the same model (EMEP) at both 50 km resolution and 5 km resolution (see Vieno et€ al., 2009). The increased resolution affects both the detail of the simulation, but also the �location of the deposition. Deposition over hillsides more closely reflects the patterns of precipitation in the United Kingdom, and thus becomes more comparable to the results obtained by the UK CBED methodology (Smith et€al., 2000). This improvement partly reflects improved modelling of dispersion, but also partly improved meteorological modelling.

14.4╇ Comparison with observations Although models are essential for mapping deposition, their trustworthiness can only be assessed by comparison with measurements. Unfortunately, as noted in Section 14.1, observations are lacking for many important aspects of the deposition process, so comprehensive evaluations are impossible. However, even routine measurements of parameters such as air concentrations, or concentrations in deposition, give valuable information.

Figure 14.9 Calculated dry deposition of reduced nitrogen for UK, calculated from EMEP model runs at two resolutions:€50 km (left) and 5 km (right). Units mg(N) m−2 (100 mg(N) m−2 = lkg(N) ha−1). Source:€EMEP4UK model (Vieno et€al., 2009a,b).

306

David Simpson

It should also be noted that global and regional scale models cannot be expected to reproduce small-scale variations in deposition regimes, caused by such factors as local emissions (especially important for NH3 close to agricultural sources, see Sutton et€al., 1998), topography (which has strong effects on rainfall amount and deposition, see Dore et€ al., 1992; Fowler et€al., 1988; Hertel et€al., 2011, Chapter 9, this �volume), or where processes not included in the model (e.g. occult deposition) are important. These problems are difficult to address, but by comparison with measurements we can make an assessment of the degree of agreement between the model and observed values. Here we focus on the evaluation of European-scale deposition estimates, but start with a brief introduction to the EMEP measurement network (Section 14.4.1) and an evaluation of the air concentrations (Section 14.4.2). Reliable modelling of gas and aerosol air concentrations is a necessary (but not sufficient) prerequisite for reliable modelling of atmospheric inputs to ecosystems and seas.

(a)

(b)

NO3

TNH4

14.4.1╇ The EMEP network The main measurement network providing European-scale data on reactive nitrogen concentrations and deposition is the EMEP network (EMEP, 2010). As discussed in detail in Fagerli and Aas (2008), 24 EMEP sites have reported nitrate and ammonium in precipitation from around 1980, with a good coverage of North and Central Europe and partly Eastern Europe. The measurements of these compounds in air did not start until the end of the 1980s. Nineteen sites reported long-term data series, but the majority of these sites were located in Nordic countries. In general, few long-term measurements are available from the south-east of Europe. Nearly all of the air measurements conducted within the EMEP network are made using the filter pack method. It is well known that this method is biased for separate gas and particulate nitrogen compounds (EMEP, 1996). Ammonium nitrate on the aerosol filter may dissociate into gaseous nitric acid and ammonia that will be captured by the impregnated filters in the filter pack sampler. This causes a negative interference on the particle filter and positive interferences on the impregnated filters. The opposite may happen if ammonia or nitric acid is captured on the front aerosol filter. An artifact free separation of these gases and particles can be achieved using denuders, but only two EMEP sites had used this method at the time of the Fagerli and Aas (2008) study. EMEP has an extensive quality control of the data that are included in the database. Laboratories that fail badly in fieldand lab inter-comparisons (Aas and Hjellbrekke, 2005) are flagged. The data sets are graded according to their quality, and for model evaluation it is clearly best to use the data with the best quality (Fagerli and Aas, 2008).

14.4.2╇ Air concentrations In this section we illustrate the performance of chemistry transport models for reactive nitrogen using the ensemble of models introduced in Section 14.3.2. Figure 14.10 compares the

Figure 14.10 Modelled and measured seasonal variation of particulate nitrate and ammonium for 2001. The data represent the average monthly mean values for Ns stations over Europe. Number of stations (Ns) is indicated in Table 14.3. Units:€μg m−3. From Schaap et€al. (2010).

mean seasonal variation of the ensemble mean model and its members to the observed variation for particulate nitrate and ammonium from the EMEP network. The spread of the models is a measure of gaps and uncertainty in our knowledge. For example, Figure 14.10 indicates a higher uncertainty for nitrate than for ammonium. These EURODELTA results indicate that in general, the models are able to capture the seasonal variation of the single components, but with significant uncertainty. Table 14.3 shows the relative root mean square error (RRMSE) and mean correlation of concentrations found from this comparison. RRMSE values are higher for the nitrogen components than for sulphate, with largest values seen for total nitrate (TNO3). This latter finding is partly an artefact though, related to the number and distribution of the measurement sites over Europe for different compounds. Nitrate (only six sites here) is measured in north-western and central Europe, at sites characterised by flat-terrain, continental meteorology and high pollution levels. By contrast, total nitrate (21 sites here) is mostly measured in less polluted areas near the sea and/or in areas with complex terrain. Low pollution levels and complex terrain are generally associated with lower model skills. All models show this characteristic for all species. Indeed, in north western and central Europe RMSE values for TNO3 tend to be lower than for NO3, reflecting the sensitivity of the nitrate partitioning to ambient conditions and precursor gas concentrations.

307

Atmospheric transport and deposition of reactive nitrogen Table 14.3 Comparison of modelled and observed inorganic species for seven CTMs and their ensemble mean. Tables gives RRMSE value (%) and (in parentheses) temporal correlation coefficients between daily modelled and measured values, averaged over a number of stations (Ns) in Europe

Model

SO4

SO2

NO3

TNO3

NH4

NHx

Ns

36

27

6

21

╇ 8

19

EMEPv3.1

54 (0.61)

101 (0.52)

69 (0.59)

LOTOS-EUROS

51 (0.54)

85 (0.51)

62 (0.55)

106 (0.50)

50 (0.61)

62 (0.49)

87 (0.44)

52 (0.53)

56 (0.35)

MATCH

62 (0.62)

99 (0.57)

52 (0.56)

88 (0.51)

44 (0.61)

50 (0.57)

CHIMERE

59 (0.45)

139 (0.52)

62 (0.53)

84 (0.37)

46 (0.54)

62 (0.41)

RCG DEHM

57 (0.55)

107 (0.43)

53 (0.62)

75 (0.43)

48 (0.55)

74 (0.38)

68 (0.55)

83 (0.51)

71 (0.38)

160 (0.43)

49 (0.54)

54 (0.49)

TM5

59 (0.50)

n.a

97 (0.59)

143 (0.50)

62 (0.57)

67 (0.43)

Ensemble

44 (0.68)

91 (0.58)

46 (0.66)

╇ 92 (0.56)

40 (0.66)

50 (0.54)

Notes:€RRMSE is the relative root-mean square error, i.e. RMSE divided by the observations and in %; Ns is the number of stations; SO4 is particulate sulphate; NO3 and NH4 are particulate NO–3 and NH+4↜; TNO3 is the sum of HNO3+NO–3; NHx is the sum of NH3+ particulate NH+4.

The lower RRMSE values for NH3 and NHx compared to the oxidised compounds reflect the fact that the majority of the ammonium is bound to sulphate. The skill of the ensemble average is generally higher than the skill of the individual models. The better skill of the ensemble average or median has been shown earlier in several recent studies for air quality (Delle Monache and Stull, 2003; Pagowski et€al., 2005; McKeen et€al., 2007; van Loon et€al., 2007; Vautard et€al., 2008; Schaap et€al., 2010) as well as for transport of passive tracers (Galmarini et€al., 2004; Riccio et€al., 2007). In climate assessments (IPCC, 2007) for example, model ensembles has become essential to evaluate the state of the knowledge of the scientific community and the spread of its uncertainty.

14.4.3╇ Wet deposition Comparison of model results for wet deposition or concentrations in precipitation is in many ways trickier than comparing gas concentrations. Hertel et€al., 2011 (Chapter 9, this volume) discusses the physical/chemical processes controlling wet-deposition of Nr compounds, and the important role that topography can play in enhancing deposition rates. As noted in van Loon et€al. (2004), the most important issue concerning the wet removal of species in CTM models is probably the meteorological input; model performance for wet deposition fluxes or concentrations in precipitation is strongly limited by the quality of the NWP models providing meteorological data. For example, models generally have problems with sub-grid precipitation, simulating precipitation more often, but in lower amounts, than reality. As precipitation scavenging is a complex and non-linear process (Barrie, 1992), such issues will cause errors in modelled wet deposition that are difficult to evaluate. There are also many uncertainties inherent in the deposition monitoring methods themselves (Draaijers and Erisman, 1993; Erisman et€al., 2005). The precipitation amount may vary quite a lot over short distances, especially in mountain areas, and the sites are not always representative for the average gridded precipitation amount. For the EMEP network, the agreement between

308

precipitation measured at EMEP sites and the EMEP model is within 30% at almost all sites. Some of this discrepancy is of course due to uncertainties with the NWP model, but some is also due to precipitation sampling problems. In an early intercomparison of six different CTMs used in Europe, van Loon et€al. (2004) found very poor model performance for the wet deposited components, despite fair to good performance for airborne components. A clear result of this study was that no model achieved good correlation coefficients (the best was just r = 0.35) for wet components, and bias and RMSE values could be very substantial (up to 60%–70% for wet deposition fluxes) relative to observed values. These results were much worse than equivalent results for concentrations in air. The models used in this study have been improved to some extent since this intercomparison, but it seems likely that a study using today’s models would still show discrepancies of up to 50%. The relatively poor agreement between modelled and observed wet deposition fluxes is not a specific feature of this inter-comparison or these models. Large differences between models were also found in the global models participating in the COSAM study, in which the wet deposition efficiency ranged over a factor of 4 (Roelofs et€al., 2001). A similar spread was also found for global models by Dentener et€al. (2006) and Textor et€al. (2006). The EMEP model seems to have been subject to most Â�evaluation against observed wet deposition estimates. Standard scatter plots showing the performance of the model against observed concentrations of NO–3 and NH+4 can be found in the yearly EMEP status reports; see Fagerli and Hjellbrekke (2008) and Berge and Hjellbrekke (2010). The model has also been compared to observed wet deposition for nitrogen from the ICP-forest network (Simpson et€al., 2006b). Differences in mean values between modelled and observed (ICP-forest) SO2â•›–4, NO–3 and NH+4 total and wet deposition were within 20% in 1997 and 30% in 2000, with the EMEP model showing slightly lower values than the observations (Simpson et€al., 2006b). Modelled and observed concentrations of SO2â•›–4, NO–3 and NH+4 in precipitation were very similar on average (differences of 0%–14%),

David Simpson Figure 14.11 Comparison of modelled and observed annual wet deposition of (a) NHx and (b) NO –3 (HNO3 and aerosol nitrate). Data are for 2001 in the EMEP model with observations. The bullets depict observations with the same colour bar as the modelled field. Measured annual deposition is calculated by using the measured precipitation amount and the nitrate and ammonium concentration in precipitation.

(a) 100 90 80 70 60

1100 1000

50

900 800 700

40

600

30

500

20

400

10

200

300 100 20

30

40

50

60

70

80

90

100

110

120

130

(b) 100 90 80 70 60

1100 1000

50

900 800 700

40

600

30

500

20

400

10

200

300 100 20

30

40

50

60

70

80

90

100

110

120

and the correlation between modelled and observed data is rather high for this type of comparison (between r2 = 0.4–0.8 for most components and years). Figure 14.11 compares measured wet deposition of oxidised and reduced nitrogen against results from the EMEP model. In these plots the measured deposition is calculated using the measured precipitation amount and the nitrate and ammonium concentration in precipitation. For reduced nitrogen, Figure 14.11 a reveals good agreement between modelled and

130

measured values, across almost all of Europe. The high modelled values near northern Italy are reflected in the measurements. Unfortunately, other regions with high predicted wet deposition have only a limited number of measurement sites (e.g. Netherlands, Belgium), and so it is difficult to evaluate model performance here. The EMEP model has a tendency to under-predict wet deposition in Nordic sites. For oxidised nitrogen (Figure 14.11b), five sites stand out with much higher measured wet deposition than modelled.

309

Atmospheric transport and deposition of reactive nitrogen Table 14.4 Comparison of observed and modelled (EMEP) contributions (%) of dry and wet deposition of oxidised (OXN) and reduced nitrogen (RDN) to total N-deposition (OXN+RDN) at Speulderbos forest, Netherlands, 1995. From Simpson et€al. (2006a)

Observed Dry+Wet

Modelled Dry+Wet

Dry

Wet

Dry+Wet

Dry

Wet

OXN

18

11

29

22

╇ 9

31

RDN

47

24

71

54

15

69

The reason for this seems to be that the observed precipitation at the sites far exceeds the modelled precipitation (e.g. by a factor of two for the Norwegian site). However, there is a very good agreement between model results and measurements at almost all other sites, which gives some confidence that the modelled budget of wet-deposition is within the uncertainty of the measured value.

14.4.4╇ Dry deposition Although wet deposition represents an important fraction of N-deposition over Europe, dry deposition is also important. Hertel et€al. (2011), (Chapter 9 this volume) discusses the physical/chemical processes controlling dry-deposition of Nr compounds. Unfortunately, the EMEP network has no specific measurements of dry deposition, so we cannot present maps of modelled versus observed dry deposition. However, dry deposition monitoring has been performed over many years at Speulderbos forest in the Netherlands (Erisman et€ al., 1997, 2001), the site with by far the highest deposition loads within the EU NOFRETETE project (Pilegaard et€ al., 2005). Erisman et€ al. (2001) presented estimates of wet and dry deposition of oxidised and reduced nitrogen for Speulderbos, over the period 1995–1998, and Simpson et€al. (2006a) compared EMEP model estimates against these. The modelled total deposition for the Speulderbos grid square in 1995 was 5200 mg(N) m−2, within 10% of that found in the measurements (4798 mg(N)/m−2). Table 14.4 illustrates the percentage breakdowns of this total deposition between wet/ dry/OXN/RDN components. These relative contributions are remarkably similar, with reduced nitrogen accounting for about 2/3 of total deposition, and dry deposition dominating both the oxidised and reduced-N contributions. Ongoing studies within NitroEurope (Sutton et€al., 2007) suggest problems with dry-deposition estimates, however. As part of the EU Nitro-Europe project, inferential modelling is being conducted with deposition codes from three European dry deposition models at selected sites of the NitroEurope (NEU) inferential network (Flechard et€ al., 2010). The Â�deposition modules are from the UK-CBED model (Smith et€ al., 2000), the Dutch IDEM model (Bleeker et€ al., 2004) and the EMEP scheme (Simpson et€ al., 2001, 2003). This study has suggested that NH3 is the single highest atmospheric Nr dry input in many parts of Europe. At sub-urban sites of the NEU network, HNO3 and particulate NO3− and NH+ could also contribute significant fractions of total dry deposition. There were, however, substantial discrepancies between models, with annual deposition rates varying as

310

Dry+Wet

much as two-fold between models at given monitoring sites. This highlights the variability in model parameterisations, stemming from the variability in measured deposition rates and canopy resistances. For NH3, the stomatal compensation point and the external leaf-surface (or non-stomatal) resistance are the largest sources of divergence between models. The effective annual mean deposition velocity (Vd) predicted by the CBED model is negative for the cropland and grassland sites, as a result of a non-zero compensation point for these land-use classes, but otherwise the lowest Vd for NH3 is always that predicted by the EMEP scheme. The discrepancies can be ascribed to different parameterisations for the non-stomatal resistances. Model estimates of aerosol Vd differ greatly among the Â�various modelling approaches and parameterisations (see Ruijgrok et€ al., 1997, for a review), but it is in the size range 0.1–1.0 μm that the variability and uncertainty are Â�greatest. Whereas mechanistic models predict very low deposition velocities for fine aerosols, typically of the order of 0.1â•›mmâ•›s−1, field measurements suggest that Vd is 1–3 orders of magnitude higher (Gallagher et€ al., 2002; Zhang et€ al., 2001). Still, such field measurements are also subject to great uncertainty (Pryor et€al., 2008b,c; Rannik et€al., 2003). This is especially relevant for reactive nitrogen in the aerosol phase, as NH+4 and NO−3 are mostly (>90%) present as sub-micron particles.

14.4.5╇ Evaluation of emissions Emissions are the most important input to all CTM models, essential to both good model performance against observations and to the reliability of any emission control scenarios. Sources of reactive nitrogen to the atmosphere have been discussed for instance in Hertel et€al. (2011) (Chapter 9 this volume), and uncertainties in inventories will be discussed in Section 14.5. Satellite-borne instruments, e.g. GOME (Burrows et€ al., 1999), SCIAMACHY (Bovensmann et€ al., 1999), aboard the ENVISAT satellite, or OMI (Boersma et€ al., 2007), represent an interesting possibility to assess emissions, or at least CTM results which can relate to emissions. Such satellites �provide global coverage of some compounds at a spatial resolution of a few tens of kilometres. One of the first outstanding pictures provided by the use of such data was the decreasing NO2 �column trends over North America and Europe, and the increasing trend over China (Richter et€al., 2005; van der A et€al., 2006). However, the extent to which tropospheric �columns can be used for characterising air quality, namely surface concentrations, is not

David Simpson

obvious, due to measurements uncertainty and column vs. surface representativeness. Blond et€al. (2007) showed in particular that spatial variability of surface concentrations in and near European cities are not captured by satellite measurements. Satellite measurements have also been used to constrain (Martin et€ al., 2003) or estimate NOx emissions (Leue et€ al., 2001). Konovalov et€ al. (2005) and Blond et€ al. (2007) show that the spatial distribution of tropospheric NO2 is generally well simulated by chemistry-transport models. However, sub-Â�regional model underestimations, especially in Southern Europe, are present, which suggests an underestimate of NOx emissions. Other models have been tested against satellite measurements, especially over China (Ma et€al., 2006). Konovalov et€al. (2006) attempted to invert measurements in order to obtain emissions at regional scale over Europe, using relations fitted to a chemistry-transport model. This method was also applied to the estimation of NOx emission decadal trends (Konovalov et€al., 2008). It showed marked differences between trends in ‘bottom-up’ and satellite derived NOx emission trend estimates in Southern European regions, while trends are consistent in Northern areas. Satellites show also some potential for the evaluation of modelled fields of NH3 and hence of their emissions (Beer et€al., 2008; Clarisse et€al., 2009). Current retrieval methods require improvement, however, before this potential can be realised.

14.5╇ Uncertainties and challenges Estimation of the atmospheric inputs to Nr deposition is challenging because of uncertainties in the whole chain of proÂ� cesses€ – emissions, dispersion, chemistry and deposition. Monks et€ al. (2009) have discussed many of the issues with regard to oxidised nitrogen, and reviews such as those by Bleeker et€ al. (2009) or van Pul et€ al. (2009) provide much information on reduced nitrogen. Emissions and processes have been discussed in Hertel et€al., 2011 (Chapter 9 this volume). Here we highlight those issues specific to modelling and measurements rather than processes. As noted in Hertel et€al. (2006), there is a substantial discrepancy in the relative importance of various physical and chemical processes that need to be taken into account in local and regional scale models for N deposition, due to the differences in time scale. On the local scale the dispersion of pollutants is the most important process with regards to the concentration levels, whereas beyond about 5€km it is increasingly necessary to have good descriptions of wet and dry deposition processes, and atmospheric chemistry.

Measurements Generally, there are a variety of issues related to the measurements of N compounds in the atmosphere, due to their large number, low concentrations, reactivity and gas/aerosol interactions (Laj et€ al., 2009). Highly reactive gases such as NH3 or HNO3 are very challenging to detect because of their �interaction with parts of the instruments, resulting in slow sensor response times. Their very short lifetime in the atmosphere means that they are highly variable, spatially and temporally. HONO is notoriously difficult to measure, usually with

positive bias caused by photolysis (it is thought) of nitrate on sampling lines and inlets. The measurement of aerosol compounds such as ammonium and nitrate also requires sophisticated instrumentation. Indeed, one of the critical limitations for model evaluation is the lack of good measurement data on the partitioning of oxidised nitrogen between HNO3, and fine and coarse particulate nitrate. Of the EMEP sites discussed by Fagerli and Hjellbrekke (2008), only a few sites reported results for the gaseous compounds, 15 for HNO3 and 11 for NH3, and of these only two use denuders whilst the others use filter-pack methods€– results from the latter are very uncertain. Organic N in the atmosphere is in general not measured as gas/particles with the exception of occasional measurements of PAN, amines and organic nitrates (but except for PAN not routinely monitored). The recent paper by Gonzalez Benitez et€al. (2010) shows the potential scale of the problem. Measurement technology is available to measure organic N compounds such as amines and PANs, but the technology is still quite expensive, especially when it comes to continuous measurements. Another basic problem is that the surface measurements which are typically available give only a partial picture of some important chemical components. For example, gaseous NO3 is often discussed as a potentially important loss mechanism for hydrocarbons, providing a night-time alternative to OH radicals in driving chemistry (see Wayne et€ al., 1991; Brown et€ al., 2006). Unfortunately, this compound is extremely difficult to measure, and has very large vertical gradients€– surface concentrations are both modelled (Fish et€al., 1999) and measured to be very small, even when boundary layer values are significant. Over sea areas, evaluation of models is further complicated as validation is usually against measurements at coastal sites; observations in the open sea rarely exists. Furthermore, for the Black Sea and the Mediterranean Sea, very few observations exist for any location.

Meteorology Uncertainties in the meteorological data used by models are often difficult to quantify, because many of the parameters that are critical for air pollution modelling are not measured, or only available through a series of assumptions. As well as precipitation (Section 14.4.3), an important example is the height of the boundary layer, or mixing height (Hmix), which controls the dispersion of all pollutants in the boundary layer but which is difficult to define even when radiosonde data are available (Seibert et€al., 2000; Stern et€al., 2008). Other important parameters include friction velocity and stability, both of which are crucial for deposition estimates. It can be noted that the uncertainties of meteorological inputs to CTMs receives relatively little attention, and these uncertainties are probably significant.

Emissions As noted above, emissions are the most important input to all CTM models, but they often receive little attention despite being subject to substantial uncertainties (Reis et€al., 2009). As an example, looking at two recent emission inventories EDGAR

311

Atmospheric transport and deposition of reactive nitrogen

v.4 (EDGAR, 2010), and the UK National Emission Inventory (UKNAEI, 2010) for the year 2005, emission figures for NOx and NH3 for the UK both differ by about 30%. Such uncertainties are likely ubiquitous in European inventories, and indeed greater in countries with few emission measurement activities and where emission inventory development receives little funding.

Deposition modelling There are very many uncertainties regarding the magnitude (and even direction) of surface–atmosphere exchange, especially concerning dry deposition, occult deposition and emissions from soils. Particle deposition rates are one obvious source of uncertainty (Pryor et€al., 2008a, c). As another example, the non-stomatal resistance term for NH3 is not only a function of ambient NH3 but also of the concentrations of acid gases (HNO3, SO2, HCl) which neutralise NH3 in water films on vegetation (Flechard et€al., 1999; Fowler et€al., 2009). To quantify this effect mechanistically requires dynamic chemical modelling with very short time steps, which precludes the implementation of such schemes in regional models, and some models (EMEP, IDEM) use the NH3/SO2 ratio as a proxy in empirical parameterisations. Flux networks such as Nitro-Europe will hopefully help reduce uncertainties in some parts of the Nr deposition budget, but there is a clear need for both improved instrumentation and analysis techniques before reliable estimates of Nr deposition can be made.

14.6╇ Conclusions This chapter has attempted to answer the overriding question:€what are the atmospheric inputs of reactive nitrogen (Nr) in Europe, and how well can we estimate these? We have focused mainly on presenting results from models, partly because of the limitations of measurements, but also because models allow for spatially comprehensive estimates of pollutant concentrations, and for mapping of deposition patterns over large areas. Models are also key tools to integrate our understanding of atmospheric chemistry and transport. Models address emissions, dispersion and transport over multiple scales, chemical transformation, and dry and wet removal of pollutants. This chapter, along with Hertel et€al., 2011 (Chapter 9 this volume), has also discussed many of the uncertainties associated with deposition estimates of Nr. For specific locations, and especially at fine scales, such estimates can be very uncertain, with considerable variations on spatial scales of less than a kilometre. Differences between modelled and observed wet deposition of more than a factor of two are not uncommon for specific sites, especially in regions of complex topography. However, it should be remembered that mass considerations provide a strong constraint on the uncertainties in Nr deposition. Globally, all emissions of Nr will deposit somewhere, so that uncertainties in the deposition are equal to uncertainties in the emissions. Over Europe this equivalency still holds to a large extent, as the lifetime of emitted Nr is usually less than a few days. Such considerations probably explain why the current generation of chemical transport models perform quite

312

well (within 30%–50% say) when compared to the available (albeit very limited) observational data for wet deposition. Estimation of deposition at fine scales remains however a formidable task, and this poses challenges for estimating exceedances of critical levels for sensitive ecosystems. Improvements in models, emissions, measurements and understanding of physical/chemical processes will be needed before we can map fine-scale Nr deposition with confidence. Ideally, model evaluation and improvement of deposition estimates should be guided by direct measurements of fluxes of Nr, but such data are extremely expensive. In any case, observations of airborne components can also play a strong role in improving models, as there are many aspects of atmospheric chemistry which are still not properly evaluated. The main needs for oxidised compounds are probably to evaluate how well the models capture the partitioning of Nr between gaseous HNO3 and either fine or coarse nitrate. For reduced compounds, better estimates of emissions are needed, and how these are affected by meteorological factors as well as agricultural practices, coupled with an understanding of biosphere-atmosphere exchange. Such work should benefit from detailed studies from research networks such as NitroEurope, EUCAARI and EUSAAR, and from field campaigns (Laj et€al., 2009; Kulmala et€al., 2009; Sutton et€al., 2007; Tang et€al., 2009). Long-term monitoring, and a balanced hierarchy of a limited number of so-called super-sites (level 3 in EMEP terminology) and larger numbers of simpler level 1 and 2 sites (UNECE, 2009) would still be a crucial requirement, however, in order to assess (among other things) emission inventories, atmospheric processes, and long-term model performance.

Acknowledgements This chapter was prepared with the support of the NinE Programme of the European Science Foundation, the NitroEurope IP (funded by the European Commission), COST Action 729 and EMEP under the LRTAP Convention.

References Aas, W. and Hjellbrekke, A.-G. (2005). Data Quality 2003: Quality Assurance and Field Comparisons. EMEP/CCC Report 6/2005, The Norwegian Institute for Air Research (NILU), Kjeller, Norway. Asman, W. (1998). Factors influencing local dry deposition of gases with special reference to ammonia. Atmos. Environ., 32, 415–421. Asman, W., Drukker, B. and Janssen, A. (1998). Modeled historical concentrations and depositions of ammonia and ammonium in Europe. Atmos. Environ., 22, 725–735. Barrie, L. (1992). Scavenging ratios:€black magic or a useful scientific tool? In:€Proc. of 5th Int. Conf. on Precipitation Scavenging and Atmosphere–Surface Exchange Processes, ed. S. Schwartz and W. Slinn, vol. 1, the Georgii volume; Precipitation scavenging processes, pp. 403–420, Hemisphere Publ. Corp., Washington, DC. Bartnicki, J. and Fagerli, H. (2008). Airborne load of nitrogen to European seas. Ecol. Chem. Eng. S, 15, 297–313. Beer, R., Shephard, M. W., Kulawik, S. S. et€al. (2008). First satellite observations of lower tropospheric ammonia and methanol. Geophys. Res. Lett., 35.

David Simpson Berge, E. and Jakobsen, H. A. (1998). A regional scale multi-layer model for the calculation of long-term transport and deposition of air pollution in Europe. Tellus, 50, 205–223. Berge, H. and Hjellbrekke, A.-G. (2010). Acidifying and Eutrophying Components:€Validation and Combined Maps, supplementary material to emep status report 1/2010, available online at www. emep. int, The Norwegian Meteorological Institute, Oslo, Norway. Bessagnet, B., Hodzic, A., Vautard, R. et€al. (2004). Aerosol modeling with CHIMERE€– preliminary evaluation at the continental scale. Atmos. Environ., 38, 2803–2817. Binkowski, F. and Roselle, S. (2003). Models-3 community multiscale air quality (CMAQ) model aerosol component€– 1. Model description. J. Geophys. Res., 108. Bleeker, A., Reinds, G., Vermeulen, A., de Vries, W. and Erisman,€J. (2004). Critical Loads and Present Deposition Thresholds of Nitrogen and Acidity and Their Exceedances at the Level II and Level I Monitoring Plots in Europe, ECN Report ECN-C-04–117, Energy Research Centre of the Netherlands (ECN). Bleeker, A., Sutton, M., Acherman, B. et€al. (2009). Linking ammonia emission trends to measured concentrations and deposition of reduced nitrogen at different scales. In: Atmospheric Ammonia. Detecting Emissions Changes and Environmental Impacts, ed. M. A. Sutton, S. Reis and S. M. Baker, pp. 123–180, Springer, New York. Blond, N., Boersma, K. F., Eskes, H. J. et€al. (2007). Intercomparison of SCIAMACHY nitrogen dioxide observations, in situ measurements and air quality modeling results over Western Europe. J. Geophys. Res., 112, D10 311. Boersma, K., Eskes, H., Veefkind, J. et€al. (2007). Near-real time retrieval of tropospheric NO2 from OMI. Atmos. Chem. Phys. 7, 2103–2118. Bovensmann, H., Burrows, J. P., Buchwitz, M. et€al. (1999). SCIAMACHY€– mission objectives and measurement modes. J.€Atmos. Sci., 56, 127–150. Brown, S., Ryerson, T., Wollny, A. et€al. (2006). Variability in nocturnal nitrogen oxide processing and its role in regional air quality. Science, 311, 67–70. Burrows, J. P., Weber, M., Buchwitz, M. et€al. (1999). The Global Ozone Monitoring Experiment (GOME)€–€mission concept and first scientific results, J. Atmos. Sci., 56, 151–175. Cape, J., Kirika, A., Rowland, A. et€al. (2001). Organic nitrogen in precipitation€–€real problem or sampling artefact? Scientif. World, 1, 230–237. Carmichael, G., Peters, L. and Saylor, R. (1991). The STEM-II regional scale acid deposition and photochemical oxidant model. 1. Overview of model development and applications. J. Geophys. Res., 25, 2077–2090. Carruthers, D., Holroy, D. et€al. (1994). UK-ADMS€– a new approach to modeling dispersion in the Earth’s atmospheric boundary layer. J. Wind Eng. Industr. Aerodynam., 52, 139–153. Chang, J., Brost, A., Isaksen, I. et€al. (1987). A three-dimensional Eulerian acid deposition model€–€physical concepts and formulation. J. Geophys. Res., 92, 14681–14700. Cimorelli, A. J., Perry, S. G., Venkatram, A. et€al. (2005). AERMOD:€a dispersion model for industrial source applications. 1:€General model formulation and boundary layer characterization. J. Appl. Meteorol., 44, 682–693. Clarisse, L., Clerbaux, C., Dentener, F., Hurtmans, D. and Coheur, P.-F. (2009). Global ammonia distribution derived from infrared satellite observations. Nature Geosci., 2, 479–483. Clarke, J., Edgerton, E. and Martin, B. (1997). Dry deposition calculations for the clean air status and trends network. Atmos. Environ., 31, 3667–3678.

Cofala, J., Amann, M., Kupiainen, K. and Hoglund-Isaksson, L. (2007). Scenarios of global anthropogenic emissions of air pollutants and methane until 2030. Atmos. Env., 41, 8486–8499. Collins, W., Stevenson, D., Johnson, C . and Derwent, R. (1997). Tropospheric ozone in a global-scale three-dimensional Lagrangian model and its response to NOx emission controls. J.€Atmos. Chem., 26, 223–274. de Leeuw, G., Cohen, L. Frohn, L. et€al. (2001). Atmospheric input of nitrogen into the North Sea€–€ANICE project overview. Cont. Shelf Res., 21, 2073–2094. de Leeuw, G., Skjoth, C, Hertel, O. et€al. (2003). Deposition of nitrogen into the North Sea. Atmos. Environ., 37, S145–S165. Delle Monache, L. and Stull, R. (2003). An ensemble air-quality forecast over western Europe during an ozone episode. Atmos. Environ., 37, 3469–3474. Dentener, F., Drevet, J., Lamarque, J. F. et€al. (2006). Nitrogen and sulfur deposition on regional and global scales€–€a multimodel evaluation. Glob. Biogeochem. Cycles, 20. Dore, A., Choularton, T. and Fowler, D. (1992). An improved wet deposition map of the United Kingdom incorporating the seederfeeder effect over mountanous terrain. Atmos. Environ., 26A, 1375–1381. Draaijers, G. and Erisman, J. (1993). Atmospheric sulfur deposition to forest stands€– Through-fall estimates compared to estimates from inference. Atmos. Environ., 27A, 43–55. Duyzer, J., Nijenhuis, B. and Weststrate, H. (2001). Monitoring and modelling of ammonia consequences deposition in agricultural areas of the Netherlands. Water, Air and Soil Pollut.€Focus, 1, 131–144. EDGAR (2010). v.4, http :€//edgar. j re.ec.europa.eu EIONET (2010). http://air-climate.eionet.europa.eu/databases/MDS/ index_html EMEP (1996). EMEP Manual for Sampling and Chemical Analysis, EMEP/CCC-Report 1995, last revision in 2001. Norwegian Institute for Air Research, Kjeller, Norway. Online at http ://www. nilu.no/proj ects/ccc/manual/index.html EMEP http ://ebas.nilu.no (2010). www. emep.int EPA (US Environmental Protection Agency) (2010). www.epa.gov/ scram001/aqmindex. htm Erisman, J., Draaijers, G., Duyzer, J. et€al. (1997). Particle deposition to forests€– summary of results and application. Atmos. Environ., 31, 321–322. Erisman, J. W., Hensen, A., Fowler, D. et€al. (2001). Dry Deposition monitoring in Europe, Water, Air Soil Pollut.€Focus, 1, 17–27. Erisman, J., Vermeulen, A., Hensen, A. et€al. (2005). Monitoring and modelling of biosphere/atmosphere exchange of gases and aerosols in Europe. Environ. Pollut., 403–413. Erisman, J., van Grinsven, H., Grizzetti, B. et€al. (2011). The European nitrogen problem in a global perspective. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Fagerli, H. and Aas, W. (2008). Trends of nitrogen in air and precipitation:€model results and observations at EMEP sites in Europe, 1980–2003. Environ. Pollut., 154, 448–461. Fagerli, H. and Hjellbrekke, A. (2008). Acidification and eutrophication. In: Transboundary Acidification, Eutrophication and Ground Level Ozone in Europe in 2006, EMEP Status Report 1/2008, pp. 41–56, The Norwegian Meteorological Institute, Oslo, Norway. Fagerli, H., Wind, P., Gauss, M. et€al. (2008). Improved resolution in EMEP models. In:€Transboundary Acidification, Eutrophication

313

Atmospheric transport and deposition of reactive nitrogen and Ground Level Ozone in Europe in 2006, EMEP Status Report 1/2008, pp. 89–104, The Norwegian Meteorological Institute, Oslo, Norway. Fish, D., Shallcross, D. and Jones, R. (1999). The vertical distribution of NO3 in the atmospheric boundary layer. Atmos. Environ., 33, 687–691. Flechard, C. R., Fowler, D., Sutton, M. A. and Cape, J. N. (1999). A dynamic chemical model of bi-directional ammonia exchange between semi-natural vegetation and the atmosphere. Q. J. R. Meteorol. Soc, 125, 2611–2641. Fournier, N., Tang, Y., Dragosits, U., De Kluizenaar, Y. and Sutton, M. (2005). Regional atmospheric budgets of reduced nitrogen over the British Isles assessed using a multi-layer atmospheric transport model. Water, Air Soil Pollut., 162, 331–351. Fowler, D., Cape, J., Leith, I. et€al. (1988). The influence of altitude on rainfall composition at Great Dun Fell. Atmos. Environ., 22, 1355–1362. Fowler, D., Pitcairn, C. E. R., Sutton, M. A. et€al. (1998). The mass budget of atmospheric ammonia in woodland within 1 km of livestock buildings. Environ. Pollut., 102, 343–348. Fowler, D., Coyle, M., Flechard, C. et€al. (2001). Advances in micrometeorological methods for the measurement and interpretation of gas and particle nitrogen fluxes. Plant and Soil, 228, 117–129. Fowler, D., Pilegaard, K., Sutton, M. et€al. (2009). Atmospheric composition change€–€ecosystems–atmosphere interactions. Atmos. Environ., 43, 5193–5267. Frohn, L. M., Christensen, J. H., Brandt, J. and Hertel, O. (2001). Development of a high resolution integrated nested model for studying air pollution in Denmark. Phys. Chem. Earth, B, 26, 769–774. Gallagher, M. W., Nemitz, E., Dorsey, J. R. et€al. (2002). Measurements and parameterizations of small aerosol deposition velocities to grassland, arable crops, and forest€–€influence of surface roughness length on deposition, J. Geophys. Res., 107. Galloway, J., Dentener, F., Capone, D. et€al. (2004). Nitrogen cycles:€past, present, and future. Biogeochemistry, 70, 153–226. Galmarini, S., Bianconi, R., Addis, R. et€al. (2004). Ensemble dispersion forecasting.€2.€Application and evaluation. Atmos. Environ., 38, 4619–4632. Gonzalez Benitez, J. M., Cape, J. N., Heal, M. R., van Dijk, N. and Diez, A. V. (2009). Atmospheric nitrogen deposition in south-east Scotland€–€quantification of the organic nitrogen fraction in wet, dry and bulk deposition. Atmos. Environ., 43, 4087–4094. Gonzalez Benitez, J. M., Cape, J. N. and Heal, M. R. (2010). Gaseous and particulate water-soluble organic and inorganic nitrogen in rural air in southern Scotland. Atmos. Environ., 44, 1506–1514. HELCOM (2005). Airborne Nitrogen Loads to the Baltic Sea, Helsinki Commission. Hertel, O., Christensen, J., Runge, E. et€al. (1995). Development and testing of a new variable scale air-pollution model€– ACDEP. Atmos. Environ., 29, 1267–1290. Hertel, O., Ambelas Skjoth, C., Frohn, L. M. et€al. (2002). Assessment of the atmospheric nitrogen and sulphur inputs into the North Sea using a Lagrangian model. Phys. Chem. Earth, B, 27, 1507–1515. Hertel, O., Skjoth, C., Brandt, J. et€al. (2003). Operational mapping of atmospheric nitrogen deposition to the Baltic Sea. Atmos. Chem. Phys., 3, 2083–2099.

314

Hertel, O., Skjoth, C. A., Lofstrom, P. et€al. (2006). Modelling nitrogen deposition on a local scale€– a review of the current state of the art. Environ. Chem., 3, 317–337. Hertel, O., Reis, S., Ambelas Skjoth, C. et€al. (2011). Nitrogen processes in the atmosphere. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Holmes, N. and Morawska, L. (2006). A review of dispersion modelling and its application to the dispersion of particles€–€an overview of different dispersion models available. Atmos. Environ., 40, 5902–5928. HTAP (UNECE Hemispheric Task Force on Atmospheric Pollutants) (2010). www. htap.org IPCC (2007). Climate Change 2007. The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the International Panel on Climate Change, www.ipcc.ch. Jenkin, M. E., Saunders, S. M., Wagner, V. and Pilling, M. J. (2003). Protocol for the development of the Master Chemical Mechanism, MCM v3 (Part B):€tropospheric degradation of aromatic volatile organic compounds. Atmos. Chem. Phys., 3, 181–193. Jenkin, M., Utembe, S. and Johnson, D. (2006). The simulated speciation and distribution of oxidized organic nitrogen in the atmospheric boundary layer. In:€Understanding and Quantifying the Atmospheric Nitrogen Cycle, ed. R. Cox, D. Fowler, P. Monks and P. Borrell, pp. 130–136, ACCENT, Keele University, UK. Johnson, D., Utembe, S. R. and Jenkin, M. E. (2006). Simulating the detailed chemical composition of secondary organic aerosol formed on a regional scale during the TORCH 2003 campaign in the southern UK. Atmos. Chem. Phys., 6, 419–431. Kessler, C., Brucher, W., Memmesheimer, M., Kerschgens, M. and Ebel, A. (2001). Simulation of air pollution with nested models in North Rhine-Westphalia. Atmos. Environ., 35, S3–S12. Konovalov, I., Beekmann, M., Vautard, R. et€al. (2005). Comparison and evaluation of modelled and GOME measurement derived tropospheric NO2 columns over Western and Eastern Europe. Atmos. Chem. Phys., 5, 169–190. Konovalov, I. B., Beekmann, M., Richter, A. and Burrows, J. P. (2006). Inverse modelling of the spatial distribution of NOx emissions on a continental scale using satellite data. Atmos. Chem. and Phys., 6, 1747–1770. Konovalov, I. B., Beekmann, M., Burrows, J. P. and Richter, A. (2008). Satellite measurement based estimates of decadal changes in European nitrogen oxides emissions. Atmos. Chem. Phys., 8, 2623–2641. Korhonen, H., Lehtinen, K. E. J. and Kulmala, M. (2004). Multicomponent aerosol dynamics model UHMA:€model development and validation. Atmos. Chem. Phys., 4, 757–771. Laj, P., Klausen, J., Bilde, M. et€al. (2009). Measuring atmospheric composition change. Atmos. Environ., 43, 5351–5414. Langner, J., Andersson, C. and Engardt, M. (2009). Atmospheric input of nitrogen to the Baltic sea basin€–€present situation, variability due to meteorology and impact of climate change. Boreal Environ. Res., 14, 226–237. Leue, C., Wenig, M., Wagner, T. et€al. (2001). Quantitative analysis of NOx emissions from GOME satellite image sequences. J. Geophys. Res., 106, 5493–5505. Loubet, B. and Cellier, P. (2001). Experimental assessment of atmospheric ammonia dispersion and short range dry deposition in a maize canopy. Water, Air and Soil Pollution: Focus, 1, 157–166. Loubet, B., Cellier, P., Milford, C. and Sutton, M. A. (2006). A coupled dispersion and exchange model for short-range dry

David Simpson deposition of atmospheric ammonia. Quarterly Journal of the Royal Meteorological Society, 132, 1733–1763. Loubet, B., Asman, W. A., Theobald, M. R. et€al. (2009). Ammonia deposition near hot spots:€processes, models and monitoring methods. In:€Atmospheric Ammonia. Detecting Emissions Changes and Environmental Impacts, ed. M. A. Sutton, S. Reis and S.€M.€Baker, pp. 205–267, Springer, New York. Ma, J., Richter, A., Burrows, J. P., Nuss, H. and van Aardenne, J. A. (2006). Comparison of model-simulated tropospheric NO2 over China with GOME-satellite data. Atmos. Environ., 40, 593–604. Martin, R V., Jacob, D. J., Chance, K. V. et€al. (2003). Global inventory of nitrogen dioxide emissions constrained by space-based observations of NO2 columns. J. Geophys. Res., 108, 4537. McKeen, S., Chung, S. H., Wilczak, J. et€al. (2007). Evaluation of several PM2.5 forecast models using data collected during the ICARTT/NEAQS 2004 field study. J. Geophys. Res., 112. Monks, P., Granier, C., Fuzzi, S. et€al. (2009). Atmospheric composition change€– global and regional air quality. Atmos. Environ., 43, 5268–5350. Olesen, H., Lofstrem, P., Berkowicz, R. and Jensen, A. (1992). An improved dispersion model for regulatory use€– the OML model. In: Air Pollution Modelling and its Application IX, ed. H. van Dop and G. Kallos, Plenum Press, New York. Olesen, H. R. (1995). The model validation exercise at M01:€overview of results. Int. J. Environ. Pollut. 5, 761. Pagowski, M., Grell, G., McKeen, S. et€al. (2005). A simple method to improve ensemble-based ozone forecasts. Geophys. Res. Lett., 32. Pilegaard, K., Skiba, U., Ambus, P. et€al. (2005). Factors controlling regional differences in forest soil emission of nitrogen oxides (NO and N2O). Biogeosciences, 3, 651–661. Pryor, S. C., Barthelmie, R. J., Sorensen, L. L. et€al. (2008a). Upward fluxes of particles over forests€–€when, where, why? Tellus, 60, 372–380. Pryor, S. C., Gallagher, M., Sievering, H. et€al. (2008b). A review of measurement and modelling results of particle atmosphere– surface exchange. Tellus, 60, 42–75. Pryor, S. C., Larsen, S. E., Sorensen, L. L. and Barthelmie, R. J. (2008c). Particle fluxes above forests:€Observations, methodological considerations and method comparisons. Environ. Pollut., 152, 667–678. Rannik, U., Aalto, P., Keronen, P., Vesala, T. and Kulmala, M. (2003). Interpretation of aerosol particle fluxes over pine forest€–€dry deposition and random errors. J. Geophys. Res., 108, 4544. Reis, S., Pinder, R. W., Zhang, M., Lijie, G. and Sutton, M. A. (2009). Reactive nitrogen in atmospheric emission inventories. Atmos. Chem. Phys., 9, 7657–7677. Rendell, A. R., Ottley, C. J., Jickells, T. D. and Harrison, R. M. (1993). The atmospheric input of nitrogen species to the North Sea. Tellus€B, 45, 53–63. Riccio, A., Giunta, G. and Galmarini, S. (2007). Seeking for the rational basis of the Median Model:€the optimal combination of multimodel ensemble results. Atmos. Chem. Phys., 7, 6085–6098. Richter, A., Burrows, J. P., Nuß, H., Granier, C . and Niemeier, U. (2005). Increase in tropospheric nitrogen dioxide over China observed from space. Nature, 437, 129–132. Robertson, L., Langner, J. and Engardt, M. (1999). An Eulerian limited-area atmospheric transport model. J. Appl. Meteorol., 38, 190–210. Roelofs, G., Kasibhatla, P., Barrie, L. et€al. (2001). Analysis of regional budgets of sulfur species modeled for the COSAM exercise. Tellus, 53B, 673–694.

Ruijgrok, W., Tieben, H. and Eisinga, P. (1997). The dry deposition of particles to forest canopy€–€a comparison of model and experimental results. Atmos. Environ., 31, 399–415. Sanderson, M., Dentener, R, Fiore, A. et€al. (2008). A multimodel study of the hemispheric transport and deposition of oxidised nitrogen. Geophys. Res. Lett., 35, L17 815, doi:10.1029/2008GL035389. Schaap, M., Timmermans, R. M. A., Roemer, M. et€al. (2008). The LOTOS-EUROS model:€description, validation and latest developments. Int. J. Environ. Pollut., 32, 270–290. Schaap, M., Vautard, R., Bergstrom, R. et€al. (2010). Evaluation of long-term aerosol simulations from seven air quality models and their ensemble in the EURODELTA study (submitted). Schlunzen, K. H. and Meyer, E. M. I. (2007). Impacts of meteorological situations and chemical reactions on daily dry deposition of nitrogen into the southern North Sea. Atmos. Environ., 41, 289–302. Seibert, P., Beyrich, F., Gryning, S.-E. et€al. (2000). Review and intercomparison of operational methods for the determination of the mixing height. Atmos. Environ., 34, 1001–1027. Seinfeld, J. and Pandis, S. (1998). Atmospheric Chemistry and Physics. From Air Pollution to Climate Change, John Wiley and Sons, New York. Simpson, D., Tuovinen, J.-P., Emberson, L. and Ashmore, M. (2001). Characteristics of an ozone deposition module. Water, Air Soil Pollut.€Focus, 1, 253–262. Simpson, D., Fagerli, H., Jonson, J. et€al. (2003). The EMEP Unified Eulerian Model – Model Description, EMEP MSC-W Report 1/2003, The Norwegian Meteorological Institute, Oslo, Norway. Simpson, D., Butterbach-Bahl, K., Fagerli, H. et€al. (2006a). Deposition and emissions of reactive nitrogen over European forests:€a modelling study. Atmos. Environ., 40, 5712–5726. Simpson, D., Fagerli, H., Hellsten, S., Knulst, J., and Westling, O. (2006b). Comparison of modelled and monitored deposition fluxes of sulphur and nitrogen to ICP-forest sites in Europe. Bio-geosciences, 3, 337–355. Singles, R., Sutton, M. and Weston, K. (1998). A multi-layer model to describe the atmospheric transport and deposition of ammonia in Great Britain. Atmos. Environ., 32, 393–399. Sliggers, S. and Kakebeeke, W. (eds.) (2004). Clearing the Air: 25 Years of the Convention on Long-Range Transboundary Air Pollution, United Nations, Economic Commission for Europe, Geneva, http:€//www.unece .org/env/lrtap Smith, R., Fowler, D., Sutton, M. A., Flechard, C. and Coyle, M. (2000). Regional estimation of pollutant gas dry deposition in the UK€–€model description, sensitivity analyses and outputs. Atmos. Environ., 34, 3757–3777. Sommer, S. G., and Jensen, E. S. (1991). Foliar absorption of atmospheric ammonia by ryegrass in the field. Journal of Environmental Quality, 20, 153–156. Sportisse, B. (2007). A review of current issues in air pollution modeling and simulation. Comp. Geosci., 11, 159–181. Stern, R., Builtjes, P., Schaap, M. et€al. (2008). A model intercomparison study focussing on episodes with elevated PM10 concentrations. Atmos. Environ., 42, 4567–4588. Sutton, M., Nemitz, E., Erisman, J. et€al. (2007). Challenges in quantifying biosphere–atmosphere exchange of nitrogen species. Environ. Poll., 150, 125–139. Sutton, M. A., Milford, C., Dragosits, U. et€al. (1998). Dispersion, deposition and impacts of atmospheric ammonia€–€quantifying local budgets and spatial variability. Environ. Pollut., 102, 349–361.

315

Atmospheric transport and deposition of reactive nitrogen Tang, Y., Simmons, I., van Dijk, N. et€al. (2009). European scale application of atmospheric reactive nitrogen measurements in a low-cost approach to infer dry deposition fluxes. Agricult., Ecosyst. Environ. 133, 183–195. Theobald, M. R., Millford, C., Hargreaves, K. J. et€al. (2001). Potential for ammonia recapture by farm woodlands: design and application of a new experimental facility. 1(S2), 791–801. Textor, C., Schulz, M., Guibert, S. et€al. (2006). Analysis and quantification of the diversities of aerosol life cycles within AeroCom. Atmos. Chem. Phys., 6, 1777–1813. UNECE (2009). EMEP Draft revised monitoring strategy, http:// www.unece.org/env/ documents/2009/EB/ge1/ece.eb.air. ge.1.2009.15.e.pdf UKNAEI (UK National Emission inventory) (2010). www. uknaei. org. uk van Aardenne, J. A., Dentener, F. J., Olivier, J. G. J., Goldewijk, C. G. M. K. and Lelieveld, J. (2001). A 1 degrees × 1 degrees resolution data set of historical anthropogenic trace gas emissions for the period 1890–1990. Glob. Biogeochem. Cycles, 15, 909–928. van der A, R. J. M., Peters, D. H. M. U., Eskes, H. et€al. (2006). Detection of the trend and seasonal variation in tropospheric NO2 over China J. Geophys. Res., 111, D12317. van Jaarsveld, J. A. (2004). The Operational Priority Substances Model – Description and Validation of OPS-Pro 4.1, RIVM Report 500045001/2004, RIVM, Bilthoven, The Netherlands. van Loon, M., Roemer, M. and Builtjes, P. (2004). Model Intercomparison in the Framework of the Review of the Unified EMEP Model, TNO-report R2004/282, TNO, TNO Environment and Geosciences, Utrecht, The Netherlands, http://www.tno.nl/ content.cfm? context=markten&content=publicatie&laagl=186&l aag2=l&item_ id=737&Taal=2 van Loon, M., Vautard, R., Schaap, M. et€al. (2007). Evaluation of longterm ozone simulations from seven regional air quality models and their ensemble. Atmos. Environ., 41, 2083–2097.

316

van Pul, A., Jaarsveld, H. V., Meulen, T. v. d. and Velders, G. (2004). Ammonia concentrations in the Netherlands€–€spatially detailed measurements and model calculations. Atmos. Environ., 38, 4045–4055. van Pul, A., Hertel, O., Geels, C. et€al. (2009). Modelling of the atmospheric transport and deposition of ammonia at a national and regional scale. In:€Atmospheric Ammonia: Detecting Emissions Changes and Environmental Impacts, ed. M. A. Sutton, S. Reis and S. M. Baker, pp. 301–358, Springer, New York. Vautard, R., Builtjes, P., Thunis, P. et€al. (2007). Evaluation and intercomparison of ozone and PM10 simulations by several chemistry transport models over four European cities within the CityDelta project. Atmos. Environ., 41, 173–188. Vautard, R., Schaap, M., Bergstrom, R. et€al. (2008). Skill and uncertainty of a regional air quality model ensemble. Atmos. Environ. 43, 4822–4832. Vieno, M., Dore, A. J., Wind, P. et€al. (2009). Application of the EMEP Unified Model to the UK with a horizontal resolution of 5 × 5 km2. In: Atmospheric Ammonia. Detecting Emissions Changes and Environmental Impacts, ed. M. A. Sutton, S. Reis and S. M. Baker, pp. 367–372, Springer, New York. Vieno, M., Dore, A. J., Stevenson, D. S. et€al.€2010 Modelling surface ozone during the 2003 heat wave in the UK. Atmospheric Chemistry and Physics, 110, 7963–7978. Wayne, R., Barnes, I., Biggs, P. et€al. (1991). The nitrate radical:€physics, chemistry, and the atmosphere. Atmos. Environ. 25A, 1–203. Zhang, L., Gong, S., Padro, J. and Barrie, L. (2001). A size-segregated particle dry deposition scheme for an atmospheric aerosol module. Atmos. Environ., 35, 549–560. Zhang, L., Vet, R., O’Brien, J. M. et€al. (2009). Dry deposition of individual nitrogen species at eight Canadian rural sites. J.€Geophys. Res., 114, D02301.

Chapter

15

Geographical variation in terrestrial nitrogen budgets across Europe Lead author: Wim de Vries Contributing authors: Adrian Leip, Gert Jan Reinds, Johannes Kros, Jan Peter Lesschen, Alexander F. Bouwman, Bruna Grizzetti, Fayçal Bouraoui, Klaus Butterbach-Bahl, Peter Bergamaschi, Wilfried Winiwarter

Executive summary Nature of the problem • Nitrogen (N) budgets of agricultural systems give important information for assessing the impact of N inputs on the environment, and identify levers for action.

Approaches • N budgets of agro-ecosystems in the 27 EU countries are established for the year 2000, considering N inputs by fertiliser application, manure excretion, atmospheric deposition and crop fixation, and N outputs by plant uptake, gaseous emissions, mineralisation, leaching and runoff. • Country N budgets for agro-ecosystems are based on the models INTEGRATOR, IDEAg, MITERRA and IMAGE. Fine geographic distribution is depicted with the former two models, which have higher spatial resolution. INTEGRATOR is the only available model for calculating non-agricultural terrestrial N budgets systems.

Key findings/state of knowledge • For EU-27, the models estimate a comparable total N input in European agriculture, i.e. 23.3–25.7 Mton N yr−1, but N uptake varies largely from 11.3–15.4 Mton N yr−1, leading to total N surpluses varying from 10.4–13.2 Mton N yr−1. Despite this variation, the overall difference at EU-27 is small for the emissions of NH3 (2.8–3.1 Mton N yr−1) and N2O (0.33–0.43 Mton N yr−1) but estimates vary largely at a regional scale. The estimated sum of N leaching and runoff at EU-27 is roughly equal to the sum of NH3, N2O and NOx emissions to the atmosphere, but estimates vary by a factor two, from 2.7 to 6.3 Mton N yr−1. • Trends in N fluxes in agro-ecosystems since 1970 show an increase in N inputs by fertilisers and manure up to 1985, followed by a decrease since 1985 in response to a decrease in crop production and in animal numbers. Actually, livestock decreased since 1970, but in the period 1970–1985 the N input by manure excretion still increased due to an increase in N excretion rates. • In non-agricultural system (forests and semi-natural vegetation), the estimated total N input is near 3.2 Mton N yr−1, while the net N uptake is near 1.1 Mton N yr−1, leading to a surplus near 2.1 Mton N yr−1. Compared to agricultural systems, the estimated N fluxes in non-agricultural systems are about five times lower for N2O emissions and 10 times lower for NOx and NH3 emissions and for the sum of N leaching and runoff.

Major uncertainties/challenges • The largest uncertainties in flux values, as estimated from inter-model comparison, concerns N leaching and runoff, followed by N2O emissions, from agricultural ecosystems.

Recommendations • Future research should focus on reducing the fluxes with the most uncertainty (N leaching and runoff, followed by N2O emissions, from agricultural ecosystems), including studies on denitrification. • To improve model assessments and enable model validation, databases should be set up of:€(i) N contents in major crops/vegetation in various regions (to improve estimates of N uptake and N surplus), (ii) NH3 and N2O emissions based on inverse modelling approaches

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

317

Geographical variation in terrestrial nitrogen budgets

(to validate N emission calculations) and N concentrations in ground water and surface water (to validate N leaching and N runoff assessments). • The number of countries with estimated NH3-N emissions in 2000 exceeding the National emission ceilings for 2010 depends on the model approach and varies between 7 and 18. Exceedance of critical N concentrations in surface water is highly model-dependent. It is relevant that data use, both on activity data and emission or leaching factors is harmonised for models predicting air emissions and N loss to waters for consistent environmental decision-making relevant to air quality, ecosystem deposition and water quality.

15.1╇ Introduction The major share of new reactive nitrogen (Nr) is introduced into the environment with the purpose of producing agricultural commodities. Excess N input, however, causes a number of ecological and human health effects, like acidification, eutrophication, elevated N saturation of forest soils, climate change and biodiversity impacts (see also Grizzetti et╯al., 2011; Moldanová et╯al., 2011; Butterbach-Bahl et╯al., 2011, Dise et╯al., 2011; Velthof et╯al., 2011, Chapters 17–21, this volume). An indication of the potential impact of N inputs in agriculture can be derived by an overview of all N inputs and N outputs, here referred to as an N budget. N budgets of agro-ecosystems are generally constructed (i) to increase the understanding of nutrient cycling, (ii) for use as performance indicator and to raise awareness in nutrient management and environmental policy, and (iii) as regulating policy instrument to monitor and enforce a certain nutrient management policy in practice (Oenema et╯al., 2003). Sometimes, the term N balance is also used, but this term is consistently used in this chapter to denote the N surplus, defined as the sum of all N inputs minus N removal by feed and food, in line with its use by the Organisation for Economic Co-operation and Development (OECD) (OECD, 2001, 2007). We use the word N budget for a complete N flux assessment. In this chapter, we present N budgets of agro-ecosystems and non-agricultural terrestrial ecosystems in Europe as performance indicator, illustrating the N use efficiency of agro-ecosystems and the loss of excess N to the environment (air and water). We summarise the present knowledge on European N budgets for terrestrial ecosystems by using a range of different modelling and input data assessment approaches. This way we implicitly assess uncertainties. As a part of the budget approach, the chapter includes key N fluxes, including N inputs by manure, fertiliser, deposition and fixation, N uptake, emissions of ammonia (NH3), nitrous oxide (N2O) nitrogen oxides (NOx) and di-nitrogen (N2), and the sum of N leaching and runoff, to provide an overall picture of the N status of Europe. The assessment concentrates at discussing data at the country level with the EU-27 as geographical scope, even though the calculations are performed in many models at much higher resolution in order to cover the nonlinearity of the soil proÂ� cesses. Most data are available around the year 2000 and so most of the data presented are reflecting the situation around this year. However, we include also a discussion of the past trends of important elements in the N-budgets since 1970 onwards. In Section 15.2, we first describe the modelling approaches and input data that are available to assess terrestrial N fluxes at

318

the European scale. We then present results in terms of farm and land N budgets for agricultural systems, including trends in N budgets in the period 1970–2000 (Section 15.3) followed by land N budgets for non-agricultural terrestrial systems (Section 15.4). An overall evaluation of the results is given in Section 15.5. This includes an evaluation of the validity of the presented model approaches by comparison of model results with independent datasets, whenever available. Furthermore, the relevance of N budgets and their trends with respect to effects on ecosystems and the reliability of N budgets at various geographic scales are discussed. For a complete overview of aggregated N fluxes across media and sectors for countries throughout Europe, we refer to Leip et╯al., 2011a (Chapter 16, this volume). Details on N sources in deposition are given in Simpson et╯al., 2011 (Chapter 14, this volume).

15.2╇ Methodological approaches and input data to assess terrestrial nitrogen budgets at the European scale 15.2.1╇ Approaches to assess nitrogen budgets at regional scale While we are interested to obtain N budgets for agriculture on a regional, country or European level, we need to differentiate different budgeting approaches by the respective system boundaries used. We distinguish three basic approaches in regional N budget studies, using the farm, land or soil as the gate at which the N inputs and outputs are quantified (see Table 15.1). (1) Farm nitrogen budget (called farm-gate budget by Oenema et╯al., 2003); it records the amounts of N in all kinds of products that enter and leave the farm via the farm-gate. Throughputs, as for example uptake of grass by animals, or the application of manure, are not part of the farm N budget. The surplus/deficit, i.e. the difference between inputs and outputs, is a measure of total N losses, adjusted for possible changes in the storage of nutrients in the farming system. Examples of this approach are the now abolished MINAS (Mineral Accounting System) regulatory nutrient book-keeping system in the Netherlands (Oenema et╯al., 1998; Neeteson, 2000), and the OSPARCOM method (Oslo and Paris Conventions for the Prevention of Marine Pollution) focusing on N and P discharges into the North Sea and Baltic Sea from the surrounding countries (OSPARCOM, 1994). In the simple farm N budget, the N surplus is not further specified, whereas N (NH3, N2O, NOx and N2) losses from

Wim de Vries Table 15.1 Definition of N inputs, N outputs and N surpluses in regional farm, land and soil nitrogen budgets for agricultural systems

Budget

System boundary

Simple

Detailed

N Inputs

N Outputs

N Surplusa

Farm

Farm N budget

Agricultural system N budget

Fertiliser, feed (concentrates), external organic N sources, N fixation and N deposition, net N manure import, and withdrawals

Sold animal (meat, milk, etc.) and crop products

N (NH3, N2O, NOx and N2) emissions and N leaching/runoff from housing and manure storage systems and soil; soil N stock changes

Land

Gross N budget (OECD approach)

Land system N budget

Fertiliser, manure excretion, external organic sources, crop residues returned on soils, N fixation, N deposition, net N manure import/ export, and withdrawals

Harvest of crop products (in arable land) or above ground removal of grass, crop residues

N (NH3, N2O, NOx and N2) emissions and N leaching/runoff from housing and manure storage systems and soil; soil N stock changes

Soil

Soil N budget

Soil system N budget

Fertiliser, manure application, grazing inputs, external organic sources, crop residues returned on soils, N fixation and N deposition

Removal of crop products (in arable land) or above ground removal of grass; crop residues, soil N stock changes

N (NH3, N2O, NOx and N2) emissions and N leaching/runoff from soil

a

N surplus is specified in the detailed N-budgets

the housing and manure storage systems and from soil to the air and to aquatic systems are specified in a detailed agricultural system budget, as illustrated in Figure 15.1. An example of this approach is the CAPRI-DNDC model (Leip et╯al., 2009). (2) Land nitrogen budgets (called gross N balances by the OECD). It records all N that enters a farm land (including housing and manure storage systems) and leaves the farmland by crop products. Nitrogen inputs include fertiliser, animal manure production/excretion, biological N fixation and N deposition. This approach is used for example by the OECD as environmental performance indicator for agriculture (OECD, 2001, 2007). In the simple approach, called gross N budget (gross N balance by the OECD), the N surplus is not further specified, whereas N losses from the housing and manure storage systems and from soil to the air and to aquatic systems are specified in a detailed land system budget. This approach is used in this chapter. (3) Soil nitrogen budget (called soil surface budget by Oenema et╯al., 2003). It records all N that enters the soil and that leaves the soil via crop uptake, including nutrient gains and losses within the soil. Nitrogen inputs via animal manure are adjusted for losses of N emissions in housing and manure management systems; all other N inputs are the same as for the land N budget. Nitrogen output (defined here as output of ‘useful product’) is corrected by the changes of N storage; accumulation of N in organic matter is regarded as useful because it improves soil quality and can potentially contribute to crop growth in following

years. Soil N surplus (see Table 15.1) is then a measure for the total N loss from the soil to either the atmosphere (NH3, N2O, NOx and N2 emissions) or the hydrosphere (N€leaching to ground water and N runoff to surface water). In the soil N budget, this N surplus is not further specified, whereas in the soil system budget all N inputs and outputs, including N gains and losses within and from the soil are specified. It should be noted, that in the literature the soil N budget mostly differs from our definition, as the NH3 emission from soils is often already corrected for while the soil N changes are included in the calculation of the surplus (Oenema et╯al., 2003). The N surplus gross N budget includes the sum of all nutrient emissions from agriculture into soil, water, and air (OECD, 2007) and is thus often used as the indicator of agricultural pressure on water quality (EEA, 2005), as it allows identifying areas with high risk of N leaching. Detailed budgets are able to resolve the individual pathways of N as presented in Table 15.1. It is important to remember that different accounting methods cover different N flows. Animal housing and manure management systems are not included in the soil budgets, while they are accounted for in farm and land budgets. In the land N budgets, the N excreted in the manure is considered, while in the soil N budget only the N in applied manure, corrected for losses in housing and manure management systems, is accounted for. Manure used for other purposes (e.g. burning) is not considered in both approaches. With respect to ‘mineral N fertiliser’, the farm N budget considers fertiliser purchases, while mineral fertiliser applications are relevant for the land and soil N budgets.

319

Geographical variation in terrestrial nitrogen budgets

While for soil budgets the system boundaries are usually the top soil layer (surface to rooting depth), and covers thus only land-based agricultural production, farm and land budgets include also the livestock sector. As for the farm (and agricultural systems) budget, the boundary is the farm, they don’t consider manure and animal intake of N in fodder produced in the farm as input or output. However, if data are available, they are often quantified as N throughput. The difference in farm, land and soil budgets is illustrated further in Leip et╯al. (2010).

15.2.2╇ Modelling approaches There are several operative activities that estimate N budgets for the European Union and for Europe at various spatial resolutions. Table 15.2 gives an overview of main model approaches that have been used for assessing total agricultural emissions of different forms of reactive N for various parts of Europe (from EU15 to whole Europe), at various geographic resolutions (from grid to country) and for different time periods. The approaches included in Table 15.2 are:€(i) complete land system N budget models for agriculture, using yearly time steps (INTEGRATOR, CAPRI, IDEAg, MITERRA, IMAGE), (ii) emission factor approaches for both agricultural and total annual NH3, N2O and NOx emissions to the atmosphere (GAINS, EMEP, EDGAR, UNFCCC-IPCC) and (iii) N loss models to either surface water (GREEN) or ground water (EPIC). In the supplementary information to this chapter (SuppÂ� lementary material, Chapter 15 & 16), a description of the various models mentioned above and the meaning of their abbreviations is given. In short, the complete land system N budget models are able to calculate all N fluxes to and from a land system, as defined in Table 15.1. First of all, these models are able to assess the N surplus or gross soil N budget according to (see Table 15.1):€N surplus = input (mineral fertilisers + livestock manure excretion corrected for transport + other organic sources + left crop residues + biological fixation + atmospheric deposition)€ – total crop removal€ – total forage uptake. The models are also all able to simulate the fate of the N surplus in terms of NH3, N2O, NOx and N2 emissions from housing and manure storage systems, N accumulation in or release from the soil (not in all models) and N losses by leaching and runoff. The emission factor approach models are limited to atmospheric emissions, but unlike the land system N budget models they include all sectors, including traffic and industry. Similarly N loss models are limited to estimates of N losses to surface water and/or ground water, but they generally include all N sources, including human sewage and direct deposition inputs to surface water. In this chapter, we focus on complete N budgets for agriculture, as derived with INTEGRATOR, IDEAg (CAPRI based model), MITERRA and IMAGE. More details on these models is given in the supplementary materials at Chapter 15 and 16 and in De Vries et╯al. (2010b). We also include a comparison of results of NH3, N2O and NOx emissions with the emission factor approaches (GAINS, EMEP, EDGAR, UNFCCC-IPCC), while results of the model GREEN are shown to illustrate the impact of diffuse sources versus point sources.

320

There are also detailed ecosystem models available that provide process-level descriptions for either daily NH3, N2O and NOx emissions, such as the DNDC model (Li et╯al., 2000) or N leaching, such as the EPIC model (Bouraoui Aloe, 2007; Van der Velde et╯al., 2009) that have been applied to derive N fluxes at regional scale in Europe. The DNDC model has for example been used to assess N2O and NOx emissions for both forests (Kesik et╯al., 2005) and agricultural land (Butterbach-Bahl et╯al., 2009) at a fixed 10 km × 10 km grid, while the EPIC model that has been applied to study the effect of agricultural practices and biofuel cultivation on N leaching (Bouraoui and Aloe, 2007; Van der Velde et╯al., 2009). However, these models do not include emissions from housing systems and in case of EPIC also not explicitly from soils, and are therefore not included in the model comparison presented in this paper. Some results are, however, shown in the Supplementary material (Chapter 15 and 16).

15.2.3╇ Data sets to estimate nitrogen inputs and outputs In order to understand the operation of models, an overview of internationally coherent datasets used by the models is given. In addition to these international datasets, often national information also exists, but in general this cannot be assessed by activities operating on a European scale. Inputs of N to agricultural systems include N fertiliser, N manure due to application and grazing, N deposition and N fixation. Data sets that are relevant for the assessment of N uptake are crop yields and element contents in crops, while N and C pools are relevant for the assessment of N emission fluxes. The assessment of N fluxes to the air (emissions of NH3, N2O, NOx, and N2) and water (N leaching to ground water, N surface runoff and subsurface flow to surface water) requires data on emission and leaching parameters in the various models to make such predictions. An overview of the data used by all the four complete N budget models is given in De Vries et╯al. (2011). More information on the datasets that are used to calculate the amount of fertiliser and manure N applied to soil is given in Supplementary material Chapter€15 and 16. In biogeochemistry models, soil C and N contents often strongly determine the N2O flux. Maps of present concentrations and pools of C and N in the soil and C/N ratios in the soil distinguishing between agricultural soils and non-agricultural soils can be based on various databases, i.e. WISE/SOTER, European Soil Data Base (ESDB2) and ICP forests database. More information on approaches and results is given in the Supplementary material (Chapter 15 and 16).

15.3╇ Farm and land nitrogen budgets for agricultural systems In the following sections, data on farm and soil N budgets are presented focusing on two recently developed model systems, i.e. IDEAg and INTEGRATOR. IDEAg consists of three

Wim de Vries Table 15.2 Overview of available models approaches for assessing emissions of different forms of Nr for various parts of Europe at various geographic resolutions and for various time periods

Model approach

Element flux considered

Method

Sectors considered

Area involved

Geographic resolution

Time

Complete land N budget models INTEGRATOR (De Vries et╯al., 2010)

N2O, NOx and NH3 emission, N leaching, N runoff

Adapted MITERRA approach for agricultural systems. Statistical model for terrestrial systems

Agriculture, terrestrial systems

EU-27+3

NCUa

1970–2000

MITERRA (Velthof et╯al., 2007, 2009)

N2O, NOx and NH3 emission, N leaching, N runoff

Emission and leaching factor approach for agricultural systems

Agriculture

EU-27

NUTS2

2000

CAPRI (Britz, 2005; Britz et╯al., 2005; CAPRI 2010)

NH3, N2O, N€surplus

Mass-budget model using an emission-factor approach

Agriculture

EU-27

NUTS2

Base year currently 2002 projections up to 2012

IDEAg, (Leip et╯al., 2008)

N2O, NOx and NH3 emission, N leaching

Economic model for agriculture, linked to mechanistic model to simulate soil N€budget

Agriculture

EU-27

HSMUa

2000

IMAGE (Alcamo, 1994; Leemans et╯al., 1998; MNP, 2006; IMAGE, 2010)

N2O, NOx and NH3 emission, N leaching, N€runoff

Extended emission factor approach with consideration of mitigation technologies

All sectors

Europe Global

Country

Present, projections

N emission models to atmosphere GAINS (Höglund-Isaksson and Mechler, 2005; Winiwarter, 2005) http://gains. iiasa.ac.at/ gains/EU/index. login?logout=1

N2O, NOx and NH3 emission

Extended emission factor approach with consideration of mitigation technologies

All sectors

Europe Global

Country

Present, projections

EDGAR (Van Aardenne, 2002) http://edgar.jrc.it

NH3, N2O and NOx emission

Extended emission factor approach with consideration of mitigation technologies

All sectors

Global

1 × 1 degree. The latest version (released 11/2008) is 0.1â•›× 0.1 degree

Past and present

EMEP (Simpson et╯al., 2003, 2006; EMEP, 2010a)

NOx and NH3 emission N deposition

Emissions (disaggregated from official national inventories) and Atmospheric dispersion model

All sectors

Europe

50 km × 50€km; 5 × 5 km possible (e.g. Vieno et╯al., 2009)

Past, present and projections up to 2030

321

Geographical variation in terrestrial nitrogen budgets Table 15.2 (cont.)

Model approach UNFCC/IPCC (IPCC, 2006; UNFCCC, 2010)

Element flux considered N2O (and NOx) emission

Sectors considered

Area involved

Geographic resolution

Emission factor approach on activity data

All sectors

Europe and other ‘Annex-I’ countries (industrialised)

Country

1990– present

Method

Time

N loss models to hydrosphere GREEN (Grizzetti et╯al., 2005, 2008; Bouraoui et╯al., 2009)

Total N diffuse emissions to waters and total N runoff

Geospatial empirical regression model

Agriculture and Point Sources

Europe

Sub-catchments (average size 180 km2)

1985–2005

EPIC (Bouraoui and Aloe, 2007; Van der Velde et╯al., 2009)

NO3, NH4, total N, soluble and particulate N runoff, N€leaching

Detailed mechanistic model

Agriculture, terrestrial systems

EU-27 + Swiss

10 km × 10 km grid (including multiple crops)

1985–2005

a

HSMUâ•›=â•›Homogeneous Spatial Mapping Units; NCU = NitroEurope Calculation Units. Units refer to clusters of 1 km2 grid cells that are characterised by similar environmental and/or agronomic conditions

elements:€(i) the CAPRI-SPAT downscaling model (Leip et╯al., 2008); (ii) the DNDC-CAPRI meta-model (Britz and Leip, 2009b); and (iii) an interface combining results of the DNDCCAPRI meta-model with elements of CAPRI-SPAT, yielding a database with environmental indicators that are inherently consistent and operating at the level of individual crops. These models use the most detailed geographically explicit input data currently available, thus allowing the best way to map the various N fluxes included in the N budget. In particular, the DNDC-CAPRI meta-model is based on detailed spatial information, partly based on biophysical model simulations. A special feature of INTEGRATOR is that it includes historical data up to 1960, thus allowing the assessment of trends in N budgets. Despite the high spatial resolution of the data available in these model systems, results presented in this chapter are mainly restricted to model comparisons at the Europewide scale (tables of complete N budgets) and at the national scale (scatter plots of N fluxes). Detailed maps are limited to N input by manure and fertiliser and to NH3 and N2O, emissions from the agricultural system (both housing systems and soil) as derived by IDEAg and INTEGRATOR. Detailed maps of total N emissions divided in various sectors are further presented in Leip et╯al., 2011a (Chapter 16 this volume).

15.3.1╇ Farm nitrogen budget The IDEAg model system can be used to provide an updated picture of a farm N-budget for Europe. In IDEAg, a combination of the farm budget (animal and crop production in relation with the EU and global market) and soil N budget has been implemented (see Figure 15.1). As explained above, the farm N budget comprises as inputs feed intake and as output animal products, both driven by the economic situation of the farm (i.e. region). The N surplus is exported to manure management systems and finally applied to crops or excreted on

322

grassland by grazing animals (other uses of manure are not significant in Europe and are not considered in IDEAg). IDEAg also calculates the fate of animal and crop products and distinguishes human consumption, processing by the industry to generate feed concentrates, biofuels or other products and, inand export for each commodity considered. Also, losses at the market (and at the farm) are estimated. As a result, the IDEAg system is able to depict a detailed picture of N-flows of the agriculture sector at the European scale.

15.3.2╇ Land nitrogen budgets Detailed land nitrogen budgets at European level An overview of a detailed European (EU27) field scale (land) N budget is presented in Table 15.3. The table compares results derived with INTEGRATOR (De Vries et╯al., 2010) with information from IDEAg (Britz and Leip, 2009a), MITERRA (Velthof et╯al., 2007, 2009) and IMAGE (De Vries et╯al., 2009). Furthermore, the sum of the officially submitted data to the UNFCCC secretariat by the 27 EU countries, as reported in the Annual European Community greenhouse gas inventory, are presented (EEA, 2008). Results include N (NH3, N2O, NOx and N2) emissions from housing systems to give complete emission estimates from the agricultural system. Consequently, we include manure excretion instead of manure application as input to the system. For EU27, the four models estimate a total N input in European agriculture of 23.3–25.7 Mton N yr−1, which is mainly due to fertiliser and animal manure inputs and to a lesser extent caused by atmospheric deposition and N fixation. The N uptake varies from 11.3–15.4 Mton N yr−1 leading to total N surpluses (N input not used by the plants) varying from 10.4 to 13.2 Mton N yr−1 at EU27 level. The lowest surplus is calculated by INTEGRATOR, as it assesses the highest uptake. The various models give in general very similar results

Wim de Vries Figure 15.1 N budget for the agricultural sector in EU27 for the year 2002 as calculated by the IDEAg model.

for the emissions of NH3 (2.8–3.1 Mton N yr−1). Comparable estimates are also derived for the direct N2O emissions Â�(0.33–0.43 Mton N yr–1), but NOx emissions vary by a factor 10 (0.02–0.22 Mton N yr–1). The sum of N leaching and N runoff also varies largely. The estimates by IDEAg and IMAGE are nearly twice as large as the estimates by INTEGRATOR and MITERRA, causing a much lower estimated N2 emission by IDEAg and IMAGE as compared to INTEGRATOR and MITERRA (Table 15.3). An important difference in this context is also that both INTEGRATOR and IDEAg include mineralisation estimates, whereas this input term is neglected in MITERRA and IMAGE. In INTEGRATOR, the net release is mainly determined by the N mineralisation in drained peat soils. In IDEAg, mineralisation of all soils is obtained from the DNDC meta-model and then scaled in two steps (the second jointly with N2 flux estimates) to close the N budget. More details on the N emission sources calculated by the various models are given in Table 15.4. Results show that the difference in NH3 emissions between IDEAg versus the other three models is the result of the higher emissions from housing and manure storage systems. Another notable difference is the much higher N2O and NOx emission from grazing by IMAGE as compared to the other models (Table 15.4). Reasons for the various similarities and differences can be summarised as follows. • All model give similar results for the N inputs by fertiliser as they use the same FAO data regarding fertiliser rates. • Deviations between inputs by manure application are larger due to different sources for animal numbers, but specifically due to deviating N excretion rates. • Differences in biological N fixation mainly follow from the different values used to derive N fixation of pulses/legumes as a fraction of the harvested N amount, as summarised in

the supplementary material chapters 15 & 16 (see also De Vries et╯al. (2010b). • The NH3 emissions by INTEGRATOR, IDEAg and MITERRA are comparable as they are based on the same GAINS dataset. There are however differences in N manure and fertiliser distribution and this affects the N leaching that is affected by soil type, land use, etc. • The difference in N2O emissions is limited on a European wide scale, considering the differences in N2O emission factors used. In INTEGRATOR, these emissions are determined as a function of soil type, land use, manure type, etc. In IDEAg, results are based on the DNDC-CAPRI meta-model, whereas MITERRA uses standard emission fractions based on IPCC. These differences do, however, affect the spatial variation in N2O emissions (see Section 15.3.3). • The higher estimated sum of total N leaching and runoff by IDEAg and IMAGE are mainly due to higher leaching and runoff fractions. In IDEAg, N leaching is based on the DNDC meta-model whereas N leaching by the other models depends on various environmental factors as described in detail in De Vries et╯al. (2011). Apparently, the difference in parameterization of the factors and in geographic resolution leads to strongly different results.

Land nitrogen inputs and nitrogen surplus at country level Land N budgets at country-scale for agriculture for the year 2000 calculated by INTEGRATOR, IDEAg, MITERRA and IMAGE for the various EU countries are presented in the Supplementary materials (Chapter 15 and 16). A scatter diagram of the N inputs as calculated with the INTEGRATOR model compared to IDEAg, MITERRA and IMAGE is given in Figure 15.2. The four approaches generally agree for fertiliser input and N inputs by manure, which is logical as it has the same

323

Geographical variation in terrestrial nitrogen budgets Table 15.3 Annual N budgets of agricultural land in Europe in 2000, including N (NH3, N2O, NOx and N2) emissions from housing systems and from soil. Output terms in italic are summations of more detailed N fluxes and should not be added in the calculation of the total N output

N budget (Mton N yr−1) Source

INTEGRATOR EU 27–2000

MITERRAa EU 27–2000

IMAGEa EU 27–2000

1.0

0.8

1.4

IDEAg EU25–2002

UNFCCb EU27–2002

Input to land Biological fixation

1.3

Manure excretion

10.3

8.8

10.4

9.8

9.1

Synthetic fertiliser

11.5

11.4

11.3

11.3

10.6

2.7

2.1

2.0

2.8

—

25.7

23.3

24.5

25.3

20.8

Atmospheric deposition Total

1.1

Output from land Plant removal

15.4c

12.5

11.3

13.5

—

N accumulation

−3.3

−3.5

—

—

—

2.9

3.1

2.9

2.8

3.1

0.40

0.43

0.33

0.43

0.4

0.21

0.11

0.02

0.22

—

7.0

4.5

7.2

2.5

nd

7.6

5.1

7.8

3.1

—

Emissions of NH3 N2O

d

NO and NO2 N2 Total (De)nitrification N leaching

2.8

5.7

2.0

—

—

0.35

0.4

0.75

—

-

3.1

6.1

2.7

5.9

6.6

Total surplus

10.4

10.8

13.2

11.8

—

Total

25.7

23.3

24.5

25.3

—

N surface runoff Total leaching/runoff

Details of the comparison between MITERRA and IMAGE are described in De Vries et╯al. (2009). Source:€EEA (2008). c Uptake includes the removal from grassland, rough grazing areas and the net crop removal from arable land. d N2O emission refers to direct N2O-N emission only that is calculated by all models. a b

basis although the IDEAg N manure inputs are consistently lower (see also Table 15.4). There are relatively large differences for the other N inputs (deposition and fixation) at country level, but this hardly affects the total N inputs by the four models, which are comparable for all countries. Total N uptake is quite different between the various approaches. As with the results at European scale (see Table 15.4), INTEGRATOR results are consistently higher than the other models. The uptake mostly decreases according to INTEGRATOR > IMAGE > IDEAg > MITERRA. Furthermore, there is quite some scatter at country level. This is reflected in an even larger scatter for the N surplus per country, indicating an uncertainty near 50% for country estimates of the N surplus.

Nitrogen emissions to air and water at country level Instead of quantifying just the gross N surplus, the N excess input can be further defined in terms of N (NH3, N2O, NOx and N2) emissions to the atmosphere, N leaching and N runoff. The N budget models described before can derive such detailed agricultural N budgets not only at European level (see Section 15.2.1), but also at country level. An example of

324

such an output calculation using INTEGRATOR is given in Table 15.5. To gain insight in the comparability of the results obtained, a comparison is given of agricultural emissions of NH3-N, N2O-N and NOx-N and N leaching for 27 EU countries for the year 2000 as derived with INTEGRATOR with those obtained by the complete N budget models (IDEAg, MITERRA and IMAGE). Furthermore, results for the N emissions were compared with standard activity data-emission factors approaches (UNFCC/IPCC, 2010; GAINS, 2010; OECD, 2010; EDGAR, 2010; and EMEP 2010b). Data used for the results of the various models for NH3-N, N2O-N and NOx-N are found in the Supplementary data for Chapter 15. A comparison of country emissions for NH3-N, N2O-N and NOx-N and of N leaching plus runoff (kton N yr−1) within EU 27 as derived with INTEGRATOR with the various other approaches is given in Figure 15.3. Results show comparable estimates for NH3 emissions, which is due to the use of comparable databases for the estimation. Both INTEGRATOR and MITERRA use the N excretion and NH3 emission constants derived by GAINS and consequently, the differences should be

Wim de Vries Table 15.4 Annual N emissions from agriculture in Europe for the year 2000

N emissions in 2000 (kton N yr−1) N source

Emission source

INTEGRATOR EU 27–2000

NH3

Housing and storage

1189

Fertiliser application

1413

b

Manure application

N2O

MITERRAa EU 27–2000

IMAGEa EU 27–2000

1428

1279

1048

678

540

798

759

823

683

Grazing

271

201

231

319

Total agriculture

2873

3066

2873

2848

Housing and storage

55

48

54

52

242

316

208

289

Grazing

124

67

66

92

Indirect emissions

43

80

51

Total agriculture

401 (444)

431 (531)

328 (379)

N application

NO and NO2

IDEAg EU 25–2002

c

d

76 d

434 (510)d

Housing and storage

20

32

36

0

N applicationc

123

16

25

23

Grazing

63

59

32

196

Total agriculture

207

108

93

219

Details of the comparison between MITERRA and IMAGE are described in De Vries et╯al. (2009). b Includes emissions through soil inputs by fertiliser and manure application. c Includes emissions through soil inputs by fertiliser and manure application, deposition, mineralisation, fixation and crop residues. d The value in brackets are the total N2O emissions calculated by INTEGRATOR, IDEAg, MITERRA and IMAGE including also indirect N2O emissions due to N leaching and NH3 and NOx emissions. a

small and are mainly due to the use of different statistics for animal numbers. Furthermore, all models use comparable statistics for N fertiliser use and NH3 emissions from manure. The differences in different N2O emissions, however, are much larger, reflecting the larger variation in model approaches, specifically the use of N2O emission factors. For example, a comparison of INTEGRATOR results with the N2O emissions reported by the EU countries to the UNFCCCIPCC shows quite a disagreement. For MITERRA, there is a good agreement with estimated N2O emission from manure management, and direct soil N2O emission (Velthof et╯al., 2009), since both methods are based on the same N2O emission fractions as a function of N inputs. Deviations between UNFCCC figures and MITERRA are thus only due to differences in activity data and the use of specific emission factors by some countries. By contrast, INTEGRATOR uses emission factors that depend on N source and environmental conditions. In both INTEGRATOR and MITERRA, the estimated indirect N2O emission (not shown here) are much smaller than those reported to the UNFCCC, owing to both a lower N2O emission factor and a lower N leaching fraction. Firstly, the revised IPCC emission factor for N leaching (IPCC, 2006) was used in both INTEGRATOR and MITERRA-EUROPE (i.e. 0.0075 kg N2O-N for each kg N that leaches), whereas the values of the UNFCCC for most countries were obtained using the former emissions factor of 0.025 kg N2O-N per kg N leached (IPCC, 1997). Secondly, IPCC uses a simple method to calculate leaching, i.e. 30% of the total N input via fertiliser, manure, grazing

and other sources leaches to ground water and surface water (Mosier et╯al., 1998). INTEGRATOR and MITERRA use a different approach to calculate N leaching which resulted in leaching losses of 11% of the total N input in EU-27. The NOx emissions appear to be very uncertain (see Figure€15.3). This is in line with results obtained by ButterbachBahl et╯al. (2009), who applied the approach used in IMAGE and three other empirical emission models, using the same input data for all models. More information on that approach and related results is given in the Supporting material in Chapters 15 and 16. The sum of N leaching plus runoff also varies largely within EU 27 and is systematically higher for IDEAg and IMAGE as compared to INTEGRATOR and MITERRA, in line with the results at European level. This implies that the used N leaching factors are highly uncertain and need further refinement.

15.3.3╇ Mapping the European agricultural nitrogen fluxes The national N inputs and N outputs presented in Section 15.3.2 do not show the regional differences in N fluxes. In this section we provide maps showing such differences, focusing on presentations with IDEAg and INTEGRATOR for agricultural ecosystems in EU-27 for the year 2000. These two models were used to illustrate the geographic variation in model results, because of their highly disaggregated model input data. With respect to the emission of greenhouse gases, such as N2O, it is

325

Geographical variation in terrestrial nitrogen budgets

Figure 15.2 A comparison of country N inputs by fertiliser, manure, other inputs (deposition and fixation), total N inputs, total net N uptake and N surplus within EU27 as derived with INTEGRATOR, IDEAg, MITERRA and IMAGE for the year 2000 (IDEAg is 2002).

crucial to know whether total emissions for the area considered are correct, whereas accurate information on the spatial distribution of the emissions is less relevant. The latter aspect is, however, crucial when assessing the risk of elevated NH3 emissions, and related N deposition, and of N leaching and

326

N runoff in view of eutrophication impacts on terrestrial and aquatic ecosystems. Here, aggregation of input data for large areas may cause accurate average N deposition and N leaching levels, but a strong deviation in the area exceeding critical N deposition loads or critical N concentrations in ground water

Wim de Vries Table 15.5 N emissions to air and water calculated at country level with INTEGRATOR for the year 2000

N output fluxes (kg N ha−1 yr−1) Countrya

Area (Mha)

Emission NH3

Emission N2O

Emission NOx

Emission N2

Leaching + runoff

Austria

3.336

13.2

2.1

0.9

23.1

11.1

Belgium

1.779

41.6

5.6

2.2

70.3

32.6

Bulgaria

6.816

4.7

0.9

0.4

16.3

4.4

Czech. Rep

4.776

11.1

2.5

1.0

38.7

18.4

Denmark

3.273

19.9

1.8

0.9

27.5

25.4

Estonia

1.846

3.8

1.1

0.5

25.5

5.4

Finland

6.914

2.0

0.4

0.1

17.6

4.6

France

35.346

14.8

2.6

1.2

30.8

13.7

Germany

21.566

20.2

2.4

1.1

42.4

19.5

Greece

8.404

6.2

1.2

0.7

20.9

10.4

Hungary

6.739

8.6

1.5

0.7

38.1

10.1

Ireland

5.043

15.5

5.4

2.8

37.3

11.5

18.434

17.6

1.9

0.9

33.3

19.1

Latvia

3.343

2.7

0.6

0.3

15.9

6.0

Lithuania

4.246

5.4

1.2

0.5

32.5

17.9

Luxembourg

0.144

20.8

6.9

0.0

34.6

13.9

Netherlands

2.491

52.6

4.8

2.4

94.3

45.0

Italy

Poland

20.265

10.6

1.3

0.4

32.0

15.6

Portugal

5.411

8.1

1.1

0.6

30.1

14.4

Romania

14.517

7.9

1.1

0.5

23.8

8.1

Slovakia

2.664

9.4

1.5

0.8

21.4

13.5

Slovenia Spain Sweden UK

0.779

20.5

2.6

1.3

28.2

12.8

35.027

7.2

0.8

0.4

16.2

7.6

7.914

4.2

0.6

0.3

12.9

5.8

16.237

15.2

4.1

1.8

39.1

15.3

EU-27

237.310

EU-27b

237.310

a b

12.1 2873

1.9 444

0.9 207

29.3 6965

13.2 3136

Data for Cyprus and Malta are not included. Data given in kton N yr-1.

and surface water (De Vries et╯al., 2009). For this reason, it is relevant to make use of models with the highest level of spatial detail with respect to inputs and outputs, such as IDEAg and INTEGRATOR. The datasets mentioned in the Supplementary materials in Chapters€15 and 16 in combination with various downscaling techniques have been used to ‘regionalise’ the agricultural N inputs from statistical data at national or subnational level to the NCU or HSMU level.

Nitrogen inputs Inputs by manure and fertiliser Input of mineral N fertiliser and manure N as derived by IDEAg and INTEGRATOR are shown in Figure 15.4. The legend of 170 kg N is chosen as this is the maximum allowed manure N input in the EC, with

the exception of a derogation (accepted after 2000) of 250 kg N for the Netherlands and 230 kg N for Denmark, Germany and Austria. High manure N application rates occur in areas of high livestock density in Europe and include parts of Denmark, the Netherlands, Belgium, Wales, Ireland, Catalonia and Galicia in Spain, and the north of Italy. Regions of high N fertiliser input can be identified in most intensive agricultural areas in Europe, again including Denmark, Belgium, the Netherlands, UK and Ireland, Brittany (France) and the Po Valley (Italy). Results show that an exceedance of the N manure input of 170 kg N occurs mainly in various dense livestock population areas, such as the Netherlands, where even the derogation of 250 kg N is often exceeded in the year 2000. There is a

327

Geographical variation in terrestrial nitrogen budgets

(a) NH3-N emission agriculture (kton.yr –1) 800

IDEAg Miterra

600

IMAGE GAINS

400

EDGAR 200

0

200

0

600

400

800

NH3-N emission agriculture INTEGRATOR

(b) N2O-N emission agriculture (kton.yr –1) 100

IDEAg Miterra

80

IMAGE

60

GAINS

40

EMEP

20

OECD

0

EDGAR UNFCC/ IPCC 0

20

40

60

80

100

N2O-N emission agriculture INTEGRATOR (kton.yr –1)

(kton.yr –1)

(c) NOx-N emission agriculture (kton.yr –1)

(d) N leaching (kton.yr –1) 1500

70 IDEAg

60 50

Miterra

40

IMAGE

30

EDGAR

20

IDEAg

1000

Miterra IMAGE

500

EMEP

10

0

0 0

10

20

30

40

50

60

70

NOx-N emission agriculture INTEGRATOR (kton.yr –1)

0

500

1000

N leaching INTEGRATOR

1500

(kton.yr –1)

Figure 15.3 A comparison of country emissions for NH3-N, N2O-N and NO-N and of the sum of N leaching and runoff for the year 2000 within EU 27 as derived with INTEGRATOR and with various other model approaches (IDEAg, MITERRA, IMAGE, GAINS, EDGAR, EMEP and UNFCC/IPCC).

clear difference between IDEAg and INTEGRATOR in western France, where the latter model calculates much higher N manure inputs. The reason for this difference is seemingly a different disaggregation of animal numbers. In general N application by mineral fertiliser is higher in IDEAg, specifically in Western Europe, but also in the Nordic countries where it is possibly an artefact due to division of N inputs by very small areas of agricultural land (Figure 15.4a, b). Inversely, N application by animal manure, including grazing, is generally higher in INTEGRATOR, except for parts of the Netherlands and Denmark. INTEGRATOR shows hot-spots, e.g. in parts of France and Eastern Europe that are not resulting from IDEAg (Figure 15.4c, d). A comparable picture for the estimated N inputs by mineral fertilisers and animal manure for the year 2000 in EU25 is given by Grizzetti et╯al. (2007), using a 10 km × 10â•›km resolution. Details on the approach, combining agricultural statistics on administrative basis and geographic land cover information, are given in Grizzetti et╯al. (2007).

328

NH3 and N2O emissions Calculations by both INTEGRATOR and IDEAg show that the regional variation in total NH3 and N2O emissions is large (Figure 15.5). Hot spots are located in areas with intensive animal husbandry in the Eastern and central part of Ireland, in England and Wales, in the Netherlands, Belgium, Denmark, in north-western and southern Germany, in the north of Italy and in the Catalonia region in Spain. In general, NH3 emissions calculated by IDEAg are higher than by INTEGRATOR in Western and Central Europe, but the reverse is true for the Nordic countries (Figure 15.5a, b). Inversely, N2O emissions calculated by IDEAg are higher everywhere, specifically in the Nordic countries, where the high emissions might be an artefact of the extremely high N fertiliser input but lower in the UK and Ireland (Figure 15.5c, d). The variation in NH3 and N2O emissions is in general comparable with the geographic variation in N surpluses, which in turn are strongly related to the variation in manure N inputs. The high correlation between

Wim de Vries

(a) N fertiliser (kg N/ha/yr) 0 - 25 25 - 50 50 - 100 100 - 170 170 - 250 >250

(c) N manure (kg N/ha/yr) 0 - 25 25 - 50 50 - 100 100 - 170 170 - 250 >250

(b) N fertiliser (kg N/ha/yr) 0 - 25 25 - 50 50 - 100 100 - 170 170 - 250 >250

(d) N manure (kg N/ha/yr) 0 - 25 25 - 50 50 - 100 100 - 170 170 - 250 >250

Figure 15.4 Nitrogen application from mineral fertiliser (a, b) and manure, including grazing (c, d) in the year 2000 in EU-27. Calculation with IDEAg on the geographic resolution of HSMUs (left) and with INTEGRATOR on the geographic resolution of NCUs (right). Grey shading in the EU-27 denotes non-agricultural areas. Countries outside EU-27 are also included by a grey shade.

329

Geographical variation in terrestrial nitrogen budgets

(a) NH3 emission (kg N/ha/yr) 0-5 5 - 10 10 - 15 15 - 20 20 - 40 >40

(c) N2O emission (kg N/ha/yr) 0-1 1-2 2-4 4-6 6-8 >8

(b) NH3 emission (kg N/ha/yr) 0-5 5 - 10 10 - 15 15 - 20 20 - 40 >40

(d) N2O emission (kg N/ha/yr) 0-1 1-2 2-4 4-6 6-8 >8

Figure 15.5 Total NH3 emissions (a, b) and N2O emissions (c, d) from agriculture in the year 2000 in EU-27. Calculation with IDEAg on the geographic resolution of HSMUs (left) and with INTEGRATOR on the geographic resolution of NCUs (right). Grey shading in the EU-27 denotes non-agricultural areas. Countries outside EU-27 are also included by a grey shade.

330

Wim de Vries

N surplus, being a main driver for N emissions and manure application is illustrated in detail by Leip et╯al. (2011b).

Nitrogen losses to ground water and surface water Nitrogen losses to either ground water or surface water can be achieved using models, which include the major N inputs and the main processes of N transport and transformation, including surface runoff (overland flow) and runoff (interflow) to surface water and leaching to ground water. Various models have been developed and applied to address the issue of N fate in the river basin, and they vary for process description, scale of study and data requirement (http://euroharp. org). On a European wide scale, both detailed (EPIC) and simple process based models (INTEGRATOR, IDEAg) and statistical models (GREEN) are available (see Table 15.2). Here, we show results derived with both INTEGRATOR and IDEAg and with GREEN. The estimated regional variation N losses from soil to both ground water and surface water in 2000 as derived with IDEAg and INTEGRATOR is given in Figure 15.6. It should be emphasised that INTEGRATOR estimates are only slightly influenced by meteorological data, since the model uses N leaching fractions that depend on soil type, land use, soil organic content, precipitation surplus, temperature and rooting depth (Velthof et╯al., 2009). In IDEAg, however, N leaching from soils is based on the DNDC-CAPRI meta-model (Britz and Leip, 2009a), which in turn is derived from CAPRI-DNDC model simulations using meteorological

data to asses water fluxes and related N leaching fluxes. In this context, use is made of the JRC-MARS database, being a spatial interpolation of more than 1500 weather stations across Europe onto a 50 km × 50 km grid (Orlandi and Van der Goot, 2003). In line with Table 15.4, results obtained by IDEAg show a much higher N leaching rate all over Europe, as compared to INTEGRATOR. Most likely, the N leaching by IDEAg is an overestimation, since there is a reasonable comparison between measured NO3 concentrations in ground water and those estimated by the MITERRA model, being the agricultural module in INTEGRATOR in an adapted form (see Section 15.5.1 on model evaluation). Figure 15.7 (left) shows an estimate of N diffuse losses to surface water for the year 2000 for Europe (Grizzetti et╯al., 2008; Bouraoui et╯al., 2009), based on the GREEN model taking into account N sources, river network and climate conditions. According to these estimates, the regions affected by higher N losses to surface waters include Belgium, the Netherlands, the Po Valley (Italy), the Brittany region (France), which are already totally or partially designated as Nitrates Vulnerable Zones (Nitrate Directive). Figure 15.7 (right) shows the estimated N source apportionment per sub-catchment for Europe for the year 2000. This map provides a picture of the relative contribution of diffuse sources (mainly agriculture) and point sources (mainly urban settlements) to the water N pollution. According to these estimates, agriculture is the main

Figure 15.6 Regional pattern of N leaching plus runoff in the year 2000 in EU-27 based on calculations with IDEAg on the geographic resolution of HSMUs (left) and with INTEGRATOR on the geographic resolution of NCUs (right). Grey shading in the EU-27 denotes non-agricultural areas. Countries outside EU-27 are also included by a grey shade.

331

Geographical variation in terrestrial nitrogen budgets

N Arctic Ocean

N Arctic Ocean

N Diffuse emissions from agriculture (kg N/ha) 0–2

0 – 25

3 – 10

26 – 50 51 – 75

11 – 20 21 – 30

Atlantic Ocean

76 – 85

Atlantic Ocean

86 – 100

> 30 North Sea

North Sea

English Channel

English Channel

Black Sea

Bay Of Biscay

250

500

1,000 Kilometers

Black Sea

Bay Of Biscay

Mediterranean Sea

0

N Diffuse sources (%) (per sub-catchment)

Mediterranean Sea

0

250

500

1,000 Kilometers

Figure 15.7 Regional pattern of N loads to surface water as diffuse emissions (left) and N source apportionment (right) for Europe in year 2000, based on calculations with the GREEN model on a geographic resolution of sub-catchments (average size 180 km2).

contributor of N for surface waters in most of the river basins, while in Mediterranean catchments point sources have a relative higher contribution, which is probably due to a less effective implementation of waste water treatments and the lower precipitation and thus N losses to surface waters.

15.3.4╇ Trends in nitrogen fluxes since 1970 Trends in N fluxes since 1970 up to the year 2000 are derived on the basis of INTEGRATOR using the following. • Data on N fertiliser use, animal numbers and crop yields from the FAO database. • Scaled N excretion rates to those used for 2000 on the basis of RAINS/GAINS data. The scaling is based on a simple N excretion model described by Witzke and Oenema (2007), using the milk production as a scaling factor for dairy cattle and the meat production as a scaling factor for other cattle, pigs and poultry. Data on the milk and meat production per country in the period 1970–2000 were taken from the FAO database. • N deposition history based on historical NOx emissions by EMEP and NH3 emissions by INTEGRATOR, while adding non-agricultural sources from IMAGE and using an emission-deposition matrix based on the EMEP model (EMEP, 2009). • Constant N fixation rates for the grassland and arable land, but using FAO data on trends in the area of dry pulses and soy beans, mainly affecting N fixation. Information on trends in data for alfalfa and clovers, affecting the estimate for biological fixation by grasslands are missing and consequently we assumed no trends in N fixation rates by grassland. • Scaled N contents in crops, based on a change in N availability (this is automatically calculated in INTEGRATOR).

332

• Trends in NH3 emission factors in view of changes in housing systems and manure application techniques. For the year 2000, GAINS data are used for the fraction of housing systems and manure application techniques with high, medium and low emissions per country. For the period 1970–1980, we assumed that all emission fractions were high and in the period 1980–2000, we assumed a linear interpolation from high emissions to the present emission percentage. Note that the available data on both crop yields and N fertiliser use in the FAO databases include trends in N use efficiency, which is mostly defined as the crop yield divided by the N input by fertiliser (Bouwman et╯al., 2005). Results derived by INTEGRATOR for the trends in all N inputs, N surplus and N outputs, in terms of N emissions to the atmosphere and N leaching to ground water and surface water, for the period 1970–2000 are given in Figure 15.8. The results show a steady increase of N inputs by fertilisers in the period 1970–1985, followed by a decrease since then, mainly in response to the increased or decreased crop production in those periods (or vice versa). Despite a slight decrease in cattle, the N input by manure excretion has increased up to 1985 due to an increase in N excretion rates, related to an increase in milk production, followed by a slight decrease in response to the decrease in livestock and the relatively constant excretion rates. The trend is also influenced by the increase in pigs anol poultry between 1970–2000 (see Oenema et╯al., 2007), but the dominant effect is that of changes in N excretion rates by dairy cattle. There is a more clear increase in the average N input in agricultural systems than in the total N input, due to a decrease in agricultural area. This holds also for the trends in the total N uptake and the related N surplus for the period 1970–2000. Results show a slightly declining trend in NH3 emission in response to a decline in livestock since 1990, but the trends in N2O and NOx emissions and N leaching are almost constant.

Wim de Vries

(a)

(b)

Balance (kg N.ha–1.yr–1)

(Input (kg N.ha–1.yr –1)

150

150 Animal manure Fertiliser 100

100

Deposition

Total input

Fixation 50

Uptake

Mineralisation

50

Surplus

Total 0 1970 1975 1980 1985 1990 1995 2000

0 1970

1975

1980

1985

Year

1990

1995

2000 Year

(c) Emission and leaching (kg N.ha–1.yr–1) 40 NH3

30

N2O NOx

20

N2 10

0 1970

Leaching

1975

1980

1985

1990

1995

2000 Year

Figure 15.8 Trends in average N inputs (a), and average N surplus (b) and average N outputs (c) at EU-27 level for the period 1970–2000 as estimated by INTEGRATOR. NB:€leaching stands for leaching plus runoff.

Trends in N2 emissions, being most uncertain, are clearly increasing up to 1985 and declining afterwards (Figure 15.8).

15.4╇ Land nitrogen budgets for non-agricultural systems 15.4.1╇ Detailed land nitrogen budgets at European level An overview of the land N budget for all terrestrial non�agricultural systems (forests and semi-natural vegetations) at the European scale (EU-27) as calculated with INTEGRATOR is given in Table 15.6. For non-agricultural systems, there is no differentiation between land and soil N budgets as all fluxes are related to the soil system. N deposition is derived with an emission deposition matrix, using NOx and non-agricultural NH3 emissions from EMEP and NH3 emission estimates from agriculture by INTEGRATOR as inputs. The N manure input to semi-natural vegetations is mainly due to rough grazing, but it also includes some manure application being calculated in the MITERRA sub-model of INTEGRATOR. For forests, rough

grazing is assumed to be negligible. Net N immobilisation (accumulation) in both forests and semi-natural vegetations is calculated as a fraction of the net N input, which is dependent on the C/N ratio of the soil, using an approach described in De Vries et╯al. (2006). NH3 emissions in forests are background emissions due to wild animals derived from Simpson et╯al. (1999), whereas the NH3 emission from short vegetations is calculated as a fraction of the N manure input by grazing animals. In forests, the estimated N2O, NO and N2 emissions by INTEGRATOR are derived with a statistical relationship with environmental factors based on results of a European wide application of the process oriented biogeochemical model Forest-DNDC (Li et╯al., 2000) by Kesik et╯al. (2005). Apart from this meta-model of Forest-DNDC, INTEGRATOR includes an empirical relationship with various environmental factors, based on hundreds of measurements assessed in the literature (Bloemerts and de Vries, 2009). In short vegetations, the N2O and NO emissions are calculated as a fraction of the N input, using emission factors that are a function of N source, soil type, pH, precipitation and temperature (see Supplementary materials Chapter 15 and 16). Finally, N leaching is assessed by multiplying the net N input by an N leaching factor and N2 emissions

333

Geographical variation in terrestrial nitrogen budgets

to the size of the country. There is also a large uncertainty in the N flax, specifically in the N2O and NOx emissions, as discussed below by comparing results of various model approaches.

Table 15.6 The annual N budget of forest soils and semi-natural vegetation (EU 27) in Europe in 2000, as derived with INTEGRATOR.

N budget (kton N yr−1) Semi-natural vegetations

Total nature

1003

1003

1367

345

1712

271

214

485

1638

1562

3200

Net uptake

302

779

1081

N accumulation

729

−26

703

Source

Forests

Inputs Manure input (grazing) Deposition Biological N fixation Total

—

Outputs

0

Emissions of

0

NH3

21

221

242

N2O

45

37

82

NOx

13

18

31

256

431

687

N2 N leaching + runoff Total

272

113

385

1638

1572

3210

are then calculated as N input minus all N output terms. In forests, N2 emission is already calculated and N leaching is calculated as all N€input minus all N output terms. The results show that while the total N input is comparable in forests and semi-natural vegetations, N deposition dominates the N input in forests, whereas manure input by grazing animals dominates the N input in semi-natural vegetations. This high manure input also causes a much larger NH3 emission in semi-natural vegetations as compared to forests. Compared to semi-natural vegetations, net N uptake and N2 emissions are lower in forests, whereas N accumulation (net N immobilisation) and N leaching are higher. In semi-natural vegetation, net N growth uptake is set equal to N excretion by grazing animals, since these animals continually remove the vegetation, but also excrete nearly the same amount on the field. In percentage of the N surplus (N input minus N uptake), the N leaching and runoff is approximately 20% from forests and 8% from seminatural vegetations, being (much) lower than the default IPCC factor of 30%.

15.4.2╇ Nitrogen budgets at country level and regional level N budgets calculated at country level An overview of the N budget for forests for the EU-27 countries, based on INTEGRATOR results, is given in Table 15.7. In this table, removal refers to the net N removal due to wood harvesting and accumulation stands for the N pool change in the soil. Results show large variations in all N fluxes, related partly

334

N2O emissions and NO emissions at country level and regional level A comparison of the results per country by the original Forest-DNDC model with those obtained by the meta-model in INTEGRATOR is presented in Figure 15.9. For regionalisation purposes, Forest-DNDC was coupled to GIS with a resolution of 50â•›km by 50 km holding all relevant information for initialising (soil and forest stand properties) and driving the model (atmospheric input, daily meteorological data). Before application of Forest-DNDC on a regional scale, the model was evaluated for its suitability by applying it to different field sites of the NOFRETE project, which were well distributed across Europe. For further details, we refer to Kesik et╯al. (2005). Results of INTEGRATOR are based on the application of meta-models for N2O and NO from DNDC at NCU level, while making checks on the N balance. We checked whether the N input by deposition and fixation, minus the net N uptake by trees, minus the calculated total N emission and N immobilisation is above a minimum N leaching rate (near zero kg N). If this is not the case, both N emission and N immobilisation are reduced, assuming that these terms are more uncertain than the estimated N deposition and N uptake. Only in cases where zero N emission and N immobiliÂ� sation still leads to a leaching rate below the minimum value, the N fixation is increased. The rationale behind this check is that in low N input systems, where trees take all the N to maintain growth, there is not enough N€available for N emissions, unless there is net N mineralisation (e.g. drained forest on peat soils). The results with the meta-model for N2O are quite comparable with the original DNDC model (Figure 15.9), except for two countries (Sweden and Finland), where the original DNDC model predicted an N2O emission of 11.9 and 10.5 kton N yr−1, whereas the meta-model predicted an N2O emission of 0.7 and 2.3 kton N yr−1, respectively. This large difference is due to the check on the N balance. In these Nordic countries with low N inputs, N is simply not available for large N2O emissions. The results with the meta-model for NOx are generally lower than the original model and this holds again specifically for Sweden and Finland but also for other countries such as Germany and France. Apart from the N balance checks, the differences are also due to the large dependence of the NOx emissions on soil properties, such as pH, being differently used in the INTEGRATOR meta-model application that in the original DNDC model. Regional patterns of the N2O and NO emissions for forests calculated with INTEGRATOR are presented in Figure 15.10. Regional patterns obtained with Forest-DNDC are presented in the supporting material in Chapters 15 and 16, showing higher N2O and NO emissions calculated by Forest-DNDC, as compared to INTEGRATOR, in the Nordic countries for Â�reasons given above.

335

−1

Data given in kton N yr .

135.342

a

EU-27a

1.773

135.342

EU-27

United Kingdom

2.583

Romania

1.895

9.117

Portugal

Sweden

0.306

Poland

1.115

2.658

Netherlands

Spain

0.091

Luxembourg

6.757

1.804

Lithuania

25.247

8.195

Latvia

Slovenia

0.300

Italy

Slovakia

1.717

11.243

France

Ireland

2.036

Finland

3.369

0.358

Hungary

10.261

Denmark

Estonia

19.606

2.549

Czech. Rep.

14.636

3.418

Bulgaria

Germany

0.611

Greece

3.698

Belgium

Area (Mha)

Austria

Country

1367

10.1

10.5

69.8

66.8

0.8

4.1

28.0

1.6

424.3

3.5

20.5

9.9

2.7

433.6

1.4

6.0

1.7

11.7

15.2

37.3

41.8

0.5

18.0

9.4

21.5

16.5

Deposition

271

2.0

2.0

26.7

20.2

0.1

0.6

5.2

0.6

59.6

0.2

2.0

2.0

0.6

54.7

0.3

1.0

0.5

1.0

2.6

19.3

11.4

0.1

2.0

2.0

2.0

2.0

Fixation

1638

12.1

12.5

96.4

87.0

0.8

4.7

33.2

2.2

483.8

3.7

22.5

11.9

3.4

488.3

1.8

7.0

2.2

12.7

17.8

56.5

53.1

0.5

20.0

11.4

23.5

18.5

Total

N input fluxes (kg N ha−1 yr−1)

302

2.2

2.2

20.9

26.9

0.1

0.8

9.7

0.0

40.8

0.4

5.1

1.6

0.8

121.7

0.4

1.2

0.1

1.6

3.4

13.5

13.4

0.1

3.2

0.4

4.1

4.3

Removal

Table 15.7 N budgets calculated at country level for forests with INTEGRATOR for the year 2000

729

5.4

3.2

39.9

23.6

0.4

2.3

10.4

1.3

248.5

2.1

9.9

5.1

1.2

251.2

0.2

2.8

1.2

6.7

7.5

22.4

9.0

0.3

10.9

5.4

11.5

8.1

Accumulation

21

0.15

0.35

0.93

1.12

0.00

0.02

0.36

0.07

5.97

0.05

0.13

0.00

0.00

7.05

0.01

0.11

0.02

0.24

0.19

0.54

0.21

0.03

0.20

0.26

0.37

0.12

Emission NH3

45

0.33

0.13

5.38

4.40

0.00

0.03

1.13

0.23

12.61

0.08

0.39

0.54

0.14

6.79

0.03

0.18

0.06

0.13

0.54

1.91

1.48

0.01

0.02

0.18

0.31

0.03

Emission N2O

13

0.10

0.01

1.09

0.25

0.00

0.00

0.02

0.21

10.57

0.14

0.09

0.27

0.06

0.47

0.00

0.01

0.00

0.14

0.05

0.19

0.21

0.00

0.01

0.00

0.23

0.01

Emission NOx

N output fluxes (kg N ha−1 yr−1)

256

1.9

3.0

26.2

27.4

0.1

0.5

7.1

0.2

36.2

0.0

2.6

1.4

0.4

68.8

0.2

1.3

0.6

0.8

3.0

8.3

8.4

0.0

0.9

2.8

1.9

2.5

Emission N2

272

2.0

3.6

2.0

3.2

0.2

1.0

4.5

0.2

129.2

1.0

4.2

3.0

0.8

32.3

0.9

1.4

0.2

3.1

3.2

9.6

20.4

0.1

4.7

2.3

5.1

3.5

Leaching + Runoff

Geographical variation in terrestrial nitrogen budgets

Figure 15.9 A comparison of country emissions for N2O emissions (left) and NOx emissions (right) from forests in EU 27 for the year 2000, calculated by DNDC, as estimated by Kesik et╯al. (2005), and calculated with INTEGRATOR.

Figure 15.10 Regional pattern for the emissions of N2O (left) and NO (right) from forest soils in EU 27 in the year 2000 as derived with INTEGRATOR. Grey shading in the EU27 denotes non-agricultural areas. Countries outside EU27 are also included by a grey shade.

Nitrogen losses to ground water and surface water The geographic variation in estimated NO3-N leaching and runoff from forest soils and short vegetations (with rough grazing) in 2000, as derived with INTEGRATOR, is shown in Figure 15.11. In line with the high N deposition inputs, N leaching below forests is high in the Netherlands and Germany and low in the Nordic countries and in Spain. In the

336

Nordic countries, N leaching does not reflect the N deposition pattern, mainly due to impacts of temperature. In the north, growth is very limited owing to low temperatures, this leading to extremely low N uptake rates. N leaching from seminatural vegetations reflects the high N manure input regions due to rough grazing, mainly occurring in western UK and central Europe.

Wim de Vries

Figure 15.11 Regional pattern of N leaching and surface runoff to ground water and surface water (left) and semi-natural vegetation (right) from forest soils in EU-27 in the year 2000 as estimated by INTEGRATOR. Grey shading in the EU-27 denotes non-agricultural areas. Countries outside EU-27 are also included by a grey shade.

15.5╇ Discussion and conclusions 15.5.1╇ Model evaluation Comparability of model results In general, results of various N budget models (INTEGRATOR, IDEAg, MITERRA and IMAGE) in terms of annual N inputs and N fluxes on a European (EU27) wide scale are reasonably comparable for the year 2000. This holds specifically for N fertiliser inputs that are all based on the same source and to a lesser extent for N manure input where livestock sources are mostly comparable, but where N excretion rates differ. Despite the overall comparability, the estimated geographic variation in N inputs differs considerably between models. A comparison of agricultural emissions of NH3-N, N2O-N and NOx-N for all the 27 EU countries as derived with the four complete N budget models and standard activity data-emission factors approaches (UNFCC/IPCC, GAINS, OECD, EDGAR and EMEP) also shows comparable estimates for NH3. The differences in N2O emissions, however, are much larger, while NOx emissions are most uncertain. This holds both on a European wide scale and with respect to the geographic variation in the emissions. Very uncertain are also the N leaching and runoff estimates, which show a very large deviation between models. This holds both for the European wide estimates and for the geographic variation. Most uncertain are also N2 emissions that are often calculated as a rest term from all other N inputs and outputs in

a budget approach. It is important to mention that this seemingly simple compound is almost not measurable and model results are quite speculative as they cannot be validated. The N2 release can be derived from radioactive labelling and there are only a handful of studies focusing on N2 measurements. In view of a complete N budget, it would be worthwhile to put more emphasis on the measurement of N2.

Comparison of results with inverse modelling results for nitrous oxide emissions Inverse modelling is an important tool for regional emission estimates and independent verification of international agreements on emission reductions, such as the United Nations Framework Convention on Climate Change (UNFCCC) and the Kyoto Protocol (IPCC, 2001; Bergamaschi, 2007). Atmospheric measurements combined with inverse atmospheric models can provide independent ‘top-down’ emission estimates of atmospheric trace gases. Inverse modelling has been widely applied for CO2 and CH4 (IPCC, 2007), while only relatively few studies are available for N2O. The first inverse analysis of the global N2O cycle was presented by Prinn et╯al. (1990) based on a 9-box model and atmospheric observations from the ALE-GAGE network for 1978–1988. They concluded that beside the use of fertiliser and fossil fuel combustion in mid latitudes, tropical sources (probably from tropical land use change) are likely to play an important role for the global budget and the observed N2O increase (32%–39% for 1978–1988). The more recent studies of Hirsch et╯al. (2006)

337

Geographical variation in terrestrial nitrogen budgets Table 15.8 N2O emissions for the year 2000 for Ireland and UK and for Western Europe as derived with INTEGRATOR and based on results by the 222Rn tracer method and the inverse model NAME (after Messager et╯al., 2008)

N2O emissions (kg N2O-N ha−1 yr−1)

Area Ireland + UK Western Europe

a

INTEGRATOR

Rn Tracer method

Inverse model NAME

Agriculture

Totalb

8.3–9.8

9.0–11.1

6.8

11.2

6.6–8.9

7.5–10.2

4.7

7.7

222

T he sum of emissions from France, Germany, the Netherlands, Belgium, Luxembourg and United Kingdom. b Total emissions by INTEGRATOR were derived by multiplying the agricultural N2O emissions with the ratio of total/agricultural N2O emissions based on GAINS. a

and Huang et╯al. (2008), based on 3D global inverse models suggest an even larger contribution of the tropical sources between 0 and 30oN. First inverse modelling estimates of European N2O emissions were provided by Ryall et╯al. (2001) and Manning et╯al. (2003), using N2O observations from Mace Head and the NAME Lagrangian particle model. Their estimates for North West European countries showed an agreement within ~30% or better with emissions reported to UNFCCC. Another example is downscaled emissions for parts of Europe based on the NAME model and a model-independent approach using the 222Rn tracer method, presented by Messager et╯al. (2008). A comparison of N2O emissions derived by INTEGRATOR with those estimates is given in Table 15.8. Results show that the comparison is reasonable. It needs to be emphasised, however, that top-down approaches generally estimate total emissions, while emission reported to UNFCCC cover only anthropogenic emissions. Hence, for quantitative comparisons good bottom-up estimates of the natural sources are needed. While the above European top-down emission estimates are based on one single station only (Mace Head), improved emission estimates require the use of further atmospheric measurements, to provide a better coverage of the European domain. Additional continuous N2O measurements are now available from the European RTD project CHIOTTO (‘Continuous HIgh-precisiOn Tall Tower Observations of greenhouse gases’) for 2006, which has set up a European network of tall towers for GHG measurements. The measurements from the CHIOTTO towers and further monitoring stations are currently used in the NitroEurope project to provide European N2O emission estimates using five independent inverse models. A particular challenge constitutes the fact that measurements from different stations / networks may have small calibration offsets, hence requiring sophisticated bias correction procedures in the inverse modelling systems. Results from the NitroEurope inverse modelling will be available early 2011. There are also great opportunities for constraining NH3 or NOx emissions by independent datasets based on wet concentration measurements and satellite measurement (Gilliland et╯al., 2003; Konovalov et╯al., 2010). Whenever such datasets come available, they will be used for independent model validation.

338

Comparison of results with measurements for nitrate concentrations in ground water and N concentrations in surface water Use was made of data on NO3 concentration measurements in groundwater in the period 2000–2003 (EC, 2007) to validate the results of the MITERRA-Europe model. The measurements of NO3 concentration showed that 17% of EU-27 monitoring stations had NO3 concentrations above 50 mg NO3 l−1, 22% were in the range of 25 to 50 mg NO3 l−1 and 61% of the groundwater stations had a concentration below 25 mg NO3 l−1 (EC, 2007). A preliminary validation of the MITERRA model on these NO3 concentration measurements (Velthof et╯al., 2009) showed that the distribution of calculated mean NO3 concentrations in NUTS2 regions of EU-27 according to MITERRAEUROPE agrees very well with the distribution of the means of measured NO3 concentrations in the EU-27. For the year 2000, MITERRA estimates that 16% of the NUTS2 regions had NO3 concentrations above 50 mg NO3 l−1, 20% were in the range of 25 to 50 mg NO3 l−1, and 65% had a concentration below 25 mg NO3 l−1. The calculated NO3 concentrations were also in the same range of the means of measured NO3 concentrations in groundwater bodies. For Belgium, Czech Republic, Denmark, the Netherlands and Poland, the calculated NO3 concentrations appear somewhat higher than the measured NO3 concentrations. Possible reasons for these apparent differences are that monitoring stations measure NO3 concentrations at various depths, while MITERRA-EUROPE estimates NO3 concentration in the soil water at uniform depth (below rooting zone). Moreover, monitoring stations may include forests and natural land, whereas MITERRA-EUROPE only calculates NO3 concentration for agricultural land. Finally, it has to be realised that the model results refer to the NO3 concentration in leachate to ground water and not to the concentrations in ground water as measured in the ground water stations.

15.5.2╇ Nitrogen budgets and effects on ecosystems There is an increasing demand by policy makers for easy to interpret and understand indicators that assess the environmental performance and ‘sustainability’ of agriculture. Results presented before thus need to be interpreted in view of possible

Wim de Vries Table 15.9 Variation in number of countries with estimated NH3-N emissions in 2000 exceeding the National emission ceilings for 2010, depending on the model approach

Model

Number of countries exceeded

Percentage countries exceededa

Total exceedance kton NH3-N yr−1

INTEGRATOR

7

28

103

IDEAg

7

28

264

MITERRA

9

36

75

IMAGE

7

29

167

GAINS

10

40

109

EDGAR

18

72

1269

EMEP

14

56

261

OECD

12

63

245

a

T he countries included in the calculation were 25 (EU27 minus Cyprus and Malta) for INTEGRATOR, MITERRA, GAINS, EDGAR and EMEP, 24 for IDEAg and IMAGE and 19 for OECD. The percentage equals the number exceeded divided by these totals.

effects to be of use in policy making. Below, we discuss various options for performance indicators, based on either gross or detailed N budget approaches, including the exceedance of the following. • Maximum N manure inputs and NH3 emission ceilings. Note that these are policy criteria based on impacts but not critical levels related to actual impacts. • Critical NH3 concentrations and critical N loads in view of biodiversity impacts and in view of elevated N saturation of forest soils, associated with damage by plagues and diseases. • Critical NO3 concentrations in ground water in view of health effects and critical N concentrations in surface waters in view of eutrophication of terrestrial ecosystems. The assessment is focused on the year 2000. Trends in the changes of risks can be derived from the trends in N fluxes since 1970, as presented earlier.

Nitrogen surpluses and manure nitrogen inputs as performance indicators In the Pan European initiative, SEBI2010, which stands for Streamlining European 2010 Biodiversity Indicators, the agricultural N balance (implying the N surplus) is one of the 26 indicators that are developed to monitor progress towards the European target to halt the loss of biodiversity by 2010 (see: http://biodiversity-chm.eea.europa.eu/information/indicator/ F1090245995). The N surplus is, however, a typical pressure indicator and not an effect indicator, since agro-ecosystems and environment both have a strong impact on the actual N emissions to the atmosphere and the N (NH4 and NO3) concentrations in leaching and runoff water, being relevant for the effects that may occur. For example, ammonia losses from agriculture are associated predominantly with animal production systems. Nitrate concentrations in the leachate to groundwater depend not only on N balance (N surplus) but also on climate (excess rainfall which dilutes the concentration), and soil type, affecting denitrification. As a result, the relationship between N surplus and N fluxes to the air and to water is diffuse.

Because of this complexity and variability, there are very few common and accepted reference levels against which to evaluate nutrient surpluses. In the Netherlands, the regulatory policy instrument MINAS has been used in the past in which reference values for N surpluses have been set tentatively at 60 and 100 kg per ha for arable land on sandy soils and clayey soils, respectively, and at 140 and 180 kg per ha for grassland on sandy soils and clayey soils, respectively. At present, N surplus is not used as a performance indicator in policy making. Instead, use is made of a maximum N application rate by animal manure of 170 kg N with exceptions (so-called derogations) of 250 kg N for the Netherlands and 230 kg N for Denmark, Germany and Austria (after the year 2000). Maps of the N input by animal manure for the year 2000 (Figure 15.4) indicate that there still exist a number of areas in Europe where this limit is exceeded.

Ammonia emission and related ammonia concentrations and nitrogen deposition as performance indicators The variation in NH3 emissions will affect the N deposition on terrestrial ecosystems. Plant species diversity of terrestrial ecosystems is affected largely by N deposition and in this context empirical and model based critical N loads have been derived. Specifically in intensive livestock areas with high NH3 emissions, the resulting N deposition may lead to an exceedance of critical N loads. In this context, national emission ceilings (NEC) have been set. A comparison of NECs for 2010 (EEA, 2010) and results of total NH3 emissions by the various models described in this chapter is given in Table 15.9. For INTEGRATOR, IDEAg, MITERRA, and IMAGE, the estimated agricultural NH3 emissions per country were multiplied by a factor 1.07, since approximately 7% of the NH3 emissions come from non-agricultural sources. The number of countries with estimated NH3-N emissions in 2000 exceeding the National emission ceilings for 2010 depends on the model approach and varied between 7 and 18, while the total exceedance varied between 75 and 1269 kton NH3-N yr−1. The large exceedances derived by EDGAR are clearly deviating from all other model approaches. The lowest emission exceedances

339

Geographical variation in terrestrial nitrogen budgets

diversity, despite the limited emission reductions in NH3 (see also Figure 15.8 lower graph for the period 1980–2000). This effect is specifically due to NOx emission reductions in that period.

Nitrogen leaching and nitrogen runoff as performance indicators

Figure 15.12 A comparison of the estimated national emissions and national emission ceilings for NH3, derived with INTEGRATOR and various other model approaches (IDEAg, MITERRA, IMAGE, GAINS, EDGAR, EMEP and OECD/IPCC) for the year 2000.

are estimated by INTEGRATOR, MITERRA and GAINS, all being based on the same animal numbers and NH3 emission factors. The variation in NH3-N emission exceedances, limited to those countries where all models calculate an exceedance is illustrated in Figure 15.12. For most countries, the exceedance is comparable, but for some countries the variation is considerable up to a fourfold variation. Insight in the actual risk of elevated NH3 emissions on terrestrial ecosystems can amongst others be derived by comparing either the actual NH3 concentration with a critical NH3 concentration in view of plant species diversity impacts. Recently updated critical levels are 1 µg.m−3 for lichens and bryophytes and 3 µg.m−3 for herbaceous plants (Cape et╯al., 2009). A comparison of EMEP model predicted NH3 concentrations with these critical levels during the last 15 years show that NH3 concentrations violate the limit for lichens and bryophytes except for Fennoscandia and Scotland, as presented in Moldanová et╯al., 2011 (Chapter 18, this volume). The limit for herbaceous plants is also exceeded in parts of Western Europe and Northern Italy. Indirectly, insight in the actual risk of elevated NH3 emissions on terrestrial ecosystems can also be derived by comparing present N depositions, which are largely determined by NH3 emissions together with NOx emissions, with the critical N deposition at the European scale. The critical N deposition is related to impacts on plant species diversity and is either derived from empirical field data or by model assessments, as discussed in Dise et╯al., 2011 (Chapter 20 this volume). The exceedance of critical N loads in view of impacts on plant species diversity is one of the 26 performance indicators in SEBI 2010. A comparison of exceedances of critical N loads in 1980 and in 2010 is given in Dise et╯al., 2011 (Chapter 20 this volume), showing that the N emission reductions in the past three decades has led to a significant reduction in the risk of N affecting plant species

340

Critical NO3 concentrations in ground water in view of health effects and critical N concentrations in surface waters in view of eutrophication of aquatic ecosystems are also important targets to evaluate the N leaching and N runoff fluxes on a European wide scale. A critical NO3 concentration in view of health impacts is set at 50 mg NO3â•›l−1. Eutrophication is the result of nutrient (both N and P) enrichment in the aquatic system, but the severity of the phenomenon largely depends on the specific regional characteristics, climate, morphology, water residence time, nutrients ratio, tropic web status, and generally on the ecosystem resilience. Therefore, similar nutrient loads may produce different effects in reason of the regional sensitivities. Similarly, the impacts are related not only to N loads, but rather to its specific synergies with the availability of other elements, such as carbon, phosphorus and silica (see also Billen et╯al., 2011; Grizzetti et╯al., 2011, Chapters 13 and 17 this volume). Nevertheless, N concentrations in surface waters, being a major driving force of the problems, are used as a proxy to evaluate the risk for water eutrophication. A critical limit of 0.5–1.0 mg N l−1 has been proposed by Camargo and Alonso (2006) based on an extensive study on the ecological and toxicological effects of inorganic N pollution in aquatic ecosystems. At present, N concentrations are generally exceeding those limits (see also Grizzetti et╯al., 2011, Chapter 17 this volume).

15.5.3╇ Conclusions and recommendations Key findings regarding the temporal and geographic variation in N budgets in agricultural and other terrestrial ecosystems over Europe are as follows. • Trends in N fluxes in agro-ecosystems since 1970 show an increase in N inputs by fertilisers and manure up to 1985, followed by a decrease since 1985 in response to a change in crop production and in animal numbers. Actually, livestock decreased since 1970, but in the period 1970– 1985 the N input by manure excretion still increased due to an increase in milk production and related N excretion rates. • For EU-27, the models estimates a total N input in European agriculture for the year 2000 of 23.3–25.7 Mton N yr−1 which is mainly due to fertiliser and animal manure inputs and to a lesser extent by atmospheric deposition and N fixation. Total N inputs at EU-27 level are comparable for all models, since they all use comparable basic data on fertiliser use and animal numbers. There exist a number of areas in Europe where a maximum N application rate by animal manure of 170 kg N is exceeded. The N uptake varies from 11.3–15.4 Mton N yr−1 leading to total N surpluses varying from 10.4–13.2 Mton N yr−1 at EU-27 level.

Wim de Vries

• The four complete N budget models for agro-ecosystems give in general very similar results for the emissions of NH3 (2.8–3.1 Mton N yr−1) and N2O (0.33–0.43 Mton N yr−1) but vary largely for NOx (0.02–0.23 Mton N yr−1). Similar results and differences are found when including standard activity data-emission factors approaches (UNFCC/IPCC, GAINS, OECD, EDGAR and EMEP). • Even though NOx emissions are more uncertain, the uncertainty in the NH3 emissions is more important for the overall uncertainty in the reactive N budget, since NOx contribute little to the overall N budget. The contribution of agriculture to total NOx emissions is less than 5%, while the contribution of agricultural NH3 emissions is more than 90%, making the variation in NH3 emissions more important. The uncertainty is illustrated by the number of countries with estimated NH3-N emissions in 2000 exceeding the National emission ceilings for 2010. Depending on the model approach, this number varies between 7 and 18, while the total exceedance varied between 75 and 1269 kton NH3-N yr−1. • The estimated sum of N leaching and runoff at EU 27 is roughly equal to the sum of NH3, N2O and NOx emissions to the atmosphere, but estimates vary by a factor two, from 2.7–6.3 Mton N yr−1. This strongly affects the area with N concentrations exceeding critical N concentrations in surface water. • In non-agricultural system (forests and semi-natural vegetation), the estimated total input is near 3.2 Mton N yr−1, while the net N uptake is near 1.1 Mton N yr−1, leading to a surplus near 2.1 Mton N yr−1. Compared to agricultural systems, the estimated N fluxes in non-agricultural systems are about 5 times lower for N2O emissions and 10 times lower for NOx and NH3 emissions and for the sum of N leaching and runoff. • The regional variation in N fluxes is mainly determined by N inputs, being highest in areas with high livestock density and intensive agricultural crop production areas, while land/soil characteristics and climate are secondary factors influencing the magnitude of N fluxes. Recommendations that can be made based on this assessment are as follows. • Future research priorities should focus on major uncertainties, in particular N2O emissions and N leaching and runoff from agricultural ecosystems. Furthermore, studies on denitrification are needed to reduce the large uncertainty in this process at the European scale. • A database should be set up of N contents in various plants and in various regions to improve estimates of N uptake and N surplus at the European scale. • Information on NH3 concentrations in air should be used in inverse modelling approaches to derive independent datasets to validate the various NH3 emission calculations. • A European-wide monitoring network of ground- and surface water, using standardised methods and covering a range of habitats, should be initiated to provide consistent

and reliable information on the long-term effects of air pollution on water quality, to be used for validation of N budget models. • It is relevant that data use is harmonised for models predicting air emissions and N loss to waters for consistent environmental decision-making relevant to air quality, ecosystem deposition and water quality.

Acknowledgements This chapter was prepared with the support of the NinE Programme of the European Science Foundation, the NitroEurope IP (funded by the European Commission), the COST Action 729. We also thankfully acknowledge Suvi Monni (JRC) for sending updated EDGAR emission data.

Supplementary materials Supplementary materials (as referenced in the chapter) are available online through both Cambridge University Press: www.cambridge.org/ena and the Nitrogen in Europe website: www.nine-esf.org/ena.

References Alcamo, J. (ed.) (1994). IMAGE 2.0. Integrated modeling of global climate change. Water, Air and Soil Pollution, 76, 1–318. Bergamaschi, P. (ed.) (2007). Atmospheric Monitoring and Inverse Modelling for Verification of National and EU Bottom-up GHG Inventories€– report of the workshop ‘Atmospheric Monitoring and Inverse Modelling for Verification of National and EU Bottom-up GHG Inventories’ under the mandate of Climate Change Committee Working Group I, Casa Don Guanella, Ispra, Italy (08–09 March 2007). European Commission Joint Research Centre, Institute for Environment and Sustainability. Billen, G., Silvestre, M., Grizzetti, B. et╯al. (2011). Nitrogen flows from European regional water sheds to coastal marine waters. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J.€W. Erisman et╯al., Cambridge University Press. Bloemerts, M. and de Vries, W. (2009). Relationships between Nitrous Oxide Emissions from Natural Ecosystems and Environmental Factors. Alterra Wageningen UR, Wageningen, The Netherlands. Bouraoui, F. and Aloe, A. (2007). European Agrochemicals Geospatial Loss Estimator:€Model Development and Application, EUR Report 22690 EN. Office for Official Publications of the European Communities, Luxembourg. Bouraoui, F., Grizzetti, B. and Aloe, A. (2009). Nutrient Discharge from River to Seas for Year 2000. Office for Official Publications of the European Communities, Luxembourg. Bouwman, A. F., van Drecht, G. and van der Hoek, K. W. (2005). Nitrogen surface balances in intensive agricultural production systems in different world regions for the period 1970–2030. Pedosphere, 15, 137–155. Britz, W. (ed.) (2005). CAPRI Modelling System Documentation. Common Agricultural Policy Regional Impact Analysis: “Development of a Regionalised EU-wide Operational Model to Assess the Impact of Current Common Agricultural Policy on Farming Sustainability”. J05/30/2004€– Deliverable 1, Bonn. Britz, W., Heckelei, T. and Kempen, M. (2005). Description of the CAPRI Modeling System, final report of the CAPRI-DynaSpat

341

Geographical variation in terrestrial nitrogen budgets Project. Institute for Food and Resource Econommics, University of Bonn, Bonn, Germany. Britz, W. and Leip, A. (2009a). Development of marginal emission factors for N losses from agricultural soils with the DNDC-CAPRI meta-model. Agriculture, Ecosystems and Environment, 133, 267–279. Britz, W. and Leip, A. (2009b). Development of marginal emission factors for N losses from agricultural soils with the DNDC-CAPRI meta-model. Agriculture Ecosystems and Environment, 133, 267–279. Butterbach-Bahl, K., Kahl, M., Mykhayliv, L. et╯al. (2009). A European-wide inventory of soil NO emissions using the biogeochemical models DNDC/Forest-DNDC. Atmospheric Environment, 43, 1392–1402. Butterbach-Bahl, K., Nemitz, E., Zaehle, S. et╯al. (2011). Nitrogen as a threat to the European greenhouse balance. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W.€Erisman et╯al., Cambridge University Press. Camargo, J. A. and Alonso, A. (2006). Ecological and toxicological effects of inorganic nitrogen pollution in aquatic ecosystems:€a global assessment. Environment International, 32, 831–849. Cape, J. N., Van der Eerden, L. J., Sheppard, L. J., Leith, I. D. and Sutton, M. A. (2009). Evidence for changing the critical level for ammonia. Environmental Pollution, 157, 1033–1037. CAPRI (2010). http://www.ilr1.uni-bonn.de/agpo/rsrch/capri/ capri_e.htm De Vries, W., Reinds, G. J., Gundersen, P. and Sterba, H. (2006). The impact of nitrogen deposition on carbon sequestration in European forests and forest soils. Global Change Biology, 12, 1151–1173. De Vries, W., Kros, J., Voogd, J. C. et╯al. (2009). Comparing Predictions of Nitrogen and Greenhouse Gas Fluxes in Response to Changes in Live stock, Land Cover and Land Management Using Models at a National, European and Global Scale. Alterra Wageningen UR, Wageningen, The Netherlands. De Vries, W., Kros, J., Reinds, G. J., Wieggers, R., Velthof, G., Oudendag, D., Oenema, O., Nabuurs, G. J., Schelhaas, M. J., Perez Soba, M., Rienks, W., de Winter, W., van den Akker, J., Bakker, M., Verburg, P., Eickhout, B. and Bouman, L. (2010). INTEGRATOR:€A Modelling Tool for European-wide Assessments of Nitrogen and Greenhouse Gas Fluxes in Response to Changes in Land Cover, Land Management and Climate – Calculation Procedures, Application Methodology and Examples of Scenario Results. Alterra, Wageningen. Alterra Report (in preparation). De Vries, W., Leip, A., Reinds, G. J. et╯al. (2011). Comparison of land nitrogen budgets for European agriculture by various modeling approaches. Environmental Pollution (in press). Dise, N. B., Ashmore, M., Belyazid, S. et╯al. (2011). Nitrogen as a threat to European terrestrial biodiversity. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. EC (2007). Report from the Commission to the Council and the European Parliament on Implementation of Council Directive 91/676/EEC Concerning the Protection of Waters against Pollution Caused by Nitrates from Agricultural Sources for the Period 2000– 2003. Commission of the European Communities, Brussels. EDGAR (2010). http://www.mnp.nl/edgar/model/v32ft2000edgar/ EEA (2005). Agriculture and Environment in EU15: The IRENA Indicator Report. European Environment Agency, Copenhagen. EEA (2008). Annual European Community Greenhouse Gas inventory 1990–2006 and Inventory Report 2008, Submission to the UNFCCC Secretariat. European Environment Agency, Copenhagen.

342

EEA (2010). http://www.eea.europa.eu/highlights/more-eu-memberstates-to-miss-2010-air-pollutant-limits/nec-status-preliminaryresults-2008-data.pdf EMEP (2009). Status Report 1/09 Transboundary Acidification, Eutrophication and Ground Level Ozone in Europe in 2007. Joint MSC-W & CCC and CEIP Report, Norwegian Meteorological Institute, Oslo, Norway. (available through www.emep.int). EMEP (2010a). http://www.emep.int EMEP (2010b). http://www.ceip.at/ GAINS (2010) http://www.iiasa.ac.at/web-apps/apd/gains/EU/index. login?logout=1 Gilliland, A. B., Dennis, R. L., Roselle, S. J. and Pierce, T. E. (2003). Seasonal NH3 emission estimates for the eastern united states based on ammonium wet concentrations and an inverse modeling method. Journal of Geophysical Research, 108, 4477. Grizzetti, B., Bouraoui, F., de Marsily, G. and Bidoglio, G. (2005). A statistical method for source apportionment of riverine nitrogen loads. Journal of Hydrology, 304, 302–315. Grizzetti, B., Bouraoui, F. and Aloe, A. (2007). Spatialised European Nutrient Balance. Report EUR 22692 EN. Grizzetti, B., Bouraoui, F. and De Marsily, G. (2008). Assessing nitrogen pressures on European surface water. Global Biogeochemical Cycles, 22, GB4023, doi:10.1029/2007GB003085. Grizzetti, B., Bouraoui, F., Billen, G. et╯al. (2011). Nitrogen as a threat to European water quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Hirsch, A. I., Michalak, A. M., Bruhwiler, L. M. et╯al. (2006). Inverse modeling estimates of the global nitrous oxide surface flux from 1998–2001. Global Biogeochemical Cycles, 20, GB1008, doi:€10.1029/2004GB002443. Höglund-Isaksson, L. and Mechler, R. (2005). The GAINS Model for Greenhouse Gases, Version 1.0:€Methane (CH4), Interim Report IR-05–54. International Institute for Applied Systems Analysis. Huang, J., Golombek, A., Prinn, R. et╯al. (2008). Estimation of regional emissions of nitrous oxide from 1997 to 2005 using multinetwork measurements, a chemical transport model, and an inverse method. Journal of Geophysical Research 113, doi:10.1029/2007JD009381. IMAGE (2010), http://www.mnp.nl/en/themasites/image/ index.html IPCC (1997). Revised 1996 IPCC Guidelines for National Greenhouse Gas Inventories. UK Meteorological Office, Bracknell, UK. IPCC (2001). Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories. Published for the IPCC by the Institute for Global Environmental Strategies, Japan. IPCC (2006). 2006 IPCC Guidelines for National Greenhouse Gas Inventories, prepared by the National Greenhouse Gas Inventories Programme, Eggleston, ed. H. Simon et╯al. Institute for Global Environmental Strategies, Japan. IPCC (2007). Climate Change 2007:€The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press. Kesik, M., Ambus, P., Baritz, R. et╯al. (2005). Inventories of N2O and NO emissions from European forest soils. Biogeosciences, 2, 353–375.

Wim de Vries Konovalov, I. B., Beekmann, M., Richter, A., Burrows, J. P. and Hilboll,€A. (2010). Multi-annual changes of NOx emissions in megacity regions:€nonlinear trend analysis of satellite measurement based estimates. Atmospheric Chemistry and Physics Discussions, 10, 10925–10968. Leemans, R., Kreileman, E., Zuidema, G. et╯al. (1998). The IMAGE User Support System:€Global Change Scenarios from IMAGE 2.1. National Institute for Public Health and the Environment, Bilthoven The Netherlands. Leip, A., Marchi, G., Koeble, R. et╯al. (2008). Linking an economic model for European agriculture with a mechanistic model to estimate nitrogen and carbon losses from arable soils in Europe. Biogeosciences, 5, 73–94. Leip, A., Weiss, F. and Britz, W. (2009). Agri-environmental nitrogen indicators for EU27. In: Proceedings of the Conference on Integrated Assessment of Agriculture and Sustainable Development:€Setting the Agenda for Science and Policy (AgSAP 2009), Egmond aan Zee, The Netherlands, 10–12 March 2009, ed. M. K. Van Ittersum, J. Wolf and H. H. Van Laar. Wageningen University and Research Centre, Wageningen, The Netherlands, pp. 184–185. Leip, A., Achermann, B., Billen, G. et╯al. (2011a). Integrating nitrogen fluxes at the European scale. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Leip, A., Britz, W., Weiss, F. and de Vries, W. (2011b). Farm, land, and soil nitrogen budgets for agriculture in Europe. Environmental Pollution (in press). Li, C., Aber, J., Stange, F., Butterbach-Bahl, K. and Papen, H. (2000). A process oriented model of N2O and NO emissions from forest soils:€I. Model development. Journal of Geophysical Research, 105, 4369–4384. Manning, A. J., Ryall, D. B., Derwent, R. G., Simmonds, P. G. and O’Doherty, S. (2003). Estimating European emissions of ozonedepleting and greenhouse gases using observations and a modeling back-attribution technique. Journal of Geophysical Research, 108, 4405. Messager, C., Schmidt, M., Ramonet, M. et╯al. (2008). Ten years of CO2, CH4, CO and N2O fluxes over Western Europe inferred from atmospheric measurements at Mace Head, Ireland. Atmospheric Chemistry and Physics Discussions, 8, 1191–1237. MNP (2006). Integrated Modelling of Global Environmental Change: An Overview of IMAGE 2.4. Netherlands Environmental Assessment Agency (MNP), Bilthoven, The Netherlands. Moldanová, J., Grennfelt, P., Jonsson, Å. et╯al. (2011). Nitrogen as a threat to European air quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Mosier, A., Kroeze, C., Nevison, C. et╯al. (1998). Closing the global N2O budget:€nitrous oxide emissions through the agricultural nitrogen cycle (OECD/IPCC/IEA phase II development of IPCC guidelines for national greenhouse gas inventory methodology). Nutrient Cycling in Agroecosystems, 52, 225–248. Neeteson, J. J. (2000). Nitrogen and phosphorus management on Dutch dairy farms:€legislation and strategies to meet the regulations. Biology and Fertility of Soils, 30, 566–572. OECD (2001). Environmental Indicators for Agriculture, Volume 3, Methods and Results. Organisation for Economic Co-operation and Development, Paris, France. OECD (2007). Environmental Indicators for Agriculture, Volume 4, Organisation for Economic Co-operation and Development, Paris, France.

OECD (2010). http://stats.oecd.org/wbos/Index.aspx/datasetcode= ENVPERFINDIC_TAD_2008 and http://www.oecd.org/document /56/0,3343,en_2649_33793_40374392_1_1_1_1,00.html Oenema, O., Boers, P. C. M., van Eerdt, M. M. et╯al. (1998). Leaching of nitrate from agriculture to groundwater:€the effect of policies and measures in the Netherlands. Environmental Pollution, 102, 471–478. Oenema, O., Kros, H. and de Vries, W. (2003). Approaches and uncertainties in nutrient budgets:€implications for nutrient management and environmental policies. European Journal of Agronomy, 20, 3–16. Oenema, O., Oudendag, D. and Velthof, G. L. (2007). Nutrient losses from manure management in the European Union. Livestock Science, 112, 261–272. Orlandi, S. and Van der Goot, E. (2003). Technical Description of Interpolation and Processing of Meteorological Data in CGMS, European Commission, DG JRC, Agrifish. Unit, http://mars.jrc.it/ mars/content/download/640/4574/file/GridWeather.doc OSPARCOM (1994). OSPARCOM Guidelines for Calculating Mineral Balances. Working Group on Nutrients, NUT 94/8/1-E, Bern, Switzerland. Prinn, R., Cunnold, D., Rasmussen, R. et╯al. (1990). Atmospheric emissions and trends of nitrous oxide deduced from 10 years of ALEGAGE data. Journal of Geophysical Research, 95, 18369–18385. Ryall, D. B., Derwent, R. G., Manning, A. J., Simmonds, P. G. and O’Doherty, S. (2001). Estimating source regions of European emissions of trace gases from observations at Mace Head. Atmospheric Environment, 35, 2507–2523. Simpson, D., Winiwarter, W., Börjesson, G. et╯al. (1999). Inventorying emissions from nature in Europe. Journal of Geophysical Research, 104, 8113–8152. Simpson, D., Fagerli, H., Jonson, J. E. et╯al. (2003). Transboundary Acidification, Eutrophication and Ground Level Ozone in Europe PART I. Unified EMEP Model Description. Norwegian Meteorological Institute, Oslo, Norway. Simpson, D., Butterbach-Bahl, K., Fagerli, H. et╯al. (2006). Deposition and emissions of reactive nitrogen over European forests:€a modelling study. Atmospheric Environment, 40, 5712–5726. Simpson, D., Aas, W., Bartnicki, J. et╯al. (2011). Atmospheric transport and deposition of nitrogen in Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. UNFCCC (2010). www.unfccc.int UNFCC/IPCC (2010). http://unfccc.int/ghg_data/ghg_data_unfccc/ items/4146.php Van Aardenne, J. A. (2002). Uncertainties in emission inventories. PhD thesis, Wageningen University, Wageningen. Van der Velde, M., Bouraoui, F. and Aloe, A. (2009). Pan-European regional-scale modelling of water and N efficiencies of rapeseed cultivation for biodiesel production. Global Change Biology, 15, 24–37. Velthof, G., Oudendag, D. and Oenema, O. (2007). Development and Application of the Integrated Nitrogen Model MITERRA-EUROPE: Alterra report 1663.1. Alterra, Wageningen, The Netherlands. Velthof, G. L., Oudendag, D. A. and Witzke, H. P. (2009). Integrated assessment of nitrogen emission losses from agriculture in EU-27 using MITERRA-EUROPE. Journal of Environmental Quality, 38, 1–16. Velthof, G., Barot, S., Bloem, J. et╯al. (2011). Nitrogen as a threat to European soil quality. In:€The European Nitrogen Assessment,

343

Geographical variation in terrestrial nitrogen budgets ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Vieno, M., Dore, A. J., Wind, P. et╯al. (2009). Application of the EMEP Unified Model to the UK with a horizontal resolution of 5 × 5 km2 Atmospheric Ammonia. In: Detecting Emissions Changes and Environmental Impacts, ed. M. A. Sutton, S. Reis and S.€M.€Baker, Springer, New York, pp. 367–372.

344

Winiwarter, W. (2005). The GAINS Model for Greenhouse Gases – Version 1.0:€Nitrous Oxide (N2O). IIASA Interim Report IR-05–55, International Institute for Applied Systems Analysis. Witzke, H. P. and Oenema, O. (2007). Assessment of Most Promising Measures Task 3 Service Contract “Integrated Measures in Agriculture to Reduce Ammonia Emissions”. Alterra, Wageningen, The Netherlands.

Chapter

16

Integrating nitrogen fluxes at the European scale Lead author: Adrian Leip Contributing authors: Beat Achermann, Gilles Billen, Albert Bleeker, Alexander F. Bouwman, Wim de Vries, Ulli Dragosits, Ulrike Döring, Dave Fernall, Markus Geupel, Jürg Herolstab, Penny Johnes, Anne Christine Le Gall, Suvi Monni, Rostislav Nevečeřal, Lorenzo Orlandini, Michel Prud’homme, Hannes I. Reuter, David Simpson, Guenther Seufert, Till Spranger, Mark A. Sutton, John van Aardenne, Maren Voß and Wilfried Winiwarter

Executive summary Nature of the problem • Environmental problems related to nitrogen concern all economic sectors and impact all media:€atmosphere, pedosphere, hydrosphere and anthroposphere. • Therefore, the integration of fluxes allows an overall coverage of problems related to reactive nitrogen (Nr) in the environment, which is not accessible from sectoral approaches or by focusing on specific media.

Approaches • This chapter presents a set of high resolution maps showing key elements of the N flux budget across Europe, including N2 and Nr fluxes. • Comparative nitrogen budgets are also presented for a range of European countries, highlighting the most efficient strategies for mitigating Nr problems at a national scale. A new European Nitrogen Budget (EU-27) is presented on the basis of state-of-the-art Europe-wide models and databases focusing on different segments of Europe’s society.

Key findings • From c. 18 Tg Nr yr−1 input to agriculture in the EU-27, only about 7 Tg Nr yr−1 find their way to the consumer or are further processed by industry. • Some 3.7 Tg Nr yr−1 is released by the burning of fossil fuels in the EU-27, whereby the contribution of the industry and energy sectors is equal to that of the transport sector. More than 8 Tg Nr yr−1 are disposed of to the hydrosphere, while the EU-27 is a net exporter of reactive nitrogen through atmospheric transport of c. 2.3 Tg Nr yr−1. • The largest single sink for Nr appears to be denitrification to N2 in European coastal shelf regions (potentially as large as the input of mineral fertilizer, about 11 Tg N yr–1 for the EU-27); however, this sink is also the most uncertain, because of the uncertainty of Nr import from the open ocean.

Major uncertainties • National nitrogen budgets are difficult to compile using a large range of data sources and are currently available only for a limited number of countries. • Modelling approaches have been used to fill in the data gaps in some of these budgets, but it became obvious during this study that further research is needed in order to collect necessary data and make national nitrogen budgets inter-comparable across Europe. • In some countries, due to inconsistent or contradictory information coming from different data sources, closure of the nitrogen budget was not possible.

Recommendations • The large variety of problems associated with the excess of Nr in the European environment, including adverse impacts, requires an integrated nitrogen management approach that would allow for creation and closure of N budgets within European environments. • Development of nitrogen budgets nationwide, their assessment and management could become an effective tool to prioritize measures and prevent unwanted side effects.

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

345

Integrating nitrogen fluxes at the European scale

16.1╇ Introduction The concept of the nitrogen cascade was introduced to describe the ‘[…] multiple linkages among the ecological and human effects of reactive nitrogen molecules as they move from one environmental system to another’ (Galloway et€al., 2003). The quantification of the nitrogen cascade requires accurate estimation of the fluxes across the sectoral and media boundaries for a large geographic entity, from regional to national, continental and global scale. Such a complete nitrogen (N)-budget was first presented for Europe by van Egmond et€al. (2002). In this chapter, an update is given of the information on complete N-budgets, including all major N2 and reactive nitrogen (Nr) fluxes, both at country and at continental level. The European Nitrogen Assessment (ENA) provides an overview of the processes and pathways associated with the cascade of Nr through the environment, and also the order of magnitude of the associated problems, based on recent scientific literature and latest available model results. Each of the chapters focuses on one specific sector (for example de Vries et€al., Chapter 15, on the N fluxes from agriculture and natural ecosystems and Svirejeva-Hopkins et€al., Chapter 12, on the effect of urbanization), on one specific medium (such as for example Simpson et€ al., Chapter 14, on the transport of Nr in the atmosphere and Billen et€al., Chapter 13, looking at nitrogen from the perspective of watersheds or on one specific aspect in the nitrogen cascade (for example the transformation processes in soils in Butterbach-Bahl et€ al., Chapter 6, or chemical reactions occurring in the atmosphere in Hertel et€al., Chapter 9). The inter-connections between these specific assessments are manifold and reflect the interactions that nitrogen undergoes in the environment across the borders of scales, sectors and media. The present chapter stands at the interface between the sections describing nitrogen issues and those explaining nitrogen problems and suggesting nitrogen solutions. Two objectives are identified. (i) To give an overview of the most important N-fluxes in Europe in a gridded representation, i.e. compiling a number of ‘key maps’ that help the understanding of regional differences of the main N-indicators. (ii) To show aggregated fluxes of nitrogen across media and sectors, i.e. integrated national N-budgets for those countries where they have already been established and a new ‘European Nitrogen Budget’ based on the evidence compiled and the filling in of gaps according to the latest scientific knowledge. The second goal, in particular, is a challenging one, as most research is done in individual disciplines and only a few ‘integrated’ models exist today that are able to give a comprehensive overview of the nitrogen budget at a national scale. However, while some decades ago the focus of research was on individual fluxes (e.g. nitrate concentrations in rivers), and specific tools (e.g. models of N2O fluxes from agricultural soils), in recent

346

years progress has been made in developing tools and databases which cross sector- and media-boundaries and are able to consider effects such as ‘pollution swapping’ and to evaluate tradeoffs. The need to mitigate environmental problems related to nitrogen in an integrated way has led to the development of ‘national nitrogen budgets’ aimed at helping to find the most efficient and cost-effective solutions to abate these problems. Still, the establishment of a nitrogen budget requires (i)€the co-operation of experts from different disciplines and/or (ii)€ the integration of various dedicated models. Nitrogen budgeting at the national scale often relies on the first solution, as the density of experimental observations at the national scale might be sufficient to come up with good estimates of nitrogen fluxes between sub-systems, and model-results can (if needed) be used to fill gaps. Such national or (in the case of Europe) supra-national nitrogen budgets are increasingly recognized to be a very useful tool for visualizing the complexity of nitrogen issues, also in relation to society as a whole. These national nitrogen budgets help to support prioritization of policies and provide a first assessment of the impact a policy might have at various points in the nitrogen cascade. In order to understand the fluxes that are included in the establishment of European and national nitrogen budgets, we employ a number of models covering partial aspects of an overall budget. Each model has its strengths and weaknesses. Therefore combining the best features of these models provides a means for the construction of a cross-sector, crossmedia European Nitrogen Budget. The following sections give an overview of the main data sources and models used for the key-maps presented in Section 16.3 as well as for the European Nitrogen Budget (for the EU-27) presented in Section 16.4, where we also present national N-budgets developed by country experts on the basis of national data sources.

16.2╇ Data sources For the assessment of key nitrogen fluxes at the European scale and the development of a European Nitrogen Budget, one cannot rely on statistical or observational data as they do not exist for most of the fluxes that need to be considered. Observational data are scarce and unevenly distributed over the European area, so that a statistical up-scaling is often not possible. Instead, models are needed that extrapolate nitrogen (and other) fluxes at large scales on the basis of existing environmental or statistical information. Here, we make use of the results of such models. The models were selected on the basis of the following criteria:€(i) applicability at the European scale; (ii) a high spatial data resolution, the European scale notwithstanding; and (iii) a focus in the parameterization of nitrogen fluxes in the compartments considered (see also de Vries et€al., 2011, Chapter€15 this volume). The CAPRI-DNDC-based Integrated Database for European Agriculture (IDEAg) gives currently the most complete information on the flow of nitrogen into and through the agricultural sector in Europe, calculating also reactive nitrogen and greenhouse gas fluxes (Leip et€al., 2008; Leip et€al., 2010a). The Indicator Database for European Agriculture builds mainly

Adrian Leip

on the results from the economic model for agriculture CAPRI (Britz and Witzke, 2008) and the biophysical model for soil nutrient turnover DNDC (Li, 2000) through a meta-modelling approach (Britz and Leip, 2009). IDEAg covers all nitrogen fluxes related to agricultural activities in Europe. It has recently been extended to cover also Nr emissions from sewerage systems in accordance with the methodology developed for the IMAGE model (Bouwman et€al., 2006). The INTEGRATOR model is an integrated model specifically designed to help developing integrated policies. It has been developed to assess responses of nitrogen and greenhouse gas (GHG) emissions to European-scale changes in land use, land management and climate. INTEGRATOR links modules calculating N and GHG emissions from housing and manure storage systems, agricultural and non-agricultural soils and surface waters, while accounting for the interaction between different sources through an emission–deposition model for NH3 and NOx. It uses relatively simple and transparent model calculations based on the use and adaptation of available model approaches, including empirical model approaches and statistical relations between model outputs and environmental variables. The model focuses on the derivation of high resolution spatially explicit data (De Vries et€al., 2009). The Emission Database for Global Atmospheric Research (EDGAR) calculates emissions of air pollutants and greenhouse gases on a grid for use in atmospheric circulation models covering all relevant anthropogenic emission sectors (Olivier et€al., 2005; Van Aardenne et€al., 2001). EDGAR was used as the standard database for deriving emission estimates as it provides a consistent emission calculation of emissions for the whole territory considered and a sophisticated downscaling procedure to map emissions at high spatial resolution, including ship and aviation emissions and detailed sub-sector disaggregation. Two datasets have been applied in this report. N2O emissions have been taken from EDGARv4.0 (JRC/PBL, 2009) and the NOx and NH3 emissions are taken from the EDGAR-CIRCE dataset (Van Aardenne et€al., 2009). The Unified EMEP model is used to estimate atmospheric transport and deposition as calculated by the Europeanscale EMEP MSC-W Chemical transport model (European Monitoring and Evaluation Programme, Meteorological Synthesizing Centre€ – West). The EMEP models have been instrumental to the development of air quality policies in Europe since the late seventies, mainly through their support to the strategy work under the Convention on Long-range Transboundary Air Pollution, and became the reference atmospheric dispersion model for use in the Integrated Assessment Models supporting the development of air quality polices under the EU Commission. The Unified EMEP model is designed to calculate air concentrations and deposition fields for major acidifying and eutrophying pollutants, photo-oxidants and particulate matter (Simpson et€al., 2006). Additional information on these models can be found in the supplementary material (see supplementary material Â�Chapter€15 and 16) including also a comparison of total atmospheric Nr emissions fluxes by various data sets. Details on the data sets used by these models to estimate Nr fluxes is given in

Table 16.1 Overview table of main models used in this chapter to generate the key maps and the European Nitrogen Budget (ENB)

Model

Nitrogen fluxes estimated for the ENB

IDEAg

emissions and nitrogen leaching and run-off from agriculture exchange of nitrogen between the soil and the livestock sectors application of mineral fertilizer to agricultural soils feed and food trade land productivity, consumption of nitrogen nitrogen input to and emissions from sewage treatment systems

INTEGRATOR

emissions and nitrogen leaching from forests and rough grazing

EDGAR

emissions from stationary combustion (energy sector, industry, residential sector) and industrial processes emissions from transport nitrogen input to solid waste management emissions from solid waste management

EMEP

atmospheric deposition

de Vries et€al. (2011, Chapter 15, this volume). A summary of the models used and the main data obtained from each of these models is given in Table 16.1. Each model focuses on different sectors and the models are thus complementary. Information for agriculture is available also from INTEGRATOR and EDGAR; because the IDEAg is the most complete source of information for agriculture the data in this chapter is taken from this model. This avoids most inconsistencies between the data presented. However, inconsistencies cannot be completely excluded and are mainly due to:€ (i) different atmospheric deposition data used in the IDEAg and INTEGRATOR model and the EMEP deposition data used in this chapter; and (ii) Nr fluxes from coastal areas which are not included in any of the Europe-wide models. Covering complex processes on a continental scale, the models are bound to rely on simplifying assumptions regarding input data sets, and the parameterization of the processes and the results presented here are consequently associated with large uncertainties. A proper assessment of these uncertainties, however, is very difficult as independent data that can be used to quantify the uncertainties are missing. An attempt to quantify the uncertainty of these (and other) models is currently being done within the European integrated project NitroEurope-IP (Sutton et€al., 2007; NitroEurope, 2010). So far, the best approximation at an uncertainty assessment is done by comparing the in- and outputs of a wide range of models, as done by de Vries et€al. (2011, Chapter 15 this volume).

16.3╇ Key maps of nitrogen fluxes in Europe The purpose of this section is to present the spatial distribution of various types of key nitrogen fluxes over Europe that are

347

Integrating nitrogen fluxes at the European scale

Figure 16.1 Nitrogen input to agricultural soils in EU27 for the year 2002. The map shows total reactive N input to agricultural soils (cropland and grassland) yr€–1 for a grid at 1 km × 1 km, the values are in kg N per total pixel area [kg N km−2 yr −1 total area]. The pie diagram at the right side gives the split of N input [Gg N rounded to 10€Gg N yr –1] for EU27:€mineral fertilizer, manure (intentionally applied manure and manure deposited by grazing animals), atmospheric deposition, biological nitrogen fixation and crop residues returned to the soil. The histogram shows the split of N input [Gg N yearâ•›−1, rounded to 10 Gg N yearâ•›−1] by country. Basis:€Indicator Database for European Agriculture (IDEAg) V1, 2009. Method:€Mineral fertilizer data are obtained from FAO at the national level and are distributed to crops and regions by the CAPRI model using information from IFA/FAO. Distribution to the grid is done on the basis of estimated crop N requirements using information of the potential and water-limited yield for the soil-climate conditions and N supply from biological fixation, atmospheric deposition, and manure nitrogen supply. Manure N supply is estimated from manure availability on the basis of a livestock density map, crop demand and typical share of nitrogen supply by organic nitrogen. The data are net of nitrogen losses occurring before the application of manure to the soil. All data are estimated in consistence with regional values using the highest posterior density approach (Heckelei et al., 2005). Nitrogen deposition data are from EMEP (2008). Biological nitrogen fixation is estimated as a crop-dependent fraction of above-ground nitrogen. Crop residues are estimated from crop-specific fraction and N-content of crop residues. Details on the distribution algorithm can be found in Leip et al. (2008) and Britz and Leip (2009).

responsible for environmental problems. Maps are derived on the basis of fine scale resolution data (1 km × 1 km). The only way to obtain data on such a high resolution was to apply models and combine their results with measurements where available. Most of the models and data sources used have already been presented and explained in detail in earlier chapters. All together, 11 such key maps are selected. They can be grouped into three categories. (i) Drivers for and pressures of nitrogen in terrestrial ecosystems, including both agricultural and nonagricultural systems. Here the total load of nitrogen on agricultural soils (Figure 16.1) and the gross soil nitrogen budget for agricultural (Figure 16.2) and non-agricultural soils (Figure 16.3) can be regarded as key-indicators. Input of nitrogen through atmospheric deposition (Figure 16.4) is a major pressure on (semi-) natural ecosystems.

348

(ii) Emissions of reactive nitrogen to the atmosphere and to the hydrosphere. This is a very important pressure for environmental problems related to nitrogen. We show separate maps on NH3 (Figure 16.5), NOx (Figure 16.6), and N2O (Figure 16.7) emissions across Europe. Each of these compounds is dominated by different source categories (energy/transport for NOx, the livestock sector for NH3, soils for N2O), so that the distinction between these compounds gives also an idea of the spatial distribution of the main driving forces for reactive nitrogen generation. Emissions of nitrogen towards aquatic systems are presented in Figure 16.8. (iii) Secondary nitrogen indicators. Three indicators have been selected:€the total productivity of agricultural land (Figure 16.9), and the total consumption of reactive nitrogen by humans (Figure 16.10) and by animals

Adrian Leip

Figure 16.2 Soil system nitrogen surplus for agricultural soils in EU27 for the year 2002. The map shows reactive N surplus for a grid of 1 km × 1 km, the values are in kg N per total pixel area [kg N km−2 yrâ•›−1 total area]. The pie diagram at the right side gives the split of surplus [Gg N yearâ•›−1, rounded to 10 Gg N yearâ•›−1] for EU27 into the loss pathways:€NH3 emissions from soils, NOx emissions from soils N2O emissions from soils N2 emissions from soils, N leaching and runoff. The histogram shows the split of the N surplus [Gg N yr−1] into different loss pathways by country. Basis:€Indicator Database for European Agriculture V1, 2009. Method:€Nitrogen input to agricultural soils is estimated as given in Figure 16.1. Removal of nitrogen by crops and harvested or grazed grass is estimated from regional and national Eurostat statistics, downscaled to the grid on the basis of potential yield from (Genovese et al., 2007). Total N surplus at the grid scale is split into individual fluxes on the basis of the MITERRA approach (NH3 emissions and run-off) as implemented in CAPRI (Britz and Witzke, 2008; Velthof et al., 2009) and the DNDC-CAPRI meta-model (Britz and Leip, 2009) for N2, N2O, NOx and N-leaching. The spatial distribution is done on the basis of the nitrogen input data. Changes in soil-nitrogen stocks are also estimated with the DNDC-CAPRI meta-model; according to the soil-system approach they are nitrogen output as thus not included in the split of the N-surplus. A closed nitrogen budget is obtained according to the method described by Leip et al. (2009a).

(Figure€16.11). While the first indicator shows the potential of the land to feed its population in Europe, the other two indicators give a good idea of the ‘life style’ of citizens. Taken together, these indicators give information on the sustainability of land use and are the basis for the watershed assessment discussed in Chapter 13 (Billen et€al., 2011, Chapter 13 this volume).

16.3.1╇ Drivers for and pressures of nitrogen in terrestrial ecosystems We select two indicators describing drivers for the environmental load of Nr and two indicators for the pressure of Nr on the environment. The total Nr input to agricultural soils includes intentionally applied (organic or mineral) fertilizer and manure from grazing livestock as well as biological nitrogen-fixation and atmospheric deposition. Also crop residues returned to the soil are included in total Nr-inputs. Reactive nitrogen additions

are required to fulfil the needs of plants without compromising the productivity of the soil. At the same time, however, excessive Nr additions to agricultural soils lead to high pressures on the environment. Atmospheric deposition is the main source of Nr for natural land and forests. Atmospheric deposition is fuelled mainly by the emissions of NOx from energy-related sources and NH3 lost from agriculture. There is limited capacity in natural ecosystem to absorb Nr. While initial Nr addÂ� itions to forests can lead to a stimulation of plant growth and the build-up of soil organic matter, additions to Nr-saturated systems have adverse effects on the system’s functioning and most Nr is leached (Aber, 1992; Butterbach Bahl et€ al., 2011, Chapter 6 this volume). We define nitrogen surplus as the difference between total Nr inputs to a system and the useful outputs following the definition of the soil-system approach (de Vries et€al., 2011, Chapter 15 this volume). Inputs are total Nr-input as defined above, while useful Nr outputs are harvested crops, including crop residues, and grazed grass. Changes of

349

Integrating nitrogen fluxes at the European scale

Figure 16.3 Soil system nitrogen surplus for forest soils (forests, scrublands, heather) in EU-27 for the year 2000. The map shows total reactive N surplus for a grid of 1 km × 1 km, the values are in kg N per total pixel area [kg N km−2 yrâ•›−1 total area]. The pie diagram at the right side gives the split of surplus [Gg N yearâ•›−1, rounded to 10 Gg N yearâ•›−1] for EU27 into the loss pathways:€emissions of NH3, NOx, N2O and N2 and N leaching. The histogram shows the split of the N surplus [Gg N yrâ•›−1] into the loss pathways by country. Basis:€INTEGRATOR, 2009. Method:€Nitrogen surplus is estimated as the sum of gaseous nitrogen fluxes and nitrogen leaching. Nitrogen leaching is calculated in INTEGRATOR from the difference of total N input, via N-deposition, biological nitrogen fixation and manure input where relevant, and the previously estimated nitrogen losses, via uptake by plant growth, NH3 losses, nitrification/denitrification gas losses, and net nitrogen immobilization. However, a minimum nitrogen leaching rate is postulated which is obtained from the water flux and a concentration of 0.02 mg N l−1 (Stoddard, 1994). A check is made if the minimum N-leaching rate is achieved; otherwise this is obtained following pre-defined rules as described by de Vries et al. (2009).

soil Nr stocks can occur in both a positive direction (filling-up the nitrogen pool) and negative direction (depletion of the Nr pool). This is a transient process and can be reversed by chang� ing farm management. According to the soil-system approach, changes in soil Nr stocks adjust the accountable quantity of useful outputs. They have an equal impact on the Nr surplus, however, not being part of a detailed split of the fate of N-surplus (for a detailed discussion see Leip et€al., 2010a). Nitrogen surplus on forest soils is defined by analogy:€Nr-inputs are fertilizer application, atmospheric nitrogen deposition and biological nitrogen fixation, while Nr-outputs are nitrogen uptake by the plants and Nr immobilization in the soil. The input of Nr to agricultural soils is dominated by the input of mineral fertilizer. Worldwide, the production of mineral fertilizer is the most important source (about 65%) of the net increase of Nr in the environment. While already for the global nitrogen cycle, human influence is larger than the natural dimensions of the nitrogen cycle (Galloway and Cowling, 2002).

350

In Europe the anthropogenic effect is even stronger. Application rates of mineral fertilizer and manure per hectare of utilized agricultural land is shown in Figure 15.7 (de Vries et€al., 2011, Chapter 15 this volume). The pattern in the map in Figure 16.1, which shows the input of nitrogen for total surface area, is different from the maps presented in Chapter 15 as it gives an idea of the share of utilized agricultural area (UAA) across Europe. High shares of UAA up to more than 90% are found in intensive farming areas, such as the Po Valley in Italy, Central Spain, Western France and Romania (Leip et€ al., 2008). For other regions, such as Finland, the Baltic countries or mountainous regions, low shares of agricultural land of generally below 10% yield low N input data even though the application rates per hectare of cultivated land can reach high values, as is the case in certain Finnish regions.The range of manure input is very large, and covers values from 7 kg N per hectare UAA in Romania, to over 230 kg N (ha UAA)−1 in the Netherlands. For some countries, extensive rearing of ruminant animals predominates and

Adrian Leip

Figure 16.4 Atmospheric nitrogen deposition to in EU-27 for the year 2001. The map shows total reactive N deposition for a grid at 1 kmÂ€× 1 km, the values are in kg N per total pixel area [kg N km−2 yrâ•›−1 total area]. The pie diagram at the right side gives the split of N deposition to the different ecosystems [Gg N yrâ•›−1, rounded to 10 Gg N yrâ•›−1] for EU27:€coniferous forests, deciduous forests, cropland, seminatural land, and inland water surfaces as well as deposition to the coastal shelf and the deep ocean, which are not shown in the map. The histogram shows the split of N deposition [Gg N yrâ•›−1] by country. Basis:€EMEP MSC-W model, rv3_3, 2009. Method:€Atmospheric N-deposition is calculated with the European-scale EMEP MSC-W Chemical transport model (European Monitoring and Evaluation Programme, Meteorological Synthesizing Centre€– West). The EMEP model was designed primarily for the calculation of acidifying substances, ozone and particles over Europe (Simpson et al., 2003, see also www.emep.int; Simpson et al., 2011, Chapter 14). The chemical scheme uses about 140 reactions between 70 species (see Andersson-Sköld et al., 1999, and references therein), and makes use of the Equilibrium Simplified Aerosol Module (EQSAM) of Metzger et al. (2002) to describe equilibria between the inorganic aerosol components. Routine N-deposition fields from the EMEP model are available at ww.emep.int. The model uses a sub-grid calculation procedure (so-called ‘mosaic’ approach) to calculate deposition separately to 19 different land-cover categories, taking into account vegetation cover, phenology and surface-characteristics. Calculations of forest-specific deposition estimates, also exploring the role of forest soil-NO emissions from Kesik et al. (2005), were presented in Simpson et al. (2006).

the input of organic nitrogen occurs mainly through deposition of manure by grazing animals, e.g. 86% for Ireland and Greece according to CAPRI model estimates while this is only 20% and less in countries such as Poland, Slovenia and Denmark. Atmospheric deposition and biological nitrogen fixation account together for only 12% of total Nr-input to agricultural soils. Generally, biological N-fixation decreases with increasing Nr input due to increasing competitiveness of non-leguminous crops (Weigelt et€al., 2009). However, as CAPRI estimates biological N-fixation to be a constant fraction of above-ground crop Nr uptake (75% for leguminous crops and 5% for grass), this effect is not considered and leads to a likely over-�estimation of biological N-fixation in intensive regions such as North France and the Netherlands in comparison to extensive grassland areas such as in Poland and Romania. The size of the N surplus in the agricultural sector (Figure€16.2) is a measure of the sustainability of the agricultural production

process, since a surplus will eventually lead to shifting the environmental problems to other places outside of the agricultural sector or abroad, including a possible time lag. The contribution of nitrogen leaching to the fate of total N-surplus varies from 26% to 73%. N-leaching is mainly a function of soil texture:€heavy clay soils in Central and South Europe offer larger opportunities for denitrification than soils with high organic carbon content and sandy soils, which are less resistant to nitrogen losses to the water. Consequently, losses of N2 are negatively correlated to N-leaching. Fluxes of N2 are very difficult to measure and therefore treated in most models as residual loss-pathway. In the IDEAg, N2 flux estimates are based on results of the DNDC model, but constrained by estimates of NH3 fluxes calculated as in the MITERRA model (Velthof et€al., 2009, see also Chapter 15 this volume) and estimates of N-leaching, N2O and NOx fluxes from the same DNDC metamodel. Nevertheless, the ratio of N2/N2O has a range between

351

Integrating nitrogen fluxes at the European scale

Figure 16.5 Total NH3 emissions in EU27 around the year 2000. The map shows the sum of NH3 emissions from terrestrial ecosystems, industry and waste management for a grid of 1 km × 1 km, the values are in kg N per total pixel area [kg N km−2 yrâ•›−1 total area]. The pie diagram at the right side gives the split of total NH3 emissions [Gg N yrâ•›−1, values for agriculture are rounded to 10 Gg N yrâ•›−1] for EU27:€agricultural soils including manure application, manure in housing systems and manure management systems excluding manure application, forest soils, emissions from waste, mainly composting of solid waste, energy, and the chemical industry. The histogram shows the split of NH3 emissions [Gg N yrâ•›−1] by country. Basis:€Agriculture:€Indicator Database for European Agriculture V1, 2009; forest soils:€INTEGRATOR, 2009; industrial processes and waste management:€EDGARCIRCE (Van Aardenne et al., 2009). Method:€(i) Agriculture:€emissions are estimated for manure and mineral fertilizer as described by Weiss (2010). Manure emissions are estimated for animal housing and manure management systems and following application on the basis of a mass-conserving approach. NH3 loss factors are taken from the GAINS model for liquid and solid manure and the emission is reduced according to an assumed implementation level for NH3 emission reduction measures using again default GAINS data (Klimont and Brink, 2004; Velthof et al., 2009). Emissions from mineral fertilizer nitrogen are calculated separately for urea and non-urea fertilizers. (ii)€Forest soils:€emissions are calculated using a constant natural background flux (after Simpson et al., 1999). (iii) Industrial processes:€production data are from US geological survey statistics, UN industrial commodity statistics and data from SRI Consulting (2005). The emission factors are EMEP/EEA (2009). Emissions from industrial processes are allocated spatially based on point source maps for the most important source categories, and using population density for the remaining categories. (iv) Waste management:€the amount of solid waste composted is estimated based on national reports to the UNFCCC (2008) and on data from European Compost Network (ECN, 2008). The emission factor is from EMEP/EEA (2009). Emissions from the waste sector are spatially distributed based on human population density.

5 and 30 (EU27-average 11.6) in-line with current understanding of the nitrogen cycle (Butterbach-Bahl et€al., 2011, Chapter 6 this volume; Seitzinger et€al., 2006). Fluxes of NH3 contribute between 6% and 17% to the total soil N-surplus and depend on the type of manure or mineral fertilizer nitrogen applied, which is also country-specific. Countries with a high share of urea applied and/or a high livestock density have high losses of nitrogen to the atmosphere as NH3. Unlike agricultural soils, the most important nitrogen change for forest soils and soils under semi-natural land (see Figure 16.3) is Nr accumulation in the soil (about 50% for forest soils and 30% for semi-natural land). The most important Nr loss-pathways for forest soils are N2 and nitrogen leaching. As the map shows the N-surplus calculated per square kilometre total area, it shows the spatial

352

variation of two important factors:€the N surplus per hectare of forest area, which is mainly affected by the water balance and soil properties, and the forest area itself, which is particularly high in mountains and in Northern Europe. Therefore, the spatial pattern of nitrogen surplus in forests mimics to a certain degree the topography of Europe, and is as such in contrast to agricultural N surplus. NH3 emissions are not a significant loss pathway of Nr from forest soils, as the input of mineral fertilizer and manure to forest soils is negligible in many countries, but it accounts for one third of the Nr losses from rough grazing land. NH3 emissions include volatilization from urine, livestock manures (slurry and solid manure) and background emissions (see Simpson et€al., (1999)).

Adrian Leip

Figure 16.6 Total NOx emissions in EU-27 around the year 2000. The map shows the sum of NOx emissions from agriculture (both agricultural soils and manure in housing and manure management systems), forest soils, industrial processes, combustion (stationary and mobile) sources, and waste management (incineration) for a grid at of 1 km × 1 km. The values are in kg N per total pixel area [kg N km−2 yrâ•›−1 total area]. The pie diagram on the right side gives the split of total NOx emissions [Gg N yrâ•›−1, rounded to 10 Gg N yrâ•›−1] for EU27:€agriculture, forests, combustion in industry and residential combustion, industrial emissions, road and other transport. The histogram shows the split of NOx emissions [Gg N yrâ•›−1] by country. The map shows also the emissions from international aviation (red) and navigation (blue) which are not included in the national totals. Values are less than 40 kg N km−2 yrâ•›−1. Basis:€Agriculture:€Indicator Database for European Agriculture V1, 2009; forest soils:€INTEGRATOR, 2009; industrial processes, combustion and fugitive emissions, and waste management:€EDGAR-CIRCE (Van Aardenne et al., 2009). Method:€(i) Agriculture:€emissions are calculated with the DNDC meta-model as described in Britz and Leip (2009). In the Indicator Database for European Agriculture, a correction of the NOx fluxes is applied only if a closed N-budget cannot be obtained through adjustment of N2 fluxes (considered as the weakest term in the DNDC meta-model) and N-leaching within the bounds set. Then, the loss terms NOx, N2O, N2, and N-leaching are scaled to obtain a closed N-budget. (ii) Forest soils:€based on results with the model PnET-N-DNDC for European forest (Kesik et al., 2005) an NOx /N2O ratio of 1.25 is used. N2O emissions are estimated from a meta-model based on PnET-N-DNDC simulations (Kesik et al., 2005). (iii) Industrial processes:€production data are from statistics of the US geological survey, UN industrial commodity statistics and data from SRI Consulting (SRIC, 2005). The emission factors are from EMEP/EEA (2009). emissions from industrial processes are allocated spatially based on point source maps for the most important source categories, and using population density for the remaining categories. (iv)€Combustion and fugitive emissions:€fuel consumption data by sector and fuel type (stationary) or transport mode and fuel type (mobile) is obtained from International Energy Agency (IEA) statistics (IEA/OECD, 2007). Production of crude oil is estimated based on IEA statistics (IEA/OECD, 2007), and venting/flaring is estimated based on data from the Carbon Dioxide Information Analysis Center (CDIAC, 2008), supplemented by reporting of the countries to the United Nations Framework Convention on Climate Change (UNFCCC). The emission factors are based on IPCC (2006)), EMEP/EEA (2009) and Amann et al. (2007). The emissions from stationary combustion are spatially distributed using point source maps for power plants, steel production plants, and oil refineries, and maps on urban and rural population density for the other sectors. (v) Combustion transport:€emissions from transportation include road and rail transportation, domestic and international navigation, domestic and international aviation and other transportation. The fuel use by each transport mode and fuel type is from IEA statistics (IEA/OECD, 2007). The fuel use in international navigation is divided between sea and port activities of 15 ships types based on Dalsøren et al. (2009). Fuel consumption in aviation is divided between landing and take-off; climbing and descent; cruise; and super-sonic based on gridded data from the AERO2K project (Eyers et al., 2004). A detailed split of the fuel used in road transportation is used in the EDGAR database considering heavy and light duty vehicles, passenger cars, buses, mopeds, and motorcycles by applying country-specific fleet distribution calculated based on registration, number of vehicles, and driven vehicle kilometres from International Road Federation (IRF, 2007). The impact of emission control measures is calculated based on European emissions standards (EURO 0€– EURO 4) and other regional standards, with data from CONCAWE (2001) and EMEP/EEA (2009). The emission factors are based on EMEP/EEA (2009), EIPPC BREF, IPCC (2006) and scientific literature. Emissions from aviation are spatially allocated based on AERO2K project, and presented separately for domestic and international aviation. Emissions from road transportation are spatially allocated using road density map, weighted with population in the case of passenger cars. Emissions from international navigation are spatially allocated using a ship traffic density map of Wang et al. (2007). Other transport emissions are gridded using population density. (vi) Waste management:€the amount of solid waste incinerated without energy recovery is estimated based on the reporting of the countries to the UNFCCC. The emission factors for solid waste are from EMEP/EEA (2009). Emissions from the waste sector are spatially distributed based on human population density.

353

Integrating nitrogen fluxes at the European scale

Figure 16.7 Total N2O emissions in EU-27 around the year 2000. The map shows total N2O emissions from terrestrial ecosystems, industry, energy and waste for a grid of 1 km × 1 km. The values are in kg N per total pixel area [kg N km−2 total area]. The pie diagram at the right side gives the split of total N2O emissions [Gg N yearâ•›−1, rounded to 10 Gg N yearâ•›−1] for EU-27:€agricultural soils, manure management excluding manure spreading on soils, forest soils, energy (large scale and domestic), industry (mainly chemical industry), waste (wastewater treatment and other waste) and transport (road and non-road). The histogram shows the split of N2O emissions [Gg N yr−1] by country. Basis:€Agriculture:€Indicator Database for European Agriculture V1, 2009; forest soils:€INTEGRATOR, 2009; industrial processes, combustion, and solid waste management:€EDGARv4 (JRC/PBL, 2009); waste water systems:€Indicator database for European Agriculture V1, 2009. Method:€(i) Agriculture:€emissions are calculated with the DNDC meta-model as described in Britz and Leip (2009). In the Indicator Database for European Agriculture, N2O fluxes are corrected to obtain a closed N-budget only in case the correction of N2 fluxes (considered as the weakest term in the DNDC metamodel) and N-leaching alone is not possible within the bounds set. In this case, the loss terms NOx, N2O, N2, and N-leaching are scaled to obtain a closed N-budget. (ii) Forest soils:€emissions are estimated from a meta-model based on simulation results for European forest soils with the model PnET-N-DNDC (Kesik et al., 2005). (iii) Industrial processes:€production data are from statistics of the US geological survey, UN industrial commodity statistics and data from SRI Consulting (SRIC, 2005). Abatement of N2O emissions from nitric acid and adipic acid emissions is included based on the reporting of countries to the UNFCCC. The emission factors are from IPCC (2006). Emissions from industrial processes are allocated spatially based on point source maps for the most important source categories, and using population density for the remaining categories. (iv) Combustion and fugitive emissions:€fuel consumption data by sector and fuel type (stationary) or transport mode and fuel type (mobile) is obtained from International Energy Agency (IEA) statistics (IEA/OECD, 2007). The emission factors are based on IPCC (2006) and other sources. The emissions from stationary combustion are spatially distributed using point source maps for power plants. (v) Combustion transport:€emissions from transportation include road and rail transportation, domestic and international navigation, domestic and international aviation and other transportation. The fuel use by each transport mode and fuel type is from IEA statistics (IEA/OECD, 2007). The fuel use in international navigation is divided between sea and port activities of 15 ships types based on Dalsøren et al. (2009). Fuel consumption in aviation is divided between landing and take-off; climbing and descent; cruise; and super-sonic based on gridded data from the AERO2K project (Eyers et al., 2004). A detailed split of the fuel used in road transportation is used in the EDGAR database considering heavy and light duty vehicles, passenger cars, buses, mopeds, and motorcycles by applying country-specific fleet distribution calculated based on registration, number of vehicles, and driven vehicle kilometres from International Road Federation (IRF, 2007). The impact of emission control measures is calculated based on European emissions standards (EURO 0€– EURO 4) and other regional standards, with data from CONCAWE (2001) and EMEP/EEA (2009). The emission factors are based on IPCC (2006). Emissions from aviation are spatially allocated based on AERO2K project, and presented separately for domestic and international aviation. Emissions from road transportation are spatially allocated using road density map, weighted with population in the case of passenger cars. Emissions from international navigation are spatially allocated using a ship traffic density map of Wang et al. (2007). Other transport emissions are gridded using population density. (vi) Solid waste management:€the amount of solid waste composted and incinerated is estimated based on the reporting of the countries to the UNFCCC. Composting data are complemented with information from European Compost Network (ECN, 2008). The emission factors for solid waste are from IPCC (2006). Emissions from the waste sector are spatially distributed based on human population density. (vii) Waste-water systems:€emissions from nitrogen in effluents (0.005 kg N2O-N kg−1 N) as well as emissions from advanced sewage treatments systems (3.2 g N2O person−1 yrâ•›−1) are from IPCC (2006).

354

Adrian Leip

Figure 16.8 Total reactive N input to the hydrosphere (rivers and groundwater) in EU-27 for the year 2002. The map shows total Nr point sources from sewerage systems and diffuse sources from agriculture and forest soils and atmospheric Nr deposition to inland water surfaces for a grid at of 1 km × 1 km. The values are in kg N per total pixel area [kg N km−2 yrâ•›−1 total area]. The pie diagram at the right side gives the split of N input to the hydrosphere [Gg N year−1, rounded to 10 Gg N yearâ•›−1] for EU27:€point sources (sewage systems) and diffuse sources (agriculture leaching, run-off, and forest soils). The histogram shows the split of Nr input to the hydrosphere [Gg N yearâ•›–1] by country. Basis:€Sewage systems and agriculture:€Indicator Database for European Agriculture V1, 2009; forest soils:€INTEGRATOR, 2009. Method:€(i) Sewage systems:€nitrogen in agricultural products is estimated with CAPRI, N from fish is obtained from Eurostat fish statistics and a mean N-content of 2.6%. Non-consumed proteins (waste) are assumed to be 30%. N retention in sewage systems is calculated from the percentage of people connected to sewage systems (for rural and urban population) with mechanic, biological or advance treatment and corresponding retention efficiencies is obtained from Van Drecht et al. (2009). According to IPCC (2006), industrial wastewater is assumed to be 25% of domestic nitrogen (in advanced treatment systems). Spatial downscaling is done according to the population density. (ii) Agriculture:€total nitrogen leaching and runoff of nitrogen from agricultural soils is estimated from the DNDC-CAPRI meta-model (2009) and integrated into the IDEAg according to Leip et al. (2009b). In addition to the data presented in Figure 16.2, the data presented here include run-off from livestock housing and manure management systems, which are estimated using the MITERRA approach (Britz and Witzke, 2008; Velthof et al., 2009). Spatial distribution is in accordance with the livestock density per grid cell by animal group. (iii) Forest soils:€nitrogen Nr leaching from forest soils is estimated as described in Figure 16.3. (iv) Atmospheric deposition:€Atmospheric Nr-deposition is calculated with the European-scale EMEP MSC-W Chemical transport model (European Monitoring and Evaluation Programme, Meteorological Synthesizing Centre€– West) as described in Figure 16.4.

N2O emissions take a larger share of the total denitrification losses with generally narrow N2/N2O ratios around 3–4 in forest soils, but higher values for rough grazing. The forest type is very important in determining the rate and also the type of Nr emissions, due to its impact on litter quality and soil pH. Pilegaard et€ al. (2006) carried out a detailed study of Nr fluxes in 15 forest sites throughout a year and across Europe and found that coniferous forest soils had much higher NO emissions than deciduous forest soils. On the other hand, N2O emissions were slightly higher in deciduous forests compared to coniferous ones. Atmospheric transport and atmospheric deposition is discussed in detail by Simpson et€al. (2011, Chapter 14 this volume). In contrast to the figures presented there, we show here

absolute deposition fluxes per area of grid cell (Figure 16.4). For the contributions of N-deposition by ecosystem shown in the pie diagram and the histogram, the share of the various ecosystems at the grid scale is taken into account also. Even though cropland covers a smaller area in EU27 than forests, about 40% of deposition fluxes on the continent go on cropland, more than on forest land, because European forests are predominantly located in areas with smaller atmospheric Nr concentrations. Forests are the main receptor ecosystem for atmospheric Nr deposition in Scandinavian countries (Finland 70% and Sweden 66%) and alpine regions as Austria (55%). Deposition over semi-natural land is important in Mediterranean countries (Greece 35%, Spain 30%), but also in countries where conditions are too wet for other land uses

355

Integrating nitrogen fluxes at the European scale

Figure 16.9 Total land productivity of agricultural land in EU-27 for the year 2002. The map shows total agricultural land productivity for a grid of 1 km × 1 km. The values are in kg N per total pixel area [kg N km−2 total area]. The pie diagram at the right side gives the split of N productivity [Gg N yrâ•›−1, rounded to 10 Gg N yearâ•›−1] for EU27:€crop products, fodder (fodder maize fodder beet and other fodder on arable land), grass and other (flowers, nurseries, etc.). The histogram shows the split of N productivity [Gg N yrâ•›−1] by country. Basis:€Indicator Database for European Agriculture V1, 2009. Method:€Land productivity is based on CAPRI regional statistics on agricultural and grassland yield with crop-specific nitrogen contents. The data include crop residues that are returned to the soil. As statistics on grassland yields are scarce, these are estimated within CAPRI on the basis of the energy requirement of livestock and energy supply from other feed available (concentrates, fodder). Spatial allocation is done on the basis of simulation results for crop-potential yields (Genovese et al., 2007) under given climatic and soil conditions.

(UK 54%, Ireland 76%). Croplands are the main receptor for atmospheric Nr deposition in Denmark (58%), Poland (52%) and Lithuania (51%). The map shows also deposition fluxes over European shelf regions and deep ocean waters, which are of equal magnitude. Together they receive about 40% of the Nr deposited over the continent. The size of the total deposition is largely controlled by agricultural emissions of NH3, which are transported over shorter distances than NOx and strongly influence local deposition rates. High deposition regions with deposition rates over 20 kg N ha−1 yr−1 are therefore associated with intensive livestock production, such as in the Netherlands and the Po Valley in Northern Italy. High deposition rates of >10 kg N ha−1 yr−1 are found almost throughout Central Europe.

16.3.2╇ Emissions of reactive nitrogen to the atmosphere and hydrosphere Emissions of reactive nitrogen to the atmosphere and the hydrosphere are caused by both agricultural and other land-use

356

activities as well as by fuel combustion and industrial processes. Detailed maps of the spatial distribution of total emissions of Nr to these media are helpful for understanding the occurrence of hot-spots and could be the first step in identifying appropriate and well-targeted mitigation measures. We show here the most important fluxes of Nr, i.e. total emissions of NH3 (Figure€16.5), total emissions of NOx (Figure 16.6) and total emissions of N2O (Figure 16.7) to the atmosphere, as well as total Nr emissions to the hydrosphere (Figure 16.8). About 95% of NH3 emissions originate from the agriculture sector. The contribution of agriculture is rather stable across the countries in EU27, with a highest contribution in countries such as Ireland, Spain or Hungary (98%). The importance of soil/field emissions (from field application of manures, grazing and fertilizers) versus emissions from manure management systems (animal housing and manure storage) varies between 40% of emissions from soils in Denmark and 67% of emissions from soils estimated for Ireland. The reason is, of course, the importance as well as the structure of the livestock sector (grazing versus housing of the animals). As manure has higher

Adrian Leip

Figure 16.10 Total human N consumption of reactive nitrogen in EU-27 for the year 2002. The map shows human nitrogen consumption of agricultural products including food waste for a grid at of 1 km × 1 km. The values are in kg N per total pixel area [kg N km−2 total area]. The pie diagram at the right side gives the split of human N consumption [Gg N yrâ•›−1, rounded to 10 Gg N yearâ•›−1] for EU-27:€crop products, animal products, fish products (incl. shellfish), and the N in non-edible products (peelings, bones) and food waste. The histogram shows the split of human N consumption [Gg N yrâ•›−1] by country. Basis:€Indicator Database for European Agriculture V1, 2009. Method:€Consumption is based on CAPRI regional statistics. Spatial allocation is done on the basis of livestock density.

volatilization fractions for NH3 than mineral fertilizer, we find hotspots of NH3 fluxes where the livestock density is high, both in intensive production systems with housed animals such as in the Po Valley, Italy, Denmark, and the Netherlands, as well as in regions with a predominance of grazing animals such as in Ireland. Intensive animal production systems have mainly developed in the vicinity of large metropolitan areas, as the banlieue of the Paris area, but also west of Berlin. The metropolitan areas themselves however, are usually low emitters of NH3. Hotspots of NH3 fluxes are found in the major European plains such as the Po Valley, North Germany, the Netherlands and Bretagne, but also in hilly regions such as the Alps in Southern Germany or also, for example, the north of Andalucia. NOx emissions arise mainly from industrial and energysources accounting for 96% of emissions. Only 4% of NOx emissions are formed biogenically, with 3% from agriculture and 1% from forests, according to the estimates presented here. Butterbach-Bahl et€al. (2008) give a range of 48.8–128.9 Gg N yr−1 for NOx emissions from agricultural soils depending on the methodology used, which matches well with the presented number of 76 Gg N yr−1 from agricultural soils (including

emissions from pasture), to which about 32 Gg N yr−1 emissions of NOx from manure management systems are added. For forest soils, Butterbach-Bahl et€al. use the process-based model Forest-DNDC as the only approach and give a number of 75 Gg N yr−1, which is about twice as large as the 32 Gg N yr−1 in our estimate from the INTEGRATOR model. The numbers do not include NOx emissions from forest fires and burning of agricultural residues, which are available in the Global Fire Emissions Database (GFED) (van der Werft et€al., 2010). According to these data, about 11 Gg of NOx-N are released to the atmosphere by burning of woodland and forests (about 75%) and agricultural waste (25%). The bulk of NOx emissions originate from combustion processes€– about equal amounts of NOx-N are emitted from stationary combustion (industry and residential combustion) and mobile combustion (mainly road transport) with 42% of total emissions or about 1.5 Tg N yr−1 each. It is important to note that emissions from aviation and navigation are divided between domestic and international transport. While the emissions from international navigation/aviation are included in the spatially allocated emissions, they do not appear in national

357

Integrating nitrogen fluxes at the European scale

Figure 16.11 Total livestock N consumption of reactive nitrogen in EU-27 for the year 2002. The map shows human nitrogen consumption of agricultural products including food waste for a grid at of 1 km × 1 km. The values are in kg N per total pixel area [kg N km−2 total area]. The pie diagram at the right side gives the split of livestock N consumption [Gg N yrâ•›−1, rounded to 10 Gg N yearâ•›−1] for EU-27:€grass (cutting and grazing), fodder (fodder maize, fodder beet and other fodder on arable land), crops (cereals and other non-fodder crops, mainly leguminous crops and oilseeds), concentrates (energy-rich and protein-rich concentrates, oilseed cakes, milk powder, molasse, etc.) and other (straw, animal products). The histogram shows the split of livestock N consumption [Gg N yrâ•›−1] by country. Basis:€Indicator Database for European Agriculture V1, 2009. Method:€Consumption is based on CAPRI regional statistics and Eurostat national statistics for fish products with crop-specific nitrogen contents. Spatial allocation is done on the basis of population or livestock density for the consumption of food or feed, respectively.

totals. Military emissions, which are estimated to be small, are not included. Additional details on the transport sector are given in Section 16.4. The highest share of industry-combustion to total NOx emissions is observed in Central-Eastern European countries such as Romania (43%), Poland (47%), or the Czech Republic (48%). In Poland, residential combustion is also considerable (21% of total NOx emissions). The importance of residential NOx emissions depends on the fuel mix and the energy efficiency; the share of residential emissions to total NOx emissions in Italy and France are relatively high with 15% and 20%, respectively, and low in Finland and Estonia (both 7%). Accordingly, the map of total NOx emissions shows high values in centres of energy-intensive industry, such as Sachsen-Anhalt in Germany, North Italy, the Netherlands, or along intensive traffic lines. In many cases, both have been developed along major river streams as can be observed for the Rhine where we find large industry complexes but also an important traffic axis. The main sources of N2O are biogenic sources including agricultural soils, manure management, as well as forest soils and

358

the waste sector, accounting together for 74% of all N2O fluxes in EU27. Processes leading to N2O formation in soils, as well as the upscaling of N2O fluxes from these sources for both agricultural and non-agricultural terrestrial sources, are discussed in Chapters 6 and 15 (Butterbach-Bahl et€ al., 2011, this volume; De Vries et€al., 2011, this volume). Agricultural soils are the main source contributing 73% of biogenic N2O sources with manure management systems, forests and the waste sector contributions at 7%, 8% and 2%, respectively. The contribution of the waste sector to biogenic N2O emissions ranges from 1%, for example in Finland, the Netherlands and Poland, to 16% in Denmark. Agricultural soils contribute to almost 90% of biogenic N2O emissions in Hungary, Greece and Finland, while the smallest contribution of soils being estimated for Estonia (45%). The most important non-biogenic sources of N2O are industrial processes, which are not caused by fuel combustion, but by the industrial processes themselves. The chemical industry is an important source of N2O emissions accounting for 20% of EU27 N2O emissions. Globally, nitric acid production is the most important N2O source within the chemical industry,

Adrian Leip

followed by adipic acid, caprolactam and glyoxal production. As industrial plants are point sources with an uneven distribution, the significance of industry ranges from 0% of national N2O emissions in Poland, Portugal and the UK to more than 60% in Denmark. Additionally, N2O is used in anaesthesia and in aerosol spray cans. The spatial structure of N2O emissions is thus a combination of the one observed for NH3 (mainly livestock) and NOx (mainly energy and transport). Thus, we find those areas which were already identified for both other gases such as the Po Valley, the Netherlands, and Sachsen-Anhalt, but also identify high fluxes from rural agricultural areas such as Hungary or Poland. The map of N2O fluxes is further complicated by its large dependency on environmental conditions. The soil type is a particularly important factor, which tends to increase fluxes in Northern Europe, where soils with a high content of organic carbon prevail, and leads to lower fluxes in Southern Europe with soils of lower organic carbon content, but this trend is overlaid with the influence of soil moisture and temperature (see Leip et€al., 2010b). The input of reactive nitrogen to European aquatic systems is dominated by point sources through sewage systems, including industrial fluxes and diffuse sources from agriculture. Thus, the Nr load to rivers is highly correlated with population density as can be seen from Figure 13.10 in Billen et€al. (2011, Chapter 13 this volume). This is superimposed on the pattern of agricultural nitrogen leaching, which is similar to the spatial pattern of agricultural nitrogen surplus shown in Figure 16.2. Runoff in agricultural systems from stables or manure management systems and leaching from forest soils are estimated to be of minor importance. Data on the distribution of sewage systems are available for most European countries from EUROSTAT, EEA (1998), Wieland (2003) and Jeppsson et€al. (2002). The overall values for these removal fractions for a country are calculated as the weighted average of the four classes as compiled for the IMAGE model by Van Drecht et€al. (2009). In the IMAGE model it is assumed that Nr emitted by people not connected to sewerage systems will be retained and does not enter the hydrosphere. In rural areas of Europe, however, the majority of human wastes will be discharged to unmonitored small sewage plants, or to septic tank/soakaway facilities. In the UK, for example, these are the dominant point sources in most rural catchments, in comparison to the major sewage treatment plants in larger towns or cities. We assumed that the Nr from such unmonitored small sewage plants or septic tanks/soakaways undergoes a ‘biological-treatment’ like transport to the river/groundwater system. A detailed study on the nitrogen removal efficiency of sewage treatment systems in the UK (Johnes, 1996) takes in to account the fact that although the process is optimized in larger sewage treatment plants in urban areas, in many of the older treatment systems, in rural areas and in most of the coastal towns and villages, volumes treated have exceeded initial design capacity through local population growth or migration. As a consequence, only mechanical treatment is available for part of the Nr, with little biological treatment of wastes, particularly

during cold, wet periods. The same is true for peak seasons, for example in tourist regions (e.g. alpine ski resorts in the winter; beach and lake vacation regions in the summer and spring), when the Nr input to sewage systems exceeds their capacity.

16.3.3╇ Secondary nitrogen indicators In this section we present secondary reactive nitrogen indicators that answer two important questions:€what is the amount of proteins (nitrogen) that European crop- and grasslands can currently produce and what is the amount of proteins (nitrogen) that European inhabitants (humans and livestock) consume? The first map shows total land productivity (Figure 16.9), defined as the sum of harvested crops, and grazed biomass according to the soil system budget. Included in the total productivity are crop residues that are or are not used for other purposes (as animal feed or bedding material, biofuels or burned), which is in contrast to the definition of nitrogen autotrophy (see Billen et€ al., 2011, Chapter 13 this volume), which gives its production of food and feed only (harvested and grazed products; Billen et€al., 2007, 2008). The second and third map show human consumption of nitrogen (Figure 16.10) and the consumption of nitrogen by livestock (Figure 16.11). Both are split by the main proteinsources, i.e. crops, livestock products and fishery products for human consumption and grass, crops/fodder and concentrates for livestock consumption. The sum of both maps gives the nitrogen heterotrophy in Europe. A large part of rural Europe is characterized by a high degree of regional specialization of agricultural activities. In most traditional agrarian systems in Europe, livestock farming used to be a critical component, providing a way to ensure cropland fertility by bringing to it, Nr extracted from seminatural N2 fixing areas in the form of manure. During the past half century, in parallel with increasing urbanization, many lowland rural areas of Europe have shifted either towards exclusive crop production, with very little cattle breeding, or to intensive livestock farming, supported to a large extent by feed importation. Mixed farming areas are restricted to highland or mountainous regions. As a result of this specialization, the exchange of food and feed over long distances has considerably increased and now often represents quite a significant share in the nitrogen budget of regions, or even of countries. Still, the total productivity of agricultural land is particularly high in the hinterland of large metropolitan areas (e.g. Paris, Berlin, London) and in areas of intensive animal production (the Netherlands, Belgium, Po Valley, Italy, many areas of England and Ireland). About 45% of the productivity yields crop products, while the other 50% of nitrogen are distributed over dedicated fodder production (20%, fodder maize and fodder beet) and grassland (25%). The significance of grass varies largely between 5% in Denmark and 60% in Ireland while the importance of feed production varies between 10% (Bulgaria, Greece) to over 40% (Sweden, Estonia). Human consumption obviously peaks in metropolitan areas like Paris, London and Berlin which are clearly visible in the map and consume more than 15 Mg N yr−1 per square kilometre. Outside

359

Integrating nitrogen fluxes at the European scale

these hotspots, human Nr-consumption is generally between 200 and 2000 kg N yr−1 km−2. There is about 36% consumption of crop products and 33% consumption of animal products, with a small fraction (3%) of fish products consumed. The last third is estimated to be either non-edible or wasted. In many central-western countries intake from animal proteins dominales while in countries like Greece, Romania and Bulgaria about 70% of protein intake is from vegetable sources. Livestock Nr-consumption follows the livestock density and thus the same regions with high/low rates can be identified that were already visible particularly in the NH3 map (Figure€16.5). The consumption of crop products by livestock is more than double the consumption of crop products by the human population, a value that increases to 3.8 and 9.3 in Ireland and Denmark. About the same amount of concentrates is being fed to animals as crop products, and about 40% of animal feed is stemming from grass and fodder. However, we find large differences in animal nutrition across the countries with some countries feeding almost 50% with concentrates, as in the Netherlands, Portugal and Denmark, while this share is only 12%–13% in Latvia, Romania countries and the Czech Republic. The share of grass is naturally high in mountainous countries like Austria, but also in lowland countries such as Ireland and the UK. In all cases, grass contributes to more than 40% of the protein requirements of animals. Urban areas, where food is consumed but not produced, are obviously heterotrophic. Rural regions specialized into crop production are autotrophic and export nitrogen as food and/ or feed, while those characterized by intensive animal farming sustained by imported feed are usually heterotrophic as in those regions the import of feed is usually not balanced by exported food products.

16.4╇ Integrated nitrogen budgets Integrated nitrogen budgets are defined here as the quantification of all major nitrogen fluxes across sectors and media within given boundaries, and fluxes across these boundaries, on an annual basis. They provide a valuable tool for optimizing the benefits of policies addressing imbalances evident in the nitrogen cascade. These policies have often been designed to achieve a specific goal, neglecting unwanted side effects such as pollution swapping. Integrated nitrogen budgets per se do not directly give a quantification of the risk of such pollution swapping effects, as these are determined by mechanistic effects and require an understanding of the dynamics of the nitrogen fluxes. However, in many instances, integrated nitrogen budgets can give a good indication of where Nr pollution is most severe and where swapping problems from one medium to another might occur. Here, a standardized integrated Nitrogen Budget (iNB) approach has been implemented to derive a suite of national integrated nitrogen budgets (NiNB) as well as the European Nitrogen Budget (ENB). Each of the NiNBs has been compiled by national experts from each country, using data available at national scale. For each iNB, five sectors are differentiated:€industry and energy, transport, agriculture, forestry and

360

natural terrestrial ecosystems, and waste. Each sector has a pool of Nr and is connected to the others by three transport media:€ the atmosphere, the hydrosphere, and the consumers who transport, for example, agricultural products to the waste management systems. Pools of Nr are hard to quantify and are often not incorporated; however stock changes are important indicators to detect possible accumulation or depletion of Nr and their implications for soil productivity, biodiversity or the development of human health problems. Conceptually, the ENB contains the same kind of information as NiNBs and they can thus be discussed together. From a methodological point of view, however, the ENB is largely model-based (see Section 16.4.2) as robust and consistent estimates for many elements of a NiNB at the European level are not currently available.

16.4.1╇ National integrated nitrogen budgets National integrated nitrogen budgets (NiNB) help in visualizing the main elements of the N cascade within a country into a figure that might transmit its main messages at a quick glance, but nevertheless contains sufficient detailed information for further analysis. Therefore, a NiNB is regarded as a very efficient policy instrument and an important tool to help prioritize policies. In particular, NiNBs can serve five objectives:€(i) they are an efficient instrument for visualizing the N cascade and its potential impact and thus help to raise awareness; (ii) NiNBs provide policy makers with information for developing efficient emission reduction measure; (iii) more importantly, they can provide a tool for monitoring the impact and environmental integrity of implemented policies; (iv) NiNBs are useful for comparisons across countries; and (v) they can help pinpoint knowledge gaps and thus contribute to improving our scientific understanding of the N cascade. Often, NiNBs have to rely on information of different origin and quality, and therefore it may not be possible to ‘close’ the budget for one or several sectors. N fluxes presented in NiNBs are ideally based on a sufficiently dense network of observational data or on detailed models calibrated and validated on national conditions; however, often data gaps have to be filled from simpler models of a broader scope such as the models used for the European Nitrogen Budget. To build a NiNB is thus a challenging task and many elements of a budget will only be quantifiable within a very high uncertainty range, for example the amount of nitrogen denitrified and released as the stable and harmless N2 gas; or sedimented and stored for potential future release in the oceans. The magnitude of the uncertainty itself is usually unquantified. Despite these difficulties, NiNBs have been developed for some countries or are in the process of being developed. • Switzerland formulated environmental targets for agriculture in 1996 based on the observation that additional efforts were required to minimize pollution of soil, air and water and to maintain biodiversity. Measures in the agriculture sector were found to be particularly cost-efficient. The recommendations built on the Swiss N-budget that had been

Adrian Leip

•

•

•

•

•

developed for the year 1994. The Swiss N-budget was updated for the year 2005 and published by the Federal Office for the Environment (BAFU, 2010). The Netherlands is a country facing significant Nr pollution problems, such as particularly high nitrate concentrations in the groundwater, as well as a decrease in biodiversity and forest vitality, high atmospheric NOx and NH3 concentrations leading to human health effects, and algal blooms in the North Sea (Erisman et€al., 2005). The Dutch nitrogen budget was estimated by Erisman et€al. (2005) on the basis of an analysis by van Grinsven et€al. (2003) and also proposed a list of measures to address the Nr pollution problems in the Netherlands. In Germany the Federal Environment Agency (UBA) released a draft national nitrogen budget as background information to the Integrated Strategy for the Reduction of Nitrogen Emissions in April 2009 (Umweltbundesamt, 2009a,b), motivated by the fact that despite major efforts most environmental targets (halting loss of biodiversity, national emission ceilings for NOx and NH3, concentration of nitrate in drinking water, mitigation of global climate change) appeared unlikely to be met and that only an integrated approach would support the development of cost-efficient and effective solutions. Whereas the strategy contains a set of measures, the budget contains very detailed information on all quantifiable nitrogen fluxes across sectors and interfaces between environmental media above 1 Gg N yr−1. In France, the construction of the nitrogen budget has been initiated with the aim of developing an overarching vision of the nitrogen cascade between industrial sectors and environmental compartments through cooperation between various French research institutes and agencies. The development of the N-budget is ongoing and results are preliminary. The United Kingdom has built a national N-budget using the€ iNB approach, based on a detailed N-budget for agriculture following the OECD approach (DEFRA, 2008), national scale N flux modelling to freshwater and coastal systems and to and from the atmospheric N pool, and published data on non-agricultural sectoral fluxes. Further work is warranted to refine and update the initial budget presented here. The Czech Republic has launched a project to estimate all N-fluxes following the German example. To that purpose, the Czech Hydrometeorological Institute (CHMI) is cooperating with the Ministry of Agriculture and with the Central Institute for Supervising and Testing in Agriculture. CHMI is providing a range of emission and deposition data. Other institutions calculate N fluxes with regard to, for example, feed, manure, agricultural products, waste and leaching. Cooperation with the Institute of Geology and other institutes is planned.

Other countries, like Turkey have recently started developing a national N-budget.

Figures 16.12–16.17 show the national integrated nitrogen budgets for countries in Europe available to date. Each NiNB is constructed from nationally available information and thus the budgets are not directly comparable. For example, river export has not been estimated in Germany, while it constitutes a significant flux in the Netherlands and Switzerland. Additional details on the data sources of the NiNBs and the outcomes of the respective projects are given in the supplementary information (see supplementary material Chapter 16, Section A). As the NiNBs are not constructed with a harmonized or even comparable methodology, a comparison of single flux estimates must be done with care. Nevertheless, they highlight the general differences across the countries, as can be seen by ranking emissions by sector and dominant Nr form of emissions. Such an assessment is shown in Figures 16.18–16.20. For example, nitrogen fluxes in the Netherlands are dominated by industrial Nr fixation. The export of Nr, mainly as fertilizer, is by far the largest N-flux and feed-imports are higher than the input of mineral fertilizer. By contrast, in Switzerland the combined estimate of atmospheric Nr deposition plus biological N fixation to agriculture is higher than the input of mineral fertilizer and feed imports. Both countries have an important exchange of Nr with other countries or the sea through river flow. In the Netherlands, the Nr transit through the country as import and export are roughly the same. However, in Switzerland, river export is the single most important sink for the country exporting more Nr than is applied to agricultural soils as mineral fertilizer. A summary of the national N-budgets presented above by main compartment or sector (Table 16.2) shows that balance is not closed for most sectors. These N-budgets give aggregated data for countries, without a spatial dimension such as was shown in the key maps above. Not all nitrogen fluxes have been (or could be) estimated yet, and, secondly, the data for different compartments and sectors have been taken from the best available, but to some extent inconsistent, data sets. Examples include atmospheric deposition (which in many countries has been obtained from the EMEP model) and atmospheric emissions, which can have data sources which are partly inconsistent with the information used in the EMEP model. The main gap between the sum of Nr emissions to and removals from the atmosphere is due to fluxes of molecular nitrogen (N2) through N fixation and denitrification, which are difficult to estimate and have not been quantified for many countries/compartments. Therefore a negative balance is observed for most countries. The agriculture sector, which is in many countries one of the best-described sectors as gross nitrogen balance calculations have been made with the OECD methodology (OECD/ Eurostat, 2003), ideally gives a closed soil N-budget, yet the link to consumers and/or the industry has often been obtained from different sources. Un-quantified stock changes in agricultural soils can potentially account for part of the positive (accumulation) or negative (depletion) balance in agriculture. The French NiNB does not yet include fluxes from agriculture to the consumer or to the hydrosphere and the data show therefore a considerable gap.

361

Integrating nitrogen fluxes at the European scale

Figure 16.12 A national integrated nitrogen budget for Switzerland, derived from a study that was carried out between 1994 and 1996 based on a mandate of the Departments of Economic Affairs and Home Affairs of the Swiss government. The project aimed at identifying the most important N fluxes between the all compartments, at assessing the fluxes with respect to the exceedance of effects-based environmental and health quality criteria, and at elaborating a strategy for a stepwise reduction of emissions of reactive nitrogen. The results of the work of the project group are summarized in the report. “Strategy for the Reduction of Nitrogen Emissions” (BUWAL, 1996).

One of the largest fluxes is the generation of Nr in the industry and energy sector. This Nr has three important sources: (a) nitrogen fixation with the Haber–Bosch process, (b) release of nitrogen from fossil energy carriers such as coal, and (c) thermal generation of Nr at high temperatures during the burning process. For the NiNBs presented, total industrial/ energy N-fixation has been estimated as the difference between total estimated N-outputs and inputs in these sectors, to give an indication of the order of magnitude of this N-flux. A positive N-balance is found for the consumers. While food input and N-output to sewage systems is quantified in most cases, the underlying assumptions may differ. Not all biomass produced is edible; not all edible parts will be consumed; and assumptions on the fate of the produced biomass are associated with considerable uncertainty (see discussion on the European Nitrogen Budget). The waste-streams are often poorly quantified in the NiNBs presented. The consumer has a central role in national N-budgets:€with excess-supply in most European countries the incentives for ‘nitrogen-efficient’ behaviour are not currently well developed to affect consumers’ behaviour. The choices of consumers nonetheless steer the societal machinery with major consequences for the nitrogen budget. In addition to input of nitrogen as biomass, the consumer receives also significant amount of industrial nitrogen, as plastics, pigments or other chemical products. The

362

fate of these products is not yet quantified and appears as a (positive) balance as these substances accumulate in the anthroposphere. An overview of emissions to the atmosphere is given in Figure 16.18. Generally, agriculture is the main emitter of Nr to the atmosphere and to the hydrosphere (Figure 16.19), but taking all combustion sources together (incl. industry), atmospheric emissions from combustion are roughly at the same level as agricultural emissions, with the exception of the United Kingdom. The United Kingdom is the country where industrial and energy emissions play the biggest role:€emissions to the atmosphere from combustion sources are almost twice as large as the emissions from agricultural sources. Nevertheless, as extensive agricultural activities involves high mineral fertilizer input (more than 1 Tg N yr−1), many livestock and high precipitation, this country estimates the highest leaching rate to waters (almost 0.5 Tg N yr−1). Emissions to the hydrosphere have not been estimated yet in the French N-budget, however, this country is characterized by extensive agricultural activity reflected both in the input of mineral fertilizer nitrogen (2.5 Tg N yr−1) and high atmospheric emissions (> 0.7 Tg N yr−1). The high land productivity throughout almost the whole territory of France (see also Figure 16.9) makes the country independent of feed-imports (only 7% of mineral fertilizer-input or 175 Gg N yr−1).

Adrian Leip

Figure 16.13 The Dutch national nitrogen budget is based on information from different publications, mainly originating from the National Bureau of Statistics. The data shown represent the average situation of 1995, 1997, 1998 and 1999 (van Grinsven et al., 2003).

Figure 16.14 The national nitrogen budget for Germany has been calculated by the Federal Environment Agency (Umweltbundesamt, 2009). The data is compiled of official, national emission, deposition and flux data sets for the years 2000–2004. The most important fluxes are emissions to atmosphere, deposition, input into hydrosphere and following export to coastal ecosystems.

363

Integrating nitrogen fluxes at the European scale

Figure 16.15 National nitrogen budget for France calculated with data gathered between 2003 and 2007 (Personal Communication Groupe de travail français sur l’azote réactif, or French working group on reactive nitrogen). The construction of the French national N-budget is an ongoing process, and the figures are therefore preliminary results which are bound to be modified.

The split of atmospheric emissions over the three reactive nitrogen gases NOx, NH3, and N2O reflects again the weight of the energy versus the agriculture sectors of the considered countries (see Figure 16.20).

16.4.2╇ The European Nitrogen Budget (ENB) Galloway et€al. (2008) formulated five vexing questions that should guide the direction of nitrogen research in the near future. The first of them relates to the ultimate fate of reactive nitrogen, and particularly the role of denitrification in soils and freshwater systems that are not well constrained. Based on the information presented above and in previous chapters, the development of an integrated N-budget for Europe is attempted. This European Nitrogen Budget. It covers the territory of the EU27 countries, with the exception of Malta and Cyprus (limited by the availability of data from the Indicator Database for European Agriculture). The European Nitrogen Budget, in contrast to most national N-budgets, is almost completely model-based, combining a model for agriculture, forestry, industrial emissions and atmospheric deposition into a common framework as given in Table€16.1. The restriction to a few models reduces the number of conflicting data as each of the models ensures consistency in the data sets used and Nr fluxes estimated. Inconsistencies at the interfaces between the models can not be excluded, though

364

they are usually small as the models deal with complementary sectors. For example, the nitrogen deposition fluxes to agricultural and (semi-) natural soils obtained from the latest simulations of the EMEP model for the year 2000 differ from the figure used in the IDEAg and INTEGRATOR models, based on slightly older versions. This points towards the need for a better fine-tuning and integration of the models, even though the differences are not large. Surprisingly, the atmospheric compartment shows a good match of total Nr emissions with total Nr deposition plus net export of nitrogen, even though the emission data are obtained from the above-mentioned models and the EDGAR-CIRCE database for industrial and energy sources, which are to some extent independent of the emission data used for the EMEP model. We are not aware of a Europewide model estimating Nr fluxes for surface waters and coastal areas and here the ENB had to be complemented by individual flux estimates taken from literature. In this chapter, we focus on nitrogen fluxes in the present time (year 2000). Nevertheless, a European nitrogen budget for the year 1900 has been developed too and is presented in the supplementary information (see supplementary material Chapter 16, Section B). When looking at the ENB, one has to keep in mind that the numbers presented are associated with large uncertainties which are difficult to quantify. The extent to which errors or biases in input data, model assumptions and model parameterizations propagate to the aggregated model output is hard to assess. In

Adrian Leip

Figure 16.16 The integrated nitrogen budget of the United Kingdom builds on the UK TAPAS modelling for agriculture (DEFRA), providing data on mineral fertilizer application rates and manure management and on the national food and feed balance modelling (DEFRA, 2008), national N flux modelling to freshwaters and the coastal zone (Johnes and Butterfield, 2002), the UK national emissions inventory and atmospheric transport and deposition modelling and a range of literature sources for UK waters. The data sets are not, however, co-incident in time, and the present budget represents the period from 1995–2005. Work is currently underway to update the budget.

the absence of sufficient observational data that allow the construction of independent estimates, two routes are possible: (i) the comparison of model results driven by the same set of input data (see de Vries et€al., 2011, Chapter 15 this volume) and (ii) a systematic assessment of the uncertainty of input data and model structural parameters and their impact on the model outcome at the large scale with Monte Carlo analysis. Both routes are currently followed in the NitroEurope Integrated Project and the results will shed some light on the reliability of estimated Nr fluxes. Additionally, for some Nr fluxes, it is indeed possible to use data from atmospheric concentration measurements to quantify the strength of total Nr emissions for some components (e.g. N2O) using inversion tools which are independent of the models used in the construction of the ENB; also this route is followed in the NitroEurope project (Sutton et€al., 2007). Figure 16.22 gives the nitrogen budget for the main sectors and compartments considered in the ENB. The atmospheric compartment comprises only fluxes of NH3 and NOx as input and wet- and dry deposition as output. The large fluxes of molecular nitrogen, in particular N-fixation in the industry and energy sectors and denitrification in terrestrial and aquatic ecosystems are poorly quantified and would add another 20–30 Tg N yr−1 in input and output. Nitrogen fixation occurs through biological N-fixation and through the Haber–Bosch process, but there is also input of Nr from fossil energy carriers and newly formed Nr through thermal reaction. These

fluxes are included in the number presented for N-fixation in the energy and industry sector, but are not quantified for the other sectors. Consumers and the waste sector store an unquantified amount of Nr in products, but most of the consumed Nr will accumulate in the wider environment, be land-filled or incinerated. Indeed, according to the numbers presented, more than 50% of the Nr made available to consumers appears to have a purpose other than nutrition. While some information on the fate of this Nr might be available, so far we were not able integrate robust data into the European Nitrogen Budget. At the European scale, budgets of industry and energy exceed those of agriculture. Agricultural soils can act as a source or a sink for carbon and nitrogen if organic matter is being depleted or accumulated. Our data suggest that a significant part of the nitrogen lost from the agriculture sector originates from mineralization of soil organic matter. A large exchange of nitrogen takes place in coastal areas, which act as a sink of oceanic nitrogen that is denitrified to N2 and N2O in the shelf regions of Europe. As a consequence, the flux of the N2 from these regions might be the largest single nitrogen flux in absolute terms and also the estimate for the N2O flux from the shelf regions is very high and exceeds, in absolute terms, fluxes from other sectors including agriculture. The split of atmospheric emissions in EU27 countries by sectors for three reactive gases (NOx, NH3, N2O) and the total is

365

Integrating nitrogen fluxes at the European scale

800

Switzerland

700

Netherlands

United Kingdom

500

France

400

Czech Republic

300 200

Netherlands Germany United Kingdom 400

France Czech Republic

300 200 100

100

366

Switzerland

500

Germany

600

0

600

Emissions [Gg N yr-1]

Emissions [Gg N yr-1]

Figure 16.17 The integrated nitrogen budget of the Czech Republic is being constructed by the Hydrometeorological Institute (CHMI) in cooperation with the Ministry of Agriculture and with the Central Institute for Supervising and Testing in Agriculture on the basis of data from different sources (e.g. National Statistical Office) for the years 2004–2008, but mainly for 2007.

Industry +Energy

Transport Consumer Agriculture

(Semi-) natural

Waste

Aquatic

0 Industry +Energy

Transport Consumer Agriculture (Semi-) natural

Waste

Aquatic

Figure 16.18 Absolute nitrogen emissions to the atmosphere for the main sectors/compartments.

Figure 16.19 Absolute nitrogen emissions to the hydrosphere for the main sectors/compartments.

shown in Figure 16.23. The figure shows that NOx fluxes dominate the emissions from energy-related sources, NH3 fluxes are the strongest for agricultural sources; the waste sector and aquatic systems emit mainly de-nitrification products (N2, N2O). Overall, the emissions for NOx and NH3 are roughly at the same level, with 3.5 Tg NOx-N yr−1 and 3.2 Tg NH3-N yr−1, respectively. N2O contributes 1.2 Tg N2O-N yr−1. The global warming potential of this greenhouse gas is 580 Tg CO2-eq (using a GWP of 298; IPCC, 2007). This is higher than the emissions of N2O

estimated in the European GHG inventory (EEA, 2010) for the year 2000 (412 Mt CO2-eq) which decreased to 364 Mt CO2-eq in the year 2008. However, it should be kept in mind that the ENB emissions include both anthropogenic and natural sources while the greenhouse gas inventories are restricted to anthropogenic emissions only. For example, it is likely that a significant portion of the coastal N2O fluxes of 500 Gg N2O-N yr−1 (or 230 CO2-eq yr−1) originates from nitrogen in incoming oceanic water. In addition, there are methodological differences that influence calculations. IPCC methodology includes indirect emissions

Adrian Leip NOx

70%

NH3 60%

N2O

50% 40% 30% 20% 10% 0% Switzerland

Netherlands

Germany

United Kingdom

France

Czech Republic

Figure 16.20 Split of total atmospheric emissions over the three reactive gases NOx, NH3, and N2O.

from agricultural soils only, while the estimates presented here cover virtually all available land and include implicitly all indirect N2O emissions, including those caused by deposition of NOx fluxes from the combustion processes. To account for indirect emissions from industrial and energy sources, roughly 20 Mt CO2-eq should be added. Direct N2O emissions from agricultural soils for EU27 in 2002 are estimated as 380 Tg N2O, which is about 35 Tg N2O yr−1 (or 15 Tg CO2-eq yr−1) more than reported by UNFCCC for the categories ‘direct soil emissions’ (4D1) and ‘pasture, range and paddock manure’ (4D2) (EEA, 2010). Thus, agreement for EU27 is satisfying, even though differences are larger for individual countries. Most of the data used for the ENB are based on the same models as also used for the key maps in Section 16.3. Thus the spatial variability of the most important fluxes is shown in Figures 16.1–16.11. In the following sections, the main figures presented in the European Nitrogen Budget are briefly reviewed, emphasizing those numbers and sectors which have not yet been introduced in detail elsewhere in the European Nitrogen Assessment.

Industry sector The motivation for the invention of the Haber–Bosch process to synthesize reactive nitrogen (ammonia) from atmospheric molecular nitrogen was the urgent need for nitrogen to enable sufficient agricultural food production and the provision of raw material for explosives (Erisman et€al., 2008; Sutton et€al., 2011, Chapter 1 this volume). As of 2008 around 48% is the nitrogen synthesized globally by the Haber–Bosch process (121 Tg N; Erisman et€al., 2008). About 24 Tg N is used in various industrial processes and the production of non-fertilizer products (IFA Statistics, 2010). Several ammonia-based products are used as fertilizers, in industrial processes and in chemical products. Besides some ammonia salts, other ammonia-based industrial products that are not used as fertilizers include nitric acid, adipic acid, hydrogen cyanides, diisocyanates, acrylonitrile, melamine and others (Domene and Robert, 2001). For Europe, it is estimated that about 30% of the Nr fixed with the Haber–Bosch process is used for non-agricultural purposes, including about 4.5 Tg N in

Western Europe in 2007, and 0.7 Tg N in Central Europe, totalling about 5.2 Tg N (see Winiwarter et€al., 2011, Chapter€24 this volume). An accounting of the production of nitrogen-containing substances in Europe is provided in the supplementary information (see supplementary material Chapter 16, Section C). The fate of these products is unknown. The total net trade of EU27 for total N (fertilizer and non-fertilizer) is estimated to be 1631 Gg N net import. Table 16.3 shows that the trade is dominated by the import of ammonia and urea, while derived products have a slight export-surplus.

Transport sector According to the European Environment Agency (EEA) the transport sector accounts for around one third of all final energy consumption in the EEA member countries and for more than a fifth of greenhouse gas emissions (EEA, 2009b). Transport is represented by international and domestic air, sea and inland waterway, off-road and pipeline transport, rail and road transport. It is mainly characterized by the road transport sector, which, in the year 2005 contributed more than 73% to global transport fuel consumption (EDGARv4, EIA, 2007; IEA/OECD, 2007) followed by air transport (≈ 11%), sea and inland waterways (≈€9%), rail transport (≈ 4%) and other transport (≈ 3%). In Europe, road transport has been the dominating source for NOx emissions since 1970 (Vestreng et€ al., 2009). With the implementation of strict measures and action plans in the early 1990s within the framework of the Convention on Longrange Transboundary Air Pollution, European NOx emissions were continuously reduced (Pulles et€ al., 2007). These early measures in Europe complemented clean air initiatives in the US (CONCAWE, 1997) and investigations in the automobile industry enforced by legislation. The main contributors in road transport producing high NOx emissions are heavy duty vehicles (HDV) using diesel fuel and light duty vehicles using gasoline (EMEP/EEA, 2009). With the introduction of EURO standards for light (considering passenger cars) and heavy duty vehicles emission reduction has led to a substantial NOx emissions decrease for all vehicles types in Europe. In Western Europe, NOx emissions of heavy duty vehicles have been cut by 86% compared with levels in the 1990s (ACEA, 2009). In contrast, NH3 and N2O emissions generally increased in the last years due to the worldwide turnover of vehicle fleets equipped with EURO 3/III standards. Although new emissions standards introduced significant NOx emissions reductions, the age structure of a national fleet causes a significant time lag until the new standard can show an effect. Moreover, increasing diesel consumption and increasing growth rates in freight transport volume on a national base (see Lambrecht et€al., 2009) prevent further NOx emissions decreases. Furthermore, one of the reasons why some air quality problems still persist, even though vehicles have become far cleaner, is that emissions in real driving conditions tend to be higher than emissions under test conditions. Consequently, of the EU27 Member States, only 15 (up from 10 in 2007) expect to be at, or below, their respective

367

Integrating nitrogen fluxes at the European scale Table 16.2 Summary of nitrogen input, output, stock changes and the nitrogen balance for the main compartments/sectors for the National integrated Nitrogen Budgets of the Netherlands, Germany, Switzerland, France, and the United Kingdom. The numbers are summarized from the above figures, thus input and output are the sums of arrows pointing to or from the respective compartments with the exception of nitrogen fixation of N2 (both in industry and biological N-fixation), which is regarded as new Nr input. Stock changes refer only to quantified stock changes in terrestrial ecosystems (soil stock changes or standing biomass in forests) and aquatic systems (sedimentation in lakes or marine waters)

Switzerland

Input

Atmosphere

Output

Stock change

Balance

254

0

255.5

0

−2

Industry +Energy +Transport

33

53

86

0

0

Consumer

12

0

0

0

12

Agriculture

92

57

171

0

−22

(Semi-) natural land

0

0

29

0

−29

Waste

0

0

48

0

−48

112

0

100

14

−2

0

0

0

0

0

Freshwater Marine water

Output

Stock change

Balance

Netherlands

Input

New Nr

Atmosphere

791

0

2975

0

−2184

0

2600

2876

0

−276

Consumer

419

0

82

0

337

Agriculture

Industry +Energy +Transport

885

13

785

0

113

(Semi-) natural land

62

0

4

0

58

Waste

90

0

90

0

0

515

0

378

125

12

Freshwater Marine water

378

0

Germany

Input

New Nr

Atmosphere

2263

0

4096

0

−1833

0

2052

2052

0

0

Industry +Energy +Transport

0 Output

0 Stock change

378 Balance

Consumer

1010

0

498

0

512

Agriculture

2202

233

2373

0

62

0

70

118

0

−48

Waste

514

0

489

0

25

Freshwater

687

0

688

0

−1

Marine water

492

0

0

0

492

France

Input

New Nr

Output

Atmosphere

1178

0

3401.26

(Semi-) natural land

Stock change 0

Balance −2223

Industry +Energy +Transport

0

2683

2683

0

0

Consumer

0

0

0

0

0

Agriculture

2681

569

716

0

2534

0

0

88

0

−88

(Semi-) natural land Waste

0

0

42.53

0

−43

Freshwater

0

0

797

0

−797

Marine water

797

0

0

0

797

United Kingdom

Input

New Nr

Atmosphere

1110

0

2347

0

−1238

0

1417

1417

0

0

Consumer

728

0

243

0

485

Agriculture

1178

0

1478

0

−300

Industry +Energy +Transport

368

New Nr

Output

Stock change

Balance

Adrian Leip Table 16.2 (cont.)

United Kingdom

Input

(Semi-) natural land

0

0

72

0

Waste

260

0

200.7

0

59

Freshwater

719

0

721

0

−2

Marine water

605

0

0

0

605

Czech Republic

Input

Atmosphere

219

Industry +Energy +Transport

0

Consumer

46

Agriculture

360

(Semi-) natural land

0

Waste

51

Freshwater Marine water

New Nr

New Nr

Output

Output

Stock change

Stock change

Balance −72

Balance

0

629

0

−409

408

408

0

0

0

51

0

−4

38

137

0

261

0

0

0

0

0

13

0

37

0

0

70

0

−70

0

0

0

0

0

Figure 16.21 Nitrogen budget for Europe (European Nitrogen Budget) for EU-27 compiled with data for the period around the year 2000. Basis:€(i) atmospheric transport and atmospheric deposition:€EMEP Unified model, rv3.1, 2009. Atmospheric transport is obtained from the source-receptor matrix available at http://www.emep.int/ ; (ii) atmospheric emissions from industry and energy, transport and solid waste systems:€EDGAR-CIRCE (Van Aardenne et al., 2009), (iii) industrial trade and non-fertilizer products:€Prud’homme, 2009; (iv) agricultural nitrogen fluxes incl. mineral fertilizer use and food and feed trade:€Indicator Database for European Agriculture, V1, 2009; (v) (semi-)natural systems:€INTEGRATOR, 2009. Export of forestry products:€FAOSTAT; (vi) fishery data:€Eurostat, 2009 (http://epp.eurostat.ec.europa.eu/portal/page/portal/statistics/search_database); (vii) sewage system fluxes including input of nitrogen and emissions:€Indicator Database for European Agriculture, V1, 2009; (viii) fluxes to groundwater and surface water systems including flux from surface waters to coastal zones:€IMAGE, 2009; (ix) N2O emissions from coastal zones:€Bange (2008); (x) other fluxes from coastal zones; see text.

emission ceilings by 2010 (NEC Directive status report 2008, EEA, 11/2009) and NOx emissions will be higher than expected. In addition, global emissions are likely to increase due to the strong economic growth in regions such as East Asia (Streets

and Waldhoff, 2000), Central Europe and Southeast Asia (e.g. Thailand). Emissions will also increase in the Middle East and Africa, where less policy regulations are in place (Cofala et€al., 2007).

369

Integrating nitrogen fluxes at the European scale N input

30,000

N output New N

25,000

Stock changes

Emissions [Gg N yr –1]

20,000

Table 16.3 Net trade of nitrogen of Europe (EU27). Values are net export in Gg N yrâ•›−1. Negative values indicate a net import towards the European Union

Product Urea

15,000

10,000

5,000

298

AN

215

CAN

141

NH3 UAN (estimate)

–5,000 Atmosphere Industry Consumer Agriculture Forests +Energy +Transport

Waste

Freshwater Marine water

Figure 16.22 Nitrogen balance by main sectors/compartments for EU27 around the year 2000. The contribution of newly generated nitrogen to the N-input as well as the contribution of changes in nitrogen stocks to the N-output (for example sedimentation in lakes and in estuaries) are also shown.

100%

NOx

NH3

N2O

Total

26 −1536 70 −1631

globally by 9%. The corresponding values for navigation are 13.3% (Annex 1 countries) and 16.4% (worldwide). Also the emissions from railway transport will increase with increasing freight transport, which has been restructured in the past 20 years.

90%

Agriculture sector

80%

Agriculture is the sector with the largest source of reactive nitrogen emissions in Europe as a whole and for each of its countries. We find a high recycling of Nr between crop production and manure excretion; the livestock sector receives about the same amount of Nr in the form of domestically produced and imported feed than the grass- and crop-sector. About 60% of agricultural products (consumed or used in industry) originate from crop production. The nitrogen use efficiency (defined as Nr in useful products relative to Nr inputs) for cultivation on soils is about 60% (considering also N-input through atmospheric deposition and biological nitrogen fixation), while the nitrogen use efficiency for a farm N-budget including animal products drops to about 30% (Leip et€al., 2010a). Productivity in Europe is high, as has been shown; however, to supply the protein requirements of European citizens, about 400 Gg N yr−1 of agricultural products for food and about 3 Tg N yr−1 for feed or industrial use have to be imported. For comparison, Galloway et€al. (2008; UNEP and WHRC, 2007) estimate is a significant global trade in fertilizer (31 Tg N), grain (12 Tg N) and meat (0.8 Tg N), and a net import of vegetal products (2367 Gg N) and meat (110 Gg N) to Europe. The authors include in their data Eastern Europe, a region which is a large producer and exporter of mineral fertilizer. Therefore, Europe is shown to be a net exporter of fertilizer (5376 Gg N), neglecting a large internal trade in the European Union (see Table 16.5).

70% 60% 50% 40% 30% 20% 10% 0% Industry +Energy

Transport Consumer Agriculture Forests

Waste

Aquatic

Total

Figure 16.23 Split of atmospheric emissions in EU27 around the year 2000 by sector into the three reactive gases NOx, NH3, and N2O.

A particular concern is aviation, which is the fastest-�growing transport sector. This growth is partly driven by increasing wealth and low prices (for aviation, fuel tax is currently not considered), which underpin strong growth in tourism travel. Aviation now accounts for more than 10% of greenhouse gas emissions. Emission standards for ships and aviation are dealt with by the respective UN organizations (International Maritime Organization, IMO, and International Civil Aviation Organization, ICAO) and by international conventions including the Convention on Long-range Transboundary Air Pollution which also addresses other sectors in addition to transport. Present measures regulate emissions (NOx) on the Landing and Take Off cycle and were designed to address airport air quality problems. The Committee on Aviation Environmental Protection (CAEP) is pursuing new certification methodologies that also take account of the flight mode as well. Table 16.4 shows that Nr emissions from civil aviation (Annex 1 countries) increased by 3.5% from 2000 to 2005, and

370

−845

AS

NPK (estimate)

0

Net export

Forestry Forests are currently undergoing net growth with net immobilization of Nr in the soil of about 810 Gg N yr−1 and a Nr uptake into the above-ground biomass of 320 Gg N yr−1 (estimated with the INTEGRATOR model). National

Adrian Leip Table 16.4 Reactive N emission in the years 2000 and 2005 for civil aviation, global aviation, marine activities and navigation and the development in percent from 2000 to 2005

Sector

Unit

Civil aviation

Gg N yr

−1

2000

2005

98

101

Global aviation

Gg N yr

−1

820

894

9.1

Marine and navigation

Gg N yr−1

1966

2229

13.3

UNFCCC

Global shipping

Gg N yr−1

3398

3954

16.4

EDGAR-CIRCE

Table 16.5 Trade of nitrogen in fertilizer, vegetal products and meat with (West, Central and East) Europe (Gg N yr−1)

Export from Europe Fertilizer

Grain

Meat

North America

2352

South America

2268

Africa

359

SE-Asia

977

Australia

59

Import to Europe

639

North America

423

South America

2318

Africa

193

Russia

81

SE-Asia

100

South America

82

Russia

28

Source: From Galloway et al., 2008.

estimates of carbon sequestration in the land use/land use change and forestry (LULUCF) sector are about 25 Tg CO2-eq (EEA, 2009a). The main uncertainty in converting this value into sequestered Nr (or stock changes in forests) is the ratio of sequestered carbon in above-ground material and soils, which differ considerably on the C/N ratio. As a rough assumption we use 50% of carbon sequestered in soils, giving stock changes of about 300 Gg N for EU27. From FAO statistics and national data we obtain a figure of 380 million m3 of total roundwood production in EU27 for the year 2000. Converted to nitrogen uptake/removal from forests this gives approximately 190 Gg N, using a basic wood density of 0.4 for average temperate and boreal trees (see IPCC, 2006), a carbon/dry-biomass ratio of 0.5 and a C/N ratio around 400 (see Katri et€al., 2004). Part of this wood is used for domestic burning and the resulting emissions are included in the energy/industry figures calculated in the EDGAR database. The remaining biomass will be used in paper and wood products. However, these numbers might be an under-estimation of the real removal of wood, as a comparison between satellite imagery and forest statistics in Italy has shown (Corona et€al., 2007).

Change (%) 3.5

Reference UNFCCC EDGAR-CIRCE

Waste sector There is little quantitative information on sewage sludge applied to agricultural fields. The value indicated in the figure above has been obtained from the national GHG inventories of European counties to the UNFCCC. In EU15, only seven countries report that domestic or industrial sewage sludge is applied to agricultural soils. The total of 45 Gg N per year is only a small fraction of the 0.7 Tg N of sewage sludge that is assumed to enter the solid waste sector. We do not distinguish here landfilling and waste burning. The input to the waste sector is mainly determined by household wastes; the estimate for the input to the solid waste systems includes currently only agricultural wastes. The IMAGE model uses a nitrogen factor of 3–4 kg N capita−1 yr−1, a value which is also confirmed from data which gave a value of 3.94 kg N per capita for over 60 catchments in the UK (Johnes, 2007). The IMAGE data are based on measured nitrogen influent to wastewater treatment plants, divided by the number of connected people. Therefore, the estimate based on human diet may be higher. For example, Billen et€al. (2008) estimate an annual food intake of 8.2 kg N capita−1 yr−1 from national French domestic consumption data; the estimate based on household consumption figures may be higher. These include products bought but not ingested, the fate of which is solid wastes instead of wastewater. The CAPRI data used here suggest higher values with about 5 Tg N offered to the consumers. This contains 6% nitrogen in non-edible products and an assumed 30% of food wastes. Thus the actual annual nitrogen consumption in Europe including consumption by pets is estimated to be 6.3 kg N capita−1 yr−1 in the input of nitrogen from consumer to the waste water treatment systems is estimated to be 3.1 Tq N yr–1.

Aquatic systems We define coastal areas as the shelf regions with a water depth of less than 200 m (Uher, 2006). This includes most of the shallow Baltic and most of the North Sea, as well as the Adriatic Sea, but excludes most of the Mediterranean Sea such as the Balearic, Ligurian and Tyrrhenian Seas. Also only narrow strips of the Atlantic coast in Spain and Portugal are included. Budgeting nitrogen fluxes in aquatic systems is one of the most difficult parts of the European Nitrogen Budget. • We assume that Nr leached from soils enters the groundwater, however, sub-surface flow does also occur but has not yet been estimated for Europe.

371

Integrating nitrogen fluxes at the European scale

• Atmospheric Nr deposition has been estimated on an ‘area-fraction’ basis. However, Nr deposition in sealed and non-sealed urban soils will enter the aquatic system either directly or via sewage treatment systems. This has not been accounted for. • The models predict total Nr leaching fluxes; their differentiation between nitrate and organic nitrogen from diffuse sources is not possible at the moment. Voss et€ al. (2011, Chapter 8, this volume) have made an attempt to quantify the global nitrogen balance in shelf regions. Most of the flux terms are associated with considerable uncertainty. However, burial in sediments and biological nitrogen fixation are both relatively small flux terms; in Europe, biological nitrogen fixation occurs mainly in the Baltic Sea and thus the value shown in the figure above is the estimate reported for the Baltic Sea (Rahm et€al., 2000; Schneider et€al., 2003). However, the contribution of benthic nitrogen fixation is not considered in this value, as estimates are lacking. Burial in sediments is likely to be a small loss term, which we are not able to quantify for European shelf regions.

16.5╇ Conclusion Environmental problems related to nitrogen concern all economic sectors and impact all media:€atmosphere, pedosphere, hydrosphere and anthroposphere. Therefore, the integration of fluxes presented in depth in earlier chapters for individual sectors/media is needed to get a picture of the overall problem. This chapter presents a set of high resolution maps showing key elements of the N flux budget across Europe. Additionally, comparative nitrogen budgets are presented for a range of European countries. A European Nitrogen Budget is presented on the basis of state-of-the-art Europe-wide models and databases focusing on different parts of Europe’s society. Key maps of nitrogen fluxes have been plotted from five models and databases covering together all sectors and media in Europe. These models combine a large spatial extent with a high spatial resolution of the data and a focus on nitrogen fluxes. The maps show high pressure on the environment in regions used intensively for agriculture such as the Netherlands, the Po Valley, Brittany, but also in the banlieue of large metropolitan areas such as Paris, Berlin and London. These areas have high N-input and agricultural surplus as well as NH3 and N2O fluxes. NOx emissions, on the other hand, are dominated by industrial and combustion sources and their distribution reflects the degree of industrialization and population density. The map shows hotspots in centres of energyintensive industry, such as Sachsen-Anhalt in Germany, North Italy, the Netherlands, or along intensive traffic lines. The spatial distribution of N2O fluxes shows elements of both patterns, but it is further complicated by the strong dependence of N2O emissions from soil properties and meteorological conditions. Land productivity in Europe is high and could be sufficient to sustain the protein requirement of European population. However, a large part of these resources is invested to feed the livestock, which consume three times the nitrogen that

372

humans consume but deliver only about 50% of the proteins in human’s diet in EU-27. As a consequence, large amount of feedstuff must be imported to Europe. National nitrogen budgets are difficult to compile using a wide range of data sources and are currently available only for a limited number of countries. The summary of the national N-budgets shows that the balance is not closed for several sectors. This is partly because not all nitrogen fluxes have yet been (or could be) estimated, and partly because the data for different compartments and sectors have been taken from best available, but partly inconsistent, data sets. Overall, national N-budgets have already shown themselves to be useful tools to identify the most important N-fluxes in a country, to provide an efficient visualization of the complexity of a problem and to elaborate efficient mitigation strategies. Furthermore, through the integration of data from different and independent sources, data gaps and sometimes contradicting scientific understanding of processes have been highlighted. Modelling approaches have been used to fill in the data gaps in some of these budgets, but it became obvious during this study that further research is needed in order to collect necessary data and make national nitrogen budgets inter-comparable across Europe. The European Nitrogen Budget is largely modelbased and provided a challenge in combining five Europe-wide models and databases. Results suggest that European agriculture (EU-27) receives c. 18 Tg N yr−1 reactive nitrogen, out of which only c. 7 Tg N yr−1 find their way to the consumer or are further processed in industry. Some 3.7 Tg N yr−1 of reactive nitrogen are released by the burning of fossil fuel, out of about 16 Tg N yr−1 of N2 which is fixed into Nr each year in industry and energy generation. The contribution in emissions of reactive nitrogen of the industry and energy sectors is comparable to that of the transport sector. More than 8 Tg N yr−1 of reactive nitrogen are disposed of to the hydrosphere; Europe is a net exporter of Nr through atmospheric transport of c. 2.3 Tg N yr−1. The largest single sink for Nr appears to be denitrification to N2 in European shelf regions. However, this sink is also the most uncertain one as it concerns also Nr that is imported through the exchange with the open ocean which is potentially as large as the input of Nr in mineral fertilizer. In contrast to most chapters in this assessment, the current chapter presents considerable new information that has been just lately compiled, estimated or calculated with recently improved models. The European Nitrogen Assessment aims at providing a comprehensive assessment of nitrogen, its related problems and possible solutions, but also at raising awareness on nitrogen issues and promoting the idea of integrated assessment as an important prerequisite for successful solutions. The large variety of problems associated with the excess of reactive nitrogen in the European environment requires such an integrated nitrogen management approach that would allow for creation and closure of N budgets within European environments. The first steps to reach this goal have already been taken in the process of assembling the assessment, jointly with the

Adrian Leip

contemporarily formed Task Force on Reactive Nitrogen under the UN-CLRTAP and the NitroEurope-Integrated Project. The progress of this joint effort is reflected in this chapter.

Acknowledgements The authors gratefully acknowledge support from the European Commission for the NitroEurope Integrated Project, and from the European Science Foundation for the Nitrogen in Europe (NinE) programme and COST 729.

Supplementary materials Supplementary materials (as referenced in the chapter) are �available online through both Cambridge University Press:€ www.cambridge.org/ena and the Nitrogen in Europe website:€www.nine-esf.org/ena .

References Aber, J. D. (1992). Nitrogen cycling and nitrogen saturation in temperate forest ecosystems. Trends Ecol., Evol., 7, 220–224. ACEA (2009). Historical series 1990€– 2008, average car age in Europe, publications 2009, European Automobile Manufacturers Association. Available at http://www.acea.be/index.php/collection/ statistics. Amann, M., Asman, W., Bertok, I. et al. (2007). Cost-effective emission reductions to meet the Environmental Targets of the Thematic Strategy on Air Pollution under different greenhouse gas constraints NEC, Scenario Analysis Report Nr. 5. IIASA, June 2007. Andersson-Sköld, Y. and Simpson, D. (1999). Comparison of the chemical schemes of the EMEP MSC-W and the IVL photochemical trajectory models. Atmos. Env., 33, 1111–1129. BAFU 2010: Sticksloffflüsse in der Schweiz, Stoffflussanalyse für das Jahr 2005, Heldstab J., Reusimann J., Bledermann R., Len D. Im Auftrag des Bundesamis für ummett BAFU, limwett Hissen Nr. 1018. Bern. Bange, H. W. (2008). Gaseous nitrogen compounds (NO, N2O, N2, NH3) in the ocean. In:€Nitrogen in the Marine Environment, 2nd edition. Capone, D. G., Bronk, D. A., Mulholland, M. R. and Carpenter, E. J. (eds.),€Elsevier, Amsterdam, pp. 51–94. Billen, G., Garnier, J., Mouchel, J.-M. and Silvestre, M. (2007). The Seine System:€introduction to a multidisciplinary approach of the functioning of a regional river system. Sci. Total Environ., 375, 1–12. Billen, G., Barles, S., Garnier, J., Rouillard, J. and Benoit, P. (2008). The food-print of Paris:€long-term reconstruction of the nitrogen flows imported into the city from its rural hinterland. Region. Environ.Change, http://dx.doi.org/10.1007/ s10113–008–0051-y Billen, G., Silvestre, M., Grizzetti, B. et€al. (2011). Nitrogen flows from European watersheds to coastal marine waters. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W.€Erisman et€al., Cambridge University Press. Bouwman, A. F., Kram, T. and Klein Goldewijk, K. (2006). Integrated modelling of global environmental change: an overview of IMAGE 2.4, Netherlands Environmental Assessment Agency (MNP):€Bilthoven, The Netherlands.

Britz, W. and Leip, A. (2009). Development of marginal emission factors for N losses from agricultural soils with the DNDC-CAPRI meta-model. Agric. Ecosyst. Environ., 133, 267–279. Britz, W. and Witzke, H.-P. (eds.) (2008). CAPRI Model Documentation 2008:€Version 2 p. Institute for Food and Resource Economics, University of Bonn: Bonn (available at http://www. capri-model.org/docs/capri_documentation.pdf) Butterbach-Bahl, K., Kahl, M., Mykhayliv, L. et€al. (2008). A European wide inventory of soil NO emissions using the biogeochemical models DNDC/ Forest DNDC . Atmos. Environ., 43, 1392–1402. Butterbach-Bahl, K., Gundersen, P., Ambus, P. et€al. (2011). Nitrogen processes in terrestrial ecosystems. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman, et€al., Cambridge University Press. BUWAL (1996). Strategie zur Reduktion der Stickstoffemissionen. Bericht der Projektgruppe Stickstoffhaushalt Schweiz, Schriftenreihe Umwelt Nr. 273, Bundesamt für Umwelt, Wald und Landschaft (BUWAL) and Bundesamt für Landwirtschaft (BLW):€Bern, Switzerland. CDIAC (2008). Fossil-fuel CO2 emissions. http://cdiac.ornl.gov/ trends/emis/meth_reg.html. Last accessed on 11.08.2009. Cofala, J., Amann, M., Klimont, Z., Kupiainen, K. and HöglundIsaksson, L. (2007). Scenarios of global anthropogenic emissions of air pollutants and methane until 2030. Atmos. Environ., 41, 8486–8499. CONCAWE (1997). Motor vehicle emission regulations and fuel specification€– part 2, detailed information and historic reviews (1970€– 1996), report 6/1997. CONCAWE (2001). Motor Vehicle emission regulations and fuel specification€– part 2, detailed information and historic reviews (1996€– 2000), report 2/2001. Corona, P., Giuliarelli, D., Lamonaca, A. et€al. (2007). Confronto sperimentale tra superfici a ceduo tagliate a raso osservate mediante immagini satellitari ad alta risoluzione e tagliate riscontrate amministrativamente. Forest, 4 , 324–332. Dalsøren, S. B., Eide, M. S., Endresen, Ø. et€al. (2009). Update on emissions and environmental impacts from the international fleet of ships. The contribution from major ship types and ports. Atmos. Chem. Phys., 9, 2171–2194. De Vries W., Kros, J., Reinds, G. J. et€al. (2009). INTEGRATOR:€a modelling tool for European-wide assessments of nitrogen and greenhouse gas fluxes in response to changes in land cover, land management and climate – calculation procedures, application methodology and examples of scenario results, Alterra Report, Alterra Wageningen UR:€Wageningen, The Netherlands. de Vries, W., Leip, A., Reinds, G. J. et€al. (2011). Geographic variation in terrestrial nitrogen budgets across Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W.€Erisman et€al., Cambridge University Press. DEFRA (2008). Summary of Food and Farming in the UK, HMSO:€London. Domene, L. A. F. and Robert, U. A. (2001). Nitrogen’s role in industrial systems. J. Industr. Ecol., 5, 77–103. ECN (2008). European Compost Network: Webpages EU Biowaste Treatment and Country presentations. http://www. compostnetwork.info/. Last accessed on 11.08.2009. EEA (1998). Annual topic update 1997. Topic report 02/98, European Environment Agency and European Topic Centre on Coastal and Marine Environment: Copenhagen.

373

Integrating nitrogen fluxes at the European scale EEA (2009a). NEC Directive status report 2008, Reporting by the Member States under Directive 2001/81/EC of the European Parliament and of the Council of 23 October 2001 on national emission ceilings for certain atmospheric pollutants EEA Technical report series:€European Environment Agency: Copenhagen. EEA (2009b). Annual European Community greenhouse gas inventory 1990–2007 and inventory report 2009, submission to the UNFCCC Secretariat, European Environment Agency: Copenhagen. EEA (2010). Annual European Community greenhouse gas inventory 1990–2008 and inventory report 2010, submission to the UNFCCC secretariat, European Environment Agency: Copenhagen. EIA (2007). International energy statistics. U.S. Energy Information Administration: Washington DC, USA. http://www.eia.doe.gov/ emeu/international/contents.html. Last accessed on 2007. EMEP (2008). EMEP measurement database. The Co-operative Programme for the Monitoring and Evaluation of the Long-Range Transmission of Air Pollutants in Europe. Available at http://www. emep.int/. EMEP/EEA (2009). EMEP/EEA air pollutant emission inventory guidebook: 2009, EEA Technical Report. Available at http:// www.eea.europa.eu/publications/emep-eea-emission-inventoryguidebook-2009. Erisman, J. W., Domburg, P., de Haan, B. J. et€al. (2005). The Dutch nitrogen cascade in the European perspective, ECN-C€– 05–007, Energy Research Centre of the Netherlands (ECN):€Petten, The Netherlands. Erisman, J. W., Sutton, M. A., Galloway, J., Klimont, Z. and Winiwarter, W. (2008). How a century of ammonia synthesis changed the world. Nature Geosci., 1, 636–639. Eyers, C. J., Addleton, D., Atkinson, K. et€al. (2004). AERO2k Global Aviation Emissions Inventories for 2002 and 2025, QINETIQ/04/01113, December 2004. Available at http://www. cate.mmu.ac.uk/aero2kreports/AERO_2k_Global_Aviation_ Emissions_Inventories_for_2002_and_2025_V1_1.pdf. Galloway, J. N. and Cowling E. B. (2002). Reactive nitrogen and the world:€200 years of change. Ambio, 31, 64–71. Galloway J. N., Cowling E. B. and Kessler E. (2002). Reactive nitrogen. Ambio 31,€59. Galloway, J. N., Aber, J. D., Erisman, J. W. et€al. (2003). The Nitrogen Cascade. BioScience, 53, 341–356. Galloway, J. N., Townsend, A. R., Erisman, J. W. et€al. (2008). Transformation of the nitrogen cycle:€recent trends, questions, and potential solutions. Science, 320, 889–892. Genovese, G., Baruth, B., Royer, A. and Burger, A. (2007). Crop and yield monitoring activities: MARS STAT action of the European Commission. Geoinformatics, 10, 20–22. Heckelei, T., Mittelhammer, R. C. and Britz, W. (2005). A Bayesian alternative to generalized cross entropy solutions to underdetermined models. In: 89th EAAE Symposium Modelling Agricultural Policies:€State of the Art and New Challanges, 3–5 February, Parma, Italy. Hertel, O., Reis, S., Skjøth, C. A. et€al. (2011). Nitrogen turnover processes in the atmosphere. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erismon et€al., Cambridge University Press. IEA/OECD (2007). Energy balances of OECD and non-OECD countries on-line data service, 20037 http://data.iea.org. IFA (2010). Global Fertilizers and Raw Materials Supply and Supply/ Demand Balances, 2010–2014. Prud’homme, M., June 2010.

374

IPCC (2006). 2006 IPCC guidelines for national greenhouse gas inventories, prepared by the National Greenhouse Gas Inventories Programme, IGES: Tokyo, Japan. IPCC (2007). Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Working Group I€– The Physical Science Basis. Summary for policymakers. Available at http://www.ipcc.ch/ WG1_SPM_17Apr07.pdf. IRF (2007). World Road Statistics 2007. Jeppsson, U., Alex, J., Pons, M. N., Spanjers, H. and Vanrolleghem,€P.€A. (2002). Status and future trends of ICA in waste water treatment- a European perspective. Water Sci. Technol., 45, 485–494. Johnes, P. D. (1996). Evaluation and management of the impact of land use change on the nitrogen and phosphorus load delivered to surface waters:€the export coefficient modelling approach. J. Hydrol., 183, 323–349. Johnes, P. J. (2007). Meeting ecological restoration targets in European waters:€a challenge for animal agriculture. In:€Redesigning Animal Agriculture, ed. D. Swain, E. Charmley, J. Steel and S. Coffey, CAB International: Wallingford, UK, pp. 185–203. Johnes, P. J. and Butterfield, D. (2002). Landscape, regional and global estimates of N flux from land to ocean:€errors and uncertainties. Biogeochemistry, 57/58, 429–476. JRC/PBL (2009). The EDGAR database EDGAR v4.0, available at http:\\edgar.jrc.ec.europa.eu\ Katri, K., Seija, K., Pekka, S. et€al. (2004). Effect of elevated [CO2] on stem wood properties of mature Norway spruce grown at different soil nutrient availability. Glob. Change Biol., 10, 1526–1538. Kesik, M., Ambus, P., Baritz, R. et€al. (2005). Inventory of N2O and NO emissions from European forest soils. Biogeosciences, 2, 353–357. Klimont, Z. and Brink, C. (2004). Modelling of emissions of air pollutants and greenhouse gases from agricultural sources in Europe, IIASA Interim Report IR-04–048, International Institute for Applied Systems Analysis:€Laxenburg, Austria. Lambrecht, M., Erdmenger, C., Bölke, M. et€al. (2009). Strategie für einen nachhaltigen Güterverkehr (Summary:€Strategy for Sustainable Freight Transport in English), UBA Texte 18/2009. Leip, A., Marchi, G., Koeble, R. et€al. (2008). Linking an economic model for European agriculture with a mechanistic model to estimate nitrogen and carbon losses from arable soils in Europe. Biogeosciences, 5, 73–94. Leip, A., Weiss, F. and Britz, W. (2009a). Agri-environmental nitrogen indicators for EU27. In:€Proceedings of the Conference on Integrated Assessment of Agriculture and Sustainable Development:€Setting the Agenda for Science and Policy (AgSAP 2009). Egmond aan Zee, The Netherlands, 10–12 March 2009, ed. M. K. Van Ittersum, J. Wolf and H. H. Van Laar, Wageningen University and Research Centre:€Wageningen, the Netherlands, pp.€184–185. Leip, A., Weiss, F., Orlandini, L. and Marchi, G. (2009b). Nitrous oxide emissions from agriculture in Europe as calculated within the Indicator Database for European Agriculture. In: Fifth International Symposium on Non-CO2 Greenhouse Gases (NCGG-5) Science, Reduction Policy and Implementation, Wageningen, the Netherlands, 30 June€– 3 July, 2009. Leip A., Britz W., de Vries W. and Weiss F. (2011). Farm, land, and soil nitrogen budgets for agriculture in Europe. Environ. Pollu. (in press). Leip, A., Busto, M. and Winiwarter, W. (2010b). Developing stratified N2O emission factors for Europe. Environ. Pollut. (in press).

Adrian Leip Li, C. (2000). Modeling trace gas emissions from agricultural ecosystems. Nutr. Cycl. Agroecosyst., 58, 259–276. Metzger, S. M., Dentener, F. J., Lelieveld, J. and Pandis, S. N. (2002). Gas/aerosol partitioning 1. A computionally efficient model. J. Geophys. Res., 107 (ACH16). NitroEurope (2010) http://www.nitroeurope.eu/ (accessed 1 June 2010). OECD/Eurostat (2003). OECD/Eurostat gross nitrogen balances handbook. Organisation for Economic Co-operation and Development, Paris. Olivier, J. G. J., van Aardenne, J., Dentener, F. et€al. (2005). Recent trends in global greenhouse gas emissions:€regional trends 1970–2000 and spatial distribution of key sources in 2000. Environ. Sci., 2, 81–99. Pilegaard, K., Skiba, U., Ambus, P. et€al. (2006). Factors controlling regional differences in forest soil emission of nitrogen oxides (NO and N2O). Biogeosciences, 3, 651–661. Prud ’ homme, M. (2009). IFA Production and International Trade Committee, personal communication. Pulles, T., van het Bolscher, M., Brandt, R. and Visschedijk,€A. (2007). Assessment of global emissions from fuel combustions in the final decades of the 20th century DG Research, Retro Project, No. 034.64145, TNO report, 2007-A-R0132/B. p. 41. Rahm, L. et€al. (2000). Nitrogen fixation in the Baltic proper:€an empirical study. J. Mar. Sys., 25, 239–248. Schneider, B. et€al. (2003). The surface water CO2 budget for the Baltic Proper:€a new way to determine nitrogen fixation. J. Mar. Syst., 42, 53–64. Seitzinger, S., Harrison, J. A., Böhlke, J. K. et€al. (2006). Denitrification across landscapes and waterscapes:€a synthesis. Ecol. Appl., 16, 2064–2090. Simpson, D., Winiwarter, W., Börjesson, G. et€al. (1999). Inventorying emissions from nature in Europe. J. Geophys. Res., D104, 8113–8152. Simpson, D., Fagerli, H., Jonson, J. et€al. (2003). The EMEP unified Eulerian model. model description, EMEP Report 1/2003, The Norwegian Meteorological Institute,€Oslo, Norway. Simpson, D., Butterbach-Bahl, K., Fagerli H. et€al. (2006). Deposition and emissions of reactive nitrogen over European forests:€a modelling study. Atmos. Environ., 40, 5712–5726. Simpson, D., Aas, W., Bartnicki, J. et€al. (2011). Atmospheric transport and deposition of nitrogen in Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. SRIC (2005). Adipic acid, nitric acid and caprolactam production data 1974–2004, Tables from the Directory of Chemical Producers, dated 2 December 2005. SRI Consulting,€Gaithersburg, MD. Stoddard, J. L. (1994). Long-term changes in watershed retention of nitrogen:€its causes and aquatic consequences. In:€Environmental Chemistry of Lakes and Reservoirs, ed. L. A. Baker. Washington (DC):€American Chemical Society, pp.€223–284. Streets, D. G. and Waldhoff, S. T. (2000). Present and future emissions of air pollutants in China:€SO2, NOx, and CO. Atmos. Environ., 34, 363–374. Sutton, M. A., Nemitz, E., Erisman, J. W. et€al. (2007). Challenges in quantifying biosphere–atmosphere exchange of nitrogen species. Environ. Pollu., 150, 125–139.

Sutton, M. A., Howard, C. M., Erisman J. W. et€al. (2011). Assessing our nitrogen inheritance. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Svirejeva-Hopkins, A., Reis, S., Nardoto G. et€al. (2011). Nitrogen flows and fate in urban landscapes. In:€The European Nitrogen Assessment, ed. M. A., Sutton, C. M. Howard, J. W. Erisman et€al. Cambridge University Press. Uher, G. (2006). Distribution and air-sea exchange of reduced sulphur gases in European coastal waters. Estuar. Coast. Shelf Sci., 70, 338–360. Umweltbundesamt (2009a). Integrierte Strategie zur Minderung von Stickstoffemissionen, Umweltbundesamt. Available at http://www.umweltbundesamt.de/luft/downloads/emissionen/ stickstoffemissionsminderungsstrategie.pdf, last accessed on 26.11.2008. Umweltbundesamt (2009b). Hintergrundpapier zu einer multimedialen Stickstoffemissionsminderungsstrategie, Umweltbundesamt. Available at http://www. umweltbundesamt.de/luft/downloads/emissionen/ hg-stickstoffemissionsminderungsstrategie.pdf, last accessed on 26.11.2008. UNEP and WHRC (2007). Reactive nitrogen in the environment: too much or too little of a good thing: United Nations Environment Programme: Paris. Van Aardenne, J. A., Dentener, F. J., Olivier, J. G. J., Klein Goldewijk,€C.€G. M. and Lelieveld, J. (2001). A high resolution dataset of historical anthropogenic trace gas emissions for the period 1890–1990. Glob. Biogeochem. Cycles, 15, 909–928. Van Aardenne, J., Doering, U., Monni, S. et€al. (2009). Emission Inventory for period 1990–2005 on 0.1x0.1 grid. Report to the Sixth Framework Programme Project No. 036961-CIRCE, 23 January 2009. Van der Werft, G. et€al. (2010). http://www.falw.vu/~gwerf/GFED/ index.html (site accessed 1 June 2010). Van Drecht, G., Bouwman, A. F., Harrison, J. and Knoop, J. M. (2009). Global nitrogen and phosphate in urban wastewater, between 1970 and 2050. Glob. Biogeochem. Cycles, 23 (GB0A03). van Egmond, K., Bresser, T. and Bouwman, L. (2002). The European Nitrogen Case. Ambio, 31, 72–78. van Grinsven, J. J. M., van Schijndel, M. W., Schotten, C. G. J. and van Zeijts, H. (2003). Integrale analyse van stikstofstromen en stikstof beleid in Nederland, RIVM report 500003001, RIVM,€Bilthoven, The Netherlands. Velthof, G., Oudendaag, D., Witzke, H.-P. et€al. (2009). Integrated assessment of nitrogen emissions from agriculture in EU-27 using MITERRA-EUROPE. J. Environ. Qual., 38, 402–417. Vestreng, V., Ntziachristos, L., Semb, A. et€al. (2009). Evolution of NOx emissions in Europe with focus on road transport control measures. Atmos. Chem. Phys., 9, 1503–1520. Voss, M., Baker, A. Bange, H. W. (2011). Nitrogen processes in coastal and marine systems. In:€The European Nitrogen Assessment, ed. M.€A.€Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Wang, C., Corbett, J. J. and Firestone, J. (2007). Improving spatial representation of global ship emissions inventories. Environ. Sci. Technol., 42, 193–199. Weigelt, A., Weisser, W. W., Buchmann, N. and Scherer-Lorenzen,€M. (2009). Biodiversity for multifunctional grasslands:€equal

375

Integrating nitrogen fluxes at the European scale productivity in high-diversity low-input and low-diversity highinput systems. Biogeosciences, 6, 1695–1706. Weiss, F., Leip, A., 2010. Methodology for quantification of greenhouse gas and ammonia emissions from the livestock sector in EU27. In: Leip, A., Weiss, F., Wassenaar, T., Pexz, I., Fullmann,€T., Londjani, P., Tubiesso, F., Grandgirand, D., Monni,€S., Bicüa,€K. (eds.) Evaluation of the livestock sector’s contribution to the EU greenhouse gas emissions (GGELS) – final report. European Commission, Joint Research Centre.

376

Wieland, U. (2003). Water use and waste water treatment in the EU and in candidate countries. Statistics in Focus, Theme 8–13/2003, Eurostat. Winiwarter, W., Hettelingh, J. P., Bouwman, L. et€al. (2011). Future scenarios of nitrogen in Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press.

Part

IV

Managing nitrogen in relation to key societal threats

Chapter

17

Nitrogen as a threat to European water quality Lead author: Bruna Grizzetti Contributing authors: Fayçal Bouraoui, Gilles Billen, Hans van Grinsven, Ana Cristina Cardoso, Vincent Thieu, Josette Garnier, Chris Curtis, Robert Howarth and Penny Johnes

Executive summary Nature of the problem • Anthropogenic increase of nitrogen in water poses direct threats to human and aquatic ecosystems. High nitrate concentrations in drinking water are dangerous for human health. In aquatic ecosystems the nitrogen enrichment produces eutrophication, which is responsible for toxic algal blooms, water anoxia, fish kills and habitat and biodiversity loss. • The continuous nitrogen export to waters reduces the capacity of aquatic ecosystems to absorb, reorganise and adapt to external stress, increasing their vulnerability to future unexpected natural or climate events.

Key findings/state of knowledge • Nitrogen concentrations in European rivers, lakes, aquifers and coastal waters are high in many regions. In addition nitrate concentrations are increasing in groundwaters, threatening the long term quality of the resource. • In Europe, nitrogen pressures occur over large areas, implying elevated costs for meeting the long-term good chemical and ecological water quality requirements. A significant part of the European population could be potentially exposed to high nitrate values in drinking water if adequate treatments were not in place. Furthermore many of European aquatic ecosystems are eutrophic or at risk of eutrophication. • Nitrogen pressures have reduced biodiversity and damaged the resilience of aquatic ecosystems and continue to pose a threat to the aquatic environment and to the provision of goods and services from the aquatic ecosystems. • Even under favourable land use scenarios the nitrogen export to European waters and seas is likely to remain significant in the near future. The effects of climate change on nitrogen export to water are still uncertain.

Major uncertainties/challenges • Policy tools are available within the European Union and under international conventions to mitigate the nitrogen pollution in water, but their full implementation has not been achieved yet throughout Europe. • In many cases a delay in the water quality response to the implementation of measures have been observed, due to previous nitrogen accumulation in soils, sediments or aquifers or to inadequate design of the mitigation plans. • The issue of pollution swapping between environmental compartments has appeared as an important element to be considered by both the scientific and policy prospective.

Recommendations • To protect and enhance the European water resources the full implementation of the existing regulations related to nitrogen is necessary, in addition to an efficient environmental monitoring. • Moreover, positive synergies could be obtained by encouraging the integration in the sectoral policies and enhancing interdisciplinarity in the scientific research, especially in support of regional assessments and pollution swapping evaluations.

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

379

Nitrogen as a threat to European water quality

17.1╇ Introduction Human activities are responsible for consistent Nr export to the environment (Vitousek et€ al., 1997; Schlesinger, 2009; Erisman et€al., 2011, Chapter 2 this volume). The enrichment of nitrogen in the aquatic system impairs the water quality of rivers, lakes, aquifers and coastal and marine waters, and contributes to the phenomenon of eutrophication (European Environment Agency, 2001; Durand et€al., 2011, Chapter 7 this volume; Voß et€al., 2011, Chapter 8 this volume; Billen et€al., 2011, Chapter€ 13 this volume). Nr is fundamental for global food production and is still insufficient in many world regions (Sanchez and Swaminathan, 2005). However, the significant anthropogenic nitrogen mobilisation through agricultural activities, waste water discharges and fossil fuels combustion produces detrimental impacts on the aquatic environment and affects both human and ecosystem health (Lavelle et€al., 2005). Estimates for year 2000 indicate that Europe is exporting 4.7 Tg of nitrogen per year to its seas (Bouraoui et€al., 2009; Billen et€ al., 2011, Chapter 13 this volume) and trends show that the production of Nr and its emission to the environment is accelerating because of the rise of agricultural demands and commercial energy production (Galloway et€al., 2008). The aquatic ecosystems are able to remove a significant part of incoming nitrogen load but this capacity is not unlimited and strongly depends on the local ecosystem characteristics (Howarth et€al., 1996; Kronvang et€al., 1999a; Alexander et€ al., 2000; Mulholland et€ al., 2008; Durand et€al., 2011, Chapter 7 this volume). Consequently there is a lot of uncertainty on the amount of self-purification of aquatic ecosystems (Seitzinger et€al., 2006; Hejzlar et€al., 2009) and on the capacity to absorb nitrogen pollution without undergoing radical changes (Millennium Ecosystem Assessment, 2005). In addition, nitrogen can build-up slowly in soil and water systems such as aquifers and reservoirs, and actual remediation practices might produce their effects only in the long-term (Jackson et€al., 2008). Europe is thus pouring nitrogen in its water resources affecting human and ecological systems at a rate that is unlikely to reverse in the near future and with consequences that are only partially understood. The socio-ecological systems have some capacity to absorb the pollution, reorganise and adapt to the external change, but this capacity is not unlimited and a slow change of the nitrogen pool may result soon or later in some chronic or drastic unexpected effects (Carpenter and Folke, 2006). In this context of uncertainty and variability, the challenge is to better understand the extent of nitrogen enrichment in water systems and the threats it poses for human and ecosystem health in the prospective of current changing drivers, such as climate change, land use change, pollution and economic growth, and to consider which mechanisms of adaptation and mitigation the science-policy interactions need to produce. In this chapter we will try to address this challenge. The paper starts by illustrating the trend of nitrogen in European rivers, aquifer and coastal waters, in order to understand the intensity and the location of the problem. Next, the threats posed by nitrogen enriched waters to human health and aquatic ecosystem functioning are considered and then analysed further in the light of major future drivers, notably land-cover and

380

climate changes. Finally, the paper proposes a reflection on the current adaptation and mitigation strategies and on the possibilities and positive synergies of science-policy interactions.

17.2╇ Nitrogen enrichment in European waters 17.2.1╇ Nitrogen trends in European surface waters In Europe some efforts have been made during the last two decades to reduce the nutrient input from waste water discharge, but diffuse pollution from agriculture still remains a major threat for waters (European Environment Agency, 2005). Figure 17.1 shows the observed annual nitrates concentrations reported by OECD (2008) at the mouth of some major European rivers. Trend analysis indicates that in Europe between 1992 and 2005 nitrogen and phosphorus concentrations remained relatively constant in lakes while they decreased in rivers (European Environment Agency, 2009). In fact, there was a slight decrease of nitrogen concentrations in European rivers compared to the values of the 1990s, except in southern Europe, and phosphorus concentrations have significantly declined, reflecting the general improvement in wastewater treatment and the reduction of phosphates in detergents (European Environment Agency, 2005). Yet, in Europe, trends in nutrient concentrations vary according to the different regions depending on local conditions. According to the information provided by the Member States on surface water quality (European Commission, 2007 COM(2007)120), between the two periods 1996–1999 and 2000–2003, the nitrate concentrations decreased in 55% of the monitoring stations and were stable in 31%. However, in 14% of the monitoring locations nitrate concentrations were increasing. Stations reporting increasing trends were located in Luxembourg, France, United Kingdom, Portugal and Belgium, while decreasing and stable trends were found in Denmark, Austria, Ireland, Sweden, Germany and the Netherlands (European Commission, 2007 COM(2007)120). These trends need to be evaluated regionally and considering the contemporary changes in nitrogen sources. In fact, between the two above mentioned periods, nitrogen input of mineral fertiliser and manure declined by 6% and 5%, respectively (European Commission, 2007 COM(2007)120), atmospheric deposition slightly decreased (Simpson et€al., 2011, Chapter 14 this volume) and nutrients point discharges were reduced by improving waste water treatments (European Environment Agency, 2005). Therefore, in certain areas nitrate concentration in surface waters may have remained constant in spite of some reduction in nitrogen inputs. According to data provided by Member States (EU27) in the last Nitrates Directive reporting exercise covering 2004–2007, nitrate concentration is increasing in 30% of the monitoring stations, while it is stable or decreasing in 70% of the stations (European Commission 2010 COM(2010)47). The current concentrations in rivers generate significant nitrogen loads to the seas (Billen et€al., 2011, Chapter 13

Bruna Grizzetti

Figure 17.1 Annual nitrate concentrations (in mgN/l) in surface water at the mouth of some major European rivers (from OECD, 2008).

this volume). In European coastal waters, nitrate concentrations have remained generally stable in the Baltic, North and Celtic Seas and have increased in some Italian coastal areas (Figure€ 17.2; European Environment Agency, 2005). Artioli et€al. (2008) compared nitrogen budgets for European seas over three periods:€ before eutrophication, during severe eutrophication and in current situation. According to their study in the Baltic Proper nitrogen and phosphorus riverine loads remained stable since the eutrophication period (1955–1985), in the Coastal North Sea nutrient inputs have declined after the severe eutrophication period (ending around 1990), and in the Northern Adriatic Sea riverine loads have increased for nitrogen, while they have been halved for phosphorus as a result of phosphate banning policies in detergents. For more details

on nitrogen trends in the European Seas see Voss et€al., 2011 (Chapter 8 this volume).

17.2.2╇ Nitrogen accumulation in aquifers Groundwater is an important resource in Europe, providing water for domestic use for about two third of the population but groundwater is a finite and slowly renewed resource and overexploitation associated with a degradation of water quality is putting in danger an important source of drinking water. In Europe, groundwater nitrate concentrations have remained stable and high in some regions (European Environment Agency, 2005). In the Third Assessment Report on the Implementation of the Nitrates Directive, the European Commission (European

381

Nitrogen as a threat to European water quality Figure 17.2 Nitrate, phosphate and N:P ratio trends observed in the European coastal waters. Source:€EEA web site http://www.eea.europa.eu Copyright EEA, Copenhagen, 2006.

Commission, 2007 COM(2007)120) reports that for the period 2000–2003 about 17% of the wells in EU15 exhibit a concentration of nitrate above the limit of 50 mg/l. An additional 7% were in the range between 40 and 50 mg/l, while about 60% were below 25 mg/l. Analysing the trend between the third and second assessment report, the European Commission found that even though 30% of the reported wells show an improvement in their concentration of nitrate, an alarming 36% show an increasing trend (European Commission, 2007 COM(2007)120). According to the last assessment report, covering the period 2004–2007, nitrate pollution in groundwater is still observed in 34% of the monitoring stations, with 15% of the stations with nitrate concentrations above 50 mg/l (European Commission, 2010 COM(2010)47). Background concentrations of nitrate in groundwater are very low. Most of the nitrates found in groundwater are thus of anthropogenic origin and mostly related to agricultural activities. Van Drecht et€ al. (2003) estimated the total leaching of nitrogen to groundwater at 55 Tg/yr at the global scale with a contribution of 8 Tg/yr for Europe, of which 40% will reach the rivers outlets. Contribution of deep aquifers mostly affected by historical use of fertiliser was estimated at 10% of the total load of nitrogen. These calculations were made at the global scale and might hide some spatial and temporal variations, however they agree with some more detailed estimates. Behrendt et€al. (2003) estimated the groundwater contribution to total load of nitrogen to be 48% for the Danube for the period 1998–2000. Schreiber et€al. (2003) analysing all German catchments found a groundwater contribution ranging from 38% to 69%. Palmeri et€al. (2005) estimated the contribution of groundÂ�water to total nitrogen load in the Po Valley to be around 36%. Even though highly variable and dependent on the degree of agriculture intensification and hydrogeological properties of the aquifers, groundwater is a significant source of nitrogen at the catchments outlets.

382

The quantity of nitrate present in the groundwater is strongly linked to the amount of nitrogen applied in agricultural land, and to the nitrogen surplus in particular. Indeed nitrogen surplus in agricultural land can be removed by surface runoff, leaching to the aquifer, and loss to the atmosphere or can be stored in the soil–water system. Bouraoui et€al. (2009) estimated the surplus of nitrogen for Europe at 11.5 Tg for year 2000 and 10 Tg for 2005 (Figure 17.3). This surplus was calculated without considering volatilisation from manure as this pathway is an additional pressure on the environment (details on surplus computation are given in the legend of Figure 17.3). At European scale, there is a decrease of nitrogen surplus for many countries (see Figure 17.3). Dramatic decreases are observed in the Netherlands, Denmark, and Germany where the nitrogen surplus is back to the level of that of 1970. However, there is still no strong evidence that the groundwater level is responding to the decrease of nitrogen surplus. The most striking cases are those of the Eastern countries that have seen a decrease by half of the nitrogen surplus, due to the economic and political changes at the beginning of the 1990s. Improvement in the water quality observed in streams is yet to respond to these changes, as large quantities of nitrate are stored in the aquifers and are released slowly depending on groundwater residence time, which may vary from weeks to several thousands of years (Alley et€al., 2002; Schlesinger, 2009). In addition, nitrogen stored in the soil system might be released slowly due to the mineralisation process (Stalnacke et€al., 2004; Grimvall et€al., 2000), and nitrate residence time in the unsaturated zone. For example, Sileika et€al. (2006), analysing long term data on nitrate concentrations in Nemunas River (Lithuania), noted a strong increase of nitrates in surface water from the Soviet period despite the large drop in fertilisation, due among other to a large storage and accumulation of soil nitrogen during the Soviet period. Even though no clear conclusion can be drawn on the response time of aquifers to changes in fertiliser application,

Bruna Grizzetti

Figure 17.3 Estimated nitrogen surplus for European and some Mediterranean countries (kg N per ha of agricultural land). The red bar indicates the year of implementation of the Nitrates Directive (1991/676/EEC). Nitrogen surplus is computed using a simple national mass balance approach. The inputs considered include:€mineral application of nitrogen (source FAO), nitrogen from manure application calculated using animal number (source FAO) and excretion coefficients non-corrected for volatilisation, atmospheric nitrogen deposition (source EMEP), symbiotic biological fixation (calculated as the nitrogen in crop harvest for soybean and pulses), non-symbiotic fixation (estimated at 25 kg/ha for rice and 5 kg/ha for other upland crops). The output considered was crop harvest taken as the crop yield (source:€FAO) multiplied by nitrogen crop content coefficients.

383

Nitrogen as a threat to European water quality

it can be expected that past unbalanced fertiliser strategies will impact for a long time the quality of European groundwater and in turn the surface water quality. Wriedt and Bouraoui (2009) estimated the average residence time for continental Europe for elementary river basin of about 100 km2 based on river density, slope, and parent material properties. According to these estimates, northern countries are characterised by shorter residence time (less than five years) and one can expect a faster reaction of the groundwater to any improved fertilisation strategy. Longer residence time are calculated in Southern Europe, Southern England (Chalk area), Poland, and along the North Sea Coast of Europe and thus any management decision to reduce nitrate load to groundwater might take decades to see any positive effect. All this evidence indicate that in Europe the past and current anthropogenic activities are impacting and might impact water resources for the years and decades to come. Indeed, nitrogen surplus from agriculture is still high in many countries and huge quantities of nitrogen are stored in the soil or aquifers. There are some major concerns as the Eastern countries will probably intensify their agriculture and thus their fertilisation rate in the near future, and some countries of Western Europe have not seen their nitrogen surplus decrease but rather stabilise at high levels. Efforts have been taken, through conventions or the application of binding Directives, and still Europe’s waters are suffering from excess nitrogen. It is a complex task to estimate how and how long it will take to restore Europe’s waters to good quality. So it is still a priority to assess how this excess nitrogen is affecting both human and ecosystems health, and evaluate how this impact will vary in a changing environment.

17.3╇ Threats for human and ecological aquatic systems 17.3.1╇ The human–ecological system The human and the environmental systems are strictly interconnected. Humans are altering the natural nitrogen cycle to increase their benefits from nature, but they are affected by the changes they are causing in the environment. This section analyses the threats induced by the nitrogen enrichment in European waters considering humans and aquatic ecosystems as the principal receptors of impacts in a context of inherent mutual relation for which human actions affect ecosystems and the impaired ecosystems affects human health and well-being. The Millennium Ecosystem Assessment has highlighted the benefits that people obtain from nature, the ecosystem services, and the tight link between human and ecological systems (Millennium Ecosystem Assessment, 2005). Ecosystem services include provisional services, such as food, fresh water, wood and fibre; regulating services, which affect water purification and regulation of climate, flood and disease; cultural services, which provide recreational, aesthetic and spiritual benefits; and supporting services, such as nutrient cycling, soil formation and primary production (Millennium Ecosystem

384

Assessment, 2005). These services support livelihood and development of human society and their sustainable use is fundamental for the human wealth and security (Folke et€al., 2002). Water and nitrogen are directly or indirectly involved in all the ecosystem services. Nitrogen in aquifers and reservoirs impairs water quality for drinking purpose, affecting directly the human health. Nitrogen enrichment in lakes, rivers and coastal and marine waters may produce the phenomenon of eutrophication which has detrimental effects for aquatic ecosystems. Consequently, the ecosystem per se is damaged and some services such as fish provision or aesthetic and recreational uses are directly affected. Moreover, through the intensive mobilisation, the nitrogen cycle and primary production may become distorted and some regulating services may be reduced or compromised. For example evidence shows that busting denitrification through additional nitrogen input in the water system may increase the emission of N2O to the atmosphere, which acts as a strong greenhouse gas affecting the climate. Studies have highlighted that the efficiency of water purification may be reduced increasing the total amount of nitrogen input in the river system (Mulholland et€ al., 2008), and that aquatic ecosystems significantly impacted by eutrophication are more vulnerable to flood events and diseases spreads (Folke et€ al., 2004; McKenzie and Townsend, 2007). Thus, altering the nitrogen availability in water is likely to significantly affect the social and environmental system, reducing many ecosystem benefits. The socio-ecological system has some capacity, resilience, to ‘absorb disturbance and reorganise while undergoing change so as to still retain essentially the same function, structure, identity and feedbacks’ (Walker et€ al., 2004). Resilience is often associated with diversity, such as biological species and economic options, to support the ecosystem capacity to renew and reorganise into a desired state although under pressure (Carpenter and Folke, 2006). Moreover, the actors of the socio-ecological system have the possibility to adapt to ongoing changes in order to moderate the undesired effects and as well to try to reverse them. However, this ecosystem ability to absorb stress and recuperate is not linear and is possible only until a certain threshold (tipping point), beyond which the recovery is difficult or impossible, leading to a regime shift. Human degradation of the environment has reduced the ecosystems resilience, shrinking the ability to mitigate natural hazards (Carpenter and Folke, 2006) and increasing the likelihood of drastic changes to less desired capacity to generate ecosystem services (Scheffer et€al., 2001; Folke et€al., 2004). There are various ways in which nitrogen enrichment of water can affect human and aquatic ecosystem health. Direct effects from use of drinking water have been described extensively in the scientific literature and are the subject of current policies. Indirect effects such as from eutrophication are less well known. In the following part we describe the threats on human health due to nitrogen in drinking water and then the threats on aquatic ecosystems and their consequent indirect effects on human health and well being, providing, where possible, estimates at European scale.

Bruna Grizzetti

17.3.2╇ Effects of nitrogen rich drinking water on human health Currently in many countries there are strict limits on the permissible concentration of nitrate in drinking water and in many surface waters. The limit is 50 mgNO3/l in the European Drinking Water Directive (Directive 98/83/EC) and 44 mgNO3/l in the United States (equivalent to 11.3 mgN/l and 10 mgN/l, respectively). These limits are in agreement with WHO recommendations established in 1970 and recently reviewed and reconfirmed (WHO, 2007; the exact formulation of the standard is that the sum of NO3/50 + NO2/3 should not exceed€1). The European Nitrates Directive also sets a limit concentration of 50 mgNO3/l for groundwater and surface water, as a threshold value for Member States to protect water bodies. There are two main health issues related to nitrate in drinking water:€the linkage with infant methaemoglobinaemia, also known as blue baby syndrome, and with cancers, for example of the digestive tract (Ward et€al., 2005). The evidence for nitrate as a cause of these serious diseases is controversial (Powlson et€al., 2008; Salomez and Hofman, 2003). In addition there is evidence for increased cardiovascular health with increased nitrate intake (Webb et€al., 2008). Presently, it is widely accepted that methaemoglobinaemia in Europe is rare, and that in general incidence is related to presence of pathogens in drinking water rather than to nitrate or nitrite (Addiscott, 2005). The emerging and returning question is whether nitrate in drinking water is harmful to humans, and if the drinking water standard in some cases could be increased (van Grinsven et€al., 2006; L’hirondel and L’hirondel, 2002). Such an increase would have great implications for policies, measures and costs related to water treatment and to fertiliser and manure application. To answer this question, the following points are relevant. • Only a small proportion (<20%) of nitrate intake in most humans is related to drinking water, most nitrate intake comes from meat and vegetables. • Nitrate is a major component in human metabolism and harmless for humans. However, N-nitroso compounds, derived from nitrate (through nitrite), are probably carcinogenic to humans. Ingested nitrate or nitrite under conditions that result in endogenous nitrosation is probably carcinogenic to humans. The underlying mechanism is endogenous nitrosation, which in the case of nitrate must be preceded by reduction to nitrite. Nitrosating agents that arise from nitrite under acidic gastric conditions react readily with nitrosatable compounds, especially secondary amines and alkyl amides, to generate N-nitroso compounds, many of which are carcinogenic (IARC, 2006). • The WHO-standard of 50 mgNO3/l is a compromise that does not account for the possible role of nitrate as a precursor for carcinogenic N-nitroso compounds. • It is virtually impossible to relate incidence of cancers to nitrate in drinking water, in view of the presence of many nitrate sources, the complexity of nitrogen metabolism and many life style and diet factors enhancing or preventing

cancers. Epidemiological studies suggesting an association between nitrate in drinking water and various health defects are therefore rare (Ward et€al., 2005). Reasons for policy to consider health risk associated with increased nitrate in drinking water are the chronic and potentially massive exposure, the empirical evidence for increased risk and the fairly easy options to prevent exposure. An additional emerging issue is related to the formation of carcinogen N-Nitrosodimethylamine (NDMA) upon disinfection of drinking water. Disinfectants may react with other compounds (precursors) in the water, like pesticides and their degradation products, to form NDMA. The latter is a potent carcinogen that is not easily removed by normal drinking treatment procedures (Schmidt and Brauch, 2008). Depending on environmental conditions and presence of nitrosamine precursors, disinfectant treatment may even enhance NDMA formation (Zhao et€al., 2008). NDMA may become a future threat, but the role of dissolved inorganic and organic nitrogen in NDMA formation is not yet clarified.

Tentative European assessment Drinking water sources In Western Europe almost 100% of the population has access to safe drinking water, while in the Eastern part the proportions vary from 58% to 80%, with lower values for rural areas, where only 30%–40% of households have access to safe drinking water (WHO, 2007). In European countries with lower GDP or large rural areas, substantial proportions of the population have no connection to public water supply. Safety and adequacy of drinking water supply in areas with no public supply or small local facilities are expected to be less systematically monitored and therefore not guaranteed. In Europe, water for drinking use is abstracted from aquifers, rivers or reservoirs, according to the regional availability of the resource, and then treated for human consumption. According to the information reported by the European Commission (2007) for the period 2002–2004, the relative shares of surface water, groundwater and other sources to the production of drinking water were 66%, 33% and 4%, respectively. These values represent an average of the information reported by 14 EU Member States and refer only to larger water supply zones, which served about 78% of the total population living in the EU14 countries. Figure 17.4 reports the relative use of water resource for drinking purposes per country (European Commission, 2007). Groundwater is the main source for drinking water in Austria, Denmark, Italy, Spain, Germany, France, Belgium and the Netherlands, while the use of surface water is dominant in Finland, Czech Republic, Estonia, Ireland, Portugal and United Kingdom. Large parts of the European population consume bottled mineral water and beverages (UNESDA, 2009), which, in general, are low in nitrate and nitrite (Griesenbeck et€ al., 2009). In EU27, the average annual consumption of bottled drinking water increased from around 87 l/capita in 2001 to 106 l/ capita in 2007. Consumption varies in the different countries,

385

Nitrogen as a threat to European water quality Figure 17.4 Relative contributions of surface water, groundwater and other sources to the production of drinking water in 14 EU Member States. The production of drinking water is expressed as population served per country by larger water supply zones (source:€European Commission, 2007).

mainly as a consequence of specific cultural and market driven lifestyles. Drinking water quality The EU Drinking Water Directive (Directive 98/83/EC) requires Member States to report drinking water quality to the Commission every three years. In the synthesis report for 2002–2004 (European Commission, 2007) nitrate is listed as one of the parameters most often causing non-compliance in the larger water supply zones (serving more than 5000 persons or producing over 1 million litres per day). The synthesis report points out that it is not easy to draw conclusions on trends in water quality for individual Member States and for Europe in view of the often very incomplete or incompatible country reports. Percentage of non-compliance is no direct indicator of potential exposure to drinking water exceeding the nitrate or nitrite limit, in fact non-compliance may be incidental and restricted to specific water supply zones and the mandatory reporting does not cover small supply zones. For example in Denmark about 30% of the population is served by smaller facilities, in Austria, France, Germany and Ireland the proportion is close to 25% (European Commission, 2007). Non-compliance for nitrate or nitrite is reported regularly but rarely exceeds 4% and is restricted to a similar proportion of monitored and reported supply zones. Assessment of potential exposure Considering data from 12 of the EU15 Member States, van Grinsven et€al. (2006) estimated that almost 3% of the population using drinking water from groundwater resources is potentially exposed to concentrations exceeding 50 mgNO3/l, and 5% to concentrations exceeding 25 mgNO3/l (see also van Grinsven et€ al., 2010). De Roos et€ al. (2003) and Gulis et€ al. (2002) found evidence of 25 mgNO3/l as a critical concentration for health effects. Exceedance is caused both by untreated private supply and insufficiently treated public supply. Combining the information on population density and estimates of nitrate concentration in European surface waters (Figure 17.5), it results that around half of the European population lives in areas with nitrate concentrations higher than 25€mgNO3/l, and about one fifth in areas with values higher that 50 mgNO3/l, with variations by country (Figure 17.6).

386

In the USA, Dodds et€al. (2009) have analysed the potential economic damages caused by eutrophication in American freshwaters with regard to drinking water use, considering the spending for bottled water. The estimated associated cost for bottled drinking water consumption in the EU27 is significant (Figure 17.7). However, people consume bottled drinking water not only because of concern about drinking quality but also because of social–cultural preferences. In addition to drinking water treatments, other costs are encountered by the society related to the health problems caused by nitrogen enrichment in drinking water. However, they are difficult to estimate because of the scarcity of targeted epidemiological studies. Van Grinsven et€ al. (2010) provide an assessment of social cost of colon cancer due to nitrate in European drinking water. They estimate a 3% increase of incidence for 11 Western European countries, corresponding to a health loss of 2.9 euro/capita per year. For a more detailed discussion on costs related to nitrogen pollution see Brink et€al., 2011 (Chapter 22, this volume).

17.3.3╇ Effects of nitrogen on the aquatic ecosystems health The anthropogenic increase of nitrogen and phosphorus in lakes, reservoirs, rivers and coastal waters is the main cause of eutrophication. The phenomenon was observed in temperate lakes during the 1960s (OECD, 1982) but it rapidly expanded to estuaries and coastal seas. In general, in aquatic ecosystems the increase of nutrient concentration promotes the phytoplankton growth and generates an imbalance between algal production and consumption. As a result, biomass sedimentation and microbial decomposition are enhanced and a large part of bottom-water oxygen is consumed (Durand et€ al., 2011, Chapter 7, this volume; Voss et€al., 2011, Chapter 8, this volume). As nitrogen is a crucial element for photosynthesis and primary production, its significant enrichment in the water medium directly alters fundamental processes of the aquatic ecosystem. The role of nitrogen in water eutrophication depends on its relative availability with respect to other elements, such as carbon, phosphorus and silica (Billen and Garnier, 2007). Scientists use the concept of the ‘limiting

Bruna Grizzetti

Figure 17.5 Estimated nitrate concentration in surface waters for year 2000. The spatial units of analysis are sub-catchments of 180 km2 average size (data are derived from the estimates of Bouraoui et€al., 2009).

Figure 17.6 Percentage of population living in areas with different average nitrate concentration in surface waters. The values derive from the overlay between population density per subcatchment and estimated nitrate concentration in the relative surface water (shown in Figure 17.5).

element’, which is the element that limits the potential rate of primary production (Howarth, 1988). It has often been observed that freshwaters are limited by phosphorus, whereas marine waters by nitrogen. However, this is a general simplification and each aquatic system has its peculiarities. In fact, in freshwaters, estuaries and coastal marine ecosystems the limiting role between nitrogen and phosphorus may change seasonally and spatially (Conley et€al., 2009). For a thorough

discussion on limiting element, nutrient ratios and their effect on the quality of freshwaters and coastal and marine waters see Durand et€al. (2011), Voss et€al. (2011) and Billen et€al. (2011, Chapters 7, 8 and 13 this volume). Eutrophication causes many negative effects on the aquatic ecosystem (for extensive review see Carpenter et€ al., 1998; Cloern, 2001; Smith, 2003; Smith and Schindler, 2009). Table€17.1 summarises some of the main effects.

387

Nitrogen as a threat to European water quality Figure 17.7 Estimated cost for bottled drinking water per capita per country. Estimates refer to year 2001. Data on packaged water consumption per capita are taken from UNESDA (2009). An average cost of 0.15 euro/l is assumed for the estimation.

Table 17.1 Adverse effects caused by eutrophication in aquatic ecosystems (from Smith and Schindler, 2009, and references therein)

Effects of eutrophication Increased biomass of phytoplankton and macrophyte vegetation Increased biomass of consumer species Shifts to bloom-forming algal species that might be toxic or inedible Increases in blooms of gelatinous zooplankton (marine environments) Increased biomass of benthic and epiphytic algae Changes in species composition and macrophyte vegetation Decline in coral reef health and loss of coral reef communities Increased incidence of fish kills Reduction in species diversity Reduction in harvestable fish and shellfish biomass Decreases in water transparency Taste, odour and drinking water treatment problems Oxygen depletion Decreases in perceived aesthetic value of the water body

Excessive nutrient loading produces an increase of algal biomass and a change in the species composition of the algal communities (Smith, 2003). Eutrophication can induce a progressive selection in phytoplankton towards fast-growing species. Bloom-forming algae (Figure 17.8) may be directly toxic for the components of the aquatic ecosystem and for humans and impairs waters for fisheries, recreation and drinking (Carpenter et€ al., 1998). The consequent sedimentation and decomposition of the dead algae depletes oxygen in bottom waters, especially in ecosystems with low rate of water turnover. Anoxic conditions kill benthos and fishes and cause extensive loss of habitats. Increasing coastal zones with significant oxygen depletion (dead zones) were reported in the last half of

388

Figure 17.8 Picture of Phaeocystis foam in Texel, the Netherlands. (Source:€Gilles Billen.)

the twentieth century in the Adriatic Sea, the Black Sea, the Kattegat and the Baltic Sea (Diaz and Rosenberg, 2008). Some algae can produce toxins that can be adsorbed or ingested via drinking waters or feeding by aquatic animals and humans. Cyanobacteria, dinoflagellates and diatoms are among the main taxonomic groups contributing to toxic algae blooms and evidence shows that their occurrence can be stimulated by nitrogen pollution (Camargo and Alonso, 2006). There are clear indications that nutrient enrichment enhances harmful algal blooms (HABs), and some evidence of a connection between increased urea fertiliser use and HABs (Glibert et€al., 2006). In all cases, the linkages between HABs and eutrophication are complex and include nutrient enrichment as well as indirect pathways (Anderson et€al., 2002). The explosive growth of undesired algae, with the consequent oxygen depletion, may induce important changes in the habitat quality and affects both the species abundance and community composition, introducing some alterations in the trophic web. For example, in shallow coastal areas the changes from perennial macroalgaes and seagrasses to �ephemeral macro� algaes, produced by nutrient enrichment, may cause loss of

Bruna Grizzetti

habitat for aquatic animals (Burkholder et€al., 2007). Similarly, proper habitat for macroalgae or benthos may shrink where water transparency or oxygen availability decline. Evidence has shown that eutrophication reduces the richness and abundance of aquatic species. ‘Nutrient enrichment can cause a change in the selective forces that regulate the biological diversity at all trophic levels within the coastal food webs’ (Cloern, 2001). Moreover, eutrophication and human actions promote the invasion of exogenous species, which further contribute to reducing the native biodiversity. According to Lotze et€al. (2006), who performed a study considering 12 temperate estuarine and coastal ecosystems covering Europe, North America and Australia from prehistory to actual times, by the end of the twentieth century, 7% of the recorded species were extinct and 91% were depleted. Moreover, land reclamation, eutrophication, diseases and destruction were responsible for the destruction of 67% of wetlands, 65% of seagrasses, and 48% of submerged aquatic vegetation. However, major impacts on diversity, structure and functioning of the coastal ecosystems were due to overexploitation and habitat destruction, and eutrophication was only part of the problem (Lotze et€al., 2006). Diversity enhances the ecosystem stability and supports ecosystem services:€fisheries, provision of nursery habitats and filtering and detoxification services (Worm et€al., 2006). Large collapses of marine fisheries occur in species-poor ecosystems (Worm et€al., 2006). Collapse of fish stock is related as well to unsustainable fishing and loss of habitat. Nitrogen pollution can also more directly affect fish populations and fishery yields, with both positive and negative consequences (National Research Council, 2000). Up to a point, increased loads of nitrogen to coastal marine ecosystems can increase secondary production€– including fish production€– as primary production increases and more energy cascades up the food web (Nixon, 1988). However, with further inputs of nitrogen to the ecosystem, hypoxia, anoxia, and other changes in the ecosystem can lead to lower production of fish and less fish harvest. Caddy (1993) hypothesised that the effects would be seen on demersal fish, due to greater adverse consequences of eutrophication on the benthos, with adverse consequences on pelagic fish occurring only at greater nutrient loads. Because of the high variability in fish populations and fish harvests, collecting data to demonstrate the responses suggested by Caddy (1993) has been difficult. However, recently, Oczkowski and Nixon (2008) showed that for the fisheries of the Nile delta, the relationship between nitrogen load and fish does indeed follow a threshold response:€fish catch increase with nitrogen load to a point, but beyond the threshold, fish catches fall significantly. Recent research studies have raised the question about the influence of nitrogen enrichment in aquatic ecosystem and the dynamics of parasitic and infectious diseases. Pathogen infections directly threaten humans, animals and aquatic ecosystem health. Although direct experiments are rare in literature, the available evidences confirm a significant positive correlation between increased nutrients and diseases (McKenzie and Townsend, 2007; Hartikainen et€al., 2009). Eutrophication can contribute to infectious diseases spread directly, enhancing the

replication rate of aquatic pathogens, or indirectly, influencing the abundance and distribution of pathogens hosts and vectors. Through experimental mesocosmos studies, Johnson et€ al. (2007) have shown that eutrophication enhances amphibian disease, increasing the density of infected snail hosts and the per-snail production of parasites. Camargo and Alonso (2006) have reported several studies describing algal blooms associated with cholera outbreaks and showing positive correlation between inorganic nutrient concentrations and larval abundances of mosquitoes, which are potential carriers of pathogenic micro-organisms. More studies are needed to understand the possible links between nitrogen enrichment and infection disease risk, especially in the presence of other environmental stressors (Smith and Schindler, 2009). This could be relevant for European aquatic ecosystems, which are likely to face in the near future temperature increases, with a consequent change in geographic distribution of insects and mosquitoes. The effects of eutrophication on the availability and the biochemical cycle of non-nutrient pollutants, such as heavy metals, pesticides, pharmaceutical and hormones have not been thoroughly explored yet, especially for emerging pollutants. Altering the physico-chemical properties of the water medium, like hindering light penetration and oxygen availability, eutrophication may influence the biodegradation of some compounds or create more favourable conditions for pollutants release from sediments. Supplies of nitrogen and phosphorus support bacterial growth and as a consequence can enhance the biodegradation of chemicals, such as aromatic hydrocarbons and pesticides (Smith and Schindler, 2009). Accelerating the phytoplankton and benthic growth, eutrophication increases the exchanges with water and bio� accumulation through the trophic chain, but it could produce as well a growth dilution effect. Therefore, the actual synergies between eutrophication and non-nutrient contaminants strictly depend on the characteristics of the chemicals, the physico-chemical conditions of the system and the local specific food web functioning. In addition to the effects caused by eutrophication, the role of nitrogen deposition in contributing to surface water acidification is well recognised (Durand et€ al., 2011, Chapter 7 this volume). Sulphur has been the main agent contributing to acidification but due to large reductions in sulphur dioxide emissions over Europe since 1980, nitrogen has become more important. Although sulphur deposition is still a main agent of acidification around Europe, some regions (such as parts of the UK, Italy and the Alps) experience such high levels of nitrogen deposition that even if sulphur deposition could be eliminated completely, nitrogen deposition alone would still cause critical load exceedance and long-term acidification (Curtis et€al., 2005a, b).

Tentative European assessment Eutrophication is the result of nutrient enrichment in the aquatic system, but the severity of the phenomenon largely depends on the specific regional characteristics, climate, morphology, water residence time, nutrient concentration and ratio, trophic web status, and generally on the ecosystem resilience. Therefore,

389

Nitrogen as a threat to European water quality

similar nutrient loads may produce different effects because of the regional sensitivities, and in particular of the relative availability of the other nutrients. Nevertheless, nitrogen and phosphorus concentrations, being the major driving forces of the problems, can be used as a proxy to evaluate the risk of water eutrophication (Phillips et€ al., 2008). According to Camargo and Alonso (2006) who performed an extensive study on the ecological and toxicological effects of inorganic nitrogen pollution in aquatic ecosystems, total nitrogen levels lower than 0.5–1.0 mg/l might prevent aquatic ecosystems from developing eutrophication and acidification. This does not apply for ecosystems naturally rich in nitrogen. The total nitrogen concentration was used as a proxy to evaluate the potential risk of surface water eutrophication Table 17.2 Classes of potential risk of eutrophication for surface water related to total nitrogen concentrations

Level of potential risk of euthropication

Total nitrogen concentration (mg/l)

Low

< 0.5

Medium

0.5–1.5

High

> 1.5

Values are based on literature (Guidance Document on Eutrophication from Intercalibration Group; Vollenweider et€al., 1976; OECD, 1982; Cardoso et€al., 2001).

at European scale. Three classes of potential risk were established based on the literature (Vollenweider et€al., 1976; OECD, 1982; Cardoso et€ al., 2001) and on the information available from the Water Framework Directive Intercalibration Exercise (Guidance Document on Eutrophication from Intercalibration Group). The classification of total nitrogen concentration in classes of potential risk is shown in Table 17.2. These values, which are in agreement with the ecological thresholds reported in other studies (Durand et€al., 2011, Chapter 7 this volume), were combined with estimates of total nitrogen concentrations in European surface waters (Bouraoui et€ al., 2009) to derive a European map of potential risk of eutrophication related to nitrogen (Figure 17.9). The map (Figure 17.9) shows that large parts of European water courses may be threatened by potential risk of eutrophication due to nitrogen concentrations. The risk varies with countries but is generally lower in the Scandinavian region (Figure 17.10). Figure 17.11 reports the potential risk of eutrophication per country for European inland wetlands using the same nitrogen concentration classification. The Water Framework Directive requires that Member States evaluate the ecological status of their surface waters based on the deviation in the status of a number of biological elements (such as phytoplankton and macrophytes) from the water body type specific reference condition. The division of water bodies into types and the establishment of type specific reference conditions should allow for better resolving the effects on the biological elements (biological Figure 17.9 Map of potential risk of eutrophication for surface freshwater based on estimated total nitrogen concentrations. The map shows three classes of nitrogen concentration estimated in surface waters associated with a potential risk of eutrophication. The related total nitrogen concentration per class of risk is reported in Table 17.2.

390

Bruna Grizzetti Figure 17.10 Percentage of surface of inner waters per class of potential risk of eutrophication related to total nitrogen concentration (see Table€17.2) per country. Inner waters refer to Corine Land Cover 2000 classes 40 and 41 (water courses and water bodies, respectively).

Figure 17.11 Percentage of surface of inner wetlands per class of potential risk of eutrophication related to total nitrogen concentration (see Table 17.2) per country. Inner wetlands refer to Corine Land Cover 2000 classes 35 and 36 (inland marshes and peat bogs, respectively).

communities) from the variation associated to natural pressures (such as natural water fluctuations) and thus a more precise evaluation of the ecological condition. Figure 17.12 shows the ecological status of the European lakes, based on data that have been collated as part of the Water Framework Directive Intercalibration exercise (2004–2007). The ecological status of lakes was evaluated by chlorophyll-a values by the Mediterranean, Atlantic, Central/Baltic and Northern Geographic Intercalibration Groups (GIG), while for the Northern GIG the ecological status evaluation included chlorophyll-a values and indicators of the status of macrophytes communities’ composition. Concerning water acidification, within Europe, nitrogen and sulphurs emission ceilings are set to prevent critical load

exceedance under both the Gothenburg Protocol of the UN-ECE Convention on Long-range Transboundary Air Pollution (CLRTAP) and the EU National Emissions Ceiling Directive (NECD) (see Hettelingh et€al., 2007, 2008). Five countries currently submit freshwater critical loads data to the international mapping and modelling programme under the CLRTAP; the UK, Norway, Sweden, Finland and Switzerland (Canton Ticino). However, impacted acid sensitive lakes and streams are present in many other countries including France, Spain, Italy, Austria, Poland, Slovakia, Czech Republic, Romania, Bulgaria and Germany (Curtis et€al., 2005b; Evans et€al., 2001). The critical load of total nitrogen deposition, when excluding the effects of sulphur deposition, is < 400 eq/ha per year across large areas of Scandinavia and parts of the UK and the

391

Nitrogen as a threat to European water quality Figure 17.12 Ecological status of European lakes according to the harmonised methodology of the Water Framework Directive Intercalibration Group (in most of the region the ecological status is evaluated by chlorophyll-a values, explanation in the text). The lakes are subdivided in three main classes of High, Good and Less Good ecological status (Lake Intercalibration data provided by the Lake Geographical Intercalibration Groups and available at CIRCA folders:€http://circa.europa.eu/ Members/irc/jrc/jrc_eewai/library. Data organised by Sandra Poikane, Joint Research Centre).

Swiss Alps and these deposition levels are exceeded across large regions of Europe (EMEP, 2009).

17.4╇ Prospects for European water quality 17.4.1╇ Impact of climate change on freshwater quality Most striking expected climate change is a continental global warming from 1 °C to 4 °C more pronounced during winter in Eastern Europe and in summer for Southern and Western Europe. Precipitation exhibits a strong north to south gradient with an increase in northern Europe and a decrease going south (Alcamo et€al., 2007) and a more pronounced seasonality. In addition there is a high probability of increase of extreme events. These changes will affect directly and indirectly the quality of freshwater and coastal and marine waters in Europe (Battarbee et€al., 2008; European Environment Agency, 2008). Indeed the predicted climate change will affect the nutrient cycling in the water bodies, but will also affect the generation and transport processes of nitrogen originating from land based sources. The impact of climate change on water quantity has been studied extensively while fewer researches have focused on the impact of water quality. Water quality and nutrient concentrations in particular, will be affected by water quantity due to dilution or concentration effects, mobility, residence time and water temperature impact (Ducharne et€ al., 2007; Whitehead et€al., 2009). Direct impacts of increased temperatures will be

392

accelerated nutrient cycling (Whitehead et€al., 2009; Murdoch et€al., 2000). Furthermore, climate change or global change in general often results in synergistic effects of the different stressors (temperature increase, precipitation and land use change). It is thus very difficult to isolate the single effect of each stressor on the fate of nitrogen. In southern European countries it is expected that climate change will result in lower runoff, exacerbated by increased water abstraction for irrigation purposes and growing human consumption. An increase in the number of intermittent streams and drought periods is also expected. The higher population release of nutrients due to eating habits with an associated increased industrial production will release increased amount of nutrients in the lower water volume resulting in both higher nutrient loads but also higher nutrient concentrations in the water courses due to point source discharges, unless additional treatment is applied (Eisenreich, 2005). Higher nutrient concentrations downstream of point discharges (industries, concentrated animal management operations) are also predicted by Whitehead et€al. (2009) and Murdoch et€ al. (2000). Concerning agriculture, even though in Southern countries the decrease in water precipitation is expected to result in lower nutrient loads, the concentration of nutrients might increase due to lower water volume circulating in the streams (Mimikou et€ al., 2000). Therefore, for southern regions, an increased contribution of point sources relative to diffuse sources is anticipated with associated management problems.

Bruna Grizzetti

In northern regions, higher precipitation may also increase nutrient loads, and higher temperatures are likely to enhance nitrogen mineralisation. Ducharne et€al. (2007), studying the Seine river basin, predicted that an increase of temperature by 2.3 °C will result in an increased soil nitrogen mineralisation from 8% to 26%. Taking into account all processes enhancing or decreasing nitrogen production from mineralisation, they expected a net increase of 20% for nitrate leaching and nitrate in streams if appropriate farming practices are not implemented. Bouraoui et€ al. (2002), evaluating the impact of six climate change scenarios for the River Ouse (UK), found that climate change will increase nutrient losses to surface water due to the combined effects of increased mineralisation and the increased amount of water circulating through the soil profile. Wilby et€al. (2006), who studied the impact of climate change on water resources in the River Kennet (UK), also predicted increasing concentrations of nitrates and ammonium until 2050. They estimated that due to the increased occurrence of summer drought, nitrates will build up in the soils and will be flushed into the streams when droughts break. Bouraoui et€al. (2004) evaluated the impact of observed climate change occurring in Finland and predicted that the increase in temperature associated with the increased winter runoff due to accelerated snowmelt results in higher nitrogen loading by about 3% with pronounced seasonality. They found increased losses of nitrogen during the winter and reduced loads during the traditional snow melt period. Similar conclusions were reached by Arheimer et€ al. (2005) when studying the Ronnea catchment (Sweden). The authors predicted an increased nitrogen loading due to increased winter precipitation and increased mineralisation from soil organic matter despite an increase in surface and groundwater nitrogen retention. The increased retention under the climate change scenarios is attributed to increased temperature, increased nitrogen concentration and drier summer conditions yielding a longer transit time in the summer, the most active season for nitrogen removal process. They also noticed an extension of the areas contributing to the total nitrogen loads, with obvious management implications. Assessing the impacts of climate change on nutrient loads or concentration in freshwater and coastal ecosystems is not straightforward as many other factors impact their status and health. Eutrophication is not only controlled by nutrient availability but is also affected by light conditions, temperature, residence time, flow regime. Invasion of non-native species will obviously depend on temperature, flow conditions but also on nutrient availability and cycling (Murdoch et€al., 2000; Rahel and Olden, 2008).The impact of climate change on freshwater ecosystems has been studied in controlled experiments in mesocosms tanks (Feuchtmayr et€ al., 2007; Christoffersen et€al., 2006; Moss et€al., 2003). Feuchtmayr et€al. (2007) showed in their experiment that higher nutrient concentrations would result in increased population of four out of eleven of the macroinvertebrates used in their study. McKee et€al. (2003) measured an increase in zooplankton abundance under increased nutrient addition. Moss et€al. (2003) showed that nutrient addition (unlike warming) increased phytoplankton chlorophyll a concentration and total algal biovolume, and did not affect the

number of species. In northern boreal regions where lentic and lotic ecosystems are oligotrophic, an increase of nutrient loads might lead in earlier stages to an increase of biodiversity. Conversely, in southern boreal countries where freshwater ecosystems are eutrophic, an increase in nutrient loads could result in a decrease of biodiversity. Under the combination of increased temperature and nutrient inputs, arctic lakes will exhibit an increase algal production and biomass leading to a potential colonisation of predatory fish (Flanagan et€al., 2003). The impact of climate change on ecosystems has been studied through the detailed analysis of dry years and even droughts. Many authors report that under increased temperature, blooms of the harmful of cyanobacteria will likely increase (Johnk et€al., 2008; Paerl and Huisman, 2008). Many reasons explain this increase in bloom:€higher temperatures favour the cyanobacteria growth, the warming of the surface water will reducing vertical mixing, and there will be an increase of the growth period to an earlier shift of stratification in spring and late destratification in autumn. Mooij et€ al. (2005) reviewing the impact of climate change on lakes in the Netherlands also report the increased presence of cyanobacteria and their predominance in the phytoplankton community. The increased loading of nutrients for instance under severe storms or wet winters followed by dry or drought condition and associated increased residence time will increase algae blooms (Paerl and Huisman, 2008). Arheimer et€ al. (2005) concludes also that under climate change, the increased amount of inorganic nitrogen entering the Ronnae lake (Sweden) will stimulate algae growth resulting in increased concentrations of cyanobacteria, zooplankton and detritus. Similar effects are reported by Whitehead et€al. (2009). For additional information on the effects of climate change on coastal and marine waters and the implication for the link between nitrogen and carbon cycle see Voß et€al., 2011 (Chapter 8 this volume). Studies at the European scale have shown that climate changes are expected to alter the soil nitrogen cycle with great regional variability (Bouraoui and Aloe, 2007), and to exacerbate the problem of eutrophication, enhancing algal blooms and new harmful invasive algae (project EURO-LIMPACS, Battarbee et€al., 2008). All these studies are affected by uncertainty, but some of the climate change effects are already being observed in European water ecosystems (Battarbee et€al., 2008; European Environment Agency, 2008).

17.4.2╇ Nitrogen fate under future land use changes Due to the interdependence of the world economic and ecological systems, prospective scenarios of human activities in watersheds should be conceived at the global level. The Millenium Ecosystem Assessment has provided the story lines of four scenarios of the world future named Global Orchestration (GO), Order from Strength (OS), Technogarden (TG), and Adapting Mosaic (AM) (Alcamo et€al., 2006). The scenarios differ in terms of environmental management (pro-active or reactive) and in the degree of connectedness among and within institutions across country borders (globalisation or regionalisation). Technogarden and

393

Nitrogen as a threat to European water quality Table 17.3 Total nitrogen and total phosphorus river load to European coastal waters calculated by the GlobalNEWS models for 2000 and for 2030–2050 according to the two extreme Millenium Ecosystem Assessment scenarios

Total nitrogen load (TgN/yr)

Total phosphorus load (TgP/yr)

2000 reference

4.04

0.59

2030 Global Orchestration

4.01

0.56

2050 Global Orchestration

3.99

0.54

2030 Adapting Mosaic

3.64

0.55

2050 Adapting Mosaic

3.50

0.53

Scenario

Adapting Mosaic were developed assuming pro-active environmental management, while Order from Strength and Global Orchestration assume reactive environmental management. Global Orchestration and Technogarden reflect trends towards globalisation, while regionalisation is assumed in Order from Strength and Adapting Mosaic. The IMAGE and GlobalNEWS models of watershed nutrient fluxes have been used to calculate nitrogen and phosphorus delivery at the outlet of 5700 world watersheds (Seitzinger et€ al., 2009). The results for European watersheds are shown in Table 17.3 for the two more contrasted scenarios GO and AM. The trends shown are the results of several opposite drivers. The number of inhabitants with a sewage connection will further increase in all scenarios, but the removal of nitrogen in wastewater treatment also increases. On the other hand, the overall nitrogen use efficiency in agricultural production in Europe increases in all scenarios, particularly Adapting Mosaic (AM), and this leads to an overall reduction of nitrogen river export. In the GO scenario there is also an increasing production. In AM, with a faster population growth than in GO between 2000 and 2030, the nitrogen export will decrease. This is caused by a combination of lower meat consumption and a major effort in better incorporating animal manure in the agricultural system. Although global scenarios are best able to take into account the interconnected nature of world economies, there is also a need to increase the spatial resolution of their simulated results, in order to examine them at the sub-regional scale, closer to that at which management decisions are to be taken. With this in mind, Thieu et€al. (2010a) proposed an interacting approach in which a sub-regional watershed model makes use of a background of ‘large-scale-scenario’ constraints provided by global models and enhances them by integrating sub-regional dynamics. Applying this approach to downscale the Global-NEWS predictions of the Millennium Ecosystem Assessment scenarios to the Seine, Somme and Scheldt watersheds, they predicted an overall increase of nitrogen delivery to the Southern Bight of the North Sea at the 2050 horizon in the GO scenario while the predictions indicate a decrease by about 20% in the AM scenario. Several studies have looked at the impact of future land use and land management on nutrient losses. Bouwman et€al.

394

(2005a) predicted that from 1995 to 2030 in Western Europe grassland area will decrease from 60 Mha to 53 Mha and arable land will decline from 86 Mha to 76 Mha. On the other hand for transition countries (including new member states and the old states of the USSR) for the same period grassland will increase from 90 Mha to 96 Mha and arable land areas from 266 Mha to 273 Mha. At the same time, the nitrogen recovery has steadily increased for Western Europe from 44% in 1970 to a predicted 58% in 2030. A similar trend is foreseen for transition countries with a predicted change of nitrogen recovery from 38% in 1970 to 58% in 2030. This indicates an increase in productivity for both Western Europe and transition countries. Indeed for Western Europe, the level of fertiliser use in 2030 is predicted to be similar to that of 1995. For transition countries, there should be an increase of nitrogen fertiliser use by about 25%. Similar conclusions are reported by the EFMA in its 2008 outlook for 2018 (EFMA, 2008). EFMA (2008) expects that crop yield will increase for all major crops including wheat, barley, maize, potatoes and oilseed rape. It also predicts a shift in the agricultural production more oriented to grain and oil seed rape productions. This should lead to an increase of nitrogen consumption of 8% and 23%, for wheat and oil seed rape respectively. EFMA (2008) expects a decrease of nitrogen consumption in fodder and grassland production. Overall it is predicted that for 2018, the nitrogen consumption will increase by 3.8% in Europe, mostly in the new Member States. Indeed, there is a continuing decreasing trend of nitrogen consumption in EU15 with the exception of Austria, Denmark and Sweden due to an increase in the area devoted to the production of energy crops. On the other hand for EU12 there should be an increase of nitrogen consumption by about 17% compared to the actual level. The impact of the changing environment on nitrogen losses are reported in Bouwman et€al. (2005b, 2005c). Because of the predicted increase in nitrogen recovery, there should be no increase of total nitrogen emissions (sum of denitrification, leaching and volatilisation) in 2030 when compared to the actual levels. Land use and climate changes are inherently interconnected by nature, and are both affected by the economic development and the mitigation and adaptation strategies adopted by the human society. Therefore future changes will be the result of complex interactions and will depend as well on societal choices. An actual example in this sense is the controversial issue of biofuels, which are motivated by economic and energy policies and present implications for both land use and climate changes, with expected negative consequences for water availability and nitrogen losses (Howarth et€al., 2009).

17.4.3╇ Possible management scenarios Global scenarios address the effects on the environment of possible future economic development at global scale. At local scale, scenarios of implementation of practical measures to reduce nitrogen pollution may be evaluated to plan effective remediation strategies at river basin level. A number of modelling studies have been published to assess the effect of measures to reduce nitrogen loads to waters, such as improved wastewater

Bruna Grizzetti

treatment, or from adoption of agro-environmental measures affecting agricultural practices or landscape management. The severely eutrophied continental coast of the Channel and the Southern North Sea is particularly well documented from that respect. Cugier et€al. (2005) explored the effect of phosphorus and/ or nitrogen tertiary treatment of urban wastewater in the Seine watershed, assuming constant agricultural diffuse sources, on the conditions of algal growth in the Seine Bight. Phosphorus treatment of wastewater (reducing the loading by 90%) appeared as quite an effective measure to reduce the potentially harmful algal blooms in the Seine Bight, with a 10-fold reduction for maximum dinoflagellates biomass with respect to reference levels. To reach a result similar to that for phosphorus, the reduction of point sources nitrogen should be carried up to 70%, which is technically feasible but extremely expensive. A reduction by 90% of point source nitrogen inputs would be required to lead the trophic state of the Seine Bight back to levels of flagellate development comparable to those of preindustrial periods. More recently, Thieu et€ al. (2010b) generalised the same approach to the three main basins responsible for nutrient enrichment of the French and Belgian coastal zone of the North Sea (Seine, Somme and Scheldt, assuming that the Rhine plume is most of the time flowing northwards along the Netherlands coast). They showed that, although phosphorus abatement from the major point sources of wastewater is a useful measure to balance phosphorus fluxes with respect to silica inputs at the coastal zone, the generalisation of denitrification in wastewater treatment will not bring a substantial benefit. Implementation of ‘good agricultural practices’ (catch crop, reduction of fertilisation, extensification of cattle farming) would lead to a significant decrease of nitrogen fluxes exported to the sea, ranging from 14% to 23% in wet and dry years respectively. The results of these scenarios were coupled with a model of Phaeocystis development in the Southern North Sea (Lancelot et€ al., 2005, 2007). The combination of wastewater treatment improvement and good agricultural practices leads to a decrease of the Phaeocystis bloom duration from 25 to 20 days, and of the biomass peak from 35·106 to 30·106 cells/l. Previous works (Rousseau et€al., 2000) showed, however, that the good status of the coastal marine ecosystem requires that the biomass of Phaeocystis cells never increases above 5·106 cells/l, a threshold for the formation of ungrazable colonies (Lancelot et€al., 2009). Nitrogen contamination of surface and groundwater as well as of coastal sea water will thus remain a major problem even if ‘good agricultural practices’ were generalised. To further reduce nitrogen export, and more specifically to satisfy to the OSPAR (2005) recommendation of a 50% reduction of nitrogen input to the sea, more drastic changes in agricultural practices should be envisaged. Essentially the same conclusion was reached by other authors working in different context and at various scales (Western France Kervidy catchment:€ Durand, 2004; England and Wales:€ Johnes et€ al., 2007 and Johnes 2007; The Elbe river:€ Kersebaum et€ al., 2003; Danish Gjern river:€ Kronvang et€al., 1999b; Netherlands:€Wolf et€al., 2005).

17.5╇ Policies for managing threats to European water quality 17.5.1╇ Policy and regulatory context At the beginning of the 1990s, the European Commission enforced various regulations to control and reduce nutrient loads into surface water, groundwater and coastal and marine water. In 1991, with the Nitrates Directive (Directive 91/676/EEC) and the Urban Waste Water Treatment Directive (Directive 91/271/EEC) the Commission started an ambitious plan to reduce the nutrient diffuse pollution originated from agriculture and the nutrient point pollution generated by urban waste water discharges. Then, the Commission established regulations for industrial emissions in 1996 (Directive 96/61/EC) and updated and reinforced the protection of drinking water in 1998 (Directive 98/83/EC). A comprehensive regulation for European water was enforced in year 2000, through the Water Framework Directive (WFD, Directive 2000/60/ EC) and its daughter directives. The WFD aims at protecting all the waters, including inland and coastal surface waters and groundwater and to achieve a good ecological status by 2015. It combines emission limit values with environmental quality standards. The WFD complements and integrates the European water legislation on nutrient reduction, in particular the Nitrate Directive and the Urban Waste Water Directive, into a coherent prospective of river basin management (Bloch, 2001). The Groundwater Directive (Directive 2006/118/EC) completes and specifies the WFD concerning the protection of aquifers, and the Freshwater Fish Directive (Directive 2006/44/ EC) sets physical and chemical parameters for fresh waters that support fish life, including among others limits for nitrite, total ammonium and non-ionised ammonia, which can be toxic for fishes. Finally, in 2008 the Marine Strategy Directive (Directive 2008/56/EC) was enforced to protect the marine environment, aiming to achieve or maintain a good environmental status for the European seas by 2021. Other European policies directly or indirectly influence the nitrogen enrichment in water, especially those related to the driving forces for nitrogen emission in the environment, such as the Common Agricultural Policy and the National Emission Ceiling Directive (Directive 2001/81/EC). The Commission supports the implementation of the WFD and related policies through the Common Implementation Strategy, a joint work programme, which involves Member States, other countries, stakeholders and NGOs to promote dialogue, common understanding and best practice exchange. An example is the Pilot River Basin activity, where, among other issues, a network of European river basins is sharing knowledge and experience in developing and implementing measures to reduce nutrient pollution (Charlet, 2007; http:// prb-water-agri.jrc.ec.europa.eu/). Similarly, Member States are working together to establish common ways to evaluate the ecological status of water bodies through the Intercalibration process (European Commission, 2008). The latter compares national ecological assessments in order to ensure that ‘good ecological status’ means the same in all the European countries. Indeed, to ensure a similar level of ambition in setting

395

Nitrogen as a threat to European water quality

the environment objectives in the European Union, the WFD mandated the intercalibration of the results of the national assessment methods. The boundary values for good ecological status for the different water types are established by the intercalibration exercise and will be the basis for setting environmental objectives for the management and programme of measures for the European river basins. The first step of the intercalibration exercise included a common agreement of reference condition criteria, and of the acceptable departures from the reference conditions. It was followed by the application of these criteria to a benchmark dataset to establish ecological rules, or if not possible statistical ones, for setting good status boundaries. An example is the data displayed in Figure 17.12 for lakes. Such data were obtained from national sources and from on-going EU projects that compiled EU wide biological datasets. (Further information can be obtained from the intercalibration technical report for lakes, the Commission decision on intercalibration and the following web address:€ http://circa.europa.eu/Members/irc/ jrc/jrc_eewai/library.) When the common implementation strategy and the dialogue fail, the Commission can use its powers under the Treaty and take the non-compliant Member States to court. In addition to European legislation, policies and interventions have been planned to protect European seas through international conventions:€HELCOM in the Baltic Sea, OSPAR in the North Sea, the Barcelona convention MEDPOL in the Mediterranean Sea and the Bucharest convention in the Black Sea. For the river basins discharging in the North Sea, the OSPAR Convention established the reduction of inputs of nitrogen and phosphorus to areas affected or likely to be affected by eutrophication in the order of 50% compared to input levels in 1985, to be achieved by 1995 (PARCOM Recommendation 88/2 and 89/4). Similarly, in the Baltic Sea within the HELCOM convention Contracting Parties committed to reduce nitrogen and phosphorus loads to the sea according to country specific reduction targets set in the recent Baltic Sea Protection Plan. A comprehensive discussion of European Directives and international actions addressing the nitrogen pollution through all the environmental compartments is presented in Oenema et€ al., 2011 (Chapter 4 this volume).

17.5.2╇ Scientific and technical knowledge The research community has developed a sound understanding of sources and pathways causing water nitrogen enrichment and of the consequent processes of transformation, transport, storage and removal (Durand et€al., 2011; Voss et€al., 2011; Chapters 7 and 8, this volume). Effective tools that the scientific community can offer for understanding the impact of human disturbance and the potential success of restoration interventions include the construction of nutrient budgets at river basins scale (Billen et€al., 2011, Chapter 13, this volume) and the spatial evaluation of pressures and sources contribution to nitrogen export (Johnes, 1996; Johnes and Heathwaite, 1997; Behrendt et€al., 2003; Grizzetti et€al., 2008) together with development and the evaluation of future scenarios (Bouwman

396

et€ al., 2005c; Bouraoui and Aloe, 2007; Velthof et€ al., 2009). Moreover, the development of integrated indicators of nutrient pressure, eutrophication status and ecosystem functioning offers useful tools to monitor in space and in time nutrient trends and measures effectiveness (European Environment Agency, 2005). Extensive knowledge is available on interventions to alleviate nitrogen (and phosphorus) pollution for the water system. Measures can be targeted on the sources, on the landscape or directly on the water body management (Novotny and Olem, 1994). They include among others:€ optimal fertilisation, where fertiliser application are reduced to match the crop requirements; spatial nutrient management, which implies lower application in areas with high erosion and runoff, use of catch crops to reduce erosion and nutrient leaching; improvement in livestock and manure surplus management; implementation of advanced treatments for waste water discharges; optimisation of sewer systems; creation of riparian strips, sedimentation ponds, appropriate drainage systems; and finally restoration of wetlands and floodplains to increase denitrification and protect wildlife habitat and biodiversity. A European initiative on this field is the COST Action 869 (COST 2010a) on mitigation options for nutrient reduction in surface water and groundwater. It provides an extended review of all the measures implemented in European countries describing for each measure main benefits, reported efficacy, region of application, likely disadvantages and potential costs (COST 2010b). Economic tools, such as cost-benefit analysis (CBA), have been developed to support policy decisions on controversial environmental issues (OECD, 2006). CBA for environmental resolutions consists of calculating, commonly in currency units, the net benefits generated by the policy or project at each point in time, based on the analysis of all benefits and costs. Although widely employed, environmental CBA is controversial (Ludwig et€al., 2005), for the way the discount rates are set and for the methods to account and monetise environmental externalities in the economic models. In fact, the evaluation of environmental effects and ecosystem services is not always possible in monetary terms, and implies a vision of the system and the actors concerned, involving definitely a discussion of values. Similarly, establishing discount rates requires a judgment in a time perspective. In economic models, discount rates indicate the way costs and benefits are weighted over time. When present benefits are weighted higher than future ones, ecosystem services are consumed faster, while the contrary produces the conservation of the natural capital with eventual wealth loss for the present generation (Ludwig et€ al., 2005). Therefore, the choice of the appropriate discount rate is not straightforward, involves a system of values and is undermined by the high uncertainty related to long-term pollution effects on ecosystems. However, evidence provides support for policies that maintain ecosystem services over the long term (Ludwig et€al., 2005). Brink et€al., 2011 (Chapter 22 this volume) provide a CBA of nitrogen in the environment with a European perspective.

Bruna Grizzetti

17.5.3╇ Effects of implementation of measures To reduce eutrophication in estuaries and coastal waters programmes of measures implemented in their river basins will be essential to restrict anthropogenic nutrient inputs (Smith and Schindler, 2009). The WFD requires Member States to prepare river basins management plans including the analysis of nutrient pressures and the plan for implementing mitigation measures, in order to achieve by 2015 a ‘good ecological status’ of all the water bodies. The latter should be achieved considering the reference status, which refer to water body conditions prior to significant anthropogenic pollution. However, the concept of good ecological status implies some room for interpretation. In fact, even consistent nutrient abatement may not lead the water bodies to the desired status as many ecosystems present a hysterisis behaviour. Duarte et€al. (2009) illustrated the trajectories of restoration in four North European coastal ecosystems (Marsdiep, Netherlands; Helgoland, Germany; Odense Fiord, Denmark and Gulf of Riga, Latvia/Estonia). They argued that in addition to nutrient enrichment other human induced changes, such as climate changes, population growth, freshwater withdrawal, may affect many fundamental factors of ecosystem functioning, producing baseline conditions different from those of the ‘reference’ status, even returning to pristine nutrient inputs. They observed that the restoration pathways of the monitored ecosystems followed dynamic trajectories of ‘regime shift and shifting baseline’. As a consequence, when setting targets of restoration, emphasis should be put on the values ensuring reliable provision of ecosystem services and good ecosystem functioning rather than focusing on particular past conditions (Duarte et€ al., 2009). This does not mean that anthropogenic nutrient input should not be reduced; on the contrary these observations show a clear awareness on how human impacts can be difficult to reverse. In Europe only 30% of surface water bodies have been identified as not being a risk of failing to achieve the WFD environmental objectives by 2015, while 40% are at risk and for the rest 30% data are not sufficient for evaluation. The lack of information regards especially coastal and transitional water (European Commission 2007 SEC(2007)362). Similar figures are reported for groundwater bodies, with 25% not at risk, 30% at risk and for the remaining 45% the evaluation is not possible for the lack of data (European Commission 2007 SEC(2007)362). Nutrient diffuse pollution has been identified as one of the most significant and widespread pressures on water ecosystems in Europe (European Commission 2007 COM(2007)128). In general during the past two decades the nutrient pollution from urban areas and point sources has been significantly reduced through the implementation of waste water treatment plants, while diffuse sources originate from agriculture remain a problem (European Environment Agency, 2005). According to the monitoring information transmitted by Member States, after almost 15 years from the enforcement of Nitrates Directive, in EU27 monitoring stations with average annual nitrate concentrations above 50 mgNO3/l were 15% for groundwater and 3% for surface water. Member States with the highest proportion of sampling points with nitrate

concentrations above 50€ mg/l were Estonia, the Netherland, Belgium, England, France, Northern Italy, North-East of Spain, Slovakia, Romania, Malta and Cyprus (European Commission 2010 COM(2010)47). In 2003, the level of compliance with the Urban Waste Water Treatment Directive in the EU15 was 79% in normal areas and 84% in sensitive areas (European Commission 2007 SEC(2007)363). There were 17 ‘big cities’ still without wastewater treatment and some countries, mainly in Southern Europe, presented areas with inadequate or lacking wastewater treatment (European Commission 2007 SEC(2007)363). Measures to reduce nitrogen and phosphorus pollution in water bodies have already been introduced in many European countries under the European legislation, international conventions and national plans. The assessment report on the implementation of the Baltic Sea Action Plan, under the HELCOM Convention, indicates that since 1990 nitrogen and phosphorus diffuse and point source loads have been slightly decreasing in the Baltic Sea catchment, however the target input levels foreseen in the Action Plan have not been met and additional reductions are needed (HELCOM, 2009). In the areas under the OSPAR Convention, the source reduction of 50% (compared to the level of 1985) has been met for phosphorus, but not completely for nitrogen. In fact, the target for nitrogen source reduction was achieved only by Denmark (in 2003), Germany and the Netherlands (both in 2005), although progress in this direction has been made also by the other Contracting Parties (OSPAR, 2008). See also Voss et€al., 2011 (Chapter 8 this volume). The implementation of programmes of measures for nutrient reduction varies in the European countries. Moreover, the water quality response to the mitigation programmes has been variable, depending on the design of the plans, the specific pollution conditions and the environmental characteristics. In general a delay has been observed between remediation actions and water response. In Denmark, the implementation of targeted regulations and nitrogen efficiency measures has reduced nitrogen loads to waters by 32%, while maintaining the crop yield and increasing the livestock production (Kronvang et€al., 2008). In Norway, results from long-term monitoring show that in spite of changes in management practices driven by subsidies and production conditions few correspondent trends were registered in nutrient losses (Bechmann et€al., 2008). In part of England, the effect of actual measures introduced in the Nitrogen Vulnerable Zones is not evident and a time lag is expected because of the specific soils and aquifers characteristics (Jackson et€al., 2008; Worrall et€al., 2009). In Finland, no clear reduction of nutrient loads or water quality improvements were observed although a large-scale programme to reduce nutrient emissions from agriculture has been introduced since 1995 (Ekholm et€al., 2007). In general, scientific evidence shows that the adopted policies to reduce anthropogenic nutrient inputs to European seas were more effective in abating point sources than diffuse sources and more successful for phosphorus rather than for nitrogen, leading to the increase of the N:P ratio in anthropogenic inputs (Artioli et€ al., 2008). Indeed, assessing policy effectiveness

397

Nitrogen as a threat to European water quality

in reducing loads of nitrogen is controversial and presents regional differences. This is related to the diffuse nature of the sources, the tight connections with lifestyles, notably human diet, and the economic implications due to the links with agriculture and livestock production. Moreover, long-retention times of groundwater may retard the system recovery (Artioli et€al., 2008). In addition, the problem of ‘pollution swapping’ needs to be taken into wider consideration from both the scientific and the policy prospective (Stevens and Quinton, 2009). The term ‘pollution swapping’ is used to indicate when a mitigation measure reduces a targeted pollutant while increasing the level of another pollutant. Unfortunately, this side effect is present in many commonly implemented measures, but weakly considered in scientific studies and management plans. For example, constructed wetlands and riparian buffer zones are widely used to remove nitrogen but at the same time they promote denitrification, increasing the emissions of N2O, a strong greenhouse gas, and thus swapping the pollution from water to the air compartment. If not correctly understood the mechanisms of pollution swapping may lead to contradicting interventions and unsuccessful regulations (Stevens and Quinton, 2009). An enlightening example in this sense is the model MITERRAEUROPE (Velthof et€ al., 2009), which provides an integrated assessment of the effect of nutrient mitigation measures in EU27 at country and regional level, considering the effects of mitigation measures contemporary on nitrogen emissions to atmosphere and to groundwater.

17.5.4╇ Opportunities for policy and science integration In developing mitigation options for nitrogen pollution in water, the way forward is to adopt a more holistic approach in the policy frame, undertaking an integration effort in legislation tools and an evaluation of the potential ‘ecosystem service swapping’, and in the scientific research, promoting interdisciplinary studies including more forms of pollutants and all the environmental compartments (Stevens and Quinton, 2009; Collins and McGonigle, 2008). Effective strategies to reduce eutrophication in aquatic ecosystems need to consider the whole land-ocean continuum and to control both nitrogen and phosphorus (Conley et€al., 2009). A stronger integration between the Marine Strategy and relÂ� evant sectoral policies, such as the Common Agricultural Policy and the Fisheries Policy, would be beneficial for the sustainable management and protection of the marine ecosystem and its resources (Salomon, 2009). Similar positive synergies could derive from an improved coordination between policies to prevent water nutrient pollution and the Common Agricultural Policy and among the different nitrogen related regulations, in order to avoid problems of pollution swapping (Oenema et€al., 2011, Chapter 4 this volume). The Commission already supports the integration between the implementation of different policies, encouraging Member States to use EU financing instruments available under the Common Agricultural Policy and the Cohesion Policy for improvements in the water field

398

(European Commission 2007 COM(2007) 128), or the further integration of climate change mitigation and adaptation strategies into the implementation of EU water policy (see European Climate Change Programme). Finally, a strategic choice for the future will be to create positive synergies between River Basin Plans foreseen by the European WFD by December 2009 and international instruments, such as the Baltic Sea Action Plan (HELCOM Convention) or the OSPAR Resolutions (OSPAR Convention). An effort to improve integration and interdisciplinarity is required also in the scientific research. Indeed, there is a need for integrated assessments to enable comparison between different regions for promoting regional assessment, and to evaluate the potential synergies between different type of policies or remediation measures, taking into account future scenarios. There is a clear need to promote science–policy–society interactions to build reciprocal trust and understanding, and to produce a new type of transdisciplinarity knowledge to support a sustainable management of the ecosystem resources. In fact, the complexity of interactions of the human–ecological system cannot be addressed by traditional disciplines, but requires a new kind of interdisciplinary science, concerned by problem-solving aspects (Carpenter et€al., 2009). This new type of knowledge production, also referred to as sustainability science, seeks to understand the interactions between nature and society, addressing the dynamics of the interactions, the longterm trends, the vulnerability and resilience of the nature–Â� society system and the opportunities for adaptive management and societal learning (Kates et€al., 2001). To address problems related to sustainable development, the integration between natural science and social science is a key point and cooperation is necessary between the academic disciplines and the different parts of society (Tappeiner et€al., 2007).

17.6╇ Conclusions and way forward From the present overview it appears that the anthropogenic increase of nitrogen in water, together with other nutrients, causes many direct and indirect biogeochemical and ecolog� ical responses in the aquatic ecosystems, most of which are un�desired and detrimental for the human-ecological system. In spite of some encouraging trends, nitrogen concentrations in rivers, lakes, aquifers and coastal waters are generally high and stable in many regions, and even increasing in some areas. In addition, evidence shows that there is gradual and increasing nitrogen enrichment of groundwater resources across Europe. This poses direct threats to human health and ecosystem functioning, reducing the actual provision and the future reliability of ecosystem services. A large part of European freshwaters and coastal waters are affected by eutrophication and current global drivers such as climate and land-use changes could exacerbate the situation in the near future. An additional challenge will be represented by the economic development of Eastern Europe, which could potentially lead to additional nitrogen loadings to the Baltic and Black Seas.

Bruna Grizzetti

Policy tools are available within the European Union and under international conventions to mitigate the nitrogen pollution in water. Their full implementation has not been achieved yet throughout Europe, but plans of measures to reduce nitrogen losses to water have already been implemented in many European countries, producing some encouraging results. However, in many cases a delay in the water quality response to the implementation of measures have been observed, due to previous accumulation of anthropogenic nitrogen in soils, sediments or aquifers or to inadequate design or targeting of the mitigation plans. At European level some regional differences have emerged in the sensitivity of coastal ecosystems to nutrient loads and on the effect of policy measures in changing the N:P ratio. Finally, the issue of pollution swapping has appeared as an important element to be considered by both the scientific and policy perspective. To support the sustainable management of the human-ecological system and promote the protection of water resources in relation to the threats posed by nitrogen in European water the full implementation of the regulations is necessary, combined with an efficient environmental monitoring. Moreover, positive synergies could be obtained by encouraging integration in the sectoral policies and enhancing interdisciplinarity in the scientific research, especially in support of regional assessments and pollution swapping evaluations. The continuous nitrogen export to waters directly or indirectly threatens the biodiversity in the aquatic ecosystems, and slowly and gradually erodes the resilience of the aquatic ecosystems, increasing their vulnerability to other unexpected stresses. Water eutrophication and aquatic biodiversity loss have economic and political implications. According to Folke et€al. (2002) two errors underpinned the past policies for managing natural resources. The first was the assumption that the ecosystem response to anthropogenic pressures is linear and the second was the lack of recognition of the mutual interdependence of the human and ecological system. Drastic changes in the ecosystem status may imply elevated costs for€the direct loss of ecological and economic recourses and for the actions required for restoration. Building and maintaining the ecosystem resilience result in a wise investment for future human wealth, especially in the context of uncertainty and global environmental changes. A policy aiming at good ecological status can certainly contribute in this direction by investing in substantially improving nitrogen use efficiency and cleaning waste waters, in spite of the recovery time and costs.

Acknowledgements This chapter was prepared with the support of the NinE Programme of the European Science Foundation, the NitroEurope Integrated Project (funded by the European Commission) and the COST Action 729.

References Addiscott, T. M. (2005). Nitrate, Agriculture and Environment, CAB International, Wallingford, UK. Alcamo, J., Van Vuuren, D. and Cramer W. (2006). Changes in ecosystem services and their drivers across the scenarios.

In:€Ecosystems and Human Well-Being:€Scenarios, ed. S. R. Carpenter et€al., pp. 279–354, Island Press, Washington, DC. Alcamo, J., Moreno, J. M., Nováky, B. et€al. (2007). Europe, Climate Change 2007:€Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK, pp. 541–580. Alexander, R. B., Smith, R. A. and Schwarz, G. E. (2000). Effect of stream channel size on the delivery of nitrogen to the Gulf of Mexico. Nature, 403, 758–761. Alley, W. M., Healy, R. W., LaBaugh, J. W. and Reilly, T. E. (2002). Flow and storage in groundwater systems. Science, 296, 1985–1990. Anderson, D. M., Glibert, P. M. and Burkholder, J. M. (2002). Harmful algal blooms and eutrophication:€Nutrient sources, composition, and consequences. Estuaries, 25, 704–726. Arheimer, B., Andreasson, J., Fogelberg, S. et€al. (2005). Climate change impact on water quality:€model results from southern Sweden. Ambio, 34(7), 559–566. Artioli, Y., Friedrich, J., Gilbert, A. J. et€al. (2008). Nutrient budgets for European seas:€a measure of the effectiveness of nutrient reduction policies. Marine Pollution Bulletin, 56, 1609–1617. Battarbee, R., Kernan, M., Livingstone, D. et€al. (2008). Euro-limpacs Position Paper: Impact of climate change on European freshwater ecosystems:€consequences, adaptation and policy. Bechmann, M., Deelstra, J., Stålnacke, P. et€al. (2008). Monitoring catchment scale agricultural pollution in Norway:€policy instruments, implementation of mitigation methods and trends in nutrient and sediment losses. Environmental Science and Policy, 11, 102–114. Behrendt, H., Bach, M., Kunkel, R. et€al. (2003). Nutrient emissions into river basins of Germany on the basis of a harmonised procedure. Federal Environmental Agency (Umweltbundesamt). Billen, G. and Garnier, J. (2007). River basin nutrient delivery to the coastal sea:€assessing its potential to sustain new production of nonsiliceous algae. Marine Chemistry, 106, 148–160. Billen, G., Silvestre, M., Grizzetti, B. et€al. (2011). Nitrogen flows from European watersheds to coastal marine waters. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Bloch, H. (2001). EU policy on nutrients emissions:€legislation and implementation, Water Science and Technology, 44, 1–6. Bouraoui, F., Galbiati, L. and Bidoglio, G. (2002). Climate change impacts on nutrient loads in the Yorkshire Ouse catchment (UK). Hydrology and Earth System Sciences, 6, 197–209. Bouraoui, F., Grizzetti, B., Granlund, K., Rekolainen, S. and Bidoglio,€G. (2004). Impact of climate change on the water cycle and nutrient losses in a Finnish catchment. Climatic Change, 66, 109–126. Bouraoui, F. and Aloe, A. (2007). European agrochemicals geospatial loss estimator:€model development and application. EUR Report 22690 EN, European Commission Joint Research Centre, Luxembourg. Bouraoui, F., Grizzetti, B. and Aloe, A. (2009). Nutrient discharge from river to seas for year 2000, EUR Report 24002 EN, European Commission Joint Research Centre, Luxembourg. Bouwman, A. F., van Drecht, G. and van Der Hoek, K. W. (2005a). Global and regional surface nitrogen balances in intensive agricultural production systems for the period 1970–2030. Pedosphere, 15, 137–155. Bouwman, A. F., van Drecht, G. and van der Hoek, K. W. (2005b). Surface N balances and reactive N loss to the environment from

399

Nitrogen as a threat to European water quality global intensive agricultural production systems for the period 1970–2030. Science in China, Series C, 48, 767–779. Bouwman, A. F., van Drecht, G., Knoop, J. M., Beusen, A. H. W. and Meinardi, C. R. (2005c). Exploring changes in river nitrogen export to the world’s oceans. Global Biogeochemical Cycles, 19, 1–14. Brink, C., van Grinsven, H., Jacobsen, B. H. et€al. (2011). Costs and benefits of nitrogen in the environment. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Burkholder, J. M., Tomasko, D. A. and Touchette, B. W. (2007). Seagrasses and eutrophication. Journal of Experimental Marine Biology and Ecology, 350, 46–72. Caddy, J. F. (1993). Toward a comparative evaluation of human impacts on fishery ecosystems of enclosed and semi-enclosed seas. Reviews in Fisheries Science, 1, 57–95. Camargo, J. A. and Alonso, A. (2006). Ecological and toxicological effects of inorganic nitrogen pollution in aquatic ecosystems:€a global assessment. Environment International, 32, 831–849. Cardoso, A. C., Duchemin, J., Magoarou, P. and Premazzi, G. (2001). Criteria for the identification of freshwaters subject to eutrophication: their use for the implementation of the Nitrates€and Urban Waste Water treatment Directives. EUR Report 19810 EN, European Commission Joint Research Centre, Luxembourg. Carpenter, S. R. and Folke, C. (2006). Ecology for transformation. Trends in Ecology and Evolution, 21, 309–315. Carpenter, S. R., Caraco, N. F., Correll, D. L. et€al. (1998). Nonpoint pollution of surface waters with phosphorus and nitrogen. Ecological Applications, 8, 559–568. Carpenter, S. R., Mooney, H. A., Agard, J. et€al. (2009). Science for managing ecosystem services:€beyond the Millennium Ecosystem Assessment. Proceedings of the National Academy of Sciences of the United States of America, 106, 1305–1312. Charlet, M. (2007). Experiences in the analysis of pressures and impacts from agriculture on water resources and developing a related programme of measures, European Commission Joint Research Centre, Luxembourg. Christoffersen, K., Andersen, N., Søndergaard, M., Liboriussen, L. and Jeppesen, E. (2006). Implications of climate-enforced temperature increases on freshwater pico- and nanoplankton populations studied in artificial ponds during 16 months. Hydrobiologia, 560, 259–266. Cloern, J. E. (2001). Our evolving conceptual model of the coastal eutrophication problem. Marine Ecology Progress Series, 210, 223–253. Collins, A. L. and McGonigle, D. F. (2008). Monitoring and modelling diffuse pollution from agriculture for policy support:€UK and European experience. Environmental Science and Policy, 11, 97–101. Conley, D. J., Paerl, H. W., Howarth, R. W. et€al. (2009). Ecology: controlling eutrophication –€nitrogen and phosphorus. Science, 323, 1014–1015. COST (2010a). http://www.cost869.alterra.nl COST (2010b). http://www.cost869.alterra.nl/Fs/List_of_options. htm#NM Cugier, P., Billen, G., Guillaud, J. F., Garnier, J. and Ménesguen,€A. (2005). Modelling eutrophication of the Seine Bight under present, historical and future Seine river nutrient loads. Journal of Hydrology, 304, 381–396. Curtis, C. J., Evans, C., Helliwell, R. C. and Monteith, D. (2005a). Nitrate leaching as a confounding factor in chemical recovery from acidification in UK upland waters. Environmental Pollution, 137, 73–82.

400

Curtis, C. J., Botev, I., Camarero, L. et€al. (2005b). Acidification in European mountain lake districts:€a regional assessment of critical load exceedance. Aquatic Sciences, 67, 237–251. De Roos, A. J., Ward, M. H., Lynch, C. F. and Cantor, K. P. (2003). Nitrate in public water supplies and the risk of colon and rectum cancers. Epidemiology, 14, 640–649. Diaz, R. J. and Rosenberg, R. (2008). Spreading dead zones and consequences for marine ecosystems. Science, 321, 926–929. Directive 91/271/EEC (1991) concerning urban waste water treatment. (OJ (1991) L271/40). (Urban Waste Water Directive). Directive 91/676/EEC (1991) concerning the protection of waters against pollution caused by nitrates from agricultural sources (OJ (1991) L375/1). (Nitrates Directive). Directive 96/61/EC (1996) concerning integrated pollution prevention and control (IPPC Directive). Directive 98/83/EC (1998) on the quality of water intended for human consumption. (Drinking Water Directive). Directive 2000/60/EC (2000) establishing a Framework for Community Action in the Field of Water Policy. (OJ (2000) L327/1). (Water Framework Directive). Directive 2001/81/EC (2001) on national emission ceilings for certain atmospheric pollutants. Directive 2006/44/EC (2006) on the quality of fresh waters needing protection or improvement in order to support fish life (OJ (2006) L264/20). Directive 2006/118/EC (2006) on the protection of groundwater against pollution and deterioration (OJ L 372, 27.12.2006, p.19). Directive 2008/56/EC (2008) establishing a framework for community action in the field of marine environmental policy (Marine Strategy Framework Directive) (OJ L 164, 25.6.2008, pp. 19–40). Dodds, W. K., Bouska, W. W., Eitzmann, J. L. et€al. (2009). Eutrophication of U. S. freshwaters:€Analysis of potential economic damages. Environmental Science and Technology, 43, 12–19. Duarte, C. M., Conley, D. J., Carstensen, J. and SanchezCamacho,€M. (2009). Return to Neverland:€shifting baselines affect eutrophication restoration targets. Estuaries and Coasts, 32, 29–36. Ducharne, A., Baubion, C., Beaudoin, N. et€al. (2007). Long term prospective of the Seine River system:€confronting climatic and direct anthropogenic changes. Science of the Total Environment, 375(1–3), 292–311. Durand, P. (2004). Simulating nitrogen budgets in complex farming systems using INCA:€calibration and scenario analyses for the Kervidy catchment (W. France). Hydrology and Earth System Sciences, 8, 793–802. Durand, P., Breuer, L., Johnes, P. et€al. (2011). Nitrogen processes in aquatic ecosystems. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. EFMA (2008). Forecast of food, farming and fertiliser use in the European Union 2007–2017. Eisenreich, S. J. (ed.) (2005). Climate Change and the European Water Dimension. European Commission, Joint Research Centre: Ispra, Italy. Ekholm, P., Granlund, K., Kauppila, P. et€al. (2007). Influence of EU policy on agricultural nutrient losses and the state of receiving surface waters in Finland. Agricultural and Food Science, 16, 282–300. EMEP (2009). Transboundary acidification, eutrophication and ground level ozone in Europe in 2007, EMEP Status Report 2009, Norwegian Meteorological Institute, Oslo, Norway.

Bruna Grizzetti Erisman, J.W., van Grinsven, H., Grizzetti, B., et€al. (2011). The European nitrogen problem in a global perspective. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J.€W. Erisman et€al., Cambridge University Press. European Commission (2007). The quality of drinking water in the European Union, Synthesis Report on the Quality of Drinking Water in the European Union period 2002–2004. (Directives 80/778/EEC and 98/83/EC). European Commission (2007) COM(2007) 120. Report from the Commission to the Council and the European Parliament on the implementation of the Council Directive 91/676/EEC concerning the protection of the waters against pollution caused by nitrates from agricultural sources for the period 2000–2003. European Commission (2007) COM(2007) 128. Communication from the Commission to the European Parliament and the Council ‘Towards Sustainable Water Management in the European Union’ First stage in the implementation of the Water Framework Directive 2000/60/EC. European Commission (2007) SEC(2007) 362. Accompanying document to the communication from the Commission to the European Parliament and the Council ‘Towards Sustainable Water Management in the European Union’ First stage in the implementation of the Water Framework Directive 2000/60/EC. European Commission (2007) SEC(2007) 363. Annex to the Communication from the Commission to the European Parliament and the Council ‘Towards Sustainable Water Management in the European Union’ First stage in the implementation of the Water Framework Directive 2000/60/ EC. 4th Commission Report (Executive Summary) on the Implementation of the Urban Waste Water Treatment Directive. European Commission (2008). Commission Decision (2008/915/EC) of 30 October 2008 establishing, pursuant to Directive 2000/60/ EC of the European Parliament and of the Council, the values of the Member State monitoring system classifications as a result of the intercalibration exercise (notified under document number C(2008) 6016). OJ. L 332/20, 10/12/2008. European Commission (2010) COM(2010) 47. Report from the Commission to the Council and the European Parliament on the implementation of the Council Directive 91/676/EEC concerning the protection of the waters against pollution caused by nitrates from agricultural sources based on Member States reports for the period 2004–2007. European Environment Agency (2001). Eutrophication in Europe’s coastal waters, EEA, Copenhagen. European Environment Agency (2005). The European environment: state and outlook 2005, EEA, Copenhagen. European Environment Agency (2008). Impacts of Europe’s changing climate, 2008 indicator-based assessment: Joint EEAJRC-WHO report. EEA Report 4/2008, EEA, Copenhagen. European Environment Agency (2009). Progress towards the European 2010 biodiversity target. EEA Report 4/2009, EEA, Copenhagen. Evans, C. D., Cullen, J. M., Alewell, C. et€al. (2001). Recovery from acidification in European surface waters. Hydrology and Earth System Sciences, 5, 283–297. Feuchtmayr, H., McKee, D., Harvey, I. F., Atkinson, D. and Moss,€B. (2007). Response of macroinvertebrates to warming, nutrient addition and predation in large-scale mesocosm tanks. Hydrobiologia, 584, 425–432. Flanagan, K. M., McCauley, E., Wrona, F. and Prowse, T. (2003). Climate change:€the potential for latitudinal effects on algal

biomass in aquatic ecosystems. Canadian Journal of Fisheries and Aquatic Sciences, 60, 635–639. Folke, C., Carpenter, S., Elmqvist, T. et€al. (2002). Resilience and sustainable development:€building adaptive capacity in a world of transformations. Ambio, 31, 437–440. Folke, C., Carpenter, S., Walker, B. et€al. (2004). Regime shifts, resilience, and biodiversity in ecosystem management. Annual Review of Ecology, Evolution, and Systematics, 35, 557–581. Galloway, J. N., Townsend, A. R., Erisman, J. W. et€al. (2008). Transformation of the nitrogen cycle:€recent trends, questions, and potential solutions. Science, 320, 889–892. Glibert, P. M., Harrison, J., Heil, C. and Seitzinger, S. (2006). Escalating worldwide use of urea: a global change contributing to coastal eutrophication. Biogeochemistry, 77, 441–463. Griesenbeck, J. S., Steck, M. D., Huber, J. C. et€al. (2009). Development of estimates of dietary nitrates, nitrites, and nitrosamines for use with the short Willet food frequency questionnaire. Nutrition Journal, 8, 16. Grimvall, A., Stalnacke, P. and Tonderski, A. (2000). Time scales of nutrient losses from land to sea: a European perspective. Ecological Engineering, 14, 363–371. Grizzetti, B., Bouraoui, F. and De Marsily, G. (2008). Assessing nitrogen pressures on European surface water. Global Biogeochemical Cycles, 22, GB4023. Gulis, G., Czompolyova, M. and Cerhan, J. R. (2002). An ecologic study of nitrate in municipal drinking water and cancer incidence in Trnava District, Slovakia. Environmental Research, A, 88, 182–187. Hartikainen, H., Johnes, P. J., Monteith, C. and Okamura, B. (2009). Bryozoan populations reflect nutrient enrichment and productivity gradients in rivers. Freshwater Biology, pp. 1–15. Hejzlar, J., Anthony, S., Arheimer, B. et€al. (2009). Nitrogen and phosphorus retention in surface waters:€an inter-comparison of predictions by catchment models of different complexity. Journal of Environmental Monitoring, 11, 584–593. HELCOM (2009). Euthrophication in the Baltic Sea: an integrated thematic assessment of the effects of nutrient enrichment and euthrophication in the Baltic Sea region: Executive Summary, Helshinki Commission, Helsinki. Hettelingh, J.-P., Posch, M., Slootweg, J. et€al. (2007). Critical loads and dynamic modelling to assess European areas at risk of acidification and eutrophication. Water, Air and Soil Pollution, 7, 379–384. Hettelingh, J.-P., Posch, M. and Slootweg, J. (2008). Critical load, dynamic modelling and impact assessment in Europe:€CCE Status Report 2008. PBL Report 500090003, www.mnp.nl/cce, Coordination Centre for Effects, Bilthoven, The Netherlands. Howarth, R. W. (1988). Nutrient limitation of net primary production in marine ecosystems. Annual Review of Ecology and Systematics, 19, 89–110. Howarth, R. W., Billen, G., Swaney, D. et€al. (1996). Regional nitrogen budgets and riverine N and P fluxes for the drainages to the North Atlantic Ocean:€natural and human influences. Biogeochemistry, 35, 75–139. Howarth, R. W., Bringezu, S., Bekunda, M. et€al. (2009). Rapid assessment on biofuels and the environment:€overview and key findings. In:€R. W. Howarth and S. Bringezu (eds.), Biofuels:€Environmental Consequences and Interactions with Changing Land Use, Proceedings of the Scientific Committee on Problems of the Environment (SCOPE) International Biofuels Project Rapid Assessment, 22–25 September 2008, Gummersbach

401

Nitrogen as a threat to European water quality Germany, pp. 1–13, Cornell University, Ithaca, NY, (http://cip. cornell.edu/biofuels/) IARC (2006). Ingested nitrates and nitrites. http://monographs.iarc.fr/ ENG/Meetings/94-nitratenitrite.pdf Jackson, B. M., Browne, C. A., Butler, A. P. et€al. (2008). Nitrate transport in chalk catchments:€monitoring, modelling and policy implications. Environmental Science and Policy, 11, 125–135. Johnes, P. J. (1996). Evaluation and management of the impact of land use change on the nitrogen and phosphorus load delivered to surface waters:€the export coefficient modelling approach. Journal of Hydrology, 183, 323–349. Johnes, P. J. and Heathwaite, A. L. (1997). Modelling the impact of land use change on water quality in agricultural catchments. Hydrological Processes, 11, 269–286. Johnes, P. J. (2007). Meeting ecological restoration targets in European Waters:€a challenge for animal agriculture. In:€Redesigning Animal Agriculture:€The Challenge for the 21st Century, pp. 185–203, CAB International, Wallingford, UK. Johnes, P. J., Foy, R., Butterfield, D. and Haygarth, P. M. (2007). Land use for Good Ecological Status:€an evaluation of scenarios for water bodies in England and Wales. Soil Use and Management, 23, 176–194. Johnk, K. D., Huisman, J., Sharples, J. et€al. (2008). Summer heatwaves promote blooms of harmful cyanobacteria. Global Change Biology, 14, 495–512. Johnson, P. T. J., Chase, J. M., Dosch, K. L. et€al. (2007). Aquatic eutrophication promotes pathogenic infection in amphibians. Proceedings of the National Academy of Sciences of the United States of America, 104, 15781–15786. Kates, R. W., Clark, W. C., Corell, R. et€al. (2001). Environment and development:€Sustainability science. Science, 292, 641–642. Kersebaum, K. C., Steidl, J., Bauer, O. and Piorr, H. P. (2003). Modelling scenarios to assess the effects of different agricultural management and land use options to reduce diffuse nitrogen pollution into the river Elbe. Physics and Chemistry of the Earth, 28,€537–545. Kronvang, B., Hoffmann, C. C., Svendsen, L. M. et€al. (1999a). Retention of nutrients in river basins. Aquatic Ecology, 33, 29–40. Kronvang, B., Svendsen, L. M., Jensen, J. P. and Dørge, J. (1999b). Scenario analysis of nutrient management at the river basin scale. Hydrobiologia, 410, 207–212. Kronvang, B., Andersen, H. E., Børgesen, C. et€al. (2008). Effects of policy measures implemented in Denmark on nitrogen pollution of the aquatic environment. Environmental Science and Policy, 11, 144–152. L’ hirondel, J. and L ’ hirondel, J. L. (2002). Nitrate and Man:€Toxic, Harmless or Beneficial?, CAB International, Wallingford, UK. Lancelot, C., Spitz, Y., Gypens, N. et€al. (2005). Modelling diatom and Phaeocystis blooms and nutrient cycles in the Southern Bight of the North Sea:€the MIRO model. Marine Ecology Progress Series, 289, 63–78. Lancelot, C., Gypens, N., Billen, G., Garnier, J. and Roubeix, V. (2007). Testing an integrated river-ocean mathematical tool for linking marine eutrophication to land use:€the Phaeocystis-dominated Belgian coastal zone (Southern North Sea) over the past 50 years. Journal of Marine Systems, 64, 216–228. Lancelot, C., Rousseau, V. and Gypens, N. (2009). Ecologically based indicators for Phaeocystis disturbance in eutrophied Belgian

402

coastal waters (Southern North Sea) based on field observations and ecological modelling. Journal of Sea Research, 61, 44–49. Lavelle, P., Dugdale, R., Scholes, R. et€al. (2005). Nutrient Cycling, Ecosystems and Human Well-Being:€Current State and Trends, Washington, DC. Millennium Ecosystem Assessment. Island Press, 23. Lotze, H. K., Lenihan, H. S., Bourque, B. J. et€al. (2006). Depletion degradation, and recovery potential of estuaries and coastal seas. Science, 312, 1806–1809. Ludwig, D., Brock, W. A. and Carpenter, S. R. (2005). Uncertainty in discount models and environmental accounting. Ecology and Society, 10(2). McKee, D., Atkinson, D., Collings, S. E. et€al. (2003). Response of freshwater microcosm communities to nutrients, fish, and elevated temperature during winter and summer. Limnology and Oceanography, 48, 707–722. McKenzie, V. J. and Townsend, A. R. (2007). Parasitic and infectious disease responses to changing global nutrient cycles. EcoHealth, 4, 384–396. Millennium Ecosystem Assessment (2005). Ecosystems and Human Well-Being:€Synthesis, Island Press, Washington, DC. Mimikou, M. A., Baltas, E., Varanou, E. and Pantazis, K. (2000). Regional impacts of climate change on water resources quantity€and quality indicators. Journal of Hydrology, 234, 95–109. Mooij, W. M., Hülsmann, S., De Senerpont Domis, L. N. et€al. (2005). The impact of climate change on lakes in the Netherlands:€a review. Aquatic Ecology, 39, 381–400. Moss, B., McKee, D., Atkinson, D. et€al. (2003). How important is climate? Effects of warming, nutrient addition and fish on phytoplankton in shallow lake microcosms. Journal of Applied Ecology, 40, 782–792. Mulholland, P. J., Helton, A. M., Poole, G. C. et€al. (2008). Stream denitrification across biomes and its response to anthropogenic nitrate loading. Nature, 452, 202–205. Murdoch, P. S., Baron, J. S. and Miller, T. L. (2000). Potential effects of climate change on surface-water quality in North America. Journal of the American Water Resources Association, 36, 347–366. National Research Council (2000). Clean Coastal Waters:€Understanding and Reducing the Consequences of Nutrient Pollution, National Academy Press, Washington, DC. Nixon, S. W. (1988). Physical energy inputs and the comparative ecology of lake and marine ecosystems. Limnology and Oceanography, 33, 1005–1025. Novotny, V. and Olem, H. (1994). Water Quality: Prevention, Identification, and Management of Diffuse Pollution, Van Nostrand Reinhold, New York. Oczkowski, A. and Nixon, S. (2008). Increasing nutrient concentrations and the rise and fall coastal fishery; a review of data from the Nile Delta, Egypt. Estuarine, Coastal, and Shelf Science, 77, 309–319. OECD (1982). Eutrophication of Water, Monitoring, Assessment and Control, Organisation for Economic Co-operation and Development, Paris. OECD (2006). Cost-Benefit Analysis and the Environment, Organisation for Economic Co-operation and Development, Paris. OECD (2008). OECD Environmental Data COMPENDIUM 2006– 2008, Organisation for Economic Co-operation and Development, Paris.

Bruna Grizzetti Oenema, O., Bleeker, A., Braathen, N. A. et€al. (2011). Nitrogen in current European policies. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. OSPAR (2005). Common Procedure for the Identification of the Eutrophication Status of the OSPAR Maritime Area, OSPAR, London. OSPAR (2008). Nutrients in the Convention Area: Assessment of implementation of PARCOM Recommendations 88/2 and 89/4, OSPAR, London. Paerl, H. W. and Huisman, J. (2008). Climate:€blooms like it hot. Science, 320, 57–58. Palmeri, L., Bendoricchio, G. and Artioli, Y. (2005). Modelling nutrient emissions from river systems and loads to the coastal zone:€Po River case study, Italy. Ecological Modelling, 184, 37–53. Phillips, G., Pietilainen, O. P., Carvalho, L. et€al. (2008). Chlorophyllnutrient relationships of different lake types using a large European dataset. Aquatic Ecology, 42, 213–226. Powlson, D. S., Addiscott, T. M., Benjamin, N. et€al. (2008). When does nitrate become a risk for humans? Journal of Environmental Quality, 37, 291–295. Rahel, F. J. and Olden, J. D. (2008). Assessing the effects of climate change on aquatic invasive species. Conservation Biology, 22, 521–533. Rousseau, V., Becquevort, S., Parent, J. Y. et€al. (2000). Trophic efficiency of the planktonic food web in a coastal ecosystem dominated by Phaeocystis colonies. Journal of Sea Research, 43, 357–372. Salomez, J. and Hofman, G. (2003). Nitrates:€a mixed blessing? Food Science and Technology, 17, 44–45. Salomon, M. (2009). Recent European initiatives in marine protection policy:€towards lasting protection for Europe’s seas? Environmental Science and Policy, 12, 359–366. Sanchez, P. A. and Swaminathan, M. S. (2005). Cutting world hunger in half. Science, 307, 357–359. Scheffer, M., Carpenter, S., Foley, J. A., Folke, C. and Walker, B. (2001). Catastrophic shifts in ecosystems. Nature, 413, 591–596. Schlesinger, W. H. (2009). On the fate of anthropogenic nitrogen. Proceedings of the National Academy of Sciences of the United States of America, 106, 203–208. Schmidt, C. K. and Brauch, H. J. (2008). N,N-Dimethylsulfamide as precursor for N-Nitrosodimethylamine (NDMA) formation€upon ozonation and its fate during drinking water treatment. Environmental Science and Technology, 42, 6340–6346. Schreiber, H., Constantinescu, L. T. et€al. (2003). Harmonised inventory of point and diffuse emissions of nitrogen and phosphorus for a transboundary river basin, http://danubis. icpot.org Seitzinger, S., Harrison, J. A., Böhlke, J. K. et€al. (2006). Denitrification across landscapes and waterscapes:€a synthesis. Ecological Applications, 16, 2064–2090. Seitzinger S. P., Mayorga E., Kroeze C. et€al. (2009). Global river nutrient export trajectories 1970–2050:€a Millennium Ecosystem Assessment scenario analysis. Global Biogeochemical Cycles, doi:10.1029/2009GB003587 Sileika, A. S., Stalnacke, P., Kutra, S., Gaigalis, K. and Berankiene,€L. (2006). Temporal and spatial variation of nutrient levels in the Nemunas River (Lithuania and Belarus). Environmental Monitoring and Assessment, 122, 335–354.

Simpson, D., Aas, W., Bartnicki, J. et€al. (2011). Atmospheric transport and deposition of nitrogen in Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Smith, V. H. (2003). Eutrophication of freshwater and coastal marine ecosystems:€a global problem. Environmental Science and Pollution Research, 10, 126–139. Smith, V. H. and Schindler, D. W. (2009). Eutrophication science:€where do we go from here? Trends in Ecology and Evolution, 24, 201–207. Stalnacke, P., Vandsemb, S. M., Vassiljev, A., Grimvall, A. and Jolankai, G. (2004). Changes in nutrient levels in some Eastern European rivers in response to large-scale changes in agriculture. Water Science and Technology, 49, 29–36. Stevens, C. J. and Quinton, J. N. (2009). Policy implications of pollution swapping. Physics and Chemistry of the Earth, 34, 589–594. Tappeiner, G., Tappeiner, U. and Walde, J. (2007). Integrating disciplinary research into an interdisciplinary framework: A€case€study in sustainability research:€introduction to the special issue. Environmental Modeling and Assessment, 12, 253–256. Thieu, V., Mayorga, E., Billen, G. and Garnier, J. (2010a). Subregional and downscaled global scenarios of nutrient transfer in river basins:€Seine–Somme–Scheldt case study. Global Biogeochemical Cycles, 24, GB0A10. Thieu, V., Garnier, J. and Billen, G. (2010b). Assessing impact of nutrients mitigation measures along rivers continuum to southern bight of the North Sea. Science of the Total Environment, 408, 1245–1255. UNESDA (2009). Union of European Beverages Associations. van Drecht, G., Bouwman, A. F., Knoop, J. M., Beusen, A. H. W. and Meinardi, C. R. (2003). Global modeling of the fate of nitrogen from point and nonpoint sources in soils, groundwater, and surface water. Global Biogeochemical Cycles, 17, 26–31. van Grinsven, H., Ward, M. H., Benjamin, N. and de Kok, T. M. C. M. (2006). Does the evidence about health risks associated with nitrate ingestion warrant an increase of the nitrate standard for drinking water? Environmental Health, 5, 5–26. van Grinsven, H., Rabl, A. and de Kok, T. M. C. M. (2010). Estimation of incidence and social cost of colon cancer due to nitrate in drinking water in the EU: Quantative cost–benefit assessment. Environmental Health, 9(58). Velthof, G. L., Oudendag, D., Witzke, H. P. et€al. (2009). Integrated assessment of nitrogen losses from agriculture in EU-27 using MITERRA-EUROPE. Journal of Environmental Quality, 38, 402–417. Vitousek, P. M., Aber, J. D., Howarth, R. W. et€al. (1997). Human alteration of the global nitrogen cycle:€sources and consequences. Ecological Applications, 7, 737–750. Vollenweider, R. A. (1976). Advances in defining critical loading levels for phosphorus in lake eutrophication. Memorie dell’ Istituto Italiana di Idrotiologia, 33, 53–83. Voss, M., Baker, A., Bange, H.W. et€al. (2011). Nitrogen processes in coastal and marine systems. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Walker, B., Holling, C. S., Carpenter, S. R. and Kinzig, A. (2004). Resilience, adaptability and transformability in social-ecological systems. Ecology and Society, 9(2).

403

Nitrogen as a threat to European water quality Ward, M., de Kok, T., Levallois, P. et€al. (2005). Drinking water nitrate and health: recent findings and research needs. Environmental Health Perspectives 113, 1607–1614. Webb, A. J., Patel, N., Loukogeorgakis, S. et€al. (2008). Acute blood pressure lowering, vasoprotective, and antiplatelet properties of dietary nitrate via bioconversion to nitrite. Hypertension, 51, 784–790. Whitehead, P. G., Wilby, R. L., Battarbee, R. W., Kernan, M. and Wade, A. J. (2009). A review of the potential impacts of climate change on surface water quality. Hydrological Sciences Journal, 54,€101–123. WHO (2007). Public Water Supply and Access to Improved Water Sources,World Heath Organization, Geneva. Wilby, R. L., Whitehead, P. G., Wade, A. J. et€al. (2006). Integrated modelling of climate change impacts on water resources and quality in a lowland catchment:€River Kennet, UK. Journal of Hydrology, 330, 204–220.

404

Wolf, J., Rotter, R. and Oenema, O. (2005). Nutrient emission models in environmental policy evaluation at different scales: experience from the Netherlands. Agriculture, Ecosystems and Environment, 105, 291–306. Worm, B., Barbier, E. B., Beaumont, N. et€al. (2006). Impacts of biodiversity loss on ocean ecosystem services. Science, 314, 787–790. Worrall, F., Spencer, E. and Burt, T. P. (2009). The effectiveness€of€nitrate vulnerable zones for limiting surface water nitrate concentrations. Journal of Hydrology, 370, 21–28. Wriedt, G. and Bouraoui, F. (2009). Towards a large scale assessment of water availability in Europe, European Commission Joint Research Centre, Luxembourg. Zhao, Y.-Y., Boyd, J. M., Woodbeck, M. et€al. (2008). Formation of N-nitrosamines from eleven disinfection treatments of seven eifferent surface waters. Environmental Science and Technology, 42, 4457–4862.

Chapter

18

Nitrogen as a threat to European air quality Lead authors: Jana Moldanová, Peringe Grennfelt and Åsa Jonsson Contributing authors: David Simpson, Till Spranger, Wenche Aas, John Munthe and Ari Rabl

Executive summary Nature of the problem • Atmospheric emissions of nitrogen oxides and ammonia are contributing to a number of negative effects to human health and ecosystems. These effects include both effects of the primary emissions but more importantly through actions of secondary pollutants such as ground level ozone (O3) and secondary particulate matter (PM). • The main air pollution effects include effects of nitrogen dioxide to human health, effects from ground level ozone to human health and vegetation and effects from particulate ammonium and nitrate to human health. There is a difficulty of ascribing health effects to NO2 per se at ambient levels rather than considering NO2 as a surrogate for a traffic-derived air pollution mixture.

Approaches • The chapter gives a brief review of our current understanding of the mechanisms and processes regarding N-containing air pollutants and their effects on human health, vegetation (effects of reactive nitrogen on ecosystems through eutrophication and acidification is treated in Dise et€al., 2011; Velthof et€al., 2011, Chapters 20 and 21, this volume) and materials. It presents historical development, current situation and outlines future perspectives of reactive nitrogen related air pollution and its effects in Europe in relation to national and EU legislation on emission limitation and air quality control.

Key findings/state of knowledge • In the EU-27 countries, 60% of the population lives in areas where the annual EU limit value of NO2 is exceeded. Air quality standards for nitrogen dioxide are exceeded mainly in urban areas. Concentrations have decreased since 1990, although the downward trends have been smaller or even disappeared after 2000. • Episodic ozone concentrations have decreased over Europe since 1990 due to VOC and NOx control. In the same time tropospheric background and continental background concentrations have increased. Present concentrations are still a threat to both human health and vegetation. • Ammonium and nitrate comprise substantial fractions of PM10 and PM2.5 (sometimes more than 1/3 and control of these compounds is important for meeting air quality standards). • It is very likely that sensitive species are and will be negatively affected by emissions of ammonia almost everywhere in western, central and parts of southern Europe, at least in areas with intensive animal husbandry.

Major uncertainties/challenges • There are large uncertainties with respect to further developments of continental-background ozone concentrations in the atmosphere due to uncertainties in future emissions of methane and nitrogen oxides. • The role of particulate ammonium and particulate nitrate regarding human health effects is still under discussion. The long-term effects of NO2 on human health found in epidemiological studies reflect rather effects of combustion or traffic related air pollution than effects of NO2 per se, making it difficult to evaluate the health effects of NO2 as such.

Recommendations • The role of ammonia and nitrogen oxides regarding PM exposures needs to be further investigated, in particular with respect to their importance for health effects. • The low success in controlling ammonia emissions needs to be further assessed, in particular in connection with the development of new agricultural policies.

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

405

Nitrogen as a threat to European air quality Table 18.1 The role of N containing compounds and ozone in air pollution effects. The threats to ecosystems from N deposition are discussed in Grizzetti et€al., 2011 and Dise et€al., 2011 (Chapter 17 (threats to water)) and Chapter 20 (threats to biodiversity))

Effects Compounds

Human health

Nitrogen dioxide

X

Ammonia

Ecosystems X

Particles NH4 /NO3

X

Ozone

X

+

−

N deposition

Visibility

X

X

X X

X

X

X

X (acidification, eutrophication)

18.1╇ Introduction Air pollution is a major threat to human health and ecosystems in Europe. The EU Thematic Strategy on Air Pollution (TSAP) (CEC, 2005) estimated that air pollution in the year 2000 caused between 300 000 and 400 000 premature deaths, mainly due to particles but also with a significant contribution from ozone. Reactive nitrogen contributes significantly to formation of both these air pollutants. European ecosystems are also threatened from air pollution through deposition of N and S containing compounds and through direct effects on vegetation. In addition to the effects on human health and ecosystems there are also significant air pollution effects on materials an visibility. The European Union has through several directives regulated emissions of air pollution. These regulations include directives on emission from combustion plants, motor vehicles, off-road machinery, industrial processes, etc., but also the emissions ceilings directive. Air quality is further regulated through air quality standards (EC, 2008a, b, c). In addition there is a protocol on national emissions ceilings under the Convention on Long-Range Transboundary Air Pollution (LRTAP Convention) regulating emissions of sulphur dioxide, nitrogen oxides, ammonia and volatile organic compounds. All these efforts have caused a decrease in emissions since 1990. Even if the emissions are going down and are expected to be further reduced over the next decade, air pollution will still be a significant threat to European population and ecosystems over the next decade. Nitrogen oxides and ammonia emissions play an important role in these effects both directly through the action of the primary emissions but also indirectly through actions of secondary air pollutants and through their deposition to the ground (Table 18.1). In this chapter we will mainly assess the role of nitrogen in threats to human health, direct effects on vegetation (effects on acidification and eutrophication of ecosystems are treated in Dise et€ al., 2011, Chapter 20 this volume) and effects on materials. Combustion processes, e.g. road traffic and industry are large contributors to emissions of nitrogen oxides (NOx), mainly in the form of nitric oxide, NO. As mentioned in Hertel et€al., 2011 (Chapter 9, this volume), NO is rapidly oxidised to nitrogen dioxide, NO2, in the atmosphere. Nitrogen dioxide is a

406

Materials

strong oxidant that absorbs visible light and hence may form a brownish red colour layer during high concentration episodes. Additionally, nitrogen dioxide is a toxic gas that can cause both long term and short term effects on health. As already mentioned in Hertel et€al., 2011 (Chapter 9, this volume), the threats of nitrogen oxides to air quality do not only concern NO and NO2 themselves. Emission of NOx also contributes to the formation of secondary pollutants, i.e. pollutants that are formed in the atmosphere, such as ozone (O3) and secondary particulate matter (PM). The exact formation pathways are covered in detail in Hertel et€ al., 2011 (Chapter 9, this volume), and hence only briefly presented here. Ozone is formed photochemically in the presence of NO2 and volatile organic compounds (VOC). However, ozone can also be destroyed by reaction with NO and therefore low ozone concentrations are often observed close to the NOx sources while the high concentrations are observed further from the sources in urban background air. Ozone is one of the most important of the global air pollutants in terms of impacts to human health, croplands and natural plant communities, and may become more important in the future. Tropospheric ozone has also impacted on climate; according to IPCC (2007) the year 2005 radiative forcing caused by ozone formed from anthropogenic emissions was the third largest (0.35 W/m2) after that of anthropogenic CO2 and methane (1.66 and 0.45 W/m2 respectively). Particulate nitrate can be formed from oxidation of NO2 to nitric acid (HNO3) that can further react with ammonia to form ammonium nitrate or can be absorbed on existing particulate matter. Nitrate and ammonium are two of the major inorganic components in urban aerosol particles. In the atmosphere NH3 reacts not only with HNO3 but also with other acid gases such as H2SO4 and HCl, and aerosols, forming ammonium (NH4+) containing particles. Oxides of nitrogen can also contribute to formation of Secondary Organic Aerosol particles (SOA) in photochemical smog. Atmospheric particles have an adverse impact on both climate and health. The climate effect is both direct via absorbing terrestrial radiation and scattering solar radiation and indirect, e.g. by influencing clouds. Both effects lead to cooling of the climate (IPCC, 2007) and will be covered in more detail in Butterbach-Bahl et€al., 2011 (Chapter 19, this volume). The fact that particles scatter/absorb light also affects visibility in cities and scenic areas.

Jana Moldanová, Peringe Grennfelt and Åsa Jonsson

Particulate matter (PM) is the most important contributor to adverse health effects of air pollution (WHO, 2005a). Before assessing effects of reactive nitrogen through contribution to the PM-mass we will give a brief overview of PM properties, sources and sinks. Particles can be both natural and anthropogenic in origin. Soil erosion, sea spray, volcanic eruptions and oxidation of biogenic VOC are examples of natural sources, while e.g. biomass burning and fossil fuel combustion are anthropogenic sources. Particles are also addressed as being primary or secondary depending on how they arise in the atmosphere. If the particles are emitted directly from its source they are referred to as primary, whereas the term secondary is used for particles that are formed in the atmosphere via gas-to-particle conversion, often induced by chemical reactions. The formation of particulate nitrate is an example of a secondary particle formation process initiated by the oxidation of NO2 to HNO3. Particles are not only classified regarding their origin but also by size. The particle size range is divided into coarse particles, i.e. particles with a diameter of >2.5 μm, and fine particles < 2.5 μm. The fine fraction is further divided into accumulation mode (100 nm–2.5 μm), ultrafine mode (10–100 nm) and nucleation mode particles (<â•›10 nm). Regulation and guidelines are using the concept of PM10 and PM2.5, which means particulate mass of particles with less than 10â•›μm or 2.5 μm in aerodynamic diameter, respectively. The reason for choosing 10 μm is that it includes the inhalable particles, i.e. those that are small enough to reach the thoracic region. PM2.5 includes only fine particles, hence excluding the coarse particle fraction, as this fine fraction has a higher probability to penetrate deeper into the lungs, reaching the alveolar region. However, as air quality with respect to particles is assessed on mass based metrics, pollution by ultrafine particles (UFP) is to a large extent hidden (a particle with 2.5â•›μm diameter has ~16€000 times higher mass than a particle of same density and 100â•›nm diameter). The number of particles is a better metric if effects of ultrafine particles are considered. The size of the particles determines their transport and removal processes in the atmosphere. Deposition of very small particles (ultrafine or nucleation range) is driven by diffusion and these particles can also grow to larger size fraction by condensation and coagulation. Coarse particles are removed from the atmosphere at a rather fast rate by deposition which is mainly driven by settling/ sedimentation. Growth of these particles by condensation or coagulation in terms of increase of their diameter is not very effective due to the large mass needed for any further growth. Fine particles in the accumulation mode (100 nm€– 2.5 μm) have the lowest deposition velocity, as neither diffusion nor sedimentation are effective, grow slowly and hence tend to accumulate in the atmosphere. During a rain event particles are scavenged by rain droplets and removed through wet deposition. Lifetime of aerosol particles with respect to precipitation scavenging varies from days to minutes, depending on diameter of scavenged particles and type and intensity of precipitation. Particles in size- range 100 nm–1 μm, i.e. those in the accumulation mode, are scavenged with the least efficiency.

18.2╇ Effects of air pollutants associated to reactive nitrogen 18.2.1╇ International legislation By the late 1970s, air pollution had become the main environmental problem faced by many countries in Europe and North America, harming people’s health and damaging ecosystems, historic buildings and monuments. In 1979, the Member States of the UN Economic Commission for Europe adopted the Convention on Long-range Transboundary Air Pollution (UN/ECE, 1979) which was the first international environmental agreement to address this threat to human health and wellbeing. The Convention is of a rather general nature and specific reductions are given in eight Protocols to the Convention. Emissions of nitrogen species are treated by the Sofia Protocol from 1988 that entered into force in 1991 (UN/ECE, 1988) and the Gothenburg Protocol from 1999 (UN/ECE, 1999). The Gothenburg Protocol came into force in 2005 and set emission ceilings for European emission of NOx which should be reduced by 41â•›% (VOC by 40%, NH3 by 17%) by 2010 compared to the level of 1990 (see Jensen et€al., 2011, Section 3.3 of Chapter 3 this volume). Additionally, the EU introduced a National Emission Ceiling (NEC) Directive 2001/81/EC (EC, 2001a), concerning NOx and other pollutants. Compared to the Gothenburg Protocol, this directive puts more pressure on some of the member states. The emission ceilings must be attained by 2010. In the European Union several directives have set requirements and standards for NOx emissions from all kinds of combustion sources:€ Directive 96/61/EC concerning integrated pollution prevention and control (IPPC) (EC 1996a), Directive 2000/76/EC on the incineration of waste (WID) (EC, 2000), Directive 2001/80/EC on the limitation of emissions of certain pollutants into the air from Large Combustion Plants (LCP) (EC, 2001b), Directive 97/68/EC regulating emissions of gases and particulate pollutants from non-road mobile machinery (EC, 1997a). The LCP Directive and the WID Directive were incorporated into a new EU Directive on industrial emissions concerning integrated pollution prevention and control 2008/1/EC (EC, 2008a) that replaced the old IPPC Directive 96/61/EC. Motor vehicle emissions have originally been regulated by Directive 70/220/EEC (light-duty vehicles) (EEC, 1970) and 88/77/EC (heavy-duty vehicles) (EEC, 1988) and amendments to those directives. In 1992 Euro 1 (for light duty vehicles, petrol and diesel) and Euro I (for heavy duty vehicles) standards entered into force (91/441/EEC; 93/59/EEC) (EEC, 1991, 1993) and in 1996 the Euro 2 Euro II standards (94/12/EC; 96/69/ EC) (EC, 1994, 1996b). These standards involved improved combustion and catalysts. The Auto-Oil Programme, which focused on the emissions of gases and particles, resulted in the Euro 3 and Euro 4 stages for light-duty vehicles (Directive 98/69/EC) (EC, 1998a) and in the Euro III and IV standards for heavy duty vehicles (Directive 99/96/EC now repealed) (EC, 1999a), as well as the fuel quality Directive 98/70/EC (EC, 1998b). Further reductions of emissions from light duty

407

Nitrogen as a threat to European air quality

vehicles were set in the Euro 5 stage that entered into force in 2009 (EC Regulations 715/2007 and 692/2008) (EC, 2007, 2008b) and which focused on emissions of particulate matter from diesel cars. Euro 6 is scheduled by the same regulations to enter into force in January 2014 and will mainly further reduce the emissions of NOx from diesel cars. Directive 2005/55/EC (EC, 2005a) on Emission from Ignition Engines in Heavyduty Vehicles (HDV Directive) replaced Directive 99/96/EC and set the Euro IV and V standards that were implemented by Directive 2005/78/EC (EC, 2005b). Euro IV entered into force October 2005 and Euro V October 2008. Euro VI stage was implemented by EC Regulation 595/2009 (EC, 2009) that replaced the HDV directive and scheduled Euro VI to enter into force in 2013. Emissions from international maritime shipping contribute significantly to air pollution in Europe. Legislation is in force to control the emissions through Annex VI of the Marine Pollution Convention (MARPOL) that was adopted in 1997 by Marine Environmental Protection Committee (MEPC) of the International Maritime Organisation (IMO) and that came into force in May 2005 (IMO, 2006). Annex VI with its amendment from October 2008 put progressive limits on emissions of SO2 and NOx globally and contains provisions allowing establishment of Emission Control Areas (ECA) with more stringent reductions of emissions of SOX, particulate matter, NOx or all three pollutants (IMO, 2009). In Europe, Baltic Sea was established as SOX ECA in 2005 and North Sea in 2006. In these areas also the most stringent controls for emissions of NOx applies on ships constructed on or after 1 January 2016. Considering the long lifetime of ship engines, this legislation will impact the NOx emissions only in the distant future. Further, the expected increase in the volume of ship movements will compensate for the environmental benefits of these measures and will lead to a continued growth in ship emissions. Drivers for the aviation industry to reduce its emissions are the International Civil Aviation Organisation’s (ICAO) aircraft engine emissions standards. In 1981, ICAO published its Annex 16, Volume II:€Environmental Protection€– Aircraft Emissions Standards; The Convention on International Civil Aviation. These standards covered the limits for emissions of HC, CO, and NOx. The ICAO aircraft engine NOx emissions standards have gradually been tightened. In 1993 ICAO reduced the permitted levels by 20% for newly certificated engines, with a production cut-off on 31 December 1999. In 1999 the ICAO further tightened the NOx standard by about 16% for engines newly certified from 31 December 2003 and in October 2004, the 1999-standards for NOx were further tightened by 12% for engines certified in 2008. It should be noted that the long lifetime of the aviation fleet causes a lag of NOx emission reductions behind introduction of the ICAO standards. Concerning air quality legislation, the LRTAP Convention and its Protocols were translated into a series of EC directives. In 1996, the Environment Council adopted a Framework Directive on Ambient Air (96/62/EC) (EC, 1996c) which addresses ambient air quality assessment and management. This framework directive includes a series of daughter directives, which set the numerical limit values for atmospheric

408

pollutants. Two of them concern reactive Â�nitrogen and its secondary products in air. The first daughter directive (1999/30/ EC) (EC, 1999b) relates to limit values for among others oxides of nitrogen, nitrogen dioxide and particulate matter (PM10) in ambient air and a date when they must be met. The third daughter directive (2002/3/EC) (EC, 2002) established target values and long term objectives for the concentration of ozone in air. Council Decision 97/101/EC and Commission Decision 2004/461/EC (EC 1997b, 2004) established a reciprocal exchange of information and data on ambient air pollution within the Member States and annual reporting on ambient air quality under the framework directive and its daughter directives. The framework for environmental policy-making in the European Union for the period 2002–2012 was set out in the Sixth Environment Action Programme (EAP) of the European Community which was adopted by the European Parliament and the Council in 2002. The EAP includes Environment and Health as one of the four main target areas requiring greater effort and air pollution is one of the issues highlighted in this area. The Sixth EAP aims to achieve levels of air quality that do not result in unacceptable impacts and risks to human health, paying particular attention to sensitive populations. The Clean Air For Europe (CAFE) initiative of the European Commission provided the scientific background and set out the objectives and measures for this phase of European air quality policy. In its Thematic Strategy on Air Pollution (TSAP) (CEC, 2005), the European Commission has established health and environmental interim objectives for the year 2020 to guide the ambition level of further measures to reduce the impacts of air pollution in Europe. Acknowledging the preliminary nature of some of the input data that have been used for the CAFE analysis the approach in the TSAP is in terms of relative improvements compared to the situation as assessed with the same methodology for the year 2000. The health objectives of TSAP are a reduction of life years lost (YOLLs) from air pollution by particulate matter and a reduction of premature mortality cases from ozone (Table 18.2). Objectives of TSAP regarding effects on ecosystems are reductions of ecosystem areas where deposition of eutrophying and acidifying species exceeds critical loads of these areas (Table 18.2). In order to incorporate the latest health and scientific developments, and objectives of the EAP, First, Second and Third daughter directives and the Council Decision 97/101/EC were replaced in 2008 by a single Air Quality Directive 2008/50/EC (EC, 2008c) that merges existing legislation for all groups of air pollutants and also contains a new regulation for PM2.5, including a limit value, target value and an Exposure Concentration Obligation. The health-related concentration limits of the Air Quality Directive 2008/50/EC for primary and secondary air pollutants related to reactive nitrogen are presented in Table€18.3. The vegetation-related concentration limits for primary and secondary air pollutants related to reactive nitrogen are presented in Table 18.4. The long-term objectives define the level to be attained eventually, save where it is not achievable through proportionate measures, with the aim of providing

Jana Moldanová, Peringe Grennfelt and Åsa Jonsson Table 18.2 Environmental objectives of the Thematic Strategy expressed as percentage improvements relative to the situation in the year 2000 (from Amann et€al., 2008)

Indicator

Unit of the indicator

Percentage improvement in 2020 compared to the situation in 2000

Loss of life expectancy as a result of exposure to PM

YOLLsa

47%

Acute mortalities from exposure to ozone

Nr. of premature deaths

10%

Area of forest ecosystems where acid deposition exceeds the critical loads for acidificationb

km

2

74%

Area of freshwater ecosystems where acid deposition exceeds the critical loads for acidification

km2

39%

Ecosystems area where nitrogen deposition exceeds the critical loads for eutrophicationb

km2

43%

Area of forest ecosystems where ozone concentrations exceed the critical levels for ozone

km2

15%

a b

Years Of Life Lost See glossary for ’Critical load’.

Table 18.3 Health-related limit- and target concentration values from EU-Directive 2008/50/EC. For PM2.5 the exposure concentration limit value is also shown. This is determined as Average Exposure Indicator (AEI) which is a 3-year running annual mean PM2.5 concentration averaged over the selected monitoring stations in agglomerations and larger urban areas set in urban background locations

Limit/target values for the protection of human health Parameter NO2 PM10 PM2.5

Ozone

Averaging period

Target/limit value µg/m3

Permitted exceedances each year

Date when the limit/target value should be met

1 hour

200

18

Limit value 1/1 2010

1 year

40

—

Limit value 1/1 2010

24 h

50

35

Limit value in force since 2005

1 year

40

—

Limit value in force since 2005

1 year

25

—

Target value 1/1 2010 Limit value 1/1 2015

1 year

20

—

Limit value 1/1 2020

1 year

20 (AEI)

—

Limit value 2015

Maximum daily 8 hour mean

120

25 days per year averaged over 3 years.

Target value 1/1 2010

effective protection of both human health and the environment (see glossary for ‘Critical level’). The European air quality legislation is built on the principle that EU Member States divide their territory into a number of air quality management zones and agglomerations. In these zones and agglomerations, the Member States should assess the air quality using measurements, modelling or other empirical techniques. Where air quality levels/ concentrations are elevated, the EU Member States have to prepare an air quality plan or programme to ensure compliance with the limit value before the date when the limit value formally enters into force. In addition, information on air quality should be disseminated to the public. The EU Member States submit annually their air quality data to AirBase, the European air quality information system, and report on air quality in the form of a predefined questionnaire (Decision 2004/461/EC) to the EU Commission.

European Environmental Agency (EEA) on behalf of the EU Commission and supported by EEA’s European Topic Centre on Air and Climate Change (ETC/ACC) maintain AirBase and produces technical papers for each reporting year with overviews and analyses of the submitted information concerning data quality and zone exceedances in the EU Member States. On the UN-ECE level monitoring and evaluation of the international protocols negotiated within the LRTAP Convention is handled by EMEP programme (Co-operative Programme for Monitoring and Evaluation of the Long-Range Transmission of Air Pollutants in Europe) that was set-up by the first protocol to the LRTAP Convention in 1988. Main elements of EMEP are:€collection of emission data, measurements of air and precipitation quality, modelling of atmospheric transport and deposition of air pollutions and from year 1999 also integrated assessment modelling.

409

Nitrogen as a threat to European air quality Table 18.4 Critical level, target value and long-term objective from EU-Directive 2008/50/EC related to the protection of vegetation and ecosystems

Parameter

Averaging period

Value

NOx Ozone

a

1 year

30 µg/m

24 h

75 µg/m3

AOT40a from 1 h means during May–July averaged over 5 years

18 000 µg/m3*h

Target value (to be reached by 1/1 2010)

AOT40a, from 1 h means during May–July

6 000 µg/m3*h

Long-term objective (no defined attainment date)

Critical level (in force)

Accumulated exposure to ozone Over a Threshold of 40 ppb; see glossary for definition.

Table 18.5 Overview of nitrogen related health impacts

Health impacts and routes

Health impacts

NOx

Inhalation - direct impacts ╇ of NO2 - impacts via O3 - impacts via PM

Asthma, respiratory ╇disorder, inflammation of air ways, reduced lung functions, bronchitis, cancers

NH3

Inhalation: - direct impacts ╇ (negligible) - impacts via PM Odour

See NOx Small as odour ╇contribution by NH3 is modest

N2O

Health impacts ╇ from global ╇warming, often enhanced by eutrophication

Enhancement of vectors ╇for infectious diseases (malaria) and frequency of infestations (HAB*, insects)

Pollutant

* Harmful Algal Bloom

18.2.2╇ Health effects There are several routes by which nitrogen air-born pollutants can affect human health leading to a variety of impacts (Table 18.5; see also Townsend et€al., 2003). In the sequel dose–response relations will be discussed for all listed impacts, except, due to lack of information, for the effects of odour and global warming.

Ammonia Health effects of ammonia are indirect through contribution of NH4+ to particulate matter. Scientific findings of the CAFE programme pointed to the fact that ammonia emissions significantly contribute to the formation of secondary particulate matter in the atmosphere (~ 20% by mass). The main source of ammonia in the atmosphere is agriculture (cattle, pig and poultry farming and use of N-fertilisers).

Nitrogen dioxide (NO2) NO2 is a toxic gas that has adverse health effects both in the long term (chronic) and short term (acute). Examples are:€ asthma, respiratory disorder, inflammation of the airways, reduced lung function, increased bronchitic symptoms

410

Threshold type 3

and cancer. The major cause of these effects is oxidative stress, i.e. formation of radicals that are very reactive and destructive to cell tissues. This oxidative stress can be generated by gaseous compounds such as NO2 and ozone or by increased production and release of oxygenated compounds in the body due to inflammation. Epidemiological studies on the health effects of exposure to nitrogen dioxide have been extensively reviewed by the WHO (WHO, 2000, 2003, 2004, 2006). A 1 hour guideline of 200 μg/m3 and an annual mean of 40 μg/m3 were recommended in the second edition of ‘Air quality guidelines for Europe’ (WHO, 2000). These values have been adopted as limit values in the EU’s air pollution directive 2008/50/EC that should be met by all Member States in 2010 (Table 18.3). Nitrogen dioxide is strongly related to PM, as both come from the same combustion sources, and it is converted to nitrates and contributes per se to fine particle mass. Several studies have noted a high correlation between nitrogen dioxide levels and suspended PM generated from the same combustion sources. At a given site, a high correlation exists between nitrogen dioxide and organic and elemental carbon, inorganic acids, PM2.5 and ultrafine particles so that nitrogen dioxide may be considered a very good indicator of the complex gas–particle mixture that originates from vehicular traffic (Gauderman et€ al., 2000; Seaton and Dennekamp, 2003) but it is very difficult to differentiate the effects of nitrogen dioxide from those of other pollutants in epidemiological studies (WHO, 2006). Factors such as temperature and humidity are also important when assessing the toxicity of a gaseous mixture (WHO, 2003). The spatial variability of NO2 is greater that that of PM. Short-term exposure studies have shown that daily average concentrations of nitrogen dioxide are significantly associated with increased overall, cardiovascular and respiratory mortality. The effect estimate for all-cause mortality derived from a well-conducted meta analysis is a 2.8% increase per 24€ppb NO2 (24 h mean). Adjusting for the effect of PM reduces the effect estimate to 0.9%, and the lower confidence interval of the effect estimate includes zero (Stieb et€al., 2002). In the European multi-city studied (Samoli et€al., 2003), the effect of PM on daily mortality was greater in areas with high nitrogen dioxide levels. Hospital admission due to respiratory causes increased by 0.4%–0.5% per 10 µg/m3 24 h average NO2 (Bylin and Forsberg 2009). However, it is important to remember that not all people are affected the same. For instance, studies

Jana Moldanová, Peringe Grennfelt and Åsa Jonsson

indicate that children and people with asthma are more sensitive to NO2 exposure. Data from Europe suggested that long-term concentrations of nitrogen dioxide or nitrogen oxides (NO + NO2) were associated with an increased risk of all-cause mortality (Filleul et€al., 2005; Hoek et€al., 2002; Nafstad et€al., 2004). However, none of the studies found evidence that nitrogen dioxide per se, but rather particulate pollution especially from traffic sources, seemed to be responsible for the observed associations. Other long-term effects of increased concentrations of NO2 include association with childhood cancer and lung cancer in adults, association with decreased lung function in children and in adults and association with incidence of asthma in children (WHO, 2006, and references there in). None of these long-term exposure studies is able to attribute the effects to NO2 concentrations per se.

Ozone Ozone is a highly reactive gas that triggers oxidative stress when it enters the airways. Excessive ozone in the air can have a marked effect on human health. It can cause breathing problems, trigger asthma, reduce lung function and cause lung diseases. As to short-term exposures, recent epidemiological studies have strengthened the evidence that daily exposures to ozone increase mortality and respiratory morbidity rates. The risk of effects increases in proportion to the ozone level, with a significant increase in mortality observed above 50–70 μg/m3 (measured as a 1 or 8 hour average) (WHO, 2008). In shortterm studies ozone appears to have effects independent of other air pollutants such as particulate matter (PM). This notion that ozone may act independently is strengthened by controlled human studies and experimental animal studies showing the potential of ozone per se to cause adverse health effects, especially in vulnerable people. Controlled human studies on PM and ozone combined corroborate this view (WHO, 2008). The relative increase of risk of short-term mortality or hospital admission from increase in maximum ozone 1 h or 8 h average concentrations are shown in Table 18.6. As to long-term exposures, new epidemiological evidence and experimental animal studies indicate effects of long-term exposure to ozone. A recent study of long-term air pollution exposure effects on mortality found that the long-term ozone exposure was associated with the risk of death from respiratory causes whereas PM2.5 was associated with the risk of death from cardiovascular causes. The estimated increase in relative risk of death from respiratory causes that was associated with an increment in ozone concentration of 10 μg/m3 was 2% (Jerrett et€al., 2009) (Table 18.6). With these long-term effects on respiratorycaused mortality O3 would be one of the most important air pollutants associated with health in Europe and the currently implemented policies would not be sufficient to reduce the impacts significantly in the next decade (ECE, 2009). The epidemiological studies pointed at existence of a threshold in the dose–response functions relating the short term ozone exposure to adverse health effects. The European APHEA2 study found a significant increase in risk of dying when ozone maximum daily 1â•›h average concentrations

exceeded 50–60 μg/m3 (Katsouyanni et€al., 2001). Study of Bell et€al. (2006) from US cities found that the estimates of increased risk of mortality were statistically significant above daily average ozone concentration of 80 μg/m3 and were stable for concentrations over 70 μg/m3. WHO recommended a cut-off value of 70 μg/m3 for integrated assessment modelling (WHO, 2008). In implementing the cut-off, no effects of ozone on health are calculated on days with a maximum daily 8 h average below 70 μg/m3 and for days with ozone concentrations above 70 μg/m3 as a maximum daily 8 h average, only the increment exceeding 70 μg/Â�m3 is used to calculate effects. Owing to the linearity of the concentration–response curve, the accumulated impact estimate is proportional to the sum of excess of maximum daily 8€h averages over the cut-off of 70 μg/m3 calculated for all days in a year. This indicator of cumulative annual exposure is called SOMO35 (Sum Of Means Over 35 ppb, 35 ppb = 70 μg/m3). For assessing ozone exposure in urban areas, urban background concentrations should be used and in line with most of the evidential health studies, it was regarded as sufficient to use one average ozone concentration per city. These recommendations were accepted by the 23rd session of the Working Group on Effects and used in the health impact assessment of CAFE (WHO, 2008). SOMO35 should not, however, be seen as a new, universal or lasting index. In particular, it reflects the evidence base as it was in 2004, i.e. that (a) effects from studies of long-term exposure were not well enough established to be quantified; and (b) there were substantial uncertainties about the slope of the concentration–response function at lower concentrations, say below 70 μg/m3. The SOMO35 index should be reconsidered if and when that evidence base changes to an important degree. Another uncertainty in use of SOMO35 for health impact assessment of CAFE is related to the grid size of EMEP model used for calculation of ozone concentrations. Ozone concentrations in urban areas are usually substantially lower than those in rural areas because of its reaction with primary emissions of nitric oxide, and it is not possible to capture this in a model with grid sizes measuring tens of kilometres. In addition, ozone formation is a non-linear process, and finer grid sizes in the model would presumably lead to a more accurate treatment of the ozone chemistry.

Particulate matter (PM) Emissions of ammonia and contribute significantly to the formation of secondary particulate matter in the atmosphere. Oxides of nitrogen can be converted to nitrates which also contribute to fine particle mass. Reactive nitrogen thus contributes to particle mass and consequently also to the adverse health effects caused by the PM. Since current knowledge does not allow specific quantification of the health effects of individual PM components, it is appropriate that current risk assessment practices consider particles of different sizes, from different sources and with different composition, as equally hazardous to health (WHO, 2007). As mentioned in the previous section, particles with a diameter less than 10â•›µm are inhalable. At which part of the

411

Nitrogen as a threat to European air quality Table 18.6 Estimates of the relative increase of all-cause and cause-specific short-term mortality and respiratory hospital admissions and their confidence intervals (CI) attributable to an increase of 10â•›μg/m3 in daily, maximum 1 h or maximum 8 h average ozone (from WHO, 2008, their Table 2.1) and of relative increase of all-cause and cause-specific long-term mortality attributable to an increase of 10 μg/m3 in maximum 1 h average ozone concentrations (Jerrett et€al., 2009)

Age group (years)

Percentage increase in risk per 10 μg/m3 ozone (95% CI)

Nr. of studies analysed

All-cause mortality, all seasons

All ages

0.4 (0.3–0.9)

32

All-cause mortality, summer

All ages

0.7 (0.4–1.01.1)

10

Cardiovascular mortality, ╇ all seasons

All ages

0.5 (0.3–0.8)

18

Cardiovascular mortality, summer

All ages

1.2 (0.4–2.0)

4

All ages

0.2 (0.1–0.3)

43

All ages

0.2 (0.2–0.3)

46

All-cause mortality

All ages

0.3 (0.1–0.4)

15

Respiratory mortality

All ages

0.0 (-0.4–0.5)

12

Cardiovascular mortality

All ages

0.4 (0.3–0.5)

13

Respiratory hospital admissions

0–14 15–64 ≥65

Not observed 0.1 (-0.9–1.2) 0.5 (-0.2–1.2)

3 5 5

All-cause mortality

≥30

Not observed

1

Respiratory

≥30

2.0 (0.7–3.4)

1

Cardiovascular

≥30

Not observed

1

Meta-analysis /outcome/disease Short-term mortality Daily average (Bell et€al., 2007)

Maximum 1 hr average (Ito et€al, 2005 ) All-cause mortality Maximum 1 h average (Levy et€al., 2005) All-cause mortality Maximum 8 h average (WHO, 2005b)

Long-term mortality Maximum 1 h average, summer half-year (Jerrett et€al., 2009)

respiratory tract the particles deposit mostly depends on their size. The upper airways (nasal and extrathoracic region) with high air velocities collect efficiently large particles in coarse mode and down to diameter of c. 1 µm by impaction process. Collection efficiency of bronchial region is low and only the smallest particles with diameter < 10 nm deposit efficiently in this region by diffusion process. Particles with diameter below 10 µm can reach the alveolar region but the collection efficiency there is the highest for ultrafine particles, i.e. those with diameter < 100 nm (Hinds, 1999). Adverse health effects regarding particles are related both to respiratory and to cardiovascular systems. Examples of adverse health effects concerning the respiratory tract are inflammatory, exacerbation of existing airway disease and impairment of pulmonary defences (WHO, 2005). Cardiovascular effects may be variability in heart rate, arrhythmia and cardiac infarction (Bylin and Forsberg, 2009). The short term respiratory mortality due to particles is larger than the cardiovascular mortality

412

on a percent basis, however, the total deaths from cardiovascular causes outnumber the respiratory (WHO, 2005). Not all people are affected by air pollution to the same extent. Susceptibility is dependent on personal characteristics, i.e. age, health status, etc., and exposure characteristics. Which of these characteristics is most important is not yet fully clear. People with heart or lung problems, as well as the elderly, have been shown to be more sensitive to PM exposure. This fact is important to consider when future impacts of air pollution are predicted, e.g. for an ageing European population. Experimental studies have shown that people with asthma are more sensitive than people without problems with the respiratory system. It has been determined that increased particle concentrations cause increased occurrence of asthma-like conditions as well as increased respiratory hospital admissions (Bylin and Forsberg, 2009). During the last decade, human experiments regarding particle exposure have been performed. Results from these studies have shown that particle concentrations similar to real

Jana Moldanová, Peringe Grennfelt and Åsa Jonsson

world pollution situations can cause respiratory inflammation. However, in epidemiological studies it can be hard to differentiate between effects from PM and other co-emitted compounds. Exposure of particles together with other co-emitted gaseous pollutants may cause inflammatory effects at lower particle concentrations compared to particles alone (Bylin and Forsberg, 2009). Threshold levels for PM concentrations below which no adverse effects occur have not been identified, which makes it difficult to recommend limits. There is some indication that particles of different size fractions may affect health in different ways. Coarse particles (diameter > 2.5 μm) may preferentially affect the airways and lungs, while fine particles (diameter > 0.1 μm) may preferentially affect the cardiovascular system. Ultrafine particles (UFP, diameter > 0.001 μm) may also migrate via the lung to other organs, including the liver, spleen, placenta and foetus, or via the nerve system to the brain. The health implications of these observations remain unknown since there are not yet enough epidemiological studies to be able to determine the exposure-response relationship for fine and ultrafine particles. This is why there are currently no guidelines for UFP exposure. Smaller particles have larger relative surface areas and therefore commonly induce more inflammation (Diociaiuti et€ al., 2001; Pozzi et€al., 2003). Health risks of PM in terms of increase of the diurnal average PM10 by 10 μg/m3 are, according to an analysis done by the WHO (2000), an increase in relative risk of mortality by 0.6%– 1.6%, an increase in occurrence of asthma related problems and medication usage by 3%–5% and increase of the number of daily hospital admission due to respiratory causes by 0.8%. As the long-term exposure to PM results in a substantial reduction in life expectancy, the long term effects clearly have greater significance to public health than the short-term effects. PM2.5 shows the strongest association with mortality indicating a 6% increase in the risk of deaths from all causes per 10â•›μg/m3 increase in long-term PM2.5 concentration. The estimated relative risk amounts to 12% for deaths from cardiovascular diseases and 14% for deaths from lung cancer per 10â•›μg/m3 increase in PM2.5 (Poppe et€ al., 2002, 2004) (Table 18.7). Other effects related to long-term exposure include increases in lower respiratory symptoms and chronic obstructive pulmonary disease and reductions in lung function in children and adults. Studies on large populations show a strong effect of PM2.5 on mortality, and have been unable to identify a threshold concentration below which ambient PM has no effect on health:€a no-effect level. After a thorough review of recent scientific evidence, a WHO working group therefore concluded that, if there is a threshold for PM, it lies in the lower band of currently observed PM concentrations in the European Region. The chemical composition of particles may also influence their health effects. The primary, carbon-centred, combustionderived particles have been found to have considerable inflammatory potency (Armstrong et€al., 2004; Mudway et€al., 2004). One of the hypotheses considered for PM’s mechanisms of action is the oxidative potential of the particles or specific components. PM from traffic sites seems to have high oxidative

Table 18.7 Estimates of increase of long-term mortalities attributable to an increase of the PM exposure by 10 μg/m3 and their confidence intervals (CI) (Poppe et€al.., 2002)

Long-term mortality

Age group (years)

Relative risk per 10 μg/m3 (95% CI)

Mortality, cardiopulmonary

≥30

1.08 (1.02€– 1.14)

Mortality, lung cancer

≥30

1.13 (1.04€– 1.22)

Mortality, total â•… (excluding violent death)

≥30

1.06 (1.02€– 1.10)

activity, and emissions from road traffic have been linked with a wide range of health effects, including effects on the cardiovascular and respiratory systems, and on atopic sensitisation to allergens in outdoor air. There is substantial epidemiological evidence of associations between health and sulphates that suggest that if sulphates are reduced (as part of the reduction of a mixture) then there will be real benefits to health. There is not much evidence for toxicity of airborne nitrates, which may be at least partly due to difficulties with measuring nitrates. Problem arises also with difficulty of epidemiological studies to distinguish effects of different pollutants in ambient air and of toxicological studies to describe effects across all sensitive groups in the population. The situation is very well summed up by this passage from a recent review paper by Reiss et€al. (2007):€‘For nitrate-containing PM, virtually no epidemiological data exist. Limited toxicological evidence does not support a causal association between particulate nitrate compounds and excess health risks. There are some possible indirect processes through which sulfate and nitrate in PM may affect health-related endpoints, including interactions with certain metal species and a linkage with production of secondary organic matter. There is insufficient evidence to include or exclude these processes as being potentially important to PM-associated health risk.’

18.2.3╇ Effects on vegetation Emissions of reactive nitrogen lead to increased atmospheric deposition into ecosystems. Various effects such as acidification and eutrophication of soils and waters, reduced biodiversity and formation of marine algal blooms are summarised in Butterbach-Bahl et€al., Durand et€al., Voß et€al., Grizzetti et€al., Dise et€al. and Velthof et€al., 2011 (Chapters 6, 7, 8, 17, 20 and 21, this volume). These long-term effects are mostly associated with critical loads of nitrogen, 10–100 years deposition levels of total Nr that are set as upper thresholds below which negative effects do not occur in specific ecosystems. In this chapter only the direct effects of the primary and secondary gas-phase pollutants related to emissions of Nr to the air, i.e. NO, NO2, NH3 and ozone, will be assessed. These effects are associated with critical levels of individual air pollutants, i.e. short-term air concentration levels (1 h to 1 year means) ‘fixed on the basis of scientific knowledge, above which direct adverse effects may occur’ (EC, 2008c). Legislation for the protection of natural vegetation against the direct effects from air pollution has developed together

413

Nitrogen as a threat to European air quality

with legislation for air pollution effects on human health. The �vegetation-related concentration limits for primary and secondary air pollutants related to reactive nitrogen as set in Directive 2008/50/EC were presented in Table 18.4.

Ammonia, NH3 Exposure to ammonia leads to a mixture of direct, indirect, primary and secondary effects on vegetation and ecosystems (Cape et€ al., 2009; Sutton et€ al., 2009). There are clear indications that the pathway of uptake of ammonia via leaf uptake from atmosphere, i.e. the direct effect of ammonia, is dominant (as opposed to the indirect effect which is via root system from the soil) (Sutton et€al., 2009). Ammonia acts as a macro-nutrient and at low exposure levels plants respond by increasing their biomass production. Growth stimulation is also considered as potentially adverse for (semi-) natural vegetation because plant growth is often limited by the supply of nutrient nitrogen, and so any increases in growth may lead to negative effects on community composition. The fertilisation effect can at higher exposure levels lead to secondary long-term adverse effects including increased susceptibility to abiotic (drought, frost) and biotic stresses. In addition, various primary toxic effects are known (Cape et€al., 2009; Sutton et€ al., 2009). The critical levels have been revised in recent years on the basis of experimental data (Table 18.8). Annually averaged concentrations below 1 µg/m³ will protect (1) sensitive lichen communities and bryophytes and (2) ecosystems where sensitive lichens and bryophytes are an important part of the ecosystem integrity. Based on data from heathlands and forest ground flora, 3 µg/m³ (uncertainty estimate 2–4 µg/m³) are assumed to protect higher plants. The critical levels given in Table 18.8 apply for native and forest species. A monthly average critical level 23 µg/m³ was retained to deal with the possibility of high peak emissions during periods of manure application.

Oxides of nitrogen, NOx Oxides of nitrogen can have a fertiliser effect, but can also be toxic to plants, depending on concentrations. The critical levels for NOx are based on the sum of the NO and NO2 concentrations because there is insufficient knowledge to establish separate critical levels for the two pollutants, although some evidence indicates that at low concentrations typical of ambient conditions, NO is more phytotoxic than NO2 (Mills, 2004). Since the type of response varies from a fertiliser effect to toxicity depending on concentration, all effects have been considered to be adverse. As for ammonia, the growth stimulation was also considered as potentially adverse for (semi-) natural vegetation owing to potential negative effects on community composition. In the past, the critical level for nitrogen oxides referred only to NO2. However, because of new evidence of the toxicity of nitric oxide (NO), the critical level now refers to NOx, defined as the combined concentrations of NO and NO2. The critical level value remains 30 μg/m3 (as NO2 equivalent) as an annual mean and 75â•›μg/m3 as a 24 hour mean. As for ammonia, UN/ECE Working Group on Effects strongly recommended the

414

Table 18.8 Critical levels (CL) of ammonia (Cape et€al., 2009)

Averaging period

Critical value µg/m3

1 month

23

Provisional value for all ╇ plants

1 year

1

Lichen communities and ╇bryophytes; ecosystems where these are a key part of ecosystem integrity

1 year

3 (2–4)

Higher plants (heath ╇land, grassland and forest ground flora)

Receptor

use of the annual mean value, as this parameter is much more reliable than shorter-term averages, and the long-term effects of NOx are thought to be more significant than the short-term effects (Mills, 2004).

Ozone Ozone damage to vegetation has been recognised and studied for many decades (Benton et€ al., 2000; Matyssek and Innes, 1999; Skärby et€ al., 1998). Today ozone is considered to be the most important gaseous pollutant causing effects on vegetation in Europe. It enters plants through leaf stomata and oxidises plant tissue, causing changes in biochemical and physiological processes and eventually death of the injured plant cells. Besides visible injuries on leafs and needles, ozone also causes premature leaf loss, reduced photosynthesis and reduced leaf, root, and total dry weights in sensitive plant species. This leads to significant decrease in productivity of some agricultural crops and to reduced forest production. In addition, many native plants in natural ecosystems are sensitive to ozone. The developments in ecotoxicology led to development of critical levels for ozone effects on plants. Traditionally, these were related to ozone concentrations, using simple mean values for different time windows (day, month, etc.), as still is the case for other gaseous air pollutants (NOx, SO2, NH3). In 1990s a new kind of critical level for ozone was elaborated, based on accumulated exposure over a concentration threshold, following the rationale that higher ozone concentrations are believed to be more damaging to plants. Example of such metrics is AOT40 used in Europe (Fuhrer et€al., 1997). The LRTAP Convention’s mapping manual (Mills, 2004) suggested two main metrics for use in performing regional scale risk assessments of ozone damage:€the AOT40 index and the flux based AFstY approach (Accumulated stomatal Flux over thresholds of Y nmol/m2/s). The flux-based approach relates risk to the absorbed ozone dose rather than ambient ozone concentration, through the use of stomatal conductance algorithms. These flux-based metrics have been defined in detail in the revised UN-ECE Mapping Manual (Mills, 2004) and are strongly recommended rather than use of AOT40. Table 18.4 presents a summary of the limit and target values of the EU Directive 2008/50/EC related to effects on vegetation.

Jana Moldanová, Peringe Grennfelt and Åsa Jonsson

18.2.4╇ Effects on materials Atmospheric pollution is an important factor in material deterioration including degradation of systems used for material protection and cultural heritage materials. Corrosion of materials was originally mostly associated with air pollution by sulphur dioxide (SO2); however the more recent studies have shown that nitric acid (HNO3), ozone and particulate matter contribute significantly to the negative effect of air pollution on materials. The lifetime of technological products is shortened because of air pollution. Buildings and other structures, as well as objects of cultural heritage, exposed to the atmosphere deteriorate more rapidly. The resulting physico-chemical and economic damage can be significant, not to mention the loss of unique parts of our cultural heritage and hazards due to decreased reliability of complicated technological devices. Also, as the result of weathering, especially that caused by acidifying pollutants, a significant part of the metals used in construction and manufactured products are released to the biosphere with a potential hazard to the environment. Deterioration rates can be calculated using dose–response functions. The recommended functions have been derived from field research programmes undertaken as part of the UN-ECE ICP Materials Exposure Programmes. Two sets of dose–response functions have been derived. One was developed for SO2-dominated situations taking into account the synergic effect of exposure to ozone and the effect of acid rainfall in combination with climatic parameters. The second was developed for multi-pollutant situations combining effects of gaseous SO2, NOx, O3, HNO3 and particulate matter together with acid rainfall in combination with climatic parameters. The impact of wet deposition of acidic species on sensitive materials is considered as an effect of the total load of H+ deposition, and impact of deposition of the sea salt as an effect of the total load of Cl– deposition. The dose–response functions use annual average concentrations of air pollutants and are available for limestone, sandstone, copper, bronze, zinc, steel and aluminium (ICP, 2010). The recommended unit of corrosion attack is the surface recession R (in μm) with the exception of aluminium, where the mass loss ML (in gm–2) should be used instead. Table 18.9 summarises the available dose–response functions for materials and lists the dependencies of these functions. For zinc, bronze and limestone the multi-pollutant function should be used when levels of HNO3 and/or particulate matter are expected to be high, as is the case with most European cities today with pollution dominated by traffic. Dose–response functions for paint coatings are also available, expressed as lifetime equations for the coatings. Because atmospheric deterioration of materials is a cumulative, irreversible process, which proceeds even in the absence of pollutants, ‘critical’ values are not as easily defined as for some natural ecosystems. Some rate of deterioration must be defined which may be considered ‘acceptable’ based on technical and economic considerations. This approach provides the basis for mapping ‘acceptable areas’ for corrosion, and deriving areas where the acceptable pollution level/load is exceeded, in an analogous way to the maps produced for natural ecosystems.

The term ‘acceptable’ is reserved for materials used in technical constructions, while ‘tolerable’ is used in connection with the degradation of cultural heritage. Based on maintenance intervals and tolerable corrosion attack before maintenance for cultural heritage objects, tolerable corrosion rates have been determined. These corrosion rates are 2.5 times higher than background corrosion values and values of tolerable corrosion for the first year exposure of some materials are listed in Table 18.9. Particles also contain soiling materials, and the tolerable PM10 level for soiling of three selected materials is 12–22 μg/m3 based on reasonable cleaning intervals.

18.3╇ Historical trends in air pollution and their current effects on health, vegetation and materials After an improvement of the air pollution situation in Europe at the beginning of nineties the trends in concentrations have been more or less stagnating for many air pollutants during the decade beginning at late 1990s. Overall exposure of Europe’s population to pollutants with a health impact has not improved since the late 1990s; however, there have been some pollutant-specific exceptions. Figure 18.1 shows that increasing part of urban population is exposed to ozone and PM concentrations over the target values. Whilst exposure to high levels of NO2 has steadily decreased, up to 30% of Europe’s urban population may still be exposed to concentrations in excess of limit values. Thus, determined effort is still required if ambient air concentration and exposure targets are to be met. For ozone there was considerable variation between years. Usually, a maximum of 25% of the urban population was exposed to concentrations above limit values. In 2003€– a year with extremely high ozone concentrations€ – this fraction increased to approximately 60%. For PM10 the urban population potentially exposed to ambient air concentrations in excess of the EU limit value varied between 23% and 45% between 1997 and 2004. There was no discernible trend over the period (EEA, 2010). Acidifying emissions in Europe have declined substantially since 1990. As sulphur emissions have fallen, nitrogen has become the predominant acidifying agent. Ozone concentrations have remained largely unchanged in recent years, even though emissions of precursor gases have been falling. Exposure of vegetation to ozone exceeds criteria for protection over very large areas of central and southern Europe. Declining concentrations of acidifying air pollutants has resulted in decreased observed corrosion rates of materials at the ICP Materials sites, by about 50% on average in the period 1987–1997. The corrosion rate of carbon steel decreased further in 1997–2003 (Figure 18.2), though the rates for zinc and limestone increased slightly. Nitric acid and particulate matter currently contribute to corrosion, in addition to sulphur Â�dioxide. Exceedances of tolerable levels of corrosion for cultural heritage materials were frequent. For 1990, it was estimated that air pollution caused € 1.8 billion of materials damage. Emission reductions envisaged under the Gothenburg Protocol are

415

416 

Multi

a

If RH < 60%, counts as RH = 0.

SO2 



multi

Sandstone



SO2

Limestone





multi SO2

Aluminium



SO2

Bronze





Multi SO2



SO2

Copper

Zinc



SO2

Steel

μg/m3

SO2

Situation

Material



O3







PM10





HNO3



















mg/m2/ year

H+





Cl−



















%

a

RH

C





















o

T

7

8

0.22

0.6

0.8

1.1

20

μm/year

Tolerable corrosion

Table 18.9 Functional dependencies of the dose–response functions for materials (UN/ECE ICP-materials, 2007). Situation (of air pollution) is SO2-dominated (SO2) or multi-pollutant (multi); dependencies of dose–response functions on concentrations of gaseous pollutants, wet deposition of acidity (H+), chlorides (Cl−), relative humidity (RH) and temperature (T) are based for annual mean values of these environmental variables

Jana Moldanová, Peringe Grennfelt and Åsa Jonsson

and 3 μg/m3 for herbaceous plants; Cape et€ al., 2009) reveals the actual risk of elevated NH3 emissions on terrestrial ecosystems. Despite uncertainties, it is very likely that bryophytes and lichens are negatively affected almost everywhere in Western, Central and parts of Southern Europe, and herbaceous plants in areas with intensive animal husbandry (NW France, Flanders, Netherlands, Denmark, parts of Germany, Switzerland and Northern Italy). Not surprisingly, these are also the areas with highest nitrogen deposition rates and exceedances of critical loads, with additional adverse effects on sensitive ecosystems (see Dise et€al., 2011, Chapter 20 this volume). The contribution of ammonium to the PM concentrations can be estimated from measurements of PM chemical composition. Putaud et€al. (2004) reviewed composition of PM measured over the last decade at 24 sites situated in natural, rural, near-city, urban, and kerbside areas in Europe. Figure 18.4 shows the contribution of ammonium to PM2.5, PM10 and coarse PM at the different types of locations. The figure indicates that contribution of ammonium to PM2.5 in Europe is around 8%. The ammonium contribution to PM is by Amann et€al. (2008) assumed proportional to the PM-related health effects calculated for the National Emission Ceilings analysis. Loss in statistical life expectancy attributable to the exposure of PM2.5 is between 6 and 36 months in central Europe (Figure 18.51).

expected to improve materials damage across Europe by more than € 1 billion (UN/ECE ICP-materials, 2007).

18.3.1╇ Ammonia Regional ammonia (NH3) concentrations in Europe have been calculated with the EMEP MSC-West model; Figure 18.3 shows the yearly mean concentrations for the years 1990, 1995, 2000 and 2005. Note that with a grid size of 50 km × 50 km, the model cannot capture the large gradients in ammonia that exist at small spatial scales. The model overestimates low ammonia concentrations (for instance at the forest sites from the EU project NOFRETETE) and underestimates high concentrations (e.g. EMEP sites, which are situated in rural areas, often surrounded by agricultural activities). However, the model calculations did not show any systematic deviation for ammonia with respect to seasons. Comparing the calculated NH3 concentration with the updated critical levels (1 μg/m3 for lichens and bryophytes % of urban population 100 80 60

18.3.2╇ Oxides of nitrogen

40

The concentrations of NO2 are spatially very variable in the urban environment, depending on time of the day, season, reactivity and meteorological factors. The natural background annual mean of NO2 is between 0.4–9.4â•›μg/m3, the urban annual mean is usually 20–90â•›μg/m3 (WHO, 2006) and for the rural background it is 15–30â•›μg/m3. Hence the WHO guideline, as well as the EU limit value concerning NO2, is exceeded in many larger cities and, as emissions of NOx are strongly traffic-related, there is a rising concern regarding NO2 concentrations in growing cities with high traffic density. For instance in a megacity like Beijing the annual average NO2 concentration in 2002 was 76 μg/m3 (Molina and Molina, 2004). In the EU-27

20 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 NO2

PM10

O3

Figure 18.1 Percentage of the urban population in Europe (EEA member countries) potentially exposed to pollutant concentrations over selected limit/ target values (only pollutants (partly) related to NR) (NO2:€annual mean of 40 μg/m3 NO2 not to be exceeded; PM10:€24 h average of 50 μg/m3 not to be exceeded more than 35 times a calendar year; O3:€daily maximum of 8 h mean of 120 μg/m3 O3 not to be exceeded more than 25 days per calendar year, averaged over three years) (EEA, 2010a).

80

60 50 40 30 20 10

Lisbon (P)

Chaumont (F)

Lahemaa (Est)

Moscow (Rus)

Madrid (E)

Toledo (E)

Lincoln Cath. (GB)

Aspvreten (S)

Birkenes (N)

Stockholm (S)

Oslo (N)

Venice (I)

Milan (I)

Rome (I)

Cassacia (I)

Bottrop (D)

Langenfeld (D)

Ahtäri (Fin)

Waldhof (D)

Kopisty (Cz)

0 Prague (Cz)

m/year

Figure 18.2 Carbon steel corrosion rates for sites in 21 European cities for years between 1987 and 2005 (see legend). The grey dashed line is the tolerable corrosion rate 20 μg/year (figure prepared from data in UN/ECE ICP-Materials, 2006, and 2007).

1987/88 1992/93 1994/95 1996/97 1997/98 2000/01 2002/03 2005/06

70

417

Nitrogen as a threat to European air quality

1990

1995

8

8

4

4

3

3

2

2

1

1

0.5

0.5

2005

2000

8

8

4

4

3

3

2

2

1

1

0.5

0.5

Figure 18.3 NH3 concentrations (in μg/m3over Europe calculated with the EMEP model for the years 1990, 1995, 2000 and 2005 using a legend with limits of 1â•›μg/m3 for lichens and bryophytes (≥ darkgreen area) and of 3 μg/m3 for herbaceous plants (≥ dark orange area).

countries, 60% of the population lives in areas where the annual EU limit value of NO2 is exceeded. The exceedance is primarily associated with the urban environment and its local traffic (Mol et€al., 2010). Urban pollution is assessed in detail in SvirejevaHopkins et€al., 2011 (Chapter 12 this volume). Also in suburban areas some Member States are still having problems with attaining the annual limit value as can be seen in Figure 18.5. In Europe (the EEA32 countries) the NOx emission have decreased by 18% in the period 1997–2006; traffic related

418

emissions showed an even larger decrease of 28%. In line with this, the ambient NOx concentrations showed decreasing trend throughout the region with decrease of about 22% at rural stations and 27% at urban/traffic stations between 1997 and 2007 (Figure 18.6). While the decrease in NO2 concentrations at rural and urban stations was quite similar to the decrease in NOx during this period, reduction in NO2 concentrations at traffic stations was small (6%) (Figure 18.6; Mol et€al., 2009). This can be explained by two processes that are important only

Jana Moldanová, Peringe Grennfelt and Åsa Jonsson

very close to the sources. Firstly, at a constant oxidant level the NO2/NOx ratio increases with lowering NOx concentration due to the NO titration. Secondly, there are clear indications that the fraction of directly emitted (primary) NO2 in the total NOx emission from road transport is increasing. The main reason is high proportion of the primary NO2 in emissions from the growing fleet of diesel cars of Euro 3 and III standard and

Figure 18.4 Contribution of ammonium to PM2.5, PM10 and coarse PM (=PM10−PM2.5) at natural and rural background, near city and urban background, and kerbside measurement stations in Europe (24 in total). Data for years 1991–2001 were analysed after the data from Putaud et€al. (2004).

higher, equipped with oxidation catalysts or particle traps incorporating oxidation catalysts (Sjödin et€al., 2009).

18.3.3╇ Ozone Rural O3 concentrations had doubled from about 10–15 ppb in rural Europe at the end of the nineteenth century to 20–30 ppb in the 1980s. Volz and Kley (1988) were able to show that by analysing the more than 100 years old ozone data from around the turn of the nineteenth and twentieth centuries made at Montsouris by old-day measurement methodologies and comparing it with more modern techniques. Since the 1980s, rural O3 concentrations have increased in many areas (Staehelin and Schnadt Poberaj, 2008) with quite different rates of change at different locations (The Royal Society, 2008). The majority of ozone precursor emissions originate from anthropogenic sources. A recent review of the Gothenburg Protocol (TFIAM, CIAM, 2007) showed that the emissions of the O3 precursors NOx and NMVOC have declined substantially as a result of emissions controls. In 2005, NOx and NMVOC emissions were 30% and 38% lower than 1990 levels for the European countries within the Protocol. Figure 18.5 Annual mean concentration map of NO2 (μg/m3) in 2008; the two highest concentration classes correspond to the limit value (40 μg/m3) and limit value plus margin of tolerance (44 μg/m3), respectively (Mol et€al., 2010).

419

Nitrogen as a threat to European air quality

The decrease in NOx and NMVOC emissions in Europe has resulted in a reduction in the magnitude of short-term peak O3 concentrations during episodes, with declines in daily peak concentrations of around 30 ppb. Reductions in peak O3 concentrations have been observed widely in Europe, both in urban and rural areas. The temporal pattern of O3 concentrations, however, reveals several additional changes during the period in which emissions of O3 precursors over Europe have declined. In particular, the lower percentiles of the �frequency

Figure 18.6 Relative changes in annual mean concentrations of NO2 and NOX to year 2007 for the three stations types of the AirBase network:€rural (a), urban (b) and traffic (c) (data from Mol et€al., 2009).

(a)

distribution and even the mean concentration at many sites have been growing. These effects are illustrated in Figure 18.7a and Figure 18.7b, from Jenkin (2008). This figure shows the changes in O3 concentrations at an urban site (Leeds) and a rural site (Lullington Heath) in the UK, and show similar trends to those observed at other sites across the UK, and more widely in Northern Europe (The Royal Society, 2008).

Effects on health The highest levels of exposure to ozone are estimated for southern Europe, with the highest levels found in northern Italy. The estimated population exposure indicates that large regions fail to meet environmental objectives and a notable fraction of the urban population, typically around 25%, is exposed to elevated ozone. Extreme conditions in 2003 pushed this to approximately 60%. Regional differences in exposure levels across Europe are shown in Table 18.10. These are expected to diminish in the next decade. Exposures in continental Europe are projected to fall by 20%–30% in southern France, Germany, northern Italy and Switzerland and to rise in the United Kingdom and Scandinavia (WHO, 2008). It can be expected that population exposure will increase regardless of currently planned precursor emission reductions, owing to increasing background levels and reduced ozone depletion in urban areas. Current exposures to ozone in Europe are associated with premature mortality and morbidity. Effects include 21 000 premature deaths, 14 000 hospital admissions for respiratory disease and more than 100 million person-days of restricted activity per year in the EU 25 (WHO, 2008). These figures are underestimates, as they do not account for possible effects at levels below 70 μg/m3. The heat wave of summer 2003 gave an indication of the potential impact of a future warmer climate on air-pollution related health effects. Bell et€al. (2007) investigated how climate

(b) 350

200 Leeds Centre

300

160

250

120

95 %-ile 90 %-ile

80

75 %-ile 40

50 %-ile 25 %-ile 10 %-ile 5 %-ile 1995

2000

2005

[ozone] (µg m–3)

[ozone] (µg m–3)

max

0 1990

Lullington Heath

max

200 150 120

95 %-ile 90 %-ile 75 %-ile 50 %-ile 25 %-ile 10 %-ile 5 %-ile

100 80 60 40 20 0 1990

1995

2000

2005

Figure 18.7 Changes in ground-level O3 concentrations at (a) urban (Leeds centre) and (b) rural (Lullington Heath) sites in the UK, showing a decline in peak values and increases in the mean and lower percentiles of the distribution. Trend in hourly mean O3 distributions based on data over the periods 1993–2006 and 1990–2006, respectively. The solid lines are linear regressions of data indicating the average trend over the period (Jenkin, 2008). With permission from Elsevier.

420

Jana Moldanová, Peringe Grennfelt and Åsa Jonsson Table 18.10 Fractions of populations exposed to ozone levels exceeding the EU directive target value of 120 μg/m3 for more than 25 days a year, by region (WHO, 2008)

Regiona

2002

2003

Northern Europe

0%

0%

Northwestern Europe

0–10%

40–50%

Central and eastern Europe

20–30%

80–90%

Southern Europe

60–70%

60–70%

a

 orthern Europe:€Denmark, Estonia, Finland, Iceland, Latvia, Lithuania, N Norway, Sweden; Northwestern Europe:€Belgium, France (north of 45ºN), Ireland, Luxembourg, Netherlands, United Kingdom; Central and eastern Europe:€Austria, Czech Republic, Germany, Hungary, Poland, Slovakia, Switzerland; Southern Europe:€Cyprus, France (south of 45ºN), Greece, Italy, Malta, Portugal, Slovenia, Spain.

change could affect ambient ozone concentrations, using an hourly concentration model for 50 United States cities for 1990 and 2050. Future concentrations were based on the IPCC A2 scenario and the impact of altered climate on ozone was estimated. The maximum 1 h ozone levels were estimated to increase on average by almost 10 μg/m3 (maximum 19.2 μg/m3), the highest increases occurring in cities with current high pollution levels.

Effects on vegetation Simpson et€al. (2007) used the EMEP chemical transport model to map the different indicators of ozone damage across Europe for two illustrative vegetation types, wheat and beech forests. Figure 18.8a illustrates the AOT40 index for forests as calculated with the EMEP model for the year 2000, with Figure 18.8b showing the ratio of this AOT40 value to the recommended critical level (CL), 5000 ppb.h. Firstly, we can note that the spatial gradients of AOT40 are very large, with typically a factor 10 difference between AOT40 values in southern Europe and those in the Nordic countries. Exceedance of the 5000 ppb.h CL for AOT40 is widespread, with only a few areas (mainly in Northern Europe) experiencing lower values. Relative exceedances of the CL of more than a factor of 10 occur in southern Europe (Figure 18.8b). Figure 18.8c,d present the corresponding results for AFst1.6, the relevant flux-based statistic for deciduous forests. The spatial pattern of AFst1.6 is rather different from that of AOT40. Whereas AOT40 clearly shows maxima in southern Europe, with much lower values in the Nordic countries, the spatial gradients in AFst1.6 are much smaller. Although the highest AFst1.6 values are still seen in parts of southern Europe, the difference between the Mediterranean and southern Sweden or Finland is typically less than a factor of two. Indeed, for the great majority of Europe, AFst1.6 values lie between 8 and 16â•›mmol/m2. The suggested CL of 4 mmol/m2 for AFst1.6 seems to be exceeded over essentially all of Europe. As noted in the Royal Society’s synthesis on ozone (The Royal Society, 2008), there is a substantial body of evidence from North America and Europe, supported by some work in Asia, Africa and Latin America, that elevated O3 levels cause reductions in the yield of sensitive crop species (Mauzerall and

Wang, 2001; Emberson et€al., 2003), and some estimates have been made of the economic impacts of crop loss due to ambient O3 levels. The annual cost of arable crop production lost due to O3 was estimated to be $2–4 billion in the USA in the 1980s, with an equivalent estimate for the EU of €6.7 billion (90% confidence interval €4.4–9.3 billion per year) in 2000 (Holland et€al., 2006). This is equivalent to 2% of arable agricultural production, but does not account for a range of other effects, including those on crop quality, visible injury, and susceptibility to pests and diseases. The greatest economic losses in Europe were predicted to be in Mediterranean countries, together with France and Germany, because the assessment used a concentrationbased exposure index. Wheat, tomatoes, vegetables and potatoes were the crops with the greatest yield losses. Van Dingenen et€al. (2009) recently provided the first global estimate of crop yield loss for four major commodities (wheat, rice, maize, soybean), of $14–26 billion in the year 2000. This is significantly higher than present day losses to crops projected to occur as a result of climate change. The Royal Society (2008) report further noted that all of these estimates are primarily based on data from field chamber experiments which may under-estimate the real effects of O3 in the field. For example, the decrease in soybean yield under open-air conditions in the Soy Free Air Concentration Enrichment (FACE) experiment, conducted in the USA, was greater than predicted by a synthesis of previous chamber studies (Morgan et€al., 2006). In one season, this was partly because O3 exposure increased the impact of a major defoliating hail event. More field release experiments, in which O3 is released over a crop which is not enclosed in chambers, are therefore needed to reduce the uncertainty in future estimates of loss in crop productivity. These need to be in a range of locations and to cover different cropping systems. According to EEA (2007) large parts of the EEA 32 countries currently exceed exposure criteria for forests. More than half of the agricultural area exceeds criteria for crop protection, total crop yield losses reaching an estimated €3 billion per annum in 2000. Since 2000 the exposure of crops has not been reduced. Furthermore, in a number of areas ozone concentrations have actually increased in recent years as can be seen in Figure 18.9. Adverse meteorology and the changing balance of airborne pollutants lie behind this.

18.3.4╇ PM Particles have been addressed as being one of the most important air pollutants regarding adverse human health impacts. The largest health impact estimates are for long term effects of PM2.5. De Leeuw and Horalek (2009) estimated the number of premature deaths in EU-27 countries from PM2.5 to be almost 500 000. The number of deaths that can be attributed to long-term PM10 exposure in the EU-25 countries is about 350 000– 370 000 premature (CEC, 2005; EEA, 2009a,b) (impacts from PM10 and PM2.5 are not additive). Figure 18.10 shows the fraction of the European population that is exposed to a certain concentration range of PM10 (both annual mean and the 36th highest daily mean) and PM2.5. WHO (2006) set an air quality guideline

421

Nitrogen as a threat to European air quality (a)

(b)

100

100 60000

10

40000

80

80

8

20000

6

10000

60

60

4

5000 3 3000

40

40

2

1000

1.5

500

20

20 1.2

200

40

60

80

100

120

40

(c)

60

80

100

120

(d)

100

100 24 16

80

10 80

8

12 8

60

6 60

4

4 2

40

3 40

2

1 0.5

20

0.2

40

60

80

100

120

1.5 20 1.2

40

60

80

100

120

Figure 18.8 Calculated metrics for ozone damage to forests as calculated with the EMEP model for the year 2000:€(a) AOT40 values (ppb.h), (b) ratio of calculated AOT40 to critical level of 5000╛ppb.h, ( c) AftY1.6 values (mmol/m2), (d) ratio of AfstY1.6 to critical level of 4 mmol/m2 (from Simpson et€al., 2007).

of 10 μg/m3, the lowest level at which total, cardiopulmonary and lung cancer mortality have been shown to increase with confidence in response to PM2.5. Only about 9% of the population is exposed to concentration below this guideline. Besides the guideline, the WHO has defined three interim targets, 15, 25 and 35 μg/m3. The air quality guideline for PM10 is 20 μg/ m3 (WHO, 2000). About ¾ of the total European population is exposed to concentrations that are above this guideline for annual mean of PM10. Figure 18.11a shows the annual mean concentrations of PM10. A statistical analysis of the monitoring data indicated that the daily PM10 limit value corresponds with an annual mean of 31 μg/ m3, although regional differences may occur (Mol et€al., 2010, and references there in), so both the exceedances of the annual limit

422

value and of the short-term (daily) limit value can be derived from the figure. The map indicates that both limit values have been exceeded in many countries across Europe. Figure 18.11b shows the annual mean concentrations of PM2.5 and enables a comparison with the PM2.5 target value of 25 μg/m3. Data from the AirBase network show exceedance of PM10 limit values of both daily and annual means at all types of stations with increasing numbers from rural background to urban background to traffic stations. The extent of exceedance of the daily limit value is larger than of the limit value for annual mean. The daily limit value is frequently exceeded at urban background stations (about 28% of stations) and at traffic stations (more than 32% of stations) (Mol et al., 2010). Regarding PM2.5, the AirBase data show that at 6%, 14%, 5%

Jana Moldanová, Peringe Grennfelt and Åsa Jonsson O3 AOT40 (May-July)

30000

(a)

PM10 annual mean

AOT40 ( g/m3.h)

24000 18000

EU Target level

<10 g/m3 10-20 g/m3

12000 6000

20-40 g/m3

EU long-term objective

40-45 g/m3 >45 g/m3

0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

Figure 18.9 Annual variation in the ozone AOT40 value (May–July in μg/ m3.h). Average values over all AirBase rural stations which reported data over at least six years in the period 1996–2006. The orange line corresponds to the 5 year averaged value (EEA, 2010b).

and 10% of the rural, (sub)urban background and traffic stations and industrial sites the PM2.5 target value of 25 μg/m3 has been exceeded (Mol et al., 2010). There are regional and seasonal variabilities in PM10 and PM2.5. In the Mediterranean region high concentrations of PM are often associated with inter-continental transport during the spring/summer (Saharan dust and Asian continental outflow) and it is therefore more common that PM concentrations are exceeded during spring and winter. For the rest of Europe, where PM is dominated by regional sources, winter time is the most common season for high PM concentrations which are a combination effect of higher emissions and less effective dispersion in wintertime. The annual average PM concentrations in continental Europe are shown in Table 18.11 for different environments. Highest concentrations are often measured at the road side. The Europe-wide tendencies in annual mean PM10 concentrations for the time period 1997–2007 are shown in Figure 18.12. The following observations can be made:€ PM10 concentrations in 2004 were approximately 25% lower than in 1997. However, an actual net tendency in PM10 concentrations from 1997 onwards cannot be discerned due to the large inter-year variations over the entire period. Variations in meteorological conditions between years can explain part of these variations. Urban and rural background concentrations trends follow each other closely. The rural background concentration provides the dominating contribution to total urban PM10. PM10 concentrations at street level are on average approximately 8 μg/m3 higher than the average concentrations measured at 301 urban stations in 19 countries. Figure 18.S1 in supplement shows map with loss in statistical life expectancy attributable to the exposure to PM2.3 which is highest in central Europe with statistical life expectancy loss between about 6 and 36 months.

Chemical composition of particulate matter Nitrate and ammonium contribute significantly to the particulate matter in Europe. From the EMEP measurements (rural stations, EMEP, 2008), NO3− and NH4+ contributed by between 6%–19% and 5%–9% respectively to the PM10. This is in line with earlier assessments by Putaud et€al. (2004). The

(b)

PM10 36th highest daily mean

<20 g/m3 20–30 g/m3 30–50 g/m3 50–65 g/m3 >65 g/m3

(c)

PM2.5 annual mean

<10 g/m3 10-15 g/m3 15-25 g/m3 25-35 g/m3 >35 g/m3

Figure 18.10 Exposure of European inhabitants in year 2005 to (a) PM10, annual mean value; (b) PM10, 36th highest daily mean value; (c) PM2.5, annual mean value concentrations (sources:€(a) and (b) EEA, 2009a; (c) de Leeuw and Morálek, 2009).

relative quantity of nitrate tends to increase with increasing PM10 resulting in a higher relative contribution on exceedance days (daily average above 50 μg/m3) compared to annual averages. Putaud et€al. (2004) found that on these days nitrate was a major component of PM10 and PM2.5 together with organic matter. The more than proportional rise in nitrate levels as a function of PM10 appears to be valid for the whole of north-

423

424 (b)

Figure 18.11 (a) Annual mean concentration map of PM10 (μg/m3); the two highest concentration classes corresponds to the annual limit value (40 μg/m3) and to a statistically derived level (31 μg/m3) corresponding to the short-term limit value). (b)€Annual mean concentrations of PM2.5 (from Mol et al., 2010).

(a)

Jana Moldanová, Peringe Grennfelt and Åsa Jonsson Table 18.11 Average PM10 and PM2.5 levels in Europe for 2002 (WHO, 2006)

Type of PM value PM10 annual average

PM10 daily average

PM2.5 annual average

μg/m3

Site

21.7

rural

26.3

Urban background

32

In street

38.1

rural

43.2

Urban background

51.8

In street

11–13

Rural background

15–20

Urban background

20–30

In street PM10, Traffic PM10, Urban PM10, Rural

125 100 75 50 1996

1998

2000

2002

2004

2006

2008

Figure 18.12 Inter-annual variation of PM10, 1997–2007 annual means (from EEA, 2009a, b).

western Europe. The reason for the extra high nitrate levels on exceedance days is that these days are often associated with very stagnant conditions. Ammonium nitrate, which has a more local character than sulphate, can build up fast, whereas the levels of pollutants from long range transport are not specifically enhanced. There are no clear long-term trends in NO3− concentrations. NO3− concentrations do not follow general PM10 and PM2.5 trends (Putaud et€al., 2004). The highest concentrations found in Europe are in the Po Valley, where there are large concentrations of NOx as well as NH3. Some of the smallest concentrations of NO3− are found in rural and natural backgrounds that may be due to lack of local sources of NOx. However, the concentrations near cities are often higher than at kerbside sites, which may be due to the time needed to form NO3− from NOx or due to low concentrations of NH3 (Figure 18.13).

18.4╇ Future perspectives, national ceilings 18.4.1╇ Ammonia In its Thematic Strategy on Air Pollution (TSAP), the European Commission outlined the strategic approach towards cleaner air in Europe (CEC, 2005) and established interim environmental objectives for the year 2020. Current emission ceilings for ammonia that should be met in 2010 were developed by 2006 (Klimont et€al., 2007). In order to meet new objectives for eutrophication, acidification and for particulate matter the following policies were suggested. (1) A possible extension of the Integrated Prevention and Pollution Control (IPPC) Directive, to include installations

Figure 18.13 Mean contributions of NH4+ and NO3− ions to PM10 (a) and PM2.5 (b) at 16 measurement sites evaluated by Putaud for time period 1991–2001 and in case of PM10 also at EMEP sites in 2006 (12 sites with NH4+ data and 25 sites with NO3− data, EMEP 2008).

for intensive cattle rearing and a possible revision of the current thresholds for installations for the intensive rearing of pigs and poultry. (2) In the context of the current rural development regulation and the Commission proposals for rural development for 2007–2013, the Commission encourages the Member States to make full use of the measures related to farm modernisation, meeting standards and agro-environment to tackle ammonia emissions from agricultural sources. The cost-effective emission ceilings for ammonium that would lead to achievement of the agreed interim objectives in 2020 and that are based on energy projections corresponding to the recent Climate and Energy Package of the European Commission and the national projections of agricultural activities were examined by Amann et€al. (2008). The ‘Current policy’ (CP) scenario was adopted as the starting point for the optimisation of additional emission-control measures to achieve the TSAP objectives. The ‘Current policy’ scenario considers implementation rates of already decided emission control legislation as currently laid down in national laws and implementation of the recent Commission proposals on the introduction of EURO-VI standards for heavy duty vehicles (EC, 2009) and on the revision of the Integrated Pollution Prevention and Control (IPPC) Directive for large stationary sources (EC, 2008a). To achieve the environmental objectives of the TSAP in 2020, further emission reduction measures on top of those considered in the ‘Current policy’ scenario would be needed to increase the reduction efforts for NH3 emissions from 8% to 22%. Figure 18.14a shows the national emissions of NH3 (Gg/y) in 2007, the national emissions ceilings for year

425

Nitrogen as a threat to European air quality (a)

800 2007 NEC 2010

600

NEC 2020

Gg NH3

TSAP 2020 400

200

Germany

Finland

Spain

EU-27/10

Germany

Finland

Spain

EU-27

Austria

Romania

Latvia

Greece

France

Italy

Slovakia

UK

Denmark

Sweden

Lithuania

Netherlands

Poland

Estonia

Ireland

Slovenia

Belgium

Luxembourg

Hungary

Malta

Bulgaria

Cyprus

Portugal

Czech Rep.

0

NEC 2010

(b)

Figure 18.14 (a) National emissions of NH3 (Gg/y) in 2007 (EMEP, 2010), the national emissions ceilings for year 2010 (NEC 2010) and the ‘Current policy’ scenario for 2020 for the 27 EU MemberStates. The scenario with measures optimised to achieve the TSAP objectives in 2020 (TSAP 2020, the cost-effective scenario) is also shown. The EU-27 emission totals are divided by 10. (b ) Distances of the year-2007 national NH3 emissions from values for year 2007 on the straight-line emission trajectories from year 2000 to year 2010 (NEC 2010, brown) and to year 2020 (‘Current policy’ scenario, blue), respectively, expressed as difference in percent of the year2007 value on the respective emission trajectory. The areas with green smiles indicate EU countries with NH3 emissions that in 2007 were below the emission trajectory to NEC 2010 (afore the brown line) and to ‘Current policy’ 2020 (blue line), respectively. The total EU-27 NH3 emissions in year 2007 were below the emission trajectory to NEC 2010 but above the emission trajectory to ‘Current Policy’ 2020. All emission scenarios come from Amann et€al. (2008).

"Current policy" 2020 40% 20% 0% –20%

2010 and the ‘Current policy scenario’ for 2020, representing suggested NEC 2020, for the 27 EU Member States’. Distances of the year-2007 national NH3 emissions from the emission trajectories going from the individual Member States emissions in 2000 to their respective NEC values are shown both for the NEC for 2010 and for the suggested NEC for 2020 in Figure 18.14b. It can be seen that only five Member States are above their emission trajectories to the 2010 NEC while more than a third are above their trajectories to the 2020 NEC. According to the national projections for 2010 published in the NEC directive status report for 2008 (EEA, 2009b) will only two Member States, Germany and Spain, miss their national ceilings.

18.4.2╇ Oxides of nitrogen The decrease in NOx emissions during the 1990s is partly due to the introduction of emission standards for road transport in Europe that included both light-duty vehicles (petrol and �diesel) and heavy duty vehicles, but also regulations targeting other sectors have contributed. In May 2005 the 1999 Gothenburg Protocol came into force regarding emission ceilings for (inter

426

Austria

Romania

Greece

Latvia

France

Italy

Slovakia

Denmark

UK

Sweden

Lithuania

Netherlands

Poland

Estonia

Ireland

Slovenia

Luxembourg

Belgium

Hungary

Bulgaria

Malta

Cyprus

Czech Rep.

–60%

Portugal

–40%

alia) NOx. Compared to the level of 1990, Europe’s emissions of NOx should by 2010 be reduced by 41% (VOC 40%, NH3 17%). Additionally, the EU introduced a national emission ceilings (NEC) Directive, 2001/81/EC, concerning NOx and other pollutants. Compared to the Gothenburg Protocol, this directive puts more pressure on some of the member states than others. The emission ceilings must be attained by 2010. While land-based NOx emissions are expected to decrease in the coming decades, emissions from shipping are projected to increase. In 2000 the emissions from international maritime shipping in the seas surrounding the European Union (i.e. Baltic, North Sea, Northeast Atlantic, and Mediterranean Sea) amounted to approximately 30% of the land-based emissions in the EU-25. Legislation is in force to control emission of SO2 and to some extent also NOx from international shipping; however, Annex 6 of the MARPOL protocol regulates NOx emissions only on newly installed engines after 2016. Considering the long lifetime of ship engines, this legislation will impact the NOx emissions only in distant future. Further, the expected increase in the volume of ship movements will compensate for the environmental benefits of these measures and will lead to

Jana Moldanová, Peringe Grennfelt and Åsa Jonsson 12 Land-based sources, CP scenario

Mt NOx/year

10

Land-based sources, TSAP target International shipping baseline

8 6 4

International shipping MTFR

2

National shipping baseline

0 2000

2005

2010

2015

2020

Figure 18.15 Emissions of nitrogen oxides from shipping (baseline scenario considering MARPOL Annex 6) compared with the emissions from landbased sources in the EU25 (‘Current policy scenario’). Emissions at Maximal Technology-Feasible Reduction are shown for 2020 (after Cofala et€al., 2007).

a continued growth in ship emissions. By 2020 emissions from maritime activities would come close to the projected baseline emission levels from land-based sources and surpass the target levels established by the European Commission for land-based sources (Figure 18.15). Airport studies confirm that aircraft continue to be a relatively small contributor to regional pollution although aircraft-related NOx contributions could increase as air traffic increases and other non-aircraft emission sources become progressively cleaner. The European Union have decided on a 10% substitution of traditional fuels in the road transport sector (petrol and conventional diesel) by alternative fuels before the year 2020 (Directive 2003/30/EC) in order to meet the challenges with increased transportation, decreased oil resources and enhanced greenhouse gas emissions. Impact of this directive on the NOx emissions have not been fully investigated yet as the emission factors for many alternative fuels and their blends are not known. Up-coming Euro 5 and 6 standards (EC 715/2007, 1999/96/EC), embodying after-treatment technologies will be implemented 2010–2015 with the aim of further reductions of the NOx emissions. The usage of diesel is increasing in Europe and this may have implications on the NO/NO2 ratio as diesel vehicles, especially when equipped with modern technology such as diesel particulate filters, emit more NO2 compared to petrol vehicles. This effect should be pronounced only close to the source region. In some regions it is evident today that NO2 is not decreasing at the same pace as NOx. Combined measurement and modelling efforts have not been able, however, to prove that the potentially increasing share of NO2 in NOx emissions due to the emissions from diesel cars could alone explain this trend. Still, in future assessment strategies concerning vehicle emissions and NOx this has to be taken into account. Additionally, it takes time to change the on-road fleet, hence, even though the latest vehicle technology means reduction in some pollutants, it may take time before that becomes a reality. Projections of future NOx emissions and their impact on future air pollution situation in Europe have been assessed by Amann et€ al. (2008) with GAINS model. The ‘Current policy’ scenario, i.e. scenario implementing all agreed legislation, projects a decrease of emissions of NOx by about 48% in 2020 compared to the year 2000. This scenario approaches emission

levels of the National Emission Ceilings. These emission ceilings must be attained by the EU member states in 2010. In Figure 18.16 is shown comparison of the NEC values for the Member States to the national emission values reported to EMEP for the year 2007. Comparison with target values of the TSAP is shown as well. Figure 18.16 indicates that almost half of the Member States were in 2007 above their emission trajectories leading from their year-2000 NOx emissions to their NEC and large majority were above their respective trajectory to the planned NEC for 2020, in figure represented by the ‘current policy’ scenario for 2020 (scenarios from Amann et€al., 2008). The national projections for 2010 published in the NEC directive status report for 2008 (EEA, 2009b) give a similar picture of attainment of the 2010 NEC in the EU-27 Member States.

18.4.3╇ Ozone Owing to the presence of stringent emission control legislation, ozone precursor emissions are expected to decline in the EU over the coming decade. VOC emissions are expected to decrease in EU15 by 33% in 2010 and 41% in 2020 compared to 2000. However, a lack of equivalent legislation will not prevent further increases in precursor emissions in other countries that are Parties to the Convention on LRTAP. This growth in emissions is expected to increase hemispheric ozone background concentrations. Furthermore, climate change could lead to higher biogenic emissions in the future (The Royal Society, 2008). At the urban scale, projected O3 precursor emissions controls (especially reductions in NOx emissions) and changes in background O3 concentrations will lead to changes in urban O3 concentrations, with potentially large increases by the end of the century depending on the future scenario. As described earlier, the changes in urban O3 concentrations result partly from reductions in the titration of urban O3 by reaction with emitted NO, but are also influenced by changes in the background O3 concentrations. This is because the reduction in the titration effect increases O3 concentrations up to the background level (Jonson et€al., 2006). Figure 18.17 shows projected exposure trends as expected changes in SOMO35 for 2020 compared to similar calculations with 2000 emissions. In some areas, such as parts of the Russian Federation and Scandinavia, SOMO35 levels are seen to increase in 2020 compared to 2000, although absolute levels are relatively low (around 2000 μg/m3·days) in both cases. In those areas that had high levels of SOMO35 in 2000 (e.g. Italy and much of southern Europe), 2020 levels are seen to be significantly lower than 2000 levels, although this still leaves levels of around 4000–5000 μg/m3 days in these areas (WHO, 2008). Reduction in ozone exposure resulting from current policies, and thus in the health impact by 2020, is estimated to be small. Population ageing will increase susceptible groups and background risks in Europe in the foreseeable future.

18.4.4╇ PM Projections for the future have been based on current legislation, i.e. ‘business–as-usual’, and emissions of PM2.5 are

427

Nitrogen as a threat to European air quality

(a)

1600 2007

1400

NEC 2010

1200

NEC 2020 TSAP 2020

Gg NOx

1000 800 600 400 200

Malta

Austria

EU-27

Austria

EU-27

France

Spain

Ireland

France

Ireland

Malta

Belgium Belgium

Spain

UK

Greece

UK

Greece

Italy

Denmark Denmark

Poland

Italy

Lithuania

Germany

Latvia

Bulgaria

Hungary

Romania

Czech

Cyprus

Finland

Estonia

Portugal

Sweden

Slovakia

Slovenia

Netherlands

Luxembourg

0

Figure 18.16 (a) National emissions of NOx (Gg/year) in 2007 (EMEP, 2010), emissions ceilings for year 2010 (NEC 2010) and ‘Current policy’ scenario for 2020 (representing suggested NEC 2020) for the 27 EU Member States. The scenario with measures optimized to achieve the TSAP objectives in 2020 (TSAP 2020, the cost-effective scenario) is also shown. The EU-27 emission totals are divided by 10. (b) Distances of the year 2007 national NOx emissions from values for year 2007 on the straight-line emission trajectories from year 2000 to year 2010 (NEC 2010, brown) and to year 2020 (‘Current policy’ 2020, blue), respectively, expressed as difference in percent of the year-2007 value on the respective emission trajectory. The areas with green smiles indicate EU countries with NOx emissions that in 2007 were below the emission trajectory to NEC 2010 (before the brown line) and to ‘Current policy’ 2020 (blue line), respectively. The year 2007 NOx emissions of Germany, Denmark and Belgium that are above the NEC 2010 trajectory fall below the emission trajectory for ‘Current policy’ 2020. All emission scenarios come from Amann et€al. (2008).

NEC 2010

(b) "Current policy" 2020 75% 50% 25% 0%

Poland

Lithuania

Germany

Bulgaria

Latvia

Hungary

Romania

Cyprus

CzechRep.

Estonia

Finland

Portugal

Slovakia

Sweden

Slovenia

Netherlands

–50%

Luxembourg

–25%

Figure 18.17 Calculated changes in SOMO35 values (μg/m3 × days) by 2020 compared with 2000 emissions (meteorology from 1997) (WHO, 2008).

428

projected to decrease by about 45% in 2020 compared to the year 2000. The main reason for this projected decline is expected to be improved vehicle technology, implemented in Euro standards 5 and 6. The PM10 is projected to decrease by 39% for the same time period which is mainly as a result of reductions in the transportation sector and power generation (CEC 2005). Subsequently, the projected life years lost due to PM2.5 is decreased by around 32% and premature deaths by 21% (CEC 2005). However, PM2.5 is still projected to account for 271â•›000 premature deaths in 2020. Table 18.12 summarises the estimated effects of air pollution on health. For 2020 it is projected that PM will cause less harm regarding both mortality and morbidity. Loss in statistical life expectancy attributable to the exposure to PM2.5 in Europe projected for year 2020 in the current policy scenario is shown together with the year 2000 situation in Figure 18.S1 in the supplementry material. Taking the projected small reductions in ammonia comparing to SO2 and NOx, it can be expected that contribution of ammonia to the secondary (and total) PM will increase.

Jana Moldanová, Peringe Grennfelt and Åsa Jonsson Table 18.12 Estimations of the effect of air pollution on health for the years 2000 (baseline) and 2020 (Current legislation in 2020 including Climate Policy) (Pye€and Watkiss 2005).

Health end-point

Units (1000)

Pollutant

2000

2020

Acute mortality

Premature deaths

O3

21,4

20,7

Respiratory hospital admissions Minor Restricted Activity Days (MRADs)

Cases

O3

14

20

Days

O3

53,924

42,227

Respiratory medication use (Children)

Days

O3

21,413

12,897

Respiratory medication use (Adults)

Days

O3

8,837

8,136

Cough and ╇ lower respiratory symptoms (children)

Days

O3

108,056

64,955

Chronic mortalitya

Life years losta

3,001

1,900

Chronic mortality

Premature deaths

PM

288

208

Infant mortality

Premature deaths

PM

0.6

0.3

Chronic bronchitis

Cases

PM

135,7

98,4

Respiratory hospital admissions

Cases

PM

51,4

32,6

Cardiac hospital admissions

Cases

PM

31,7

20,1

Restricted activity days (RADs)

Days

PM

288,292

170,956

Respiratory medication use (children)

Days

PM

3,510

1,549

Respiratory medication use (adults)

Days

PM

22,990

16,055

Cough and lower respiratory symptoms (children)

Days

PM

160,349

68,819

Lower respiratory symptoms ╇ (adults with chronic symptoms)

Days

PM

236,498

159,724

a

a

PM a

N  ote two alternative metrics are used for the presentation of chronic mortality from PM. Firstly in terms of years of life lost and secondly in terms of numbers of premature deaths. These are not additive.

18.5╇ Conclusions As shown in this chapter, nitrogen air pollution is a serious threat to human health and ecosystems, both through direct effects from the emitted compounds but more seriously through the secondary compounds formed in the atmosphere. Legislation to improve air quality is in place which results in improving air quality, however, there are still large exceedances of air quality standards and critical levels in Europe. Even if present and proposed legislation is fully implemented, emissions of nitrogen oxides and ammonia will still pose a problem in Europe. • Air quality standards for nitrogen dioxide are exceeded in many urban areas in Europe. The exceedance is primarily associated with the local traffic. The year-2007 NO2 emissions of 12 out of 27 EU Member States were above the emission trajectory from year 2000 to the NEC for 2010. • NEC for ammonia will be met in 2010 by most of the Member States; the ‘Current policy’ scenario, i.e. the scenario where all already agreed legislation has been implemented, is, however, above the emission level that would meet the TSAP objectives for 2020. It is very likely that sensitive species are and will be negatively affected by emissions of ammonia almost everywhere in Western, Central and parts of Southern Europe, at least in areas with intensive animal husbandry.

• There is still a need for better data on emissions and atmospheric concentrations, in particular for ammonia. • Reduction in PM2.5 mortality in 2020 for the ‘Current policy’ scenario gives lower than the TSAP goal for 2020, which is 47% reduction of the long-term exposure mortality taking year 2000 as a base. Considering the high contribution of Nr to PM2.5, further reductions of Nr emissions could have a positive effect on reducing PM-related health effects and would contribute to meet the TSAP goal. Further, deposition of Nr in Europe also leads to exceedances of critical loads for nutrition nitrogen in large parts of Europe, both under the current situation and under the ‘Current policy’ scenario in 2020. • The new findings on long-term exposure effects of ozone on mortality with the relative risk increase per additional 10 μg/m3 ozone adding on the relative risk increase per 10 μg/m3 PM2.5 would make it difficult to meet the TSAP target of 47% reduction in mortality. As only small decreases are projected for ozone concentrations, the mortality from ozone would remain high in the ‘Current policy’ scenario in 2020. • The uncertainties of the health impacts of nitrogen species (NO2, NH4+, NO3−, HNO3 and others) are very large and Â�further epidemiological and toxicological research is needed to obtain more reliable exposure–response functions. In particular, the damage cost of NH3 emissions is extremely uncertain.

429

Nitrogen as a threat to European air quality

Acknowledgements The authors gratefully acknowledge support from the Nitrogen in Europe (NinE) programme of the European Science Foundation, from the NitroEurope IP funded by the European Commission and from the COST Action 729.

Supplementary materials Supplementary materials (as referenced in the chapter) are� available online through both Cambridge University Press:€ www.cambridge.org/ena and the Nitrogen in Europe website: www.nine-esf.org/ena.

References Amann, M., Bertok, I., Cofala, J. et€al. (2008). National Emission Ceilings for 2020 Based on the 2008 Climate and Energy Package, NEC Scenario Analysis Report No. 6. International Institute for Applied Systems Analysis (IIASA), Laxenburg, Austria, http:// www.iiasa. Armstrong, B., Hutchinson, E., Unwin, J. and Fletcher, T. (2004). Lung cancer risk after exposure to polycyclic aromatic hydrocarbons. Environmental Health Perspectives, 112, 970–978. Bell, M., Peng, R. and Dominici, F. (2006). The exposure–response curve for ozone and risk of mortality and the adequacy of current ozone regulations. Environmental Health Perspectives, 114, 532–536. Bell, M. L., Goldberg, R., Hogrefe, C. et€al. (2007). Climate change, ambient ozone, and health in 50 US cities. Climate Change, 82, 61–76. Benton, J., Fuhrer, J., Gimeno, B. et€al. (2000). An international cooperative programme indicates the widespread occurrence of ozone injury on crops. Agriculture, Ecosystems and Environment, 78, 19–30. Butterbach-Bahl, K., Gundersen, P., Ambus, P. et€al. (2011). Nitrogen processes in terrestrial ecosystems. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Bylin, G. and Forsberg, B. (2009). Utmhusmiljön: påverkan på allergi och astma. In: Allergi och astma, ed. G. Hedlin and K. Larsson, Studentlitteratur, Lund, Sweden, pp. 159–174. CEC (2005). The Communication on Thematic Strategy on Air Pollution and the Directive on Ambient Air Quality and Cleaner Air for Europe. Impact Assessment. Sec. (2005) 1133. http://ec.europa.eu/governance/impact/ia_carried_out/docs/ ia_2005/sec_2005_1133_en.pdf. Cape, J. N., Van der Eerden, L. J., Sheppard, L. J., Leith, I. D. and Sutton, M. A. (2009). Evidence for changing the critical level for ammonia. Environmental Pollution, 157, 1033–1037. Cofala, J., Amann, M., Heyes, C. et€al. (2007). Final Report:€Analysis of Policy Measures to Reduce Ship Emissions in the Context of the Revision of the National Emissions Ceilings Directive, International Institute for Applied Systems Analysis (IIASA),€Laxenburg, Austria, www.iiasa.ac.at. de Leeuw, F. and Horálek, J. (2009). Assessment of the Health Impacts of Exposure to PM2.5 at a European Level, ETC/ACC Technical Paper No. 2009/1. Diociaiuti, M., Balduzzi, M., De Berardis, B. et€al. (2001). The two PM(2.5) (fine) and PM(2.5–10) (coarse) fractions:€evidence of different biological activity. Environmental Research, 86, 254–262. Dise, N. B., Ashmore, M., Belyazid, S. et€al. (2011). Nitrogen as a threat to European terrestrial biodiversity. In:€The European Nitrogen

430

Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Durand, P., Breuer, L., Johnes, P. et€al. (2011). Nitrogen processes in aquatic ecosystems. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. EC (1994). Directive 94/12/EC of the European Parliament and the Council of 23 March 1994 relating to measures to be taken against air pollution by emissions from motor vehicles and amending Directive 70/220/EEC. http://ec.europa.eu/enterprise/sectors/ automotive/documents/directives/motor-vehicles/index_en.htm EC (1996a). Directive 1996/61/EC on Integrated Pollution Prevention and Control (ICCP). http://eur-lex.europa.eu/LexUriServ/ LexUriServ.do?uri= CELEX:31996L0061:EN:NOT EC (1996b). Directive 96/69/EC of the European Parliament and of the Council of 8 October 1996 amending Directive 70/220/EEC on the approximation of the laws of the Member States relating to measures to be taken against air pollution by emissions from motor vehicles. http://ec.europa.eu/enterprise/sectors/automotive/ documents/directives/motor-vehicles/index_en.htm EC (1996c). Directive 96/62/EC on ambient air quality assessment and management (Air Quality Framework Directive). http://ec.europa. eu/environment/air/legis.htm EC (1997a). Directive 97/68/EC of the European Parliament and of the Council of 16 December 1997 on the approximation of the laws of the Member States relating to measures against the emission of gaseous and particulate pollutants from internal combustion engines to be installed in non-road mobile machinery. http:// ec.europa.eu/environment/air/transport/non_road.htm EC (1997b). Council Decision 97/101/EC establishing a reciprocal exchange of information and data from networks and individual stations measuring ambient air pollution within the Member States (EoI Decision). http://ec.europa.eu/environment/ air/legis.htm EC (1998a). Directive 98/69/EC of the European Parliament and of the Council of 13 October 1998 relating to measures to be taken against air pollution by emissions from motor vehicles and amending Council Directive 70/220/EEC. http://ec.europa.eu/ enterprise/sectors/automotive/documents/directives/motorvehicles/index_en.htm EC (1998b). Directive 98/70/EC of the European Parliament and of the Council of 13 October 1998 relating to the quality of petrol and diesel fuels and amending Council Directive 93/12/EEC. http:// ec.europa.eu/environment/air/legis.htm EC (1999a). Directive 1999/96/EC of the European Parliament and of the Council of 13 December 1999 on the approximation of the laws of the Member States relating to measures to be taken against the emission of gaseous and particulate pollutants from compression ignition engines for use in vehicles, and the emission of gaseous pollutants from positive ignition engines fuelled with natural gas or liquefied petroleum gas for use in vehicles and amending Council Directive 88/77/EEC. http://ec.europa.eu/enterprise/ sectors/automotive/documents/directives/motor-vehicles/ index_en.htm EC (1999b) Directive 1999/30/EC relating to limit values for sulphur dioxide, nitrogen dioxide and oxides of nitrogen, particulate matter and lead in ambient air (First Daughter Directive). http://ec.europa.eu/environment/ air/legis.htm EC (2000). Directive 2000/76/EC of the European Parliament and of the Council of 4 December 2000 on the incineration of waste. http://ec.europa.eu/environment/air/legis.htm

Jana Moldanová, Peringe Grennfelt and Åsa Jonsson EC (2001a). Directive 2001/81/EC on National Emission Ceilings (NEC) http://ec.europa.eu/environment/air/legis.htm EC (2001b). Directive 2001/80/EC of the European Parliament and of the Council of 23 October 2001 on the limitation of emissions of certain pollutants into the air from large combustion plants. http:// ec.europa.eu/environment/air/legis.htm EC (2002). Directive 2002/3/EC of the European Parliament and of the Council relating to ozone in ambient air (Third Daughter Directive). http://ec.europa.eu/environment/air/legis.htm EC (2004). Commission Decision 2004/461/EC laying down a questionnaire for annual reporting on ambient air quality assessment under Council Directives 96/62/EC and 1999/30/EC and under Directives 2000/69/EC and 2002/3/EC of the European Parliament and of the Council. http://ec.europa.eu/environment/ air/legis.htm EC (2005a). Directive 2005/55/EC of the European Parliament and of the Council of 28 September 2005 on the approximation of the laws of the Member States relating to the measures to be taken against the emission of gaseous and particulate pollutants from compression-ignition engines for use in vehicles, and the emission of gaseous pollutants from positive-ignition engines fuelled with natural gas or liquefied petroleum gas for use in vehicles. http://ec.europa.eu/enterprise/sectors/automotive/ documents/directives/motor-vehicles/index_en.htm EC (2005b). Directive 2005/78/EC of 14 November 2005 implementing Directive 2005/55/EC of the European Parliament and of the Council on the approximation of the laws of the Member States relating to the measures to be taken against the emission of gaseous and particulate pollutants from compressionignition engines for use in vehicles, and the emission of gaseous pollutants from positive ignition engines fuelled with natural gas or liquefied petroleum gas for use in vehicles and amending Annexes I, II, III, IV and VI thereto. http://ec.europa.eu/ enterprise/sectors/automotive/documents/directives/motorvehicles/index_en.htm EC (2007). Regulation (EC) No 715/2007 of the European Parliament and of the Council of 20 June 2007 on type approval of motor vehicles with respect to emissions from light passenger and commercial vehicles (Euro 5 and Euro 6) and on access to vehicle repair and maintenance information. http://ec.europa.eu/ enterprise/sectors/automotive/documents/directives/motorvehicles/index_en.htm EC (2008a). Directive 2008/1/EC of 15 January 2008 concerning integrated pollution prevention and control. http://ec.europa.eu/environment/air/legis.htm EC (2008b). Commission Regulation (EC) No 692/2008 of 18 July 2008 implementing and amending Regulation (EC) No 715/2007 of the European Parliament and of the Council on type-approval of motor vehicles with respect to emissions from light passenger and commercial vehicles (Euro 5 and Euro 6) and on access to vehicle repair and maintenance information. http://ec.europa. eu/enterprise/sectors/automotive/documents/directives/motorvehicles/index_en.htm EC (2008c). Directive 2008/50/EC of the European Parliament and of the Council of 21 May 2008 on ambient air quality and cleaner air for Europe. http://ec.europa.eu/environment/air/legis.htm EC (2009). Regulation (EC) No 595/2009 of the European Parliament and of the Council of 18 June 2009 on type-approval of motor vehicles and engines with respect to emissions from heavy duty vehicles (Euro VI) and on access to vehicle repair and maintenance information and amending Regulation (EC) No 715/2007 and Directive 2007/46/EC and repealing Directives 80/1269/EEC, 2005/55/EC and 2005/78/

EC. http://ec.europa.eu/enterprise/sectors/automotive/documents/ directives/motor-vehicles/index_en.htm ECE (2009). Health risks of air pollution from biomass combustion. Report by the Task Force on Health, ECE/EB.AIR/WG.1/2009/12, 9 July 2009. EEA (2007). Air Pollution in Europe 1990–2004. EEA Technical Report No. 2/2007, EEA, Copenhagen. EEA (2009a). Spatial Assessment of PM10 and Ozone Concentrations in Europe 2005. EEA Technical Report No. 1/2009, EEA, Copenhagen. EEA (2009b). NEC Directive status report 2008. EEA Technical Report No. 11/2009, EEA, Copenhager. EEA (2010a). http://www.eea.europa.eu/data-and-maps/figures/ percentage-of-urban-population-resident-in-areas-wherepollutant-concentrations-are-higher-than-selected-limit-targetvalues-eea-member-countries-1997 EEA (2010b) http://eea.europa. eu/data-and-maps/figures/annualvariation-in-the-ozone-aot40-value-may-july-1996-2006 EEC (1970). Council Directive 70/220/EEC of 20 March 1970 on the approximation of the laws of the Member States relating to measures to be taken against air pollution by gases from positive-ignition engines of motor vehicles. http://ec.europa.eu/enterprise /sectors/automotive/documents/directives/motor-vehicles/ index_en.htm EEC (1988). Council Directive 88/77/EEC of 3 December 1987 on the approximation of the laws of the Member States relating to the measures to be taken against the emission of gaseous pollutants from diesel engines for use in vehicles. http://ec.europa.eu/ enterprise/sectors/automotive/documents/directives/motorvehicles/index_en.htm EEC (1991). Council Directive 91/441/EEC of 26 June 1991 amending Directive 70/220/EEC on the approximation of the laws of the Member States relating to measures to be taken against air pollution by emissions from motor vehicles. http://ec.europa. eu/enterprise/sectors/automotive/documents/directives/motorvehicles/index_en.htm EEC (1993). Council Directive 93/59/EEC of 28 June 1993 amending Directive 70/220/EEC on the approximation of the laws of the Member States relating to measures to be taken against air pollution by emissions from motor vehicles. http://ec.europa. eu/enterprise/sectors/automotive/documents/directives/motorvehicles/index_en.htm Emberson, L., Ashmore, M. and Murray, F. (2003). Air Pollution Impacts on Crops and Forests: A Global Assessment, Imperial College Press, London. EMEP (2008). Transboundry Particulate Matter in Europe: Status Report 2008. EMEP Report 4/2008. http://tarantula.nilu.no/ projects/ccc/reports/emep4–2008.pdf EMEP (2010). www.emep.int Filleul, L., Rondeau, V., Vandentorren, S. et€al. (2005). Twenty five year mortality and air pollution:€results from the French PAARC survey. Occupational and Environmental Medicine, 62, 453–60. Fuhrer, J., Skärby, L. and Ashmore, M. (1997). Critical levels for ozone effects on vegetation in Europe. Environmental Pollution, 97, 91–106. Gauderman, W. J., McConnell, R., Gilliland, F. et€al. (2000). Association between air pollution and lung function growth in southern California children. American Journal of Respiratory and Critical Care Medicine, 162, 1383–1390. Grizzetti, B., Bouraoui, F., Billen, G. et€al. (2011). Nitrogen as a threat to European water quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press.

431

Nitrogen as a threat to European air quality Hinds, W. C. (1999). Aerosol Technology, Properties, Behavior, and Measurement of Airborne Particles, 2nd ed. New York WileyInterscience, 442 pp. Hertel, O., Reis, S. and Ambelas Skjøth, C. (2011). Nitrogen processes in the atmosphere. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Hoek, G., Brunekreef, B., Goldbohm, S., Fischer, P. and van den Brandt, P. (2002). Association between mortality and indicators of traffic-related air pollution in the Netherlands:€a cohort study. Lancet, 360, 1203–1209. Holland, M., Kinghorn, S., Emberson, L. et€al. (2006). Development of a framework for probabilistic assessment of the economic losses caused by ozone damage to crops in Europe, CEH Project No. C02309NEW, Department of Environment, Food and Rural Affairs, London. ICP (2010). www.corr-institute.se/ICP-Materials IMO (2006). MARPOL, (consolidated edition). IMO, ISBN 978–92– 801–42167, www.imo.org IMO (2009). Revised MARPOL Annex VI: Regulations for the Prevention of Air Pollution from Ships, 2008. IMO, ISBN 978–92– 801–42433, www.imo.org IPCC (2007). Climate Change 2007:€The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, ed. S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K. B. Averyt, M. Tignor and H. L. Miller, Cambridge University Press. Ito, K., DeLeon, S. F. and Lippmann, M. (2005). Associations between ozone and daily mortality:€analysis and meta-analysis. Epidemiology, 16, 446–457. Jenkin, M. E. (2008). Trends in ozone concentration distributions in the UK since 1990:€local, regional and global influences. Atmospheric Environment, 42, 5434–5445. Jensen, L. S., Schjoerring, J. K., van der Hoek, K. et€al. (2011). Benefits of nitrogen for food fibre and industrial production. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Jerrett, M., Burnett, R. T., Pope, C. A. III, et€al. (2009). Long-term ozone exposure and mortality. New England Journal of Medicine, 360, 1085–1095. Jonson, J. E., Simpson, D., Fagerli, H. and Solberg, S. (2006). Can we explain the trends in European ozone levels? Atmospheric Chemistry and Physics, 6, 51–66. Katsouyanni, K., Touloumi, G., Samoli, E. et€al. (2001). Confounding and effect modification in the short term effects of ambient particles on total mortality:€results from 29 European cities within the APHEA2 project. Epidemiology, 12, 521–531. Klimont, Z., Asman, W. A. H., Bertok, I. et€al. (2007). Measures in Agriculture to Reduce Ammonia Emission. International Institute for Applied Systems Analysis (IIASA), Laxenburg, www.iiasa.ac.at. Levy, J. I., Chemerynski, S. M. and Sarnat, J. A. (2005). Ozone exposure and mortality:€an empiric bayes metaregression analysis. Epidemiology, 16, 458–468. Matyssek, R. and Innes, J. (1999). Ozone: a risk factor for trees and forests in Europe? Water, Air and Soil Pollution, 116, 199–226. Mauzerall, D. and Wang, X. (2001). Protecting agricultural crops from the effects of tropospheric ozone exposure:€reconciling science and standard setting in the United States, Europe and Asia. Annual Review of Energy and Environment, 26, 237–268.

432

Mills, G. (ed.) (2004). Mapping critical levels for vegetation. In: Mapping Manual 2004, UN/ECE Convention on Long-range Transboundary Air Pollution, Manual on Methodologies and Criteria for Mapping Critical Loads and Levels and Air Pollution Effects, Risks and Trends (UBA ed.), http://icpmapping.org/cms/ zeigeBereich/5/manual-und-downloads.html Mol, W. J. A., van Hooydonk, P. R., and de Leeuw, F. A. A.€M. (2009). European Exchange of Monitoring Information and State of the Air Quality in 2007. ETC/ACC Technical paper No.2009/3, http://air-climate.eionet.europa.eu/reports/ ETCACC_TP_2009_3_EoI_AQ_meta_info2007 Mol, W. J. A., van Hooydonk, P. R. and de Leeuw, F. A. A. M. (2010). European Exchange of Monitoring Information and State of the Air Quality in 2008, ETC/ACC Technical Paper No.2010/1, http://air-climate.eionet.europa.eu/reports/ ETCACC_TP_2009_3_EoI_AQ_meta_info2007 Molina, M. J. and Molina, L. T. (2004). Megacities and atmospheric pollution. Journal of the Air and Waste Management Association, 54, 644–680. Morgan, P. B., Mies, T. A., Bollero, G. A., Nelson, R. L. and Long, S. P. (2006). Season-long elevation of ozone concentration to projected 2050 levels under fully open-air conditions substantially decreases the growth and production of soybean. New Phytologist, 170, 333–343. Mudway, I. S., Stenfors, N., Duggan, S. T. et€al. (2004). An in vitro and in vivo investigation of the effects of diesel exhaust on human airway lining fluid antioxidants. Archives of Biochemistry and Biophysics, 423, 12. Nafstad, P., Haheim, L. L., Wisloff, T. et€al. (2004). Urban air pollution and mortality in a cohort of Norwegian men. Environmental Health Perspectives, 112, 610–615. Poppe, C. A., Burnett, R. T., Thun, M. J. et€al. (2002). Lung cancer cardiopulmonary mortality, and long-term exposure to fine particulate air pollution. Journal of the American Medical Association, 287, 1132–1141. Poppe, C. A., Burnett, R. T., Thurston, G. D. et€al. (2004). Cardiovascular mortality and long-term exposure to particulate€matter air pollution:€epidemiological evidence of general pathophysiological pathways of disease. Circulation, 109, 71–77. Pozzi, R., De Berardis, B., Paoletti, L. and Guastadisegni, C. (2003). Inflammatory mediators induced by coarse (PM2.5–10) and fine (PM2.5) urban air particles in RAW 264.7 cells. Toxicology, 183, 243–254. Putaud, J. P., Raes, F., van Dingenen, R. et€al. (2004). A European aerosol phenomenology-2:€chemical characteristics of particulate matter at kerbside, urban, rural and background sites in Europe. Atmospheric Environment, 38, 2579–2595. Pye, S. and Watkiss, P. (2005). CAFE CBA:€Baseline Analysis 2000 to 2020, AEAT Report AEAT/ED51014/Baseline Scenarios, http://www.iiasa.ac.at/ docs/HOTP/Mar05/cafe-cba-baselineresults.pdf. Reiss, R. Anderson, E. L., Cross, C. E. et al. (2007) Evidence of health impacts of sulfate- and nitrate- containing particles in ambient air inhalation. Toxicology, 19, 419–449. Samoli, E., Touloumi, G., Zanobetti, G. et€al. (2003). Investigating the dose–response relation between air pollution and total mortality in the APHEA-2 multicity project. Occupational and Environmental Medicine, 60, 977–982. Seaton, A. and Dennekamp, M. (2003). Hypothesis:€ill health associated with low concentrations of nitrogen dioxide€– an effect of ultrafine particles? Thorax, 58, 1012–1015.

Jana Moldanová, Peringe Grennfelt and Åsa Jonsson Simpson, D., Emberson, L., Ashmore, M. and Tuovinen, J. (2007). A comparison of two different approaches for mapping potential ozone damage to vegetation: a model study. Environmental Pollution, 146, 715–725. Sjödin, Å ., Jerksjö, M., Sandström, C. et€al. (2009) Implementering av ARTEMIS Road Model i Sverige, IVL Report No.B1886, www.ivl.se. Skärby, L., Ro-Poulsen, H., Wellburn, F. and Sheppard, L. (1998), Impacts of ozone on forests:€a European perspective. New Phytologist, 139, 109–122. Staehelin, J., and Schnadt Poberaj, C. (2008). Long-term tropospheric ozone trends:€a critical review. Advances in Global Change Research, 33, 271–282. Stieb, D. M., Judek, S. and Burnett, R. T. (2002). Meta-analysis of time-series studies of air pollution and mortality:€effects of gases and particles and the influence of cause of death, age, and season. Journal of the Air and Waste Management Association, 52, 470–484. Sutton, M. A., Reis, S. and Baker, S. M. H. (eds.) (2009). Atmospheric Ammonia: Detecting Emission Changes and Environmental Impacts, Results from a CLRTAP Expert Workshop, Springer, New York. Svirejeva-Hopkins, A., Reis, S., Magid, J. et€al. (2011). Nitrogen flows and fate in urban landscapes. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. TFIAM, CIAM (2007). Review of the Gothenburg Protocol. Taskforce on Integrated Assessment Modelling of the UNECE Convention on Long-range Transboundary Air Pollution (TFIAM) and the Centre for Integrated Assessment Modelling (CIAM). CIAM Report No. 1/2007. The Royal Society (2008). Ground-Level Ozone in the 21st Century: Future Trends, Impacts and Policy Implications, Science Policy Report No. 15/08, The Royal Society, London. www.royalsociety.org. Townsend, A. R., Howarth, R. W., Bazzaz, F. A. et€al. (2003). Human health effects of a changing global nitrogen cycle. Frontiers in Ecology and The Environment, 1, 240–246. UN/ECE (1979). 1979 Convention on Long-Range Transboundary Air Pollution, www.unece.org/env/lrtap/lrtap_h1.htm UN/ECE (1988). Protocol to the 1979 Convention on Long-range Transboundary Air Pollution on the Reduction of Emissions of Nitrogen Oxide or their Transboundary Fluxes by at least 30 per cent, www.unece.org/env/lrtap/ nitr_h1.htm UN/ECE (1999). Protocol to the 1979 Convention on Long-Range Transboundary Air Pollution to Abate Acidification, Eutrophication and Ground-level Ozone, www.unece.org/env/lrtap/multi_h1.htm UN/ECE ICP-Materials (2006). Trends in Corrosion and Air Pollution in 1987–2003, Programme Centre of the International Cooperative Programme on Effects of Air Pollution on Materials, including Historic and Cultural Monuments (ICP Materials), ECE/EB.AIR/WG.1/2006/7. UN/ECE ICP-Materials (2007). Dose–response functions for the multipollutant situation. Water, Soil and Air Pollution Focus, 7, 249–258.

Van Dingenen, R., Dentener, F. J., Raes, F. et€al. (2009). The global impact of ozone on agricultural crop yields under current and future air quality legislation. Atmospheric Environment, 43, 604–618. Velthof, G., Barot, S., Bloem, J. et€al. (2011). Nitrogen as a threat to European soil quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Volz, A. and Kley, D. (1988). Evaluation of the Montsouris series of ozone measurements made in the 19th century. Nature, 332, 240–242. Voss, M., Baker, A., Bange, H. W. (2011). Nitrogen processes in coastal and marine systems. In:€The European Nitrogen Assessment, ed.€M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. WHO (2000). Air Quality Guidelines for Europe, 2nd ed., WHO Regional Publications, European Series, No. 91., WHO Regional Office for Europe, Copenhagen, www.euro.who.int/__data/assets/ pdf_file/0005/74732/E71922.pdf WHO (2003). Health Aspects of Air Pollution with particulate Matter, Ozone and Nitrogen Dioxide. Report on a WHO Working Group, Bonn, Germany, 13–15 January 2003, WHO Regional Office for Europe, Copenhagen, document EUR/03/5042688, www.euro. who.int/document/e79097.pdf. WHO (2004). Health Aspects of Air Pollution: Answers to Follow-up Questions from CAFE Report on a WHO Working group, Bonn, Germany, 15–16 January 2004, WHO Regional Office for Europe, Copenhagen, document EUR/04/5046026, www.euro.who.int/ document/E82790.pdf. WHO (2005a). Particulate Matter Air Pollution: How It Harms Health. Fact sheet of WHO Regional Office for Europe, Copenhagen, document EURO/04/05 WHO (2005b). Effects of Air Pollution an Children’s Health and Development: A Review of the Evidence. Copenhagen, WHO Regional Office for Europe. http://www.euro.who.int/document/ E86575pdf. WHO (2006). Air Quality Guidelines Global Update 2005, WHO Regional Office for Europe, Copenhagen, www.euro.who.int/__ data/assets/pdf_file/0005/78638/E90038.pdf. WHO (2007). Health Relevance of Particulate Matter from Various Sources, Report on a WHO Workshop, Bonn, Germany, 26–27 March 2007, WHO Regional Office for Europe, Copenhagen, document EUR /07/5067587, www.euro.who.int/__data/assets/ pdf_file/0007/78658/E90672.pdf WHO (2008). Health Risks of Ozone from Long-Range Transboundary Air Pollution, WHO Regional Office for Europe, Copenhagen, www.euro.who.int/__data/assets/pdf_ file/0005/78647/E91843.pdf

433

Chapter

19

Nitrogen as a threat to the European greenhouse balance Lead authors: Klaus Butterbach-Bahl, Eiko Nemitz and Sönke Zaehle Contrubuting authors: Gilles Billen, Pascal Boeckx, Jan Willem Erisman, Josette Garnier, Rob Upstill-Goddard, Michael Kreuzer, Oene Oenema, Stefan Reis, Martijn Schaap, David Simpson, Wim de Vries, Wilfried Winiwarter and Mark A. Sutton

Executive summary Nature of the problem • Reactive nitrogen (Nr) is of fundamental importance in biological and chemical processes in the atmosphere–biosphere system, altering the Earth’s climate balance in many ways. These include the direct and indirect emissions of nitrous oxide (N2O), atmospheric Nr deposition and tropospheric ozone formation (O3), both of which alter the biospheric CO2 sink, Nr supply effects on CH4 emissions, and the formation of secondary atmospheric aerosols resulting from the emissions of nitrogen oxides (NOx) and ammonia (NH3). • Human production and release of Nr into the environment is thus expected to have been an important driver of European greenhouse balance. Until now, no assessment has been made of how much of an effect European Nr emissions are having on net warming or cooling.

Approaches • This chapter summarizes current knowledge of the role of Nr for global warming. Particular attention is given to the consequences of atmospheric Nr emissions. The chapter draws on inventory data and review of the literature to assess the contribution of anthropogenic atmospheric Nr emissons to the overall change in radiative forcing (between 1750 and 2005) that can be attributed to activities in Europe. • The use of Nr fertilizers has major additional effects on climate balance by allowing increased crop and feed production and larger populations of livestock and humans, but these indirect effects are not assessed here.

Key findings/state of knowledge • Due to its multiple, complex effects on biospheric and atmospheric processes, the importance of Nr for the European greenhouse gas balance has so far received insufficient attention. • The main warming effects of European anthropogenic Nr emissions are estimated to be from N2O (17 (15–19) mW/m2) and from the reduction in the biospheric CO2 sink by tropospheric O3 (4.4 (2.3–6.6) mW/m2). The main cooling effects are estimated to be from increasing the biospheric CO2 sink by atmospheric Nr deposition at −19 (−30 to −8) mW/m2 and by light scattering effects of Nr containing aerosol (−16.5 (−27.5 to −5.5) mW/m2), in both cases resulting from emissions of NOx and NH3. • The production of O3 from European emissions of NOx is estimated to have a modest warming effect (2.9 (0.3–5.5) mW/m2), which is largely offset by the cooling effect of O3 in reducing the atmospheric lifetime of CH4 (−4.6 (−6.7 to −2.4) mW/m2), giving an uncertain net warming of +1.7 (−6.4 to +3.1) mW/m2). • Overall, including all of these terms, European Nr emissions are estimated to have a net cooling effect, with the uncertainty bounds Â�ranging from a substantial cooling effect to a small warming effect (−15.7 (−46.7 to +15.4) mW/m2).

Major uncertainties/challenges • The largest uncertainties concern the aerosol and Nr fertilization effects, and the estimation of the European contributions within the global context. • Published estimates suggest that the default N2O emission factor of 1% used by IPCC for indirect emissions from soils following Nr deposition is too low by at least a factor of two. • The wider effects of fertilizer Nr, in allowing increased biospheric C cycling, food and feed production and populations of livestock and humans are a major uncertainty. Industrial production of Nr can be considered as having permitted increased overall consumption (of food, feed and fuel) with major net warming effects. These interactions remain to be investigated. The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

434

Klaus Butterbach-Bahl, Eiko Nemitz and Sönke Zaehle

Recommendations • The contribution of anthropogenic alteration of the nitrogen cycle to the radiative balance needs to be specifically accounted for in any greenhouse gas reporting (e.g., UNFCCC). • Although individual components of Nr emissions have cooling effects, there are many opportunities for ‘smart management‘ in linking the N and C cycles. These can help mitigate greenhouse gas emissions, while reducing the other Nr-related environmental threats such as eutrophication, acidification, air quality and human health.

19.1╇ Introduction and objectives This chapter aims to characterize how inputs of reactive nitrogen compounds (Nr) to the biosphere have affected the concentration of atmospheric trace substances and particles that are important for the radiative balance of the earth system. Based on our current understanding, and with a specific focus on Europe, this chapter furthermore evaluates how historic, present day and future changes in biospheric Nr inputs have and will feedback on the European greenhouse gas (GHG) balance. By including the additional cooling effects of aerosol, we extend the GHG estimates to assess the overall effect on radiative balance. The pathways of Nr input to the biosphere and how they are influencing atmospheric composition, and thus the radiative balance, are complex. They involve microbiological, plantphysiological, animal-physiological and physico-chemical processes, as well as manure management, industrial processes or atmospheric chemistry (Figure 19.1). Sources of anthropogenic Nr additions to the global biosphere are primarily related to fertilizer production, combustion processes, including the transport sector, or cultivation of leguminous plants (Galloway et€al., 2004). Once Nr has entered the biosphere, it can directly or indirectly affect the radiative balance of the earth by various processes. Direct effects are generally related to the formation of N2O, a greenhouse gas which is approximately 296 times as powerful as CO2 on a 100-yr timescale and per unit of weight (IPPC, 2007). The dominant source of both, natural or anthropogenic emissions is microbial production by nitrification and denitrification (see Butterbach-Bahl et€ al., 2011, Chapter 6, this volume). Indirect effects of Nr additions on the radiative balance involve a multitude of mainly biological processes on the ecosystem scale, but also physicochemical processes in the atmosphere (Figure 19.1), with the most prominent ones being as follows. (a) Changes in ecosystem C fluxes and C sequestration, affecting CO2 exchange. (b) Changes in ruminant and ecosystem CH4 production and consumption. (c) Changes in N2O production and emission. (d) Changes in atmospheric chemistry and specifically nitrogen oxides (NOx), ammonia (NH3) increasing aerosol formation and associated changes in the oxidative capacity of the troposphere with relevant feedbacks on biospheric processes, e.g., tropospheric ozone (O3) and plant growth. The primary effects of Nr inputs (which are easy to understand, but not to quantify) are increased emissions of N trace gases

(N2O, NH3, NOx) to the atmosphere. The processes driving the biosphere-atmosphere exchange of these compounds, such as nitrification and denitrification (N2O and NO) or volatilization (NH3) depend significantly on the availability of Nr in the plant–soil system. Thus, increased Nr inputs to agricultural systems (with livestock farming systems having the highest Nr use intensity in Europe) lead to increased losses of N trace gases (NH3, NOx, N2O) at the site of Nr input. However, following the cascade of nitrogen downwind or downstream into other ecosystems, N2O emissions affect a broader regional scale (Davidson, 2009; Oenema et€al., 2009). Thus, Nr trace gas emissions from natural and semi-natural terrestrial ecosystems, as well as emissions from water bodies, such as lakes, rivers or coastal waters, need to be considered (Figure 19.1). As a macro-nutrient, Nr positively affects photosynthesis and thus, the assimilation of atmospheric CO2 in plant biomass (Figure 19.2) (Liu and Greaver, 2009). Furthermore, Nr can stimulate the growth of the soil microbial community and in particular stimulate low affinity CH4 oxidation in rice paddies or inhibit high affinity CH4 oxidation in upland soils (Bodelier and Laanbroek, 2004; Figure 19.2), i.e., the availability of Nr affects the tendency or strength of binding of the enzyme CH4-monooxygenase for catalyzing the oxidation of CH4 to methanol. Even the process of methane production in anaerobic sediments (but also in the enteric fermentation in ruminants) is affected by the addition of Nr, as the stimulation of plant growth is a positive feedback upon rhizodeposition of C compounds and plant litter production, both of which serve as substrates for methanogenesis. Since Nr is highly mobile within the biosphere, it also affects aquatic ecosystems, e.g., with regard to eutrophication and the biosphere–atmosphere exchange of CO2 and CH4 or by changing the source strength of coastal waters for N2O. Besides these biological processes driving biosphere–Â� atmosphere exchange of CO2, CH4 and N2O (NOx), the importance of NH3 volatilization and industrial processes, as well as soil NOx emissions for particle formation and tropospheric O3 concentrations needs to be considered. Similarly, feedback loops need to be addressed for the quantification of the net effect of Nr on the GHG balance. In the case of O3, not only is it important as a greenhouse gas, but it also has a detrimental effect on plant productivity and, thus on atmospheric CO2 removal by terrestrial ecosystems, which needs to be accounted for. Therefore, to characterize and quantify effects of Nr on the radiative balance at continental to global scales, it is necessary to evaluate the effects of Nr on N2O-, CO2-, CH4-exchange as well as on exchange of NOx- and NH3 and their consequences for aerosol and O3 formation.

435

Nitrogen as a threat to the European greenhouse balance

Atmosphere

–

OH -

CH4

N2O

–

strat. O3

+

+

trop. O3

CO2

Aerosols

+/–

+

CO2

CH4

N2O

CO2 – +

+/– Growth

Nr

– + +/– Growth +/– + Soil R h

CO2

+

–

Fossil Fuel Burning CH4

+/–

+/–

Nr

Fertiliser Production

Growth + +/– Soil R h

Industry & Transport

Nr

+/–

CH4 CO2

Agricultural Ecosystems

CH4

(Semi-)Natural Ecosystems

Wetlands & Reparian Areas

Nr

Sewage Plants

+/– R h

Nr Transport in Rivers & Groundwater Figure 19.1 Effects of reactive nitrogen (Nr) on various biospheric processes in terrestrial and aquatic ecosystems and on atmospheric chemistry. Feedbacks on the production and consumption of atmospheric compounds directly or indirectly affecting the global radiative balance are indicated by arrows, where the thickness of the arrow gives an indication about the relative importance of a particular process. (Black arrows:€Nr fluxes; green arrows:€effects (in terms of positive or negative feedbacks); Radiatively active compounds red or fluxes (arrows) of them are marked red (blue) if they tend to increase (decrease) the radiative forcing. Dashed arrows:€Direct effects from anthropogenic additions of Nr; Black arrows:€Nr fluxes; Green arrows:€Effects (in terms of positive or negative feedback); Red arrows:€compounds that increase the radiative forcing (warming); Blue arrows:€compounds that decrease the radiative forcing (cooling).)

This chapter estimates, for the first time, the effect of European nitrogen usage on the climate system, taking into account:€ (i) direct emissions of the long-lived GHG nitrous oxide (N2O), (ii) the effect of Nr on the biospheric control of other GHGs, and (iii) the effect of Nr emissions on long-lived (e.g., methane, CH4) and short-lived radiative forcing agents (e.g., O3, particles). Several different metrics are used to quantify the climate effect of a change, e.g., in atmospheric composition or land cover. The most commonly used are the radiative forcing (RF) and the global warming potential (GWP). The RF is the global, annual mean radiative imbalance to the Earth’s climate system caused by human activities. The GWP of a trace gas is defined as the instantaneous mass emission of carbon dioxide that gives the same time-integrated radiative forcing as the instantaneous emission of unit mass of another trace gas (such as CH4 or N2O), when considered over a given time horizon. Thus, this metric is particularly useful in quantifying and comparing the future climate impact that is due to current emissions of longlived GHGs. However, GWP is less suited to quantifying the impact of short-lived agents. These radiative forcing metrics do not account for the climate sensitivity to the forcing. For example, climate sensitivity of radiative forcing due to changes

436

in O3 may not be the same as for RF due to CO2 and the sensitivity might differ geographically (Hansen et€al., 1997; Joshi et€al., 2003). Both GWP and RF are used in this chapter:€ the GWP is applied to assess the impact of Nr on current GHG emissions in Europe, and the RF concept to assess the indirect effects on air chemistry and to quantify the integrated effect that European Nr emissions have had on the global climate system.

19.2╇ Effects of reactive nitrogen on net N2O exchange

The major driver for changes in atmospheric N2O concentrations is the increased use of Nr fertilizer, which on the one hand allowed humans to dramatically increase global agricultural production and, thus, to feed the current global world population (Erisman et€al., 2008; Jensen et€al., 2011, Chapter 3, this volume), but on the other hand increased Nr availability and thus microbial N2O production. Owing to cascading of applied Nr onto landscape, regional and even global scales following the volatilization of NH3 and NOx, leaching of nitrate to water bodies or erosion processes, fertilizer Nr has also affected the source strength of non-agricultural terrestrial and aquatic

Klaus Butterbach-Bahl, Eiko Nemitz and Sönke Zaehle

CO2 CO2

N 2O

CH4 Photosynthesis (+)

ANPP (+) C allocation

Autotrophic respiration (+)

N input

BNPP (+/–) Root uptake

C supply (+/–) C : N (–)

N toxicity (+) Denitrifying bacteria (+)

Heterotrophic respiration (+/–)

c

i ob

r Ae

– +/

e ( ob al) r c r i M ene (g

CO2

)

Organic matter

Me

An

C substrate supply (+/–)

ae

rob tha ic arc nog ha eni ea c (+)

N2 NH4+

CH4

Aerobic Methanotropic bacteria (–)

N 2O

NO

Anaerobic –

NH2ONO

NO2

NO3–

N2O Nitrifying bacteria (+)

Aerobic

C cycle N cycle SOC

DIC/DON

DIN/DON

+ Positive feedback – Negative feedback

Figure 19.2 Potential mechanisms regulating the terrestrial ecosystem responses of CO2, CH4 and N2O production and consumption to increased availability of Nr (figure adapted from Liu and Greaver, 2009). (ANPP, aboveground net primary productivity; BNPP, belowground net primary productivity; SOC, soil organic carbon; DOC, dissolved organic carbon; DIN, dissolved inorganic nitrogen; DON, dissolved organic nitrogen.)

systems for N2O (Galloway et€al., 2004; Butterbach-Bahl et€al., 2011, Chapter 6, this volume). Also the emission of Nr from industry and combustion (including transport) has directly and indirectly contributed to changes in the global atmospheric N2O source strength. In addition, emissions of NOx from all combustion processes has resulted in huge increases in the atmospheric loading of Nr, with consequences for Nr deposition to terrestrial and aquatic ecosystems and thus also for Nr availability for microbial processes and finally N2O production (Sutton et€al., 2007; Simpson et€al., 2011, Chapter 14, this volume). This section evaluates separately each source category for N2O and how the source strength may have changed with time.

19.2.1╇ Direct N2O emissions from agricultural activities N2O emissions from the agricultural sector are mainly related to direct N2O emissions from soils following the application of Nr€– either in the form of synthetic fertilizer or in the form of manure€– or from N2O emissions related to livestock production,

specifically during manure storage, livestock grazing or from paddocks. The IPCC (2006) guidelines specifically list all of these sources and provide emission factors for estimating N2O emissions from them. Other top-down or bottom-up approaches have investigated only some of these sources or have amalgamated several sources together. Therefore, we provide a general overview of results for different approaches in Table 19.1. N2O emissions from soils are mostly estimated by using emission factor approaches (EFs), expressing proportionality between N2O efflux and fertilizer N input rate. However, it needs to be noted that these factors have a wide range of uncertainty (Eggleston et€al., 2006), and their use can underestimate the cumulative effect fertilizer N production may have on worldwide N2O formation. While using the IPCC default factor (1% of Nr applied being directly emitted as N2O plus indirect N2O emissions following Nr cascading downwind/ downstream of ecosystems due to volatilization/deposition, leaching/run-off or sewage emissions, see below) may still reflect the average situation at the plot scale, its overall effect on the global emission situation may be underestimated as the increase in atmospheric concentrations is observed to be more

437

438

52

Oenema et€al. (2007)

72.0

59.2

71.1

24.6

Sewage plants and waste disposal

30.7

<0.1

Biomass burning

215.8

Industry & transport

419.8

819 (833)b

Total

b

a

EEA (2009). Total (incl. not listed other sources, e.g., LULUCF). c Emissions of N2O from application of manure, i.e. does not include losses from synthetic fertilizer application. d Total N2O emissions for forest soils in EU15 only, i.e. also including background emissions. Calculations were done with the biogeochemical model Forest-DNDC. e Total N2O emissions include natural background emissions from rivers, shelf and estuarine areas (NE Atlantic between 45 and 66N, Baltic Sea, Mediterranean Sea and Black Sea). f Assuming a percentage loss of N2O of 2% (1.96%–2.1%) from manure and of 2.5% (2.37%–2.7%) for fertilizer nitrogen and using values for manure production (11â•›302 Gg N for EU-27) and fertilizer N (10â•›678 Gg N for EU-27) as given in Oenema et€al. (2009). g Assuming a percentage loss of N2O from new N fixation of 3%–5%. Calculations are based on total area of agricultural land in EU 27 (171 Mha, Oenema et€al., 2009) and mean N inputs to agricultural land in EU 27 in the year 2000 by synthetic fertilizer (66 kg N ha−1 yr−1) and biological N fixation (5 kg N ha−1 yr−1) (Table 7, Velthof et€al., 2009). h Minimum of the Crutzen et€al. (2008) estimate plus 215 Gg N from Industry & Transport sector. i Σ UNFCC (column:€1–3; Industry), indirect (Kesik et€al.)+estuaries (Bange) +biomass (EDGAR).

579h–1438i Gg N2O-N (or 269–669 Tg CO2-Equivalents)

364–607

Crutzen et€al. (2008)g

N2O effect on EU-27 radiative balance

475–526

Davidson (2009)f

330–670

77–87d

34.6

N2O emissions from EU rivers, shelfs & estuaries

Bange (2006)e

62

65.3

Indirect N2O emissions (N- deposition & leaching)

290

377

61c

224.2

272.5

N2O emissions (pasture, range-land & paddocks)

Seitzinger and Kroeze (1998)e

Kesik et€al. (2005)

Oenema et€al. (2009)

Direct N2O emissions from soils following Nr use

All numbers in Gg N / yr

EDGAR (2009)

UNFCCC

a

N2O emissions (manure management,: housing & storage)

Table 19.1 Sources of anthropogenic N2O emissions in EU-27 in the year 2000

Klaus Butterbach-Bahl, Eiko Nemitz and Sönke Zaehle

in the order of 2%–2.5% of manure and fertilizer Nr (Davidson, 2009) or 3%–5% of newly produced Nr (Crutzen et€al., 2008; see further below). In many regions of Europe, Nr added to soils exceeds the removal via crop and animal products. Velthof et€ al. (2009) published the Nr losses to the environment for EU-27, using a deterministic and static N cycling model that calculates Nr emissions from agricultural activities on an annual basis (MITERRA-EUROPE). Based on their evaluation, 171 Mha of agricultural land in EU-27 receives on average 143 kg Nrâ•›ha−1â•›yr−1, most of which is in the form of mineral fertilizer (46%) and manure (43%). On average, N2O emissions have been estimated at 2 kg N ha−1 yr−1 agricultural land (EU-27), which amounts to approx. 377â•›Ggâ•›N in the year 2000 for the agricultural sector (Oenema et€ al., 2009; see also Leip et€ al., 2011, Chapter 16, this volume). Total denitrification losses including both N2O and N2 as estimated by a N emission factor and mass balance approach were however much larger:€44 kg N ha−1 yr−1 agricultural land on average (range:€8–183 Nâ•›ha−1â•›yr−1 agricultural land) or 7486 Gg N for EU-27 (Velthof et€al., 2009; Oenema et€al., 2009), indicating that emissions of N2 are dominating. N2O emissions from the agricultural sector of EU-27 as reported by the EU member states in the framework of the Kyoto protocol are in the same range as reported by Oenema et€al. (2009), i.e., approx. 400 Gg N in the year 2000 for direct soil N2O emissions and emissions of N2O from pastures and manure management (EEA, 2009). The figures presented above are mainly based on IPCC emission factor (EF) approaches. However, Crutzen et€al. (2008) compared bottom-up and top-down approaches and suggested that the EF factors as recommended by IPCC may miss part of the effects over the lifetime of an Nr molecule. As a consequence, the overall emissions from releasing Nr to the environment as a fertilizer are not fully characterized by the 1.4% estimated according to IPCC (1% direct + 0.4% indirect; IPCC, 2006; note that 1% direct emissions is a global average which is based on field measurements, see e.g., database by Stehfest and Bouwman, 2006), but 3%–5% for synthetic fertilizer Nr or otherwise newly created Nr (Crutzen et€ al., 2008). In a recent re-evaluation of available data on temporal trends in global fertilizer and manure use and changes of atmospheric N2O concentrations over the past 150 years, Davidson (2009) suggested that during manure or mineral fertilizer use, 2% or 2.5% of the nitrogen is converted to N2O, respectively, and that double accounting of newly fixed N may be appropriate if the same N atom is first applied as newly fixed synthetic mineral fertilizer to produce animal feed and later gets recycled onto the soil as manure. It should be noted that the EF of Davidson and Crutzen et€ al. include both direct and indirect emissions. When using the Davidson (2009) approach and figures for EU-27 for total manure application (11â•›302 Gg€N) and fertilizer use (10â•›678€Gg€N) (Oenema et€al., 2009), this would result in total N2O emissions from the agricultural sector of 493 Gg N. If an overall conversion factor is applied of 3%–5% from newly fixed N (synthetic fertilizer N plus biological N fixation) to N2O-N, including indirect sources from cascading Nr (Crutzen et€al., 2008), then N2O emissions from EU-27 due to agricultural activities would range from

364–607 Gg N for the year 2000 (Table 19.1) (see also De Vries et€al., 2011, Chapter 15, this volume). Drainage of peatlands for agricultural use or for improving forest growth is widely practiced in Northern Europe. The resulting emissions of greenhouse gases from such organic soils may constitute a significant contribution to national GHG emissions. For example, Kasimir-Klemedtsson et€ al. (1997) estimated that CO2 and N2O emissions from farmed organic soils in the Netherlands, Sweden and Finland, though representing a minor fraction of arable soil, may contribute 3%–10% of total national greenhouse gas emissions (see also Freibauer, 2003). However, since land use conversion to arable land cannot directly be attributed to Nr use this source has not been considered in this study.

19.2.2╇ Direct N2O emissions from livestock farming and feedlots Livestock production systems and management of livestock manure exert various influences on the environment and make up a relatively large share of the total emissions of nitrous oxide (N2O). The influences on the environment greatly depend on the livestock production system itself, the management and the environmental conditions. Much of the influence of livestock Â�systems on the environment occurs via its effects (direct and indirect) on land use (changes) and nutrient element cycling. These effects have increased greatly over the last decades, Â�particularly in response to the current trends in livestock production:€Â�up-Â�scaling, intensification, specialization and regional conglomeration (Tamminga, 2003; Foley et€ al., 2005; Naylor et€ al., 2005; Steinfeld et€ al., 2006). These trends are facilitated by the availability of cheap energy, transport infrastructure and cheap Nr fertilizers for boosting the production of animal feeds. Though livestock consumes less than 3% of the global net primary production (Smil, 2002), its contribution to the global burden of NH3, CH4 and N2O in the atmosphere ranges between 10% and 40% (Bouwman et€ al., 1997; Oenema and Tamminga, 2005). Globally, livestock excrete about 100 (70–140) Tg Nr per year, but only 20%–40% of this amount is recovered and applied to crops (Sheldrick et€al., 2003; Oenema and Tamminga, 2005). The total amount of Nr excreted by livestock in EU-27 was about 7–8 Tg Nr in the early 1960s and increased to 11 Tg in the late 1980s. Thereafter it tended to decrease again. These amounts are in the same order of magnitude as the Nr fertilizer use. Fertilizer Nr use was 4 Tg in 1960, peaked at 12 Tg per year in the late 1980s, and was about 10.5 Tg in 2002 (FAOSTAT, 2006). Losses of N via N2O emissions from manure management are presented in Table 19.2. The table shows that soil-based N2O emissions (application and grazing) were higher than N2O emissions from housing and storage.

19.2.3╇ Direct N2O losses from sewage treatment and waste disposal Both heterotrophic denitrification and nitrification (more specifically nitrifier denitrification, see Butterbach-Bahl et€al., 2011,

439

Nitrogen as a threat to the European greenhouse balance Table 19.2 Total excretion of Nr by livestock and emission of N2O (kton) from animal manure management systems in EU-27 in 2000 (Oenema et€al.., 2009)

Livestock category Dairy cattle

Housing & Storage

Land Application

Grazing

Total

N2Oa

N 2O

N2O

N2O

2670

18

18

12 27

Nr excreted

Other cattle

3210

18

14

Pigs

1687

9

17

26

Poultry

1750

7

9

16

59

Other

1055

2

3

24

29

Total

10 372

54

61

62

177

a

â•›N2O emissions given here are the sum of emissions from housing systems and storage and do not include losses from mineral fertilizer applications to soil.

Chapter 6 this volume) are responsible for N2O emission during the process of wastewater treatment (Kampschreur et€ al., 2009). Under either totally anoxic or fully oxic conditions, N2O production is rather limited; however, it increases significantly under low oxygen partial pressures (typically in the range 0.3% to 1.5% O2 saturation) (Tallec et€al., 2006, 2008). Under actual operating conditions of wastewater aerobic activated sludge treatment N2O emission can represent 0.1% to 0.4% of the NH4+-N load oxidized (Tallec et€ al., 2006). Similarly, nitrate removal through denitrification at zero Â�oxygen could result in N2O emissions representing 0.4±0.3% of the NO3–N eliminated (Tallec et€ al., 2008). The latter figure is in the range of 0–1% mentioned by Hanaki et€al. (1992) for a bench-scale study of sewage plants with sludge ages of more than 2.5 days and influent chemical oxygen demand/NO3–N ratios higher than 2.5. One basic difficulty is that it is not always easy to distinguish clearly wastewater treatment from other waste management, e.g. the burning of sewage sludge, both industrial and domestic. EDGAR (2009) estimates annual emissions from waste water handling at about 25.7â•›Ggâ•›N representing ~5% of total European N2O emissions. For this emission source, the size of the population in the European countries is a good measure for the amounts emitted. Solid waste disposal includes the incineration of solid wastes (mandatory for future years according to the Waste Incineration Directive, 2000/76/EC; OJ L332, P91–111), as well as its combustion for power and heat generation, and finally its disposal in landfills. In total, solid waste disposal sources contributed about 4.98 Gg N2O-N in the year 2000, representing 0.8% of European emissions (EDGAR, 2009).

19.2.4╇ Direct N2O emission fluxes from the energy sector, industry, transport, etc. Like NOx, N2O may be formed as a side product during combustion. Temperatures favouring N2O formation are someÂ� what lower than those of NOx, so that N2O emission factors in Â�medium-temperature installations (500–600 °C) are clearly higher than at higher temperatures. Fluidized bed boilers, which employ lower temperatures (partly to abate NOx) are thus an N2O source. N2O may also be formed during reduction of NOx, during selective non-catalytic reduction, as well as in

440

48

three-way catalysts used for vehicles. In conclusion, NOx abatement is often a cause of N2O formation. Referring to figures on combustion emissions from EU-27, reveals that combustion is nevertheless only a minor source for N2O emissions. For 2007, EEA (2009) reports 7.6 Tg CO2-eq from power plants, and 13.4 Tg CO2-eq from transport (diesel and gasoline fuel), which is small fraction of the agricultural emissions. The only major point sources of N2O are large industrial facilities. The production of nitric acid via the oxidation of ammonia (a process that is also employed to produce caprolactam) and especially the application of concentrated nitric acid as an oxidizer for production of a handful of special chemicals (mainly adipic acid but also glyoxylic acid/glyoxal) are associated with high N2O concentrations downstream and in consequence have high emission factors. Nitric acid plants exist typically in each medium sized European country, with several plants in large countries. By contrast, installations to produce the specialized chemicals are limited to very few sites in Europe:€ adipic acid is produced in three installations in Germany and one each in the UK and France. Two small facilities also exist in the Ukraine with one in Italy. Glyoxal, another compound along the nitric acid pathway, is produced in France only, while caprolactam is mainly produced in Belgium and the Netherlands. The very limited number of installations and the high concentrations in the exhaust gas make it possible to monitor the emissions, so that the reports of EEA (2009) can be considered reliable. Furthermore, N2O mitigation measures have become available at very low costs or could even be integrated into the overall process as product recovery. In consequence, already before 2000, all large adipic acid plants were equipped with mitigation devices, such that in EU-27 N2O emissions dropped from 60 Tg CO2-eq in 1990 to 9 Tg CO2-eq, changing from a major to a minor source of N2O emissions. Within the EU, only the Italian plant followed relatively late, after 2005. At the same time, the N2O emissions from the other small industrial products mentioned decreased markedly (from 4.7 to 2 Tg CO2-eq). The N2O emissions from nitric acid production are still relevant. For 2007, EEA (2009) reports emissions of 40 Gg CO2-eq, about one quarter of the direct soil emissions.

Klaus Butterbach-Bahl, Eiko Nemitz and Sönke Zaehle

Abatement is technically possible and has been successfully applied in at least one nitric acid plant, but as plume concentrations of N2O are lower it is more expensive than in other processes. Nevertheless, the perspective of emission trading seems to make it financially attractive for European nitric acid producers to apply abatement measures as needed. According to industry reports, the phase-out of substantial N2O emissions is under way. The estimate for the UNFCCC reporting is 261 Gg N for the industry and transport sectors (EEA, 2009) is significantly lower than estimates by EDGARv4 with 461 Gg N (Reis et€al., 2009). Given that emissions from the chemical industry, as well as from the energy and transport sectors are well understood, differences between UNFCCC and EDGAR data are most likely due to different emission factors (or activity rates) in both approaches (Reis et€al., 2009).

19.2.5╇ Indirect N2O emissions from natural/ semi-natural terrestrial ecosystems Nitrogen deposited in natural and semi-natural ecosystems via atmospheric input mainly comes from two sources:€ agricultural activities and associated NH3 volatilization and NOx emissions through the burning of fossil fuels, from the industrialized continents, shipping and aircraft or biomass burning, lightning and production from soil microbes (Simpson et€al., 2006). The N demand of a growing forest is approx. 5–10 kg Nr/ ha/yr (Scarascia-Mugnozza et€ al., 2000). Thus, Nr deposition > approx. 10 kg N will lead to an enrichment of the ecosystem with Nr, i.e., increased Nr availability in the soil-plant system. Indications for Nr enrichment are a reduction of the C:N ratio of the litter, forest floor or mineral soil, increased concentrations of nitrate and ammonium in the soil solution (Kristensen et€al., 2004; Mannig et€al., 2008), as well as increased emissions of N2 and Nr trace gases from the soil. Several studies show that Nr deposition and N2O as well NO emissions from forest soils are positively correlated (Brumme and Beese, 1992; Brumme et€ al., 1999; Papen and Butterbach-Bahl, 1999; Van Dijk and Duyzer, 1999; Butterbach-Bahl et€ al., 1997; Pilegaard et€ al., 2006; Skiba et€al., 2006). The observed stimulation of fluxes is mainly attributed to the increased availability of Nr (as NH4+ and NO3−) for the microbial processes of nitrification and denitrification (Rennenberg et€al., 1998; Corré et€al., 1999), i.e., the key microbial processes responsible for N trace gas production in soils. A possible further explanation for increased N2O emissions due to ecosystem Nr enrichment was recently provided by Conen and Neftel (2007). They speculated that increased Nr availability may have reduced N2O reduction in soils via denitrification, i.e., that the ratio of N2O:N2 increases with increasing Nr availability. Since increased Nr deposition also affects nitrate leaching and runoff (Dise et€al., 1998; Borken and Matzner, 2004), indirect N2O emissions from water bodies due to Nr deposition to natural systems also need to be considered. However, a thorough evaluation and quantification of Nr deposition effects on soil N trace gas emissions remains difficult, since environmental conditions, such as meteorology or

soil and plant properties, significantly affect the magnitude, temporal course and composition of the emitted N gases. Having in mind these difficulties, there have been several attempts to estimate the stimulating effect of Nr deposition on N2O emissions from forest soils. Skiba et€al. (2006) used a gradient approach, with measuring sites located in a mixed forest at increasing distances from a poultry farm, i.e., a strong NH3 source. They estimated that >3% of the N deposited to the woodland sites was released as N2O. Butterbach-Bahl et€al. (1998) used a regression type approach, time series of nitrogen deposition throughfall data and continuous N2O and NO emission measurements at the long-term monitoring site at Höglwald Forest for estimating Nr-deposition driven N2O losses. Their estimate is comparable to that in the study by Skiba et€ al. (2006), i.e., 1.4% for coniferous forests and 5.4% for deciduous forest. Also, a literature review by Denier van der Gon and Bleeker (2005) showed that Nr deposition to forests stimulates N2O emissions within the same range; they concluded that the stimulating effect was higher for deciduous forests (5.7% of deposited N is lost as N2O) than for coniferous forests (3.7%). In a scenario study at the EU scale, Kesik et€al. (2005) estimated Nr deposition effects on forest soil N2O emissions by running the biogeochemical model Forest-DNDC either with best estimates of atmospheric Nr deposition or by assuming that Nr deposition was zero. The results indicated that, across Europe, 1.8% of atmospheric Nr deposition was returned to the atmosphere as N2O. All published estimates, therefore, show that the default N2O emission factor of 1% used by IPCC for indirect emissions from soils following N deposition (Mosier et€al., 1998; IPCC, 2006) is most likely too low by at least a factor of two.

19.2.6╇ Indirect N2O emissions from riparian areas, rivers and coastal zones Although direct emission from agricultural soils is the dominant process responsible for N2O emission by the agricultural sector, indirect emissions linked to the cascading of agricultural Nr ‘downstream’ from the fields might also play a significant role. Reactive nitrogen inputs to rivers and coastal waters include both natural and anthropogenic components; the latter is dominated by applied fertilizer Nr lost through leaching and runoff, followed by sewage and atmospheric sources (Seitzinger and Kroeze, 1998). Most of the Nr may already be denitrified in riparian areas in the direct vicinity of the sites of Nr application (e.g., arable fields). From a review of available data on N2O emissions from riparian wetlands, Groffman et€al. (2000) concluded that although current data are inadequate to propose a quantitative emission factor for Nr entering riparian areas, these emissions are likely to be significant in many regions. A nitrogen budget of the Seine hydrographical network (Billen et€al., 2001, 2007, 2009), reveals that up to 25%–30% of the Nr input to surface water from agricultural soils is denitrified to N2O and N2 in riparian zones, compared to only 5%–10% in-stream. If the percentage of N2O loss with respect to nitrate denitrified in riparian zones is the same as in the drainage network, N2O

441

Nitrogen as a threat to the European greenhouse balance

emissions from riparian zones would represent about 10% of the estimated total direct N2O emissions from the agricultural soils of the watershed (Garnier et€al., 2009). Also with regard to N2O losses from coastal areas, estimates are highly uncertain. The total flux of Nr and the fraction of fertilizer N reaching coastal waters is both variable and difficult to estimate. Early studies of individual rivers and/or �estuaries, targeted both measured and estimated inputs from various land use activities such as Nr in sewage and atmospheric �deposition (Billen et€ al., 1985; Larsson et€ al., 1985; Jaworski et€al., 1992; Boynton et€al., 1995; Nixon et€al., 1995; Howarth et€ al., 1996). Other work examined dissolved Nr export in relation to specific watershed characteristics, such as human population and energy use (Cole et€al., 1993), and point and non-point sources (Cole and Caraco, 1998). Another study modelled river and estuarine N2O production globally, using functions of nitrification and denitrification that were related to external N loading rates derived by adapting local/ regional models of watershed environmental parameters to global databases (Seitzinger and Kroeze, 1998). The results indicated (i) that ~ 8% of the Nr input to terrestrial ecosystems is exported as dissolved inorganic nitrogen (DIN) in rivers; (ii) the DIN export to estuaries globally (year 1990) is ~20.8 Tg Nr per yr; (iii) about 1% of the Nr input from fertilizers, atmospheric deposition, and sewage to watersheds is lost as N2O in rivers and estuaries; hence rivers and estuaries might account for 20% of current global anthropogenic N2O emissions and are thus similar in magnitude to previously identified sources such as direct anthropogenic N2O emissions from soils (Seitzinger and Kroeze, 1998). Such model-derived estimates should, however, be approached with caution. First, for denitrification the heterogeneity of microbial ecosystem structure (Rich and Myrold, 2004), oxygen status (Helder and De Vries, 1983) and physicochemical aspects such as sediment porosity and grain size (Garcia-Ruiz et€al., 1998), ambient temperatures, pH and water content (Berounsky and Nixon, 1990), and levels of suspended particulate matter (Owens, 1986) all co-vary to constrain N2O production. Second, the simple linear functions relating DIN loading to N2O in the global-scale models (Seitzinger and Kroeze, 1998; Seitzinger et€ al., 2000) are not well supported by individual studies, which reveal a wide range. For example, previous work in the Humber estuary (UK) implied that ~25% of the terrestrial DIN input converts to N2O via sediment denitrification (Barnes and Owens, 1999), a far larger conversion than the mean of about 0.15% employed in the global-scale models (Seitzinger and Kroeze, 1998; Seitzinger et€ al., 2000). By contrast, a dynamic model of estuarine DIN cycling in the Tyne estuary (UK) showed only 3.9% of the DIN load to be nitrified, in comparison to a value of 60% assumed in the global scale models, with only 0.009% of the DIN load converted to N2O (Rodrigues et€al., 2007). A corresponding value for the Scheldt estuary was 0.17% of nitrified Nr being converted to N2O, much closer to the global scale average (Rodrigues et€al., 2007). Nevertheless, this study concluded that the amount of atmospheric N2O derived from agricultural sources in general, including estuarine transformations of N, might need to be

442

revised downward, consistent with constraints set by atmospheric N2O growth. Although the global models (Seitzinger and Kroeze, 1998; Seitzinger et€ al., 2000) included some major European river/ estuaries in their development, direct comparisons of DIN load to N2O production were not readily available. Given the variability among the few direct comparisons that have been made and ranges of more than an order of magnitude in both DIN inputs and nitrification rates in European estuaries (Rodrigues et€ al., 2007), a meaningful representative mean value for the ratio for DIN input to N2O production for European estuaries appears to remain not well constrained. Seitzinger and Kroeze (1998) presented the first comprehensive estimate of N2O emissions from terrestrial and aquatic sources at the scale of Europe (more specifically the watersheds of North Eastern Atlantic, the Baltic, the Black Sea and the Mediterranean). They calculated N2O emissions from rivers, estuaries and shelf areas as a percentage (0.3 or 3%) of nitrification and denitrification rates in rivers and estuaries which were, in turn, calculated as a function of external inputs of dissolved inorganic nitrogen (DIN). Although the method is rather rough and questionable, it provides a first order of magnitude of the fluxes at European scale:€0.23 Tg N2O-N per yr from rivers and 0.11 Tg N2O-N per yr from estuaries and shelves. Based on a compilation of available data of N2O concentration in estuarine and coastal marine waters, Bange (2006) arrived at overall higher figures, based on component estimates of 0.13–0.16 and 0.20–0.41 Tg N2O-N per year for European shelf and estuarine waters, respectively. However the major merit of the approach used by Seitzinger and Kroeze (1998) is to explicitly link N2O emissions from aquatic systems to the nitrogen load reaching surface waters from agricultural sources, as is suggested by the site specific data presented above. Even if one assumes that 50% of emissions are natural, the remaining magnitude of N2O emissions from coastal zones in Europe would still be at least in the same magnitude as indirect emissions from leaching and Nr deposition (Table 19.1).

19.3╇ Effects of Nr on net CH4 exchange

Reactive nitrogen availability has been reported to affect both the capacity of upland soils to serve as a sink for atmospheric CH4 and as a source of CH4 emissions from wetlands (Figure 19.2). Furthermore, indirect effects of Nr on animal feed quality, in this case, changes in indigestible carbohydrates and crude protein contents, can also affect ruminant CH4 emissions. These three aspects are briefly discussed here with their potential feedback on the EU radiative balance. A summary of all effects is presented in Table 19.3.

19.3.1╇ Net CH4 oxidation by upland soils

Globally, biological methane oxidation is estimated at 17 ± 9€Tg CH4-C per yr (Dutaur and Verchot, 2007). By comparison, Boeckx and Van Cleemput (2001) estimated the total EU15 CH4 oxidation (grassland and agricultural land) to be 0.2 Tg C per yr. In general, higher uptake rates than for grasslands and agricultural land are reported for other upland ecosystem

443

<0.31

Tg CO2-Equivalents: 8.3 to 23

?

−34.1 to −163

b

a

European Environment Agency (2006). Boeckx and Van Cleemput (2001). c Based on the linear equation by Liu and Greaver (2009) and an average N deposition to EU forests and semi-natural land of 13 kg Nr yr−1 or 10 kg Nr yr−1, respectively. d Assuming that rates would equal those of upland forest soils if Nr use would be abandoned. e Saarnio et€al. (2009). f Assuming a mean atmospheric Nr deposition of 10 kg N ha−1 yr−1 and using the Nr response factor as provided by Liu and Greaver (2009) together with the total area of wetlands in EU-27. g EEA (2009); includes only the values for enteric fermentation (year 2000). h The effects of Nr supply on forage digestability and CH4 emissions per animal, and the effect of Nr in allowing increased animal stocking rates and overall animal numbers are not quantified here. i Schulze et€al. (2009), all terrestrial ecosystems (forest only = −109 Tg C yr−1) j De Vries et€al. (2007) and section 19.4 (this volume). k See Section 19.5.2 (this volume).

<0.01

Tg C yr−1: 0.27 to 0.75

?

?h

−9.3 to −44.5

<0.01f

Nr effect on EU-27 radiative balance

<0.01c

4.98g

Terrestrial C sequestration

5.0–14.4k

0.2d (0.18–0.48)

−0.07 (0.06–0.17)

3.92e

Ruminants

O3 effects on ecosystem carbon sequestration (Tg C yr-1)j

0.09d (0.09–0.27)

−0.11 (−0.08–0.26)

−2.7 (−2.2–6.3)

46 915

Wetlands

−36.6±12.9j

0.02±0.006c

Effect of Nr on CH4 exchange (Tg CH4-C yr−1)

−0.12 (−0.04–0.25)

−1.0 (−0.7–2.2)

264 932

Other semi natural land

Nr effect on ecosystem CO2 fluxes (Tg C yr−1)

−0.27 (−0.23–0.65)

Total CH4 exchange EU24 (Tg CH4-C yr−1)

−1.6 (−0.5–3.1)

1 174 325

Arable

−125i

−2.7 (−2.2€– 6.3)

CH4 exchange (kg CH4-C ha−1yr−1)b

820 219

Grassland

Ecosystem CO2 fluxes (Tg C yr−1)

1 030 635

Total area (km2)a

Forest

Table 19.3 Changes in CH4 fluxes in EU-24/EU-27 and C sequestration by terrestrial ecosystems due to the use of Nr and its dispersal in the environment (positive values are emissions and negative values are uptake)

Nitrogen as a threat to the European greenhouse balance

types such as forests and heathlands in the temperate and boreal zones (Smith et€ al., 2001; Butterbach-Bahl, 2002; Dutaur and Verchot, 2007). Uptake of CH4 by upland soils in EU25 has been approximately calculated as 4 Tg CH4-C per yr (Schulze et€al., 2009). This sink strength, however, has been and still is affected by Nr (Figure 19.2) and land use change. Increased atmospheric Nr deposition increases [NH4+] in the soil and usually decreases CH4 uptake by well-drained soils (Steudler et€al., 1989; Gulledge and Schimel, 1998). Three mechanisms have been postulated for the partial inhibition (slowing down) of CH4 uptake by well-drained soils in response to increased Nr input:€(a) competitive inhibition of the methane mono-oxygenase by ammonia, (b) inhibition of methane consumption by toxic intermediates and end products of methanotrophic ammonia oxidation such as hydroxylamine and nitrite, or (c) osmotic stress due to high concentrations of nitrate and/ or ammonium (Schnell and King, 1996; Bradford et€al., 2001; Bodelier and Laanbroek, 2004; Reay and Nedwell, 2004). Changes in land use affect soil methane oxidation, for example by the demand for increased production for food, fibre and biofuels. Generally CH4 oxidation in upland soils sharply decreases with intensity of landuse, i.e. CH4 oxidation will be highest in forests and natural grasslands, somewhat lower in managed grasslands and negligible in arable soils. On the basis of a meta-data analysis of published studies on Nr effects on CH4 uptake by upland soils, Liu and Greaver (2009) estimated the detrimental effects of Nr addition to soils on CH4 uptake as 0.012 ± 0.006 kg CH4-C ha−1 yr−1 for upland agricultural fields and as 0.016 ± 0.004 kg CH4-C ha–1 yr–1 for non-agricultural ecosystems, per 1 kg Nr ha−1 yr−1 added to the ecosystem. Comparable values for the inhibitory effect of Nr addition to forest ecosystems in Europe via Nr deposition on CH4 uptake are reported by Butterbach-Bahl et€ al. (1997) and De Vries et€al. (2008). They reported a factor of 0.031 kg CH4-C ha−1 yr−1 per 1 kg N ha−1 yr−1 in the form of NH4+. If one assumes that the total molar ratio of reduced to oxidized atmospheric Nr deposition in Europe is approx. 1:1 (which is in accordance with Simpson et€al., 2006) this would result in precisely the same number as in the Liu and Greaver study, assuming oxidized N does not affect CH4 oxidation. These findings can be used to approximate the effect of Nr additions via atmospheric deposition or fertilizer application on CH4 uptake for natural/semi-natural ecosystems in Europe. For€grassland and agricultural land, an approach which considers Nr application and uses the regression line provided by Liu and Greaver (2009) yields somewhat unrealistic values if the Nr use intensity in Europe is taken into account. However, for agricultural land one can assume that CH4 uptake activity may recover to a maximum of values observed in natural upland systems (if no Nr is applied and assuming that no management alteration). On the basis of an extensive literature review on existing measurements of methane uptake in Europe and elsewhere, Boeckx and van Cleemput (2001) summarized ranges for CH4 uptake for forest soils of (2.2–6.3 kg CH4-C ha−1 yr−1, mean:€2.7 CH4-C ha−1 yr−1), for grasslands (0.5–3.1 kg CH4-C ha−1 yr−1, mean:€1.6 CH4-C ha−1 yr−1) and for arable soils (0.7– 2.2 kg CH4-C ha−1 yr−1, mean:€1.0 CH4-C ha−1 yr−1). These ranges

444

are in general agreement with a more recent global analysis of CH4 uptake on upland soils by Dutaur and Verchot (2007). If one assumes that CH4 uptake would, at maximum, equal the rates of uptake in forest soils, then agricultural activity and use of Nr has decreased CH4 uptake by on average 0.1 Tg CH4-C per yr for grassland soils and by 0.2 Tg CH4-C per yr for arable soils, respectively (at the scale of EU-24), see Table 19.3. For forest the formula as provided by Liu and Greaver (2009) was used. Average atmospheric Nr deposition in Europe was assumed to be on average 13 kg Nr ha−1 yr−1 for forests (Simpson et€al., 2006) or 10 kg Nr ha−1 yr−1 for other semi-natural upland systems (e.g., heathland, macchia), respectively (Table 19.3). With this approach we do not consider that large parts of forests in Northern Europe receive <13 kg N ha−1 yr−1, whereas forests in Central Europe are exposed to values of atmospheric N deposition >13 kg N ha−1 yr−1, i.e. the spatial variability of Nr deposition was disregarded for this calculation. However, the effect of Nr atmospheric deposition on rates of atmospheric CH4 uptake by forest soils and soils of other semi-natural land uses at the scale of EU-27 can be seen to be small at <0.05 Tg CH4-C yr−1. In summary, accounting for all land uses, the effect of all Nr in suppressing CH4 uptake by soils in Europe was estimated at 0.27–0.75 Tg CH4 yr−1 (Table 19.3).

19.3.2╇ Net CH4 emissions from wetlands

The effect of Nr additions to wetland ecosystems will largely depend on the N status of the respective ecosystems. For highly managed wetland ecosystems, such as rice paddies, that receive high loads of Nr fertilizers, inorganic Nr additions have been shown to result in reductions of CH4 emissions. This effect is most likely due to a stimulation of CH4 oxidizing bacteria following increased Nr availability (Bodelier and Laanbroek, 2004) or due to a slower decrease in redox potential and thus delayed onset of methanogenesis (if nitrate fertilizers or ammonium sulphate fertilizers are used). On the other hand, Nr additions will increase plant productivity and thus also rhizodeposition of C substrates, which can fuel methanogenesis and can lead to a net increase in CH4 emissions, as has been shown for rice paddies and natural wetlands (Fumoto et€al., 2008). Furthermore, significant rates of Nr addition to wetlands (>approx. 10–20 kg Nr ha−1 yr−1) promotes the growth of vascular plants (to the detriment of moss numbers) and thus potentially increases evapotranspiration€ – which may lower the water table€ – and stimulates CH4 oxidation in the rhizosphere (Berendse et€ al., 2001; Bouchard et€al., 2007). However, reports on the effect of atmospheric Nr deposition on CH4 emissions from wetlands are scarce (Dise and Verry, 2001), so that the possible consequences of Nr additions to wetlands and on CH4 fluxes are difficult to assess. In their metaanalysis of data sets, Liu and Greaver (2009) therefore included studies on Nr addition effects on CH4 emissions from rice paddies, because of scarcity of relevant datasets and because, in their evaluation, the response to Nr addition was not different between the two ecosystems. The calculated response factor suggests that, per kg Nr, CH4 emissions are stimulated by 0.008 ± 0.004 kg CH4-C ha−1 yr−1. Compared to the overall CH4

Klaus Butterbach-Bahl, Eiko Nemitz and Sönke Zaehle

emissions from wetlands and water bodies in Europe, which were estimated at 3.92 Tg CH4-C yr−1 (Saarnio et€al., 2009), the Nr deposition effect (assuming a mean value of 10 kg N ha−1 yr−1) on total emissions is negligible and <0.01 Tg CH4-C yr−1.

19.3.3╇ CH4 emissions from ruminants and their relationship to nitrogen Ruminants are an important source of CH4, which is a Â�product of microbial degradation of feed constituents. In the long term, some feeding practices on intensively managed farms have improved considerably and can be expected to continue to do so, mainly because of the gradual improvement of ruminant genoÂ�types which requires a permanent adaptation of nutritional strategies. In the case of extensive feeding systems, changes over time in diet composition were much smaller. There are several interactions between Nr management and ruminants CH4 emissions. These include the use of Nr fertilizer to supply increased amounts of feed, allowing larger Â�livestock populations, the effects of forage and feed quality on CH4 Â�emissions, and management interactions concerning the fate of livestock excreta. Possibly the largest effect of Nr supply on ruminant CH4 emissions in Europe is by providing fertilizer to grow crops used as forage and feed for ruminants. Without substantial Nr inputs€– with a significant share coming from feed imports€– also Europe would not be able to support so many livestock, especially in the dairy, pig and poultry sectors. By comparison, beef cattle and especially sheep are typically grazed on more extensive systems, often with much lower Nr fertilizer inputs. In broad terms, Erisman et€ al. (2008) estimated that around half of the world population depends on Nr fertilizers and this percentage is expected to be higher for Europe. In this case, a minimum of 50% of European ruminant CH4 emissions (i.e. ~2.5 Tg CH4-C per year) could therefore be attributed to Nr supply. However, this figure remains very uncertain due to the different mixtures of land use, forage and feed (including feed imports) used to support the European ruminant population. In the following paragraphs, we therefore focus on the direct interactions of Nr with CH4 emissions from ruminants. The key for a high N use efficiency of the animal and thus a low N emission potential of the manure lies in the optimization of the rumen microbial protein synthesis. These microbes require both degradable protein and carbohydrates (starch, sugars). The most prominent change in the last decades in European feeding practices concerns changes in the carbohydrate sources. This is through the gradual exchange of part of the forage (rich in fibre) by concentrate (rich in starch and often also protein). Under certain conditions (e.g., sufficient ruminal availability of the concentrate’s carbohydrates, adjusted amino acid supply) this improves N use efficiency by up to 50% and mitigates potential N emissions accordingly. By contrast, forages from extensive grasslands are often low in digestibility (few ruminally degradable carbohydrates) and those from intensively managed grasslands are excessive in protein, thus being inferior to mixed forage-silage maize-concentrate diets in that respect.

The consequences for methane emission of these changes are variable; however, a proportionate decline in methane formation per unit of digestible carbohydrates, and therefore especially per unit of livestock-source food, can still be expected from the use of mixed diets (Beauchemin et€ al., 2008). Mitigating effects of concentrate on proportionate methane emission range between 0 and 70%. Hindrichsen et€ al. (2006) noted a reduction of methane emission from 36 to 19 and 14 g CH4 per kg of milk in cows yielding 10, 20 and 30 kg milk day–1, respectively, showing that the efficiency gain is much lower when starting from a higher level of yield. Additionally, any reduction with this diet change may be at least partially compensated for by a concomitantly increased methane emission from the manure (Hindrichsen et€al., 2006). The second most important measure to improve N use efficiency in livestock is given by a reduction of excessively high (crude) protein contents of the diet. This is very effective in reducing potential N emission (Külling et€al., 2001), but effects on methanogenesis are small and inconclusive (Külling et€al., 2001, 2003). The only exception arises when a dietary protein reduction below a critical value adversely affects fibre degrading rumen microbial species. These species are also those providing most of the hydrogen required by the methanogens. This situation rarely occurs in Europe, except in very extensive systems. In developing countries, the use of either ureamolasses licking blocks, or ammonia/urea treated straw is a major option for dealing with the limitation in degradable N. The reduced methane emission resulting from this strategy is associated with a massive increase in performance (Moss et€al., 1994; Islam and Begum, 1997). The source of ruminally degradable protein (grass, oil meals, urea, etc.) has no direct influence on the ruminal N use efficiency and therefore does not affect either N or methane emissions in a substantial way. However, the metabolic N use efficiency of the animal for milk and meat production can be improved by using ruminally-undegradable protein sources when dietary protein is concomitantly reduced (Kröber et€al., 2000). Such protein sources include naturally and artificially rumen-protected protein or even single protected amino acids. As protein is not a major substrate for the methanogens, again this set of feeding measures would only affect Nr emissions and not methanogenesis. New attempts in the mitigation of CH4 and associated noxious gases are based on feeding with low dietary concentrations of effective secondary plant metabolites (Beauchemin et€ al., 2008). There is a huge variability in compounds and plant species available, and some of them have already been processed to marketable products, others might be grown as forages by European farmers. Effective plants are often rich in tannins, saponins, essential oils and sulphur-containing compounds. As these ingredients may partially inhibit ruminal protein degradation to ammonia and at the same time the methanogens, these measures are particularly promising. However, they are rarely cost-effective unless mitigation of noxious gases also improves animal growth or milk yield or is included in a payment scheme.

445

Nitrogen as a threat to the European greenhouse balance

Methane emissions per unit of food have substantially declined with increasing feed quality over the past decades, but also due to improvements in the genetic characteristics of the herd and the improvements in management. However, this change was not accompanied by a corresponding reduction in the number of animals needed to produce our food, as eating habits simultaneously increased towards higher consumption of these now cheaper foods. This means that global methane emission from ruminant husbandry has continued to increase during the last decades, as fueled by substantial Nr fertilizer inputs, and thus contributes to the ongoing raise in atmospheric CH4 concentration.

19.4╇ Effects of reactive nitrogen on ecosystem net CO2 exchange

Reactive nitrogen is a key nutrient for both vegetation and soil biota and because of the limited (natural) supply it is a limiting factor for plant growth and soil organic matter decomposition in many terrestrial ecosystems (Vitousek and Howarth, 1991; LeBauer and Treseder, 2008). Field studies have demonstrated a positive effect of low to medium level Nr additions on plant growth and carbon accumulation (Vitousek and Howarth, 1991; Aber et€al., 1993; Bergh et€al., 1999; Franklin et€al., 2003). Nr additions affect vegetation growth by increasing tissue N content and leaf-level photosynthesis, as well as decreasing the (relative) investment into below-ground carbon allocation (Poorter and Nagel, 2000; Magill et€ al., 2004). Both mechanisms generally increase above-ground productivity and lead to a higher accumulation of above-ground woody biomass. However, the effectiveness of these mechanisms is limited when other processes such as water limitation, micronutrient availability, and competition for light become more limiting than Nr availability. For example, a 15-year-long Nr amendment study in Harvard Forest, USA, showed that the large increases in carbon accumulation predicted using a linear relationship between leaf Nr content and photosynthesis, failed to materialize in the field at least for red pine (Bauer et€ al., 2004). As a consequence, the vegetation response is strongest in young fast growing forest ecosystems (Oren et€ al., 2001) and boreal forest ecosystems (Bergh et€al., 1999; Jarvis and Linder, 2000), in which Nr is the primary constraint of growth. Furthermore, the Nr addition effect is expected to saturate or even decline in ecosystems with high Nr input ecosystems (Aber et€ al., 1998; Brumme and Khanna, 2008). Soil respiration rates, both from autotrophic and heterotrophic sources, have been shown to be generally reduced under elevated Nr (Fog, 1988; Agren et€ al., 2001; Hagedorn et€ al., 2003; Knorr et€ al., 2005; Olsson et€ al., 2005). The reason for the decline is likely to be an alteration of the microbial decomposition of organic matter by uncoupling the degradation of polysaccharides and polyphenols (Sinsabaugh et€ al., 2002, 2005). Increased Nr availability stimulates cellulolysis, which tends to accelerate the decomposition of labile litter, and inhibits the expression of oxidative enzymes required for the breakdown of lignin and other secondary compounds. In consequence labile organic matter stocks may turnover more

446

rapidly, thus shrink in abundance, while humified fractions accumulate (Sinsabaugh et€al., 2005). The net effect on C accumulation in soil depends on whether changes in the decomposition rate or increased C inputs from increased biomass and litter production dominate (Schulze et€al., 2000). Rainfed, ombotrophic bogs are a special case in the response of soil organic matter to Nr addition. Low Nr inputs promote the growth of the peat building Sphagnum plants; however, higher Nr availability generally favors the growth of vascular plants (Berendse et€al., 2001; Bubier et€al., 2007). Because of (co-)limitation with phosphorus (P) and potassium, Nr input does not necessarily increase growth (Limpens et€ al., 2004). However, even where growth is increased, the net effect on C accumulation may still be zero or even negative because of the higher degradability of vascular plant litter (Gunnarsson et€al., 2008). Nr additions generally increase the decay of dead material in peatlands (i) due to reduced microbial Nr limitation and (ii) indirectly due to improved litter quality (Bragazza et€al., 2006). Enhanced decomposition in peatlands due to Nr may however, be limited, as several studies with high Nr inputs show signs of P limitation (Limpens et€al., 2004; Bragazza et€al., 2006). At the whole ecosystem scale, the rate of C accumulation in response to Nr addition is determined by the fate of the Nr, and the stoichiometry of vegetation and soil organic matter (Nadelhoffer et€al., 1999, 2004). For example, in forests, because of the high C:N ratio of woody biomass, N stored in wood will involve a much stronger C accumulation than in N storage in soil organic matter. Depending on the fate of the added Nr, the C accumulation per unit Nr could vary between zero (in ecosystems with no or little N retention) and several hundred (where most N accumulates in woody tissue). A number of recent studies have aimed at quantifying the response of C accumulation to additions based on fertilizer trials (Hyvonen et€al., 2007), application of 15N tracer (Nadelhoffer et€al., 1999, 2004), observations at long-term monitoring plots (De Vries et€al., 2006; Solberg et€al., 2009; Laubhann et€al., 2009), as well as interpretation of net ecosystem production data from eddycovariance CO2 measurements (Magnani et€ al., 2007; Sutton et€ al., 2008). The published estimates of the response of carbon sequestration to N addition in above-ground biomass and in soil organic matter for forests and heathlands have recently been summarized by de Vries et€ al. (2009) (Table 19.4). The results of the various studies are in close agreement and show that above-ground accumulation of carbon in forests is generally within the range 15–40 g C per g Nr. In heathlands, a range of 5–15 g C per g Nr has been observed based on low-dose Nr fertilizer experiments. The uncertainty in C sequestration per kg Nr addition in soils is larger than that for above-ground biomass and varies on average between 5–35 g C per g Nr for both forests and heathlands. All together these data indicate a total carbon sequestration range of 5–75 g C per g Nr deposition for forest and heathlands, with a most common range of 20–40 g C per g Nr. Such low values are to be expected as 15N tracer studies suggest that most added Nr becomes stored in soil organic matter (Nadelhoffer et€al., 1991; Tietema et€al., 1998). The results are in line with a meta-analysis of studies on CO2 fluxes from Nr additions in multiple terrestrial and

Klaus Butterbach-Bahl, Eiko Nemitz and Sönke Zaehle Table 19.4 Estimated ranges in carbon sequestration per kg nitrogen addition in above-and-below ground biomass in forest at various scales (after de Vries et€al., 2009)

Carbon sequestration (kg C per kg Nr)

Scale of application

Above ground

Below ground

Total

Correlation between NEP and total N deposition

—

—

68–177a

Chronosequences (5) in boreal and temperate forests of Eurasia and North America

Magnani et€al. (2007) as re-evaluated by Sutton et€al. (2008)

Correlation between the average growth increase of nearly 400 Intensive Monitoring plots and N deposition in a multivariate analysis

15–38

—

—

Nearly 400 Intensive forest Monitoring plots

Solberg et€al. (2009), Laubhann et€al. (2009)

Extrapolation of 15N experimental data with average C/N ratios of forest ecosystem compartments

30–70

11–18

41–88

One forest site in Sweden

Melin et€al. (1983)

Extrapolation of 15N experimental data with average C/N ratios of forest ecosystem compartments

25

21

46

Generic average

Nadelhoffer et€al. (1999)

Extrapolation of 15N experimental data with site specific data at 6000 plots in Europe

33

15

48

European average

De Vries et€al. (2006)

Average results from 30 year low dose (34 kg Nr /ha/yr) fertilizer experiments

25

—

—

Forest plot in Sweden

Högberg et€al. (2006)

Average results from 14–30 year fertilizer experiments

25

11

36

Two forest plots in Sweden and Finland

Hyvönen et€al. (2008)

Average results from 10 year chronic N addition (30 kg Nr /ha/yr) experiments

17

23

40

Four forest plots in the USA

Pregitzer et€al. (2008)

Range in results of three process-based models (high Nr supply)

—

—

10–30

One forest site in Sweden

Levy et€al. (2004)

Range in results of three process-based models (medium Nr supply, 9–25 kg Nr / ha/yr)

—

—

43–75

One forest site in Sweden

Sutton et€al. (2008)

Range in results of five process-based models

15–25

—

—

Two forest plots in UK

Rehfuess et€al. (1999)

Average result of the process-based model EFM

—

—

41–54

22 forest plots in Europe

Milne and van Oijen (2005); Sutton et€al. (2008)

Range in results of the process based model SUMO

20–30

—

—

Dutch forests

Wamelink et€al. (2009a)

Range in average results per latitude of the process based model chain SMART2SUMO2

3–12

5–11

7–24

166 forest plots in Europe

Wamelink et€al. (2009b)

37 (2–79)

87 forest plots in Europe and Northern America

Zaehle and Friend (2010)

Approach

Authors

Forests Empirical field data

N experimental data

15

Results of fertilizer experiments

Results of model simulations

Range in results based on the processbased model O-CN

447

Nitrogen as a threat to the European greenhouse balance Table 19.4╇ (cont.)

Carbon sequestration (kg C per kg Nr)

Scale of application

Above ground

Below ground

Total

Results from 5–11 year N fertilizer experiments at 20–120 kg Nr /ha/ yr

5–15

20–34

25–49

2 heathland plots

Evans et€al. (2006); Evans, pers. comm.

Model simulations for the N fertilizer experiment sites

-

21–32

-

3 heathland plots

Evans et€al. (2006); Evans, pers. comm.

Approach

Authors

Heathlands

a

â•›The high value assumes no covariation between Nr and climate drivers; the low value attributes variation first to climate (growing degree days).

wetland ecosystem types by Liu and Graever (2009). The analysis included 68 publications that contained 208 observations across North and South America, Europe and Asia. The overall results showed that the effect on net ecosystem CO2 exchange for non-forest ecosystems (grassland, wetland and tundras) was not statistically significant (very large differences), while the effect on ecosystem net carbon storage for forest ecosystems showed on average a statistically significant 6% increase, with annual Nr additions ranging from 25 to 200 kg Nr ha−1 yr−1. On average, forest ecosystems sequestered 24.5 ± 8.7 kg CO2-C ha−1 yr−1 per kg Nr ha−1 yr−1 (−89.8 ± 32.0 kg CO2 equivalents ha−1 yr−1) added to the ecosystem. Note that these results cannot be extrapolated to systems with very high Nr inputs, nor to other ecosystems such as peatlands, where the impact of Nr is much more variable, and may range from C sequestration to C losses. It should be noted that although a study of forest chronoÂ� sequences (Magnani et€al., 2007) might appear to give a very high response of ~400 g CO2-C per kg Nr (Hogberg, 2007), part of this high value can be related to the need to account for dry Nr deposition, in which case the results show a prima facae response of ~177 g CO2-C per g Nr depositon (de Vries et€al., 2008; Sutton et€al., 2008). While this remains a high value, it may be explained by spatial covariation with climatic differences between sites, accounting for which gave a smaller estimate of 68 g CO2-C per g Nr deposition (Sutton et€al., 2008). Nevertheless, even that lower estimate is high compared with the other forest studies shown in Table 19.4. Overall, the mean forest C responses in Table 19.4 for above- and below-ground are 25 (20–30) and 15 (14–17) g CO2-C per g Nr input, respectively, amounting to an overall response of 41 (35–47) g CO2-C per g Nr input. (This would equate to 47 (33–61) g CO2-C per g Nr if the total responses shown were used, including the higher dry-deposition-corrected estimate of Magnani et€al., 2007, and Sutton et€al., 2008.) Taking the spatial distribution of forests Nr deposition into account, the total (wet+dry, reduced+oxidized) Nr deposition over European forests has increased from 1860 to 2000 by 1.5 Tg Nr yr–1 over a forest area of 188 Mha (Dentener et€al., 2006, Zaehle et€al., 2010). This forest area is somewhat larger than reported in the forest statistics for EU-27 as this estimate

448

accounts for all forested and woody land cover types, including those not reported in forest statistics. The increase in Nr deposition implies a mean increase of ~7.9 kg Nr ha–1 yr–1. Note that this estimate refers to above-canopy deposition velocities, not Nr catch by an ecosystem. Based on this Nr deposition rate, and the above reported NEP response (24.5 ± 8.7 g C g−1 Nr from Liu and Graever, 2009), C sequestration due to Nr deposition would average 36.6 ± 13.0 Tg C yr−1 for Europe. This figure is in good agreement with recent simulations by Zaehle et€al. (unpublished) using the OCN model (Zaehle et€al., 2010) and same rates of Nr deposition and forest cover changes. Such a calculation resulted in a net forest uptake rate due to Nr deposition of 23.5 ± 8.5 Tg C yr−1 (mean and standard deviation for the years 1996–2005), compared to an estimated net C uptake of 140 Tg C yr−1 resulting from the historical changes in atmospheric [CO2], climate, Nr addition (deposition and fertilizer application) and land-cover change. In addition, the Nr effect on unmanaged grasslands accounts for a further sink of 2.8 Tg C yr−1. This mechanistic model implies an average C storage of 37 g CO2-C g−1 Nr across the sites considered by Magnani et€al. (2007), with a range of 2–72 g CO2-C g−1 Nr, and a comparably low average response for European forests of 14.5 CO2-C g−1 Nr, due to the interaction with other growth limitations in this model (e.g., from water shortage in Southern Europe). The simulations with OCN suggest that Nr deposition has played only a minor role in terrestrial C cycling prior to the 1950s, after which the effect increased to the mid 1980s. The effect has thereafter remained relatively constant with some inter-annual variations related mainly to the interactions of Nr availability with climatic variability. It is obvious that there remains significant uncertainty in the overall response of ecosystem carbon sequestration to Nr inputs. While the very high forest C response to Nr observed by Magnani et€al. (2007) is an outlier, there remain significant differences between the estimate of Liu and Greaver (2009) at 24.5 ± 8.7 g C g–1 Nr and the mean of the values summarized in Table 19.4, at 41 (35–47) g CO2-C per g Nr input, indicating the need for further research. Key uncertainties for forests Â�concern the sensitivity of the C response to the amount, form and manner of Nr input (i.e., non-linear response with dose, dosing frequency, NOy vs NHx, above- vs below-canopy addition, etc.) and the

Klaus Butterbach-Bahl, Eiko Nemitz and Sönke Zaehle

successional status of the forest, including the biophysical and biochemical legacies of prior land-use history. Nitrogen additions also alters carbon cycling in agricultural systems due to positive effects on biomass yields, crop residues inputs to soil or stimulation of humus formation at high Nr availability (Christopher and Lal, 2007). On the other hand, �fertilizer N has been shown to promote the decomposition of crop residues and soil organic matter (Khan et€ al., 2007). In their metadata-analysis Liu and Greaver (2009) found that Nr additions to agricultural soils may slightly increase soil organic carbon stocks by an average of 2%. However, this estimate is based on a limited number of studies and, thus, cannot consider the variability of the chemical composition of the SOC pool across sites and regions, so that the estimate remains highly �uncertain. Only if N is added to the soil in the form of manure soil C€ sequestration may significantly increase (Christopher and Lal, 2007). In view of the rather uncertain mineral N effects on soil C stocks we therefore assumed that they are neutral.

19.5╇ Effects of reactive nitrogen through atmospheric chemistry Emissions of Nr compounds, in particular of NOx from combustion sources and fertilized soils, and ammonia (NH3) from livestock wastes and fertilizers, exercise multiple indirect effects on the climate system, through their participation in atmospheric chemistry. Several recent studies have attempted to attribute climate forcing components to emissions of pollutants, including NOx and NH3. The IPCC (2007) report does this in a simplified manner, without taking into account chemical interactions, while the more recent study of Shindell et€al. (2009) provides an improved allocation to emissions.

19.5.1╇ NOx and changes in atmospheric oxidation capacity Nitrogen oxides affect the oxidation capacity of the atmosphere:€NO2 undergoes photolysis which results in the formation of ozone (O3). Tropospheric O3 represents one of the most important greenhouse gases. The radiative forcing (RF) due to long-lived greenhouse gases (CO2, CH4, N2O) does not vary significantly with location of its emission, so that the RF can be estimated from the emissions alone. By contrast, the lifetime of O3 is much shorter and highly variable (from days to many weeks). Furthermore, the O3 production per molecule of NOx emitted depends non-linearly on other emissions, primarily of CO and non-methane hydrocarbons (NMHCs). To complicate matters further, the RF of O3 also depends on its vertical distribution. As a consequence, the efficiency of NOx in producing RF through O3 production depends greatly on where this NOx is emitted. Increases in O3 (associated with NOx emissions) also lead to increases in OH. Similarly, NO reacts with HO2 to form OH directly. The concentration of OH governs the lifetime of CH4, and thus the increase in OH has had a negative RF effect by reducing the CH4 burden. Similarly, the increase in the CH4 burden since pre-industrial times has increased the photochemical

production of O3, an effect which the IPCC attributes to CH4 (rather than O3) and which accounts for about 20% of the RF of CH4. The net RF of NOx is therefore the result of competing effects and will differ depending on where and when the NOx is emitted, which makes it much harder to estimate the contribution of continents to the RF. For example, aircraft emissions appear to be particularly efficient in generating ozone, while spiking experiments have found the net RF from NOx spikes is negative for January and July, and positive for April and October (Berntsen et€al., 2005). A few modelling studies have tried to quantify the response of greenhouse gases to changes in NOx emissions on different continents. Some of these were perturbation experiments where the aim was to find out the response to relatively modest changes in present-day emissions, to estimate the efficacy of abatement strategies. The extrapolation to zero anthropogenic emissions is somewhat uncertain due to non-linearities in the chemical system. Derwent et€ al. (2008) estimated that a January emission pulse of 2 Tg NO2 in Europe results in global time-integrated emission responses of −0.6 and +0.013 ppb years for CH4 and O3, respectively. Thus, a sustained cut of all anthropogenic NOx emissions in Europe (5.0 Tg Nr yr−1) would increase the global average concentration of CH4 by approximately +6.5 ppb years and decrease that of O3 by −0.14 ppb years. Applying a relationship between RF and concentration of 0.37 mW m−2ppb−1 for CH4 (Schimel et€al., 1996) results in an estimate of the RF of current European NOx emissions of€–2.4 mW m−2 through the effect on reducing global CH4 and a net effect of +1.4 mW m−2 (+2.0 mW m−2 short-term and −0.6 mW m−2 long-term) through the effect on O3. Model results indicate that the sensitivity of CH4 and O3 to NOx increases with increasing NOx concentrations and thus the extrapolation of a 2 Tg NO2 pulse to a€–5 Tg yr−1 would tend to underestimate the overall effect. Also, emission spikes in one single month (January) may not be representative for the average effect over the year (Berntsen et€al., 2005). It should be noticed that industrial NOx emissions originate from the same sources as emissions of CO and NMHC. We here estimate the effect of the NOx emissions alone. However, without these emissions, emissions of CO and NMHC would also be lower and the O3 RF forcing responds very sensitively to the combined change. Table 19.5 summarizes the different estimates of the global average RF due to European anthropogenic emissions of NOx through their effect on tropospheric O3 and CH4 lifetime. Here, the O3 RF is the relatively small difference between the shortterm cooling effect (through reducing O3 in the short term) and the long-term warming effect (through reducing the oxidative capacity in the longer term; see Derwent et€al., 2008).

19.5.2╇ NOx and tropospheric O3 feedbacks on plant growth Elevated tropospheric ozone, resulting largely from NOx and VOC emissions, is well-known to cause reductions in the growth

449

Nitrogen as a threat to the European greenhouse balance Table 19.5 Summary of estimates of the global RF due to the effect of European anthropogenic NO x emissions on CH4 and O3 (mW m−2)

Model

Study

CH4 RF

O3 RF

Total

Comment

LMDzINCA

Berntsen et€al. (2005)

−5.0

+4.3

−0.7

Calculated as −5× estimated response to 1 Tg pulse

Oslo-CTM2

Berntsen et€al. (2005)

−4.2

+5.5

+1.3

Calculated as −5× estimated response to 1 Tg pulse

MOZART-2

Naik et€al. (2005)

−6.7

+0.3

−6.4

Derived as 10× the estimated response to a 10% cut

STOCHEM

Derwent et€al. (2008)

−2.4

+1.4

−1.0

Derived here (see text)

−4.6

+2.9

−1.7

Mean

of crops and tree species, and changes in species composition in experimental grasslands (cf. Karnosky et€ al., 2007; Fuhrer, 2009). By inhibiting wood production, ozone directly leads to reduced aboveground C storage in forests (Wittig et€al., 2009) and through its effects on plant C assimilation and C allocation belowground, ozone could affect soil carbon sequestration, and thus soil C storage. Ozone may change decomposition proÂ� cesses in the soil through effects of ozone on plant residue mass and on the concentration of nutrients, secondary metabolites, lignification and/or the C/N ratio of above- and below-ground plant parts, in combination with indirect effects on soil microbial communities (Kim et€al., 1998; Kanerva et€al. 2008; Chen et€al., 2009). Results from short-term experiments are, however, not consistent, and it is thus difficult to draw a consistent picture of the impact of ozone on belowground C cycle processes. While reduced relative C allocation to roots leading to lower root litter production is generally observed, consequences of ozone stress for long-term C stabilization involving microbial processes are less clear. In aspen (Populus tremuloides) and in mixed aspen-birch (Betula papyrifera) stands, Loya et€ al. (2003) observed that ozone strongly inhibited extra stable soil C formation from elevated CO2. Under wheat and soybean, elevated ozone caused a change in soil C quality towards high molecular weight and more aromatic components (Islam et€al., 1999). While biogeochemical model simulations consistently project reduced soil C sequestration and C stocks (Ren et€al., 2007; Sitch et€al., 2007), long-term field studies of soil C under elevated ozone are lacking. Sitch et€al. (2007) estimated that global gross primary production is projected to decrease in 2100 as compared to 1901 by 8%–23% owing to plant ozone damage. However, there are still large uncertainties since counteracting effects of changes in atmospheric O3 and CO2 concentration cannot easily be delineated. As a first approach we assumed that by the year 2000 ecosystem sequestration in Europe was already reduced by 4%–11.5% due to increased atmospheric O3 concentration, thereby neglecting regional differences in C sequestration and atmospheric O3. We applied this assumption to ecosystem CO2 fluxes in EU 25, which were approximately −125 Tg C per yr in the period 2000–2005 (Schulze et€al., 2009). Thus, contemporary O3 concentrations may already have reduced total C sequestration in Europe by 5–14 Tg C per yr as an upper bound (Table 19.3).

450

19.5.3╇ Nitrous oxide and stratospheric ozone Forster et€ al. (2007) assess the total radiative forcing from observed changes in stratospheric ozone to be −0.05+/−0.1 W m−2 (medium confidence), subject to the spatial and temporal distribution of the changes, with mean decreases relative to pre-1980s values of 6% and 3%, in the Southern and Northern Hemisphere, respectively. Only a rather small fraction of this decline could potentially be attributed to the observed increases in atmospheric N2O concentrations, given the model experiments by Nevison and Holland (1997), however, such an assertion is fraught with uncertainty due to the complexities of stratospheric ozone chemistry. Given that the EU-27 currently emits around 10.6% of the global N2O (Table 19.7) the approximate effect of anthropogenic Nr increase from the EU-27 is uncertain, but likely to be comparatively small.

19.5.4╇ Nitrogen containing aerosols Aerosols are known to have a cooling effect on the climate through direct scattering of sunlight (the direct effect). Emissions of NH3 and NOx are associated with the formation of aerosol sulphates and nitrates. Ammonia neutralizes sulphuric acid (H2SO4) to produce ammonium bisulphate and ammonium sulphate aerosols, which are stable salts (although they may revolatilize through cloud processing (Bower et€al., 1997). Oxidation of NOx results in the production of HNO3 which interacts with NH3 to form ammonium nitrate (NH4NO3), which establishes a dynamic equilibrium with the gas precursors:€ the aerosol phase favoured in cold, humid conditions and high gas concentrations. Any NH4NO3 may therefore revolatilize into NH3 and HNO3 if temperatures rise, relative humidity drops or if the gas concentrations decrease. Sulphates and NH4NO3 form in the accumulation mode (0.1–1 μm). This aerosol mode undergoes the slowest removal from the atmosphere and therefore survives longest, providing the largest surface area for the condensational processes to occur. This size range is highly efficient in scattering light and therefore contributes to the direct forcing of aerosols. Nitric acid also interacts with coarse (sub-micron) crustal and seasalt aerosol to form calcium and sodium nitrate (Ca(NO3)2, NaNO3), respectively. Thus there are several proÂ� cesses by which the emission of Nr compounds impacts on the direct RF of aerosol (Table 19.6).

Klaus Butterbach-Bahl, Eiko Nemitz and Sönke Zaehle Table 19.6 Summary of different estimates of the anthropogenic direct aerosol forcing global and averaged over Europe. These values implicitly include neutralization by ammonia

Global (W m−2) Model

Study

Sulphate

Nitrate

GCM II-prime

Adams et€al. (1999)

−0.96

−0.19

GATORG

Jacobson (2001)

−0.32

−0.05

GISS

Liao and Seinfeld (2005)

−0.57

GEOS-CHEM v5.03

Martin et€al. (2004)

−0.36

−0.069

Oslo CTM-2

Myhre et€al. (2004, 2005)

−0.37

−0.02

GISS

Bauer et€al. (2007)a

−0.34

−0.11

a

European average (W m−2)

European contribution to global (mW m−2)

Sulphate

Nitrate

Sulphate

Nitrate

−0.34

−26.5

−11.1

−1.36

−0.81

European values:€S. Bauer, personal communication.

Modulation of H2SO4 production from SO2 As discussed above, NOx emissions affect OH concentrations and therefore the main source of H2SO4 in the atmosphere, production through SO2 oxidation. Shindell et€al. (2009) attribute a global RF of −0.13 W m−2 to the effect of NOx emissions on SO42−. Regional contributions of this effect do not appear to have been derived. A first estimate of the European effect is here derived by multiplication of this RF estimate with the ratio of the total European/global SO42− RF. In large parts of North America and China the aerosol is often observed to be acidic (i.e. sulphate exists as bisulphate or even sulphuric acid), and the same holds for European areas affected by higher SO2 emissions or subject to warm temperatures (e.g., Eastern and Southern Europe or Northern Scandinavia; Nemitz et€al., 2010). By contrast, in Central and Western Europe sulphate now tends to be fully neutralized by NH3 and in these areas, sulphuric acid competes with nitric acid for the ammonia. As a consequence a reduction in H2SO4 would increase the amount of NH3 available for NH4NO3 formation, which would at least partially offset the reduction in (NH4)2SO4. Thus the total aerosol RF over Europe is likely to be less sensitive to H2SO4 production than at the global average.

Neutralization of sulphuric acid Sulphuric acid condenses to form aerosol even in the absence of NH3. However, any NH3 present is readily incorporated into this aerosol neutralizing sulphuric acid to bisulphate and eventually to sulphate. These three forms of sulphate have different optical properties and can hold different amounts of water, leading to different effective aerosol sizes. Jacobson (2001) attributed a global tropopause forcing of +0.06 W m−2 to anthropogenic ammonium, by reducing the water holding capacity of the aerosol compared with H2SO4 at a given RH. Martin et€ al. (2004) modelled the optical properties of the NH4+–SO42−–NO3− system based on the two different branches of the deliquescence/ efflorescence curve. Because the atmospheric lifetime of sulphate is quite long (days to weeks; Garland, 2001), regional sulphate fields cannot easily be linked to regional emissions. As a

first estimate, we derive the contribution of European anthropogenic emissions by multiplying the global value by the ratio of European to global anthropogenic SO2 emissions (Smith et€al., 2001), which does not consider the fact that the deposition and oxidation of SO2 will vary in different parts of the world.

Coarse nitrate formation Adding only a thin layer to the existing aerosol, the reaction of HNO3 with coarse aerosol does not significantly change the radiative properties of this size mode, which is already less important for the climate system.

Ammonium nitrate formation Formation of NH4NO3 exercises an RF contribution that can be clearly attributed to Nr emissions. Fewer model studies have attempted to estimate the direct forcing of nitrates, compared with sulphates, and the general picture is that, globally, the nitrate effect is about ¼ of the sulphate effect. NH4NO3 forms where large emissions of NH3 and NOx (from agricultural and combustion sources, respectively) co-exist at relatively low temperatures, and it volatilizes in remote areas with an atmospheric life time of hours to days. Thus, fine nitrate RF is regionally variable, with Europe representing one of the major global hotspots, together with China and parts of the USA. Figure 19.3 demonstrates that the nitrate field over Europe can be clearly attributed to Europe itself and concentrations decrease towards the edge with little intercontinental transport. This allows the regional nitrate RF to be linked to the European emissions. Thus the regional top-of-the-atmosphere forcing of nitrate averaged over Europe, scaled to the globe, provides a reasonable estimate of the contribution of European NH3 and NOx to the global RF, although there is likely to be some net export of NO3− to the east, which is not captured by this approach. See also Figure 19.4.

Ammonium and nitrate indirect effects In addition, to the direct effect, aerosols exercise a number of indirect effects, e.g., by modulating the number of cloud condensation nuclei, with associated increases in cloud albedo

451

Nitrogen as a threat to the European greenhouse balance

NO3 5 4 3 2 1 0.5

Figure 19.3 Modelled fine nitrate (NO3−) concentration over Europe (in μg m−3), from Schaap et€al.. (2004). Estimates of the European forcing from sulphates and nitrates are summarized in Table 19.6. While several studies have quantified the effect at the global scale, only the authors of the study of Bauer et€al. (2001) provided European averages on request, the fields of which are shown in Figure 19.4. The comparison with Figure 19.3 Illustrates that the global coarse resolution models at present have limited skill in reproducing the European nitrate fields and the European forcings therefore have to be treated with some caution.

(1st indirect effect), the cloud lifetime/thickness (2nd indirect effect) or changing the albedo of snow (mostly important for black carbon). These effects are poorly quantified for sulphate aerosol and no global quantification appears to exist for the contribution of nitrate aerosol. Li et€al. (2009) recently simulated the effect of nitrate indirect forcing for China and estimated a RF of −1.63 W m−2 in January and€–2.65 W m−2 in July. In addition to the effect on the Earth’s radiation balance, aerosols are thought to stimulate plant growth and thus carbon storage. Through scattering of light, fine particulates tend to increase diffuse radiation. This may result in an increase in ecosystem production, since photosynthesis seems to be more efficient under diffuse light conditions (Mercado et€al., 2009). Furthermore, increased availability of Nr also seems to have a positive impact on the reflectivity (albedo) of vegetation. This would imply an additional cooling effect, although the mechanism is yet not clear (Ollinger et€al., 2008). These two latter effects were not considerd here.

19.6╇ Integration:€comparing present trends with the past This section attempts to answer the question:€‘What is the overall climate effect of European nitrogen emissions?’ Although several approaches could be taken to answer this question, here the following constraints were considered:€(a) since the overall effect includes short-lived agents, the quantification should be based on the RF metric; (b) since the overall effect includes long-lived GHGs, the estimate needs to consider the effect on the global rather than European regional RF; and (c) the

452

estimate should illustrate how European N management could change the European contribution to climate change. In extension to the IPCC approach, the question is therefore refined to:€‘What has been the contribution of European anthropogenic nitrogen emissions to the atmosphere to the overall change in global radiative forcing (between 1750 and 2005) that can be attributed to European activity?’. To answer this question poses two additional challenges:€(i) the European contribution of the IPCC RF bar graph needs to be isolated, which does not appear to have been done before, and (ii) the present-day GWPs and RFs derived above need to be turned into RFs integrated over the 1750–2005 time window.

19.6.1╇ European contribution to the change in global RF 1750–2005 The RF of European GHG emissions is estimated by Â�applying the total global RF components in relation to the presentday contribution of European to total global emissions (Table€ 19.7). This does not take into account that this ratio has evolved over the time-scale that affects present-day concentrations of GHGs. Approximately 23% of the global jet fuel is currently sold in Europe (source:€USEIA, 2009) and we have therefore assumed that the same proportion of contrails can be attributed to European activity. The European effect of black carbon (BC) on snow was similarly estimated from the present day ratio of European to global BC emissions (Bond et€al., 2004). The change in surface albedo due to landuse change has been highly variable geographically, and largest albedo increases have been derived for North East US, Europe (with a possible decrease in the Iberian Peninsula), East Asia and Brazil (Mhyre et€al., 2005). However, once the geographical variability in available solar radiation to calculate the effect of RF is included, the effect is weighted towards the low latitudes. There is substantial disagreement in the spatial patterns of the RF due to surface albedo changes (cf. Hansen et€ al., 1998; Mhyre et€ al., 2005; Betts et€ al., 2006), but qualitative agreement that the albedo increase in Europe has been larger than the global average. Thus, we derive a first estimate of the RF due to European landuse change as twice the value expected by ratioing European to global landmass.

19.6.2╇ Conversion of nitrogen-related present day RF and GWP to change in RF for 1750–2005 Estimating the radiative forcing (RF) due to Nr deposition effects on terrestrial C storage cannot be done directly from present-day net exchanges, because of the long life-time of CO2 in the atmosphere. By contrast, the spatial patterns of net exchanges are of lesser importance. This implies that the combined impact of the historical net terrestrial, marine and fossil C exchanges with the atmosphere on atmospheric [CO2] need to be taken into account. Calculation based on the historical emission estimates from fossil fuel statistics (Boden et€al.,

Klaus Butterbach-Bahl, Eiko Nemitz and Sönke Zaehle Table 19.7 Contribution of Europe (OECD states) to total global anthropogenic emissions of CO2, CH4 and N2O and importance of Nr use for GHG emissions in Europe. Global and European (OECD states) data were taken from the EDGAR (2009) emissions database

CO2

CH4

N2O

Tg CO2

Tg CH4

Tg CO2-equiv.

Gg N2O

Tg CO2-equiv.

Global

29 913

321

7 382

12 208

3 613

EU (OECD countries)

3 861

16.4

376.4

1 299

384.6

%-share EU

12.9

5.1

5.1

10.6

10.6

Nr effect on GHG emissions (Table 19.3)

−34 to −163

0.4–1.0

8.3–23

910–2260

269–669

%-share in relation to total-EU emissions

0.01–0.05

2.2–6.1

100

100

Figure 19.4 Example model result of NO3− and SO42− top-of-the-troposphere forcing in W /m2 at the global and zoomed-in European scale (based on the data by Bauer et€al. (2007 kindly replotted by the authors)).

453

Nitrogen as a threat to the European greenhouse balance Table 19.8 Summary of best estimates of Nr global RF attributed to European anthropogenic emissions, and their uncertainty ranges (in mW m−2)

Effect

Best estimate

Min

Max

â•…â•…â•… Increase in terrestrial C sequestration due to atmospheric Nr depositiona

−19

−30

−8

â•…â•…â•… Decrease in terrestrial C sequestration due to tropospheric O3 from NOx

4.4

2.3

6.5

â•…â•…â•… Decrease in CH4 soil uptake due to atmospheric Nr deposition

+0.13

+0.03

+0.24

â•…â•…â•… Increase in atmospheric [N2O]

+17

14.8

19.1

Gas phase chemistry

−1.7

−6.4

+3.1

Biosphere interactions

N2O

â•…â•…â•… Reduction in CH4 lifetime

−4.6

−6.7

−2.4

â•…â•…â•… Tropospheric O3 production€– radiative effect

+2.9

+0.3

+5.5

Aerosol direct effects

−16.5

−27.5

−5.5

â•…â•…â•… Total sulphate effect

−5.4

−9.4

−1.4

â•…â•…â•… Increase in H2SO4 production from SO2

−10.1

N/A

N/A

â•…â•…â•… Neutralization of H2SO4

+4.7

N/A

N/A

â•…â•…â•… Coarse nitrate production

negligible

â•…â•…â•… NH4NO3 direct effect

−11.1

−18.1

−4.1

Aerosol indirect effects

No estimate

â•…â•…â•… Effect on cloud albedo

No estimate

â•…â•…â•… Effect on precipitation / cloud lifetime

No estimate

â•…â•…â•… Stimulation of plant growth through diffuse radiation

No estimate

a

This estimates accounts for the atmospheric Nr deposition to forests and unmanaged grasslands; it does not include the effect of agricultural fertilization with Nr.

2009) and marine (Le Quere et€al., 2009) and terrestrial forest (Zaehle et€al., 2010) modelling, suggests a net radiative cooling of 19 mW m−2 due to atmospheric Nr deposition effects on forests and natural grasslands from EU-27, compared with a total cooling effect of 74 mW m−2 from EU-27 terrestrial carbon storage, which is roughly an order of magnitude less than the radiative forcing from EU-27 fossil fuel emissions (Zaehle, unpublished results) (Table 19.8). The predicted C storage due to Nr deposition is estimated to be 23.48 Tg C yr–1 by forests and 2.79 Tg C yr−1 by unmanaged grasslands, which is lower than the estimate of 34–163 Tg C yr–1 derived for forests on the basis of Section 19.4 (cf. Table 19.7). This reflects the fact that the Nr-sensitivity in this mechanistic model lies at the lower end of the values found in the literature (Section 19.4). Consequently, the radiative effect predicted here must therefore be considered conservative. It should also be noted that the estimate of the carbon response of managed (agricultural) ecosystems is not included. A similar modelling approach was not available to estimate the effect of Nr related increases in ozone concentrations on plant growth over the 1750–2005 time period. A first estimate of the effect of +4.4 mW m−2 was derived by comparison of the reduced C storage of −5 to 14 Tg C per yr estimated in Section 19.5.2 to the increased C storage due to Nr deposition predicted by the OCN model of +36.6 Tg C per yr (Section 19.4). This assumes that the time-lines in Nr and associated O3 damage have been similar.

454

Again in the absence of a time-integrating model calculation, a first estimate of the radiative forcing associated with reduced CH4 uptake by soils was derived from the ratio of the estimated reduction in CH4 uptake to the anthropogenic European emissions (Table 19.7), which led to a small warming of 0.13 mW m−2. The current European RF of Nr-related aerosol is taken as an approximation for the 1750–2005 change. Clearly, some Nr-related aerosol effects would have been in place in 1750 due to NOx and NH3 emissions from the natural biosphere and lightning. The resulting estimates of the European contribution to the change in global RF and the effects of Nr are summarized in Figure 19.5 and Table 19.8. The European contribution to the global RF is estimated to be about 410 mWâ•›m−2 equating to about 27% of the global anthropogenic change in RF. European Nr emissions to the atmosphere cause warming and cooling effects, which add to a net cooling effect of −15.7 mWâ•›m−2 which equates to a reduction of the European RF of about 4%. Because the indirect aerosol effects (e.g., cloud albedo effect) are not contained in this figure, the best estimate of the net cooling effect of Nr would be larger. However, since the errors associated with the individual components are significant, the range of possible values is estimated to be −46.7 to +15.4€mW€m–2, i.e. −11% to +4% of the total contribution. The estimates of the effects of Nr emissions to the atmosphere on (i) carbon sequestration, (ii) aerosols (sum of nitrate and sulphate) and (iii) N2O emissions are each estimated to be

Klaus Butterbach-Bahl, Eiko Nemitz and Sönke Zaehle

426 [382 to 469]

Fossil fuel & land use change CO 2 Biospheric CO 2 (incl. atmos. fertilization & O3 effect) Long-lived greenhouse gases

-74 [-86 - -62] -19 [-30 - -8 ] 4.4 [2.3 - 6.5] 24.5 [22-27 ] -4.6 (-6.7 - -2.4) 0.13 [0.03 - 0.24]

CH4 (decreased atmospheric lifetime & and decreased soil uptake)

N 2O Halocarbons Stratospheric

17.0 [14.8 - 19.1] 17.0 [14.8 - 19.1] 7.5 [4.5 - 10.5] <- 8 <- 1

Ozone Tropospheric Stratospheric water vapour from CH4

5.0 [2.0 - 8.0] 2.9 [0.3 - 5.5] 3.6 [1.0 to 6.1]

Land use

-38.3 [-76.6 - 0.0]

Surface albedo Black carbon on snow Sulphates Total aerosol

(SO2 oxidation & aerosol neutralization)

Direct effect Cloud albedo effect

9.9 [0 - 19.8 ] -26.5 [-16.5 to -36.5] -5.4 [-9.4 to -1.4]

Nitrate

-11.1 [-18.1 - -4.1] -11.1 [-18.1 - -4.1]

?

? <2

Linear contrails

409.7 [336.9 - 557.8] -15.7 [-46.7 - +15.4]

Total anthropogenic

-80

-60

-40

-20

0

20

100 200 300 400 500 600

European contribution to global radiative forcing [mW m -2 ] Figure 19.5 A first estimate of the change in global radiative forcing (RF) due to European emissions and the effect on European anthropogenic Nr emissions to the atmosphere. The RF components due to European anthropogenic activity have been derived as described in the text. The Nr effect is taken from Table 19.8. Red bars:€positive radiative forcing; light green bars:€positive radiative forcing due to direct/indirect effects of Nr; blue bars:€negative radiative forcing; dark green bars:€negative radiative forcing due to direct / indirect effects of Nr. For biospheric CO2, the dark green bar represents the additional CO2 sequestered by forests and grasslands due to Nr deposition, while the light green bar represents the decrease in productivity due to effects of enhanced O3 caused by NOx emissions. For CH4 the positive (not visible) and negative contributions represent the effects of Nr in reducing CH4 uptakes by soil and the decreased atmospheric lifetime, respectively.

about +/−â•›18 mW m−2 and are therefore of similar magnitude. While the emission of reactive nitrogen (NOx and NH3) has a net cooling effect, the effect of N2O emissions is obviously warming. However, it should be noted that some of this N2O emission is due to previous deposition of Nr from NOx and NH3 emissions. For comparison, IPCC (2007, Figure 2.21 and Table 2.13) represented a first estimate of the global RF attributed to global emissions (and their changes) of individual anthropogenic precursors). This includes the effects of N2O and NOx emissions

on direct GHG concentrations and atmospheric chemistry, which are quantified as +0.14 and −0.21 W m−2 respectively, but does not include the effect of NH3 emissions. A more detailed study (Shindell et€al., 2009) attributed a total cooling of −0.29 and −0.09 W m−2 to global emissions of NOx and NH3, respectively, the sum of which roughly balances the warming effect of N2O emission at the global level. However, unlike the estimate presented here, neither of these two global studies considered biosphere feedbacks through the promoting CO2 sequestration and ozone damage.

455

Nitrogen as a threat to the European greenhouse balance Table 19.9 Impacts of legislation on nitrogen-triggered atmospheric radiation effects and expectations beyond current implementation

Primary effect on

Relevance for radiative budget

Nitrate Directive

Water quality

High (limits N inputs to soils, thus also N2O emissions)

Water Framework Directive

Water quality

Emission Ceilings Directive/Gothenborg Protocol

Air quality€– NOx

Medium (ozone precursor)

Further abatement is technically feasible

Air quality€– NH3

Medium (precursor of cooling aerosol, but opportunities in coupling of Nr and C measures, e.g. improving N use efficiency)

Further abatement is technically feasible

Air quality€– PM (under discussion€– not implemented yet)

Low and highly uncertain (may be either increasing or decreasing radiative effects)

Extends beyond current legislation

Common Agricultural Policy

Market regulation instrument

Medium (decreased agricultural overproduction affects Nr compounds, C and net radiative forcing)

Decreasing agricultural production subsidies may give more weight to environmental initiatives

Kyoto Protocol/UNFCCC

Greenhouse gas emissions

High (but focus currently on CO2; potential for including Nr management options)

More direct focus on agriculture as a low-cost mitigation option seems realistic

See water framework directive Waters directive as such is challenging to implement

19.7╇ Future trends and mitigation opportunities Reactive nitrogen compounds will continue to affect the formation and burden of radiatively active compounds in the atmosphere of the future. Scenarios of these effects deal with the trends leading to the release of precursor compounds and with measures aimed at limiting environmental effects (see Winiwarter et€al., 2011, Chapter 24 this volume). The climate-relevant effects of Nr compounds have been outlined in the current chapter. They comprise not only the effects of atmospheric N2O, but also the indirect effects of Nr on atmospheric CH4, CO2, O3 and particles. Effects of any policy mitigating Nr emissions will therefore have legacy effects also for these other climatically relevant compounds. Conversely, because of the multiple interactions between these compounds, future emission limitation for any of these substances also relates to the respective contribution of Nr. The same is the case for future mitigation beyond current national or international agreements. In Table 19.9, we summarize the current international policy instruments, and their possible efficacy in relation to radiative effects and foreseeable future developments beyond current agreements. Overall, it should be noted that while reduction of Nr emissions as N2O is of direct benefit for European climate forcing, the situation is more complex for NOx and NH3 emissions. In the case of NOx, a number of cooling effects (e.g., N fertilization and aerosol effects) and warming effects (tropospheric ozone, indirect N2O emissions) apply, with Figure 19.5 suggesting an overall net cooling effect. For NH3 emissions, the main effects are cooling (N fertilization and aerosol effects), which will be larger than the warming effect (indirect N2O emissions).

456

Extension beyond current legislation

Legislation/policy

In principle, therefore, control of NOx and NH3 emissions for other environmental reasons, such as threats to biodiversity and air quality, needs to be considered against a ‘climate penalty’, as is already well established for the control of SO2 emissions. In this situation it is important to quantify the trade-offs between the different environmental effects of the Nr forms (see Brink et€al., 2011, Chapter 22 this volume). Nevetheless, it must be emphasized that, dispite the overall cooling effect of European NOx and NH3 emissions, it is expected that there are opportunities for reducing emissions of these pollutants that can lead to net climate benefits. This may be achieved where the measures concerned provide simÂ� ultaneous reductions for several pollutants. In the case of NOx emissions, the challenge is to identify measures that reduce both CO2 and NOx emissions, especially reducing overall fuel consumption. In the case of agriculture, reducing NH3 and nitrate losses is central to improving nitrogen use efficiency and to reduce soil N2O and NO emissions, with the potential to reduce fertilizer Nr consumption and the CO2 emissions associated with its manufacture. Similarly, measures to trap biogas from stored manures have co-benefits for both NH3 and CH4 emissions. These interactions highlight the need for integrated approaches to manage to Nr, climate and other issues, as discussed by Oenema et€al. (2011, Chapter 23 this volume). It should be noted that, a complete balance on the effect of European activities on radiation budgets should ideally also cover the consequences of European demand on the release of Nr compounds outside Europe and their conversion to substances affecting the radiation budget. Considering this ‘footprint’ would allow us to monitor leakages, i.e., export of emission intensive activities to other countries outside Europe while importing the products for less problematic further action. In

Klaus Butterbach-Bahl, Eiko Nemitz and Sönke Zaehle

many respects, the future development is easier to understand for the whole globe than for a sub-region (see Erisman et€al., 2008, who have projected the global future of the application and release of nitrogen). This is beyond the possible discussion here, especially as abatement measures beyond Europe cannot be covered.

19.8╇ Research needs This chapter demonstrates that Nr emissions significantly affect biospheric and atmospheric processes that are of importance for the Earth’s radiative balance in a number of ways. The effects of Nr additions to our biosphere on radiative balance are either direct€– as in the case of N2O€– or (in most cases) indirect. That is to say that, in addition to the stimulation of N2O emissions from Nr use (e.g., from soils, water bodies, or during sewage and waste treatment), other effects of Nr for ecosystem productivity and, hence, biosphere–atmosphere exchange of CO2, biosphere–atmosphere CH4 exchange and atmospheric chemistry (aerosol production, O3 chemistry, atmospheric lifetime of CH4) are likely even more important in terms of their effect on the European radiative forcing balance. The estimates of the individual components presented here have a large uncertainty. Taking the importance and contribution of individual processes to the European radiative balance into account, the uncertainty may be ranked as follows:€aerosol Nr effects > N2O emissions from soils and water bodies > Nr effects for biospheric CO2 sequestration > Nr effects for O3 chemistry > Nr effects on CH4 exchange and atmospheric lifetime. These large uncertainties can only be reduced through targeted research, thereby overcoming classical sectorial research, e.g., by linking research communities working on atmospheric chemistry, terrestrial processes or aquatic nutrient cycling. In particular, although we have focused on the role of atmospheric emissions, Nr has even wider effects on society. Anthropogenic Nr affects world food and feed supplies, livestock numbers and human population, all of which would need to be considered in addressing the overall effect of Nr on climate balance. It is essential that future research projects should explicitly address the importance of Nr for the radiative balance. This implicitly includes that a new view and perception of the importance of nitrogen for the radiative balance needs to be established, e.g., by specifically account for Nr effects in any greenhouse gas reporting (e.g., Kyoto protocol). Until now, the focus has been predominantly on carbon. Yet, it is obvious that nitrogen is a key threat for the European and global climate and, therefore, Nr research should become more fully integrated into the scientific assessment, societal debate and environmental policies.

Acknowledgements Several of the authors acknowledge research funding from the European Commission for the integrated project NitroEurope IP, together with supporting travel funds from the Nitrogen in Europe (NinE) programme of the European Science Foundation and the COST Action 729. The work was co-funded by many national organizations, including the German National Science

Foundation, and the UK Department for Environment Food and Rural Affairs, the Karlsruhe Institute of Technology, the Max Planck Society and the UK Natural Environment Research Council.

References Aber, J. D., Magill, A., Boone, R. et€al. (1993). Plant and soil responses to chronic nitrogen additions at the Harvard Forest, Massachusetts. Ecological Applications, 3, 156–166. Aber, J., McDowell, W., Nadelhoffer, K. et€al. (1998). Nitrogen saturation in temperate forest ecosystems:€hypotheses revisited Bioscience, 48, 921–934. Adams, P. J., Seinfeld, J. H. and Koch, D. M. (1999). Global concentrations of tropospheric sulfate, nitrate, and ammonium aerosol simulated in a general circulation model. Journal of Geophysical Research – Atmospheres, 104, 13â•›791–13â•›823. Agren, G. I., Bosatta, E. and Alison, H. M. (2001). Combining theory and experiment to understand effects of inorganic nitrogen on litter decomposition. Oecologia, 128, 94–98. Bange, H. W. (2006). Nitrous oxide and methane in European coastal waters. Estuarine, Coastal and Shelf Science, 70, 361–374. Barnes, J. and Owens, N. J. P. (1999). Denitrification and nitrous oxide concentrations in the Humber estuary, UK, and adjacent coastal zones. Marine Pollution Bulletin, 37, 247–260. Bauer, G. A., Bazzaz, F. A., Minocha, R. et€al. (2004). Effects of chronic N additions on tissue chemistry, photosynthetic capacity, and carbon sequestration potential of a red pine (Pinus resinosa Ait.) stand in the NE United States. Forest Ecology and Management, 196, 173–186. Bauer, S. E., Kouch, D., Unger, N. et€al. (2009). Nitrate aerosols today and in 2030: a global simulation including aerosols and tropospheric ozone. Atmospheric Chemistry and Physics, 7, 5043–5059. Beauchemin, K., Kreuzer, M., O’Mara, F. and McAllister, T. (2008). Nutritional management for enteric methane abatement:€a review. Australian Journal of Experimental Agriculture, 48, 21−27. Berendse, F., Van Breemen, N., Rydin, H. et€al. (2001). Raised atmospheric CO2 levels and increased N deposition cause shifts in plant species composition and production in Sphagnum bogs. Global Change Biology, 7, 591–598. Bergh, J., Linder, S., Lundmark, T., and Elfving, B. (1999). The effect of water and nutrient availability on the productivity of Norway spruce in northern and southern Sweden. Forest Ecology and Management, 119, 51–62. Berntsen, T. K., Fuglestvedt, J. S., Joshi M. M. et al. (2005). Response of climate to regional emissions of ozone precursors:€sensitivities and warming potentials. Tellus, 57B, 283–304. Berounsky, V. M. and Nixon, S. W. (1990). Temperature and the annual cycle of nitrification in waters of Narrragansett Bay. Limnology and Oceanography, 35, 1610–1617. Betts, R. A., Goldewijk, K. K. and Rmankutty, N. (2006). Radiative forcing by anthropogenic surface albedo change before and since 1750. Hadley Centre Technical Note 70. The Met Office, Brackwell, UK. Billen, G., Somville, M., DeBecker, E. and Servais, P. (1985). A€nitrogen budget of the Scheldt hydrographical basin. Netherlands€Journal of Sea Research, 19, 223–230. Billen, G., Garnier, J., Ficht, A. and Cun, C. (2001). Modeling the response of water quality in the Seine river estuary to human activity in its watershed over the last 50 years. Estuaries, 24, 977–983.

457

Nitrogen as a threat to the European greenhouse balance Billen, G., Garnier, J., Nemery, J. et€al. (2007). A long-term view of nutrient transfers through the Seine river continuum. Science of the Total Environment, 375, 80–97. Billen, G., Thieu, V., Garnier, J. and Silvestre, M. (2009). Modelling the N cascade in regional watersheds:€the case study of the Seine, Somme and Scheldt rivers. Agriculture, Ecosystems and Environment, 133, 234–246. Bodelier, P. L. E. and Laanbroek, H. J. (2004). Nitrogen as a regulatory factor of methane oxidation in soils and sediments. FEMS Microbiology and Ecology, 47, 265–277. Boden, T., Marland, G. and Andres, R. J. (2009). Global CO2 Emissions from Fossil-Fuel Burning, Cement Manufacture, and Gas Flaring:€1751–2006. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, Oak Ridge, TN. Boeckx, P. and Van Cleemput, O. (2001). Estimates of N2O and CH4 fluxes from agricultural land in various regions of Europe. Nutrient Cycling in Agroecosystems, 60, 35–47. Bond, T. C., Streets, D. G., Yarber, K. F. et€al. (2004). A technologybased global inventory of black and organic carbon emissions from combustion. Journal of Geophysical Research, 109, D14203. Borken, W. and Matzner, E. (2004). Nitrate leaching in forest soils:€an analysis of long-term monitoring sites in Germany. Journal Plant Nutrition and Soil Science, 167, 189–196. Bouchard, V., Frey, S. D., Gilbert, J. M. and Reed, S. E. (2007). Effects of macrophyte functional group richness on emergent freshwater wetland functions. Ecology, 11, 2903–2914. Bouwman, A. F., Lee, D. S., Asman, W. A. H. et€al. (1997). A global high-resolution emission inventory for ammonia. Global Biogeochemical Cycles, 11, 561–587. Bower, K. N., Choularton, T. W., Gallagher, M. W. et€al. (1997). Observations and modelling of the processing of aerosol by a hill cap cloud. Atmospheric Environment, 31, 2527–2544. Boynton, W. R., Garber, J. H., Summers, R. and Kemp, W. M. (1995). Inputs, transformations and transport of nitrogen and phosphorus in Chesapeake Bay and selected tributaries. Estuaries, 8, 285–314. Bradford, M. A., Ineson, P., Wookey, P. A. and Lappin-Scott, H. M. (2001). The effects of acid nitrogen and acid sulphur deposition on CH4 oxidation in a forest soil:€a laboratory study. Soil Biology and Biochemistry., 33, 1695–1702. Bragazza, L., Freeman, C., Jones, T. et€al. (2006). Atmospheric nitrogen deposition promotes carbon loss from peat bogs. Proceedings of the National Academy of Sciences of the USA, 103, 19386–19389. Brink, C., van Grinsven, H., Jacobsen, B. H. et€al. (2011). Costs and benefits of nitrogen in the environment. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Brumme, R. and Beese, F. (1992). Effects of liming and nitrogen fertilization on emissions of CO2 and N2O from a temperate forest. Journal of Geophysical Research, 97, 851–858. Brumme, R. and Khanna, P. K. (2008). Ecological and site historical aspects of N dynamics and current N status in temperate forests. Global Change Biology, 14, 125–141. Brumme, R., Borken, W. and Finke, S. (1999). Hierarchical control on nitrous oxide emission in forest ecosystems. Global Biogeochemical Cycles, 13, 1137–1148. Bubier, J. L., Moore, T. R. and Bledzki, L. A. (2007). Effects of nutrient addition on vegetation and carbon cycling in an ombrotrophic bog. Global Change Biology, 13, 1168–1186. Butterbach-Bahl, K. (2002). Exchange of trace gases at the forest floor atmosphere interface:€CH4. In:€Trace Gas Exchange in Forest Ecosystems, R., Gasche, H. Papen, and H. Rennenberg, (eds.) Kluwer, Dordrecht, The Netherlands, pp. 141–156.

458

Butterbach-Bahl, K., Gasche, R., Breuer, L. and Papen, H. (1997). Fluxes of NO and N2O from temperate forest soils impact of forest type, N deposition and of liming on the NO and N2O emissions. Nutrition Cycles and Agroecosystems, 48, 79–90. Butterbach-Bahl, K., Gasche, R., Huber, C., Kreutzer, K. and Papen, H. (1998). Impact of nitrogen input by wet deposition on N-trace gas fluxes and CH4-oxidation in spruce forest ecosystems of the temperate zone in Europe. Atmospheric Environment, 32, 559–564. Butterbach-Bahl, K., Gundersen, P., Ambus, P. et€al. (2011). Nitrogen processes in terrestrial ecosystems. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Chen, Z., Wang, X., Feng, Z., Xiao, Q. and Duan, X. (2009). Impact of elevated O3 on soil microbial community function under wheat crop. Water, Air and Soil Pollution, 198, 189–198. Christopher, S. F. and Lal, R. (2007). Nitrogen management affects carbon sequestration in North American cropland soils. Critical Reviews in Plant Science, 26, 45–64. Cole, J. J. and Caraco, N. F. (1998). Atmospheric exchange of carbon dioxide in a low-wind oligotrophic lake measured by the addition of SF6. Limnology and Oceanography, 43, 647–656. Cole, J. J., Peierls, B. L., Caraco, N. F. and Pace, M. L. (1993). Nitrogen loading of rivers as a human-driven process. In: M. J. McDonnell, and S. T. A. Pickett (eds.), Humans as Components of Ecosystems, Springer-Verlag, Berlin pp. 141–157. Conen, F. and Neftel, A. (2007). Do increasingly depleted δ15N values of atmospheric N2O indicate a decline in soil N2O reduction? Biogeochemistry, 82, 321–326. Corré, M. D., Pennock, D. J., van Kessel, C. and Elliott D. K. (1999). Estimation of annual nitrous oxide emissions from a transitional grassland-forest region in Saskatchewan, Canada. Biogeochemistry, 44, 29–49. Crutzen, P. J., Mosier, A. R., Smith, K. A. and Winiwarter, W. (2008). N2O release from agro-biofuel production negates global warming reduction by replacing fossil fuels. Atmospheric Chemistry and Physics 8, 385–389. Davidson, E. A. (2009). The contribution of manure and fertilizer nitrogen to atmospheric nitrous oxide since 1860. Nature Geoscience, 2, 659–662. De Vries, W., Reinds, G. J., Gundersen, P. and Sterba, H. (2006). The impact of nitrogen deposition on carbon sequestration in European forests and forest soils. Global Change Biology, 12, 1151–1173. De Vries, W., Oenema, O., Butterbach-Bahl, K. and Denier van der Gon, H. (2007). The impact of atmospheric nitrogen deposition on the exchange of carbon dioxide, nitrous oxide and methane from European forests. In:€Greenhouse Gas Sinks, ed. D. Reay, N., Hewitt, J. Grace, and K. Smith, CAB International, Wallingford, UK, pp. 249–283 De Vries, W., Solberg, S., Dobbertin, M. et€al. (2008). Ecologically implausible carbon response? Nature, 451, E1–E3. De Vries, W., Solberg, S., Dobbertin, M. et€al. (2009). The impact of nitrogen deposition on carbon sequestration by European forests and heathlands, Forest Ecology and Management, 258, 1814–1823. De Vries, W., Leip, A., Reinds, G. J. et€al. (2011). Geographic variation in terrestrial nitrogen budgets across Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Denier van der Gon, H. A. C. and Bleeker, A. (2005). Indirect N2O emission due to atmospheric N deposition for the Netherlands. Atmospheric Environment, 39, 5827–5838.

Klaus Butterbach-Bahl, Eiko Nemitz and Sönke Zaehle Dentener, F., Drevet, J., Lamarque, J. F. et€al. (2006). Nitrogen and sulfur deposition on regional and global scales:€a multimodel evaluation, Global Biogeochemical Cycles, 20. Derwent, R. G., Stevenson, D. S., Doherty, R. M. et€al. (2008). Radiative forcing from surface NOx emissions:€spatial and seasonal variations, Climatic Change, doi:€10.1007/s10584–007–9383–8. Dise, N. B. and Verry, E. S. (2001). Suppression of peatland methane emission by cumulative sulfate deposition in simulated acid rain. Biogeochemistry, 53, 143–160. Dise, N. B., Matzner, E. and Gundersen, P. (1998). Synthesis of nitrogen pools and fluxes from European forest ecosystems. Water, Air and Soil Pollution, 105, 143–154. Dutaur, L. and Verchot, L. V. (2007). A global inventory of the soil CH4 sink. Global Biogeochemical Cycles, 21, GB4013. EDGAR (2009). Emission Database for Global Atmospheric Research (EDGAR), release version 4.0, European Commission, Joint Research Centre (JRC)/Netherlands Environmental Assessment Agency (PBL). http://edgar.jrc.ec.europa.eu (site accessed January 2010). EEA European Environment Agency (2009). http://dataservice.eea. europa.eu (Site accessed, January 2010.) Eggleston, H. S., Buendia, L., Miwa, K., Ngara, T. and Tanabe, K. (2006). IPCC Guidelines for National Greenhouse Gas Inventories, prepared by the National Greenhouse Gas Inventories Programme, IGES, Hayama, Japan. Erisman, J. W., Sutton, M. A., Galloway, J., Klimont, Z. and Winiwarter, W. (2008). How a century of ammonia synthesis changed the world. Nature Geoscience, 1, 636–639. European Environment Agency (2006). Land accounts for Europe 1990–2000. Towards integrated land and ecosystem accounting, EEA Technical Report No. 11/2006, EEA, Copenhagen. European Environmental Agency (2008). Annual European Community greenhouse gas inventory 1990–2006 and inventory report 2008, EEA Technical Report No 6/2008, EEA, Copenhagen. Evans, C. D., Caporn, S. J. M., Carroll, J. A. et€al. (2006). Modelling nitrogen saturation and carbon accumulation in heathland soils under elevated nitrogen deposition. Environmental Pollution, 143, 468–478. FAO (2006). FAOSTAT. http://apps.fao.org (site accessed January 2010). Fog, K. (1988). The effect of added nitrogen on the rate of decomposition of organic-matter. Biological Reviews of the Cambridge Philosophical Society, 63, 433–462. Foley, J. A., Defries, R., Asner, G. P. et€al. (2005). Global consequences of land use. Science, 309, 570–574. Forster, P., Ramaswamy, V. and Artaxo, P. (2007). Changes in atmospheric constituents and in radiative forcing. In: Climate Change 2007:€The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, ed. S. Solomon et€al., Cambridge University Press. Franklin, O., Hogberg, P., Ekblad, A. and Agren, G. I. (2003). Pine forest floor carbon accumulation in response to N and PK additions:€Bomb C-14 modelling and respiration studies. Ecosystems, 6, 644–658. Freibauer, A. (2003). Regionalised inventory of biogenic greenhouse gas emissions from European agriculture. European Journal of Agronomy, 19, 135–160. Fuhrer, J. (2009). Ozone risk for crops and pastures in present and future climates. Naturwissenschaften, 96, 173–194. Fumoto, T., Kobayashi, K., Li, C., Yagi, K. and Hasegawa, T. (2008). Revising a process-based biogeochemistry model (DNDC) to

simulate methane emission from rice paddy fields under various residue management and fertilizer regimes. Global Change Biology, 14, 382–402. Galloway, J. N., Dentener, F. J., Capone, D. G. et€al. (2004). Nitrogen cycles:€past, present and future. Biogeochemistry, 70, 153–226. Garcia-Ruiz, R., Pattinson, S. N. and Whitton, B. A. (1998). Kinetic parameters of denitrification in a river continuum. Applied and Environmental Microbiology, 64, 2533–2538. Garland, J. A. (2001). On the size dependence of particle deposition. Water, Air and Soil Pollution, Focus, 1, 323–332. Garnier, J., Billen, G., Vilain, G. et€al. (2009). Nitrous oxide (N2O) in the Seine river and basin:€Observations and budgets. Agriculture Ecosystems and Environment, 133, 223–233. Groffman, P. M., Gold, A. J. and Addy, K. (2000). Nitrous oxide production in riparian zones and its importance to national emission inventories. Global Change Science, 2, 291–299. Gulledge, J. and Schimel, J. P. (1998). Moisture control over atmospheric CH4 consumption and CO2 production in diverse Alaskan soils. Soil Biology and Biochemistry, 8/9, 1127–1132. Gunnarsson, U., Bronge, L. B., Rydin, H. and Ohlson, M. (2008). Near-zero recent carbon accumulation in a bog with high nitrogen deposition in SW Sweden, Global Change Biology, 14, 2152–2165. Hagedorn, F., Spinnler, D. and Siegwolf, R. (2003). Increased N deposition retards mineralization of old soil organic matter, Soil Biology and Biochemistry, 35, 1683–1692. Hanaki, K., Hong, Z. and Matsuo, T. (1992). Production of nitrous oxide gas during denitrification of wastewater. Water Science and Technology, 26, 1027–1036. Hansen, J., Sato, M. and Ruedy, R. (1997). Radiative forcing and climate response. Journal of Geophysical Research, 102, 6831–6864. Hansen, J. E., Sato, M., Lacis, A. et€al. (1998). Climate forcings in the Industrial era. Proceedings of the National Academy of Sciences of the USA, 95, 12753–12758. Helder, W. and De Vries, R. T. P. (1983). Estuarine nitrite maxima and nitrifying bacteria (EMS-Dollard estuary). Netherlands Journal of Sea Research, 17, 1–18. Hindrichsen, I. K., Wettstein, H. R., Machmüller, A. and Kreuzer, M. (2006). Methane emission, nutrient degradation and nitrogen turnover in dairy cows and their slurry at different milk production scenarios with and without concentrate supplementation. Agriculture, Ecosystems and Environment, 113, 150−161. Hogberg, P. (2007). Nitrogen impacts on forest carbon. Nature, 447, 781–782. Högberg, P,. Fan, H., Quist, M., Binkley, D. and Tamm, C. O. (2006). Tree growth and soil acidification in response to 30 years of experimental nitrogen loading on boreal forest Global Change Biology, 12, 489–499. Howarth, R. W., Billen G., Swaney, D. et€al. (1996). Regional nitrogen budgets and riverine N and P fluxes for the drainages to the North Atlantic Ocean: Natural and human influences. Biogeochemistry, 35, 75–139. Hyvonen, R., Agren, G. I., Linder, S. et€al. (2007). The likely impact of elevated [CO2], nitrogen deposition, increased temperature and management on carbon sequestration in temperate and boreal forest ecosystems:€a literature review. New Phytologist, 173, 463–480. Hyvönen, R., Persson, T., Andersson, S. et€al. (2008). Impact of longterm nitrogen addition on carbon stocks in trees and soils in northern Europe. Biogeochemistry, 89, 121–137. IPCC (2002). Climate Change 2001 Synthesis Report. Cambridge University Press.

459

Nitrogen as a threat to the European greenhouse balance IPCC (2006). IPCC Guidelines for National Greenhouse Gas Inventories, prepared by the National Greenhouse Gas Inventories Programme, IGES, Hayama, Japan. IPCC (2007). Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the International Panel on Climate Change Cambridge University Press. Islam, K. R., Mulchi, C. L. and Ali, A. A. (1999). Tropospheric carbon dioxide or ozone enrichments and moisture effects on soil organic carbon quality. Journal of Environmental Quality, 28, 1629–1636. Islam, M. R. and Begum, J. (1997). Short review of global methane situation and of facilities to reduce in ruminants in third world countries. Asian-Australasian Journal of Animal Science, 10, 157–163. Jacobson, M. Z. (2001). Global direct radiative forcing due to multicomponent anthropogenic and natural aerosols. Journal of Geophysical Research, 106, 1551–1568. Jarvis, P. and Linder, S. (2000). Botany:€constraints to growth of boreal forests. Nature, 405, 904–905. Jaworski, N. A., Groffman, P. M., Keller, A. A. and Prager, J. C. (1992). A watershed nitrogen and phosphorus balance:€the upper Potomac River Basin. Estuaries, 15, 83–95. Jensen, L. S., Schjoerring, J. K., van der Hoek, K. et€al. (2011). Benefits of nitrogen for food fibre and industrial production. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Joshi, M., Shine, K., Ponater, M. et€al. (2003). Comparison of climate response of different radiative forcings in three general circulation models towards an improved metric climate change. Climate Dynamics, 20, 843–854. Kampschreur, M. J., Temmink, H., Kleerebezem, R., Jetten, M. S. M. and van Mark Loosdrecht, C. M. (2009). Nitrous oxide emission during wastewater treatment. Water Research (in press). Kanerva, T., Palojärvi, A., Rämö, K. and Manninen, S. (2008). Changes in soil microbial community structure under elevated tropospheric O3 and CO2. Soil Biology and Biochemistry, 40, 2502–2510. Karnosky, D. F., Skelly, J. M., Percy, K. E. and Chappelka, A. H. (2007). Perspectives regarding 50 years of research on effects of tropospheric ozone air pollution on US forests. Environmental Pollution, 147, 489–506. Kasimir-Klemedtsson, A., Klemedtsson, L., Berglund, K. et€al. (1997). Greenhouse gas emissions from farmed organic soils:€a review. Soil Use and Management, 13, 245–250. Kesik, M., Ambus, P., Baritz, R. et€al. (2005). Inventories of N2O and NO emissions from European forest soils. Biogeosciences, 2, 353–375. Khan, S. A., Mulvaney, R. L., Ellsworth, T. R. and Boast, C. W., (2007). The myth of nitrogen fertilization for soil carbon sequestration. Journal of Environmental Quality, 36, 1821–1832. Knorr, M., Frey, S. D. and Curtis, P. S. (2005). Nitrogen additions and litter decomposition a meta-analysis. Ecology, 86, 3252–3257. Kristensen, H. L., Gundersen, P., Callesen, I. and Reinds, G. J. (2004). Throughfall nitrogen deposition has different impacts on soil solution nitrate concentration in European coniferous and deciduous forests. Ecosystems, 7, 180–192. Kröber, T. F., Külling, D. R., Menzi, H., Sutter, F. and Kreuzer, M. (2000). Quantitative effects of feed protein reduction and methionine on nitrogen use by cows and nitrogen emission from slurry. Journal of Dairy Science, 83, 2941–2951. Külling, D. R., Menzi, H., Kröber, T. F. et€al. (2001). Emissions of ammonia, nitrous oxide and methane from different diet types of dairy manure during storage as affected by dietary protein content. Journal of Agricultural Science, 137, 235−250. Külling, D. R., Menzi, H., Sutter F., Lischer, P. and Kreuzer, M. (2003). Ammonia, nitrous oxide and methane emissions from differently

460

stored dairy manure derived from grass- and hay-based rations. Nutrient Cycling in Agroecosystems, 65, 13−22. Larsson, U., Elmgren, R. and Wulff, F. (1985). Eutrophication and the Baltic Sea:€–causes and consequences. Ambio, 14, 9–14. Laubhann, D., Sterba, H., Reinds, G. J. and de Vries, W. (2009). The impact of atmospheric deposition and climate on forest growth in European monitoring plots:€an individual tree growth model. Forest Ecology and Management, 258, 1751–1761. LeBauer, D. S. and Treseder, K. K. (2008). Nitrogen limitation of net primary productivity in terrestrial ecosystems is globally distributed. Ecology, 89, 371–379. Leip, A., Achermann, B., Billen, G. et€al. (2011). Integrating nitrogen fluxes at the European scale. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Le Quere, C., Raupach, M. R., Canadell, J. G. et€al. (2009). Trends in the sources and sinks of carbon dioxide. Nature Geoscience, 2, 831–836. Levy, P. E., Wendler, R., Van Oijen, M., Cannell, M. G. R. and Millard,€P. (2004). The effect of nitrogen enrichment on the carbon sink in coniferous forests:€uncertainty and sensitivity analyses of three ecosystem models, Water, Air and Soil Pollution, Focus, 4, 67–74. Li, S., Wang, T., Zhuang, B., and Han, Y. (2009). Indirect radiative forcing and climatic effect of the anthropogenic nitrate aerosol on regional climate of China. Advances in Atmospheric Sciences, 26, 354–362. Liao, H. and Seinfeld, J., (2005). Global impacts of gas-phase chemistry aerosol interactions on direct radiative forcing by anthropogenic aerosols and ozone. Journal of Geophysical Research, 110, D18208. Limpens, J., Berendse, F. and Klees, H. (2004). How phosphorous availability affects the impact of nitrogen deposition on Sphagnum and vascular plants in bogs. Ecosystems, 7, 793–804. Liu, L. and Greaver, T. L. (2009). Nitrogen addition stimulates emissions of biogenic greenhouse gases in terrestrial and wetland ecosystems. Ecology Letters, 12, 1103–1117. Loya, W. M., Pregitzer, K. S., Karberg, N. J., King, J. S. and Giardina, C. P. (2003). Reduction of soil carbon formation by tropospheric ozone under increased carbon dioxide levels. Nature, 425, 705–707. Magill, A. H., Aber, J. D., Currie, W. S. et€al. (2004). Ecosystem response to 15 years of chronic nitrogen additions at the Harvard Forest LTER, Massachusetts, USA. Forest Ecology and Management, 196, 7–28. Magnani, F., Mencuccini, M. et€al. (2007). The human footprint in the carbon cycle of temperate and boreal forests. Nature, 447, 849–851. Mannig, P., Saunders, M., Bardgett, R. D. et€al. (2008). Direct and indirect effects of nitrogen deposition on litter decomposition. Soil Biology Biochemistry, 40, 688–698. Martin, S. T., Hung, H.-M., Park, R. J. et€al. (2004). Effects of the physical state of tropospheric ammonium-sulfate-nitrate particles on global aerosol direct radiative forcing. Atmospheric Chemistry and Physics, 4, 183–214. Melin, J., Nommik, H., Lohm, U. and Flower-Ellis, J. (1983), Fertilizer nitrogen budget in a Scots pine ecosystem attained by using rootisolated plots and 15N tracer technique,. Plant and Soil, 74, 249–263. Mercado, L. M., Bellouin, N., Sitch, S. et€al. (2009). Impact of changes in diffuse radiation on the global land carbon sink. Nature, 458, 1014–1017. Myhre, G., Stordal, F., Berglen, T. F., Sundet, J. K. and Isaksen, I. S. A. (2004). Uncertainties in the radiative forcing due to sulphate aerosols. Journal of Atmospheric Science, 61(5), 485–498. Mhyre, G., Kvalevag, M. M. and Schaaf, C. B. (2005). Radiative forcing due to anthropogenic vegetation change based on MODIS surface albedo data. Geophysical Research Letters, 32, L21410.

Klaus Butterbach-Bahl, Eiko Nemitz and Sönke Zaehle Milne, R. and Van Oijen, M. (2005). A comparison of two modelling studies of environmental effects on forest carbon stocks across Europe. Annals of Forest Science, 62, 911–923. Mosier, A., Kroeze, C., Nevison, C. et€al. (1998). Closing the global N2O budget:€Nitrous oxide emissions through the agricultural nitrogen cycle, OECD/IPCC/IEA Phase II:€development of IPCC guidelines for national greenhouse gas inventory methodology. Nutrition Cycles Agroecosystems, 52, 225–248. Moss, A. R., Givens, D. I. and Garnsworthy, P. C. (1994). The effect of alkali treatment of cereal straws on digestibility and methane production by sheep. Animal Feed Science and Technology, 49, 245−259. Nadelhoffer, K., Giblin, A., Shaver, G. and Linkins, A. (1991). Microbial processes and plant nutrient availabilities in arctic soils. In:€Arctic Ecosystems in a Changing Climate, ed. F. C. Chapin, Academic Press, San Diego, CA, pp.€281–300. Nadelhoffer, K., Emmet, B., Gunderson, P. et€al. (1999). Nitrogen deposition make a minor contribution to carbon sequestration in temperate forests. Nature, 398, 145–148. Nadelhoffer, K. J., Colman, B. P., Currie, W. S., Magill, A. and Aber,€J.€D. (2004). Decadal-scale fates of N-15 tracers added to oak and pine stands under ambient and elevated N inputs at the Harvard Forest (USA). Forest Ecology and Management, 196, 89–107. Naik, V., Mauzerall, D., Horowitz, L. et€al. (2005). Net radiative forcing due to changes in regional emissions of tropospheric ozone precursors. Journal of Geophysical research - Atmospheres, 110, D24306. Naylor, R., Steinfeld, H., Falcon, W. et€al. (2005). Losing the links between livestock and land. Science, 310, 1621–1622. Nemitz, E., Dorsey, J. R., Flynn, M. J. et€al. (2009). Aerosol fluxes and particle growth above managed grassland. Biogeosciences, 6, 1627–1645. Nevison, C. D. and Holland, E. (1997). A reexamination of the impact of anthropogenically fixed nitrogen on atmospheric N2O and the stratospheric O3 layer. Journal of Geophysical Research, 102, 25519–25536. Nixon, S. W., Granger, S. L. and Nowicki, B. L. (1995). An assessment of the annual mass balance of carbon, nitrogen, and phosphorus in Narragansett Bay. Biogeochemistry, 31, 15–61. Oenema, O. and Tamminga, S. (2005). Nitrogen in global animal production and management options from improving nitrogen use efficiency. Science in China, Series C, 48, 871–887. Oenema, O., Oudendag, D. and Velthof, G. L. (2007). Nutrient losses from manure management in the European Union. Livestock Science, 112, 261–272. Oenema, O., Witzke, H. P., Klimont, Z., Lesschen, J. P. and Velthof,€G.€L. (2009). Integrated assessment of promising measures to decrease nitrogen losses from agriculture in EU-27. Agriculture Ecosystems and Environment, 133, 280–288. Oenema, O., Salomez, J., Branquisho, C. et€al. (2011). Developing integrated approaches to nitrogen management. In:€European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Ollinger, S. V., Richardson, A. D., Martin, M. E. et€al. (2008). Canopy nitrogen, carbon assimilation, and albedo in temperate and boreal forests:€functional relations and potential climate feedbacks. Proceedings of the National Academy of Sciences of the USA 105, 19336–19341. Olsson, P., Linder, S., Giesler, R. and Hogberg, P. (2005). Fertilization of boreal forest reduces both autotrophic and heterotrophic soil respiration, Global Change Biology, 11, 1745–1753.

Oren, R., Ellsworth, D. S., Johnsen, K. H. et€al. (2001). Soil fertility limits carbon sequestration by forest ecosystems in a CO2-enriched atmosphere. Nature, 411, 469–472. Owens, N. J. P. (1986). Estuarine nitrification: a naturally occuring fludized bed reaction? Estuarine, Coastal and Shelf Science, 38, 31–44. Papen, H. and Butterbach-Bahl K. (1999). A 3-year continuous record of nitrogen trace gas fluxes from untreated and limed soil of a N-saturated spruce and beech forest ecosystem in Germany, 1. N2O emissions. Journal of Geophysical Research, 104, 18487–18503. Pilegaard, K., Skiba, U., Ambus, P. et€al. (2006). Factors controlling regional differences in forest soil emission of nitrogen oxides (NO and N2O). Biogeosciences, 3, 615–661. Poorter, H. and Nagel, O. (2000). The role of biomass allocation in the growth response of plants to different levels of light, CO2, nutrients and water:€a quantitative review. Australian Journal of Plant Physiology, 27. Pregitzer, K. S., Burton, A. J., Zak, D. R. and Talhelm, A. F. (2008). Simulated chronic nitrogen deposition increases carbon storage in Northern Temperate forests. Global Change Biology, 14, 142–153. Reay, D. S. and Nedwell, D. B. (2004). Methane oxidation in temperate soils. Effects of inorganic N. Soil Biology and Biochemistry, 36, 2059–2065. Rehfuess, K. E., Agren, G. I., Andersson, F. et€al. (1999). Relationships between recent changes of growth and nutrition of Norway spruce, Scots pine and European beech forests in Europe In:€RECOGNITION€– Working Paper 19, European Forest Institute, Joensuu, Finland. Reis, S., Pinder, R. W., Zhan, M., Lijie, G. and Sutton, M. (2009). Reactive nitrogen in atmospheric emission inventories. Atmospheric Chemistry and Physics, 9, 7657–7677. Ren, W., Tian, H., Liu, M. et€al. (2007). Effects of tropospheric ozone pollution on net primary productivity and carbon storage in terrestrial ecosystems of China. Journal of Geophysical Research D., Atmospheres, 112, D22S09. Rennenberg, H., Kreutzer, K., Papen, H. and Weber P. (1998). Consequences of high loads of nitrogen for spruce (Picea abies)€and beech (Fagus sylvatica) forests. New Phytologist, 139, 71–86. Rich, J. J. and Myrold, D. D. (2004). Community composition and activities of denitrifying bacteria from adjacent agricultural soil, riparian soil, and creek sediment in Oregon, USA. Soil Biology and Biochemistry, 36, 1431–1441. Rodrigues, P. P. G. W., Barnes, J. and Upstill-Goddard, R. (2007). Simulating estuarine nitrous oxide production by means of a dynamic model. Marine Pollution Bulletin, 54, 164–172. Saarnio, S., Winiwarter, W. and Leitão, J. (2009). Methane release from wetlands and watercourses in Europe. Atmosph. Environm., 43, 1421–1429. Scarascia-Mugnozza, G., Bauer, G. A., Persson, H., Matteucci, G. and Masci, A. (2000). Tree biomass, growth and nutrient pools. In: (ed.) E. D. Schulze, Carbon and Nitrogen Cycling in European Forest Ecosystems, Springer, Berlin, pp. 49–60. Schaap, M., van Loon, M., Ten Brink, H., Dentener, F. J. and Bultjes, P.€J. H. (2004). Second inorganic aerosol simulations for Europe with special attention to nitrate. Atmospheric Chemistry and Physics, 4, 857–874. Schimel, D. S., Braswell, B. H., McKeown, R. et€al. (1996). Climate and nitrogen controls on the geography and timescales of terrestrial biogeochemical cycling. Global Biogeochemical Cycles, 10, 677–692. Schnell, S. and King, G. M. (1996). Responses of methanotrophic activity in soils and cultures to water stress. Applied Environmental Microbiology, 62, 3203–3209.

461

Nitrogen as a threat to the European greenhouse balance Schulze, E. D., Högberg, L., van Oene, H. et€al. (2000). Interactions between the carbon and nitrogen cycle and the role of biodiversity:€a synopsis of a study along a north–south transect through Europe. In:€Carbon and Nitrogen Cycling in European Forest Ecosystems, (ed.) E. D. Schulze, Springer, Berlin, pp.€468–492. Schulze, E. D., Luyssaert, S., Ciais, P. et€al. (2009). Importance of methane and nitrous oxide for Europe’s terrestrial greenhouse-gas balance. Nature Geoscience 2, 842–850. Seitzinger, S. P. and Kroeze, C. (1998). Global distribution of nitrous oxide production and N inputs in freshwater and coastal marine ecosystems. Global Biogeochemical Cycles, 12, 93–113. Seitzinger, S. P., Kroeze, C. and Styles, R. V. (2000). Global distribution of N2O emissions from aquatic systems:€natural emissions and anthropogenic effects. Global Change Science, 2, 267–279. Sheldrick, W., Syers, J. K. and Lingard, J. (2003). Contribution of livestock excreta to nutrient balances. Nutrient Cycling in Agroecosystems, 66, 119–131. Shindell, D. T., Faluvegi, G., Koch, D. M. et€al. (2009). Improved attribution of climate forcing to emissions. Science, 326, 716–718. Simpson, D., Butterbach-Bahl, K., Fagerli, H., Kesik, M. and Skiba, U. (2006). Deposition and emissions of reactive nitrogen over European forests:€a modelling study. Atmospheric Environment, 40, 5712–5726. Simpson, D., Aas, W., Bartnicki, J. et€al. (2011). Atmospheric transport and deposition of nitrogen in Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Sinsabaugh, R. L., Carreiro, M. M. and Repert, D. A. (2002). Allocation of extracellular enzymatic activity in relation to litter composition, N deposition, and mass loss. Biogeochemistry, 60, 1–24. Sinsabaugh, R. L., Gallo, M. E., Lauber, C., Waldrop, M. P. and Zak, D. R. (2005). Extracellular enzyme activities and soil organic matter dynamics for northern hardwood forests receiving simulated nitrogen deposition. Biogeochemistry, 75, 201–215. Sitch, S., Cox, P. M, Collins, W. J. and Huntingford, C. (2007). Indirect radiative forcing of climate change through ozone effects on the landcarbon sink. Nature, 448, 791–794. Skiba, U., Dick, J., Storeton-West, R. et€al. (2006). The relationship between NH3 emissions from a poultry farm and soil NO and N2O fluxes from a downwind forest. Biogeosciences, 3, 375–382. Smil, V. (2002). The Earth’s Biosphere:€Evolution, Dynamics and Change. MIT Press, Cambridge, MA. Smith, J. S., Pitcher, H. and Wigley, T. M. L. (2001). Global and regional anthropogenic sulfur dioxide emissions. Global Planetary Change, 29, 99–119. Solberg, S., Dobbertin, M., Reinds, G. J. et€al. (2009). The impact of changes in atmospheric deposition and climate on forest growth in European monitoring plots:€an empirical stand growth model. Forest Ecology and Management, 258, 1735–1750. Stehfest, L. and Bouwman, L. (2006). N2O and NO emission from agricultural fields and soils under natural vegetation: summarizing avialable measurement data and modelling of global annual emissions. Nutrient Cycling in Agroecosystems, 74, 207–228. Steinfeld, H., Gerber, P., Wassenaar, T. et€al. (2006). Livestock’s Long Shadow: Environmental Issues and Options, FAO, Rome, Italy. Steudler, P. A., Bowden, R. D., Melillo, J. M. and Aber, J. D. (1989). Influence of nitrogen fertilization on methane uptake in temperate forest soils. Nature, 341, 314–316.

462

Sutton, M. A., Nemitz, E., Erisman, J. W. et€al. (2007). Challenges in quantifying biosphere–atmosphere exchange of nitrogen species. Environmental Pollution, 150, 125–139. Sutton, M. A., Simpson, D., Levy, P. E. et€al. (2008). Uncertainties in the relationship between atmospheric nitrogen deposition and forest carbon sequestration. Global Change Biology, 14, 2057–2063. Tallec, G., Garnier, J., Billen, G. and Gousailles, M. (2006). Nitrous oxide emissions from secondary activated sludge in nitrifying conditions of urban wastewater treatment plants:€effect of oxygenation level. Water Research, 40, 2972–2980. Tallec, G., Garnier, J., Billen, G. and Gousailles, M. (2008). Nitrous oxide emissions from denitrifying activated sludge of urban wastewater treatment plants, under anoxia and low oxygenation. Bioresource Technology, 99, 2200–2209. Tamminga, S. (2003). Pollution due to nutrient losses and its control in European animal production. Livestock Production Science, 84, 101–111. Tietema, A., Emmett, B., Gundersen, P., Kjnaas, O. and Koopmans, C. (1998). The fate of 15N-labelled nitrogen deposition in coniferous forest ecosystems. Forest Ecology and Management, 101, 19–28. USEIA (2009). US Energy Information Administration. http://tonto. eia.doe.gov/cfapps/ipdbproject/iedindex3.cfm Van Dijk, S. M. and Duyzer, J. H. (1999). Nitric oxide emissions from forest soils. Journal of Geophysical Research, 104, 15955–15961. Velthof, G. L., Oudendag, D., Witzke, H. P. et€al. (2009). Integrated assessment of nitrogen losses from agriculture in EU-27 using MITERRA-EUROPE. Journal of Environmental Quality, 38, 402–417. Vitousek, P. M. and Howarth, R. W. (1991). Nitrogen limitation on land and in the sea:€how it can occur. Biogeochemistry, 13, 87–115. Wamelink, G. W. W., van Dobben, H. F., Mol-Dijkstra, J. P. et€al. (2009a). Effect of nitrogen deposition reduction on biodiversity and carbon sequestration. Forest Ecology and Management, 258, 1774–1779. Wamelink, G. W. W., Wieggers, R., Reinds, G. J. et€al. (2009b). Modelling impacts of changes in carbon dioxide concentration, climate and nitrogen deposition on growth and carbon sequestration of Intensive Forest Monitoring plots in Europe. Forest Ecology and Management, 258, 1794–1805. Winiwarter, W., Hettelingh, J. P., Bouwman, L. et€al. (2011). Future scenarios of nitrogen in Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Wittig, V. E., Ainsworth, E. A., Naidu, S. L., Karnosky, D. F. and Long, S. P. (2009). Quantifying the impact of current and future tropospheric ozone on tree biomass, growth, physiology and biochemistry:€a quantitative meta-analysis. Global Change Biology, 15, 396–424. Zaehle, S. and Friend, A. D. (2010). Carbon and nitrogen cycle dynamics in the O-CN land surface model:€1. Model description, site-scale evaluation and sensitivity to parameter estimates, Global Biogeochemical Cycles, 24, GB 1005. Zaehle, S., Friend, A. D., Dentener, F. et€al. (2010). Carbon and nitrogen cycle dynamics in the O-CN land surface model:€2 The role of the nitrogen cycle in the historical terrestrial carbon balance, Global Biogeochemical Cycles, 24, GB 1006.

Chapter

20

Nitrogen as a threat to European terrestrial biodiversity Lead author: Nancy B. Dise Contributing authors: Mike Ashmore, Salim Belyazid, Albert Bleeker, Roland Bobbink, Wim de Vries, Jan Willem Erisman, Till Spranger, Carly J. Stevensand Leon van den Berg

Executive summary Nature of the problem • Biodiversity is the variability among living organisms, from genes to the biosphere. The value of biodiversity is multifold, from preserving the integrity of the biosphere as a whole, to providing food and medicines, to spiritual and aesthetic well-being. • One of the major drivers of biodiversity loss in Europe is atmospheric deposition of reactive nitrogen (Nr).

Approaches • This chapter focuses on Nr impacts on European plant species diversity; in particular, the number and abundance of different species in a given area, and the presence of characteristic species of sensitive ecosystems. • We summarise both the scientific and the policy aspects of Nr impacts on diversity and identify, using a range of evidence, the most vulnerable ecosystems and regions in Europe.

Key findings / state of knowledge • Reactive nitrogen impacts vegetation diversity through direct foliar damage, eutrophication, acidification, and susceptibility to secondary stress. • Species and communities most sensitive to chronically elevated Nr deposition are those that are adapted to low nutrient levels, or are poorly buffered against acidification. Grassland, heathland, peatland, forest, and arctic/montane ecosystems are recognised as vulnerable habitats in Europe; other habitats may be vulnerable but are still poorly studied. • It is not yet clear if different wet-deposited forms of Nr (e.g. nitrate, NO3− versus ammonium, NH4+) have different effects on biodiversity. However, gaseous ammonia (NH3) can be particularly harmful to vegetation, especially lower plants, through direct foliar damage. • There are some clear examples of reductions in faunal diversity that can be linked to Nr deposition, but overall, our knowledge of faunal effects is still limited. Changes to above-ground faunal communities probably occur primarily through changes in vegetation diversity, composition or structure. • Evidence is strong that ecological communities respond to the accumulated pool of plant-available N in the soil. Thus the cumulative load of enhanced Nr impacting an ecosystem is probably highly important. • Because of this response to cumulative inputs, it is likely that biodiversity has been in decline in Europe for many decades due to enhanced Nr deposition. Equally, full recovery in response to reduced Nr deposition is likely to be slow, especially in highly impacted ecosystems. In some cases recovery may require management intervention. • Exceedence of critical loads for nutrient nitrogen is linked to reduced plant species richness in a broad range of European ecosystems.

Major uncertainties/challenges • It is very likely that Nr deposition acts synergistically with other stressors, in particular climate change, acid deposition, and ground-level ozone; these synergies are poorly understood. • The nature and rate of recovery of biodiversity from nitrogen pollution is not well understood. The optimal strategy to restore a habitat, and exactly what this ‘restored’ habitat constitutes, are both hard to define. • As with many disciplines, communicating biodiversity science to stakeholders, and communicating stakeholder needs to scientists, requires continuing effort and improvement.

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

463

Nitrogen as a threat to biodiversity

Recommendations (research / policy) • Future research should focus on understanding the extent of the problem of atmospheric Nr-driven biodiversity decline both within and outside Europe, synergistic interactions between Nr deposition and other drivers (particularly climate change, land use, and other pollutants), the relative effects of reduced and oxidised N, rates of recovery, and cascades of impacts through the vegetation, soil biota (especially microbes), and above-ground fauna. • Nitrogen manipulation experiments should be continued, and new experiments initiated in vulnerable habitats, paying particular attention to areas with low Nr deposition. • A European-wide monitoring network covering a range of habitats should be initiated to provide information on the long-term effects of air pollution on biodiversity.

20.1╇ Overview Together with habitat conversion and climate change, the atmospheric deposition of reactive nitrogen (abbreviated Nr or N) has been recognised as one of the most important threats to global biodiversity (Sala et€al., 2000), and this threat is explicitly or implicitly a main driver behind many nitrogen pollution control policies. Nitrogen deposition can directly damage vegetation, eutrophy ecosystems, alter nutrient ratios in soil and vegetation, increase soil acidity, and exacerbate the impact of other stressors such as pathogens or climate change. These stressors in turn can reduce the abundance of susceptible flora and fauna and change the community composition in favour of more tolerant species, resulting in a reduction, or even loss, of some species from the local habitat. This chapter summarises the processes, evidence, models and policies concerning biodiversity reduction due to Nr in vulnerable terrestrial ecosystems in Europe. We will focus on vegetation because of the extensive body of research on plants and the general conclusions that can be drawn from these studies; however, effects on fauna will also be considered. Other chapters in the ENA describe impacts of nitrogen on the biodiversity of water bodies (Grizzetti et€al., 2011, Chapter 17 this volume), soil organisms, and agricultural ecosystems (both Velthof et€al., 2011, Chapter 21 this volume). We begin the chapter (Section 20.1) with an overview of the various levels of biodiversity and the main threats to terrestrial diversity, including nitrogen. We then briefly discuss the types of ecosystems in Europe that are the most vulnerable to biodiversity loss through atmospheric nitrogen deposition. In Section 20.2 we examine the processes by which nitrogen changes vegetation composition and diversity. Section 20.3 uses evidence from a variety of approaches to determine in more detail how nitrogen impacts sensitive terrestrial ecosystems in Europe, and the extent to which changes may have already occurred. Section 20.4 turns to the development of predictive models for testing the implications on biodiversity of different future nitrogen pollution and climate scenarios, using one model chain as an example. Finally, in Section 20.5 we briefly describe the major European legislation on biodiversity protection and on air pollution control, and evaluate whether the current habitat-based pollution control policies in Europe may be appropriate to encompass protection of biodiversity.

20.1.1╇ What is biodiversity? ‘Biodiversity’, a contraction of ‘biological diversity’, first appeared in print only in 1986 (Wilson, 1988). Since then, the term has achieved global recognition, with 2010 being designated by the

464

United Nations as the International Year of Biodiversity (CBD, 2010). At its simplest, biodiversity is ‘the variety of life, in all its many manifestations’ (Gaston and Spicer, 2004). This variety includes the diversity of genes, populations, species, communities and ecosystems (Mace et€al., 2005). The 1992 Convention on Biological Diversity states: Biological diversity means the variability among living organisms from all sources including, inter alia, terrestrial, marine and other aquatic ecosystems and the ecological complexes of which they are part; this includes diversity within species, between species and of ecosystems. (CBD, 1992)

Genetic diversity provides the basis for biological diversity. A level up, organismal, or species diversity is the most common application of the term ‘biodiversity’ (Figure 20.1). Species diversity can also be described at the compositional (e.g. families, orders), structural, or functional levels (Gaston, 1996). Finally, ecological or ecosystem diversity describes niches, habitats, ecosystems, biomes and, ultimately, the whole Earth (Gaston and Spicer, 2004). Most people would agree that species-rich, diverse ecosystems are intrinsically valuable. Increasingly, however, natural habitats are explicitly valued for the benefits or ‘services’ they provide to humans. These include provisioning services, such as food and water; regulating services, such as moderation of the impacts of floods, drought, land degradation, and disease; supporting services, such as soil formation and nutrient cycling; and cultural services, such as recreational, spiritual, religious and other non-material benefits (Mace et€al., 2005). Biodiversity has been particularly recognised as a repository of genetic diversity for future medical, industrial and food products, for its aesthetic value, and for the stability and resilience against stress that may be conferred by diversity in an ecosystem (Ehrlich and Ehrlich, 1992; Tilman and Downing, 1994; Tilman et€al., 1996). No single measure encompasses all of the elements of biodiversity:€ the measure used is appropriate to the scale of the investigation (e.g. within a single species, across ecosystems) and the purpose of the study (e.g. conservation of a rare species, defining threats to a biome). At the organismal level, species richness (the number of species in a defined area) integrates many different levels of biodiversity, and it is relatively easy to measure even at large scales (Gaston and Spicer, 2004; Figure€20.1). However, it does not account for how closely related species are to each other, nor the number or spatial distribution of individuals:€equal weight is given to a species occurring just once and a species that is dominant (Gotelli and Colwell, 2001). Since species richness usually refers to the mean number of species in a particular sampling unit, reduced species richness does not necessarily indicate a local extinction of any particular species,

Nancy B. Dise

across scales, such as being highest at the equator, or declining with increasing altitude (Gaston, 1996). However, drivers such as air pollution or land conversion can modify or reverse these trends. In Europe, species- and family richness increase as one moves south from tundra to boreal forest in Scandinavia, reaching its highest levels in the temperate broadleaf mixed forest of central and southern Europe, then declining slightly in the Mediterranean forest, woodland and scrub biomes (Williams et€al., 1997; Mace et€al., 2005).

20.1.2╇ Threats to terrestrial biodiversity in addition to atmospheric nitrogen deposition€– a brief overview

Figure 20.1 (Top) Species-rich mesotrophic grassland, Cricklade meadows, United Kingdom. (Bottom) Determining the number of species per quadrat in the same grassland. Photos:€N. Dise.

but it does mean that fewer individuals of some species occur in the landscape under investigation. Considering species richness alone may lead to overly optimistic conclusions about the health of an ecological community, because simply counting species does not identify the replacement of characteristic or protected species, and because in habitats where plant species numbers are typically low (such as heathlands or bogs) the number of species may not decline greatly under even highly unfavourable conditions. Species abundance describes how common a species is in an area, but it can be time-consuming to measure. For vegetation, both biomass and cover (the proportion of a defined area occupied by a particular species) can be useful measures of abundance. Indices such as the Shannon diversity index incorporate both species richness (with increasing values as the number of species increases) and relative abundance (with higher values in communities where species have similar abundance, as opposed to a small number of dominant species). Generally, species richness increases with increasing temperature and precipitation; this leads to predictable patterns

Within the timescale of relevance to ourselves (outside geologi� cal-scale events), terrestrial biodiversity is threatened almost exclusively by direct or indirect human activity. In addition to nitrogen deposition, other air pollutants can impact diversity, including ozone, and the deposition of sulfur, metals, and other acidifying compounds. Pollutants can also leach into groundwater or runoff and damage downstream ecosystems, sometimes for many years. Semi-natural habitats converted to agriculture are also often fertilised, and these nutrients can persist in the soil long after a site has been taken out of cultivation. Biodiversity is strongly threatened by habitat conversion. Direct habitat loss and degradation through human population growth and industrial expansion continue on a broad scale. In the UK, a widespread reduction in the frequency of birds, butterflies and plants has been explicitly related to loss of habitat (Thomas et€al., 2004). Consequences are not just confined to the immediately impacted area:€fragmentation of habitats has important consequences for biodiversity in the surrounding region (Bender et€al., 1998). Species adapted to disturbance, or invasive alien species (including pathogens) can change species composition, cause local extinction of native species, and alter habitats. In Europe there are a number of such problematic alien species, including rhododendron (Rhododendron ponticum) and mink (Mustela vison) (Usher, 1986). Impacts of climate change on species composition are already being detected in Europe, particularly in shifts northward of the range of many species, and a reduction in abundance of species adapted to colder climates (Walther et€al., 2002). These changes will likely accelerate. Many European plant species are at risk from climate change (Thuiller et€al., 2005); indeed Thomas et€al. (2004) use climate envelope models to predict that between six and eight percent of plant species in Europe could become extinct on the continent by 2050 due to the changing climate. The Mediterranean region is particularly vulnerable to climate change, especially increases in the frequency and severity of drought (Hampe and Petit, 2005; Thuiller et€al., 2005).

20.1.3╇ Ecosystem sensitivity and vulnerability to N deposition Ecosystems can be defined by both their sensitivity and their vulnerability to a stress such as enhanced nitrogen deposition.

465

Nitrogen as a threat to biodiversity

Sensitivity measures responsiveness:€in this case, how strongly an ecosystem responds to a particular level of nitrogen deposition. Vulnerability is the likelihood of response given the existing level of the driver:€a sensitive habitat in a high-N deposition region is more vulnerable than a sensitive habitat in a low-N deposition region. The major impacts of N deposition on terrestrial ecosystem diversity are through (1) eutrophication, (2) acidification, (3) direct foliar impacts, and (4) exacerbation of other stresses. Here we provide a brief overview of these impacts and introduce the major sensitive ecosystems in Europe; these are expanded upon in Sections 20.2 and 20.3. Since nitrogen limits (or, with P, co-limits) primary productivity in many terrestrial ecosystems in Europe, habitats most likely to be sensitive to eutrophication due to N include those with low levels of nitrogen in their soils and those characterised by stress-tolerant species unable to compete well with species better adapted to take advantage of additional nutrients (Bobbink et€ al., 1998). Ecosystems occurring on weakly buffered soils are most sensitive to acidification from nitrogen deposition. Lower plants that depend on atmospheric inputs as their primary source of nutrients, such as mosses (particularly Sphagnum), lichens, or liverworts, can be highly sensitive to direct impacts of N. Finally, foliar enrichment in nitrogen can leave a species vulnerable to predation or disease. Habitats at risk for biodiversity change through N deposition are sensitive to one or more of these processes. Many semi-natural grassland communities in Europe are dominated by species with low nutrient requirements, are sensitive to acidification, eutrophication, or both, and occur in areas with elevated N deposition. Ecosystems of cold climates, including montane, boreal, tundra, subarctic, and arctic habitats, are also vulnerable to nitrogen deposition. Many of these ecosystems are dominated by bryophytes and lichens, which can be highly sensitive to direct foliar nitrogen deposition. Nitrogen deposition can pose a threat even in remote areas through factors such as orographic enhancement of deposition at high altitudes, concentration of pollutants in fog or mist, or high levels of N deposited over a short period in snowmelt (Taylor et€al., 1999). Heathland communities are highly sensitive to N deposition due to nutrient-poor acidic soils, although their distribution, primarily in low N-deposition regions, generally makes them less vulnerable than other habitats. However, where their occurrence coincides with elevated N deposition, such as in the Netherlands, they have shown dramatic responses, with extensive loss of heather, and conversion to grassland (Bobbink et€al., 1998) (Figure 20.2). It is possible that the low occurrence of heathland in areas with elevated N deposition may in part reflect these ecosystems already converting to grassland. Wetland communities vary in their sensitivity to atmospheric nitrogen deposition depending on their water source. With a high dependence on rainwater and a very low nutrient status, ombrotrophic bogs and nutrient-poor fens can be very sensitive (Bergamini and Pauli, 2001). In Europe, these peatlands mainly occur in the north, and are vulnerable in areas of their range where N deposition is elevated, such as southern Scandinavia and northern continental Europe. Climate change

466

Figure 20.2 Nitrogen-degraded heath, Veluwe, Netherlands, with encroachment by the grass Molinia caerulea. Photo:€R. Bobbink.

is also predicted to be more pronounced in high-latitude regions (IPCC, 2007), potentially exacerbating the impact of Nr. Forests throughout Europe can be highly vulnerable to nitrogen deposition, particularly if they are on nutrientpoor soil:€ they have shown evidence for changes in biomass (Nellemann and Thomsen, 2001) and ground flora composition (Pitcairn et€ al., 1998) that can be related to ecosystem nitrogen enrichment due to elevated N deposition. The composition of the flora and fauna in many coastal habitats is mainly driven by salinity, making them normally less sensitive than other habitats to atmospheric nitrogen. However, as with other ecosystems, this depends in part upon the relative source of nutrients:€sand dune vegetation (receiving proportionally less N from the sea and more from the atmosphere) is known to be sensitive to nitrogen deposition (van den Berg et€al., 2005b) and is likely to be vulnerable in areas of higher N deposition. Shrubland communities typical of the Mediterranean region have not been well studied, although there are indications that their vegetation is sensitive to N deposition in combination with other factors such as drought or disturbance (Calvo et€al., 2005). The Mediterranean Basin is Europe’s only biodiversity hotspot (Myers et€al., 2000). The number of endemic plants and amphibians is very high in this region, and yet the impact of nitrogen deposition on typical ecosystems has received little research attention to date. In the mid 1990s, deposition of 10 kg N ha−1yr−1, widely considered as a threshold for nitrogen impacts (Bobbink et€ al. 2003), was exceeded in approximately 12% of the region; by 2050 it is estimated to be as high as 69% under a business-as-usual scenario (Phoenix et€al., 2006).

20.2╇ Processes A complex series of events occurs when N inputs increase in a region with initially low background N deposition, with many ecological processes interacting at different temporal and spatial scales (Figure 20.3). The main types of impacts are described

Nancy B. Dise

Indicates an increase Indicates a decrease Disturbance and stress factors Direct toxicity (NO2, NH3, NH4+) Susceptibility to pathogens and herbivory

Figure 20.3 Schematic of the main impacts of enhanced N deposition on ecosystem processes and species richness. Stress is considered to occur when external constraints limit the rate of production of vegetation; disturbance consists of mechanisms that affect plant biomass by causing its partial or total destruction.

Competition for light and resources Increased nitrogen deposition and concentration

Availability of N

N mineralisation

productivity

Litter production and quality Soil acidification

Species richness

N limitation → P limitation

Suppressed nitrification and + accumulation of NH4 Decrease in base cations and 3+ increase in metals (Al )

in this section, as well as the overall processes that characterise ecosystem responses. We conclude with a brief consideration of interactions between N deposition and other major humaninfluenced environmental drivers on biodiversity.

20.2.1╇ Direct and indirect impacts of N The severity of the impacts of atmospheric N deposition on a species or community depends upon a number of factors, primarily:€(1) the duration and total amount of the N input, (2) the chemical and physical form of N, (3) the intrinsic sensitivity of the plant and animal species present, (4) the local climate and other abiotic conditions (e.g. soil acid neutralising capacity, availability of other nutrients), and (5) the past and present land use or management. As a consequence, despite the same potential impacts (Figure 20.3), different ecosystems can show wide variability in sensitivity to atmospheric N deposition (Bobbink et€al., 2010).

Direct foliar impacts At high concentrations, nitrogenous gases, aerosols, and dissolved compounds can be directly toxic to the above-ground parts of plants: NO,€NO2, NH3 and NH4+ are especially phytotoxic. Studies have mostly concentrated on crops and saplings, but studies with native herbaceous or shrub species in open-top chambers have also demonstrated leaf injury, changes in physiology, and growth reductions at high concentrations of airborne N pollutants (Pearson and Stewart, 1993; Grupa, 2003). Direct foliar damage is due to high short-term concentrations of N, rather than broader ecosystem-scale changes (eutrophication, acidification) arising from the accumulation of nitrogen in the soil from long-term loads of N. Lichens are the most sensitive group of vegetation to direct toxicity from dry-deposited N, especially in the form of NH3 (Hallingbäck, 1992; Van Herk et€al., 2003). Direct toxic effects of wet-deposited N (primarily as NH4+) at fairly low deposition rates

have also been reported for bryophytes and lichens (Bates, 2002). Direct foliar impacts on trees were observed or inferred in some highly polluted forests in Europe in the 1980s (Nihlgård, 1985), but have become rare due to the closing or modernisation of local industrial sources and the success of pollution control legislation. However, concentrations of nitrogen-based air pollutants are increasing in parts of Asia (primarily in China and India), possibly leading again to direct foliar impacts.

Eutrophication N is the limiting nutrient for plant growth in many natural and semi-natural terrestrial ecosystems, especially under oligotrophic and mesotrophic conditions in Europe (globally, either N or P limitation, or co-limitation by both nutrients, is widespread; Elser et€al., 2007). In the short term (years), enhanced N deposition causes an increase in the availability of inorganic N in the topsoil. This leads to an increase in plant productivity in N-limited vegetation, both through increased growth of existing species and by invasion of new, more productive species. Over the long term (years to decades) litter production increases as a result. Because of this, the rate of N mineralisation will gradually increase, which may further increase plant productivity. This is a positive feedback, because higher N mineralisation leads to higher N uptake, etc. The rate of N cycling in the ecosystem is therefore accelerated, although the response time to enhanced N inputs can be long in organic soils with high C:N ratios, or in any soil with large potential N sinks. Above a certain level of primary productivity, local species diversity can decline as the production of a few species able to exploit the available N greatly increases. Competitive exclusion (‘overshading’) of characteristic species of oligotrophic or mesotrophic habitats by relatively fast-growing nitrophilic species occurs, with rare species at low abundance especially at risk (Bobbink et€al., 1998; Suding et€al., 2005) (Figure 20.4). Changes in species richness and composition are generally

467

Nitrogen as a threat to biodiversity Figure 20.4 A chalk grassland vegetation (Mesobromion erecti) in the Netherlands (left) without N addition and (right) after three years of N addition (100 kg N ha−1y−1 as NH4NO3) (from Bobbink, 1991).

long-term effects, although they may be induced by relatively large doses of nitrogen applied over a few years. When N is no longer limiting in these ecosystems, plant growth becomes limited by other resources such as water or phosphorus (P). In this situation, vegetation productivity will not increase with further increases in N. Nitrogen concentrations in plant tissues will, however, often increase because N availability still increases. This may affect the palatability of the vegetation for herbivores, or its sensitivity to pathogens. Even in ecosystems that are limited by other nutrients, long-term N inputs can lead to nutrient imbalances (e.g. anomalous N:P ratios) which could ultimately change plant species composition.

Acidification Both oxidised and reduced N can acidify soils:€ oxidised N through acting as a mobile anion accompanying basic cations leached from soil (Johnson and Cole, 1980), and reduced N through the acidifying effects of both nitrification and root exchange of NH4+ for H+ (Bolan et€ al., 1991). Soil acidification, or the loss of acid neutralising capacity, triggers many long-term changes (Ulrich, 1983). Owing to their high buffering capacity, calcareous soils will not at first change pH when exposed to acid (N and/or S) deposition:€pH generally remains above 6.5 until the soil calcium carbonate exposed to the acid is nearly depleted. In soils dominated by silicate minerals (pH 6.5–4.5), buffering is taken over by cation exchange processes on soil adsorption sites. In mineral soils with a large cation exchange capacity and high base saturation, this buffering may continue for several decades, even at relatively high acid inputs. Eventually, however, exchangeable basic cations can be depleted, leading to a decline in the soil pH to below 5. This causes the breakdown of clay minerals and the dissolution of hydrous oxides of several metals, resulting in elevated levels of these metals, especially reactive aluminium (Aln+), in the soil solution and soil exchange complex. As soil acidity increases and pH declines, the ecosystem’s capacity to remove nitrogen is compromised through reduced nitrification or plant uptake rates, often resulting in the accumulation of ammonium (NH4+) (Roelofs et€al., 1985). High soil acidity also reduces the decomposition rate of organic material, leading to increased accumulation of litter (Van Breemen et€al., 1982; Ulrich, 1983). As a result of this cascade of changes, plant growth and the species composition of sensitive vegetation can

468

be damaged:€acid-resistant plant species become dominant, and species typical of intermediate- and higher-pH soil disappear.

Susceptibility to secondary stress and disturbance The sensitivity of plants to stress (defined here as external constraints, such as drought, frost, pathogens or herbivores, that limit dry matter production rate) or disturbance (mechanisms causing the destruction of plant biomass) may be significantly affected by N deposition. With increasing N deposition, susceptibility to fungal pathogens and attacks by insects is enhanced. This is probably due to reduced concentrations of phenolic compounds (leading to lower resistance) and higher levels of soluble nitrogen compounds such as free amino acids (leading to higher palatability), together with the overall lower vitality of individual plants exposed to air pollution (Flückiger et€al., 2002). Increased levels of pathogenic fungi have been found for several tree and shrub species in N-addition experiments and field surveys, but for most ecosystems data are lacking and the influence of pathogens on diversity is still unclear (Bobbink et€al., 2003; Flückiger et€al., 2002). Herbivory in general is affected by the palatability of the plant material, which is primarily determined by its N content (Throop and Lerdau, 2004). Data on relationships between herbivory intensity and N deposition are scarce, but a link has been demonstrated in dry Calluna heathlands. Outbreaks of heather beetle (Lochmaea suturalis), which forages exclusively on the green parts of Calluna vulgaris, can occur in dry lowland heaths. Attacks of the beetle lead to the opening of closed C. vulgaris canopy, increasing light penetration in the vegetation and enhancing the growth of understorey grasses such as Deschampsia flexuosa or Molinia caerulea (Figure 20.2). The frequency and intensity of these outbreaks are clearly related to atmospheric N inputs and N concentrations in the heather, although the exact controlling processes are not clear (Brunsting and Heil, 1985; Berdowski, 1993; Bobbink and Lamers, 2002). N-related changes in plant physiology, biomass allocation (root/shoot ratios) and mycorrhizal infection can also influence the sensitivity of plant species to drought or frost stress, leading to reduced growth of some species and potential changes in plant interactions.

Sensitivity to nitrogen form Plant species composition may be affected by a change in the dominant form of nitrogen the ecosystem receives in deposition,

Nancy B. Dise

which may favour species better able to use that form. Since pollution control measures in Europe have been more successful in controlling emissions of oxidised N than reduced N (Oenema et€al. 2011, Chapter 4 this volume), in regions with a high rate of N deposition most of the nitrogen originates from NHx from agricultural activity (Asman et€ al., 1998; Fowler, 2002; Sutton et€al., 2008). This could cause a shift from NO3– to NH4+ in the soil, especially in habitats with low nitrification rates. Species of calcareous or moderately acidic soils are able to use NO3−, or a combination of NO3− and NH4+, as a nitrogen source, whereas early studies showed that species of acid habitats generally use NH4+ (Gigon and Rorison, 1972; Kinzel, 1982), because at least some of these plants do not have nitrate reductase (Ellenberg, 1996). Laboratory and field studies demonstrate that most understorey species of deciduous forests in southern Sweden are favoured when both NH4+ and NO3− can be taken up instead of only NH4+ (Falkengren-Grerup, 1998; Olsson and Falkengren-Grerup, 2000). Increased NH4+ uptake can lead to reduced uptake of basic cations (K+, Ca2+ and Mg2+), and exchange of these cations from the plant to the rhizosphere. Ultimately this can lead to severe nutritional imbalances, which have been implicated in the decline in tree growth in areas with high deposition of reduced N (Nihlgård, 1985; Van Dijk et€al., 1990; Bobbink et€al., 2003). High NH4+ concentrations in the soil solution or leaf water layer can also be toxic to many sensitive plant species, causing disturbed cell physiology, cell acidification, accumulation of N-rich amino acids, poor root development, and inhibition of shoot growth (Nihlgård, 1985). Strong evidence exists that several rare or threatened plant species of grassland, heathland, moorland, and soft-water lakes are intolerant to increased concentrations of reduced N and to high NH4+/NO3− ratios (De€Graaf et€al., 1998; Paulissen et€al., 2004; Kleijn et€al., 2008; Van den Berg et€al., 2008) (Figure 20.5).

20.2.2╇ Interactions between N deposition, other air pollutants and climate change Interactions with the effects of SOx deposition The acidifying effects of both S and N on soils and water may lead to the same pathway of changes, and the effects are difficult to separate for each pollutant in areas with both high S and N deposition. Thus, in many cases, observed increases in acid-resistant species and declines in acid-sensitive species can be caused by both airborne components. ‘Legacy’ pollution from high levels of sulfate deposition in the past can also predispose an ecosystem to greater sensitivity to N deposition. S deposition peaked in Europe in the 1980s and has since declined dramatically across the continent. However, some soils in highly impacted ecosystems, particularly in central and Eastern Europe, continue to show elevated levels of acidity and heavy metals, and depleted concentrations of basic cations, resulting from prolonged exposure or high loads of earlier pollution. In addition to the direct effects of acidity and metal toxicity on plant species composition, nutritional imbalances due to increasing N deposition will occur sooner and

Figure 20.5 Characterisation of habitats of common (blue diamonds) and rare (red squares) species typical of Dutch heathland, matgrass swards and fen meadows by means of pH and molar NH4+/NO3− ratio in the soil. Symbols indicate mean ± standard error. Almost all rare species occur only at low NH4+/ NO3− ratios (From Kleijn et al., 2008).

at lower N deposition rates in soils that have been depleted of basic cations. Finally, if rates of nitrification or plant uptake are impacted by soil acidity, the negative impacts of reduced N will be much larger, because inputs of NH3 and NH4+ will remain in the reduced form for a longer time in the soil solution, soil exchange complex, or surface water/groundwater.

Interactions with the effects of ozone A major pathway for the formation of tropospheric ozone (O3) is photochemical reaction with NOx; therefore, the two pollutants are closely interlinked. The complex chemical transformations that characterise the atmospheric reactions between NOx and O3 are outside of the scope of this chapter (see Hertel et€al., 2011, Chapter 9 this volume), but illustrate that the transformations and fates of atmospheric pollutants can rarely be considered in isolation. The impacts of O3 alone on the biodiversity of semi-natural and natural ecosystems are not well studied, but O3 fumigation has been shown to reduce the productivity of semi-natural vegetation, in some cases together with changes in species composition (Ashmore, 2005). Data on interactions between N deposition and O3 are scarce, and this is a major gap in knowledge. There is, however, at least one field investigation of these interactions. In a three-year experiment in sub-alpine grasslands in Switzerland, N deposition stimulated the productivity of the vegetation and altered the functional group composition, but O3 did not (Bassin et€al., 2007). Only one significant interaction between N deposition and O3 was found:€although N addition increased the chlorophyll content of the vegetation, this effect was counterbalanced by accelerated leaf senescence under high O3 concentrations. Over the longer term this interaction may express itself as an overall reduction in the growth and C assimilation of the community, but such higher-level impacts are still not demonstrated. Both N deposition and O3 may also reduce C transport to the roots, leading to a (possibly additive) lowering of the root:shoot ratios of plants (Ashmore, 2005), but again, these interactions have not yet been quantified in the field.

469

Nitrogen as a threat to biodiversity

Interactions with the effects of climate change Temperature and precipitation are the main determinants of the distribution of plants and animals. If climate changes, the biotic composition of ecosystems will also change. N deposition impacts will act together with changes in climate, but there is a major gap in knowledge concerning their interactive effects. In addition to changes in air temperature, which affect major ecosystem characteristics such as vegetation composition and productivity, shifts in the intensity and occurrence of precipitation, drought, frost, and fire will all interact with N deposition to impact diversity (Wiedermann et€al., 2007; Gerdol et€al., 2007). Many of these climate-related drivers that interact with N are also the most uncertain factors in climate change modelling. Even a superficial treatment of the potential additive, synergistic or antagonistic effects of climate change and nitrogen pollution on vegetation biodiversity is beyond the scope of this chapter. We do, however, return to this topic in Section 20.4 as we use an ecosystem model to explore the potential of these drivers separately and together to change the species composition of one habitat over time.

20.3╇ Evidence of change This section evaluates the evidence that Nr, at historic and current levels, both can and has caused a loss of biodiversity in vulnerable terrestrial ecosystems in Europe. We first consider the types of evidence linking N deposition with biodiversity change, and describe the most vulnerable European regions and habitats based on this evidence. We then use this evidence to demonstrate relationships and thresholds of biodiversity loss, discuss changes in diversity over the past 70 years, highlight some of the more sensitive species, and consider the rate and extent to which recovery may occur when N deposition levels decline.

20.3.1╇ Types of evidence There are three major types of evidence available to relate N deposition to biodiversity for terrestrial ecosystems. The first is from manipulation experiments, in which nitrogen deposition is increased, normally by application of NH4+ and/or NO3– in artificial rainwater. If significant changes are detected in the experimental treatments and not in the controls, it can be inferred with some confidence that N deposition is a primary driver of the change. Experiments can provide information on how long it takes for different components of a system to respond to N addition, and can be designed to assess interactions, for example with management intensity, temperature, or drought. Experiments can also identify thresholds for effects on biodiversity. However, experimental studies typically assess relatively short-term responses (even the longest experiments seldom exceed 20 years) and often use high concentrations of the applied pollutant, which may influence the response of the vegetation. In addition, it may be difficult to identify thresholds of response from experiments in areas with a relatively long history of elevated N deposition, where there may already have been significant impacts of N deposition on biodiversity. Finally, site-specific factors such as previous management might explain part of the observed response.

470

A second approach is through spatial field surveys of sites covering a gradient of nitrogen deposition. Targeted surveys (explicitly designed to test N deposition impacts) may use short but steep gradients of N deposition (e.g. close to intensive animal units) or have a regional, national or even continental focus. Surveys can provide insight into longer-term responses, can cover a wider range of nitrogen deposition than experiments, and avoid experimental artefacts. Since gradients of N deposition may be correlated with those of other potential drivers (e.g. S deposition, climate, or management intensity), these other drivers need to be measured and considered in analyses and interpretation. In addition, because they are correlative, targeted surveys cannot prove causality, but can often determine the statistical significance of N deposition as a potential driver of changes in biodiversity. Ecological surveillance networks can also be analysed for spatial relationships between diversity and N deposition. Surveillance surveys typically record the presence or absence of species in larger areas (e.g. 10 × 10 km squares). As they are usually not designed to specifically identify nitrogen deposition (or even pollution) impacts, such studies reflect the influence of land use and a range of climatic, edaphic and management factors. Attribution of any change to nitrogen deposition, therefore, can be even more difficult than in targeted surveys. However, surveillance surveys usually cover a wide region, and so can potentially detect signals of change in biodiversity at the national level, including effects on rare and scarce species. A third type of evidence to identify changes in community composition through N deposition is re-surveys over time of previous vegetation studies. Collating from the literature data collected over time from the same sites or from repeated surveys would also be included in this type of evidence, even if no new fieldwork is undertaken. The original survey may have been conducted for a variety of reasons, and since detection of N deposition impacts on biodiversity is rarely one of the reasons, re-surveys are often limited by the confounding influence of other factors. Attributing causes to vegetation changes detected in re-surveys is particularly vulnerable to changes in land use and, increasingly, climate, that have occurred over the intervening period. It also may be challenging to identify the exact sites that were studied many years ago. However, given the limited duration of most experiments, re-surveys are the only type of evidence that can directly identify changes occurring over long periods of time, and so are an essential component of the strategy to characterise N deposition impacts on vegetation community composition and diversity. Each of the above approaches has strengths and weaknesses, and these are often complementary. Multiple strands of evidence from a variety of approaches thus provide the most convincing support for N-driven changes in biodiversity.

20.3.2╇ Evidence of change by ecosystem As described in Sections 20.1 and 20.2, a wide range of ecosystems across Europe are sensitive to adverse effects of N deposition on biodiversity, particularly habitats with characteristically nutrient-poor conditions. Table 20.1 summarises the effects on plant

Nancy B. Dise Table 20.1 Effects of nitrogen deposition on plant biodiversity reported across the major bio-climatic zones in Europe

Habitat

Observed effects on plant biodiversity

Grassland

-Acid grassland:€Reduced species richness, particularly of forbs. -Calcareous grassland:€Change in species composition; reduced species richness in some experiments.

Forest

-Temperate:€Invasion of nitrophilic species; loss of epiphytic lichen species. -Boreal:€Decreased cover of ericaceous shrubs; decline of characteristic bryophytes.

Peatland

-Decline of characteristic bryophyte species. -Loss of sundew.

Heathland

-Loss of characteristic lichen species. -Invasion of nitrophilic acid grassland species. -Reduced species richness, particularly of bryophytes.

Arctic and montane

-Grasslands:€Increased cover of sedges, reduced proportional cover of grasses and forbs. -Heaths:€Reduction in cover and richness of lichens, and cover of mosses.

Coastal dune

-Reduction in species richness. -Increased grass growth. -Loss of lichen species.

Mediterranean ecosystems

-Forest:€Loss of sensitive lichen species. -Grassland and shrub:€Loss of native forb species.

Tundra

-Reduced cover of lichens. -Increased cover of vascular plants. -Changes in bryophyte species composition.

{ { {

{

{ { { {

Key references

Overall weight of evidence in Europe

Stevens et al. (2006s) Maskell et al. (2010s) Duprè et al. (2010r) Bobbink (1991e)

Strong for species-rich acid and calcareous grasslands of temperate regions; limited for others.

Nordin et al. (2005e, 2006e) Makipaa and Hiekkinen (2003s) Brunet et al. (1998s) Mitchell et al. (2005s)

Strong for boreal and temperate forests; limited for other forests.

Redbo-Tortensson (1994e) Mitchell et al. (2002e) Wiedermann et al. (2009e) Limpens et al. (2004e)

Strong, with a range of studies.

Barker (2001e) Heil and Diemont (1983e) Caporn et al. (2006e) Maskell et al. (2010s) Edmondson et al. (2010s)

Strong for temperate dry heaths, limited for others.

Bassin et al. (2007e) Pearce and van der Wal (2002e) Britton and Fisher (2007e)

Intermediate.

Jones et al. (2004s) Remke et al. (2009s) van den Berg et al. (2005ae)

Limited to a small number of studies.

Fenn et al. (2003s*, 2008s*) Weiss (1999s*)

Very limited.

Gordon et al. (2001e) Arens et al. (2008e)

Limited to a small number of studies.

e = Evidence from N-manipulation experiment, s = Evidence from spatial survey, r = Evidence from temporal re-survey, *Study is from outside Europe.

biodiversity in these environments, and gives examples of key studies that provide evidence of effects. In cases for which there is limited European evidence (i.e. Mediterranean ecosystems), Table 20.1 includes examples from similar systems in North America. Table 20.1 shows that there is a large amount of evidence of damage to European terrestrial biodiversity due to elevated nitrogen deposition, particularly for grassland, forest, peatland and heathland ecosystems. However, for some specific habitats, and for much of southern and eastern Europe, the evidence is very limited. The most impacted plant functional types are forbs, bryophytes, lichens and nutrient-poor shrubs; graminoids adapted to higher nutrient levels are the main beneficiaries of elevated N deposition. Many different environmental conditions are likely to modify the impacts of N deposition within these habitats. For instance, Clark et€ al. (2007) analysed 23 N-addition experiments in North America and found that species richness reduction was greatest where cation exchange capacity was low, temperature was low and the increase in primary production in response to N was greatest. Both Stevens et€al. (2004)

and Duprè et€al. (2010) showed that the relationship between N deposition and diversity in acid grasslands is modified by soil pH. A further critical factor may be co-limitation by other nutrients. In tundra ecosystems, for example, responses to enhanced N deposition are usually only observed when the P limitation typical of these systems is released (Gordon et€ al., 2001; Madan et€al., 2007), although Arens et€al. (2008) showed significant effects of N deposition alone. There is reason to believe that the evidence summarised in Table 20.1 provides a conservative estimate of the long-term impact of N deposition on European ecosystems. Much of the evidence (particularly N-addition experiments) originates from areas that have received elevated N deposition over the past 50–60 years, and where this cumulative high deposition may already have significantly affected biodiversity. Pardo et€al. (2010), assessing critical thresholds of N deposition for loss of biodiversity in North America, identified in many cases lower threshold values than for the equivalent ecosystems in Europe. This may reflect the lower rates of N deposition, especially in

471

Nitrogen as a threat to biodiversity

Species richness (n)

30

(a)

25 20 15 10 5 0 0

10 20 30 N deposition (kg N ha–1yr –1)

40

Species richness (n)

40 35

(b)

30 25 20 15

10 5 0 0

10

20 N deposition (kg N

30

40

50

ha–1yr –1)

Figure 20.6 (a) Species richness versus N deposition in a targeted survey of one acid grassland community in Great Britain (per 2 x 2 metre quadrat; linear and power functions shown). Adapted from Stevens et al. (2004). (b) Species richness in all acid grassland communities versus N deposition for the wide-ranging UK Countryside Survey (●  ) (linear regression râ•›2 = 0.09, p<0.001), shown together with the targeted survey of Stevens et al. (2004) (râ•›2 = 0.55, p<0.0001) (◯). From Stevens et al. (2009).

remote areas of the North American continent, which provide a stronger basis for identifying the full range of species that existed before significant N deposition began. It may also reflect the much longer history of intensive human use of European ecosystems, which could have caused a historical ‘nitrogen Â�deficit’ in soils long exploited for agriculture or forestry.

20.3.3╇ Field surveys:€identifying spatial relationships with N deposition Regional surveys can reveal spatial relationships between N deposition and biodiversity. As described above, surveys may or may not be specifically targeted to identify N impacts, and on their own cannot prove causality. However, by statistically accounting for the potential influence of other driving variables, and by linking results of spatial surveys to evidence from field manipulations and other sources, attribution of impacts to N can be made to varying degrees of confidence. The first regional survey specifically designed to identify potential pollutant impacts on vegetation species richness was carried out in acid grasslands in Great Britain by Stevens et€al. (2004). The species composition of 68 sites in a specific acid grassland community across Great Britain was related to 20 potential drivers on diversity measured on site or collated from available datasets.

472

Stevens et€al. (2004) showed a strong pattern of declining species richness with increasing nitrogen deposition (Figure 20.6a). Forbs were particularly affected:€over the deposition range measured across Great Britain, the species richness and cover of forbs declined by an average of 75%, from approximately 8 species/20% cover per m2 at low rates of N deposition to 1–2 species/5% cover at the highest rates of N (Stevens et€al., 2006). Grass species richness also declined with N deposition, and the cover of grasses showed a non-significant increasing trend (i.e. a higher abundance of fewer species). There was no relation with N deposition in either richness or cover for bryophytes. A subsequent comparison of these results with findings from a UK ecological surveillance survey (Maskell et€al., 2010) showed a similar significant relationship with N deposition for all UK acid grasslands (Figure 20.6b):€although the surveillance data, as expected, show greater scatter, the relationship with N deposition is highly significant and comparable to the targeted survey (Stevens et€al., 2009). The study of Stevens et€ al. (2004) was recently expanded to encompass acid grassland habitats across western Europe, and showed the same pattern of species richness decline with increasing N deposition (Stevens et€ al., 2010). Again, the decline in species richness was strongest for forbs, but grasses and bryoÂ�phyes showed stronger negative trends than in Great Britain alone (Figure 20.7). In addition to acid grassland, gradient surveys in Europe, primarily the UK (both targeted and surveillance), show significant negative relationships between N deposition and some component of biodiversity for forest, peatland, heathland, coastal dune, tundra, and arctic/montane ecosystems (Table 20.1; RoTAP, 2010), but the contribution of other drivers on diversity of these habitats has not yet been fully investigated. Evidence from gradient studies also suggests that the diversity of Mediterranean forests and grasslands in the US is impacted by N deposition, although these ecosystems have not been well studied in Europe. Recent research in the UK indicates that calcareous grasslands show changes in species composition, but not in species richness, over the N deposition gradient in the UK (van den Berg et€al., 2010; Maskell et€al., 2010). Some N-addition experiments have, however, also induced a species richness change in calcÂ� areous grasslands (e.g. Bobbink 1991, Figure 20.4). An important question in evaluating the overall weight of evidence for effects of N deposition on biodiversity in Europe is whether there is consistency between the findings from the different types of study identified. Although few direct comparisons exist, these suggest that the results are comparable. The underlying relationships between the three surveys of acid grasslands described above (Stevens et€ al., 2004, 2009, 2010) were all very similar, despite wide variability in the type of survey, specific community studied, and geographic region. For peatlands, Wiedermann et€ al. (2009) recently showed that the effects of N deposition in a three-year field experiment€– Â�reducing the cover of Sphagnum and increasing that of vascular plants€– were consistent with effects observed in a targeted survey along a national gradient of N deposition from the south to north of Sweden. This consistency of results across a variety of approaches provides strong support

Nancy B. Dise Figure 20.7 Species richness as a function of total inorganic nitrogen deposition for 153 acid grasslands across western Europe. Sites in green show Stevens et€al. (2004) survey. From Stevens et€al. (2010).

that N deposition is driving a reduction in diversity in these ecosystems. Because of variability in the data, particularly at low levels of N deposition, many of the relationships identified so far can be modelled equally well with a linear, power, or step function. The latter two functions often depict a greater rate of reduction in species richness with increasing N in less polluted environments (e.g. compare solid curve with dotted in Figure 20.6a). This better reflects the findings of manipulation experiments on ecosystems historically receiving relatively low levels of N deposition (Clark and Tilman, 2008; Bobbink et€al., 2010). The form of the relationship is important:€if the rate of species richness loss is higher at lower N deposition levels, pollution control policies to protect biodiversity should aim for limits on new sources of

atmospheric N, and a reduction of existing N sources, in areas where N deposition is currently low to intermediate.

20.3.4╇ Re-surveys:€change in diversity over time Several long-term ecological surveillance studies in Europe report a decline in species characteristic of low-nutrient conditions and an increase in nitrophilic plant species over recent decades, including botanical inventories in the UK, Spain and Portugal (Gimeno, 2009; Preston et€al., 2002). In the UK, the Countryside Survey, designed to investigate changes across the rural environment over time, has provided detailed information on vegetation across the nation since 1978 (Carey et€al., 2008). In three broad habitats (woodlands, grasslands, and

473

Nitrogen as a threat to biodiversity

Figure 20.8 Relationships between vascular plant species richness and cumulative N deposition (in kmol N ha−1) from analyses of acid grassland plot data from the 1940s onward in Germany, the Netherlands and Great Britain. From Duprè et al. (2010). Plots are standarised for comparison.

heathland/bogs), species richness was already significantly lower in areas with higher rates of N deposition in 1978. There has been little change in the subsequent 30 years, although some further decreases in biodiversity in areas with higher N deposition were found between 1978 and 1998 (RoTAP, 2010). Other UK studies that have re-surveyed sites with historical vegetation data have shown similar results, suggesting that a decline in biodiversity was already under way several decades ago. The exceptions to this are surveys in Scotland (RoTAP, 2010) which show a reduction in species richness of grasses and lichens between the 1960s/1970s and 2005 at sites with higher N deposition. This is consistent with the hypothesis that plant biodiversity is related to cumulative, rather than current, rates of deposition, and hence thresholds for significant loss of diversity are reached earlier in areas of high N deposition than in areas such as Scotland, with lower N deposition. A large amount of ecological information has been collected from ecosystems across Europe over many years for a variety of different purposes. Although a challenge to collate in a consistent format, these data can provide valuable information on changes in plant communities. In the Netherlands, for example, Tamis et€al. (2005) analysed trends for 83 ecological groups of species in 10 million vascular plant records over the course of the twentieth century. The most important trend in the data was a decline in those groups associated with nutrient-poor sites, although there was some reversal of this

474

trend after 1980. Jenssen (2008) assessed data from about 1500 vegetation relevés in forests in north-eastern Germany since 1960, before which N deposition was estimated to be less than 10 kg ha−1 yr−1 (i.e. below the critical load; see Section 20.5). Although overall species richness increased over this period, the frequency and cover of red-listed species, especially those adapted to low nutrient availability, declined. Duprè et€al. (2010) analysed data from about 1100 unfertilised acid grassland plots across Great Britain, the Netherlands, and Germany dating from the 1940s to the present. After adjusting for plot size and accounting for drivers such as climate, the species richness at each site was significantly negatively related to the estimated level of N deposition at that site since 1939 (Figure 20.8). Cumulative N deposition was more correlated to species number than cumulative S deposition for all regions. In Great Britain and Germany, cumulative N deposition was strongly related to a decline in the proportion of dicot species and an increase in the number of grass species. Bryophytes also significantly declined with increasing cumulative N in Great Britain (not measured in the other countries).

20.3.5╇ Identification of sensitive species and thresholds Figures 20.9–20.12 provide examples of species and genera from different habitats or plant functional groups that have been

Nancy B. Dise

Figure 20.9 (Top) Survival of Succisa pratensis in a hydroculture experiment grown at different NH4+ concentrations in application. From van den Berg et€al. (2005a). (Bottom) Succisa pratensis. Photo:€P. Vergeer.

found to be particularly sensitive to N deposition. Evidence comes from a variety of sources, with N addition experiments particularly useful in identifying the species that are most affected by nitrogen enrichment. In some cases these experiments allow the development of dose–response functions, and the identification of thresholds for adverse effects. In grasslands, several studies have shown a reduction in forb species richness or cover with increasing N deposition, leading to a decrease in the diversity, the attractiveness, and the range of ecosystem services offered by these communities. In Europe, one of the attractive forb species identified as adversely affected by N deposition is Devilsbit scabious (Succisa pratensis) (Figure 20.9). Several N-fertilisation experiments have shown a negative effect on this species in areas of high N deposition (Pauli et€al., 2002; Vergeer et€al., 2003), and laboratory experiments show a strong decline in this species with increasing NH+4 concentration in soil solution (Fig 20.9, van den Berg et€al., 2005a). Heathlands are dominated by ericaceous shrubs such as the common heather, Calluna vulgaris. Over the past 30–40 years, there is evidence from a number of countries (Netherlands, UK, Denmark, Norway, Germany) that heaths dominated by Calluna

Figure 20.10 (Top) Density of Drosera rotundifolia in four years (1988–91) in relation to four levels of nitrogen fertilisation (Redbo-Tortensson, 1994). (Bottom) Drosera rotundifolia. Photo:€A. Britton.

have been invaded by acid grassland species adapted to higher nutrient availability (Bobbink et€ al. 1998), with an associated reduction in the abundance of other characteristic heathland species (see Figure 20.2). Calluna was also identified as occurring at lower frequency in high-N sites in the UK acid grasslands survey of Stevens et€al. (2004). The mechanisms underlying this change in the dominant species are complex, but breakdown of the canopy by heather beetles (Lochmaea suturalis), which increase under higher N deposition, is an important factor in forming gaps where invasion by grasses can occur. In bogs, Drosera (sundew) are one of the largest families of carnivorous plants, which trap insects to provide an additional source of nitrogen and other nutrients to supplement the low nutrient status of their characteristic habitats. The number of individuals of Drosera rotundifolia, one of the three European sundew species, in a Swedish bog was reduced within one year when the N deposition rate was artificially increased from 10 to 20 kg N ha−1 yr−1 (Figure 20.10; Redbo-Tortensson, 1994). Racomitrium moss heaths are an important habitat of montane and arctic areas of Europe. In the UK, this habitat has been identified as being of high conservation value, but it has declined over recent decades, with heath species out-competed

475

Sphagnum NPP (g C m –2 yr –1)

Nitrogen as a threat to biodiversity C

N

Figure 20.12 (Top) Net primary production of Sphagnum magellanicum determined after 3.5 growing seasons of NH4NO3 treatments (40 kg N€ha−1y−1) to 10 cm cores is significantly reduced in 4 out of 6 field sites (abbreviated on the x-axis). Negative values for Re for both N and control treatments were caused by drought. Adapted from Limpens et al. (2004). (Bottom) Sphagnum magellanicum. Photo:€J. Limpens. Figure 20.11 (Top) Racomitrium lanuginosum. Photo:€A. Britton. (Bottom) (a) growth, and (b) percent cover of Racomitrium lanuginosum (grey bars) and graminoids (white bars) treated with either high (40 kg N ha−1yr−1) or low (10 kg N ha−1yr−1) nitrogen fertiliser over two summers, and measured at the end of the second growing season. From Pearce and van der Wal (2002).

by grass species (van der Wal et€al., 2003). While increased sheep grazing is one factor associated with this decline, several field experiments have demonstrated that relatively low levels of N deposition can cause a rapid decline in the growth and cover of Racomitrium (Figure 20.11). In these experiments there was no difference in effect between reduced and oxidised N, and only a slight difference between the lower and higher doses of N. Peatlands dominated by Sphagnum mosses are a major element of boreal and sub-arctic regions of Europe. They are an important global carbon sink because of their characteristic low decomposition rates, due both to low oxygen levels and to the

476

chemical characteristics of Sphagnum. A shift from Sphagnum to vascular plants reduces the size of this carbon sink. A reduction in the abundance of Sphagnum and other moss species can also lead to erosion, landscape degradation, deterioration in water quality, and reduced water retention. Experimental and field evidence in northwest Europe clearly show that relatively low rates of N deposition can increase N availability below the moss layer, increase growth of vascular plants, and change the water balance (Gunnarsson et€al., 2002; Bragazza and Limpens, 2004; Malmer et€al., 2003), all of which can impact Sphagnum. At higher rates of N deposition, sensitive Sphagnum species can be directly impacted (Figure 20.12). Taken together, the experiments, spatial gradient surveys and temporal re-surveys described in Sections 2.3.3 to 2.3.5 portray the long-term impact of atmospheric N deposition in

Nancy B. Dise

Europe over the second half of the twentieth century as reducing the presence of terrestrial plant species adapted to low�nutrient and poorly-buffered habitats. However, the strongest evidence is confined to grassland, forest, heathland, and peatland in northwest Europe, and we still lack a comprehensive understanding of the effects of N deposition on the biodiversity of habitats throughout the European continent.

20.3.6╇ Significance of the form of N deposition The evidence evaluated above considers only total N deposition. However, in any integrated assessment of the nitrogen problem in Europe, the chemical and physical form of N deposition is important, as these relate to different sources. From our understanding of processes as described earlier, and from empirical evidence (see Kleijn et€al., 2008 (Figure 20.5); De Graaf et€al., 1998, 2009), elevated concentrations of ammonium in the soil are more likely to have adverse effects on biodiversity than elevated soil concentrations of nitrate. However, there is little current evidence of any differences in effect between the deposition of reduced and oxidised nitrogen. While a small number of experimental studies have shown greater effects on species composition of wet-deposited reduced N over oxidised N (Twenhöven, 1992; van den Berg et€al., 2008), other long-term experimental studies show little evidence of differential effects and, overall, the experimental evidence is limited. Survey data are also not well suited to distinguish between the relative impacts of the same deposition rate of reduced versus oxidised nitrogen, as both natural gradients in deposition and spatial associations with other potential drivers differ between the two forms in the field. Ammonium deposition is also difficult to model on a regional scale because of its localised, diffuse sources and its relatively complicated transport, chemical reactivity, and deposition dynamics. Better understood is whether the gaseous and aerosol components of N deposition have different effects on vegetation compared to wet deposition. In drier regions of Europe, gaseous and aerosol inputs are likely to be a dominant form of N deposition, and in high concentrations these can cause direct foliar damage. A significant input of deposited N is in the form of nitric acid (HNO3) aerosol, but there is no evidence of the effects of this component on vegetation diversity. The direct effects of gaseous ammonia (NH3) are more well-known. The only major field experiment to directly compare the effects of gaseous NH3 with wet-deposited NH+4 at the same level showed much greater adverse effects from the gaseous form of reduced N on sensitive shrub, bryophyte and lichen species (Sheppard et€al., 2009). Although this comparison is for an ombrotrophic bog only, there is a considerable body of field evidence to demonstrate the local effects of NH3 in different regions of Europe, especially on lichens (Table 20.2). Much of this evidence relates to surveys around point sources of NH3, but wider-scale surveys in agricultural areas also provide evidence of effects on biodiversity (Rihm et€ al., 2009). Ammonia can affect epiphytic lichens both because it increases nitrogen availability and because, as a basic gas, it can increase tree bark pH, thus adversely affecting acidophytic species.

Table 20.2 Summary of field studies on effects of gaseous ammonia on plants

Location

Effect observed

Reference

UK

Decrease in nitrophobic epiphytic lichen species.

Sutton et al. (2009) Wolsely et al. (2009)

UK

Decreased cover of nitrophobic vascular plants.

Pitcairn et al. (1998, 2009)

Netherlands

Decrease in presence of acidophytic epiphytic lichen species.

Van Herk et al. (2003)

Italy

Increase in strictly nitrophytic lichen species

Frati et al. (2007)

Portugal

Increase in strictly nitrophytic lichen diversity and decrease in oligotrophic lichen diversity.

Pinho et al. (2009)

Switzerland

Increase in frequency of nitrophytic lichens relative to nitrophobic lichens.

Rihm et al. (2009)

In summary, different forms of N deposition are likely to have different effects on biodiversity because they affect the processes described in Section 20.2 in different ways, and because plants vary in both their sensitivity and their use of these forms. However, evidence of differential effects in the environment is limited, and the nature of any effect will be modified by local soil conditions and vegetation composition. Deposition of gaseous ammonia is more likely to cause loss of diversity than is the equivalent rate of wet reduced N deposition, especially in lichen- and bryophyte-dominated communities.

20.3.7╇ Evidence for impacts of N deposition on fauna Research on the effects of increased N inputs on faunal diversity in semi-natural and natural ecosystems is mostly lacking. Establishing effects on fauna can be difficult, as animals are usually mobile, and different species use the landscape at different spatial scales and over different times. There is, however, some evidence of impacts of N deposition on fauna through changes in food and environmental conditions (including micro-climate), and through the vegetation structure and landscape heterogeneity needed by animal species to complete their life cycles (Throop and Lerdau, 2004). Changes in both vegetation nutrient content and plant species composition can impact the fauna dependent on that vegetation. It is likely, for instance, that the frequency of caterpillars, and therefore butterflies and moths, has declined in areas of high N deposition due to both intrinsic vegetation changes and community composition changes (Weiss, 1999; Ockinger et€al., 2006). However, some butterfly or moth species may profit from N deposition if the preferred plant species of their larval forms becomes more dominant

477

Nitrogen as a threat to biodiversity

through N inputs. In a re-survey of 13 grassland sites in southern Sweden, for example, Ockinger et€al. (2006) found that local extinctions of butterflies between 1980 and 2000 were greater for those species whose larval host plants were adapted to low nutrient conditions; conversely, new colonisations were reported for butterfly species whose larval host plants were adapted to nutrient-rich conditions. Changes in plant nutrient content and species composition can filter down to impact detritivores feeding on the organic matter originating from this vegetation. Elevated N deposition can also cause the heterogeneity of the vegetation to decline due to factors such as extensive grass intrusion. The occurrence of animal species is related to landscape heterogeneity by at least three mechanisms. First, species may depend on specific conditions that are only present in transitions between habitats. Second, many animal species require different parts of the landscape for activities such as reproduction, resting, and foraging. Finally, heterogeneity creates the possibility of risk spreading, leading to a higher persistence of populations of animal species. Thus, N deposition affects faunal diversity not only directly (e.g. changes in the food quality and micro-climate), but also indirectly through changes in the configuration and heterogeneity of habitats. We illustrate this with two examples. The ground beetle (Carabidae) assemblages of dry open coastal grasslands are characterised by species preferring drought and relatively high temperatures. N deposition can lead to grass encroachment in the dune vegetation (see Bobbink et€ al., 2003). The invasion of extensive, relatively dense grasses changes the characteristic micro-climate of the open dunes (warm during daytime, but fairly cold at night and continuously dry) to a more buffered, continuously cool and moist micro-climate. This then cascades to the fauna. A comparison of the ground beetle assemblages between 15 coastal dune grasslands on the Waddensea islands Ameland and Terschelling showed that encroachment of the grasses Calamagrostis epigejos and Ammophila arenaria resulted in a change from the warmth- and drought-preferring Carabidae species dominating in intact dry dune grasslands to a beetle assemblage dominated by moisture-preferring species (Nijssen et€al., 2001). Thus, N deposition changes the vegetation composition, which in turn changes the composition of the fauna associated with that vegetation. The decline of the red-backed shrike (Lanius collurio) illustrates how the effects of increased atmospheric N deposition can cascade through the food web (Beusink et€al., 2003) (Figure 20.13). This bird species declined from 1950 onwards throughout Western Europe. Much of this loss has been attributed to direct habitat degradation, but in less developed areas, such as the coastal dunes of northern Germany, of southern Denmark, and of the Netherlands (where the shrike has disappeared), �direct habitat conversion cannot be the main reason. In these dune habitats, the pattern in population trends can be related to the rate of N deposition. Shrikes feed on large insects and small vertebrates such as lizards, and carry only a single prey to the nest at a time. To ensure a constant and sufficient energy supply for nestlings and over the breeding period, they

478

Figure 20.13 The red-backed shrike (Lanius collurio), a species that has been indirectly impacted by chronically elevated nitrogen deposition (photograph:€E. Dirksen).

require a high diversity of large prey species, and this requires a heterogeneous landscape. However, N deposition has made Dutch coastal dunes much more homogeneous, with widespread encroachment of tall grasses and bushes leading to a decrease of open sandy areas and a loss of species-rich succession stages. The N-driven decline in landscape heterogeneity greatly reduces the availability of large prey species, and this has been identified as the main factor in the decline of the shrike population in these regions (Esselink et€al., 2007). In other situations, the decline of a specific element of the landscape may be associated with changes in bird species. For example, a reduction in numbers of a rare bird species, the dotterel (Eudromias morinellus), in Scotland has been associated with a decline in Racomitrium heath, its preferred habitat, to which elevated N deposition has probably contributed.

20.3.8╇ Evidence of recovery from biodiversity loss Based on committed emission reductions, a decline in both the area and extent of nitrogen deposition in Europe from its 1980s peak is expected by 2010, with further reductions expected by 2020 (Section 20.5). Whether, and how quickly, this will lead to recovery from adverse effects on biodiversity is uncertain. The term ‘recovery’ can have different meanings, but here we define it as the return of the species composition of an ecosystem to its pre-N pollution state. Relevant evidence comes from experiments in which applications of N have ceased and from field observation in areas of Europe (such as the Netherlands and Denmark), where large reductions in emissions, especially from agriculture, have been achieved since the 1980s. However, the evidence for recovery of biodiversity in many areas of Europe is confounded by the simultaneous recovery from acidification due to the more rapid decline in sulfur deposition. Rates of recovery have been most intensively studied in grasslands, and many studies suggest that recovery is a slow process. For example, a study in the Netherlands showed that vascular plant species numbers in heavily fertilised

Nancy B. Dise

(150–250€ kgâ•›Nâ•›ha−1yr−1) grasslands changed slowly and were not stable 14€years after cessation of the N treatments (Olff and Bakker, 1991). Similarly, Clark and Tilman (2008) showed that the start of recovery of species numbers in plots treated with 10–95 kg N ha−1yr−1 for 23 years in a North American grassland could only be detected 13 years after the end of the experimental treatment. Hegg et€al. (1992) found that an effect of 40 kg N ha−1 yr−1 on species composition of an alpine pasture in Switzerland was still visible almost 40 years after the last application. However, faster rates of recovery have also been reported in grasslands. Results from an experiment on an unimproved grassland in England suggested that recovery of vegetation biodiversity (to a state comparable to that prevailing in the central plots) took from 3 to 5 years after cessation of the nitrogen treatment (5 years of 25 kg N ha−1yr−1) (Mountford et€al., 1996). In the Netherlands, the species composition of an actively managed calcareous grassland was similar to that of control plots within 10 years of the cessation of 8 years of treatment with 115–170 kg N ha−1yr−1 (Smits et€al., 2008). Fairly long timescales of recovery have been reported in non-grassland ecosystems. For example, both Power et€ al. (2006) (after application of 15.4 kg N ha−1yr−1 for 7 years) and Strengbom et€al. (2001) (after application of 34–108 kg N ha−1yr−1 for a period of 18 years) found no signs of recovery of vegetation after cessation of N application for 8 and 9 years in a heathland and boreal forest, respectively, although biogeochemical recovery was more rapid. From the experiments described above it is clear that the rates of recovery vary considerably. It is likely that the rates depend on a number of factors, including the intrinsic ecosystem sensitivity and buffering capacity, the ambient N input, management type and intensity, and the amount of accumulated N in the soil as a result of both long-term N deposition and the experimental application. All of these experiments only removed additional experimental N inputs, so the plots still received ambient N deposiÂ� tion. Few studies have examined the effect of a reduction from current rates of N deposition to pristine deposition rates. The best example of this at an ecosystem level is the NITREX network, in which transparent roofs were built over European conifer forests (Wright and van Breemen, 1995). In more polluted forests, ambient precipitation was cleaned of pollutants via ion exchange and returned, as ‘pre-industrial’ deposition, to the forest. Forests in less impacted regions were treated with elevated levels of nitrogen deposition. In a Dutch experiment within this network, Boxman et€al. (1998) showed, after 6 years, improved growth of pine trees, an increased number of sporocarps of mycorrhizal fungi, and a decline in the number of nitrophilic species (notably the fern Dryopteris dilatata) in plots from which wet N deposition was replaced by artificial clean rainwater. A different approach was taken by Jones et€al. (2005), who removed vegetated cores from an acid grassland community in Wales to an artificial enclosure, where they irrigated the cores with deionised water (removing the input of N in deposition). In response, the cover of the sensitive moss Racomitrium increased, but there was little change in vascular plant cover.

In areas that have been highly impacted by NH4+ deposition, evidence shows that recovery of the original vascular plant species diversity may only be possible with active management intervention. In the Netherlands, for example, where large reductions in NH4+ deposition have been achieved in some areas, soil conditions still prevent the establishment of sensitive heathland and acid grassland species, and management by nutrient removal and lime application is first needed to restore the necessary biogeochemical conditions (Kleijn et€al., 2008). In summary, species that are impacted by direct deposition of N, such as lichens, fungi, and bryophytes, may rapidly recover once N deposition has been reduced. In contrast, recovery of vascular plant diversity may take several decades, and may require significant biogeochemical recovery to precede it. In cases where cumulative N loads are high or damage is severe, active management intervention may be needed to restore the full range of species that were originally present. The reduction in diversity of vascular plant species owing to N deposition is probably a cumulative progression occurring over several decades; likewise, recovery of biodiversity is likely to be a slow process.

20.4╇ Models This section provides an overview of modelling approaches to describe nitrogen deposition impacts on the biodiversity of terrestrial ecosystems and to predict future change. These models build mathematical representations of complex phenomena either by developing equations that simulate the main underlying processes, developing empirical relationships, or combining these approaches. The models may be used for a variety of purposes, including hypothesis testing, risk mapping, policy/management recommendations, scenario testing, and future predictions. We first describe the major approaches to modelling, and introduce the leading models used in Europe today to evaluate the impact of nitrogen deposition on natural ecosystems. We then link two of these models together to demonstrate how models can be used to explore questions such as the relative importance of nitrogen deposition and climate change on biodiversity into the future. We conclude with a discussion of the limitations of current models and some of the improvements that are needed or are currently being developed.

20.4.1╇ Modelling approaches The simplest approach to model the relationship between nitrogen deposition or concentration and plant community composition is to use empirical relationships (e.g. see Figures 20.6–20.8) relating the two factors (Figure 20.14; left strand of diagram). With empirical models, although we might have good hypotheses about the underlying mechanisms for the relationships derived, no knowledge or assumptions about these mechanisms is required. Empirical models simply show the best relationships among the measured parameters. Such mathematical relationships often readily lend themselves to geographically large-scale extrapolations because they require few parameters for upscaling, and the models can be developed

479

Nitrogen as a threat to biodiversity Figure 20.14 Modelling approaches to predict plant species diversity response to external drivers such as nitrogen deposition. Brief descriptions of the models shown here or described in the text are in Appendix 20.1.

to use data available from national or European datasets. They need to be used with caution for predicting future trends, however, since we cannot usually be certain that the relationships developed are cause-and-effect ones. An alternative way of modelling relationships between N deposition and plant community composition is to use a deterministic, or process-based approach (Figure 20.14, right-hand strand). These simulate changes in biotic or ecological characteristics in response to environmental drivers using mathematical representations of the most important processes as we understand them. The strength of this approach is that it is based upon our best knowledge of the actual drivers and so allows us to test hypotheses and make more detailed predictions. Deterministic models are also useful for assessing time trends and response times, and to explore future scenarios. However, they often require a large amount of data for development, and a number of parameters to be set for their application. In addition, the equations used are only as good as our knowledge of the process being simulated, which is often incomplete or even poor. The most commonly used deterministic models for biodiversity are two-stage. First the biogeochemistry of the ecosystem is simulated as a function of drivers such as nitrogen deposition using a deterministic model. The predicted ecosystem biogeochemistry is then used as input to into either an empirical model or another deterministic model to simulate the composition of the plant community that corresponds to that biogeochemistry (De Vries et€al., 2007, 2010). The major deterministic models used in the first, biogeochemical stage (e.g. SMART2, MAGIC, ForSAFE) were all

480

initially developed to predict a forested ecosystem’s response to acid deposition. The models differ in aspects such as the relative importance of different processes, the detail in which processes are represented, and the scale at which the models function (see Appendix 20.1). Each model conceptualises differently the way in which an ecosystem responds to long-term inputs of nitrogen, has different input requirements, and provides as an output different representations of soil N status (e.g. soil C/N, soil available N) and soil acid status (e.g. pH, base saturation) at different scales. The empirical vegetation models are based on a large number of field surveys. From these, species-response curves are derived for many higher and lower plant species. Since they are based on many sites, the empirical vegetation models provide an excellent picture of the current composition of vegetation communities, and relationships between composition and drivers such as soil chemistry, climate and management. Instead of developing relationships between biogeochemistry and community composition using tens of thousands of observations, deterministic vegetation models such as Veg and BERN simulate changes in community composition using dynamic simulations of processes (Appendix 20.1).

20.4.2╇ Using models Models are properly validated with independent data sets. This allows the accuracy and uncertainty of the model to be evaluated, and can identify how sensitive different parts of the model are to variations in the input data. There is a fairly

Nancy B. Dise

Figure 20.15 Predicted change over time in vegetation cover of 49 forest stands in Sweden, comparing the scenarios of no controls enacted on N deposition in the 1980s with maximum feasible future reductions in N deposition, both under the IPCC ‘middle’ climate scenario A2. The median (50%) estimate is shown, with the 10th and 90th percentiles shaded below and above the median, respectively.

good record in the literature of validation of both types of deterministic models (biogeochemistry and vegetation) and of empirical biogeochemistry models (see De Vries et€ al., 2010; Belyazid et€ al., 2006; van der Salm et€ al., 2007; Dise et€al., 2009). Validation of empirical vegetation models are at an earlier stage, probably because the models themselves are fairly new, although there is evidence from acid grasslands that equations developed on independent sets of data are very similar (Stevens et€al., 2009). Once a model has been satisfactorily validated, it can be used to explore the implications of different scenarios. As an example, we use the combined ForSAFE-Veg model chain (both deterministic, Figure 20.14, right side) to investigate the combined impact of climate change and nitrogen deposition on the biogeochemistry and ground vegetation composition of 49 forest sites. The forests are 17 boreal and cool temperate coniferous stands in Sweden (dominated by spruce and pine) and 32 Swiss forests, both from the northern Swiss plains (mostly deciduous and mixed) and from the Alps (mostly coniferous); the years from 1950 to 2100 are modelled. ForSAFE (Forest SAFE; Soil Acidification in Forest Ecosystems; Wallman et€al., 2005) merges a geochemical model for soil solution chemistry, weathering and cation exchange (SAFE; Alveteg, 1998) with a forest simulator for plant growth, litterfall, and organic matter decomposition. The model requires information on environmental drivers such as pollutant deposition, climate, and management, and dynamically simulates changes in the ecosystem based on changes in these drivers. The output of ForSAFE (ecosystem parameters such as soil chemistry and moisture, shading, and temperature) is then linked with Veg (Vegetation model:€Belyazid et€al., 2006), which simulates changes in the composition of the ground vegetation in response to changes in these biotic and abiotic factors, using plant species-specific information on habitat preferences. We first use this model chain to evaluate the effectiveness of pollution controls, with the model run using either (1)€the scenario of maximum feasible reduction of N deposition, or

(2) N deposition if there had been no controls on emission. The simulation assumes that most of the “maximum feasible reduction” has been accomplished through existing pollution legislation, so it is essentially an evalution of the effectiveness of current N emissions reduction policies (Section 20.5). ForSAFE-Veg suggests that, had the European legislation of the late twentieth century to reduce the peak of N deposition (UNECE, 2010) not been enacted, 20% of the ground vegetation of these forests (by cover) would have shifted to a new type by 2100 (Figure 20.15). The above simulation assumes, however, that the climate is changing over time. Comparing scenarios with and without climate change, the model suggests that global warming alone will cause a 40% change in the cover of ground vegetation in these Swedish and Swiss forests from 1950 to 2100, even with nitrogen control policies in place. This change is primarily due to the direct effects of a warmer climate, and secondarily to soil moisture changes and increased nitrogen status from mobilising soil N. The accumulated soil nitrogen, in turn, is in part the legacy of enhanced N deposition since the early twentieth century. One could conclude from this exercise that, although the composition of the ground vegetation of European forests is likely to significantly change due to a warming climate (partly from mobilising accumulated soil N), this change would be even greater in the absence of nitrogen pollution control policies enacted since the 1980s. However, these results should be simply taken as tools to explore potential implications of nitrogen deposition and climate change on vegetation diversity, rather than predictions of the actual future. We have explored only one model chain:€ different models would give different results and, since our knowledge of both current processes and the future environment is limited, no model is currently ‘the correct’ one. The above exercise describes the application of a deterministic model which can be used dynamically to evaluate the impact of various scenarios, but for a limited number of sites. Empirical models are generally simpler and require fewer data, and if these data are available on a large scale they can be useful for scaling-up assessments for regional, national, continental or even global applications. In Section 20.5 we will explore how one such empirical model can be used to develop maps of biodiversity change in response to N deposition on a European scale.

20.4.3╇ Future improvements in models The range of models available for simulating the effects of N deposition on terrestrial biogeochemistry and ecology provides a useful toolbox for a variety of biodiversity applications. Empirical and deterministic methods can be used complementarily and also as counter-checks. However, while the potential impact of different scenarios can be tested, and maps of biodiversity risk developed, more work is needed. In particular, more refinement and testing on validation datasets of existing models are needed to improve predictions. In addition, models need to simulate delay times in the response of plant

481

Nitrogen as a threat to biodiversity

communities to changes in N deposition. As discussed previously, experimental evidence suggests that there are likely to be considerable delays in the response of the ground vegetation community to changes in N emissions and deposition due to their response to accumulated, rather than current, N deposition, as well as ecological effects such as recruitment. These lags need to be understood and incorporated into models to assess the effectiveness of air pollution legislation, in particular how soon we are likely to see the effects of N emission reductions. Another limitation with the models concerns the distinction between reduced and oxidised nitrogen, and dry and wet N deposition. Whereas, as described previously, experiments have shown that reduced and oxidised N in soil have different effects on plants, and dry-deposited gaseous NH3 is particularly harmful as a direct pollutant, models do not yet incorporate these processes at the necessary level of detail. Furthermore, as shown in the example above, any reasonable estimate of the future changes in plant communities needs to incorporate climate change effects. Climate drivers are increasingly included in dynamic modelling simulations, but are not yet incorporated in empirical models to any extent. This limits the degree to which empirical estimates of critical N deposition loads on plant communities can be extrapolated into the future. However, even if climate change is incorporated in models, our understanding of the impact of global warming on particular regions and ecosystems is still highly uncertain. So, although the models serve as useful guides, they will not tell us for certain what the future will bring.

20.5╇ Policy and critical loads In this final section we describe the development of current European policies on biodiversity conservation and air pollution (including nitrogen) abatement, the relative success of these policies, and how the two policy aims of biodiversity protection and pollution abatement may be connected through the concept of critical loads.

20.5.1╇ Biodiversity legislation Serious attention to halting and reversing biodiversity loss began in 1992 with the UN Convention on Biological Diversity (CBD, 1992) in which the international community committed itself to addressing biodiversity protection and enhancement via a legally binding global treaty. The Convention has three objectives:€the conservation of biodiversity, the sustainable use of its components, and the fair and equitable sharing of benefits arising out of the utilisation of genetic resources. The CBD produced a strategic plan set out by the 6th Conference of the Parties of the Convention (2002, COP). The COP’s mission was ‘to achieve, by 2010, a significant reduction of the current rate of biodiversity loss at the global, regional and national level, as a contribution to poverty alleviation and to the benefit of all life on Earth’ (Balmford et€al., 2005). Europe responded to the CBD in 1995 through the endorsement of the Pan-European Biological and Landscape Diversity Strategy by the more than 50 countries covered by the United Nations Economic Commission for Europe (UNECE). This

482

strategy provided the only platform for cross-European cooperation on halting biodiversity loss. In the European Union, the EC Biodiversity Conservation Strategy (ECBS) was adopted in 1998 in response to the requirements of the CBD. Four biodiversity action plans€– natural resources, agriculture, fisheries, and development€– were adopted in 2001 and laid out actions to implement the strategy. A review of the implementation of ECBS was initiated in 2004, and led to the EC Communication on Halting the Loss of Biodiversity by 2010 (CEC, 2006). Political agreement on the 2010 target has been accompanied by a growing consensus on the need for long-term, structured, global and European coÂ�ordination of biodiversity monitoring, indicators, and reporting efforts on a sound funding basis. Outside of these formal conventions and communications, protection and restoration of biodiversity has been increasingly prominent in EU strategy and research framework programmes. The objective of ‘Managing natural resources more responsibly:€to protect and restore habitats and natural systems and halt the loss of biodiversity by 2010’ was first adopted by the EU in its Strategy for Sustainable Development in 2001 (CEC, 2001). The conservation of biodiversity is also one of the four main issues to be addressed along with ‘climate change’, ‘environment and health issues’, and ‘preserving natural resources and waste management’ within the EU Sixth Environmental Action Programme ‘Our Future, Our Choice’ (CEC, 2002). At the national level, several countries have included the EU 2010 target as part of their national biodiversity strategies. However there has been only limited ‘trickling down’ of these policies to the local scale, where managers or conservation authorities of nature reserves, recreation areas, etc., may have (explicit or implicit) biodiversity action plans that are specific to a particular location. There are some strategies designed to integrate local plans into the European scale (e.g. Fauna, Flora and Habitats (FFH) Directive; Natura2000 network). However, there is a long way to go before achievement of an integrated biodiversity strategy encompassing the local, national and European levels. The progress toward achieving the 2010 European biodiversity target was assessed by the European Environment Agency (EEA, 2009). The overall conclusion was that, whereas progress has been made in some areas, the status of most species and habitats still gives rise to concern.

20.5.2╇ Nitrogen as a recognised threat to biodiversity In 2004 the COP identified nitrogen deposition as one of 17 biodiversity ‘headline’ indicators for assessment of progress in achieving the 2010 goal (COP, 2004). The same framework of 17 headline indicators was adopted at the European level in 2005 (PEBLDS, 2005). The ‘Streamlining European 2010 Biodiversity Indicators’ (SEBI2010) process was set up to oversee implementation of the framework at both the EU and pan-European level. Thus, the COP has specifically recognised N deposition as both a threat to biodiversity and a useful indicator of that threat, being relatively straightforward to estimate via well-Â�established deposition models at both the national and Â�cross-European

Nancy B. Dise

scales (e.g. the European Monitoring and Evaluation Programme€ – EMEP). Other SEBI2010 indicators directly linked to nitrogen include nutrients in transitional, coastal and marine waters, and the Agricultural Nitrogen Balance (input vs. output of nitrogen in the agricultural system). The development and use of the nitrogen deposition headline indicator is overseen by the International Nitrogen Initiative (INI), and is based on critical load exceedance for nitrogen (see below). Monitoring is done via data generated within the UNECE Convention on Long-range Transboundary Air Pollution. Such recent advances support current policy and will no doubt contribute toward future policy development. Although it is increasingly recognised at the European scale, air pollution is still often not explicitly taken into account in biodiversity action plans, especially at the local or regional levels. As a consequence, nitrogen deposition may not be assessed as a threat in a way that is consistent with its known impacts on biodiversity. For instance, N deposition effects were not taken into account when selecting Natura2000 areas (Slootweg et€al., 2007). The potential impacts of N deposition on Natura2000 sites, and implications for compliance with the Habitats Directive, was the subject of a workshop held in May 2009 (Sutton et al., 2010).

20.5.3╇ Air pollution legislation and critical loads As described in Oenema et€al., 2011 (Chapter 4 this volume), air pollution legislation in Europe is implemented within the framework of the Convention on Long-range Transboundary Air Pollution (CLRTAP, EU-NEC Directive 2001) and its eight subsequent protocols. The Gothenburg Protocol (1999) sets emission ceilings for 2010 for four pollutants:€SOx, NOx, VOCs and NH3. As a result of the protocol, it is expected that Europe’s sulfur emissions will be reduced by at least 63%, its NOx emissions by 41%, its VOC emissions by 40% and its ammonia emissions by 17% in 2010 as compared to 1990 (CLRTAP, 2010). European emissions controls for sulfur and nitrogen are based on the critical loads concept, an effects-based approach (Spranger et€al., 2008). A critical load is defined as: A quantitative estimate of an exposure to one or more pollutants below which significant harmful effects on specified sensitive elements of the environment do not occur according to present knowledge (Nilsson and Grennfelt, 1988).

In theory, if pollutant deposition rates are lower than critical loads, ecosystems maintain their long-term elastic stability against stress. Among other things, this means that the development potential, sustainable use, and persistence of ecological functions of ecosystems are protected. The concept of critical loads was developed in response to acid deposition, and the ‘harmful effects’ are usually biogeochemical (e.g. soil acidification) rather than ecological. The acidity critical load encompasses both sulfur and nitrogen deposition. Critical loads have also been defined for eutrophication of terrestrial ecosystems, as well as for heavy metal pollution of terrestrial and surface water ecosystems. One limitation of the critical load approach is that, because they relate to a long-term sustainable state of an ecosystem, critical loads do not change over time and thus do

not allow a prognosis of on-going ecosystem status at any point in time. Dynamic models must be used to predict such changing states over time (see Section 20.4). Several decades of critical loads-based pollution control policies have resulted in reductions in the emissions of all targeted air pollutants, most dramatically sulfur, and clear improvement in areas such as soil and water acidification. The difference between the critical load for a pollutant and the actual deposition of the pollutant on an area is the critical load exceedance. There is still a substantial area of semi-natural ecosystems in Europe where critical loads of nutrient nitrogen deposition are exceeded, including over 50% of the forests of Europe (Hettelingh et€al., 2008b). However, the level of exceedance has declined considerably over all habitats over the past 30 years (Figure 20.16). Thus, because of the success of pollution control legislation, we are progressively narrowing the gap between the nitrogen deposition and the critical load of nitrogen that is considered to pose a minimal threat to the health and integrity of the ecosystem. Increasingly, the focus of European policymakers has shifted to human health and biodiversity effects of air pollution (e.g. in the EU Thematic Strategy on Air Pollution of 2005, revised CLRTAP Protocols, and the EU National Emissions Ceilings Directive). Critical loads for eutrophication and acidification are now being evaluated for how well they can be linked to impacts on biodiversity, and how this may be translated into policy. The underlying assumption is that the critical N deposition load for minimising biogeochemical damage is similar to that at which biodiversity impacts are minimised. This explicit linking of critical loads to biodiversity is receiving increasing attention (e.g. the Workshop on Nitrogen Deposition, Critical Loads and Biodiversity, Edinburgh, November 2009). Critical loads for N are also being used for the protection of biodiversity at local or regional levels in several European countries, as well as in the US (e.g. in US National Park policy; Porter and Johnson, 2007).

20.5.4╇ Applying the critical loads concept to biodiversity protection Evaluation of the appropriateness of nitrogen critical loads for biodiversity is at a fairly early stage, and has focused on the use of€established ‘empirical’ critical loads for eutrophication. These are based on evidence from field experiments and targeted surveys to identify threshold rates of N deposition for effects on ecosystem structure or function (Bobbink et€al., 2003). Table 20.3 summarises the evidence that the biodiversity of sensitive habitats in Europe is affected when the nutrient nitrogen critical load is exceeded. In many cases effects are similar to the overall impacts described in Table 20.1, but here the focus is on changes observed in the field in relation to the critical load. For those habitats described as sensitive to atmospheric N in Table 20.1 but not listed in Table 20.3 (tundra, coastal dune, Mediterranean), there is either no critical load or no field evidence to evaluate whether critical load exceedance is related to observed effects on biodiversity. The critical load ranges reflect the range of sensitivity of different habitats within the broad ecosystem types. Evidence

483

Nitrogen as a threat to biodiversity

Figure 20.16 Exceedance of critical loads for eutrophication by deposition in 1980 and (projected) 2010 under current legislation. Map made using the official data of the National Focal Centres (NFCs) on critical N loads for each country (map of 2010 from Hettelingh et al., 2008b; map of 1980:€M. Posch, personal communication).

in Table 20.3 is primarily based on ecological surveillance datasets; in addition, a synthesis of data from experimental studies (Bobbink, 2008, described below) is included. In most of the non-experimental studies, specific relationships with critical load exceedance were not evaluated€– rather, the evaluation identifies ecological changes at sites or in areas where the critical load is exceeded. For the ecosystems listed, Table 20.3 provides broad support for the use of empirical critical loads for protection of biodiversity in addition to ecosystem structure and function. Further evidence for the appropriateness of critical loads based on biogeochemical change as a useful index for terrestrial biodiversity comes from recent ‘inverse modelling’ work, based on van Dobben et€ al. (2006) and updated by de Vries et€al. (2010). First, the MOVE model (see Figure 20.14) was run inversely to produce the critical biogeochemical conditions (e.g. lowest pH range, highest N availability range) associated with a large number of vegetation assemblages, then the SMART2 model was run inversely to determine the nitrogen deposition that would produce those conditions. That value of N deposition was then considered to be a type of ‘ecological’ critical N load for that vegetation type. These critical loads overlap with established empirical critical loads for most vegetation classes. However there are several reasons for caution in assessing biodiversity risk based on the exceedance of critical loads alone. First, as described previously, ecological surveillance datasets are not usually designed to identify N deposition impacts, and so attribution of any impact to nitrogen deposition can be difficult. Second, the relatively short-term N-addition experiments used to define empirical critical loads may not accurately reflect the impact of many decades of N deposition on the biodiversity of vulnerable ecosystems. Third, because of factors such as

484

accumulated nitrogen in the ecosystem, climate change, and potential recruitment limitations, we do not know with confidence when, or to what degree, biodiversity will recover if pollution is reduced to the level of the critical load or below. Finally, different habitats within the broad ecosystem classifications shown in Table 20.3 are differentially vulnerable to N deposition. For instance, acid grasslands are more vulnerable than calcareous grasslands, and this is reflected in a lower critical load for these ecosystems. Scaling up to regional extrapolations using critical loads based on broad ecosystem types could therefore underestimate the impact of N deposition on more vulnerable habitats. Table 20.3 also only describes impacts from N deposition loads. As discussed previously, deposition of gaseous ammonia is more likely to reduce diversity than is the equivalent rate of wet N deposition, especially in non-vascular plants. Critical levels (concentrations, rather than loads) for NH3 deposition have recently been set at 1â•›µg m−3 for lichen and bryophyte-Â�dominated communities, and at 3â•›µgâ•›m−3 for vascular plant-dominated communities (Cape et€ al., 2009). Evaluation of exceedance of these critical levels, as well as critical loads for total N deposition, should be a component of any future European assessment of the impact of N deposition on biodiversity.

20.5.5╇ Developing European-scale assessments of nitrogen deposition impacts on€biodiversity Using data from nitrogen addition experiments in the field, Bobbink (2008) derived empirical relationships between plant species richness or similarity indices and exceedance of critical

Nancy B. Dise Table 20.3 Evidence of loss in biodiversity in different habitats across Europe in relation to the critical load (CL) or critical load exceedence (CLE) for nutrient nitrogen. Critical loads from Bobbink et€al. (2003)

Critical load range (kg N ha–1 y–1)

Relationship with critical load

Source of evidence

Arctic, alpine and subalpine scrub

5–15

Species richness in experiments declines with CLE.

Bobbink (2008)

Heathlands

10–25

CLE within the Netherlands correlated with reduced species numbers in dry but not wet heaths.

Van Hinsberg et al. (2008)

Peatlands

5–10

-Reduced cover of Sphagnum species and increased cover of vascular plants above CL. -Loss of characteristic bog species on Danish ombrotrophic mires above the CL

Wiedermann et al. (2009) Aaby (1994)

Grasslands

10–30

-Decline in species richness at sites with higher N deposition above the CL. -Reduction in species richness of vascular plants, and declines in forbs relative to grasses above the CL -Species richness in experiments decline with CLE.

Maskell et al. (2010) Duprè et al. (2010) Bobbink (2008)

Forests

10–20

-Frequency of Vaccinium in Sweden decreases at N deposition above the CL. -CLE within the Netherlands correlated with reduced species numbers in forests on sandy soils. -Similarity to species composition of ‘unpolluted’ sites in boreal forest experiments declines with CLE. -Decline in frequency of red-listed species in German forests as N deposition increased above CL.

Strengbom et al. (2003) Van Hinsberg et al. (2008) Bobbink (2008) Jenssen (2008)

Habitats

Note:€critical loads are being reviewed in 2010 and some values listed in Table 20.3 may be revised accordingly.

loads for three different habitats (Figure 20.17, Table 20.3). Exceedence of critical loads was determined by subtracting the maximum of the range of the critical load for that specific ecosystem from the N addition load used in the experiment. Robust datasets were only available for grassland, arctic/alpine shrub (‘scrub’) and coniferous boreal forest. For grasslands and arctic/alpine shrub habitats, species richness declined as exceedance of the empirical critical load increased (Figure 20.17). There was no clear relationship between species richness and critical load exceedance for boreal forest ecosystems, although there was evidence of a shift in understorey species composition as exceedance increased, reflecting the replacement of species adapted to low nutrient availability by more nutrient-demanding species. This biodiversity change in forest ecosystems was described by the Sørensen similarity index (SI) instead of species richness. (The SI compares the similarity of two samples, and is defined as SI=2C/(A+B), where A and B are the species numbers in two samples, and C is the number of species in common among those samples (Sørensen, 1948)). A dose-response relationship between the SI and critical load exceedance was then developed for forests in a similar way as for grassland and shrubland. Using estimated N deposition from the EMEP model, together with the European distribution of the three broad vegetation types represented by these ecosystems (forest, shrubland and grassland), the functions can then be used to derive maps of estimated vegetation change as a function of N deposition.

Figure 20.18 shows the output of such an exercise (Hettelingh et€ al., 2008a). The modelled species richness or similarity in relation to non-affected ecosystems is compared to the estimated level in 1900 (Schöpp et al., 2003), before a significant increase in regional N deposition (Figure 20.18). Figure 20.18 suggests that the largest N-driven ecological change (in this case, community composition) of the three broad ecosystems between 1900 and 1990 has occurred in forests. This may reflect the high filtering function of forests, resulting in significantly higher levels of N deposition reaching the forest floor than in ecosystems characterised by low-stature vegetation. Shrubland is described in this up-scaling to have been impacted the least by N deposition, while grassland is intermediate. Note that a reduction in species richness of up to 10% is already estimated for shrub vegetation in 1900 – this is due to the low critical load for this vegetation type (Table 20.3), which may already have been exceeded in many areas by 1900. This exercise is primarily illustrative, and includes many simplifications. Most notably, the responses of arctic/alpine shrubland, and of coniferous boreal forests, are extrapolated across all shrubland and forests in Europe. As our knowledge and evidence base improve, so will the accuracy of the values we attribute to biodiversity reduction and risk for different ecosystems. Maps similar to the preliminary assessment shown in Figure 20.18 may ultimately provide an overview of the potential severity, extent and distribution of biodiversity change across Europe,

485

Nitrogen as a threat to biodiversity

biodiversity as a function of the critical load exceedence for a habitat, and can form the basis for extrapolation over an appropriate range of that habitat. However, because at least part of the relationship we observe today is the result of longterm accumulated nitrogen, such models should not be used to predict an instantaneous change in biodiversity as critical load exceedence changes. More research is required on rates of recovery for different ecosystems, and as this knowledge builds, dynamic and empirical approaches will no doubt play a joint role in informing policy development and assessment.

Proportional number of species: Treatment/Control

1.2

(a) 1.0

0.8 0.6 0.4

Semi-natural grassland

0.2 0

20

40

60

80

100

120

140

Critical load exceedance (kg N ha–1yr –1)

Proportional number of species: Treatment/Control

1.1

(b)

1.0 0.9 0.8 0.7 0.6

0.5

Arctic / alpine shrubland

0.4 –10

0

10

20

30

40

50

60

70

Critical load exceedance (kg N ha–1yr –1)

Figure 20.17 Relationships between relative species number and exceedance of the maximum of the critical load range in N-addition experiments for the specific habitat of relevance in (a) grassland and (b) arctic/ alpine shrub habitats. From Bobbink (2008).

which could help guide more focussed investigations of specific regions and habitats. However, being based on extrapolation of empirical models, they are not well suited for describing rates of biodiversity recovery (or loss) should N deposition decline (or increase). Since it is likely that ecosystems respond at least in part to the cumulative N deposition that has been stored in soil and vegetation over decades, it may take many years for an ecosystem to lose enough of this accumulated N (via leachate, atmospheric emission, fire, etc.) for it to recover to its preÂ�impacted state, even after N deposition itself has been reduced to below the critical load. The reduction in biodiversity from N deposition has probably occurred over many decades; the time course of recovery is also likely to be long. Even if the most stringent air pollution control policies are enacted, some ecosystems have likely been so damaged by chronic nitrogen loading that pollution reduction would not lead to full recovery within a reasonable time period. In these cases, active restoration, such as grazing, burning, mowing or cutting, could be considered as a management tool to accelerate the natural processes of nitrogen removal. The ‘Survival Plan Forest and Nature’ in the Netherlands (Overlevingsplan Bos en Natuurâ•›–â•›OBN) provides a good example of how such restoration programmes may be developed, reviewed and acted upon in the field. In practice, then, empirical models such as shown in Figure€ 20.17 may be useful to describe the current status of

486

20.6╇ Conclusions and recommendations Biodiversity is the variability among living organisms, and includes variability at the level of the individual, species, community and ecosystem. The value of biodiversity is multifold, from preserving the integrity of the biosphere as a whole, to providing human services such as food and medicines, to spiritual and aesthetic well-being. Biodiversity in Europe is threatened by many different forces, most of them driven by human population and our growing needs for more and better-quality food, housing, industry, and transport. One of the major drivers of biodiversity loss is atmospheric deposition of Nr, a product of both agricultural and industrial (primarily transport) activities. This chapter has focussed on nitrogen impacts on European plant species diversity, but its conclusions are broadly applicable to the wider biota, including fauna and below-ground organisms, and to areas outside Europe. We summarised both the scientific and the policy aspects of nitrogen impacts on diversity, including the processes and evidence, the most vulnerable ecosystems and regions, modelling approaches, and current legislation. Species and communities most sensitive to chronically elevated nitrogen deposition are those that are adapted to low nutrient levels, or occur in habitats that are poorly buffered against acidification (Section 20.1). Sensitive habitats occur in grassland, heathland, wetlands, and forests, among other ecosystems. A sensitive ecosystem in an area of high nitrogen deposition is vulnerable to biodiversity loss, and there are many such areas across Europe. As a biodiversity ‘hotspot’ containing many sensitive habitats, the Mediterranean basin is potentially highly vulnerable to nitrogen deposition. Although levels of N deposition are still relatively low in most parts of this region, they are increasing. Nitrogen impacts vegetation diversity through direct foliar damage, eutrophication, acidification, and susceptibility to stress (Section 20.2). Reduced nitrogen at high concentrations in the soil solution, or dry-deposited directly to leaf surfaces, can be particularly harmful to biota. Although knowledge of impacts on fauna is low, there are some clear examples of reductions in faunal diversity that can be linked to nitrogen deposition. Rather than direct impacts of nitrogen, damage to faunal diversity is usually a secondary result of changes in vegetation diversity, heterogeneity, composition, or structure. It is also likely that nitrogen deposition acts synergistically with other stressors, in particular climate change, acid deposition, and ground-level ozone, although these synergies have been poorly studied.

1900

1990

Grass

Shrubs

Forests

Figure 20.18 Modelled percentage of species richness in grassland (top), shrubland (middle) and similarity index in forests (bottom) for two different time scenarios:€the pre-N deposition status in 1900 (left), and 1990 levels (right) of N deposition (Hettelingh et€al., 2008a,b). Values are expressed as percentages of species number or similarity in non-N impacted ecosystems:€red = less than 80% of the non-impacted ecosystem, orange = 80%–90%, yellow = 90%–95%, light green = 95%–99%, dark green = 99%, blue = 100%.

487

Nitrogen as a threat to biodiversity

In Section 20.3 we reviewed both the evidence that biodiversity of a variety of sensitive ecosystems can be reduced due to Nr, and the evidence that biodiversity is actually reduced in Europe in response to chronic N deposition. We compared the results of long-term field manipulation experiments of nitrogen on a variety of ecosystems, regional surveys along deposition gradients, and re-surveys of surveillance sites and other studies over time. We conclude from these studies that it is highly likely that terrestrial biodiversity has been in decline in Europe for many decades due to nitrogen deposition, probably since the large-scale expansion and intensification of agriculture and transport in the second half of the twentieth century. There are several approaches to modelling changes in biodiversity with increasing loads of nitrogen, each with its advantages and disadvantages, and each best suited for a particular application (Section 20.4). Deterministic models, built from our understanding of the processes driving phenomena, allow us to test hypotheses and to predict how specific ecosystems may respond to different scenarios of N deposition (including pollution reduction), but can be very data-intensive. Empirical models€ – often based on observational data of wide extent but low resolution€– are well-suited for upscaling and extrapolation to assess the vulnerability of different habitats and regions across Europe to nitrogen-driven loss of biodiversity. However, they primarily describe observed relationships, and these may not be cause-and-effect. Efforts are focussed on combining the two approaches to utilise the best features of each to understand and predict nitrogen-driven loss of biodiversity. Policies and legislation aimed at enhancing and restoring biodiversity in Europe are framed around the 1992 Convention on Biological Diversity (CBD) and the 2002 CBD Strategic Plan to reduce and reverse loss of biodiversity by 2010 (Section 20.5). Nitrogen deposition has been adopted as a central indicator of biodiversity loss, such that restoration of diversity is explicitly linked to reductions in nitrogen input. However, N deposition is still not integrated across the biodiversity policy and management arenas in Europe. Likewise, biodiversity impacts are not explicit in European pollution control legislation such as the Convention on Long-Range Transboundary Air Pollution. The concept of critical loads for pollution deposition, first developed as a joint science–policy response to acid deposition, is a useful approach for identifying and mitigating biodiversity loss due to nitrogen pollution. Nitrogen removal experiments suggest, however, that recovery may be slow, and in some cases may require active management intervention. We suggest that future research should focus on quantifying:€(1) the extent of terrestrial biodiversity reduction due to N deposition in Europe (expanding research to all potentially vulnerable ecosystems), (2) the current extent and future threats outside of Europe, particularly in Asia, (3) synergistic interactions between N deposition and other drivers on diversity, particularly climate change, habitat conversion, and other pollutants, (4) the relative effects of reduced and oxidised N, (5) rates of recovery, and (6) cascades of impacts through the vegetation, soil biota (including microbes), and above-ground fauna.

488

Nitrogen manipulation experiments should be continued, and new experiments initiated in vulnerable habitats, particularly in areas with low N deposition, since accumulated plant-available nitrogen in the soil may have already impacted biodiversity in regions receiving elevated N deposition over many years. Manipulation studies (existing and new) should, if possible, incorporate a treatment cessation to gain new information on rates of recovery. Historical records should be further utilised to establish the rate of change that has already occurred across Europe. Finally, we suggest that a European-wide monitoring network covering a range of habitats be initiated, using consistent methods, to provide information on the long-term effects of air pollution on biodiversity.

Acknowledgements This chapter was prepared with the support of the NinE Programme of the European Science Foundation, the NitroEurope IP (funded by the European Commission), and the COST Action 729. The authors would like to thank the ENA editorial team, two internal reviewers and two external reviewers for their valuable comments on the manuscript.

Appendix 20.1 Descriptions of models introduced in Section 20.4 BERN:€ Bioindication for Ecosystem Regeneration towards Natural conditions, is a semi-empirical model for estimating plant niches based on abiotic site factors (Schlutow and Huebener, 2004). MAGIC:€ Model of Acidification of Groundwater in Catchments (Cosby et€al., 2001) MAGIC is a lumped-parameter model which simulates chemical processes in soils at catchment level to relate atmospheric deposition of acidifying compounds to soil, groundwater and surface water chemistry. MOVE (GBMOVE):€ Model of Vegetation (Latour and Reiling, 1993) MOVE predicts the occurrence of plant species based on field-based empirical relationships between species distribution and Ellenberg indicators for water and nutrient status. About 900 plant species are covered, calibrated on almost 160€000 vegetation relevés. The Ellenberg indicator values are derived from the biogeochemical status (e.g. soil pH, soil C/N) that is the output of the biogeochemical model. GBMOVE is MOVE calibrated for vegetation in Great Britain. ForSAFE:€ Forest SAFE (Wallman et€ al., 2005). ForSAFE merges SAFE (see below) with a forest simulator for plant growth, litter fall and organic matter decomposition, thus integrating the inorganic soil, organic matter and decomposition, vegetation growth, uptake and respiration, and hydrology. The model requires information on environmental drivers for deposition, climate, and management, and dynamically simulates changes in the ecosystem based on changes in these drivers. In comparison to the soil-oriented MAGIC and SMART2, ForSAFE is vegetation-driven and contains more dynamic

Nancy B. Dise

equations; MAGIC and SMART2 uses more aggregated parameters. ForSAFE has been connected with Veg, the vegetation response and composition model (see below), which reads state variables (soil chemistry and moisture, shading, and temperature) from ForSAFE and simulates the composition of the ground vegetation community. NTM:€Nature Technical Model (Wamelink et€al., 2003) NTM uses four dominant factors to characterise the environment:€ groundwater level, soil pH, soil nitrogen availability, and management. The first three factors are outputs of the biogeochemical model; the management regime is an input. In NTM the relation between vegetation and these environmental factors is determined by regression. The vegetation can be characterised on two levels:€generalised ‘potential’ biodiversity, or vegetation type. SAFE:€Soil Acidification in Forest Ecosystems is a multilayer soil geochemical model for soil solution chemistry, weathering and cation exchange (Alveteg, 1998). SMART2:€ Simulation model for acidification’s regional trends (Kros et€al., 1995) SMART2 is an extension of the one-compartment soil acidification model SMART (De Vries et€al., 1989) by including a nutrient cycling model and describing the major hydrological and biogeochemical processes in both the litter layer and mineral soil. As with SMART, it consists of a set of mass balance equations describing the soil input-output relationships, and a set of equations describing the rate-limited and equilibrium soil processes. SMART 2 is an improvement over SMART in simulating two soil layers (rather than one in SMART), and it includes a complete nutrient cycle (litterfall, mineralisation, root uptake, immobilisation, nitrification and denitrification) for base cations and N. SMART2 is designed for more regional applications than MAGIC and ForSAFE; the latter two models operate primarily at the catchment or plot scale. Veg:€ VEGetation model (Belyazid et€ al., 2006; Sverdrup et€al., 2007) Veg is a process-based model simulating changes in the composition of the ground vegetation in response to changes in biotic and abiotic factors, using plant species-specific information on habitat preferences. Veg includes an integration of the N cycle with process kinetics and feedbacks to the chemistry, organic matter decomposition and growth cycles of the vegetation. Changes in biotic and abiotic factors included are soil solution nitrogen and phosphorus concentration, soil acidity, soil moisture, light intensity at the forest floor, temperature, grazing pressure and competition between species based on height and root depth. The model combines these responses to predict how an entire plant community would evolve if one or many environmental drivers change over time.

References Aaby, B. (1994). Monitoring Danish raised bogs. In:€Mires and Man:€Mire Conservation in a Densely Populated Country€– The Swiss Experience, ed. A. Grunig, Kosmos, Birmensdorf, pp.€284–300.

Alveteg, M. (1998). Dynamics of forest soil chemistry. PhD thesis, Department of Chemical Engineering II, Lund University, Lund, Sweden. Arens, S. J. T., Sullivan, P. F. and Welker, J. M. (2008). Nonlinear responses to nitrogen and strong interactions with nitrogen and phosphorus additions drastically alters the structure and function of a high arctic ecosystem. Journal of Geophysical Research, 113, G03209. Ashmore, M. R. (2005). Assessing the future global impacts of ozone on vegetation. Plant, Cell and Environment, 28, 949–964. Asman, W. A. H., Sutton, M. A. and Schjorring, J. K. (1998). Ammonia:€emission, atmospheric transport and deposition. New Phytologist, 139, 27–48. Balmford, A., Bennun, L., ten Brink, B. et€al. (2005). The Convention on Biological Diversity’s 2010 target. Science, 307, 212–213. Barker, C. G. (2001). The impact of management on heathland response to increased nitrogen deposition. PhD thesis, University of London. Bassin, S., Volk, M., Suter, M., Buchmann, N. and Fuhrer, J. (2007). Nitrogen deposition but not ozone affects productivity and community composition of subalpine grassland after 3 yr of treatment. New Phytologist, 175, 523–534. Bates, J. W. (2002). Effects on bryophytes and lichens. In:€Air Pollution and Plant Life (2nd edition), ed. J. N. B. Bell and M. Treshow, John€Wiley and Sons, Chichester, pp. 309–342. Belyazid, S., Westling, O. and Sverdrup, H. (2006). Modelling changes in forest soil chemistry at 16 Swedish coniferous forest sites following deposition reduction. Environmental Pollution, 144, 596–609. Bender, D. J., Contreras, T. A. and Fahrig, L. (1998). Habitat loss and population decline:€a meta-analysis of patch size effect. Ecology, 79, 517–533. Berdowski, J. J. M. (1993). The effect of external stress and disturbance factors on Calluna-dominated heathland vegetation. In:€Heathlands:€Patterns and Processes in a Changing Environment, ed. R. Aerts, and G. W. Heil, Kluwer, Dordrecht, pp. 85–124. Bergamini, A. and Pauli, D. (2001). Effects of increased nutrient supply on bryophytes in montane calcareous fens. Journal of Bryology, 23, 331–339. Beusink, P., Nijssen, M., Van Duinen, G.-J. and Esselink, H. (2003). Broed- en voedselecologie van Grauwe klauwieren in intacte kustduinen bij Skagen, Denemarken. Stichting Bargerveen, Afdeling Dierecologie, K.U. Nijmegen. Bobbink, R. (1991). Effects of nutrient enrichment in Dutch chalk grassland. Journal of Applied Ecology, 28, 28–41. Bobbink, R. (2008). The derivation of dose–response relationships 4 between N load, N exceedance and plant species richness for EUNIS habitat classes. In: Critical Load, Dynamic Modelling and Impact Assessment in Europe, ed. J.-P. Hetterling, M. Posch, and J. Slootweg, Coordination Centre for Effects, Netherlands Environmental Assessment Agency, pp.€63–72. http://www.mnp. nl/en/themasites/cce/publications/cce-status-report-2008/ Bobbink, R., Hornung, M. and Roelofs, J. G. M. (1998). The effects of air-borne nitrogen pollutants on species diversity in natural and semi-natural European vegetation. Journal of Ecology, 86, 717–738. Bobbink, R. and Lamers, L. P. M. (2002). Effects of increased nitrogen deposition. In: Air Pollution and Plant Life (2nd edition), ed. J.€N.€B. Bell and M. Treshow, John Wiley and Sons, Chichester, pp.€201–235. Bobbink, R., Ashmore, M., Braun, S., Flückiger, W. and van den Wyngaert, I. J. J. (2003). Empirical critical loads for natural and semi-natural ecosystems:€2002 update. In:€Empirical Critical Loads

489

Nitrogen as a threat to biodiversity of Nitrogen, ed. B. Achermann and R. Bobbink, Swiss Agency for Environment Forests and Landscape, Bern, pp. 43–169. Bobbink, R., Hicks, K., Galloway, J. et€al. (2010). Global assessment of nitrogen deposition effects on plant terrestrial biodiversity:€a synthesis. Ecological Applications, 20, 30–59. Bolan, N. S., Hedley M. J. and White, R. E. (1991). Processes of soil acidification during nitrogen cycling with emphasis on legumebased pastures. Plant and Soil, 134, 53–63. Boxman, A. W., Blanck, K., Brandrud, T. E. et€al. (1998). Vegetation and soil biota response to experimentally-changed nitrogen inputs in coniferous forest ecosystems of the NITREX project. Forest Ecology and Management, 101, 65–79. Bragazza, L. and Limpens, J. (2004). Dissolved organic nitrogen dominates in European peat bogs under increasing atmospheric nitrogen deposition. Global Biogeochemical Cycles, 18, GB4018. Britton, A. J. and Fisher, J. M. (2007). Interactive effects of nitrogen deposition, fire, and grazing on diversity and composition of lowalpine prostrate Calluna vulgaris heathland. Journal of Applied Ecology, 44, 125–135. Brunet, J., Diekmann, M. and Falkengren-Grerup, U. (1998). Effects of nitrogen deposition on field layer vegetation in south Swedish oak forests. Environmental Pollution, 102, 35–40. Brunsting, A. M. H. and Heil, G. W. (1985). The role of nutrients in the interactions between a herbivorous beetle and some competing plant species in heathlands. Oikos, 44, 23–26. Calvo, L., Alonso, I., Fernandez, A. J. and De Luis, E. (2005). Shortterm study of effects of fertilisation and cutting treatments on the vegetation dynamics of mountain heathlands in Spain. Plant Ecology, 179, 181–191. Cape, J. N, van der Eerden, L. J., Sheppard, L. J., Leith, I. D. and Sutton,€M. A. (2009). Evidence for changing the critical level of ammonia. Environmental Pollution, 157, 1033–1037. Caporn, S. J. M., Carroll, J. A., Studhome, C. and Lee, J. A. (2006). Recovery of ombrotrophic Sphagnum mosses in relation to air pollution in the southern Pennines. Report to Moors for the Future, Edale. Carey, P. D., Wallis, S., Chamberlain, P. M. et€al. (2008). Countryside Survey:€UK Results from 2007. NERC/Centre for Ecology and Hydrology, Edinburgh. CEC (2001). EU Sustainable Development Strategy, 2006 Update and 2009 Review. http://ec.europa.eu/environment/eussd/ CEC (2002). 6th Community Environment Action Programme. http:// ec.europa.eu/environment/newprg/index.htm CEC (2006). Halting the Loss of Biodiversity by 2010€– and Beyond€– Sustaining Ecosystem Services for Human Well-Being. http:// ec.europa.eu/environment/nature/biodiversity/comm2006/ bap_2006.htm Clark, C. M. and Tilman, D. (2008). Loss of prairie grass species after chronic low-level nitrogen deposition to prairie grasslands. Nature, 712–715. Clark, C. M., Cleland, E. E., Collins, S. L. et€al. (2007). Environmental and plant community determinants of species loss following nitrogen enrichment. Ecology Letters, 10, 596–607. CLRTAP:€Convention on Long-Range Transboundary Air Pollution. (2010). http://www.unece.org/env/lrtap Convention on Biological Diversity (CBD) (1992). Text of the Convention on Biological Diversity. http://www.cbd.int/ convention/ Convention on Biological Diversity (CBD) (2010). International Year of Biodiversity. http://www.cbd.int/2010

490

Cosby, B. J., Ferrier, R. C., Jenkins, A. and Wright, R. F., (2001). Modelling the effects of acid deposition:€refinements, adjustments and inclusion of nitrogen dynamics in the MAGIC model. Hydrology and Earth System Sciences, 5, 499–517. De Graaf, M. C. C., Bobbink, R., Roelofs, J. G. M. and Verbeek, P. J.€M. (1998). Differential effects of ammonium and nitrate on three heathland species. Plant Ecology, 135, 185–196. De Graaf, M. C. C., Bobbink, R., Smits, N. A. C., Van Diggelen, R. and Roelofs, J. G. M. (2009). Biodiversity, vegetation gradients and key biogeochemical processes in the heathland landscape. Biological Conservation, 142, 2191–2201. De Vries, W., Kros, J., Reinds, G. J. et€al. (2007). Developments in Modelling Critical Nitrogen Loads for Terrestrial Ecosystems in Europe, Alterra Wageningen UR, Wageningen, the Netherlands. Report 1382. De Vries, W., Posch, M. and Kämäri, J. ( 1989). Simulation of the long-term soil response to acid deposition in various buffer ranges. Water, Air and Soil Pollution, 48, 349–390. De Vries, W., Wamelink, W., van Dobben, H. et€al. (2010). Use of dynamic soil-vegetation models to assess impacts of nitrogen deposition on plant species composition and to estimate critical loads:€an overview. Ecological Applications, 20, 60–79. Dise, N. B., Rothwell, J. J., Gauci, V., van der Salm, C. and de Vries,€W. (2009). Predicting dissolved inorganic nitrogen leaching in European forests using two independent databases. Science of the Total Environment, 407, 1798–1808. Duprè, C., Stevens, C. J., Ranke, T. et€al. (2010). Changes in species€richness and composition in European acidic grasslands€over the past 70 years:€the contribution of cumulative atmospheric nitrogen deposition. Global Change Biology, 16, 344–357. Edmondson, J. L., Carroll, J. A., Price, E. A. C. and Caporn, S. J. M. (2010). Bio-indicators of nitrogen pollution in heather moorland. Science of the Total Environment, 408, 6202–6209. EEA (2009). Progress towards the European 2010 Biodiversity Target, EEA Report No. 04/2009, EEA Copenhagen. http://www.eea. europe.eu/publications/progress-towards-the-european-2010biodiversity-target Ehrlich, P. A. and Ehrlich, A. H. (1992). Economics of biodiversity loss. Ambio, 21, 219–226. Ellenberg, H. (1996). Vegetation Mitteleuropas Mit Den Alpen, 5th edn. Stuttgart, Verlag Eugen Ulmer. Elser, J. J., Bracken, M. E .S., Cleland, E. E. et€al. (2007). Global analysis of nitrogen and phosphorus limitation of primary producers in freshwater, marine and terrestrial ecosystems. Ecology Letters, 10, 1135–1142. Esselink, H., Van Duinen, G.-J., Nijssen, M. et€al. (2007). De grauwe klauwier mist kevers door verruigende duinen. Vakblad Natuur, Bos und Landschap, 2007–4, 22–24. EU NEC 2001 Directive 2001/81/EC of the European Parliament and the Council; http://ec.europa.eu/environment/air/pdf/nec_eu_27.pdf Falkengren-Grerup, U. (1998). Nitrogen response of herbs and graminoids in experiments with simulated acid soil solution. Environmental Pollution, 102 (Suppl), 93–99. Fenn, M. E., Baron, J. S., Allen, E. B. et€al. (2003). Ecological effects of nitrogen deposition in the western United States. BioScience, 53, 404–420. Fenn, M. E., Jovan, S., Yuan, F. et€al. (2008). Empirical and simulated critical loads for nitrogen deposition in California mixed conifer forests. Environmental Pollution, 155, 492–511.

Nancy B. Dise Flückiger, W., Braun, S. and Hiltbrunner, E. (2002). Effects of air pollutants on biotic stress. In:€Air Pollution and Plant Life (2nd edition), ed. J. N. B. Bell, and M. Treshow, John Wiley and Sons, Chichester, pp. 379–406. Fowler, D. (2002). Pollutant deposition and uptake by vegetation. In:€Air Pollution and Plant Life (2nd edition), ed. J. N. B. Bell, and M. Treshow, John Wiley and Sons, Chichester, pp. 43–67. Frati, L., Santoni, S., Nicolardi, V. et€al. (2007). Lichen biomonitoring of ammonia emission and nitrogen deposition around a pig stockfarm. Environmental Pollution, 146, 311–316. Gaston, K. J. (1996). What is biodiversity? In: Biodiversity:€A Biology of Numbers and Difference, ed. K. J. Gaston,€Blackwell Science, Oxford. Gaston, K. J. and Spicer, J. I. (2004). Biodiversity:€An Introduction. Blackwell Science, Oxford. Gerdol, R., Petraglia, A., Bragazza, L., Iacumin, P. and Brancaleoni,€L. (2007). Nitrogen deposition interacts with climate in affecting production and decomposition rate in Sphagnum mosses. Global Change Biology, 13, 1810–1821. Gigon, A. and Rorison, I. H. (1972). The response of some ecologically distinct plant species to nitrate- and to ammonium-nitrogen. Journal of Ecology, 60, 93–102. Gimeno, B. S. (2009). New results on critical loads of nitrogen in southern Europe. In:€Proceedings of the workshop of the Coordination Centre for Effects, 11–15 May 2009. http://www.pbl. nl/en/themasites/cce/workshops/stockholm-2009/index.html. Gordon, C., Wynn, J. M. and Woodin, S. J. (2001). Impacts of increased nitrogen supply on high Arctic heath:€the importance of bryophytes and phosphorus availability. New Phytologist, 149, 461–471. Gotelli, N. J. and Colwell, R. K. (2001). Quantifying biodiversity: procedures and pitfalls in the measurement and comparison of species richness. Ecology Letters, 4, 379–391. Grizzetti, B., Bouraoui, F., Billen, G. et€al. (2011). Nitrogen as a threat to European water quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Gunnarsson, U., Malmer, N. and Rydin, H. (2002). Dynamics or constancy in Sphagnum dominated mire ecosystems: a 40 year study. Ecography, 25, 685–704. Grupa, S. V. (2003). Effects of atmospheric ammonia (NH3) on terrestrial vegetation: a review. Environmental Pollution,124, 179–221. Hallingback, T. (1992). The effect of air pollution on mosses in Southern Sweden. Biological Conservation, 59, 163–170. Hampe, A. and Petit, R. J. (2005). Conserving biodiversity under climate change:€the rear edge matters. Ecology Letters, 8, 461–467. Hegg, O., Feller, U., Dähler, W. and Scherrer, C. (1992). Long-term influence of fertilization in a Nardetum: phytosociology of the pasture and nutrient contents in leaves. Vegetatio, 103, 151–158. Heil, G. W. and Diemont, W. H. (1983). Raised nutrient levels change heathland into grassland. Vegetatio, 53, 113–120. Hertel, O., Reis, S. and Ambelas Skjøth, C. (2011). Nitrogen processes in the atmosphere. In:€The European Nitrogen Assessment, ed. M.€A.€Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Hettelingh, J. P., Posch, M., Slootweg, J., Bobbink, R. and Alkemade,€R. (2008a). Tentative dose–response function applications for integrated assessment. In:€Critical Load, Dynamic Modelling and Impact Assessment in Europe, ed. J.-P. Hettelingh, M. Posch,€and J. Slootweg, Coordination Centre for Effects, Netherlands

Environmental Assessment Agency, pp. 83–89. http://www.mnp. nl/en/themasites/cce/publications/cce-status-report-2008/ Hettelingh, J.-P., Posch, M. and Slootweg, J. (2008b). Status of the critical loads database and impact assessment. In: Critical Load, Dynamic Modelling and Impact Assessment in Europe. eds. J.-P. Hettelingh, M. Posch, and J. Slootweg, Coordination Centre for Effects, Netherlands Environmental Assessment Agency, pp.€15–28. http://www.mnp.nl/en/themasites/cce/publications/ cce-status-report-2008/ Intergovernmental Panel on Climate Change (IFCC) (2007). Climate Change: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, eds. S. Solomon, D. Qin, M. Manning et al. Cambridge University Press. Jenssen, M. (2008). Analysing and modelling the effects of nitrogen deposition on species diversity in North-east German forests. In:€Proceedings of the workshop of the Coordination Centre for Effects, 21–23 April 2008. http://www.pbl.nl/en/themasites/cce/ workshops/berne-2008/index.html. Johnson, D. W. and Cole, D. W. (1980). Anion mobility in soils:€relevance to nutrient transport from forest ecosystems. Environment International, 3, 79–90. Jones, M. L. M. (2005). Nitrogen deposition in upland grasslands: critical loads, management, and recovery. PhD thesis, University of Sheffield. Jones, M. L. M., Wallace, H. L., Norris, D. et€al. (2004). Changes in vegetation and soil characteristics in coastal sand dunes along a gradient of atmospheric nitrogen deposition. Plant Biology, 6, 598–605. Kinzel, S. (1982). Pflanzenökologie und Mineralstoffwechsel. Ullmer, Stuttgart. Kleijn, D., Bekker, R. M., Bobbink, R., De Graaf, M. C. C. and Roelofs,€J. G. M. (2008). In search for key biogeochemical factors affecting plant species persistence in heathland and acidic grasslands:€a comparison of common and rare species. Journal of Applied Ecology, 45, 680–687. Kros, J., Reinds, G. J., de Vries, W., Latour, J. B. and Bollen, M. J. S., 1995. Modelling abiotic site factors in response to atmospheric deposition and upward seepage. In:€Scenario Studies for the Rural Environment, ed. J. F. T. Schoute, P. A. Finke, F. R. Veeneklaas and H. P. Wolfert, Kluwer, Dordrecht, pp. 445–448. Latour, J. B. and Reiling, R. (1993). A multiple stress model for vegetation (‘MOVE’):€a tool for scenario studies and standardsetting. Science of the Total Environment, 134, 1513–1526. Limpens, J., Berendse, F. and Klees, H. (2004). How phosphorus availability affects the impact of nitrogen deposition on Sphagnum and vascular plants in bogs. Ecosystems, 7(8), 793–804. Mace, G., Masundire, H. and Baille, J. (2005). Biodiversity. In: Millenium Ecosystem Assessment Volume 1:€Current States and Trends. Island Press, Washington. Madan, N. J., Deacon, L. J. and Robinson, C. H. (2007). Greater nitrogen and/or phosphorus availability increase plant species’ cover and diversity at a High Arctic polar semidesert. Polar Biology, 30, 559–570. Makipaa, R. and Heikkinen, J. (2003). Large-scale changes in abundance of terricolous bryophytes and macrolichens in Finland. Journal of Vegetation Science, 14, 497–508. Malmer, N., Albinsson, C., Svensson, B. and Wallen, B. (2003). Interferences between Sphagnum and vascular plants:€effects on plant community structure and peat formation. Oikos, 100, 469–482.

491

Nitrogen as a threat to biodiversity Maskell, L. C., Smart, S. M., Bullock, J. M., Thompson, K. and Stevens,€C. J. (2010). Nitrogen deposition causes widespread loss of species richness in British habitats. Global Change Biology, 16, 671–679. Mitchell, E. A. D., Buttler, A., Grosvernier, P. et€al. (2002). Contrasted effects of increased N and CO2 supply on two keystone species in peatland restoration and implications for global change. Journal of Ecology, 90, 529–533. Mitchell, R. J., Truscot, A. M., Leith, I. D. et€al. (2005). A study of the epiphytic communities of Atlantic oak woods along an atmospheric nitrogen deposition gradient. Journal of Ecology, 93, 482–492. Mountford, J. O., Lakhani, K. H. and Holland, R. J. (1996). Reversion of grassland vegetation following the cessation of fertilizer application. Journal of Vegetation Science, 7, 219–228. Myers, N., Mittermeier, R. A., Mittermeier, C. G. et€al. (2000) Biodiversity hotspots for conservation priorities. Nature, 403, 853–858. Nellemann, C. and Thomsen, M. G. (2001). Long-term changes in forest growth: potential effects of nitrogen deposition and acidification. Water, Air and Soil Pollution, 128, 197–205. Nihlgård, B. (1985). The ammonium hypothesis: an additional explanation to the forest dieback in Europe. Ambio, 14, 2–8. Nijssen, M., Alders, K., Van der Smissen, N. and Esselink, H. (2001). Effects of grass encroachment and grazing management on carabid assemblages of dry dune grasslands. Proceedings Experimental and Applied Entomology. Amsterdam, 12, 113–120. Nilsson, J. and Grennfelt, P. (eds.) (1988) Critical Loads for Sulphur and Nitrogen, UNECE/Nordic Council workshop report, Skokloster, Sweden. Nordic Council of Ministers,€Copenhagen. Nordin, A., Strengbom, J., Witzell, J., Nasholm, T. and Ericson, L. (2005). Nitrogen deposition and the biodiversity of boreal forests:€Implications for the nitrogen critical load. Ambio, 34, 20–24. Nordin, A., Strengbom, J. and Ericson, L. (2006). Responses to ammonium and nitrate additions by boreal plants and their natural enemies. Environmental Pollution, 141, 167–174. Ockinger, E., Hammarstedt, O., Nilsson, S. G. and Smith, H. G. (2006). The relationship of local extinctions of grassland butterflies and increased soil nitrogen levels. Biological Conservation, 128, 564–573. Oenema, O., Bleeker, A., Braathen, N. A. et€al. (2011). Nitrogen in current European policies. In:€The European Nitrogen Assessment. ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Olff, H. and Bakker, J. P. (1991). Long-term dynamics of standing crop and species composition after the cessation of fertilizer application to mown grassland. Journal of Applied Ecology, 28, 1040–1052. Olsson, M. O. and Falkengren-Grerup, U. (2000). Potential nitrification as an indicator of preferential uptake of ammonium or nitrate by plants in an oak understory. Annals of Botany, 85, 299–305. Pardo, L. H., Robin-Abbott, M. J. and Driscoll, C. T. (eds.) (2010). Assessment of N Deposition Effects and Empirical Critical Loads of N for Ecoregions of the United States, USDA Forest Service General Technical Report. http://www.nrs.fs.fed.us/clean_air_water/ clean_water/critical_loads/local-resources/docs/Empirical_CLS_ of_N_100414.pdf Pauli, D., Peintinger, M. and Schmid, B. (2002). Nutrient enrichment in calcareous fens:€effects on plant species and community structure. Basic and Applied Ecology, 3, 255–266.

492

Paulissen, M. P. C. P., Van der Ven, P. J. M., Dees, A. J. and Bobbink,€R. (2004). Differential effects of nitrate and ammonium on three fen bryophyte species in relation to pollutant nitrogen input. New Phytologist, 164, 551–558. Pearce, I. S. K. and van der Wal, R. (2002). Effects of nitrogen deposition on growth and survival of montane Racomitrium lanuginosum heath. Biological Conservation, 104, 83–89. Pearson, J. and Stewart, G. R. (1993). The deposition of atmospheric ammonia and its effects on plants. New Phytologist, 125, 283–305. PEBLDS (Pan-European Biological and Landscape Diversity Strategy) (2005). http://www.peblds.org Phoenix, G. K., Hicks, W. K., Cinderby, S. et€al. (2006). Atmospheric nitrogen deposition in world biodiversity hotspots:€the need for a greater global perspective in assessing N deposition impacts. Global Change Biology, 12, 470–476. Pinho, P., Branquinho, C., Cruz, C. et€al. (2009). Assesment of critical levels of atmospheric ammonia for lichen diversity in cork-oak woodland, Portugal. In:€Atmospheric Ammonia, ed. M. A. Sutton, S. Reis and S. M. H. Baker, Springer, New York, pp. 109–120. Pitcairn, C. E. R., Leith, I. D., Sheppard, L. J. et€al. (1998). The relationship between nitrogen deposition, species composition and foliar nitrogen concentrations in woodland flora in the vicinity of livestock farms. Environmental Pollution, 102, 41–48. Pitcairn, C. E. R., Leith, I. D., van Dijk, N., et€al. (2009). The application of transects to assess the effects of ammonia on woodland groundflora. In:€Atmospheric Ammonia, ed. M. A. Sutton, S. Reis and S. M. H. Baker, Springer, New York, pp. 59–69. Porter, E. and Johnson, S. (2007). Translating science into policy:€using ecosystem thresholds to protect resources in Rocky Mountain National Park. Environmental Pollution, 149, 268–280. Power, S. A., Green, E. R., Barker, C. G., Bell, J. N. B. and Ashmore, M.€R. (2006). Ecosystem recovery:€heathland response to a reduction in nitrogen deposition. Global Change Biology, 12, 1241–1252. Preston, C. D., Pearman, D. A. and Dines, T. D. (eds.) (2002). New Atlas of the British and Irish Flora. Oxford University Press. Redbo-Tortensson, P. (1994). The demographic consequences of nitrogen fertilization of a population of sundew, Drosera rotundifolia. Acta botanica neerlandica, 43, 175–188. Remke, E., Brouwer, E., Kooijman, A. et€al. (2009). Even low to medium nitrogen deposition impacts vegetation of dry, coastal dunes around the Baltic Sea. Environmental Pollution, 157, 792–800. Rihm, B., Urech, M. and Peter, K. (2009). Mapping ammonia emissions and concentrations for Switzerland: effects on lichen vegetation. In:€Atmospheric Ammonia, ed. M. A. Sutton, S. Reis and S. M. H. Baker, Springer, New York, pp. 87–92. Roelofs, J. G. M., Kempers, A. J., Houdijk, A. L. F. M. and Janson, J. (1985). The effects of air-borne ammonium sulphate on Pinus nigra in the Netherlands. Plant and Soil, 42, 372–377. RoTAP (2010). Review of Transboundary Air Pollution. Acidification, Eutrophication, Ground-Level Ozone and Heavy Metals, in the UK. http://www.rotap.ceh.ac.uk. Sala, O. E., Chapin III, F. S., Armesto, J. J. et€al. (2000). Global biodiversity scenarios for the year 2100. Science, 287, 1770–1774. Schlutow, A. and Hübener, P. (2004):€The BERN Model:€Bioindication for Ecosystem Regeneration towards Natural Conditions, UBA-Texte Nr 22/2004, http://www.umweltdaten.de/publikationen/fpdfl/2784.pdf Sheppard, L. J., Leith, I. D., Crossley, A. et€al. (2009). Long-term cumulative exposure exacerbates the effects of atmospheric ammonia on an ombrotrophic bog:€Implications for critical levels.

Nancy B. Dise In:€Atmospheric Ammonia, eds. M. A. Sutton, S. Reis and S. M. H. Baker, Springer, New York, pp. 49–58. SchÖpa, W., Posch, M., Mylona, S. and Johansson, M. (2003). Longterm development of acid deposition (1880–2030) in sensitive freshwater regions in Europe. Hydrology and Earth System Sciences, 7, 436–446. Slootweg, J., Posch, M. and Hettelingh, J.-P. (eds.) (2007). Critical Loads of Nitrogen and Dynamic Modelling, CCE Progress Report 2007. http://www.mnp.nl/en/themasites/cce/publications/cceprogress-report-2007/ Smits, N. A. C., Willems, J. H. and Bobbink, R. (2008). Long-term after-effects of fertilisation on the restoration of calcareous grasslands. Applied Vegetation Science, 11, 279–286. Spranger, T., Hettelingh, J.-P., Slootweg, J. and Posch, M. (2008). Modelling and mapping long-term risks due to reactive nitrogen effects:€an overview of LRTAP convention activities. Environmental Pollution, 154, 482–487. Stevens, C. J., Dise, N. B., Mountford, J. O. and Gowing, D. J. (2004). Impact of nitrogen deposition on the species richness of grasslands. Science, 303, 1876–1879. Stevens, C. J., Dise, N. B., Gowing, D. J. G. and Mountford, J. O. (2006). Loss of forb diversity in relations to nitrogen deposition in the UK:€regional trends and potential controls. Global Change Biology, 12, 1823–1833. Stevens, C. J., Maskell, L. C., Smart, S. M. et€al. (2009). Identifying indicators of atmospheric nitrogen deposition impacts in acid grasslands. Biological Conservation 142, 2069–2075. Stevens, C. J., Dupré, C., Dorland, E. et€al. (2010). Nitrogen deposition threatens species richness of grasslands across Europe. Environmental Pollution, in press. Strengbom, J., Nordin, A., Näsholm, T. and Ericson, L. (2001). Slow recovery of boreal forest ecosystem following decreased nitrogen input. Functional Ecology, 15, 451–457. Strengbom, J., Walheim, M., Näsholm, T. and Ericson, L. (2003). Regional differences in the occurrence of understorey species reflect nitrogen deposition in Swedish forests. Ambio, 32, 91–97. Suding, K. N., Collins, S. L., Gough, L. et€al. (2005). Functional- and abundance-based mechanisms explain diversity loss due to N fertilization. Proceedings of the National Academy of Sciences of the USA, 102, 4387–4392. Sutton, M. A., Erisman, J. W., Dentener, F. and Miller, D. (2008). Ammonia in the environment:€from ancient times to present. Environmental Pollution, 156, 583–604. Sutton, M. A., Wolseley, P. A, Leith, I. D. et€al. (2009). Estimation of the ammonia critical level for epiphytic lichens based on observations at farm, landscape and national scales. In:€Atmospheric Ammonia, ed. M. A. Sutton, S. Reis and S. M. H. Baker, Springer, New York, pp. 71–86. Sutton, M. A., Whitfield, C., Bealey, H. J. and Hicks, W. K. (2010). Summary for Policy Makers. In: Proceedings from a COST 729/ Nine/ESF/CCW/JNCC workshop, Brussels, May 2009. http:// cost729.ceh.ac.uk/n2workshiop/documents. Sverdrup, H., Belyazid, S., Nihlgård, B. and Ericson, L. (2007). Modelling change in ground vegetation response to acid and nitrogen pollution, climate change and forest management in Sweden 1500–2100 A.D. Water, Air and Soil Pollution:€Focus, 7, 163–179. Tamis, W. L. M., van’t Zelfte, M., van der Meijden, R., Groen, C.€L.€G. and Udo de Haes, H. A. (2005). Ecological interpretation of changes in the Dutch flora in the 20th century. Biological Conservation, 125, 211–224.

Taylor, K., Woof, C., Ineson, P. et€al. (1999). Variation in seasonal precipitation chemistry with altitude in the northern Pennines, UK. Environmental Pollution, 104, 1–9. Thomas, C. D., Cameron, A., Green, R. E. et€al. (2004). Extinction risk from climate change. Nature, 427, 145–148. Throop, H. L. and Lerdau, M. T. (2004). Effects of nitrogen deposition on insect herbivory:€implications for community and ecosystem processes. Ecosystems, 7, 109–133. Thuiller, W., Lavorel, S., Araújo, M. B., Sykes, M. T. and Prentice, I. C. (2005). Climate change threats to plant diversity in Europe. Proceedings of the National Academy of Sciences of the USA 102, 8245–8250. Tilman, D. and Downing, J. A. (1994). Biodiversity and stability in grasslands. Nature, 367, 363–365. Tilman, D., Wedin, D. and Knops, J. (1996). Productivity and sustainability influenced by biodiversity in grassland ecosystems. Nature, 379, 718–720. Twenhöven, F. L. (1992). Competition between two Sphagnum species under different deposition levels. Journal of Bryology, 17, 71–80. Ulrich, B. (1983). Soil acidity and its relation to acid deposition. In: Effects of Accumulation of Air Pollutants in Ecosystems, ed. B. Ulrich, and J. Pankrath, Reidel Publishing, Boston, MA, pp. 127–146. UNECE (2010). Protocol concerning the Control of Emissions of Nitrogen Oxides. http://www.unece.org/env/lrtap/nitr_h1.htm. Usher, M. B. (1986). Invasibility and wildlife conservation:€Invasive species on nature reserves. Philosophical Transactions of the Royal Society of London, Series B, Biological Sciences, 314, 695–709. Van Breemen, N., Burrough, P. A., Velthorst, E. J. et€al. (1982). Soil acidification from atmospheric ammonium sulphate in forest canopy throughfall. Nature, 299, 548–550. Van den Berg, L. J. L., Dorland, E., Vergeer, P. et€al. (2005a). Decline of acid-sensitive plant species in heathland can be attributed to ammonium toxicity in combination with low pH. New Phytologist, 166, 551–564 Van den Berg, L. J. L., Tomassen, H. B. M., Roelofs, J. G. M. and Bobbink, R. (2005b). Effects of nitrogen enrichment on coastal dune grassland:€A mesocosm study. Environmental Pollution, 138, 77–85. Van den Berg, L. J. L., Peters, C. J. H., Ashmore, M. R. and Roelofs, J.€G. M. (2008). Reduced nitrogen has a greater effect than oxidised nitrogen on dry heathland vegetation. Environmental Pollution, 154, 359–369. Van den Berg, L. J. L., Vergeer, P., Rich, T. C. G, et€al. (2010). Direct and indirect effects of nitrogen deposition on species composition change in calcareous grasslands. Global Change Biology, doi: 10.1111/j.1365-2486.2010.02345.x. Van der Salm, C. A., de Vries, W., Reinds, G. J. and Dise, N. B. (2007). N leaching across European forests:€derivation and validation of empirical relationships using data from intensive monitoring plots. Forest Ecology and Management, 238, 81–91. Van der Wal, R., Pearce, I., Brooker, R. et€al. (2003). Interplay between nitrogen deposition and grazing causes habitat degradation. Ecology Letters, 6, 141–146. Van Dijk, H. F. G., De Louw, M. H. J., Roelofs, J. G. M. and Verburgh, J.€J. (1990). Impact of artificial amonium-enriched rainwater on soils and young coniferous trees in a greenhouse. Part 2 - Effects on the trees. Environmental Pollution, 63, 41–60. Van Dobben, H. F., van Hinsberg, A., Schouwenberg, E. P. et€al. (2006). Simulation of critical loads for nitrogen for terrestrial plant communities in The Netherlands. Ecosystems, 9, 32–45.

493

Nitrogen as a threat to biodiversity Van Herk, C. M., Mathijssen-Spiekman, E. A. M. and de Zwart, D. (2003). Long distance nitrogen air pollution effects on lichens in Europe. Lichenologist, 35, 347–359. Van Hinsberg, A., Reijnem, R., Goedhart, P., de Knegt, B. and van Esbroek, M. (2008). Relation between critical load exceedance and loss of protected species. In:€Critical Load, Dynamic Modelling and Impact Assessment in Europe, ed. J. P. Hettelingh, M. Posch, and J. Slootweg, Coordination Centre for Effects, Netherlands Environmental Assessment Agency. http://www.pbl.nl/cce. Velthof, G., Barot, S., Bloem, J. et€al. (2011). Nitrogen as a threat to European soil quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et€al., Cambridge University Press. Vergeer, P., Rengelink, R., Ouborg, N. J. and Roelofs, J. G. M. (2003). Effects of population size and genetic variation on the response of Succisa pratensis to eutrophication and acidification. Journal of Ecology, 91(4), 600–609. Walther, G.-R., Post, E., Convey, P. et€al. (2002). Ecological responses to recent climate change. Nature, 416, 389–395. Wallman, P., Svenssson, M., Sverdrup, H. and Belyazid, S. (2005). ForSAFE: an integrated process-oriented forest model for long-term sustainability assessments. Forest Ecology and Management, 207, 19–36. Wamelink, G. W. W., ter Braak, C. F. J. and van Dobben, H. F. (2003). Changes in large-scale patterns of plant biodiversity predicted from environmental economic scenarios. Landscape Ecology, 18, 513–527.

494

Weiss, S. B. (1999). Cars, cows, and checkerspot butterflies: nitrogen deposition and management of nutrient-poor grasslands for a threatened species. Conservation Biology, 13, 1476–1486. Wiedermann, M. M., Nordin, A., Gunnarsson, U., Nilsson, M. B. and Ericson, L. (2007). Global change shifts vegetation and plant–parasite interactions in a boreal mire. Ecology 88, 454–464. Wiedermann, M. M., Gunnarsson, U., Nilsson, M. B., Nordin, A. and Ericson, L. (2009). Can small-scale experiments predict ecosystem responses? An example from peatlands. Oikos, 118, 449–456. Wilson, E. O. (ed.) 1988. Biodiversity. National Academy Press, Washington, DC. Williams, D. H., Gaston, K. J. and Humphries, C. J. (1997). Mapping biodiversity value world-wide: combining higher-taxon richness from different groups. Proceedings of the Royal Society, Series B, 264, 141–148. Wolseley, P. A., Leith, I. D., van Dijk, N. and Sutton, M. A. (2009). Macrolichens on twigs and trunks as indicators of ammonia concentrations across the UK: a practical method. In: Atmospheric Ammonia, ed. M. A. Sutton, S. Reis and S. M. H. Baker. Springer, pp. 101–108. Wright, R. F. and van Breemen, N. (1995). The NITREX project:€an introduction. Forest Ecology and Management, 71, 1–5.

Chapter

21

Nitrogen as a threat to European soil quality Lead author: Gerard Velthof Contributing authors: Sébastien Barot, Jaap Bloem, Klaus Butterbach-Bahl, Wim de Vries, Johannes Kros, Patrick Lavelle, Jørgen Eivind Olesen and Oene Oenema

Executive summary Nature of the problem • A large part of agricultural soils in Europe are exposed to high N inputs because of animal manure and chemical fertiliser use. Large parts of the European natural soils are exposed to high atmospheric N deposition. • High N inputs threaten soil quality, which may negatively affect food and biomass production and biodiversity and enhance emissions of harmful N compounds from soils to water and the atmosphere.

Approaches • An overview of the major soil functions and soil threats are presented, including a description of the objectives of the European Soil Strategy. • The major N threats on soil quality for both agricultural and natural soils are related to changes in soil organic content and quality, soil acidification, and loss of soil diversity. These threats are described using literature.

Key findings/state of knowledge • Generally, N has a positive effect on soil quality of agricultural soils, because it enhances soil fertility and conditions for crop growth. However, it generally has a negative effect on soil quality of natural soils, because it results in changes in plant diversity. • Soil acts as a filter and buffer for N, and protects water and atmosphere against N pollution. However, the filter and buffer capacity of soils is frequently exceeded by excess of N in both agricultural and natural soils, which results in emission of N to the environment. • Pyrite containing soils are found widespread in Europe. Nitrate removal from groundwater by pyrite oxidation increases concentrations of cations, heavy metals and sulphate. This causes problems when this water is used as drinking water. • Combined application of N and C (e.g. in manures) has a positive effect on soil organic matter content in agricultural soils. There are indications that the use of only N fertilisers may result in a decline of soil organic matter under certain conditions. • Application of fertilisers and manure and atmospheric deposition causes soil acidification. Soil acidification may lead to a decrease in crop and forest growth and leaching of components negatively affecting water quality, including heavy metals. Liming is widely used to reduce acidification of agricultural soils. • Nitrogen affects populations of soil organisms, which may change N transformations in soil and emissions to the environment. • The N inputs to agricultural and natural soils have decreased during the past ten years and the implementation of environmental policies may lead to a further decrease in the future. Consequently, the threat on the quality of European soils due to N will also decrease. Model simulations indicate that most of the European forest soils could recover from their acidified state within a few decades under the current N emission reduction plans.

Major uncertainties/challenges • The effect of N on soil organic matter content and quality is uncertain. Some studies suggest that only use of N fertiliser may result in a decline of soil organic matter content. • The effect of N on diversity of soil (micro) organisms and the effects of changes of soil biodiversity on soil fertility, crop production and N emissions towards the environment are not fully understood.

Recommendations • Soil quality plays an important role in the production of food, feed and biomass, provides a habitat for biodiversity and controls emissions of pollutants to water and air. It is important to recognise the functions of soils in N cycling in both research and policy aiming at decreasing N emissions, and improving food production, and soil biodiversity. The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

495

Nitrogen as a threat to European soil quality

21.1╇ Introduction 21.1.1╇ Soil quality and functions Soil is the top layer of the earth and is composed of a mixture of mineral and organic compounds, water, air, and living organisms. The soil map (Figure 21.1) shows a large diversity in soil types in Europe. Soil types are classified on the basis of soil characteristics and properties (such as organic matter content, pH, clay content), and soil horizons (three-dimensional bodies containing one or more soil properties). Soil quality can be defined as the capacity of a specific kind of soil to function, within natural or managed ecosystem boundaries, to sustain plant and animal productivity, maintain or enhance water and air quality, and support human health and habitation (Karlen et al., 1997; Schjønning et al., 2004). Major functions of soils are (EC, 2006; Karlen et al., 1997; Schjønning et al., 2004): • food and other biomass production; • storage, filtering, buffering and transformations of natural and anthropogenic produced substances, including N;

• • • •

a biological habitat and gene reservoir; sink for C; source of raw materials; physical and cultural environment for human and human activities; and • archival function for natural history.

21.1.2╇ EU soil policy Recognising the extent of soil degradation and associated environmental and social risks in Europe, the European Commission proposed a Thematic Strategy for Soil Protection (EC, 2006). This strategy also contained a proposal for a Framework Directive. The overall objective of this European Soil Strategy is protection and sustainable use of soil, based on the following guiding principles: • preventing further soil degradation and preserving its functions • when soil is used and its functions are exploited, action has to be taken on soil use and management patterns, and Figure 21.1 Soil Map of Europe (Source:€European Soil Bureau).

496

Gerard Velthof

• when soil acts as a sink/receptor of the effects of human activities or environmental phenomena, action has to be taken at the source. • restoring degraded soils to a level of functionality consistent at least with current and intended use, thus also considering the cost implications of the restoration of soil. In the Strategy, human activities, like inadequate agricultural and forestry practices, tourism, urban and industrial activities and construction works are indicated as the main factors that have an impact on soil functions. Main degradation threats to soil quality are erosion, salinisation, compaction, loss of organic matter, landslides, contamination, and sealing (Tóth et al., 2008). Soil degradation has a direct impact on the quality of water and air and threatens food and feed safety as well. Soil protection objectives are also included in the EU Water Framework Directive, because soils act as a filter and a buffer for water bodies and are a source of point and diffuse pollution sources of aquatic ecosystems.

21.1.3╇ Effect of soil degradation on N emissions The focus of this chapter is on the threats of N on soil quality. Inversely, however, soil quality also strongly affects N cycling and soil degradation processes may enhance N emissions to the environment. In this paragraph a short overview is presented about the effects of soil compaction, erosion, salinisation, contamination, and organic matter decline on N emissions to water and air. Soil compaction occurs by mechanical stress on the soil surface by agricultural and construction machinery and by overgrazing. The threat of compaction of European soils increases because of the increasing use of heavy machinery (Van den Akker et al., 2003). Compaction leads to a decreased soil porosity and a reduced water infiltration capacity (Eckelmann et al., 2006). Low soil porosity enhances denitrification and thereby the risk of increased N2O emission (Ruser et al., 1998). The low water infiltration capacity of compacted soils also increases the risk of surface runoff of N to surface waters. Erosion leads to displacement of soil particles by water or wind (Eckelmann et al., 2006). It causes irreversible soil loss over tens or hundreds of years. The Mediterranean region is particularly prone to erosion, because of long dry periods followed by heavy rainfall, especially in regions with steep slopes. Erosion results in loss of fertile soil and pollution of surface water with N and other soil compounds. Landslides, the movement of a mass of soil induced by physical processes such as excess rainfall, also cause loss of fertile soil and may pollute surface waters with N. Salinisation is the accumulation of soluble salts in soil to the extent that soil fertility is severely reduced (Eckelmann et al., 2006). Salinity poses problems in Hungary, and many Mediterranean countries and is expected to increase owing to larger extremes in droughts and rainfall periods as a possible consequence of climate change. High salt concentrations inhibit biological N transformations in soil (Curtin et al., 1999), as well as N fixing capacity by legumes (Delgado et al., 1993). High salt contents also decrease plant growth, which may lead to a lower

N use efficiency of applied N and higher N emissions in the form of gaseous losses or leaching towards the environment. Contamination of soils with organic (micro) compounds and heavy metals may hamper crop growth and decrease N use efficiency. Moreover, biological N transformations in soils are affected by contaminants (Bååth, 1989). Around 45% of soils in Europe have a low or very low organic matter content (0%–2% organic C) and 45% have a medium content (2%–6% organic C; EC, 2006). Low organic matter contents are mainly found in Southern Europe (Figure 21.2). Organic matter decline is in particular an issue in Southern Europe, but also in parts of France, the United Kingdom, Germany, the Netherlands and Sweden. Smith et al. (2005) suggested that C stocks in European cropland decline because of changes in land use or agricultural management such as tillage practices or manure use. The inputs of N may also play a role in changes in soil organic matter contents (section 21.3.2). A decrease of soil organic matter contents negatively affects the biological, chemical, and physical soil fertility, N transformations in the soil (mineralisation, and denitrification) and biodiversity. This may result in reduced crop growth and a decrease in the buffering and filtering capacities of soils for N, leading to N losses to water and atmosphere. These examples show that soil degradation hampers growth of crops and biological N transformations in soil, which may enhance N emissions to water and air. Proper soil and water management, as proposed in the European Soil Strategy, are needed to decrease N losses induced by soil degradation.

21.2╇ Fates of excess nitrogen inputs to soils Nitrogen that enters the soils is generally biologically or chemically transformed, e.g. via mineralisation, immobilisation in soil organic matter, nitrification, and denitrification (Butterbach-Bahl et al., 2011, Chapter 6 this volume). Soils protect the quality of air and water by storage, filtering, buffering and transformations of N. This soil function is threatened when the N inputs to the soils exceeds the N output by removal of the crop, tree, or vegetation. An excess input of N to the soil changes the rate of the different N transformation processes, affects soil organic matter content, decreases soil biodiversity and decreases the filtering and buffering capacity of the soil. Moreover, transformations of N often result in soil acidification, that may be reflected in a lower soil pH. Soil acidification may decrease crop growth, change N transformations in the soil, and decrease soil biodiversity. The N surplus of the soil balance of agricultural soils is an indicator for N emissions to the environment (atmosphere and hydrosphere). The soil surface balance includes all relevant N inputs and outputs from the soil. The fate of the N surplus is controlled by a combination of factors, including type and rate of N input, soil type and properties, weather conditions, crop type, tree species, and the hydrology of the soils. Differences in fertiliser and manure use (rate and application method), soil and crop type and weather conditions cause differences in the fate of the N surplus between EU countries (Figure 21.3). The N surplus on the soil surface

497

Nitrogen as a threat to European soil quality

Soil organic matter mineral soils ton SOM / ha < 70 70 - 100 100 - 130 130 - 160 > 160 Peat

Figure 21.2 Organic matter stocks in mineral soils in Europe. The location of peat soil is indicated (Lesschen et al., 2009).

balance ranges from less than 50 kg N/ha per year to more than 200 kg N/ha per year. Most of the N surplus on the soil surface balance (on average about 50%–60% of the N surplus) is lost as the harmless gas N2, followed by NH3 emission, NO3 leaching, and emissions of N2O, and NOx. Part of N surplus may (temporally) accumulate in organic matter. More details on the fate of the N surplus are given in (De Vries et al., 2011, Chapter 15 this volume). In non-agricultural soils, the N input is limited to N deposition and N fixation. Nitrogen deposition to natural

498

ecosystems in Europe is generally higher than 5 kg N/ha per year and often exceeds 40–50 kg N/ha per ha per year in areas near intensively used agricultural production regions. Leaching of N generally increases at N deposition rates higher than about 10 kg N/ha per year (Figure 21.4; Dise et al., 1998; Gundersen et al., 1998; Butterbach-Bahl et al., 2011, Chapter 6 this volume). Elevated N deposition to forests also increases nitrification and denitrification and thereby emissions of gaseous N (N2O, NOx and N2). A large number of controlling variables and complex interactions influence the net N2O

Gerard Velthof Figure 21.3 Fate of the N surplus of the soil balance in EU-27 in 2000 in kg N per ha agricultural land calculated using the model MITERRA-EUROPE (after Velthof et al., 2009).

and a pH decrease of groundwater (Appelo and Postma, 1999). Thus, nitrate removal from groundwater by pyrite oxidation increases concentrations of cations, heavy metals, and sulphate. These high concentrations may cause problems when this water will be used as drinking water. Nitrate leached from agricultural and natural soils may thus negatively affect soil quality in pyrite containing soils. Pyrite containing soils are found widespread in Europe, and have been reported in Denmark (Jørgensen et al., 2009), England (Moncaster et al., 2000), France (Molénat et al., 2002), Germany (Eulenstein et€al., 2008), the Netherlands (Hartog et al., 2005), and Spain (Otero et al., 2009).

21.3╇ Threats of nitrogen on soil quality Figure 21.4 Leaching of total N against atmospheric deposition rate measured at 121 forest plots in Europe (after De Vries et al., 2007b).

and NOx emissions, such as N deposition rate, precipitation, temperature, pH, clay content and tree species composition are important variables (Kesik et al., 2005; Bloemerts and de Vries, 2009; Butterbach-Bahl et al., 2011, Chapter 6 this volume). Denitrification is the main mechanism of removal of nitrate in deep groundwater and subsoil. The organic C content of the subsoil is low, which limits denitrification. However, the subsoil may also contain reduced inorganic compounds, including pyrite (FeS2), siderite (FeCO3), and other ferrous containing compounds. These compounds can be used by denitrifying bacteria as energy source. Pyrites often contain trace elements, including the toxic elements nickel, arsenic, cobalt, copper, lead, manganese and zinc (Huerta-Diaz and Morse, 1992; Larsen and Postma, 1997). These trace elements are released when pyrite is oxidised during denitrification. Moreover, pyrite oxidation results in an increase of the sulphate concentration

N has several effects on the soil quality of natural ecosystems (Table 21.1). Impacts on the inorganic N concentration are presented in the previous paragraph. In this paragraph the effects of N on soil acidification (Section 21.3.1), soil organic matter (Section 21.3.2), and soil biodiversity (Section 21.3.3) are presented.

21.3.1╇ Effects on soil acidification Transformations of N are an important source and sink of hydrogen ions or protons (Tabel 21.1). Acidification considerably reduces the fertility of the soil, affects microbial transformations in the soil, and may cause depression of crop growth, and yields (Marschner, 1995; Bolan et al., 2003). It may lead to (i) less availability (or deficiencies) of nutrients such as phosphorus, calcium, magnesium and molybdenum, (ii) a release of toxic compounds, including aluminium, and manganese, and (iii) hampering of the activity of soil micro-organisms involved in N transformations, such as mineralisation of organic N and biological N fixation. Low soil pH promotes the production of N2O during nitrification and denitrification (Granli and Bøckman, 1994). Soil acidification results in leaching of catÂ� ions. In the Netherlands, the observed increase of hardness (i.e.

499

Nitrogen as a threat to European soil quality Table 21.1 Effects of N on soil parameters of natural soils, their mechanisms, and the ecosystem response

Soil parameter

Mechanism

Ecosystem response

Literature

C/N ratio

Narrows at sites with high N availability, due to the incorporation of surplus N in soil organic matter.

Plant species richness ↕ Decomposition of SOM ↕ Microbial biomass ↑

(Von Oheim et al., 2008) (Friedel et al., 2008) (Dumortier et al., 2002) (Berg, 2000)

Inorganic nitrogen concentration

Nitrogen deposition is close to or exceeds ecosystem N demand. Input of inorganic N increases soil solution concentrations.

Plant productivity ↑ Leaf/needle N content ↑ Litter decomposability ↑ Plant species richness ↓ Vascular plants in wetlands ↑ Microbial N immobilisation ↓ Nitrogen leaching ↑ Soil N2O/NO emissions ↑

(De Vries et al., 2006b) (Corré et al., 2007) (Kreutzer et al., 2009) (Gundersen et al., 2006) (Stevens et al., 2006)

Acidification and soil buffering capacity

Nitrification of deposited NH3/NH4+ leads to H+ formation. In the course of the acidification process base cations are leached.

Nutrient availab. (Ca/Mg) ↓ Al/Mn toxcity if soil pH<5.5 ↑ Biodiversity ↓ Microbial activity ↓ Root growth ↓ Nitrogen leaching ↑ DOC leaching ↓ Soil N2O/NO emissions ↑ Wetland CH4-emissions

(Matzner and Murach, 1995) (Raubuch and Beese, 2005) (Bowman et al., 2008) (Gauci et al., 2005) (Evans et al., 2008)

Soil C stocks and SOC stratification

Surplus N decreases fine root biomass and, thus, reduces belowground litter production, but increases aboveground plant production and litter fall.

Total soil C stocks ↑ Forest floor C stocks ↑ Mineral soil C stocks ↕

(Högberg, 2007) (De Vries et al., 2006b) (Hyvönen et al., 2007, 2008)

Soil aggregation

N can increase litterfall and improve litter quality and, thus, positively affect soil fauna and the formation of organo-mineral soil aggregates by e.g. earthworm activities

Soil aeration ↑ Water infiltration ↑

(Lavelle et al., 2007)

the contents of Ca and Mg) in groundwater used for drinking water has been associated with the acidification of agricultural soils (Velthof et al., 1999). Hardness is considered an aesthetic water quality factor and is not known to pose a health risk to users.

Agricultural soils Ammonium based fertilisers acidify soils, because of a combination of nitrification, ammonium uptake by plants, and/ or ammonia volatilisation (Table 21.2). Nitrate based fertilisers increase pH of soils, because of a combination of nitrate uptake by plants, and/or denitrification. The most used mineral N fertilisers in Europe (i.e. >90% of total N fertiliser consumption) are calcium ammonium nitrate, ammonium nitrate, NPK fertilisers, urea, and urean (Fertilizers Europe, 2010). All these fertilisers have an acidifying effect (Harmsen et al., 1990), indicating that the use of mineral N fertiliser causes

500

acidification in a large part of the European agricultural soils. Besides the type of N fertiliser, soil acidification is dependent on crop type, soil type, weather conditions (leaching), and other N sources (Bolan et€al., 2003). Other major inputs of N are manure, N excreted during grazing, biological N fixation, and atmospheric deposition. Manures contain high ammonium concentrations, which has an acidifying effect either by uptake or nitrification. Organic N results in alkalinisation if it is mineralised (consumption of H+), but this effect is more than compensated by nitrification (production of 2H+) and the combined process leads to acidification unless NO3 is consumed (consumption of H+) (Table 21.2). The total effect of manure on acidification depends on rates of inputs and uptake of cations and anions, specifically of NH4 and NO3, by the crop and the rate of denitrification affecting nitrate leaching. Grazing causes a heterogeneous pattern, with acidification in urine patches (nitrification) and alkalinisation in dung pats (mineralisation).

Gerard Velthof Table 21.2 Generation (acidification) and consumption (alkalinisation) of protons (H+) in N transformation processes

Process

Reactiona

H+, mol/mol N

Biological N-fixation

4ROH + 2N2 + 3CH2O → 4RNH2 +3CO2 + H2O

0

Mineralisation of organic N

RNH2 + H2O + H+ → ROH + NH4+

−1

Urea hydrolysis

(NH2)2CO + 3H2O → 2NH4+ + 2OH− + CO2

−1

Nitrification

NH4 + 2O2 → NO3 + 2H + H2O

+2

Ammonium assimilation

ROH + NH4 → RNH2 + H2O + H

+1

Nitrate assimilation

ROH + NO3 + H + 2CH2O → RNH2 + 2CO2 + 2H2O

−1

Ammonia volatilisation

NH4+ → NH3 + H+

+1

Denitrification

5CH2O + 4NO3 + 4H → 2N2 + 5CO2 + 7H2O

a

+

−

+

+

+

+

−

−

+

−1

 = in the reaction mean organic C compounds. (Based on Van Breemen et al., 1983; De Vries and Breeuwsma, 1986; R Bolan et al., 2003).

Grazing in intensively managed grassland with high N inputs results in soil acidification (Oenema, 1990). Biological N fixation, in which N2 is fixed in organic N, does not affect soil acidification (Table 21.2). However, actively N2-fixing leguminous crops acidify the soil, because of an excess uptake of cations over anions (Haynes, 1983). Atmospheric N deposition causes acidification, because most of the N is present as NH4. The content of heavy metals cadmium, zinc, and copper are high in some agricultural soil, because of long-term inputs of heavy metals via fertilisers, manures, lime and organic products. The mobility of these heavy metals generally increases when soil pH decreases (Adriano, 2001). Liming of agricultural soils reduces the mobility of heavy metals and decreases risk of crop uptake and leaching to ground and surface waters (Bolan et al., 2003). However, if agricultural soils with elevated heavy metal contents are abandoned and not limed, there is a risk that these soils acidify and heavy metals are released to the environment (Boekhold, 1992).

Soils of natural ecosystems and forests Influences of elevated atmospheric N (and S) deposition on natural terrestrial ecosystems have received particular attention since the 1980s, specifically with respect to forests. The acidifying effects of N and S include (Table 21.2) (i) loss of base cat� ions from the soil causing deficiency of these nutrients for forest trees (notably Mg), (ii) release of soluble toxic Al affecting fine root growth and inhibiting the uptake of base cations (Cronan et al., 1989; Marschner, 1995) and (iii) a decrease in pH that may affect mineralisation processes and hence nutrient availability (De Vries et al., 1995, 2000; Erisman and de Vries, 2000). In acid soils, atmospheric deposition of S and N compounds leads to elevated Al concentrations in the soil solution. Figure 21.5 shows that more than 80% of the variation in Al concentration in subsoils of European forested plots with pH below 4.5 could be explained by a variation in SO4 and NO3 concentrations. In soils with a pH above 5.0, the release of Al is generally negligible, since base cations release by weathering and cation exchange buffers the incoming net acidity in those soils.

21.3.2╇ Effects on soil organic matter Soil organic matter is important for soil fertility, and affects both crop production (food and other biomass) and N and C transformations. Soil organic matter is also a sink for C, and important from a perspective of greenhouse gas emissions. Soil organic matter content and composition (quality) affects many physical, chemical, and biological properties of soils, including soil structure, water holding capacity, aeration, compaction, risk of erosion, biodiversity, (micro)biological N transformations, and the cation exchange capacity (CEC). The C/N ratio and the degradability of the soil organic matter are major factors affecting the quality of organic matter. Soils with low C/N ratio and high degradability have a high N mineralization capacity and are often considered as fertile from an agricultural point of view. However, a high N mineralization capacity is considered negative for natural soils as a high N release may lead to a decrease in plant biodiversity. One of the major controllers of soil organic matter content is land use and land management (Guo and Gifford, 2002) rather than N input. Land use and N availability are, however, closely linked and need to be jointly considered when discussing N effects on the quality of soils. In pre-industrial times, it was common practice across Europe that natural ecosystems were exploited for nutrients in order to maintain the soil fertility of the arable land, e.g. by extracting litter from forests, forest grazing, wood harvest or by converting grassland/heathland swards to arable land (Glatzel, 1991; Sieferle, 2001). These management practices have resulted in a large-scale depletion of natural soils in nutrients and soil organic matter and degradation of soil properties (Glatzel, 1991). The situation changed with the introduction of N-fixing legumes at the end of the nineteenth century, and with the increasing availability of synthetic fertilisers at the beginning to the mid of the twentieth century (Chorley, 1981). Owing to the large scale increased accessibility and availability of N for agricultural production, the pressure on natural ecosystems to serve as a reserve for nutrients ceased. However, the impacts of former land management on nutrient availability in soils may be traced for almost 2000 years (Dupouey et al., 2002). The demand for increased

501

Nitrogen as a threat to European soil quality

Figure 21.5 Concentration of total Al against total SO4+NO3 in the subsoil of monitoring plots with a pH < 4.5 (after De Vries et al., 2007b). The solid line is a regression line being equal to:€y = −95 + 0.74x (R2 = 0.86).

agricultural production was also the major driver to drain many huge wetland areas in central and northern Europe. The change to oxidative conditions has promoted mineralisation of C and N, previously stored for thousands of years.

Agricultural soils Application of N generally increases crop production and thereby also the returns of organic matter as crop residue to the soil (Paustian et al., 1997). A review by Glendining and Powlson (1995) showed that long-term use of only inorganic N fertilisers increased soil organic C content in the majority of long-term studies, but the increases tended to be small (see Figure 21.6). Results of long-term studies in Russia of Shevtsova et al. (2003) showed that high application rates of inorganic N fertilisers lead to a decrease in soil organic C content compared with unfertilised soils. This decrease was related to an increase in C mineralisation due to changes in soil quality by the long term use of the acidifying fertilisers. Results of an experiment of Olesen et al. (2000) showed that application of only mineral N fertiliser caused higher shoot biomass and lower root biomass in cereals, compared to application of animal manure. This was probably because of the higher availability of N in the soil with mineral N fertiliser, which resulted in less development of roots in comparison to the lower N availability in the manure treatment. This effect may be an explanation for the observed small effects of mineral N fertiliser on soil organic matter content in comparison to manure N. Khan et al. (2007) and Mulvaney et al. (2009) analysed results of a long-term experiment in the USA, and concluded that 40 to 50 years of inorganic N fertilisation caused a net decline in soil C content, despite massive residue C incorporation. They indicated that mineral fertiliser N promotes decomposition of crop residues and soil organic matter. However, Powlson et al. (2010) disagreed with this conclusion and suggested that the results of the long-term experiment were misinterpreted.

502

Figure 21.6 Carbon concentration in the topsoil (0–20 cm) for different mineral fertiliser treatments in the Askov long-term experiment in 2004 after 110 years of treatment differences on a sandy loam (Christensen et al., 2006). A crop rotation with winter cereals, root crops, spring cereals and grass-clover has been subjected to a range of different fertiliser treatment over the period 1894 to 2004. Various levels (0, ½, 1, 1½, 2) of N, P and K applied as mineral fertiliser were tested in the experiment. Note that the positive effect of K on soil C content was presumably because K is an important nutrient for the grass-clover.

The combination of mineral fertiliser and animal manure is most effective in maintaining soil organic matter contents at a stable level (Glendining and Powlson, 1995). A study of Sleutel et al. (2003) in Belgium showed a decrease in soil organic C contents in agricultural soils during the period 1989–1999. This was attributed to a decrease in manure application, because of environmental legislation. Moreover, farmyard manure production in Belgium has been replaced by slurry based systems. Regular application of farmyard manure has a beneficial effect on soil organic C (see Hofman and van Ruymbeke, 1980). The studies indicate that the effects of N fertilisation on soil organic C contents are diverse and related to the initial€soil organic C content, soil type, climatological conditions, soil management, development of shoot and root biomass, N source and land use. Factors, such as soil tillage, changes in climatological conditions, and changes in land use probably have a larger effect on the soil organic matter content than the use of N.

Soils of natural ecosystems and forests Natural soils have low N contents. Increased N deposition rates have significantly changed C and N cycling and soil C/N ratios during the last decades (Vitousek et al., 1997; Gundersen et al., 2006; Corré et al., 2007). A high input of N leads to eutrophication or N saturation of the natural systems, mainly indicated by low C/N ratios of organic matter, elevated N leaching and sometimes elevated NH4/base cation ratios (Table 21.1). Eutrophication also leads to reduced plant species diversity (Dise et al., 2011, Chapter 20 this volume) and may cause damage to forests due to:€(i) water shortage, since a high N input favours growth of canopy biomass (De Visser et€al., 1994), (ii) nutrient imbalances, since the increase in canopy biomass also causes an increased demand for base cation

Gerard Velthof

nutrients (Ca, Mg,€K) (Boxman and Roelofs, 1988), and (iii) an increased sensitivity to factors such as frost (Bruck, 1985) and attacks by fungi. Several studies showed that soil respiration may decrease under conditions of increased N availability (Olsson et al., 2005), which possibly can be attributed to the formation of recalcitrant organic material with narrow C/N ratios (Table 21.1). At an ecosystem scale, a large positive effect of N input to the C sequestration capacity of boreal and temperate forests in living biomass has been demonstrated (Högberg, 2007; Magnani et al., 2007; De Vries et al., 2008). On the basis of forest inventory data and measured inputs of N, De Vries et al. (2006b) estimated that at an EU scale the average impact of an additional N input on the net C sequestration in both tree wood and soil was approximately 50 kg C/kg N deposited. A literature review of (i) empirical relations between spatial patterns of C uptake and influencing environmental factors including nitrogen deposition, (ii) 15N field experiments, (iii) long-term low dose N fertiliser experiments and (iv) results from ecosystem models indicate a total C sequestration range of 5–75 kg C/kg N deposition for forest and heathlands, with a most common range of 20–40 kg C/kg N (De Vries et al., 2009). It should be noted that the N addition effect on C sequestration will saturate with time or even decline due to detrimental effects of surplus N availability in forest soils on forest health. Effects of N deposition to peatlands systems may differ from those to upland systems such as forests and heathlands. Moderate to low N deposition rates have been shown to promote the growth of sphagnum, even though nutrient imbalances, especially P deficiencies may set close limits for accelerated growth (Berendse et al., 2001; Phuyal et al., 2008). Furthermore, at higher N deposition plant community changes may occur, with vascular plants outcompeting mosses. Consequences of N deposition for peatland C storage are variable. Increased net ecosystem productivity following N deposition and stimulated growth of sphagnum and vascular plants may be offset by increased decomposition due to improved litter quality (Gunnarsson et al., 2008; Trinder et al., 2009).

21.3.3╇ Effects on soil biodiversity Nitrogen affects the biodiversity of soil organisms. Belowground especially fungi, saprotrophic decomposers as well as mycorrhizal fungi, and N fixing bacteria are reduced by fertilisation and high N availability (Streeter, 1988; Johansson et al., 2004; De Vries et al., 2006a). Bacteria and fungi are the primary decomposers of dead organic matter such as plant residues and manure, and they release mineral nutrients by mineralisation. Soil microorganisms are consumed by microbivores such as protozoa, nematodes and mites. Microbivores, in turn, are eaten by bigger predatory soil fauna. All these links in the soil food web contribute to mineralisation and nutrient cycling. Changes in bacteria and fungi will also affect the soil fauna via the bacterial and fungal channels in the soil food web. While plants and soil microorganisms directly react to the availability of mineral N, soil fauna, such as protozoa,

nematodes, enchytraeids, collembolas, insect larva or earthworms, mostly react indirectly to N through effects on plant growth and microbial dynamics (Bardgett, 2005). Since both plant litter and microorganisms are at the base of soil detritivore food webs, this is likely to lead to bottom-up effects on the whole belowground food web, on plants and on the aboveground food web (Wardle et al., 2004). While, trophic effects are likely to be very influential (De Ruiter et al., 1994), non-trophic activities of soil fauna are also involved in soil response to inputs of mineral N (Lavelle and Spain, 2001). For example, the biomass of earthworms is likely to increase if N increases plant biomass production and plant litter production. This would finally change soil aggregation, water infiltration, and organic matter dynamics (Lavelle et al., 2007). Because of the complexity of the interactions involved in soil fauna response to increasing N inputs, it is more difficult to make general predictions on this response than on the response of soil microorganisms or plants. While it is difficult to predict the effect of N inputs on particular taxa of soil fauna, N inputs could have a clearer effect on the biodiversity of soil fauna. Since N decreases plant biodiversity, it may be suggested that this leads to a lower soil biodiversity if soil taxa are eating preferentially the organic matter coming from particular plant species. However, belowground and aboveground biodiversities do not seem to be linked in such a straightforward way (Hooper et al., 2000). Some studies have shown that soil fauna is more sensitive to the quantity than to the composition (quality) of organic matter. Thus, while N inputs often decreases plant diversity, it also tends to change the quality of organic matter, which may increase soil fauna diversity (Cole et al., 2005; Van der Wal et€al., 2009).

Agricultural soils Most of the information about effects of fertilisers on soil biodiversity comes from studies where organic farming systems were compared with conventional intensive farming systems. One has to be aware that in organic farming systems, there are usually more differences with conventional farming systems than fertilisation alone, such as the avoidance of pesticides. In a comparison of 23 farms in Estonia, animal manure increased soil microbial biomass, activity and N mineralisation, and chemical fertilisers resulted in negative effects compared to organic fertilisers (Truu et al., 2008). In long-term experiments (25 years), Ge et al. (2008) found a negative effect of mineral N (300 kg N per ha per yr) on microbial diversity (genetic diversity, number of genotypes or species of bacteria). Also Jangid et al. (2008) found higher bacterial diversity with organic manure (poultry litter) and lower diversity with mineral fertilisers. In grassland soils, fungal biomass increased with reduced N fertilisation (De Vries et al., 2007a; Van Groenigen et al., 2007). Negative effects of inorganic N on fungi were mostly attributed to changes in vegetation and organic matter characteristics (Bardgett et al., 1999; Rousk and Bååth, 2007). When fungi are affected also bacteria can be affected because of competitive interactions between these two groups.

503

Nitrogen as a threat to European soil quality

In agricultural soils, it is difficult to disentangle the effects on soil fauna of the various agricultural practices. Indeed herbicides, pesticides, tillage, organic fertilisation and mineral fertilisation impact soil fauna (Jordan et al., 2004). Large species such as earthworms are often affected by tillage that directly increases their mortality (Edwards and Lofty, 1982; Chan, 2001). The depletion in organic matter of many agricultural soils (Lal, 2004) is also a very influential factor for the whole soil detritivore food web (Lavelle and Spain, 2001), which may in turn hide effects of N. However, some studies have detected effects of mineral N fertilisation on some groups of soil fauna in crop or pasture soils. For examples, nematodes (Okada and Harada, 2007) or protozoa (Forge et al., 2005) have been shown to be impacted by mineral fertilisation, probably through a trophic impact on microorganisms and plant growth. Earthworms have also been shown to have higher densities in some N fertilised plots (Jordan et al., 2004).

Soils of natural ecosystems and forests Soils of natural ecosystems and forests usually are more dominated by fungi and fungivorous soil fauna. Probably there are no fundamental differences between the effect of N input on microbial processes in agricultural soils and other ecosystems. Usually natural soils are not fertilised, are affected by atmospheric deposition of N, and are more acid. Also in natural soils long-term high N inputs have been shown to cause changes in the structure of the microbial community (Nemergut et al., 2008). Some of the microbial community shifts could provide explanation for changes in soil organic matter structure. Steep declines in basidiomycete fungi were related to higher lignin concentrations in N-amended alpine tundra plots. Multiple factors can alter soil organic matter pools following increases in N availability, including shifts in both plant productivity and species composition. The effects of increased N inputs on decomposition of organic matter, C storage and CO2 emissions are not clear yet (Craine et al., 2007). In natural ecosystems many trophic effects of N enrichment on soil fauna have been documented. For example, in a forest soil, a reduction in the diversity of nematodes and an increase in their total abundance were attributed to an increase in the relative abundance of bacterivorous and fungivorous nematodes due to the positive impact of fertilisation on microbial biomasses (Forge and Simard, 2001). Non-trophic effects of N enrichment on soil fauna have also been documented. N inputs can lead to soil acidification which negatively impacts many taxons. Xu et al. (2009) found that N deposition both decreases soil pH, the diversity of collembolas and the density of the most abundant collembolan species in a Swiss forest.

21.4╇ Management and future perspectives 21.4.1╇ N inputs to soils The surplus of N in agricultural soils can be decreased by decreasing the total N input and/or increasing the outputs of N in harvested products. Nitrogen surpluses have been declining since the eighties in member states of the European Union

504

(Figure 21.7). In Central Europe, the economic situation in the early nineties has caused a drop in the use of N fertiliser. In the intensively managed agricultural systems in the member states in North and West Europe, the decrease in the N surplus is mainly due to reform of the agricultural policy and environmental legislation, which has enhanced a more efficient N use. The decrease in the surplus indicates that N emission to the environment has decreased. However, many regions in the EU-27 still have a significant surplus on the soil N balance (see also Figure 21.3). Measures to decrease N inputs include balanced N fertilisation, in which the N inputs to the crop are tuned to the crop demand, low protein animal feeding, and/or decrease of the number of livestock. An increase in the N output can be obtained by proper management of soils, water and nutrients and control of pests. The Nitrates Directive, which is implemented in EU-27 requires measures in nitrate vulnerable zones increases, such as balanced N fertilisation is imposed. The Gothenburg Protocol of UNECE’s Convention on Longrange Transboundary Air Pollution is under revision, and may also include balanced N fertilisation as one of the measures. It is expected that because of implementation of environmental policies, the N inputs to agricultural soils with a high N surplus will further decrease in the near future. Following a series of control measures during the last two decades, emissions of NH3 and NOX and subsequent N deposition have been reduced in Europe. The concept of critical N load has been developed for natural systems to set targets for reduction of N emission (Hettelingh et al., 2001). The area where critical N loads are exceeded clearly decreased between 1980 and 2005. Nevertheless, high exceedances for critical N loads remain widespread especially in north-Â�western European areas. Further reductions in N emissions to the atmosphere are predicted (Moldanová et al., 2011, Chapter 18 this volume).

21.4.2╇ Soil acidification The expected decrease in N inputs to soils in the future will also decrease soil acidification. Model simulations of Reinds et€al. (2009) showed that most of the European forest soils could recover from their acidified state within a few decades under the current emission reduction plans. Liming is used to decrease soil acidification and optimise plant growth in carbonate free soils. The major sources of lime used in agriculture are lime stone (CaCO3) and dolomite (CaMgCO3), but also other compounds may reduce soil acidification (burned lime, rock phosphate, sugar beet pulp). Liming is widely used in agriculture and there are no indications that acidification of agricultural soils hampers crop production in Europe. The amount of limestone used in agriculture in several western and northern European countries has decreased strongly (Figure 21.8). This coincides with the decrease in N fertiliser consumption since 1990. These results suggest that acidification of agricultural soils in Europe is decreasing since the early nineties. Dissolution of lime in soils leads to dissolution of carbonates and release of CO2. Thus, liming results in emission of

Gerard Velthof Figure 21.7 Nitrogen surplus of the soil balance of agricultural soils for selected countries (source data:€OECD, 2010).

Figure 21.8 Total amount of lime (dolomite and limestone) used for grassland and cropland in 1990, 1996, and 2006 in Austria, Germany, Denmark, Finland, France, United Kingdom, Ireland, the Netherlands and Sweden. These European countries reported their lime use to UNFCCC. Data derived from the 2008 report of Greenhouse Gas emissions to UNFCCC (UNFCCC, 2010).

CO2. The IPCC has recognised this source of CO2 in the latest update of methodology of calculating greenhouse gas emissions (IPCC, 2006). Countries are obliged to report the CO2 emission from limestone and dolomite use in agriculture in the annual inventories to UNFCCC (category 5G). The use of lime in agriculture is only a minor source of CO2 and, for example, much smaller than the N2O emission from agriculture. Liming of forest soils has been widely discussed as a method of neutralising the effect of acidification. Beier and Rasmussen (1994) conclude that it is possible to reverse the acidification processes in the soil by liming, and that it is possible to increase growth and improve the nutritional balance in the trees by fertilisation and irrigation. However, the complexity of the ecosystem and the factors controlling vitality and sustainability of the ecosystem are still not fully understood. In Sweden, Andersson and Persson (1988) recommend a liming dose of 2–5 ton/ha when improvement of soil and root environment is required. Higher doses may be needed to avoid leaching of aluminium from catchments. However, liming may have negative effects on the development of tree fine roots, particularly in areas with a high N deposition (Persson and Ahlström, 1990). Results from liming experiments in Dutch forests for the period 2000–2005 (Wolf et al., 2006) show that liming leads to an increase in plant species, especially nitrophilic species, which is considered as a non-desirable side effect. Furthermore, liming increases the decomposition of organic matter, leading to thin humus layers

and a decrease in soil biota species. Wolf et al. (2006) thus considered permanent forest liming as an undesirable management option, but it can be beneficial as a once-only event in nutrient rich deciduous forests.

21.4.3╇ Soil organic matter In agricultural soils, increasing or maintaining soil organic matter content can be obtained by strategies in which the input of organic matter to the soil is higher than the output by harvested crop and by decomposition in the soil. Sources of organic matter are crop residues, manures, and organic products like compost. There are large differences in decomposition of residues of arable crops, with highest decomposition in residues of vegetables and low decomposition in cereal straw (Velthof et al., 2002). Changing crop types and including winter crops in the rotation are options to increase input of organic matter to the soil. Grasslands have a high biomass production and roots and stubbles are a large source of organic matter. Including grassland in a crop rotation will enhance soil organic matter contents in comparison to continuous cropland (Van Eekeren et al., 2008). The organic matter in farmyard manure is less degradable than that in animal slurries and the use of farmyard manure instead of slurry has a beneficial effect on soil organic matter content of soils (Leinweber and Reuter, 1992). No tillage or reduced tillage decreases decomposition of soil

505

Nitrogen as a threat to European soil quality

organic matter. All these strategies to increase or maintain soil organic matter content of agricultural soils have also been suggested as strategies to increase C sequestration in soil (Smith et al., 2008). Thinning and removal of the top soil layer are options to avoid eutrophication of natural soils. For coniferous forests the combination of sod cutting and felling and removal of trees, can lead to an improvement of soil quality. Thinning reduces litter fall and thereby the N input to soils. In heathlands, sod cutting is an efficient measure to halt invasion of grasses and increase plant species diversity (Diemont and Oude Voshaar, 1994). The complete removal of the organic top layer, including the vegetation, ensures the removal of accumulated N. A less rigorous measure is choppering. Although considerably more nutrients were removed by sod-cutting than by choppering, nutrient output by choppering was still sufficient to compensate for about 60 years of net N input (Niemeyer et al., 2007). These types of measures may lead to changes in soil biodiversity.

21.4.4╇ Soil biodiversity The main option to reduce or prevent unwanted effects of N on soil biodiversity is the reduction of N inputs. Further, there is broad agreement on general principles to promote and maintain soil biodiversity, natural soil fertility and ecosystem services, i.e.€the input of organic matter in the soil should be sufficient to meet the C and energy requirements of the soil biota, and the nutrient requirements of the crop (Swift et al., 2004; Barrios, 2007; Brussaard et al., 2007; Kibblewhite et al., 2008). It also helps when the soil remains covered by crops, which continuously feed the soil organisms via root exudates and residues. Intensive soil tillage and the use of pesticides should be kept to a minimum.

Acknowledgements This chapter was prepared with the support of the NinE Programme of the European Science Foundation, the NitroEurope IP (funded by the European Commission) and the COST Action 729.

References Adriano, D. C. (2001). Trace Elements in Terrestrial Environments: Biogeochemistry, Bioavailability, and Risks of Metals. Springer, New York. Andersson, F. and Persson, T. (1988). Liming as a measure to improve soil and tree condition in areas affected by air pollution. National Swedish Environmental Protection Board, Uppsala. Appelo, C. A. J. and Postma, D. (1999). Geochemistry, Groundwater and Pollution, 2nd edition., Balkema, Rotterdam. Bååth, E. (1989). Effects of heavy metals in soil on microbial processes and populations:€a review. Water, Air and Soil Pollution, 47, 335–379. Bardgett, R. (2005). The Biology of Soil, a Community and Ecosystem Approach, Oxford University Press. Bardgett, R. D., Mawdsley, J. L., Edwards, S. et al. (1999). Plant species and nitrogen effects on soil biological properties of temperate upland grasslands. Functional Ecology, 13, 650–660.

506

Barrios, E. (2007). Soil biota, ecosystem services and land productivity. Ecological Economics, 64, 269–285. Beier, C. and Rasmussen, L. (1994). Effects of who-ecosystem manipulations on ecosystem internal processes. Trends in Ecology€and Evolution, 9, 218–223. Berendse, F., van Breemen, N., Rydin, H. et al. (2001). Raised atmospheric CO2 levels and increased N deposition cause shifts in plant species composition and production in Sphagnum bogs. Global Change Biology, 7, 591–598. Berg, B. (2000). Litter decomposition and organic matter turnover in northern forest soils. Forest Ecology and Management, 133, 13–22. Bloemerts, M. and de Vries, W. (2009). Relationships between nitrous oxide emissions from natural ecosystems and environmental factors. Alterra Wageningen UR, Wageningen, The Netherlands. Boekhold, A. E. (1992). Field Scale Behaviour of Cadmium in Soil. Wageningen University Wageningen, The Netherlands. Bolan, N. S., Adriano, D. C. and Curtin, D. (2003). Soil acidification and liming interactions with nutrient and heavy metal transformation and bioavailability. Advances in Agronomy, 78, 215–272. Bowman, W. D., Cleveland, C. C., Halada, Ĺ., Hreško, J. and Baron, J.€S. (2008). Negative impact of nitrogen deposition on soil buffering capacity. Nature Geoscience, 1, 767–770. Boxman, A. W. and Roelofs, J. G. M. (1988). Some effect of nitrate versus ammonium nutrition on the nutrient fluxes in pinusSylvestris seedlings:€effects of mycorrhizal infection. Canadian Journal of Botany, 66, 1091–1097. Bruck, R. I. (1985). Decline of montane boreal ecosystems in the southern Appalachian mountains. Phytopathology, 75, 1338–1348. Brussaard, L., de Ruiter, P. C. and Brown, G. G. (2007). Soil biodiversity for agricultural sustainability. Agriculture, Ecosystems€and Environment, 121, 233–244. Butterbach-Bahl, K., Gundersen, P., Ambus, P. et al. (2011). Nitrogen processes in terrestrial ecosystems. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al., Cambridge University Press. Chan, K. Y. (2001). An overview of some tillage impacts on earthworm population abundance and diversity: implications for functioning in soils. Soil and Tillage Research, 57, 179–191. Chorley, G. P. H. (1981). The agricultural revolution in northern Europe, 1750–1880:€nitrogen, legumes, and crop productivity. Economic History Review, 34, 71–93. Christensen, B. T., Petersen, J. and Trentemøller, U. M. (2006). The Askov long-term experiments on animal manure and mineral fertilisers:€The lermarken site 1894–2004. Aarhus University, Faculty of Agricultural Sciences, Tjele, Denmark. Cole, L., Buckland, S. M. and Bardgett, R. D. (2005). Relating microarthropod community structure and diversity to soil fertility manipulations in temperate grassland. Soil Biology and Biochemistry. 37, 1707–1717. Corré, M., Brumme, R., Veldkamp, E. and Beese, F. O. (2007). Changes in nitrogen cycling and retention processes in soils under spruce forests along a nitrogen enrichment gradient in Germany. Global Change Biology 13, 1509–1527. Craine, J. M., Morrow, C. and Fierer, N. (2007). Microbial nitrogen limitation increases decomposition. Ecology, 88, 2105–2113. Cronan, C. S., April, R., Bartlett, R. J. et al. (1989). Aluminum toxicity in forests exposed to acidic deposition:€the ALBIOS Results. Water, Air and Soil Pollution, 48, 181–192. Curtin, D., Steppuhn, H., Campbell, C. A. and Biederbeck, V. O. (1999). Carbon and nitrogen mineralization in soil treated with

Gerard Velthof chloride and phosphate salts. Canadian Journal of Soil Science, 79, 427–429. De Ruiter, P. C., Neutel, A.-M. and Moore, J. C. (1994). Modelling food webs and nutrient cycling in agro-ecosystems. Trends in Ecology€and Evolution, 9, 378–383. De Visser, P. H. B., Beier, C., Rasmussen, L. et al. (1994). Biological response of 5 Forest ecosystems in the Exman project to input changes of water, nutrients and atmospheric loads. Forest Ecology and Management, 68, 15–29. De Vries, W. and Breeuwsma, A. (1986). Relative importance of natural and anthropogenic proton sources in soils in the Netherlands. Water, Air and Soil Pollution, 28, 173–184. De Vries, W., van Grinsven, J. J. M., van Breemen, N., Leeters, E.€E.€J. M. and Jansen, P. C. (1995). Impacts of acid deposition on concentrations and fluxes of solutes in acid sandy forest soils in the Netherlands. Geoderma, 67, 17–43. De Vries, W., Klap, J. M. and Erisman, J. W. (2000). Effects of environmental stress on forest crown condition in Europe. Part€I:€Hypotheses and approach to the study. Water, Air and Soil Pollution, 119, 317–333. De Vries, F. T., Hoffland, E., van Eekeren, N., Brussaard, L. and Bloem,€J. (2006a). Fungal/bacterial ratios in grasslands with contrasting nitrogen management. Soil Biology and Biochemistry, 38, 2092–2103. De Vries, W., Reinds, G. J., Gundersen, P. and Sterba, H. (2006b). The impact of nitrogen deposition on carbon sequestration in European forests and forest soils. Global Change Biology, 12, 1151–1173. De Vries, F. T., Bloem, J., van Eekeren, N., Brussaard, L. and Hoffland,€E. (2007a). Fungal biomass in pastures increases with age and reduced N input. Soil Biology and Biochemistry, 39, 1620–1630. De Vries, W., van der Salm, C., Reinds, G. J. and Erisman, J. W. (2007b). Element fluxes through European forest ecosystems and their relationships with stand and site characteristics. Environmental Pollution, 148, 501–513. De Vries, W., Solberg, S., Dobbertin, M. et al. (2008). Ecologically implausible carbon response? Nature, 451, E1–E3. De Vries, W., Solberg, S., Dobbertin, M. et al. (2009). The impact of nitrogen deposition on carbon sequestration by European forests and heathlands. Forest Ecology and Management, 258, 1814–1823. De Vries, W., Leip, A., Reinds, G. J. et al. (2011). Geographic variation in terrestrial nitrogen budgets across Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al., Cambridge University Press. Delgado, M. J., Garrido, J. M., Ligero, F. and Lluch, C. (1993). Nitrogen fixation and carbon metabolism by nodules and bacteroids of pea plants under sodium chloride. Physiologia Plantarum, 89, 824–829. Diemont, W. H. and Oude Voshaar, J. H. (1994). Effects of climate and management on the productivity of Dutch heathlands. Journal of Applied Ecology, 31, 709–716. Dise, N. B., Matzner, E. and Forsius, M. (1998). Evaluation of organic horizon C:N ratio as an indicator of nitrate leaching in conifer forests across Europe. Environmental Pollution, 102, 453–456. Dise, N. B., Ashmore, M., Belyazid, S. et al. (2011). Nitrogen as a threat to European terrestrial biodiversity. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al., Cambridge University Press. Dumortier, M., Butaye, J., Jacquemyn, H. et al. (2002). Predicting vascular plant species richness of fragmented forests in agricultural landscapes in central Belgium. Forest Ecology and Management, 158, 85–102.

Dupouey, J. L., Dambrine, E., Laffite, J. D. and Moares, C. (2002). Irreversible impact of past land use on forest soils and biodiversity. Ecology, 83, 2978–2984. EC (2006). Communication from the Commission to the council, the European parliament, the European economic and social committee and the Committee of the Regions. Thematic Strategy for Soil Protection. Brussels. 22.9.2006, COM(2006)231 final and SEC(2006)1165. Eckelmann, W., Baritz, R., Bialousz, S. et al. (2006). Common Criteria for Risk Area Identification according to Soil Threats, European Soil Bureau Research Report No.20, EUR 22185 EN. Office for Official Publications of the European Communities, Luxembourg. Edwards, C. A. and Lofty, J. R. (1982). The effect of direct drilling and minimal cultivation on earthworm populations. Journal of Applied Ecology, 19, 723–734. Erisman, J. W. and de Vries, W. (2000). Nitrogen deposition and effects on European forests. Environmental Reviews, 8, 65–93. Eulenstein, F., Werner, A., Willms, M. et al. (2008). Model based scenario studies to optimize the regional nitrogen balance and reduce leaching of nitrate and sulfate of an agriculturally used water catchment. Nutrient Cycling in Agroecosystems, 82, 33–49. Evans, C. D., Goodale, C. L., Caporn, S. J. M. et al. (2008). Does elevated nitrogen deposition or ecosystem recovery from acidification drive increased dissolved organic carbon loss from upland soil? A review of evidence from field nitrogen addition experiments. Biogeochemistry, 91, 13–35. Fertilizers Europe (2010). www.efma.org Forge, T. A. and Simard, S. W. (2001). Structure of nematode communities in forest soils of southern British Columbia:€relationships to nitrogen mineralization and effects of clearcut harvesting and fertilization. Biology and Fertility of Soils, 34, 170–178. Forge, T. A., Bittman, S. and Kowalenko, C. G. (2005). Responses of grassland soil nematodes and protozoa to multi-year and single-year applications of dairy manure slurry and fertiliser. Soil Biology€and Biochemistry, 37, 1751–1762. Friedel, J. K., Ehrmann, O., Pfeffer, M. et al. (2008). Soil microbial biomass and activity:€the effect of site characteristics in humid temperate forest ecosystems. Journal of Plant Nutrition and Soil Science, 169, 175–184. Gauci, V., Dise, N. and Blake, S. (2005). Long-term suppression of wetland methane flux following a pulse of simulated acid rain. Geophysical Research Letters, 32, L12804. Ge, Y., Zhang, J.-B., Zhang, L.-M., Yang, M. and He, J.-Z. (2008). Longterm fertilization regimes affect bacterial community structure and diversity of an agricultural soil in northern China. Journal of Soils and Sediments, 8, 43–50. Glatzel, G. (1991). The impact of historic land use and modern forestry on nutrient relations of Central European forest ecosystems. Nutrient Cycling in Agroecosystems, 27, 1–8. Glendining, M. J. and Powlson, D. S. (1995). The effects of long continued applications of inorganic nitrogen fertiliser on soil organic nitrogen€– a review. In: Soil Management, Experimental Basis for Sustainability and Environmental Quality, ed. R. Lal, and B.€A. Stewart. CRC Press, Boca Raton, FL, pp. 385–446. Granli, T. and Bøckman, O. C. (1994). Nitrous oxide from agriculture. Norwegian Journal of Agricultural Sciences Supplement, 12, 7–128. Gundersen, P., Callesen, I. and de Vries, W. (1998). Nitrate leaching in forest ecosystems is related to forest floor C/N ratios. Environmental Pollution, 102, 403–407.

507

Nitrogen as a threat to European soil quality Gundersen, P., Schmidt, I. K. and Raulund-Rasmussen, K. (2006). Leaching of nitrate from temperate forests: effects of air pollution and forest management. Environmental Reviews, 14, 1–57. Gunnarsson, U., Bronge, L. B., Rydin, H. and Ohlson, M. (2008). Near-zero recent carbon accumulation in a bog with high nitrogen deposition in SW Sweden. Global Change Biology, 14, 2152–2165. Guo, L. B. and Gifford, R. M. (2002). Soil carbon stocks and land use change:€a meta analysis. Global Change Biology, 8, 345–360. Harmsen, K., Loman, H. and Neeteson, J. J. (1990). A derivation of the Pierre–Sluijsmans equation used in the Netherlands to estimate the acidifying effect of fertilisers applied to agricultural soil. Fertilizer Research, 26, 319–325. Hartog, N., Griffioen, J. and van Bergen, P. F. (2005). Depositional and paleohydrogeological controls on the distribution of organic matter and other reactive reductants in aquifer sediments. Chemical Geology, 216, 113–131. Haynes, R. (1983). Soil acidification induced by leguminous crops. Grass and Forage Science, 38, 1–11. Hettelingh, J.-P., Posch, M. and de Smet, P. A. M. (2001). Multi-effect critical loads used in multi-pollutant reduction agreements in Europe. Water, Air and Soil Pollution, 130, 1133–1138. Hofman, G. and van Ruymbeke, M. (1980). Evolution of soil humus content and calculation of global humification coeffcients on different organic matter treatments during a 12 year experiment with Belgian silt soils. Soil Science, 129, 92–94. Högberg, P. (2007). Nitrogen impacts on forest carbon. Nature, 447, 781–782. Hooper, D. U., Bignell, D. E., Brown, V. K. et al. (2000). Interactions between aboveground and belowground biodiversity in terrestrial ecosystems:€patterns, mechanisms, and feedbacks. BioScience, 50, 1049–1061. Huerta-Diaz, M. A. and Morse, J. W. (1992). Pyritization of trace metals in anoxic marine sediments. Geochimica et Cosmochimica Acta, 56, 2681–2702. Hyvönen, R., Ågren, G. I., Linder, S. et al. (2007). The likely impact of elevated [CO2], nitrogen deposition, increased temperature, and management on carbon sequestration in temperate and boreal forest ecosystems: a literature review. New Phytologist, 174, 463–480. Hyvönen, R., Persson, T., Andersson, S. et al. (2008). Impact of longterm nitrogen addition on carbon stocks in trees and soils in northern Europe. Biogeochemistry, 89, 121–137. IPCC (2006). 2006 IPCC Guidelines for National Greenhouse Gas Inventories, prepared by the National Greenhouse Gas Inventories Programme. IGES, Japan. Jangid, K., Williams, M. A., Franzluebbers, A. J. et al. (2008). Relative impacts of land-use, management intensity and fertilization upon soil microbial community structure in agricultural systems. Soil Biology and Biochemistry, 40, 2843–2853. Johansson, J. F., Paul, L. R. and Finlay, R. D. (2004). Microbial interactions in the mycorrhizosphere and their significance for sustainable agriculture. FEMS Microbiology and Ecology, 48, 1–13. Jordan, D., Miles, R. J., Hubbard, V. C. and Lorenz, T. (2004). Effect of management practices and cropping systems on earthworm abundance and microbial activity in Sanborn Field:€a 115-year-old agricultural field. Pedobiologia, 48, 99–110. Jørgensen, C. J., Jacobsen, O. S., Elberling, B. and Aamand, J. (2009). Microbial oxidation of pyrite coupled to nitrate reduction in anoxic groundwater sediment. Environmental Science and Technology, 43, 4851–4857.

508

Karlen, D. L., Mausbach, M. J., Doran, J. W. et al. (1997). Soil quality:€a concept, definition, and framework for evaluation. Soil Science Society of America Journal, 61, 4–10. Kesik, M., Ambus, P., Baritz, R. et al. (2005). Inventories of N2O and NO emissions from European forest soils. Biogeosciences, 2, 353–375. Khan, S. A., Mulvaney, R. L., Ellsworth, T. R. and Boast, C. W. (2007). The myth of nitrogen fertilization for soil carbon sequestration. Journal of Environmental Quality, 36, 1821–1832. Kibblewhite, M. G., Ritz, K. and Swift, M. J. (2008). Soil health in agricultural systems. Philosophical Transactions of the Royal Society of London, Series B, Biological Sciences, 363, 685–701. Kreutzer, K., Butterbach-Bahl, K., Rennenberg, H. and Papen, H. (2009). The complete nitrogen cycle of an N-saturated spruce forest ecosystem. Plant Biology, 11, 643–649. Lal, R. (2004). Soil carbon sequestration impacts on global climate change and food security. Science, 304, 1623–1627. Larsen, F. and Postma, D. (1997). Nickel mobilization in a groundwater well field:€release by pyrite oxidation and desorption from manganese oxides. Environmental Science and Technology, 31, 2589–2595. Lavelle, P. and Spain, A. (2001). Soil Ecology. Kluwer Academic Publishers, Dordrecht. Lavelle, P., Barot, S., Blouin, M. et al. (2007). Earthworms as key actors in self-organized soil systems. In: Ecosystem Engineers:€Plants to Protists, ed. K. Cuddington, J. E. Byers, W. G. Wilson and A. Hastings, Academic Press, London, pp. 405. Leinweber, P. and Reuter, G. (1992). The influence of different fertilization practices on concentrations of organic carbon and total nitrogen in particle-size fractions during 34 years of a soil formation experiment in loamy marl. Biology and Fertility of Soils, 13, 119–124. Lesschen, J. P., Eickhout, B., Rienks, W., Prins, A. G. and Staritsky, I. (2009). Greenhouse gas emissions for the EU in four future scenarios, WAB Report 500102 026. PBL, Bilthoven, The Netherlands. Magnani, F., Mencuccini, M., Borghetti, M. et al. (2007). The human footprint in the carbon cycle of temperate and boreal forests. Nature, 447, 848–850. Marschner, H. (1995). Mineral Nutrition of Higher Plants, 2nd edition, Academic Press, London. Matzner, E. and Murach, D. (1995). Soil changes induced by air pollutant deposition and their implication for forests in central Europe. Water, Air and Soil Pollution, 85, 63–76. Moldanová, J., Grennfelt, P., Jonsson, Å. et al. (2011). Nitrogen as a threat to European air quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al., Cambridge University Press. Molénat, J., Durand, P., Gascuel-Odoux, C., Davy, P. and Gruau, G. (2002). Mechanisms of nitrate transfer from soil to stream in an agricultural watershed of French Brittany. Water, Air and Soil Pollution, 133, 161–183. Moncaster, S. J., Bottrell, S. H., Tellam, J. H., Lloyd, J. W. and Konhauser, K. O. (2000). Migration and attenuation of agrochemical pollutants:€insights from isotopic analysis of groundwater sulphate. Journal of Contaminant Hydrology, 43, 147–163. Mulvaney, R. L., Khan, S. A. and Ellsworth, T. R. (2009). Synthetic nitrogen fertilizers deplete soil nitrogen:€a global dilema for sustainable cereal production. Journal of Environmental Quality, 38, 2295–2314.

Gerard Velthof Nemergut, D. R., Townsend, A. R., Sattin, S. R. et al. (2008). The effects of chronic nitrogen fertilization on alpine tundra soil microbial communities:€Implications for carbon and nitrogen cycling. Environmental Microbiology, 10, 3093–3105. Niemeyer, M., Niemeyer, T., Fottner, S., Härdtle, W. and Mohamed, A. (2007). Impact of sod-cutting and choppering on nutrient budgets of dry heathlands. Biological Conservation, 134, 344–353. OECD (2010). http://stats.oecd.org/ Oenema, O. (1990). Calculated rates of soil acidification of intensively used grassland in the Netherlands. Fertilizer Research, 26, 217–228. Okada, H. and Harada, H. (2007). Effects of tillage and fertiliser on nematode communities in a Japanese soybean field. Applied Soil Ecology, 35, 582–598. Olesen, J. E., Askegaard, M. and Rasmussen, I. A. (2000). Design of an organic farming crop rotation experiment. Acta Agriculturae Scandinavica, 50, 13–21. Olsson, P., Linder, S., Giesler, R. and Hogberg, P. (2005). Fertilization of boreal forest reduces both autotrophic and heterotrophic soil respiration. Global Change Biology, 11, 1745–1753. Otero, N., Torrento, C., Soler, A., Mencio, A. and Mas-Pla, J. (2009). Monitoring groundwater nitrate attenuation in a regional system coupling hydrogeology with multi-isotopic methods:€The case of Plana de Vic (Osona, Spain). Agriculture, Ecosystems and Environment, 133, 103–113. Paustian, K., Collins, H. P. and Paul, E. A. (1997). Management controls on soil carbon. In: Soil Organic Matter in Temperate Agroecosystems, ed. E. A. Paul, K. Paustian, E. T. Elliot and C. V. Cole, CRC Press, Boca Raton, fl, pp. 15–49. Persson, H. and Ahlström, K. (1990). The effects of forest liming on fertilization on fine-root growth. Water, Air and Soil Pollution, 54, 365–375. Phuyal, M., Artz, R. R. E., Sheppard, L., Leith, I. D. and Johnson, D. (2008). Long-term nitrogen deposition increases phosphorus limitation of bryophytes in an ombrotrophic bog. Plant Ecology, 196, 111–121. Powlson, D. S., Jenkinson, D. S., Johnston, A. E. et al. (2010). Comments on ‘synthetic nitrogen fertilizers deplete soil nitrogen:€a global dilemma for sustainable cereal production,’ by R. L. Mulvaney, S. A. Khan, and T. R. Ellsworth. Journal of Environmental Quality 39, 1–4. Raubuch, M. and Beese, F. (2005). Influence of soil acidity on depth gradients of microbial biomass in beech forest soils. European Journal of Forest Research, 124, 87–93. Reinds, G. J., Posch, M. and Leemans, R. (2009). Modelling recovery from soil acidification in European forests under climate change. Science of the Total Environment, 407, 5663–5673. Rousk, J. and Bååth, E. (2007). Fungal and bacterial growth in soil with plant materials of different C/N ratios. FEMS Microbiology and Ecology, 62, 258–267. Ruser, R., Schilling, R., Steindl, H., Flessa, H. and Beese, F. (1998). Soil compaction and fertilization effects on nitrous oxide and methane fluxes in potato fields. Soil Science Society of America Journal, 62, 1587–1595. Schjønning, P., Elmholt, S. and Christensen, B. T. (2004). Soil quality management: concepts and terms. In: Managing Soil Quality:€Challenges in Modern Agriculture, ed. P. Schjønning, S. Elmholt and B. T. Christensen. UK, CAB International, Wallingford, pp. 1–12. Shevtsova, L., Romanenkov, V., Sirotenko, O. et al. (2003). Effect of natural and agricultural factors on long-term soil organic matter dynamics in arable soddy-podzolic soils: modeling and observation. Geoderma, 116, 165–189.

Sieferle, R. P. (2001). The Subterranean Forest: Energy Systems and the Industrial Revolution, The White Horse Press, Cambridge, UK. Sleutel, S., de Neve, S. and Hofman, G. (2003). Estimates of carbon stock changes in Belgian cropland. Soil Use and Management, 19, 166–171. Smith, P., Andrén, O., Karlsson, T. et al. (2005). Carbon sequestration potential in European croplands has been overestimated. Global Change Biology, 11, 2153–2163. Smith, P., Martino, D., Cai, Z. et al. (2008). Greenhouse gas mitigation in agriculture. Philosophical Transactions of the Royal Society of London, Series B, Biological Sciences, 363, 789–813. Stevens, C. J., Dise, N. B., Gowing, D. J. G. and Mountford, G. O. (2006). Loss of forb diversity in relation to nitrogen deposition in the UK:€regional trends and potential controls. Global Change Biology, 12, 1823–1833. Streeter, J. (1988). Inhibition of legume nodule formation and N2 fixation by nitrate. CRC Critical Reviews in Plant Sciences, 7. Swift, M. J., Izac, A.-M. N. and van Noordwijk, M. (2004). Biodiversity and ecosystem services in agricultural landscapes: are we asking the right questions? Agriculture, Ecosystems and Environment, 104, 113–134. Tóth, G., Montanarella, L. and Rusco, E. (2008). Threats to Soil Quality in Europe. European Commission, Joint Research Centre, Institute for Environment and Sustainability. Trinder, C. J., Johnson, D. and Artz, R. R. E. (2009). Litter type, but not plant cover, regulates initial litter decomposition and fungal community structure in a recolonising cutover peatland. Soil Biology and Biochemistry, 41, 651–655. Truu, M., Truu, J. and Ivask, M. (2008). Soil microbiological and biochemical properties for assessing the effect of agricultural management practices in Estonian cultivated soils. European Journal of Soil Biology, 44, 231–237. UNFCCC (2010). http://unfccc.int/ Van Breemen, N., Mulder, J. and Driscoll, C. T. (1983). Acidification and alkalinization of soils. Plant and Soil, 75, 283–308. Van den Akker, J. J. H., Arvidsson, J. and Horn, R. (2003). Introduction to the special issue on experiences with the impact and prevention of subsoil compaction in the European Union. Soil and Tillage Research, 73, 1–8. Van der Wal, A., Geerts, R. H. E. M., Korevaar, H. et al. (2009). Dissimilar response of plant and soil biota communities to longterm nutrient addition in grasslands. Biology and Fertility of Soils, 45, 663–667. Van Eekeren, N., Bommelé, L., Bloem, J. et al. (2008). Soil biological quality after 36 years of ley-arable cropping, permanent grassland and permanent arable cropping. Applied Soil Ecology, 40, 432–446. Van Groenigen, K. J., Six, J., Harris, D. and van Kessel, C. (2007). Elevated CO2 does not favor a fungal decomposition pathway. Soil Biology and Biochemistry, 39, 2168–2172. Velthof, G. L., Beek, C. G. E. M. and van Erp, P. J. (1999). Leaching of calcium and magnesium (hardness) from arable land and maize land on non-calcareous sandy soils. Meststoffen, 1999, 60–66. Velthof, G. L., Kuikman, P. J. and Oenema, O. (2002). Nitrous oxide emission from soils amended with crop residues. Nutrient Cycling in Agroecosystems, 62, 249–261. Velthof, G. L., Oudendag, D. A., Witzke, H. P. et al. (2009). Integrated assessment of nitrogen emission losses from agriculture in EU-27 using MITERRA-EUROPE. Journal of Environmental Quality, 38, 1–16.

509

Nitrogen as a threat to European soil quality Vitousek, J. P. W., Aber, J. D., Howarth, R. W. et al. (1997). Human alteration of the global nitrogen cycle:€sources and consequences. Ecological Applications, 7, 737–750. Von Oheim, G., Hardtle, W., Naumann, P. S. et al. (2008). Long-term effects of historical heathland farming on soil properties of forest ecosystems. Forest Ecology and Management, 255, 1984–1993. Wardle, D. A., Bardgett, R. D., Klironomos, J. N. et al. (2004). Ecological linkages between aboveground and belowground biota. Science, 304, 1629–1633.

510

Wolf, R. J. A. M., Dimmers, W. J., Hommel, P. W. F. M. et al. (2006). Bekalking en toevoegen van nutriënten Evaluatie van de effecten op het bosecosysteem€– een veldonderzoek naar vegetatie, humus en bodemfauna. Alterra, Wageningen, The Netherlands. Xu, G. L., Schleppi, P., Li, M. H. and Fu., S. L. (2009). Negative responses of Collembola in a forest soil (Aptal, Switzerland) under experimentally increased N deposition. Environmental Pollution, 157, 2030–2036.

Part

V

European nitrogen policies and future challenges

Chapter

22

Costs and benefits of nitrogen in the environment Lead authors: Corjan Brink and Hans van Grinsven Contributing authors: Brian H. Jacobsen, Ari Rabl, Ing-Marie Gren, Mike Holland, Zbigniew Klimont, Kevin Hicks, Roy Brouwer, Roald Dickens, Jaap Willems, Mette Termansen, Gerard Velthof, Rob Alkemade, Mark van Oorschot and Jim Webb

Executive summary Nature of the problem • Single issue policies have been an effective means of reducing reactive nitrogen (Nr) emissions in the EU, but to make further reductions more-integrated approaches are required.

Approaches • This chapter shows how cost–benefit analysis (CBA) can provide guidance for the setting of new policy priorities for the abatement of the European Nr emissions from an integrated perspective. • Data on costs and benefits of Nr-abatement, including four national and regional case studies, are reviewed and made comparable by expression in euro per kg of added Nr (agriculture) or euro per kg of reduced Nr emission (unit cost approach). • Social cost estimates are based on Willingness to Pay (WTP) for human life or health, for ecosystem services and greenhouse gas (GHG) emission reduction.

Key findings • The total annual Nr-related damage in EU27 ranges between 70 and 320 billion Euro, equivalent to 150–750 euro/capita, of which about 75% is related to health damage and air pollution. This damage cost constitutes 1%–4% of the average European income. • Inferred social costs of health impacts from NOx are highest (10–30 euro per kg of pollutant-Nr emission). Health costs from secondary ammonium particles (2–20 euro/kg N), from GHG balance effects of N2O (5–15 euro/kg N), from ecosystem impacts via N-runoff (5–20 euro/kg N) and by N-deposition (2–10 euro/kg N) are intermediate. Costs of health impacts from NO3 in drinking water (0–4 euro/kg N) and by N2O via stratospheric ozone depletion (1–3 euro/kg N) are estimated to be low. • The first year social benefit of Nr for the farmer ranges between 1 and 3 euro per kg added N-fertilizer equivalent. Internalizing the Â�environmental costs of N-fertilization would lower the optimal N-rate for arable production in North-West Europe by at least 50 kg/ha.

Uncertainties • Major uncertainties in our approach are dose-response relationships and poor comparability of WTP studies. Also it is often not simple to identify the Nr-share in adverse impacts and in abatement measures.

Recommendations • The CBA results presented provide support for the present focus of EU and UNECE N-policies on air pollution and human health, and on reducing ammonia emissions from agriculture; the social benefits of abatement tend to exceed the additional costs. • Although options are attractive that offer simultaneous reductions of all N pollutants, the CBA points to the need to prioritize NOx and NH3 abatement over the abatement of N2O emissions. Social cost of potential increases in emissions of N2O and nitrate, when enforcing low ammonia emission techniques, are overwhelmed by the social benefits of decreased NH3-emission.

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

513

Costs and benefits of nitrogen in the environment

22.1╇ Introduction Human welfare and nitrogen In some parts of the world a considerable amount of anthropogenic reactive nitrogen (Nr) is lost to the air, water and land causing a range of environmental and human health problems while, simultaneously, in other parts of the world food production is Nr-deficient. Therefore, Galloway et╯al. (2008) conclude that ‘optimizing the need for a key human resource while minimizing its negative consequences requires an integrated interdisciplinary approach and the development of strategies to decrease N-containing waste’. In fact the question is whether the N-cycle can be changed in such a way that a human welfare improvement is achieved, i.e. the benefits of this change exceed (or at least are in balance with) the associated costs. Given that many of the environmental effects involved, have costs and benefits for the population, government intervention might be required. Decisions by policy makers can therefore be supported by an integrated assessment of all social costs and benefits of changing N-management, showing the trade-offs that are at stake.

Human welfare implications of changing nitrogen management Theoretically, the effect of a change in N-management on human welfare can be determined by comparing the social costs and benefits. The social costs are the resources a society has to give up when changing N management, e.g. the cost of an investment in sewage treatment or the income lost in agriculture if fertilizer use is limited. The social benefits are all effects that positively contribute to human welfare, e.g. the protection of threatened species or avoiding negative health impacts of NOx emissions. A change in N-management will only lead to an improvement in human welfare if the sum of the social benefits exceeds the sum of the social costs. In theory, the social optimal level of Nr-use is found where marginal social cost equals marginal social benefits. In view of the complexity of the N cycle, in practice it is, however, a difficult task to determine all human welfare effects of chang� ing N management, as will become clear in this chapter.

Policy context of social cost–benefit analyses related to nitrogen management Increasingly, decision makers demand that decisions are based on an assessment of the social costs and benefits of Nr-reduction options. Two policy stages can be distinguished, requiring complementary assessment procedures. (1) At the initial stage there is a relatively high level of environmental pollution, for which various low cost measures are available. If the pollution causes unacceptable social problems, the low cost of measures implies that further control will be beneficial for society. Therefore, in this stage policy makers are mainly interested in the identification of options that reduce pollution at the lowest costs and cost-effectiveness analysis (CEA) is a suitable tool to support policy making; (2) At a later stage the environmental pollution has been reduced considerably through the implementation of Â�

514

low-cost measures and marginal benefits of further reduction have decreased. At this stage it becomes important (and often also more difficult) for policymakers to know more precisely the marginal cost and marginal benefits of Nr-reduction in order to determine economically efficient use levels and cost–benefit analysis is helpful to support policymaking. In this chapter, Section 22.2 describes the theoretical background and main tools for integrated assessment of the social costs and benefits of Nr-reduction strategies. Section 22.3 gives examples and results of the economic valuation of the effects of Nr on (i) agriculture, (ii) ecosystems and (iii) human health. Section 22.4 addresses the cost of Nr abatement. Section 22.5 summarizes four relevant case studies of integrated assessment of costs and benefits. Section 22.6 provides some tentative CBA results from EU27 countries and the agricultural sector and discusses the limitations and potential of CBA in developing Nr-reduction policies.

22.2╇ Theoretical background from an economic perspective 22.2.1╇ General framework An economic impact assessment of policy measures to prevent and mitigate the harmful effects resulting from Nr-use requires assessment of all the welfare effects of these measures. Various types of policy measures with different impacts on social welfare are available. Figure 22.1 shows how various response options can influence the environmental impact of economic activities (agricultural production, energy use, and industrial processes) that serve to meet the demand for various goods and services. Specific policies might either change the driving forces behind the Nr-related environmental impact, i.e. the demand for goods and services, affect the economic activities (leading to reduction or prevention of emissions to the environment), work towards restoring the environmental quality, or deal with the final impacts, e.g. medical treatment or ecosystem restoration. These options have an impact on welfare in several ways. In many cases it is, however, far from a simple matter to quantify and value these welfare effects in a directly comparable way.

22.2.2╇ Methods for economic impact assessment Important economic impact assessment methods include costeffectiveness analysis (CEA) and cost–benefit analysis (CBA). The purpose of CEA is to find out how predetermined targets, e.g. threshold values for nutrients in a catchment, can be achieved in the least cost way (Brouwer and De Blois, 2008). In practice, CEA is carried out at varying levels of complexity, scale, comprehensiveness and completeness. The same applies to CBA, which can be carried out from the perspective of an investment decision of an individual company, accounting for the private costs and revenues only, but also from the perspective of society as a whole taking into account all effects that influence social welfare. The latter is also referred to as ‘extended’

Corjan Brink and Hans van Grinsven Figure 22.1 Schematic representation of general framework.

CBA (Brouwer and van Ek, 2004). We refer to CBA here primarily from a social perspective, in which case CBA compares the costs and benefits of different policy options, preferably in monetary terms. When costs and benefits of policy options occur at different points in time, they are made comparable in time through a weighting procedure called discounting (see the later section on Discounting in Section 22.2.5). The result of this analysis is a net present value, where the discounted (present) values of the costs are subtracted from the discounted (present) value of the benefits. Dividing the present value of the benefits by the present value of the costs provides a B/C ratio, which, if larger than one, indicates that the policy option is beneficial. Monetizing the impacts of public environmental policy is not always possible, however, non-monetized impacts, if considered relevant, can be included qualitatively during the discussion of the CBA results. Different approaches exist on how non-monetized impacts are included in a CBA:€they can be listed as ‘pro memoria’ items in the balance sheet, expressed in qualitative or quantitative terms or become part of a wider multi-criteria analysis (Brouwer and van Ek, 2004).

22.2.3╇ Valuing the benefits The monetary value of a positive welfare effect, e.g. an improvement in water quality, can be measured as the amount of money society is willing to give up to secure the improvement (De Zeeuw et╯al., 2008). This is called the willingness to pay (WTP). For goods and services that are traded in markets, this WTP can be derived from demand for the goods and services involved at different market prices. However, for many environmental goods and services there exist no markets and therefore no market prices are available which reflect their economic value. Therefore, alternative ways are required to estimate the monetary value of changes in environmental quality. Although it seems that the acceptability of these methods has increased due to substantial improvements in the state-of-the-art of the methodological approach, there remains discussion about whether or not we should always try to put a monetary value on all environmental goods and services (De Zeeuw et╯al., 2008). It

is perhaps also important to point out that CBA is a tool to support decision-making. Policy makers are not bound to follow the outcome of the CBA, they may apply their own weightings to the wide variety of welfare implications of public environmental policy or factor in other issues that lead them to a different conclusion than that generated by the CBA.

22.2.4╇ Calculating the cost Costs are defined as the value of the negative welfare implications of an activity or policy, resulting in the sacrifice of alternatively employable scarce resources (Markandya et╯al., 2001). Relevant costs are not only the costs that a typical operator or farmer faces when implementing an abatement measure, but include all resources a society has to sacrifice as a consequence of the N management practices. These costs include investment and operating cost (e.g. in the case of an investment in sewage treatment), but also opportunity cost, for example by loss of productivity of agricultural land when it is set aside to reduce Nr leaching, or welfare loss because of limitations to recreational activities in a certain area as a result of a measure. Like the assessment of benefits, the assessment of costs faces several challenges. • The potential for the actual response to policies tends to differ from the predicted response. For example, industries will always look for the cheapest solution to meet legislated requirements and experience shows that these may not always match the technical solutions typically included in cost assessments (see e.g. Oosterhuis, 2007). • Actually incurred costs of measures tend to be lower than predicted, owing to technological improvements, added competition and the fact that stakeholders who perceive themselves most likely to be disadvantaged by new policy are those most likely to respond to surveys on cost data (see e.g. Oosterhuis, 2006). • Cost estimates tend to focus on single pollutants or even single environmental issues (e.g. eutrophication), while substantial cost savings may be achieved when assessing the co-benefits.

515

Costs and benefits of nitrogen in the environment

Notwithstanding the above, the expenditures (and also earnings or cost savings) directly associated with the implementation of the policy option by the relevant actors (e.g. the farmers or the government) can generally be estimated. From the perspective of social welfare, costs (and savings) for all parties within a society have to be considered (Brouwer et╯al., 2008). These include the indirect costs, i.e. costs for other actors than those implementing the measure (e.g. a reduction in pig production in a given area would imply lower economic activity among people providing services and products to these farms and for those who buy their products from these farms). In practice, these costs are more difficult to estimate and require broader macro-economic models. The relevance of including the indirect costs depends on their expected size and the role they are expected to play in the decision-making procedure (Zhang and Folmer, 1998).

22.2.5╇ Use of CBA in policymaking Potential and conditions CBA has an important role in the development, design and evaluation of policies that influence the N cycle. Its central role in informing policy decisions is mainly due to its ability to develop consistent optimal policies, provide accountability to decision makers and answer questions regarding the potential alternatives. The first key benefit of CBA is that it fosters the development of a robust evaluation methodology that can be applied across a range of policy decisions. This will help to avoid inconsistencies of the type where a million is spent in one area to avoid the loss of, e.g., a life year while refusing to spend a thousand for the same in another area. A consistent approach also provides a key mechanism to allow scrutiny of policies and determine the accountability of decision makers. By producing an auditable and open CBA it demonstrates in detail to stakeholders exactly the reasoning behind the decision. This explanation allows all stakeholders to review the methodology that has been employed, the assumptions that were made and the sources of information. It should also be noted that such an approach does not preclude the influence of impacts that could not be reflected in the CBA, but does highlight these factors for scrutiny.

Scale The scale at which CBA will be performed differs between different problems, because of differences of the dose–response functions and of the social appreciation of the costs and benefits. A CBA for N effects and measures in urban air pollution will be very different from a CBA for marine eutrophication (see e.g. Jacobsen et╯al., 2007). Therefore, when applying CBA for the complete N-cycle, breaking down the problem to a smaller scale is necessary.

Dealing with uncertainties Uncertainty can become a major factor if the perceived range covers the switching level between a policy being cost beneficial or not. Uncertainties are of several types:€data uncertainties which can often be described using statistical distributions

516

and dealt with accordingly; modelling assumptions that may be approached using sensitivity analysis; and biases resulting from (e.g.) a lack of data. There is a tendency to regard uncertainties as affecting estimates of benefit more than estimates of cost. This view arises from the perspective that the costs of individual and well defined abatement measures can be described with a reasonable level of accuracy, though there is a tendency for actual costs of specific measures to be lower through improvements in efficiency and identification of other cost savings once measures start to be widely deployed. When a policy permits flexibility (e.g. through setting national emission ceilings) uncertainties in abatement costs become much more significant as experience shows that the response of industry (etc.) can be different to that originally envisaged. As it is clearly in the interests of affected parties to minimize costs once a policy is in force, the tendency is for this source of bias to exaggerate costs. For a discussion of a few ways to deal with various kinds of uncertainties in the context of using CBA in decisionmaking, see Supplementary materials, referenced at the end of the chapter.

Distributional impacts As an illustration of the relevance of distributional impacts of policy options, consider a policy that increases the wealth of 10% of the population with the highest wealth by 100 million euro at the cost of 95 million euro from the 10% of the population with the lowest wealth. Taken as an aggregate the population is 5 million euro better off financially. However, that does not necessarily mean that total social welfare has increased, and in this example it may even have been reduced. To partially reflect such considerations, some Member States, such as the UK, have produced distributional weights by income (Treasury, 2010). A second aspect relates to the population over which the impacts are dispersed. This could become a policy consideration, particularly if the costs are borne by a relatively small group, whereas the benefits are distributed over a much larger population. In addition to the specific work that has been done to try and reflect such equity considerations, there is generally also a requirement to reflect the distributional impacts. Such requirements are provided in impact assessment guidance both at the EU level and in many member states such as the UK.

Discounting Discounting is used to express time-preference and make costs and benefits that occur at different points in time comparable. It is obvious that the choice of discount rate is important for the outcome of a CBA. This is in particular the case if costs and benefits occur at very different points in time, which is often true for policy options with respect to environmental quality (e.g. climate change, see Markandya et╯al., 2001). Even for discount rates that are not excessively high, it follows that costs and benefits to future generations are practically ignored (De Zeeuw et╯al., 2008).

Side effects Measures that are aimed at reaching an environmental objective, e.g. good ecological status for rivers, will often have

Corjan Brink and Hans van Grinsven

additional effects on the environment, which might either Â�positively or negatively affect social welfare. In the literature, several terms are used to depict the associated benefits and costs that arise in conjunction with mitigation policies for a specific purpose, including co-benefits, ancillary impacts, secondary benefits and side effects (Markandya et╯al., 2001). Examples of side effects related to agricultural measures to reduce ammonia emissions are a change in greenhouse gas emissions (Brink et╯al., 2005) and reductions in odour from animal production. The cost effectiveness analysis can in some cases be more complex when both multiple primary and secondary effects are involved as well as upstream–downstream and transboundary effects are observed (Jacobsen, 2007). Although from a policy perspective policy options have primary and secondary effects, from a social welfare perspective all effects are relevant. Therefore, it is obvious that in a social CBA all (intended and unintended) effects have to be considered.

22.3╇ Valuation of nitrogen effects 22.3.1╇ Introduction Reactive N is beneficial for society as it is a key component of chlorophyll, amino acids, proteins and enzymes (see e.g. Olson and Kurtz, 1982). Sufficient supply of Nr is required for plant metabolism, and addition of Nr will essentially increase the efficiency of photosynthesis to produce carbohydrates for food, feed, fibre, etc. In view of the relatively low price of N fertilizer as compared to the value of land, labour and crops, application of Nr is beneficial for farm economy up to high rates (see also Jensen et╯al., 2011, Chapter 3, this volume). These high rates of artificial fertilizer in combination with inefficient handling and use of manures are the reasons that agriculture is now the dominant source of emissions of Nr to the environment in many parts of Europe (Leip et╯al., 2011, Chapter€ 16 this volume). The other major source of Nr is energy use, where formation and emission of nitrogen oxides is a side-effect of combustion of fossil fuels. These emissions cause social cost through impacts on ecosystems, human health and the GHGbalance. As explained in Section 22.1, the optimal level of N-mitigation for society is reached when the marginal cost of mitigation is equal to the marginal benefit of reduced environmental impacts. In the case of agriculture, social damage costs can be mitigated by reducing the Nr-input. The marginal cost of these reductions equal the benefits lost due to decreased crop yield. Von Blottnitz et╯al. (2006) determine the so-called socially and privately optimal rates of N for agricultural production; SONR and PONR (See Box 22.1). Mitigation options other than reducing the Nr-rates are not considered. The benefits of Nr for agriculture can relate both to mass and quality of the crop. Crop mass typically shows a non-linear response to the Nr input rate, with diminishing, or for some crops negative, return with increasing rates. The three major damage categories for society are (i) loss of life years and human health,

Box 22.1╇ Equations for CBA of nitrogen in agriculture (dY/dN)PONR = PN/PC (dY/dN)SONR = (PN + E)/PC where: Y = crop yield (kg/ha) N = input rate of reactive nitrogen (kg/ha) PONR = privately optimal Nr input rate (kg/ha) SONR = socially optimal Nr input rate (kg/ha) PN = price of Nr (purchase and handling; euro/kg) PC = price of crop (euro/kg) E = externalities; sum of environmental damage costs (euro/kg) E = ∑D(Ni) PD where: D = social damage caused by nitrogen (euro/impact) Ni = emission of Nr compound (kg/ha) Ni = ef A or ef N ef = emission factor A = economic activity PD = social cost of environmental damage (impact/kg Nr) UBoN = (YPONR−YN=0)*PC/PONR€– PN where: UBoN = net unit crop benefit of Nr (euro/kg N)

(ii) loss of biodiversity and ecosystem services, (iii)€ climate change. As these social impacts have multiple causes and strongly depend on resilience and resistance of, respectively, humans, ecosystems and climate, major problems exist to derive causal relations with emissions of Nr, and to value the Nr-share of the impact. An overview of unit benefits (PC) and damage costs (E) of Nr follows.

22.3.2╇ Benefits for agriculture The economic benefit of Nr for the farmer depends on prices of crops and fertilizer. Although crop prices are somewhat volatile, increasing prices of artificial fertilizer (Jensen et╯al., 2011, Chapter 3 this volume) cause the price ratio PN/PC to increase in time. As a result, the marginal net benefit of Nr for farm economy also tends to decrease. Extrapolation of the trend of the price ratio between 1995 and 2008 to 2020 would give a price ratio of 10 as compared to present values of 7. Values of PONR in 2020 would then be about 15 kg/ha lower than pres�ent values. On the other hand uncertainties about the (non-linear) response of crop yield to Nr-rate cause unit benefits of Nr (tangent in Figure 22.2) to be uncertain and consequently also the value of PONR. From the perspective of the farmer as a risk manager, this uncertainty of response in combination with uncertainty about weather conditions during the oncoming growing season (in fact there is a suite of possible response curves), may cause him to focus more on the average (chord in Figure 22.2) than on marginal (tangent in Figure 22.2) economic return on his investment in Nr. This behaviour of the farmer is amplified by the small share of costs of Nr in the total variable production costs. Pedersen et╯al. (2005) showed that the N-costs for potato farming in six out of seven

517

Costs and benefits of nitrogen in the environment Table 22.1 Unit benefits of reactive N in euro per kg of N for various crops on sandy soils as a function of reduction of N-fertilizer use relative to 2006 (in general the fertilizer recommendation) in the Netherlands (values are gross benefits as savings on fertilizer, or loss of income from manure acceptance are not considered; inferred from van Dijk et╯al., 2007)

N-fertilization relative to 2006 (= recommended) levels 100%–90%

90%–80%

80%–70%

70%–60%

60%–50%

crop benefits of N (euro per kg Nr) Edible potato

2.1

2.6

3.0

Starch potato

0.3

1.0

1.7

2.3

2.7

0.3

1.7

2.7

4.0

5.0

Silage maize

0.5

0.8

1.4

1.4

1.6

Leek

2.0

3.3

4.3

5.5

6.9

Broccoli

7.0

9.4

11.9

14.6

17.4

Cauliflower

3.3

5.4

7.4

9.5

11.9

Tulip bulb

6.3

10.0

13.5

18.0

22.3

13.2

19.4

26.1

33.2

31.3

certain. Dividing UBoN by the price of fertilizer-N yields the net financial return on the investment in fertilizer (euro per euro). At fertilizer prices in 2006 of 0.8 euro per kg Nr (calcium ammonium nitrate, CAN) N-fertilizer levels in 2006 were exceeding PONR for starch potato, sugar beet and silage maize. In addition to the tendency for the N-recommendation to exceed PONR, farmers tend to apply more N-fertilizer than recommended.

10 at PONR ∂Y/∂N = PN/PC at SONR ∂Y/∂N = (PN+E)/PC

8 Grain yield (ton/ha)

4.2

Sugar beet

Lily bulb

6 Mean N-response: (YPONR-Yo)/PONR

4

Unit benefits of N for winter wheat, oilseed rape and dairy in Europe

2 SONR

0

0

100

PONR

N-rate for maximum yield

200 N rate (kg/ha)

300

400

Figure 22.2 Example of a yield response curve for winter wheat in the UK demonstrating the marginal and average unit benefit of annual nitrogen fertilizer inputs.

European Member States did not exceed 5% of variable costs. N-cost for cereal production, however can amount to 20%.

Effect of crop type and N-level Using yield curves based on field trials, commonly used second order polynomial fits and 2006 price levels (Van Dijk et╯al., 2007) indicative unit crop benefits (euro) per kg of Nr (UBoN) can be derived for a range of arable and horticulture crops (Table 22.1). Within a range of 0% to 20% below the recommended N fertilization level, unit benefits range between 0.5 and 20 euro per kg Nr (Table 22.1) and unit benefits will increase when N-fertilization is further reduced. Unit benefits for vegetables and flowers are clearly higher than those for arable crops and return on investment in N-fertilizer for these crops is

518

3.8

Winter wheat is the major arable crop in Europe using about 25% of agricultural area and total N-fertilizer use in Europe (EU27), and was therefore selected to provide some more information on variation and uncertainty of Nr benefits. The data used represent a wide range of conditions, rotations and setups of field trials. The net unit benefit of Nr (UBoN:€ (Yopt − Yo)*PC/PONR€– PN) ranges from 0.4 euro per kg Nr in SE Europe to 2.7 euro per kg Nr in NW Europe. Oilseed rape is an emerging crop but the market and price for it is uncertain because of current developments in EU policies for climate and biofuels. Values of UBoN for oilseed rape (Figure 22.3) range between 0.3 and 2.3 euro per kg and therefore are somewhat lower than for winter wheat in view of lower yield and weaker response to Nr. The share of roughage and feed concentrates use for the dairy sector in Europe varies strongly; with high shares of roughage in regions with the largest areas of productive grassland and low land prices (northern Europe, Ireland) and lower shares in regions with intensive dairy production but high land prices (Netherlands, Flanders) or low productivity (semi-arid). Using yield response data for grassland in the UK, the Netherlands and Flanders UBoN was found to range between 1.2 and 3.3 euro per kg Nr (Figure 22.3) and therefore similar to values for wheat in NW European countries. UBoN values for production

Corjan Brink and Hans van Grinsven 12 Milk 10 OSR Frequency

8

Wheat

6 4 2 0 1

2

3

4

Economic return on N (euro/kgN) Figure 22.3 Economic return in first year an investment on nitrogen fertilizer for grain of wheat and oil seed rape (OSR), and for milk, using results of field trials (N = 23) and based on a wheat price of 125 euro/ton, a OSR price of 220 euro/ton and a milk price of 290 euro/ton and a CAN-fertilizer price of 0.8 euro/kgN.

of roughage for silage for these four cases range between 0.6 and 2 euro per kg Nr using a herbage price of 100 euro per tonne. Combining UBoN results for 23 data sets from field trials for winter wheat, oilseed rape and dairy (Figure 22.3) a median value of UBoN of 1.7 euro per kg Nr was found (see Supplementary materials). There is evidence that in the EU protein supply to livestock is in surplus. Witzke and Oenema (2007) estimated protein surplus to range between 10% and 20%. Furthermore, statistical data on N excretion of livestock show substantial variation within one member state (e.g. for fattening pigs in the Netherlands about +/− 20%) indicating there is room for efficiency gains. The implementation of low-protein animal feeding is not without cost for the farmer as this will restrict the use of certain types of inexpensive industrial residues and raw materials in the compound feed industry. Oenema et╯al. (2009) estimated the total cost of a 10% reduction of protein in feed at 10 billion euro/yr for EU27.

Uncertainties in the net unit crop benefit of N (UBoN) Values of UBoN are uncertain and change in time, which are uncertainties a farmer has to deal with when deciding on the proper Nr-rate for his crops. Uncertainties include the following. • The shape of yield response curve:€the value of PONR, and consequently also the value of UBoN, is sensitive to the shape of the response function. Gandorfer (2006) and Henke et╯al. (2007) used different equations to fit yield response of winter wheat and found a PONR range of +/− 15%. The corresponding range of UBoN is about 1 euro per kg Nr. • Annual variation of weather:€PONR for winter wheat for individual years ranged between 145 and 235 kg per ha; the resulting uncertainty of UBoN is about 3 euro per kg Nr. • Use of N in manure:€about one third of N-supply in agriculture in EU27 comes from manure-N (Velthof et╯al., 2009). However, effective supply is much less, as it is economically unattractive, both for cattle and arable farmers

to use manure-N, in view of costs of transport, application and uncertain composition and lower nutrient efficiency. It is not uncommon that in areas with high livestock densities manure N has a negative economic value (and UBoN). • Prices of food and N-fertilizer:€when farmers fertilize at the economically optimal level, UBoN is relatively insensitive to the price ratio. Unit benefits of N for crop production based on trials are underestimates, as in these trials the yield at zero N-addition is on average half (35–80% for the 23 trials in Figure 22.3) of the maximum yield. So to a large extent crops grow on N-resources from the soil and previous crops. For the case studies used for Figure 22.3, estimates of these N-resources are in the range of 70–200 kg per ha and clearly related to Nr-fertilization prior to the trial year, and to some extent on other Nr-inputs like biological fixation of N2 and atmospheric Nr deposition. Atmospheric Nr deposition on agricultural soils in the EU is about 3 Mt per yr and constitutes about 20% of the total input (Leip et╯al., 2011, Chapter 16 this volume) and a much larger share for organic farming. Including long-term effects of Nr-fertilization on crops yield would on average double the unit benefit. Using data from (rare) long term field trials, like the Broadbalk Experiment at Rothamsted (Brentrup et╯al., 2004) for continuous winter wheat, gave a UBoN value of 3.5 euro per kg Nr and therefore 1–2€euro€per kg higher than for annual trials.

22.3.3╇ Impacts on ecosystem health and function Valuation of the impacts on biodiversity A useful starting point for economic valuation of impacts of N on ecosystems is the Millennium Ecosystem Assessment’s classification of ecosystem services (MEA, 2005). This classification focuses on four ecosystem service (ESS) types provided by the natural environment that have positive or negative consequences for human populations:€ (i) provisioning services, (ii) regulating services, (iii) supporting services and (iv) cultural services (Figure 22.4). Following this framework, biodiversity is therefore valued through its impact on the different ecosystems’ ability to provide key services underpinning human well-being. Although, biodiversity valuation is inherently an interdisciplinary area of study, it is clear from review studies (Nunes et╯al., 2003; Markandya et╯al., 2008; Raffaelli et╯al., 2009) that existing studies are either dominated by a social science perspective (O’Neill, 1997; Lee and Mjelde, 2007) or a natural science perspective (Costanza, 1980; Odum and Odum, 2000).

Valuation of loss of ecosystem services due to N Although the TEEB-COPI study (Braat and ten Brink, 2008) did not make the role of Nr explicit, it lists (only) three studies that provide data to derive unit cost values for ESS linked to N. A value of 2.2 euro per kg Nr was given for the ESS ‘Water purification and waste management’ both for scrubland and grassland, and 25 euro per kg NOx-N for the ESS ‘Air quality maintenance’. Pretty et╯al. (2003) quantified costs of freshwater eutrophication in England and Wales. The problem with using this

519

Costs and benefits of nitrogen in the environment Figure 22.4 Mapping the link between biodiversity, ecosystem services and human well being (source MEA, 2005).

study for deriving unit costs is twofold:€ (1) no distinction is made between the effect of Nr and phosphorus, and (2) damage costs have a mixed background and some cases are in fact control costs. Cost items considered are reduced value of waterside properties, drinking water production, reduced recreational and amenity value, and also loss of biodiversity. They estimate total damage cost due to loss of ESS at 80–120 million euro per year (105–160 US$). Considering a total Nr-runoff (waste water and agricultural runoff) in the UK of around 300 kt per year in 2000 and attributing all loss to N, a unit cost of 0.3 euro per kg can be inferred. Söderqvist and Hasselström (2008) estimated WTP for a clean Baltic (see also Section 22.5.2) updating results from Contingency Valuation surveys in the 1990s. In this survey a random sample of respondents was questioned in the 1990s about their Willingness to Pay (WTP) for a Baltic Sea ‘undisturbed by excessive inputs of nutrients’. The causality and share of N for eutrophication of the Baltic Sea was not made explicit, but instead the WTP for the Baltic Sea objective was made equivalent to a reduction of 50% of the Nr-load. Values of WTP range between 70–160 euro per household for the Eastern European Baltic states with lower GDP, and between 500–800 euro per household WTP values for the Baltic States with high GDP (Table 22.2). Values are somewhat higher than values reported in the AQUAMONEY study for 11 river basins that ranged between 20–200 euro per household (AQUAMONEY, 2010). Assuming that eutrophication damage can be mitigated by a 50% reduction of the Nr-load to the Baltic Sea, WTP results can be converted to an average unit damage cost of 12 euro per kg Nr for the total Baltic drainage basin, and range between 2–6 euro for East European Baltic states and 23–42 euro for the NW European Baltic states (Table 22.3). Gren et╯al. (2008) report a range of unit damage costs of 12–24 euro per kg Nr, based on Söderqvist and Hasselström (2008), using different discount rates. The NEEDS project (New Energy Externalities DevelopÂ� ments for Sustainability; Econcept; Ott et╯al., 2006) is one of the few studies that has attempted to estimate the value of the loss of biodiversity due to acidification and eutrophication across European countries. The authors state that they use a restoration cost approach implicitly assuming that society is willing to bare the costs of restoration and that the cost estimates therefore offer a lower bound estimate of the benefits involved with restoration. Typical results from NEEDS are costs to restore occurrence of target species that have disappeared due

520

to atmospheric deposition of eutrophying and acidifying Nr compounds. These values were converted to provide (low) estimate of average unit damage cost for EU25 of 2.5 euro per kg for NOx-N (range 0.4–10) and 2.3€euro per kg NH3-N (range 0.1–10) (for further details see Supplementary materials).

22.3.4╇ Impacts on human health There are several routes by which N pollutants can affect human health leading to a variety of impacts (Table 22.4. see also Townsend et╯al., 2003). In the following paragraphs, dose–response relations and economic value are discussed for all listed impacts, except, due to lack of information, for odour and global warming.

Air pollution For air pollutants, NOx is a precursor of O3 which is harmful to human health (Moldanová et╯al., 2011, Chapter 18 this volume). The evidence for direct effects of NO2 is less clear and most health impact assessments (including CAFE; Holland et€ al. 2005a,b) have not assumed direct effects; instead they evaluate the health damage of NO2 by applying the dose–Â�response functions of ambient PM to the nitrates that are created in the atmosphere from NOx emissions (health impacts from secondary particulate matter are highly uncertain and debated; for more detail, see Moldanová et╯al., 2011). The monetary value includes market costs (medical treatment, wage and productivity losses, etc.), as well as non-market costs that take into account an individual’s Willingness-to-Pay (WTP) to avoid the risk of pain and suffering. If the WTP for a non-market good has been determined correctly, it is like a price, consistent with prices paid for market goods. The range of mean annual health cost in EU Member States is 2–32 euro per kg N for NOx, and 2–36 euro per kg N for NH3 (Table 22.5). The most important endpoint for air pollution by Nr is mortality from chronic exposure (to ozone and secondary particulate matter) which contributes 67% to health cost. As shown by Rabl (2003), air pollution mortality must be evaluated in terms of loss of life expectancy rather than of number of premature deaths. Thus, one needs the value of a life year (VOLY). But, by contrast to the numerous studies of so called ‘Value of Statistical Life’ (VSL; an unfortunate and often misunderstood name for what is really the ‘willingness to pay for avoiding the risk of an anonymous premature death’), there have been very few studies until now to determine VOLY. For the 1998 and 2000 reports ExternE had calculated VOLY by assuming that VSL is a discounted sum of annual VOLYs; choosing 3.4€million euro

Corjan Brink and Hans van Grinsven Table 22.2 Benefits of accomplishing a ‘clean’ Baltic Sea, converted to willingness to pay (WTP) per household of four people and to N damage costs for ecosystem aspects assuming that by a 50% reduction of the Nr-load a ‘clean’ Baltic will be accomplished

WTP benefits Euro ×106 Denmark

Nr-Load Gg

Populationa Million 2002

WTP euro/ household

Damage cost Euro per kg Nr

920

44

4.5

823

42

Estonia

60

56

1.4

168

2

Finland

610

49

5.1

475

25

Germany

530

46

2.8

746

23

Latvia

60

44

2.4

102

3

Poland

930

318

38.6

96

6

Sweden

1460

74

8.8

664

39

180

83

9.7

74

4

80

93

3.6

89

2

4830

807

76.9

251

12

Russia Lithuania Total

Note:€ In the area draining to the Baltic; data from Hannerz and Destouni, 2006. a

Table 22.3 Summary of unit cost estimates for ecosystem damage by reactive nitrogen

Study

Parameter

Valuation technique

Unit cost (euro per kg Nr)

Eutrophication in UK (Pretty et╯al., 2003)

Loss of biodiversity and services

Mixed

0.3

Eutrophication Baltic (Gren et╯al., 2008)

Clean un-eutrophied Baltic

WTP

12–24

Modelling impacts of atmospheric deposition on terrestrial systems NEEDS (2006)

Disappeared fraction of target species

Ecosystem restoration

3 (NOx) 2 (NH3)

TEEB/COPI (EC, 2008)

Loss of ecosystem services

WTP ecosystem services

2–25

for VSL (a weighted mean of European studies), this implied a VOLY of approximately 100 000 euro/life year. The Current ExternE recommendation for VOLY is 40 000 euro per life year, based on a contingent valuation by the ExternE team in nine countries of Europe with a total sample size of almost 1500 (Desaigues et╯al., 2007). The European commission currently recommends a range of 52╛000 to 1╛20╛000 euro per life year.

Nitrate in drinking water The threat to human health of nitrate in drinking water was described in an earlier chapter (Grizzetti et╯al., 2011, Chapter 17 this volume). Although a regulatory limit of 50 mg/l for nitrate in drinking water has been in place in the EU since 1980, there are no reliable dose–response functions to assess health loss or mortality due to nitrate in drinking water. Epidemiological studies providing evidence for health impacts are rare:€ the European Food Safety Authority (2008) concluded from a review of recent epidemiological studies that even ‘these were mostly studies with a weak study design and limited strength of evidence’. While there is consensus that the association between

nitrate and methaemoglobinaemia is weak, there is emerging evidence for increased incidence of colon cancer to be one of the more prominent chronic health impacts of nitrate in drinking water exceeding 25 mg/l nitrate (DeRoos et╯al., 2003; Van Grinsven et╯al., 2010; Grizzetti et╯al., 2011, Chapter 17, this volume). Using data for 11 EU member states the total population exposed to drinking water exceeding 25 mg/l nitrate was estimated at 23 million persons (6.5% of the total population) of which 8 million persons (2.3%) were exposed through public supply (Table 22.6). The associated increase of incidence of colon cancer was estimated at 3% (Van Grinsven et╯al., 2010). The total monetary value of this loss of life was estimated at 1.6 billion euro per year or 4.5 euro per average individual, using a value of 40 000 euro/yr for years lost due to premature death and 12 000 for years of health lost due to suffering from colon cancer. The mean unit damage cost for the 11 member states is estimated at 0.7 euro per kg of N-leaching (range 0.1–2.4 euro per kg Nr) when assuming that a 100% reduction of N-leaching is required to fully prevent exceedance of 25 mg/l NO3. The lowest

521

Costs and benefits of nitrogen in the environment Table 22.4 Overview of Nr related health impacts

Health impacts and routes

Health impacts

Unit damage cost

NOx

Inhalation - impacts via O3 - impacts via PM10 - direct impacts of NO2

Asthma, respiratory disorder, inflammation of airways, reduced lung functions, bronchitis, cancers

5.6 euro/kg NOx (euro price level 2000 inferred from CAFE results)

NH3

Inhalation: - direct impacts (negligible) - impacts via PM10 - odour

See NOx Small as odour contribution by NH3 is modest

9.5 euro/kg NH3 (euro price level 2000 inferred from CAFE results)

N2O

Health impacts from global warming, often enhanced by eutrophication

Enhancement of vectors for infectious diseases (malaria) and frequency of infestations (HABS, insects)

Nitrate

Drinking water intake followed by conversion to nitrite. Nitrite is a precursor for carcinogenic N-nitroso compounds and nitrite binds to haemoglobin

Cancers (e.g. colon, neural tube) and reproductive outcome from chronic exposure. Methaemoglobinaemia (blue baby disease)

Pollutant

0.7 euro/kg NO3-N For exposure through public and private wells using groundwater

Table 22.5 Unit damage costs for health impacts by airborne NOx and NH3 (euro per kg Nr; using VOLY 40 000 euro per life year and the CAFE/WHO methodology (Methodex, 2010)

NH3 euro per kg N

NOx euro per kg N

NH3 euro per kg N

Austria

15

29

Latvia

4

5

Belgium

36

17

Lithuania

2

6

Czech Republic

24

24

Luxembourg

30

29

Denmark

10

14

Netherlands

27

22

Estonia

3

3

Poland

12

13

Finland

3

2

Portugal

4

4

France

15

25

Slovakia

17

17

Germany

22

32

Slovenia

4

3

13

Greece Hungary Ireland Italy

16

22

Spain

5

9

18

Sweden

7

7

3

12

United Kingdom

21

13

13

19

values are found for Ireland, the UK and the Netherlands and could be viewed as benefits of investments in a good drinking water infrastructure and protection against nitrate pollution. Highest values are found for Austria, Denmark, Italy, France and Germany and in part reflect high nitrate leaching rates and a lower proportion of the population connected to large public drinking water supply facilities. The unit damage cost results are tentative. In view of ongoing discussions on clinical and epidemiological evidence for adverse health of nitrate, a lower bound for the unit damage of zero seems appropriate. Unit damage cost could also be higher,

522

NOx euro per kg N

for example, when other chronic health impacts are included, or by including drinking water from surface water resources and data from central and eastern Europe or when present exceedance of 25 mg/l NO3 is attributed to a smaller (e.g. 50%) share of present N leaching.

Depletion of stratospheric ozone by N2O Depletion of the stratospheric ozone layer by man-made chemicals causes skin cancers and cataract. Owing to the large reduction of emissions, such as hydrochlorofluorocarbons and halons, after implementation of the Montreal Protocol, nitrous

523

5.9

16.2

3.4

9.7

8.9

2.8

6.8

3.5

3.8

2.0

Denmark

Finland

France

Germany

Ireland

Italy

Netherlands

Spain

UK

Total

8.8

Belgium

139.2

20.4

12.6

5.5

28.3

1.1

42.0

19.8

1.2

2.0

3.7

2.5

×1000

million

Austria

Colon cancer incidence (1993– 1997)

Total population exposed to >25 mg/l NO3

5

0.2

0.2

0.1

1.0

0.0

1.9

1.0

0.0

0.2

0.1

0.1

×1000

Additional colon cancer cases due to nitrate

23

1.0

1.2

0.5

4.7

0.1

9.1

4.7

0.1

0.8

0.5

0.5

×1000

Total number of lost healthy life years before death

18

0.8

0.9

0.4

3.6

0.1

7.1

3.6

0.1

0.6

0.4

0.4

×1000

Total number of lost life years from premature death

1000

43

51

20

202

3

393

202

4

35

23

2.9

0.7

1.3

1.3

3.5

0.9

4.8

3.4

0.9

6.6

2.2

2.9

0.7

0.2

0.4

0.2

1.9

0.1

1.4

0.6

0.8

0.6

2.4

1.9

euro/kg N

euro/ capita million euro/year 23

Unit health damage cost from N-leaching agricultural land Monetary values of loss of (healthy) life years

Monetary values of loss of (healthy) life years

Table 22.6 Assessment of increased incidence and damage costs for nitrate in drinking water derived from groundwater resources, in 11 EU member states using groundwater and drinking water quality data, and cancer registration data for the mid 1900s (Van Grinsven et╯al., 2010)

Costs and benefits of nitrogen in the environment

oxide presently is the dominant ozone depleting substance (ODS; Ravishankara et╯al., 2009). Total health loss by ozone depletion in 2007 was estimated at about 500 000 disabilityadjusted life years (DALY) by Struijs et╯al. (2010). A first estimate of the unit damage cost for human health impacts by N2O was made taking an ozone depletion potential (ODP) of 0.017 as compared to CFC-11 (CFCl3). The concept of Equivalent Effective Stratospheric Chlorine (EESC, expressed in ppt), applied by Struijs et╯al. (2010) and Ravishankara et╯al. (2009) was used to estimate the cumulative EESC reduction by N2O in a scenario where all anthropogenic emissions of ODSs were halted in 2011 (Daniels et╯al., 2010). This calculation takes into account both the different ODP and the atmospheric fate of the ODSs. The cumulative EESC reduction by N2O between 2011 and 2050 was calculated at 1220 ppt yr, which is 6% of the total EESC reduction (Daniels et╯al., 2010). Taking a DALY loss of 806 DALY/ppt/year (Struijs et╯al., 2010), then yields a health loss of 24.2 DALY per kton of N2O. From this the unit damage cost was calculated at 3 euro per kg N2O€–N (taking an economic value of 40 000 euro per DALY, as approximating to the value of VOLY used above). Some major sources of uncertainties are the choice of the N2O reduction scenario and the dose–response relation for cataracts.

22.4╇ Costs of mitigation Implementation of mitigation options generally involves cost to society. To some extent, this cost is a loss in benefits achieved by current N management practices as described in the previous section (e.g. benefits for agriculture). Although the exact distinction is difficult to make (so one must be careful not to double-count), this section mainly deals with the direct cost of mitigation options as a result of additional resources that are required when implementing these options (see also Section 22.2.3).

22.4.1╇ Mitigation options for air quality The main sources of NOx and NH3 emissions are fossil fuel combustion and agriculture, respectively. Information on mitigation options for NOx and NH3 in Europe can be obtained from, e.g., the GAINS model (http://gains.iiasa.ac.at), which includes data on technical measures to reduce emissions from key sources. The GAINS databases on emission and cost parameters have been compiled over several years during national and industrial consultations accompanying preparation of the Thematic Strategy (CEC, 2005), NEC review process, and participation in the work of several UN Expert Groups on abatement technologies (Cofala and Syri, 1998; Klimont and Brink, 2004). Measures that are available for mitigating NH3 and NOx emissions are all targeting emissions at the source. For NH3 from agriculture, these include low N feed, low emission housing for cattle, pigs, and poultry, air scrubbing, covered slurry storage, low ammonia application of slurry and solid manures, incineration of poultry manure, and urea substitution (UNECE, 2007). It is important to stress that some of these options address only one ‘step’ in the emission cascade and so may move abated N from one compartment to the other, e.g.

524

from housing to storage or from storage to land application. The benefit of single ‘compartment’ options is limited except for efficient land application that is at the end of the chain. Principally the measures should be applied in combination with, e.g., low emission housing with closed storage and low emission application. NOx emissions from energy combustion can be reduced via combustion modification (in-furnace controls, e.g. low NOx burners), treatment of the flue gases (by selective catalytic reduction (SCR) or selective non-catalytic reduction (SNCR)), and measures in the transport sector (e.g. changes in engine design, fuel quality, after-treatment of the exhaust gas by various types of catalytic converters, on-board diagnostic systems, etc.). All of these technical measures are characterized by high reduction levels, ranging from 50%–60% for combustion modification to well over 90% for catalytic converters. In Europe, most of the potential for NOx reduction is expected to be exhausted in the next decades (contrary to NH3) as the currently implemented legislation, especially in the EU, requires installation of efficient technologies on stationary sources (see CEC, 2007a) and transport is asked to implement measures with reduction efficiencies over 97% (CEC, 2007b). Any remaining potential is very expensive, with marginal costs ranging from about 5 euro per kg NOx-N in some Eastern European countries to more than 15€euro per kg NOx-N in the Netherlands. Figure 22.5 shows the emission reduction potential and marginal cost for NOx and NH3 for the EU27 in 2020 in addition to the measures already implemented under current legislation (based on the data in the GAINS model). The potential and marginal cost show great variation between countries. It is important to note that some of the NOx measures have potential for increasing emissions of NH3 and N2O (e.g. catalytic converters, fluidized bed combustion) or change the ratio of NO/NO2 emitted which has implications for urban air quality. Although in recent years improvements have been made, owing to the sensitivity to some of these emissions, the issue should be monitored further. Beyond the technical measures listed above, there are a number of so-called non-technical (management) measures having potential to reduce N losses to the air (or water), e.g. timing of application and increased grazing, energy conservation, traffic restrictions and speed limits. For agriculture, Oenema et╯al. (2007) and Velthof et╯al. (2009) characterize a much more exhaustive list of abatement measures that were implemented in the MITERRA-EUROPE model (see also Oenema et╯al., 2011, Chapter 4, this volume). The costs of such non-technical mitigation options are generally more difficult to determine than the costs of technical measures.

22.4.2╇ Mitigation options for water quality Nutrient emissions to surface waters can be reduced in different ways and at different stages of the nutrients’ pathway through the environment. A number of measures reducing emissions at the source, such as reductions in fertilizer use, reductions in livestock and reducing the N content of fodder, simultaneously improve water and air quality. Measures reducing the

Corjan Brink and Hans van Grinsven

Marginal cost (euro/kg NOx-N)

50

40

30

20

10

0 0

100

200

300

400

Emissions reduction (kt NOx-N)

Marginal costs (euro/kg NH3-N)

50

40

30

20

10

0 0

500

1000

1500

Emissions reduction (kt NH3 /year) Figure 22.5 Cost curves for NOx and NH3 mitigation in EU27 countries in 2020; source:€GAINS, current legislation scenario (2010 estimates).

emissions of Nr to the air (NOx or NH3) indirectly contribute to water quality because they result in a reduction in the deposition of Nr (see previous section). In addition, water quality improvements can be achieved by reducing Nr emissions from sewage treatment plants. In the EU these emissions are subject to the Urban Waste Water Treatment Directive, which requires substantial reductions in nutrients concentrations before the effluent is discharged into surface water (EC, 1991). It is also possible to reduce the leaching of Nr from soil to water, e.g. by cultivation of catch crops and reducing discharges of Nr to surface waters by measures such as wetlands at river mouths along the coastal waters, manure-free zones along agricultural land and the use of helophyte filters (Gren et╯al., 2008). Finally, once the Nr has been released into the surface waters, for some water systems it is possible to improve the water quality by removing the Nr from the system, e.g. by ecological fish-stock management (so-called bio-manipulation,

see e.g. Lammens, 1999; Meijer, 2000; Lammens et╯al., 2004) or by dredging. The cost of measures to reduce the emission of Nr to surface waters largely depends on specific local circumstances; e.g. the price of land. Gren (2008) describes the calculated marginal cost of reducing Nr inputs to the Baltic Sea for a range of measures (Table 22.7). Van den Broek et╯al. (2007) report that wet riparian buffer strips are more effective in reducing pollution than dry riparian buffers:€if applied to larger regions wet buffers can decrease Nr emission to surface water by 15 kg Nr per ha at a cost of 37 euro per kg Nr making them a rather costly measure. Dry riparian buffers are even more expensive at an estimated 40 euro per kg Nr. Van der Bolt et╯al. (2008) report average additional cost of reducing Nr emissions to surface waters in the Netherlands of 70 euro per kg Nr for a package of measures including manurefree zones along surface water and of 45 euro per kg Nr for a package of measures including helophyte filters. Figure 22.6 is an example of a cost curve for reducing Nrleaching, based on an evaluation of the implementation of Action Plan for the Aquatic Environment II (Jacobsen et╯al., 2004). Introduction of environmental sensitive schemes and reducing livestock intensity tend to be expensive measures, whereas wetlands and increased utilization of manure are relatively inexpensive. Nitrates in drinking water can be reduced in concentration to prevent exceedance of limit values in drinking water at relatively low cost. Typical measures are mixing polluted water with clean water and biochemical water treatment. Alternatively, the infrastructure for drinking water collection can be adjusted (e.g. deeper extraction wells). Cost data are scarce but are expected to decrease with increasing scale of the drinking water production and treatment. Illustrative annual cost values are 0.5 euro per capita per yr for water treatment and mixing for the UK and the Netherlands where large aquifers are available (Pretty et╯al., 2003; Van Beek et╯al., 2006), 3 euro per capita per yr for Austria and Germany when extraction wells or drinking water infrastructure need adjustment (Ademsam et╯al., 2002; Brandt, 2002) and 15€euro per capita per yr when new private wells are installed.

22.4.3╇ Mitigation options for N2O (greenhouse gas balance) There are two strategies to decrease N2O emissions from agriculture (Oenema et╯al., 2001): • balanced N fertilization, i.e. increasing the N use efficiency together with a lowering of the total Nr input, and • decreasing the release of N2O per unit Nr from the nitrification and denitrification processes.

Increase of N-use efficiencies Improving the N use efficiency reduces both direct N2O emission from soils and indirect N2O releases associated with ammonia emission and nitrate leaching. Measures to increase the N use efficiency in crop production systems, include adjustment of Nr application rate, method, and timing relative to

525

Costs and benefits of nitrogen in the environment Table 22.7 Marginal (calculated) mitigation cost per kg Nr reduction of inputs to the Baltic Sea for a selection of emission reduction measures at sources and end of pipe (Gren et╯al., 2008)

Sewage treatment

Private sewers

Catch crop

Wetlands

euro/kg Nr reduction to coastal waters Denmark

15–35

54–60

31–32

7–18

Finland

15–45

54–77

16–34

1–15

Germany

15–48

54–82

12–35

2–3

Poland

12–48

46–81

9–11

1–1

Sweden

15–79

54–81

5–40

8–290

Estonia

12–35

46–59

6–9

5–7

Lithuania

12–41

46–83

8

2

Latvia

12–49

46–70

15–22

7–10

Russia

12–67

46–115

17–21

10–15

Decreasing the release of N2O per unit Nr

or to change them to create an environment less favourable for N2O production (decrease the N2O/N2 ratio; also see Oenema et╯al., 2001; Butterbach-Bahl et╯al., 2011, Chapter 6 this volume). Options available include the following. • Using ammonium based fertilizer (including urea) instead of nitrate fertilizer during wet conditions may significantly reduce N2O emission (Clayton et╯al., 1997; Velthof et╯al., 1997; Dobbie and Smith, 2003; Jones et╯al., 2005, 2007). This option is cost neutral but emissions of ammonia increase if urea is used without low-emission methods. • Available carbon is an important energy source for denitrifying bacteria (Tiedje, 1988). Avoiding conditions with high contents of available carbon and nitrate in the soil therefore decreases N2O emissions. The costs of these types of measures are low, because they are based on correct timing of N application and choice of fertilizer type. • Nitrification inhibitors delay the conversion of ammonium to nitrate (and possible denitrification of the produced nitrate). Fertilizer containing nitrification inhibitor costs about 1.5 to 2 times more than a common ammonium based fertilizer. • Enhancing aeration of soils by proper drainage, irrigation and soil tillage and avoiding application of N during wet conditions reduce N2O emission from soils. Associated costs are low. • Removal of crop residues from the field. The costs are relatively high, because this requires equipment to collect the residues, and additional costs for handling and storage of the residues. • Winter crops or catch crops reduce the nitrate content of the soil in the winter. Costs are related to soil tillage and seed, and are higher than costs related to improved N management (see also Section 22.4.2). • Proper mixing of the manure may decrease N2O emissions from solid manure systems (Sommer, 2001). The costs of these measures are relatively low.

Measures to reduce N2O emission have to focus on avoiding application of Nr during conditions favourable for denitrification

For further details on mitigation options for N2O, see Supplementary materials.

Marginal cost (euro/kg N-leaching)

12

10 8

6 4

2 0 0

10

20

30

40

Emission reduction (kt N-leaching) Figure 22.6 Cost curve based on the implementation of ‘Action Plan for the Aquatic Environment II’ in Denmark (Jacobsen, 2004).

the crop demand, use of soil and plant testing as a basis for N fertilization, proper manure management (including grazing systems), and accounting for mineralization of organic N. Adjustment of crops in rotation and growth of winter crops are also options to increase N use efficiency. More general measures are improved management of soils and crops. In general, there are no net costs or costs are low for these options, because they result in a higher yield and/or less use of mineral N fertilizer. In an assessment of the global potential to mitigate greenhouse gas emissions, Smith et╯al. (2008) estimated the costs of improved nutrient management at 5 US$ per ha cropland and of improved agronomy (i.e. agronomic practices to increase yields, such as changes in crop rotations) at 20 US$ per ha cropland (see also Jensen et╯al., 2011, Chapter 3, this volume).

526

Corjan Brink and Hans van Grinsven 120

Marginal cost (euro/kg N2O-N)

100

80

60 40

20

0 0

200 400 600 Emission reduction N2O (kt/yr)

800

Figure 22.7 Cost curve for emission reduction of N2O in EU27 for base year 2005; total emission is 1450 kt. Source:€GAINS, current legislation scenario; Winiwarter et╯al. (2009).

The marginal abatement costs for N2O calculated by GAINS (Figure 22.7) range from less than 1 euro per kg N2O-N for balanced fertilization and adjusted tillage, up to 50 euro per kg N2O-N for adjusted timing of fertilization and use of nitrification inhibitors (note€that values can be up to 100 times higher than values per kg N-fertilizer in view of an emission factor for N2O of about 1%).

22.4.4╇ Mitigation options for soil quality Input of N affects soil organic matter content and quality, soil acidification and soil biodiversity and through this soil functions (see also Velthof et╯al., 2011, Chapter 21 this volume):€(i) storage, filtering, buffering and transformations of N, (ii) food and other biomass production, (iii) carbon sink and (iv) biological habitat and gene pool (Dise et╯al., 2011, Chapter 20 this volume). In general, the adverse impacts of N inputs to soil quality of agricultural soils can be mitigated by modest adjustments of management of soil and crop residues. In general, Nr has a positive effect on content and quality of soil organic matter agricultural soils (Glendining and Powlson, 1995). The results of Khan et╯al. (2007) and Shevtsova et╯al. (2003) suggest that in some circumstances N fertilization may enhance the mineralization of soil organic C. However, the apparent negative effect of mineral N fertilizer on soil organic C content may not only be related to enhanced mineralization, but also to differences in the input of crop residues. Options to maintain or increase the organic matter content in agricultural soils include the use of manures, growing winter crops, improved crop residue management, and reduced tillage (Smith et╯al., 2008). The costs of such measures are relatively low (marginal cost of 0.05–0.1 euro per kg Nr for NW European countries, assuming an average N-rate of 100 kg per ha) and they may result in higher crop yields and quality. Options to improve soil quality of natural soils are based on decreasing the Nr content of the soil, such as thinning and sod

cutting, removal of the organic top layer, and choppering (see Diemont and Oude Voshaar, 1994; Niemeyer et╯al., 2007). The costs of these measures per ha can be high, but in general are only applied on a local scale. Liming is widely used to reduce acidification of agricultural soils. The average input of lime in NW Europe is 0.7 kg lime per kg Nr input. If it is assumed that the use of lime is needed to compensate the acidification caused by N-fertilizer; a rough estimate of the cost for lime use is then about 0.1 euro kg Nr. Also in natural systems, liming of soil may reverse the acidification processes (Beier and Rasmussen, 1994). However, Wolf et╯al. (2006) consider permanent liming of forests as an undesirable management option, because it increases the decomposition of soil organic matter leading to thin humus layers and a decrease in soil biota species. However, it can be beneficial as a once-only event in nutrient rich deciduous forest. Options to restore loss of soil biodiversity are related to Nr input and include the use of manures, growth of winter crops and proper soil tillage, and restricted use of pesticides (Brussaard et╯al., 2007; Kibblewhite et╯al., 2008). The costs for measures are low and there may even be benefits for the farmers, as they may increase crop yield and quality. Smith et╯al. (2008) value the costs related to soil tillage and residue management at 5 US$ per ha per year, those related to nutrient management at 5 US$ per ha per year, and those related to agronomy (such as changes in crop rotation) at 20 US$ per ha per year.

22.5╇ CBA use in policy design and evaluation:€case studies 22.5.1╇ CBA for support of European Air Quality Policy Cost–benefit analysis of European air quality policy has built on the methodological framework developed under the European Commission-funded ExternE Project (Bickel and Friedrich, 2005). The first policy applications of this approach date back to the mid-1990s when it was applied to the EU’s Acidification Strategy. Since then it has been applied in development of the EU’s air quality directives, National Emission Ceilings Directive, the Gothenburg Protocol under the Convention on Long-range Transboundary Air Pollution (CLRTAP) and various other legislation concerning industrial emissions. Methods have been refined over time to factor in new research and the views of expert bodies including World Health Organization and working groups convened under CLRTAP. The analysis principally covers effects on human health, crops and building materials. Valuation of ecosystems has yet to be achieved, so ecosystem effects are described only in terms of critical load exceedance. Figure 22.8 illustrates results for the EU’s Thematic Strategy on Air Pollution (Pye, et al. 2008), which feeds into the revision of the National Emission Ceilings Directive. The figure shows for each country the ratio of benefits to costs using a conservative estimate of health benefits. The scenario for which these data

527

Costs and benefits of nitrogen in the environment Table 22.8 Estimated costs and benefits of applying best available emission control techniques at the 10 plants in the EU26 with the largest combined SO2 and NOx baseline emission (Barrett and Holland, 2008)

NOx emission kt/yr

NOx benefit, €M/year

Rank

Country

Plant

1

Bulgaria

Maritsa II

1450

58

103

985

101

9.79

2

Spain

Puentes

1400

19

47

1357

122

11.11

3

Greece

Megalo�polis A

1400

3

285

70

4.08

4

Spain

Teruel

1050

31

77

497

65

7.62

5

Poland

Belchatow

4340

39

147

885

290

3.05

6

Bulgaria

Maritsa I

200

10

26

282

26

11.03

7

Poland

Patnow

1200

11

39

521

100

5.22

8

Spain

Compostilla

1312

31

85

340

107

3.19

9

UK

Cottam

2008

26

74

505

137

3.69

10

UK

Drax

3960

54

198

338

191

1.77

Total cost €M/year

Benefit–cost ratio

Figure 22.8 Illustration of effect of EU Thematic Strategy on Air Pollution (EU) on the ratio of health benefits and cost of measures. Results show benefit–cost ratios for each country under a scenario taking the CAFE-low case for benefit estimation. In all countries the ratio is greater than 1, which implies that a net benefit will be achieved.

100

Benefit–cost ratio

Total benefit, €M/ year

Electrical capacity, MW

10

Austria Belgium Bulgaria Cyprus Czech Denmark Estonia Finland France Germany Greece Hungary Ireland Italy Latvia Lithuania Luxembourg Malta Netherlands Poland Portugal Romania Slovakia Slovenia Spain Sweden UnitedK Total (EU27)

1

were obtained seeks to meet the environmental quality objectives of the Thematic Strategy in the most cost-efficient way, according to the GAINS model (IIASA, 2008). The results presented take account of the EU’s Climate and Energy Package, supplementing the initial analysis for the Commission for which the benefit–cost ratio was lower (demonstrating the clear co-benefits of combined climate and air quality policies). It is clearly demonstrated that the Thematic Strategy policy scenario is estimated to generate significant net benefits for the EU relative to costs. Significant variation in benefits per unit cost is clear across the EU (reasons for which are discussed in more detail below). The application of Monte Carlo analysis and sensitivity analysis demonstrated that the principal conclusion drawn from the quantification, that benefits of the policy would exceed the costs, was robust. Damage costs from different studies are often of limited comparability because of variations in views on what damage

528

should be quantified, dose–response and valuation functions, release of and exposure to air pollutants in different countries and scale. For a consistent set of assumptions, the marginal damage estimates per tonne NOx release can vary by a factor 20 (530–9600 euro per tonne), depending on the country or sea region in which the emissions occur (Methodex, 2010). Given the dominance of health impacts in this analysis the differences between countries largely reflect the probability of someone being exposed to emissions from a particular source. Emissions from countries in central Europe that are surrounded by fairly populated areas for hundreds of kilometres all round are therefore linked with greater damage than countries around the geographical fringe of Europe. Any marginal damage estimates provided without explicit mention of such assumptions are clearly difficult to understand. Similar results can be obtained for ammonia where damage costs in EU25 Member States range between 700 and 30â•›000 euro per tonne.

Corjan Brink and Hans van Grinsven

It is therefore concluded that despite the presence of signifiÂ� cant uncertainties CBA is a useful tool to support European air pollution policy development. A further practical example of the application of CBA is using cost–benefits ratios to select power plants where application of emission control techniques are expected to generate highest social benefits for the EU (Table 22.8; taken from Barret and Holland, 2008).

22.5.2╇ Integrated nutrient management of the Baltic Sea Policy background Damages from eutrophication in the Baltic Sea due to excess Nr and phosphorus loads have been documented since early 1960s by a number of different studies (see, for example, Wulff et╯al., 2001). The affected countries also showed concern by, among other things, the establishment of the administrative body Helcom in charge of policies for improving the Baltic Sea since 1974, and ministerial agreements in 1988 and 2007 (Helcom, 1993; Helcom, 2007). However, approximately 20 years after the meeting in 1988, the agreed level of nutrient reductions in 1988 is far from being reached. One important reason for the hesitation to reduce nutrient loads to the Baltic Sea is likely to be the associated costs, which now start to increase at a higher rate than earlier since the low cost options, such as improvement in nutrient cleaning at sewage treatment plants located at the coastal waters of the Baltic Sea, have been implemented in several countries. Another reason is the differences among countries in perceived benefits from nutrient reduction. Furthermore, a successful implementation of an international agreement requires a perception of fairness by involved stakeholders (Carraro, 2000; Bérubé and Cusson, 2002; Lange et╯al., 2007). In order to calculate net benefits and compare these with different fairness criteria integration is needed of (i) N and phosphorus transports, (ii) upstream and downstream located abatement measures, and (iii) economic and fairness conditions. Although there is a large literature on net benefits from international environmental agreements, there are few studies considering this together with fairness outcomes and with several interlinked pollutants and abatement options. Existing evaluations of international agreement have been made mainly for energy policies (Carraro and Buchner, 2002; Lange et╯al., 2007; Dannenberg, 2008).

Modelling and data retrieval The typical approach for evaluations for energy policies has been to calculate net benefits of mitigation and adaptation strategies and to compare these with different fairness criteria. Using this approach and cost minimization for the Baltic Seas takes into account a number of different abatement measures which either reduce nutrient loads from sources or act as sinks for nutrients. Examples of the former are the use of selective catalytic reductions at combustion sources, livestock reductions, and decreases in use of N fertilizers. Land use changes such as construction of wetlands and grass land provide examples of measures reducing downstream nutrient transports.

For each abatement measure, costs are calculated which do not include any side benefits, such as provision of biodiversity by wetlands. This implies an overestimation of abatement costs of measures implemented in the drainage basins. On the other hand, the cost estimates do not account for eventual dispersion of impacts on the rest of the economy from implementation of the measure in a sector, such as eventual increase in prices of inputs of a simultaneous implementation of improved cleaning at sewage treatment plants. Unless otherwise stated, all data and calculations are found in Gren et╯al. (2008). Given all assumptions, the calculated total nutrient loads of approximately 830€kt of Nr and 40 kt of phosphorus, which come relatively close to the estimates obtained in Helcom (2004) (for further details see Supplementary materials).

Net benefits and fairness Although there is a general consensus on the requirement of fairness for truthful implementation of cleaning plan, there is less agreement on the operational definition of fairness. Usually, a distinction is made between the processes of reaching agreements and the outcome of the agreements (Carraro, 2000; Grasso, 2007). When focusing on fairness, two principles can be distinguished; egalitarian and equity. The egalitarian principle rests on equal human rights, where citizens have the right to, for example, the same amount of emission of N and phosphorus. The equity principle, based on the capability approach suggested by Sen (1999), relates financial burdens of actions to the agents’ ability to meet them. Based on these two principles of fairness with respect to allocation of cleaning among countries, two criteria are included:€emission per capita and related to gross domestic product (GDP). Calculated net benefits from a cost effective achievement of Helcom Baltic Sea Action Plan (BSAP) and fairness outcomes are presented in Table 22.9. The results presented in Table 22.9 indicate a total net gain from the implementation of the Helcom BSAP, but also that the net benefits are unevenly distributed among the countries. However, positive net benefits for all countries can, in principle, be obtained by a suitable choice of international policy instruments. Under a nutrient trading scheme, choice of distribution of initial permits, which implies capital transfers, can be adjusted in order to affect countries’ net benefits. This will also have impact on the outcomes of the fairness criteria, which show significant differences in load per capita and load per GDP. For example, Poland has relatively low nutrient loads but also faces the largest net loss from the abatement programme. This case study points to the need of integrated assessment of net benefits and fairness criteria for truthful implementation of international water management agreements.

22.5.3╇ Cost of implementation of the Nitrate Directive in Denmark Measures and costs to fulfil the Nitrate Directive in Denmark The implementation of the Nitrate Directive in Denmark through the Action Plan for the Aquatic Environment II

529

Costs and benefits of nitrogen in the environment Table 22.9 Net benefits and fairness under cost effective achievement of the Helcom BSAP

Fairness criteria Load/capita

Net benefits Mill I$ Denmark

I$/capita

Kg N

Load/1000 I$ GDP kg P

kg N

kg P

816

177.4

9

0.12

0.26

0.003

Estonia

–111

–82.8

31

0.51

1.67

0.027

Finland

507

96.4

9

0.20

0.26

0.006

Germany

513

155.5

12

0.08

0.38

0.003

Latvia

–163

–71.2

13

0.63

0.84

0.041

Poland

–1752

-45.9

6

0.23

0.40

0.016

Sweden

1354

149.2

8

0.09

0.24

0.003

–86

–9.7

8

0.26

0.64

0.020

–270

–79.2

17

0.33

1.12

0.021

808

10.6

8

0.22

0.41

0.011

Russia Lithuania Total

In order to account for differences in purchasing power among countries, cost and benefit estimates are adjusted by the purchasing power parity (PPP) index and measured in international dollars, I$. The PPP index reflects the purchasing power of a dollar in each country, and varies between 0.7 and 1.9. This adjustment implies an upward adjustment of net benefits in countries with PPP>1 and a downward adjustment when PPP<1.

(Action Plan II) has been accepted by the EU. Based on the results from the technical evaluation of Action Plan II, the cost effectiveness of each measure is calculated (Jacobsen, 2004). The total cost connected to the area related measures (top four measures in Table 22.10) was 27 million euro per year. The reduction in cost compared to expectations is mainly due to a smaller land area with voluntary agreements. The area related measures carry half the costs, but only 16% of the reduction in N-leaching. It should be noted that the area related measures serve a range of purposes which have not been valued, such as lower phosphorus loss, lower pesticide usages and biodiversity. One of the most important farm related measures has been lowering the legal Nr application standards by 10% which is discussed below. The cheapest measures are (1) construction of wetlands, (2) better utilization of Nr in animal manure and (3) changes in feeding. The Environmental Sensitive Schemes (ESAs) and lower animal density on farms are among the most expensive measures when the cost is related to only the reduction in N-leaching. The area-related measures have not achieved objectives, mainly due to the low area involved in their application. On the other hand the reduction in N-leaching due to the farm related measures has achieved the expected level of control and on top of this come the additional measures at the farm level, which ensure that the total aim has been achieved. The total cost is 70 million euro and the cost effectiveness is approximately 2.0 euro per kg of reduced N leaching.

Lower N application€– costs and considerations When trying to estimate the costs of reducing Nr applications there is a need to look both at the change in yield and quality as well as the implications for the value to the farmer. The N application standard was introduced in Denmark in the

530

late 1980s, with maximum application equal to the economic optimum in 1991. This in itself reduced the N-application as some farmers applied more than required for the optimum. In 1998 the application standard was reduced to 10% below the economic optimum of 1991. In the year 1997/98, 38% of all farms applied 20 kg N per ha per year below the standard. These farms were organic farms, but also dairy farms where standards were not binding. This percentage dropped to 10% in the year 2000. The associated area is 1.9 million ha as compared to 2.3 million ha where the application standards were in place in 1997/1998. Another element of the Danish N-policy is that total national N-application is capped to ensure that the national application rate will not rise if, for example, crop prices increase and/or fertilizer costs fall. The total cost of a reduction of 10% in N-standards was estimated using a sector-model at 23 million euro; 10–15 million euro due to lower yield and 7–10 million euro due to loss of crop quality. These model estimates allow change of crop rotation when this is profitable (for more detail see Jacobsen et╯al., 2004). Field trials for cereals have shown that the protein content drops by 0.2% per 10 kg N reduction of the Nr-application. The economic cost of lower protein varies from crop to crop; for bread wheat the cost is fairly high, for barley and export wheat cost is average, and low for forage crops. For development of Danish N-policies environmental gains were not monetized for comparison against economic loss for the farmers. The approach is aimed at finding the most costÂ�effective measures to reach the target. The reduction of N standards to 10% below the economic optimum has increased the incentives to optimize handling of all Nr resources. Together with the required utilization of animal manure there is a large incentive to optimize Nr usage at the farm level. Changes in feeding have reduced the N-leaching more than expected and while implementation costs are limited.

Corjan Brink and Hans van Grinsven Table 22.10 Cost effectiveness (euro per kg N-leaching) for the different measures in Action Plan for the Aquatic Environment II

Area

Reduction N-leaching

Total cost

Cost effectiveness

1000 ha

kt N

million euro

euro per kg N

Wetlands

2.9

0.8

0.7c

0.9

ESA-areas

25.7

0.7

7.7

10.9

Afforestation

14.2

0.8

4.7c

5.9

111.5

3.7

14.0

3.8

3.8

5.7

1.5

0.14

1.5

10.4

3.0

6.4

2.1

10.1

6.7

0.7

12.9

22.8

1.8

35.9

70.2

2.0

Organic farming Changed feeding Lower livestock density

a

Catch crops (6%) Increased utilization N in manure (15%) Reduced N-standards (10%)

a

a

Sum b

In the technical evaluation the effect of these measures has not been divided into the effect of each measure, which is why the estimation here is somewhat uncertain. b Changes in land use and animal production are not considered. c Annualized using a 4% discount rate and infinite lifetime. Source:€Jacobsen (2004). a

22.5.4╇ Cost of implementing Nitrate Directive for Dutch agriculture Policy development since 1998 Between 1998 and 2006 the Netherlands used the Mineral Accounting System (MINAS) to implement the EU Nitrates Directive. MINAS was a system based on farm gate balances and levies, that did not strictly implement the EU application standard for N in manure of 170 kg per ha per year. For this reason the EU Court of Justice ruled in 2003 that MINAS had to be replaced by a system of more rigid application standards. Another reason was the persisting exceedance of the environmental target in groundwater of 50 mg/l nitrate in sandy and clayey soils in use for agriculture, in spite of apparent decreases of Nr surpluses and nitrate concentrations since 1998. Implementation of application standards for N included a time schedule for gradual tightening of standards based on environmental demands and economic feasibility. In the new system arable farms could accept less manure and part of the dairy farms had to dispose of manure, causing an increase of manure supply. As a result, costs to dispose of manure went up which particularly affected land-less intensive livestock farms. While arable and dairy farmers had several readily available farm measures to deal with their problems, the most feasible solution for pork and poultry farmers was manure processing and export. However, this solution was costly and stimulation of innovation required a subtle and time consuming interaction between farmers (cooperative), commercial manure processors, research and national and EU policies (criteria for export and acceptance of processed low carbon manure as mineral fertilizer when complying with legal application standards for total Nr).

Arable and (non-greenhouse) horticulture For clay soils, application standards were set for the period 2006–2009. For crops on sandy soils where the fertilizer recommendation leads to exceedance of 50 mg/l nitrate in shallow groundwater, establishment of legal standards was postponed to 2007. Modelling showed that costs for arable farming on clay soils mainly result from a reduction of income linked to limitations on manure use and not from lower crop yields due to Nr-shortage. For arable farming on sandy soils, with lower application standards and less opportunity to use manure, a reduction of application standards by 30% in 2009 relative to 2006, costs are similar to those on clay soils, but mainly caused by yield loss due to Nr-shortage. Typical compensating measures to reduce costs are the use of green manure and the application of livestock manure in spring instead of late summer. Precision application of fertilization is also an option, but rather costly and therefore more applicable for horticulture. Costs of measures were also expressed as a reduction of farm income (Table 22.11). This reduction is fairly low, but not irrelevant as profit margins for arable farming also tend to be low. The total annual national cost resulting from a 30% reduction of application standards for arable farming and horticulture on sandy soils (about 200 000 ha) was estimated at 4 million euro (20 euro per ha), which is a modest amount compared to the total cost of production or administrative costs of N-policies. Furthermore it was found that costs were restricted to a small group of farm types, in particular farms with no possibility to use manure (e.g. horticulture on sand). Both findings suggest that there is still scope for substantial and cost-efficient reduction of Nr use in arable farming.

531

Costs and benefits of nitrogen in the environment Table 22.11 Examples of reduction of economic yield due to lowering of application standards in 2009 relative to 2006

Soil type

Reduction of application standards, 2009 relative to 2006

Sector Arable

Example of reduction of application standard

Costs without compensating farm measures

Costs with compensating farm measures

kg/ha

euro/ha

euro/ha (% income)

Clay

0%

Winter wheat:€245→220

15–35

5–20 (1%–4%)

Sand and loess

−30%

Winter wheat:€19→160

0–55

0–40 (0%–5%)

Vegetables

Clay

0%

Brussel sprouts:€320→290

0–20

0–5

Sand

−30%

Brussel sprouts:€290→275

355–490

65–165 (3%–6%)

Flower bulbs

Sand

−30%

Tulip:€200→190

250–1075

0–295 (0%–18%)

Source: PBL (2007).

Table 22.12 Effect of Nr intensity on economic yield of dairy farms (inferred from Van den Ham et╯al., 2007)

Milk production

500 t/yr or 12.5 t/ha

Nitrogen intensity

High N

Low N

700 t/yr or 15.0 t/ha Difference

High N

Low N

Difference

Fertilizer Nr (kg/ha)

164

╇ 89

75

195

140

55

Manure Nr (kg/ha)

272

242

30

339

284

55

Feed Nr (kg/ha)

╇ 91

╇ 92

╇ 1

119

120

╇ 1

Soil surplus Nr (kg/ha)

226

147

79

256

168

88

Milk yield (euro/100 kg)

30.5

30.5

0.0

30.2

29.5

0.7

Dairy Negative economic impacts of the new policy for the dairy sector are modest. Costs can amount to 5–45 euro per ha and occur mainly on intensive dairy farms on sandy soils that apply grazing. These farms may have to buy additional fodder, the cost of which is not compensated by savings on fertilizer. However, most dairy farms can take measures to reduce these costs, like increasing their acreage of silage maize. It was estimated that due to the introduction of the new system the dairy sector had to dispose an additional 2.2 million tonnes of manure in 2006 at a cost 15–20 million euro (1.5–2 euro per ha). Monitoring results also indicate that there is no clear relationship between Nr intensity and economic yield from milk production (Table 22.12). For moderately intensive dairy farms there is no effect at all and for intensive farms economic loss is about 2%.

Intensive animal husbandry The major source of costs from limiting Nr application for intensive animal husbandry is storage and transport of manure (Figure 22.9; Van Grinsven et╯al., 2005). Introduction of the system of application standards, including a strict limit for N in manure (170 kg/ha for arable farms and dairy farms (10%) with no derogation and 250 kg/ha for dairy farms with a derogation (90%) increased manure costs for the pig and poultry sector by about 10 euro per tonne, amounting to national costs of 90 million euro in 2006; PBL, 2007). About half of these costs are not

532

related to storage and transport but to additional remuneration to arable farmers for accepting manure (these transfer costs are not considered for national assessments).

22.6╇ Synthesis and discussion 22.6.1╇ Costs and benefits of N on human health, ecosystems and greenhouse gas balance The results of the monetized environmental impacts for the different Nr-compounds are summarized in Table 22.13 as unit damage costs, i.e. the value of the impact, per unit of Nr, on human health, ecosystems and greenhouse gas balance. The values are presented as a range given the large uncertainties surrounding these damage cost estimates. For example, the relationship between damage and emission levels is in most cases non-linear therefore the unit damage costs (which in fact are the slopes of response function) will depend on the level of Nr emissions. Furthermore, unit damage costs vary between countries by a factor 20 to 100 due to differences in dispersion, exposure and mitigation between countries (see Sections 22.3.3 and 22.3.4). Ideally, the estimates are presented as a function of both biophysical and human population characteristics that significantly affect the size of the impact. Moreover, the estimates are based on different methods, adding to the complexity of direct comparison of

Corjan Brink and Hans van Grinsven

500 National cost of manure (million euro)

Other 400

Administration Manure disposal

300

200

100

0 1985

1990

1995

2000

2005

Figure 22.9 National costs from fertilizer and manure policies for agriculture in the Netherlands (manure disposal costs are net result of additional costs for animal production and additional income for arable production).

Figure 22.10 Low and high estimates of total social damage in EU27 as a result of environmental N-emissions in 2000.

of Nr and associated values is also affected by uncertainties, with particular problems affecting valuation of mortality and quantification of ecosystem impacts. Using economic efficiency as an evaluation criterion, the marginal abatement cost for a specific Nr-compound should not exceed the associated marginal social benefits in terms of avoided damage costs presented in Table 22.13, unless it is considered that there are significant additional elements that remain unquantified, which can often be the case. Current policy scenario studies, however, commonly consider marginal abatement and mitigation costs exceeding the values presented in Table 22.13. Only the mitigation costs for N-leaching used by Jacobsen (2004) and Gren et╯al. (2008) are somewhat lower and do not exceed the social benefits of decreased environmental damage. Aggregating the average unit damage costs presented in Table 22.13 across the EU27 using emission data provides an indication of the total damage due to the emission of Nr. This is presented in Figure 22.10. Again a lower and upper bound is presented in order to properly reflect the uncertainty underlying the damage cost estimates. Accounting for the impacts of the emissions in 2000 of N2O, NOx, NH3 to air and N to water, the total annual N-damage in the EU27 ranges between 70 and 320 billion Euro. This corresponds to a welfare loss of 150–750 euro per capita, which is in turn equivalent to 0.8–3.9% of the average disposable per capita income in the EU27 in 2000 (Eurostat 2010). About 60% of these damage costs are related to human health, 35% to ecosystem health and 5% to the effects on the greenhouse gas balance. Despite these difficulties, some provisional conclusions can be drawn, namely the following. (i) Health impacts of airborne pollution contribute most to social cost of Nr. (ii) Social cost of environmental damage by airborne pollutants NOx-N and NH3-N are similar, but those for NH3-N are more uncertain. (iii) Social cost of damage to ecosystems caused by N-runoff appears to be broadly similar to that by airborne Nr (atmospheric deposition). (iv) The social cost of the damage to aquatic ecosystems by N-emissions to water is higher than that to public health. This is as expected as nitrate pollution of drinking water is strictly regulated and most tap water is purified or blended.

22.6.2╇ Costs and benefits for agriculture the value estimates. It is important to appreciate that there are uncertainties on both sides of the cost–benefit equation. While valuation of the costs of measures to reduce the emission of different Nr compounds and mitigate their effects is fairly well developed, particularly for the air compartment, there is a tendency for costs to be overestimated as they do not account for refinement of existing approaches and the development of new ones by industry. The importance of this tendency varies with the type of policy under investigation. It can be particularly large for flexible mechanisms such as the use of emission ceilings. Quantification of the impacts

For illustration of the scope for improvement of N-management in agriculture it is more meaningful to express N-damage costs and benefits per unit of N applied to agricultural land. This is possible by combining unit damage costs for N-compounds in Table 22.13 with emission factors for N-compounds per unit of N application (see e.g. Velthof et╯al., 2009). Some indicative results are shown for Calcium Ammonium Nitrate (CAN) application (CAN is the most used chemical fertilizer in Europe) to arable land (Table 22.14). Because of the low emission factors for airborne N-compounds from CAN, unit costs by N-emissions to water are the most prominent damage items.

533

Costs and benefits of nitrogen in the environment Table 22.13 Emissions of Nr in EU27 and estimated ranges of unit damage costs for the major Nr pollutants and, between brackets, single values inferred from studies used in this assessment

Emission-EU27a

Health

Ecosystem

Climate

Total

euro/kg Nr

euro/kg Nr

Tg Nr

% agric

euro/kg Nr

euro/kg Nr

Nr to water

4.9

60

0–4 (1b)

5–20 (12d)

5–24 (13)

NH3-N to air

3.5

80

2–20 (12 )

2–10 (2 )

4–30 (14)

NOx -N to air

3.4

10

10–30 (18 )

N2O-N to air

0.8

40

1–3 (2f )

c

e

c

2–10 (2 )

12–40 (20)

e

5–15 (9g)

6–18 (11)

EU27 Emissions for year 2000 based on various sources (e.g. EMEP, MITERRA) Health damage from nitrate in groundwater based drinking water based on Grinsven et╯al. (2010). Lower limit for unit damage costs for health impacts of NO3 (colon cancer) c Based on unit damage costs damage for airborne NOx (20 euro/ kg Nr) and NH3 (12 euro/kg Nr) from ExternE (2005) after conversion of results per mass of pollutant to mass of Nr in pollutant. Range arbitrarily set at ± 10 euro/ kg Nr for both NOx and NH3. With respect to NH3 the lower bound reflects the present debate over the importance of health impacts from ammonium in airborne particulate matter. d Upper bound based on WTP for a ‘healthy Baltic’ from study of Söderqvist and Hasselström (2008) and assumption in Gren et╯al. (2008) that damage can be repaired by 50% reduction of N-load to Baltic Sea. Lower bound arbitrarily set at 25% of upper bound. e Ecosystem damage by deposition of NH3 and NOx on terrestrial ecosystem. Lower bound based on the EU NEEDS project (Ott et╯al., 2006) representing the cost for restoring biodiversity loss due to Nr. Upper bound arbitrarily set at 5 times lower bound as a possible value when using an ecosystem service approach (uncertain share of Nr). f Increased incidence of skin cancers and cataracts from depletion of stratospheric ozone. Unit damage cost is inferred from a global LCA study by Struijs et╯al. (2010). g Climate damage based on contribution of N2O-N to greenhouse gas balance and CO2-price. Uncertainty range based on variation of CO2-price since 2005 between 10 and 30 euro/t. a

b

Table 22.14 Long term social cost of adverse N-effects per kg of CANfertilizer application

N-effect

Min

Max

Nitrate groundwater

0.0

1.4

N-load surface water

0.3

4.0

NH3-emission to air

0.0

0.9

NOx-emission to air

0.0

0.2

N2O-emission to air

0.1

0.3

Total damage

0.4

6.8

Unit damage costs for N2O are small compared to other cost items, implying that N-policies for agriculture should not focus on reduction of emissions of N2O. In addition there are damage costs related to emissions of NOx, NH3 and CO2 during manufacturing of chemical fertilizer; in the range of 0.1–0.3 euro/kg Nr (Von Blottnitz et╯al., 2006). Unit (short-term) benefits of N-fertilizer for producing bulk commodities such as cereals, potatoes, sugar beet, and milk range between 0.3 and 3.3 euro per kg CAN-N (Section 22.3.2). This range is rather narrow compared to the maximum range of damage costs of 0.4–6.8 euro per kg CAN-N (Table 22.14). The ratio of marginal private benefits over marginal environmental cost of an additional kg of N then would then be around 0.5 (taking ratios of lower bounds and upper bounds respectively, assuming that WTP for agricultural products and environment in the EU are correlated). Taking into account long-term benefits of N to secure soil N-availability could nearly double the N-benefits, and would raise the ratio to a value around 1. Upscaling of marginal damage cost to EU27 gives an annual social cost of between 20 and 150 billion euro, as compared to an annual benefit of N-fertilizer for farmers between 10 and 50 billion euro, or 20 to 80 billion euro when including long-term N-benefits (Section 22.3.2).

534

Results suggest that for the present levels of N-fertilisation, the marginal environmental costs of the use of CAN tend to be close to the marginal agricultural benefit. As N-emissions and social impacts increase proportionally with the use of CAN, while effects on crop yield level off (see Section 22.3.1), the risk of externalities exceeding crop benefits will tend to increase with higher inputs. However, it should be stressed that the upper bounds of the environmental costs are indicative and have a lower probability of occurrence than the empirically based upper bounds of the agronomic benefits. Von Blottnitz et╯al. (2006) estimated environmental damage (externalities) from fertilizer application at about 0.5 euro/kg N, which corresponds to the lower bound of the range presented here (Figure 22.11). The value by von Blottnitz et╯al. (2006) is low mainly because they used low cost estimates for N-runoff based on Pretty et╯al. (2003; see Section 22.3.4). When part of the N-addition is in the form of manure, the difference between externalities and net crop benefits will increase in view of the higher emission factors for ammonia (up to 70%), and the lower fertilizing efficiency of N in manure as compared to chemical fertilizer. In view of the high unit damage cost for ammonia the use of manure-N without applying far reaching low emission techniques, therefore, would often be not beneficial for society. To a lesser extent this is also true for use of urea fertilizer, that loses around 15% of N as ammonia. First estimates of how much N application rates should be reduced are obtained by comparing the social optimal N-rate (SONR) to the farm (private) optimal N-rate (PONR; see equations given in Section 22.3.1). For winter wheat in Germany (data from Henke et╯al., 2007), and oilseed rape (using data by Sieling and Kage, 2008) SONR was between 35 and 90 kg/ha lower than the PONR. This difference corresponds rather well to results by Brentrup et╯al. (2004) who found a difference of 50–100 kg/ha by applying LCA using winter wheat data from the Broadbalk Experiment at Rothamsted in the UK.

Corjan Brink and Hans van Grinsven

Marginal value (euro/kg N-fertilizer)

8

N2O-emission to air

7

NOx-emission to air

6

NH3-emission to air

5 4

N-load surface water

3

Nitrate groundwater

2 1 0

min

max

Externalities

min

max

Long-term Neffect

Net crop benefits

8 Marginal value (euro/kg N-fertilizer)

CAN-fertilizer application

Annual

7 Long term

6

Climate

5 4

Ecosystem

3 Human Health

2 1 0

min

max

Externalities

min

max

Net crop benefits

Figure 22.11 Comparison and breakdown of low and high estimates of marginal costs and benefits per kg of applied N at present levels of N-fertilizer (as Calcium Ammonium Nitrate) application for arable agriculture on sandy loam.

Although these results are only indicative, they illustrate the extent of the social costs caused by present Nr-fertilization rates in agriculture and the benefits achievable when reducing current Nr-application levels accounting for the environmental costs. However, one also has to keep in mind the social importance and value attached to existing farming systems in Europe. Nr-benefits are incomplete as social concern about food security and the socio-economic position of farmers and rural communities are not considered here, as well as the benefits of affordable agricultural products for industry, retailers and consumers. For example, applying SONR for winter wheat in NW Europe would cause a loss of grain of 1–2 tonne/ha. This production loss of 20% would need to be compensated, e.g. by developing more N-efficient wheat varieties or increasing wheat production in SE Europe or other parts of the world.

22.6.3╇ Discussion and future challenges Air pollution CBA, using indicators such as ‘unit damage costs per tonne emission’ and ‘benefit–cost ratio’, are increasingly used to evaluate and adjust air pollution policies (see Sections 22.5.1 and 22.5.2). Presently, the EU CAFE programme is the most integrated operational approach for using CBA to support integrated N policies, but this approach does not include emissions and effects of Nr in soil and water, and focuses primarily on human health impacts (partly because of the high level of concern over them). Although cost–efficiency and cost–benefit results from the CAFE procedures are the foundation for setting National Emission Ceilings for SO2 and NOx, they are not for ammonia. In view of the linkage between the ceiling for ammonia and cost of agricultural production, setting this ceiling is primarily based on political negotiation. The resulting negotiated ceilings still cause massive exceedance of critical N deposition levels for ecosystems. ExternE used a ‘revealed preference’ approach to determine the value of acidification and eutrophication effects. This approach is based on the assumption that the decisions policy makers have made in the development of the Gothenburg Protocol under the UNECE Convention on Long-Range Transboundary Air Pollution and of the EU’s National Emission Ceilings Directive provide a proxy for the the social value of acidification and eutrophication. Whilst this method can help to assess consistency between polÂ� icies, the use of such numbers assumes that policy makers are fully informed about all impacts and their implications for social Â�welfare. These not only include environmental damages, but also a number of other factors (e.g. concerns over competitiveness and employment) come into play when decisions on permissible Â�levels of air pollution and associated damage are under considerÂ�Â� ation. All these social costs and benefits will affect the apparent relationship between damage and the costs of avoiding it. In considering how policy makers should react to the current problems with eutrophication it is worth referring back to the problem of acidification in the 1980s. This was seen as being linked only to acidification and its effects on ecosystems (particularly forests and freshwaters) and building materials (particularly for monuments). Now, however, the problem has expanded and impacts on human health are considered the dominant driving force for regional European air pollution policy. Concern about ecosystem damage is increasingly linked to Nr deposition and eutrophication, rather than to acidification. It could be that the policy response to health impacts is little different to an optimal response for ecosystem impacts, in which case the omission from valuation of the latter may be of little consequence. In terms of environmental protection it would be useful to conduct a qualitative assessment of the type of impacts that the health based policies could and could not mitigate, and the extent to which the areas that benefit most from these policies are also the ones at greatest risk from eutrophication.

Water pollution and agriculture Use of cost–benefit assessments for evaluation and design of N-policies for agriculture and aquatic ecology is still uncommon. In the case of the role of Nr for aquatic ecology major

535

Costs and benefits of nitrogen in the environment

reasons appear to be the lack of a strong causality (e.g. in view of the role of phosphorus) and public awareness. The EU Water Framework Directive, in part dealing with trans-boundary transport of Nr in watersheds, does allow exemption to member states in the event that the costs of measures are disproportionate (e.g. to economic resources of a country or industry); either by making emission or water quality objectives less stringent or allowing extension of the time period to achieve the objectives. For agriculture there is the complex balancing of the role of Nr for farm income, food security, environment and a level playing field on the world market. From a farmer’s perspective it is profitable to add additional N fertilizer as it generates robust net revenues up to high input levels and increases the chance for return on investment in other production factors. The issues of food security, farm income and competitiveÂ�Â� ness have moved up the policy agenda since the food crisis in 2008, and the ongoing financial and economic crisis. There is a wide scope for increasing Nr-benefits in agriculture, particularly in NW Europe, but this requires a new kind of international cooperation to deal with the other issues at stake. The compelling obligation to feed Europe and the world could, for example, be combined with partial internalization of environmental costs by policies stimulating an increase of agricultural production in less productive regions of central, eastern and southern Europe. Competitiveness and reduction of environmental impacts could perhaps be combined by using the N-efficiency of agricultural production as a criterion for cross compliance. The N-efficiency would a priori benefit from strong EU wide regulations with respect to application of manure and discounting manure N within the application standards for total N.

Challenges for policy and research Although integrated assessment of social cost and benefits of N in the environment is still under development it already provides guidance and useful insights, both in the domain of policy and science. In the domain of policy provisional results raise questions about the social benefits of present N-fertilization levels in agricultural production and future abatement options for air pollution by NOx and NH3. Another relevant conclusion appears to be that policy priorities for reduction of agricultural emissions of N2O are not currently supported by the expected social benefits of this reduction, when valuing N2O emission according to the CO2-trading price. Some more specific considerations for future N-policies are as follows. • Increase the role of cost–benefit assessment as a supporting tool for policy evaluation and design:€not as the economic truth about environmental policies but as a vehicle to increase transparency of policy decisions. • Take into account co-benefits and side effects beyond the N-cycle when developing integrated N-policies. • Consider to use N-efficiency of agricultural production as policy target and a criterion for cross compliance. Producing more food with less Nr is an important challenge for Europe and the world.

536

• Recognize the uncertainty in estimated abatement costs considering economy of scales and future technology improvements resulting in lower actual costs. • Stricter regulation of manure application and improved N-efficiency, in view of the robust benefits for society. Some specific issues for future research are as follows. • Harmonizing methods to quantify and combine Nr-damage functions for human health, ecosystem health and climate change. The dose–response relations available for the role and share of Nr in ecosystem service provision are still very limited. Dose–response relations are available for the domain of human health but are subject to uncertainty (particularly for health impacts from airborne secondary ammonium and nitrate salt particles and for waterborne nitrate). • Establish the long term effect of N-fertilization on crop yields. Although agronomic research into the yield response and economic benefit of Nr has a long history and a high standard we do not seem to understand the system well enough to find the key to higher N-efficiencies without affecting food security and economic vitality of the farming community. • Quantify the role of airborne ammonium containing particles for health impacts. Present policies and data cause ammonia to be one of dominant cost items, while evidence for health risk of airborne ammonium and nitrate particles remains uncertain.

Acknowledgements This chapter was prepared with the support of the NinE Programme of the European Science Foundation, the NitroEurope IP (funded by the European Commission) and the COST Action 729. The authors thank Jaap Struijs for doing the health damage assessment for N2O.

Supplementary materials Supplementary materials (as referenced in the chapter) are available online through both Cambridge University Press: www.cambridge.org/ena and the Nitrogen in Europe website: www.nine-esf.org/ena.

References Ademsam, H., Bruck, M., Geissler, S. and Fellner, M. (2002). Externe Kosten im Hochbau. Band 1. Studie im auftrag des:€Bundesministerium für Wirstschaf und Arbeit. AQUAMONEY (2010). http://www.aquamoney.ecologic-events.de/ sites/results.html Barret, M. and Holland, M. (2008). The Costs and Health Benefits of Reducing Emissions from Power Stations in Europe. The Swedish NGO Secretariat on Acid Rain, Sweden and the European Environmental Bureau (EEB), Brussels, Belgium. Beek, K. van, Hesen, P., Kappelhof, J. and Vink, K. (2006). Consequenties huidige vermesting voor de waterleidingbedrijven (in Dutch). H2O, 39, 25–28. Beier, C. and Rasmussen, L. (1994). Effects of whole-ecosystem manipulations on ecosystem internal processes. Trends in Ecology and Evolution, 9, 218–223.

Corjan Brink and Hans van Grinsven Bérubé, G. and Cusson, C. (2002). The environmental legal and regulatory frameworks: assessing fairness and efficiency. Energy Policy, 30, 291–298. Bickel, P. and Friedrich, R. (eds.) (2005). Externalities of Energy: Methodology 2005 Update. Office for Official Publications of the European Communities, Luxembourg. http://www.externe.info/. Blottnitz, H. von, Rabl, A., Boiadjiev, D., Taylor, T. and Arnold, S., (2006). Damage costs of nitrogen fertilizer in Europe and their internalization. Journal of Environmental Planning and Management, 49, 413–433. Bolt, F. van der, van Boekel, E., Clevering, O. et╯al. (2008). Ex-ante evaluatie landbouw en KRW :€effect van voorgenomen en potentieel aanvullende maatregelen op de oppervlaktewaterkwaliteit voor nutriënten (in Dutch). Alterra, Wageningen, The Netherlands. http://library.wur.nl/way/ bestanden/clc/1877408.pdf. Braat, L. and ten Brink, P. (2008). The Cost of Policy Inaction: The Case of Not Meeting the 2010 Biodiversity Target, Study for European Commission, DG Environment. Contract ENV.G.1/ ETU/2007/0044. http://ec.europa.eu/environment/nature/ biodiversity/economics/teeb_en.htm Brandt, H. (2002). Kosten und Auswirkungen der Gemeinsamen Agrarpolitik (GAP) in Deutschland. Oxfam Deutschland, Berlin. Brentrup, F., Küsters, J., Lammel, J., Barraclough, P. and Kuhlmann, H. (2004). Environmental impact assessment of agricultural production systems using the life cycle assessment (LCA) methodology II. The application to N fertilizer use in winter wheat production systems. European Journal of Agronomy, 20, 265–279. Brink, C., van Ierland, E. C., Hordijk, L. and Kroeze, C. (2005). Costeffective emission abatement in agriculture in the presence of interrelations:€cases for the Netherlands and Europe. Ecological Economics, 53, 59–74. Broek, J. A van den, van Hofwegen, G., Beekman, W. and Wottiez, M. (2007). Options for Increasing Nutrient Use Efficiency in Dutch Dairy and Arable Farming towards 2030. Statutory Research Task Unit for Nature and Environment, Wageningen, The Netherlands. http://www.wotnatuurenmilieu.wur.nl Brouwer, R. and van Ek, R. (2004). Integrated ecological, economic and social impact assessment of alternative flood control policies in the Netherlands. Ecological Economics, 50, 1–21. Brouwer, R. and De Blois, C. (2008). Integrated modelling of risk and uncertainty underlying the selection of cost-effective water quality measures. Environmental Modelling and Software, 23, 922–937. Brouwer, R., Hofkes, M. and Linderhof, V. (2008). General equilibrium modelling of the direct and indirect economic impacts of water quality improvements in the Netherlands at national and river basin scale. Ecological Economics, 66, 127–140. Brussaard, L., de Ruiter, P. C. and Brown, G. G. (2007). Soil biodiversity for agricultural sustainability. Agriculture, Ecosystems and Environment, 121, 233–244. Butterbach-Bahl, K., Gundersen, P., Ambus, P. et╯al. (2011). Nitrogen processes in terrestrial ecosystems. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Carraro, C. (ed.) (2000). Efficiency and Equity in Climate Change Policy. Kluwer, Amsterdam, The Netherlands. Carraro, C. and Buchner, B. (2002). Equity, development and climate change policy. In:€Climate Change Policy Regimes, International Trade and Growth, ed. C. Carraro and C. Kemfert, B. Buchner, CEPS-ESRI Collaboration Studies, Brussels, Belgium. CEC (2005). Communication from the Commission to the Council and the European Parliament on a Thematic Strategy on Air

Pollution, SEC(2005) 1132. Commission of the European Communities, Brussels, Belgium. CEC (2007a). Proposal for a Regulation of the European Parliament and of the Council on type approval of motor vehicles and engines with respect to emissions from heavy duty vehicles (Euro VI) and on access to vehicle repair and maintenance information, COM(2007) 851 final. Commission of the European Communities, Brussels, Belgium. CEC (2007b). Proposal for a Directive of the European Parliament and of the Council on industrial emissions (integrated pollution prevention and control), COM(2007) 844 final. Commission of the European Communities, Brussels, Belgium. Clayton, H., McTaggart, I. P., Parker, J., Swan, L. and Smith, K. A. (1997). Nitrous oxide emissions from fertilised grassland:€A 2-year study of the effects of N fertiliser form and environmental conditions. Biology and Fertility of Soils, 25, 252–260. Cofala, J. and Syri, S. (1998). Nitrogen Oxides Emissions, Abatement Technologies and Related Costs for Europe in the RAINS Model Database. International Institute for Applied Systems Analysis (IIASA), Laxenburg, Austria. Costanza, R. (1980). Embodied energy and economic valuation. Science, 210, 1219–1224. Daniel, J. S., Fleming, E. L., Portmann, R. W. et╯al. (2010). Options to accelerate ozone recovery:€ozone and climate benefits. Atmospheric Chemistry and Physics Discussions, 10, 10839–10863. Dannenberg. A. (2008). Do equity preferences matter in climate negotiations? An empirical investigation. Paper presented at the 13th EAERE conference, Gothenburg, June, 2008. DeRoos, A. J., Ward, M. H., Lynch, C. F. and Cantor, K. P. (2003). Nitrate in public water supplies and the risk of colon and rectum cancers. Epidemiology, 14, 640–649. De Zeeuw, A., In ‘ t Veld, R., Van Soest, D., Meuleman, L. and Hoogewoning, P. (2008). Social Cost Benefit Analysis for Environmental Policy-Making, Background Study. RMNO (Advisory Council for Spatial Planning, Nature and the Environment), Den Haag, The Netherlands. http://www.rmno.nl/ files_content/SCBA%20report%20webversie.pdf Desaigues, B., Ami, D., Bartczak, A. et╯al. (2007). Final Report on the Monetary Valuation of Mortality and Morbidity Risks from Air Pollution. Deliverable for WP6 of RS1b of NEEDS project. July 2007. Diemont, W. H. and Oude Voshaar, J. H. (1994). Effects of climate and management on the productivity of Dutch heathlands. Journal of Applied Ecology, 31, 709–716. Dijk, W. van, Prins, H., de Haan, M. H. A. et╯al. (2007). Economisch consequenties op bedrijfsniveau van het gebruiksnormenstelsel 2006–2009 voor de melkveehouderij en akker- en tuinbouw (in Dutch), Rapport PPO no. 365. Praktijkonderzoek Plant and Omgeving. Wageningen, The Netherlands. Dise, N. B., Ashmore, M., Belyazid, S. et╯al. (2011). Nitrogen as a threat to European terrestrial biodiversity. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Dobbie, K. E. and Smith K. A. (2003). Impact of different forms of N fertilizer on N2O emission from intensive grassland. Nutrient Cycling in Agroecosystems, 67, 37–46. EC (1991). Council of the European Communities. Directive concerning urban wastewater treatment (91/271/EEC). EC (2008). The economics of ecosystems and biodiversity (TEEB). Interim report 2008. European Commission, Brussels, Belgium. Econcept (2006). NEEDS:€Energy Externalities Developments for Sustainability. Deliverable D.4.2.-RS 1b/WP4€– July06. Econcept AG.

537

Costs and benefits of nitrogen in the environment EFMA (2008). Forecast of Food, Farming and Fertilizer Use 2008 to 2018. Volume 1:€Executive Summary and Regional Data. European Fertilizer Manufactures Association (EFMA), Zurich. European Food Safety Authority (2008). Nitrates in vegetables. Efsa Journal 689. Eurostat (2010). www.epp.eurostat.ec.europa.eu ExternE (2005). Galloway, J. N., Townsend, A. R., Erisman, J. W. et╯al. (2008). Transformation of the nitrogen cycle:€recent trends, questions, and potential solutions. Science, 320, 889–892. Gandorfer, M. (2006). Bewertung von Precision Farming dargestellt am Beispiel der teilflächenspezifischen Stickstoffdüngung. Diss. Technischen Universität, München Glendining, M. J. and Powlson, D. S. (1995). The effects of long continued applications of inorganic nitrogen fertilizer on soil organic nitrogen: a review. In:€Soil Management, Experimental Basis for Sustainability and Environmental Quality, ed. R. Lal and B. A. Stewart, CRC Press, Boca Raton, FL, pp. 385–446. Grasso, M. (2007). A normative ethical framework in climate change. Climatic Change, 81, 223–246. Gren, I.-M. (2008). Costs and Benefits from Nutrient Reductions to the Baltic Sea. The Swedish Environmental Protection Agency, Stockholm, Sweden. http://www.naturvardsverket.se/Documents/ publikationer/978–91–620–5877–7.pdf. Gren, I.-M., Jonzon, Y. and Lindqvist, M. (2008). Costs of Nutrient Reductions to the Baltic Sea: Technical Report, Working paper 2008:1. Department of Economics, SLU, Uppsala. Grinsven, H. van, van Eerdt, M., Willems, J., Hubeek, F. and Mulleneers, E. (2005). Evaluation of the Dutch manure and fertilizer policy, 1998–2002. In:€Evaluating Agri-Environmental Policies:€Design, Practice and Results, Organisation for Economic Co-operation and Development, Paris, pp. 398–410. Grinsven, H. J. M. van, Rabl, A., de Kok, T. M., Grizzetti, B. (2010). Estimation of incidence and social cost of colon cancer due to nitrate in drinking water in the EU:€a tentative cost–benefit assessment. Environmental Health. Grizzetti, B., Bouraoui, F., Billen, G. et╯al. (2011). Nitrogen as a threat to European water quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Ham, A. van den, Daatselaar, C. H. G., Doornewaard, G. J. and de Hoop, D. W. (2007). De eerste ervaringen met het Gebruiksnormenstelsel. Studie in het kader van de Evaluatie Meststoffenwet 2007. Hoofdrapport, Rapport no. 3.07.04. Landbouw Economisch Instituut (LEI), Den Haag, The Netherlands. Hannerz, F. and Destouni, G. (2006). Spatial characterization of the Baltic Sea drainage basin and its unmonitored catchments. Ambio, 35, 214–219. Helcom (1993). The Baltic Sea Joint Comprehensive Environmental Action Programme, Baltic Sea Environmental Proceedings No. 48. Helsinki Commission, Helsinki, Finland. Helcom (2004). The Fourth Baltic Sea Pollution Load Compilation (PLC-4). Helsinki Commission. Helsinki, Finland. Helcom (2007). An Approach to Set Country-Wise Nutrient Reduction Allocations to Reach Good Marine Environment of the Baltic Sea, Helcom BSAP Eutro Expo/2007. Helsinki Commission, Helsinki, Finland. Henke, J., Breusted, G., Sieling, K . and Kage, H. (2007). Impact of uncertainty on the optimum nitrogen fertilization rate and agronomic, ecological and economic factors in an oilseed rape based crop rotation. Journal of Agricultural Science, 145, 455–468.

538

Holland, M., Hunt, A., Hurley, F., Navrud, S. and Watkiss, P. (2005a). Methodology for the Cost–Benefit Analysis for CAFE:€Volume€1:€Overview of Methodology, AEA Technology Environment, Didcot, UK, http://europa.eu.int/comm/ environment/air/cafe/pdf/cba_methodology_vol1.pdf. Holland, M., Watkiss, P., Pye, S., de Oliveira, A. and van Regemorter, D. (2005b). Cost–Benefit Analysis of Policy Option Scenarios for the Clean Air for Europe Programme for Service Contract for Carrying out Cost–Benefit Analysis of Air Quality Related Issues, in Particular in the Clean Air for Europe (CAFE) Programme. http:// www.cafe-cba.org/assets/thematic_strategy_analysis_v3.pdf. IIASA (2008). National Emission Ceilings for 2020 based on the 2008 Climate and Energy Package IIASA, Laxenburg, Austria, http:// www.iiasa.ac.at/rains/reports/NEC6-final110708.pdf. Jacobsen, B. H. (2004). Økonomisk slutevaluering af Vandmiljøplan II, Rapport no. 169. Institute of Food and Resource Economics, Copenhagen University, Denmark. Jacobsen, B. H. (2007). In Search of Cost-Effective Measures, Report No. 191. Institute of Food and Resource Economics, Copenhagen University, Denmark. Jacobsen, B. H., Abildtrup, J., Andersen, M. et╯al. (2004). Omkostninger ved reduktion af landbrugets næringsstoftab til vandmiljøet: Forarbejde til vandmiljøplan III [Cost of reducing nutrient losses from agriculture€– Analyses prior to the Danish Aquatic Programme III]. Report no. 167. Institute of Food and Resource Economics, Copenhagen University, Denmark. Jacobsen, T., Refsgaard, A., Jacobsen, B. H. and Ørum, J. E. (2007). Integrated modelling of nitrate loads to coastal waters and land rent applied to catchment scale water management. Water Science and Technology, 56, 39–47. Jensen, L. S., Schjoerring, J. K., van der Hoek, K. et╯al. (2011) Benefits of nitrogen for food fibre and industrial production. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J.€W. Erisman et╯al., Cambridge University Press. Jones, S. K., Rees, R. M., Skiba, U. M. and Ball, B. C. (2005). Greenhouse gas emissions from a managed grassland. Global and Planetary Change, 47, 201–211. Jones, S. K., Rees, R. M., Skiba, U. M. and Ball, B. C. (2007). Influence of organic and mineral fertilizer on N2O fluxes from a temperate grassland. Agriculture, Ecosystems and Environment, 121, 74–83. Khan, S. A., Mulvaney, R. L., Ellsworth, T. R. and Boast, C. W. (2007). The myth of nitrogen fertilization for soil carbon sequestration. Journal of Environmental Quality, 36, 1821–1832. Kibblewhite, M. G., Ritz, K. and Swift, M. J. (2008). Soil health in agricultural systems. Philosophical Transactions of the Royal Society of London Series B,€Biological Sciences, 363, 685–701. Klimont, Z. and Brink, C. (2004). Modelling of Emissions of Air Pollutants and Greenhouse Gases from Agricultural Sources in Europe. IIASA, Laxenburg, Austria. Lammens, E. H. R. R. (1999). The central role of fish in lake restoration and management. Hydrobiologia, 395–396, 1573–1617. Lammens, E. H. R. R., Nes, E. H. van, Meijer, M.-L. and van den Berg, M. S. (2004). Effects of commercial fishery on the bream population and the expansion of Chara aspera in Lake Veluwe. Ecological Modelling, 177, 233–244. Lange, A., Vogt, C. and Ziegler, A. (2007). On the importance of equity in international climate policy:€empirical analysis. Energy Economics, 29, 545–562. Leip, A., Achermann, B., Billen, G. et╯al. (2011). Integrating nitrogen fluxes at the European scale. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press.

Corjan Brink and Hans van Grinsven Lee, C.-K. and Mjelde, W. J. (2007). Valuation of ecotourism resources using a contingent valuation method:€the case of the Korean DMZ. Ecological Economics, 63, 511–520. Markandya, A., Halsnaes K., Lanza A. et╯al. (2001). Costing methodologies. In:€Climate change 2001. Mitigation, ed. B.€Metz, O. Davidson, R. J. Swart and J. Pan, Cambridge University Press, pp.€451–98. Markandya, A., Nunes P. A. L. D., Bräuer, I. et╯al. (2008). The Economics of Ecosystems and Biodiversity€– Phase 1 (Scoping): Economic Analysis and Synthesis. European Commission, Venice, Italy. Methodex (2010). http://www.methodex.org/BeTa-Methodex%20v2.xls Millennium Ecosystem Assessment (2005). Ecosystems and Human Well-Being:€Biodiversity Synthesis. World Resources Institute, Washington, DC. Meijer, M. L. (2000). Biomanipulation in the Netherlands: 15 years of experience. PhD thesis Wageningen University, The Netherlands. Moldanová, J., Grennfelt, P., Jonsson, Å. et╯al. (2011). Nitrogen as a threat to European air quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Niemeyer, M., Niemeyer, T., Fottner, S., Härdtle, W. and Mohamed, A. (2007). Impact of sod-cutting and choppering on nutrient budgets of dry heathlands. Biological Conservation, 134, 344–353. Nunes, P. A. L. D., van den Bergh, J. C. J. M. and Nijkamp, P. (2003). The Ecological Economics of Biodiversity:€Methods and Applications, E. Elgar, Cheltenham, UK. O’Neill, J. (1997). Managing without prices:€the monetary valuation of biodiversity. Ambio, 26, 546–550. Odum, H. T. and Odum, E. P. (2000). The energetic basis for valuation of ecosystem services. Ecosystems, 3, 21–23. Oenema, O., Velthof, G. L. and Kuikman, P. J. (2001). Technical and policy aspects of strategies to decrease greenhouse gas emissions from agriculture. Nutrient Cycling in Agroecosystems, 60, 301–315. Oenema, O., Oudendag, D. A., Witzke, H. P. et╯al. (2007). Integrated Measures in Agriculture to Reduce Ammonia Emissions, Final summary report. European Commission service contract 070501/2005/422822/MAR/C1. Alterra, Wageningen, The Netherlands. Oenema, O., Witzke, H. P., Klimont, Z., Lesschen, J. P. and Velthof, G. L. (2009). Integrated assessment of promising measures to decrease nitrogen losses from agriculture in EU-27. Agriculture, Ecosystems and Environment, 133, 280–288. Oenema, O., Bleeker, A., Braathen, N. A. et╯al. (2011). Nitrogen in current European policies. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Olson, R. A. and Kurtz, L. T. (1982). Crop nitrogen requirements, utilization and fertilization. In:€Nitrogen in Agricultural Soils, ed. F.€J. Stevenson, American Society of Agronomy, Madison, WI, pp.€567–599. Oosterhuis, F. (ed.) (2006). Ex-Post Estimates of Costs to Business of EU Environmental Legislation, IVM Report E-06/08. Institute for Environmental Studies, Vrije Universiteit Amsterdam, The Netherlands. Oosterhuis, F. (2007). Cost Decreases in Environmental Technology: Evidence from Four Case Studies. IVM Report R-07/05, July 2007. Institute for Environmental Studies, Vrije Universiteit Amsterdam. Ott, W., Baur, M., Kaufmann, Y., Frishknecht, R. and Steiner, R. (2006). Assessment of biodiversity losses. Deliverable D4.2.-RS 1b/ WP4€– July 06. NEEDS (New Energy Externalities Developments and Sustainability). Project nr. 502678. Switzerland.

PBL (2007). Werking van de Meststoffenwet 2006. (In Dutch. Translated title:€Evaluation of the Fertiliser Act 2006. Transition from a system of loss standards to a system of use standards). Bilthoven, The Netherlands. Pedersen, S. M., Bizik, J., Costa, L. D. et╯al. (2005). Potato Production in Europe: A Gross Margin Analysis. Department of Economics, University of Copenhagen, Frederiksberg. Pretty, J. N., Mason, C. F., Nedwell, D. B. et╯al. (2003). Environmental costs of freshwater eutrophication in England and Wales. Environmental Science and Technology, 37, 201–208. Pye, S., Holland, M., Van Regenmorter, D., Wagner, A. and Wattiss, P. (2008). Analysis of the Costs and Benefits of Proposed Revisions to the National Emission Ceilings Directive. NEC CBA Report 3. National Emmision Ceilings for 2020 based on the 2008 Climate and Energy Package. http://www.cafe-cba.org/assets/necdcba3.pdf Rabl, A. (2003). Interpretation of air pollution mortality:€number of deaths or years of life lost? Journal of the Air and Waste Management Association, 53, 41–50. Rafaelli, D., Termansen, M., Austen, M. et╯al. (2009). Valuation of Biodiversity,€NERC, Peterborough, UK. Ravishankara, A. R., Daniel, J. S. and Portmann, R. W. (2009). Nitrous oxide (N2O):€the dominant ozone-depleting substance emitted in the 21st century. Science, 326, 123–125. Sen, A. (1999). Development as Freedom. Anchor Press, New York. Shevtsova, L., Romanenkov, V., Sirotenko, O. et╯al. (2003). Effect of natural and agricultural factors on long-term soil organic matter dynamics in arable soddy-podzolic soils:€modeling and observation. Geoderma, 116, 165–189. Sieling, K. and Kage, H. (2008). The potential of semi-dwarf oilseed rape genotypes to reduce the risk of N leaching. Journal of Agricultural Science, 146, 77–84. Smith, P., Martino, D., Cai, Z. et╯al. (2008). Greenhouse gas mitigation in agriculture. Philosophical Transactions of the Royal Society of London, Series B,€Biological Sciences, 363, 789–813. Söderqvist, T. and Hasselström, L. (2008). The Economic Value of Ecosystem Services Provided by the Baltic Sea and Skagerrak. http://www.naturvardsverket.se/Documents/publikationer/978– 91–620–5874–6.pdf Sommer, S. G. (2001). Effect of composting on nutrient loss and nitrogen availability of cattle deep litter. European Journal of Agronomy, 14, 123–133. Struijs, J., Van Dijk, A., Slaper, H. et╯al. (2010). Spatial- and timeexplicit human damage modeling of ozone depleting substances in life cycle impact assessment. Environmental Science and Technology, 44, 204–209. Tiedje, J. M. (1988). Ecology of denitrification and dissimilatory nitrate reduction to ammonium. In:€Biology of Anaerobic Microorganisms, ed. A. J. B. Zehnder, John Wiley and Sons, New€York, pp. 179–243. Treasury (2010). http://www.hm-treasury.gov.uk/data_greenbook_ index.htm Townsend, A. R., Howarth, R. W., Bazzaz, F. A. et╯al. (2003). Human health effects of a changing global nitrogen cycle. Frontiers in Ecology and the Environment, 1, 240–246. UNECE (2007). Guidance Document on Control Techniques for Preventing and Abating Emissions of Ammonia, ECE/EB.AIR/ WG.5/2007/13. UNECE, Geneva. Velthof, G. L., Oenema, O., Postma, R. and van Beusichem, M. L. (1997). Effects of type and amount of applied nitrogen fertilizer on nitrous oxide fluxes from intensively managed grassland. Nutrient Cycling in Agroecosystems, 46, 257–267.

539

Costs and benefits of nitrogen in the environment Velthof, G. L., Oudendag, D., Witzke, H. P. et╯al. (2009). Integrated assessment of nitrogen emissions from agriculture in EU-27 using MITERRA EUROPE. Journal of Environmental Quality, 38, 402–417. Velthof, G., Barot, S., Bloem, J. et╯al. (2011). Nitrogen as a threat to European soil quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Winiwarter, W., Höglund-Isakssen, L., Schöpp, W. et╯al. (2009). Emission abatement potentials and costs for non-CO2 greenhouse gases in Annex-I countries according to the GAINS model. In: Proceedings of the Fifth International NCGG Symposium, Wageningen, The Netherlands, 30 June€– 3 July 2009.

540

Witzke, H. P. and Oenema, O. (2007). Assessment of Most Promising Measures: Task 3 Service Contract – Integrate Measures in Agriculture to Reduce Ammonia Emissions. http://www.scamonia. wur.nl. Wolf, R. J. A. M., Dimmers, W. J., Hommel, P. W. F. M. et╯al. (2006). Bekalking en toevoegen van nutriënten Evaluatie van de effecten op het bosecosysteem: een veldonderzoek naar vegetatie, humus en bodemfauna. Alterra, Wageningen, The Netherlands. Wulff, F., Bonsdorff, E., Gren, I.-M., Stigebrandt, A. and Johansson, S. (2001). Integrated management of the Baltic Sea. AMBIO, 30, 254–259. Zhang, Z. X. and Folmer, H. (1998). Economic modelling approaches to cost estimates for the control of carbon dioxide emissions. Energy Economics, 20, 101–120.

Chapter

23

Developing integrated approaches to nitrogen management Lead author: Oene Oenema Contributors: Joost Salomez, Cristina Branquinho, Michaela Budňáková, Pavel Čermák, Markus Geupel, Penny Johnes, Chris Tompkins, Till Spranger, Jan Willem Erisman, Christian Pallière, Luc Maene, Rocio Alonso, Rob Maas, Jacob Magid, Mark A. Sutton and Hans van Grinsven

Executive summary Nature of the problem • Reactive nitrogen (Nr) occurs in different forms, arises from a wide range of activities and sources, and leads to environmental impacts over different spatial and temporal scales. • Integrated approaches to N management are anticipated to provide more effective (larger decreases in unwanted emissions) and /or more efficient (less side effects, less costs) policy measures than policy measures based on single sources and pollutant species. • There are many notions of integrated approaches, but as yet little consensus about the best integrated approaches. There is also little quantitative empirical evidence of the performance of these approaches in practice. • The pitfall of integrated approaches is that they may be more complex to agree, leading to a delayed implementation.

Approaches • Based on recent literature and a discussion among experts, the present chapter provides a conceptual framework for developing integrated approaches to N management. • Whilst discussing the framework, various examples of existing partially integrated N management approaches have been considered. • A package of key actions in different sectors is envisaged that, together, should contribute to further developing integrated approaches to N management in the future

Key findings/state of knowledge • The conceptual framework developed here distinguishes five dimensions of integration:€(i) vertical dimension, i.e., cause–effect relationships of N species; (ii) horizontal dimension, i.e., integration of all N species via for example N budgets; (iii) integrating N management with the management of other elements, such as SO2, P, CO2, and CH4, (iv) integrating stakeholders views, and (v) regional integration, i.e., integration over spatial scales. • The toolbox for developing integrated approaches to N management has various types of tools, including systems analyses, communication, integrated assessment modeling, N budgeting, stakeholder dialogue and chain management. • Integrated approaches may be most applicable to agriculture, because of the role of N in food, feed and fiber production and the relative large diffuse N losses from multiple sources in multiple forms.

Major uncertainties/challenges • There is as yet little empirical evidence of the perceived increased effectiveness and efficiency of integrated N management approaches relative to single Nr species and single Nr source management approaches. Potentially, integrated N management approaches may also achieve a broader set of societal targets simultaneously, but there is as yet little empirical evidence for this promise. • The ‘optimum’ level of integration likely depends on many factors, and it remains a challenge to define such optima for various situations and cases.

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

541

Developing integrated approaches to nitrogen management

Recommendations • A package of seven key actions is identified that would contribute to developing an integrated approach to better management of nitrogen in the environment. ❍⚓╇ Improving nitrogen use efficiency in crop production. ❍⚓╇ Improving nitrogen use efficiency in animal production. ❍⚓╇ Increasing the fertilizer N equivalence value of animal manure. ❍⚓╇ Low-emission combustion and energy-efficient systems. ❍⚓╇ Recycling nitrogen (and phosphorus) from waste water systems. ❍⚓╇ Societal consumption patterns:€energy and transport saving. ❍⚓╇ Societal consumption patterns:€lowering animal protein consumption.

23.1╇ Introduction This chapter discusses integrated approaches to nitrogen (N) management and explores options for further development of these approaches. The notion that N needs to be managed in a comprehensive and integrated way follows from the understanding that reactive nitrogen (Nr) once formed is involved in a sequence of transfers, transformations and environmental effects (Galloway and Cowling, 2002; Galloway et╯al., 2008), that the economic costs of emissions abatement are often high, and that the management of a single source and/or a single Nr species, especially agriculture, is not always efficient. There are three main sources of Nr emissions (agriculture, combustion and wastes), with numerous sub-sources, and five main threats of these emissions (water quality, air quality, climate change, biodiversity loss, and soil quality, with various sub threats (Sutton et╯al., 2011, Chapter 5, this volume). Managing each single sub-source of Nr emissions (mainly NH3, N2O, NOx, NO3−) with their impacts in isolation is virtually impossible, because of the bewildering number of sources and the complexity of the cause-effect relationships. The so-called ‘nitrogen cascade’ clearly illustrates the complexity of the global N cycle and also the need for an integrated approach to N management (Galloway et╯al., 2003). Fundamental arguments for using integrated approaches to N management follow also from the first and second law of thermodynamics. Basically, the first law implies that the element N can be transformed into different species, but it can not be ‘destroyed’. The second law of thermodynamics basically implies that N has the natural tendency ‘to dissipate’ into the environment. Nitrogen has even been termed ‘double mobile’, together with carbon and sulfur (Smil, 2001), because these elements are mobile in both air and water (and soil). Though there is scientifically sound underpinning for considering the management of the various N sources in a more holistic and integrated manner, there are also barriers and constraints for more integrated approaches, such as the compartmental and discipline oriented structure and organization of policy departments and science groups. There is also discussion about ‘what and how to integrate?’. In EU policy, there is an increasing tendency for developing more integrated (economicenvironmental) approaches, but many current environmental policies still have a narrow scope as regards N management (Oenema et╯al., 2011; Bull et╯al., 2011, Chapters 4 and 25, this volume). The discussion is in part also confused by lack of clear and accepted definitions about the terms ‘integrate’ and

542

‘management’ (see Supplementary materials). This discussion is not limited to N; there are also pleas for integrating natural sciences and economy in decision making so as to enhance environmental protection and resource use efficiency (Hall et╯al., 2001). The objectives of this chapter are (i) to discuss current approaches to integrated N management and to present a conceptual framework for these approaches, and (ii) to propose some further integrated approaches to N management, including a package of key actions that would contribute to improved overall N management in Europe. Integration of approaches to coordinate efforts at the levels of policy and supranational organizations are discussed in Chapter 25 (Bull et╯al., 2011).

23.2╇ Conceptual framework for integrated approaches 23.2.1╇ Integration dimensions Integration is perceived as combining separate elements and aspects in an organized way, so that the constituent units function cooperatively (see Supplementary materials). There are various integrated approaches to N management in practice, with various degrees of combining separate elements and aspects. We postulate that five different dimensions (categories of elements and aspects) of integration in N management can be distinguished conveniently, namely:€(i) vertical integration, (ii) horizontal integration, (iii) integration of other elements, (iv) integration of stakeholders’ views, and (v) regional integration (Figure 23.1). These dimensions are further discussed below, while referring to existing policies.

23.2.2╇ Vertical and horizontal integration In economy, ecology and also societies, it is helpful to think in terms of levels of organizational hierarchies. A hierarchy is defined as an arrangement into a graded series of entities. A hierarchy can link entities either vertically or horizontally. Vertical integration in economy is the linkage of upstream suppliers to downstream buyers (Figure 23.2). Vertical integration results in more control, higher production efficiency and more marketing power. Vertical integration in ecology is the functional linkage of autotrophic producers to heterotrophic consumers (including herbivores, carnivores, omnivores and saprovores), expressed in the idea of a food chain. In terms of N management, vertical integration relates to linking ‘cause and

Oene Oenema

Vertical integration Integration of other elements

Stakeholders’ views integration

Horizontal integration

Regional integration

Figure 23.1 Framework of integrated N management, with five dimensions (see text). (Source:€original material for this chapter.)

Production chain

Production chain

Retail Wholesale Processing Manufacturer b

VERTICAL INTEGRATION

Manufacturer a Suppliers

HORIZONTAL INTEGRATION

Figure 23.2 Conceptual visualization of vertical and horizontal integration of firms in production chains. (Source:€original material for this chapter.)

effect’, and ‘source and impact’. Examples of vertical integration are the ‘driving forces, pressures, state, impact and response’ framework (DPSIR-framework; see OECD, 1991; EEA, 1995) and the ‘effects- based approach’ to emissions abatement policies as applied in the Gothenburg Protocol (UNECE, 1999). Essentially, vertical integration is the basis of all current N policies in Europe, as the human health effects and ecological impacts are the legitimate of these policies, while the selection of abatement measures is based in part on the economic consequences (cost-effectiveness). Thus, the gains in human health and biodiversity are weighted against the cost of the emission abatement. A full cost–benefit analysis is still complicated, because of the difficulty of attaching monetary values to human health and ecosystems, although significant progress has been described in Chapter 22 (Brink et╯al., 2011, Chapter 22 this volume). Evidently, including cost–benefit analyses would make vertical integration of N management more complete. Horizontal organization is related to up-scaling so as to benefit from larger scale and number. Horizontal integration is the linkage of elements of similar entity, for example

when similar firms merge to benefit from the economics of scale (Figure 23.2). Also the herding of animals, schooling of fishes, flocking of birds and colonies of ants and termites can be considered as forms of horizontal integration. Horizontal integration in N management relates to combining N species, N sources and N emissions within a certain area in the management plan. Partial forms of horizontal integration are in the Gothenburg Protocol (e.g., all anthropogenic NOx sources and all NH3 sources have been included, but N2O emissions to air and N leaching to waters are not included) and the EU Nitrates Directive (all N sources in agriculture have to be considered for reducing NO3 leaching to waters, but NH3 and N2O emissions to air are not addressed explicitly). Similarly, the emission of gaseous N2 through denitrification is not considered in these policies. Although emission of gaseous N2 does not lead directly to adverse environmental effects, its release can be considered as a waste of the energy used to produce Nr, indicating the need that N2 emissions should also be addressed. Conceptually, the N cascade model (Galloway et╯al., 2003; Sutton et╯al., 2011, Chapter 5, this volume) is a nice example of horizontal integration, but this model has not been made operational for management actions yet. The N cascade is also a conceptual model for vertical integration, especially when cost–benefit analyses are included.

23.2.3╇ Integration of other elements and compounds Emissions of nitrogen oxides (NOx), ammonia (NH3) and sulphur dioxide (SO2) to air have rather similar environmental effects (air pollution, acidification, eutrophication), and that is the reason that the effects-based approach of the CLRTAP Gothenburg Protocol and the EU National Emission Ceiling Directive address each of NOx, NH3 and SO2. Similarly, emissions of Nr and phosphorus (P) to surface waters both contribute to eutrophication and biodiversity loss, and thus EU policies related to combat eutrophication of surface waters address N and P simultaneously (Oenema et╯al., 2011, Chapter 4, this volume). Further, the N and carbon (C) cycles in the biosphere are intimately linked, and the perturbations of these cycles contribute to increased emissions of CO2, CH4 and N2O to the atmosphere. Climate change policies address these greenhouse gases simultaneously. Nitrogen may also affect CO2 emissions through its effect on carbon sequestration in the biosphere and by alteration of atmospheric chemistry (Butterbach-Bahl et╯al., 2011, Chapter 19, this volume). Evidently, there are two main reasons to integrate N management with the management of specific other elements (compounds) in environmental policy, namely (i) the other elements (compounds) have similar environmental effects, and (ii) interactions between N species and these other elements and compounds. From the practitioner point of view, there can be benefits when managing N and specific other elements simultaneously. This holds for example for NOx and SO2 (and soot) from combustion sources, and N and P in agriculture and sewage waste treatment.

543

Developing integrated approaches to nitrogen management

23.2.4╇ Stakeholder involvement and integration Any N management policy, whether integrated or not, needs to be: (i) policy-relevant; i.e., address the key environmental and other issues; (ii) scientifically and analytically sound; (iii) cost effective; i.e., costs have to be in proportion to the value of environmental improvement, and (iv) politically legitimate; i.e., acceptable and fair to users. When one or more of these constraints are not fulfilled, the management policy will be less effective, either through a delay in implementation and/or through poor implementation and performance. Satisfying the aforementioned constraints requires communication between actors from policy, science and practice. Tuinstra et╯al. (2006) argue that the credibility, legitimacy and relevance of the science-policy interaction are to a large extent determined by ‘boundary’ work in an early stage of the communication process between policy and science. They analyzed the communication process between policy and science in the Convention for Long-range Transboundary Air Pollution (CLRTAP) and the EU National Emission Ceiling Directive. Boundary work is defined here as the practice of maintaining and withdrawing boundaries between science and policy, thereby shaping and reshaping the science–policy interface. Of similar importance is the communication with practitioners, i.e., the actors that ultimately have to execute management actions in practice. Integrating their views has to be done also as early as possible during the design phase of the N management plans and measures, because the practitioners, in the end, have to implement the management measures. Integrating views of practitioners may range from public consultation procedures, hearings to participatory approaches and learning; the latter take the practitioners’ perspectives fully into account and give them a say also in planning and managing. A good example of the latter approach is the EU Water Framework Directive (EC, 2010), which requires full stakeholder involvement for the establishment of water basin management plans. Integration of practitioners’ views does not necessarily lead to faster decision making; on the contrary, the decision making process often takes more time. Public consultation procedures can be very long-winded, though techniques like multi-criteria decision making (MCDM) may support Â�decision making effectively; this approach aims at deriving a way out of conflicts and to come to a compromise in a transparent process. Integration of practitioners’ views may ultimately improve the acceptance of the management strategies, and thereby facilitate the implementation of the management strategies in practice.

23.2.5╇ Regional integration Regional integration or ‘integration of spatial scales’ is considered here as the fifth dimension of integration. Regional integration aims at enhanced cooperation between regions. It relates to integration of markets and to harmonization of

544

governmental polices and institutions between regions through political agreements, covenants and treaties (Bull et╯al., 2011). Arguments for regional integration are:€ (i) enhancing markets, (ii) creation of a level-playing field, (iii) the transboundary nature of environmental pollutions and (iv) the increased effectiveness and efficiency of regional policies and related management measures. In terms of N management, regional integration relates, for example, to the harmonization and standardization of environmental policies across European Union and for air pollution in the UNECE region (Oenema et╯al., 2011; Bull et╯al., 2011). The water basin or catchment management plans developed within the framework of the EU Water Framework Directive are also a form of regional integration. Here, water quantity and quality aspects are considered in an integrated way for a well-defined catchment. The trend toward regional integration during the last decades does not necessarily mean that local management actions are less effective and/or efficient. Local actions can be made site-specific and, as a consequence, are often more effective than generic measures. This holds both for households, farms and firms, and especially when actors can have influence on the choice of actions. Also, the motivation for contributing to the local environment and nature can be larger than for contributing to the improvement of the environment in general (see Kahn, 2001).

23.3╇ Tools for integrated approaches to N management The toolbox for developing integrated approaches to N management contains tools that are uniformly applicable, as well as highly specific, suitable for just one dimension of integration. Important common tools are:€(i) systems analysis, (ii) communication, (iii) N budgeting, (iv) integrated assessment modeling and cost–benefit analyses, (v) logistics and chain management, and (vi) stakeholder dialogue. The starting point for developing integrated approaches is ‘systems analysis’, as it provides information that is needed for all dimensions of integration. Systems analysis allows for identifying and quantifying components, processes, flows, actors, interactions and inter-linkages within and between systems, and provides a practical tool for discussing integrated approaches to N management. In essence, it encompasses the view that changes in one component will promote changes in all of the components of the systems (Odum, 1996). These type of tools are being used especially by the science-policy interface. A second tool for developing integrated approaches is communication. Communication is transferring information, but at the same time the tool for raising awareness and for explaining the meaning, purpose, targets and actions of integrated approaches to N management to all actors involved. Clear communication is important, as there is often ambiguity in the use of the terms ‘integrated’ and ‘management’ and insufficient clarity about the objectives and required actions. Communication can help make the concept transparent and thereby can facilitate the adoption of targets and measures in practice.

Oene Oenema

A third type of tool is nitrogen balances, which quantifies the differences between nitrogen inputs and outputs of systems and of compartment of these systems. This is an indispensable tool for horizontal integration and in part also vertical integration; it integrates over N sources and N species for well-defined areas and/or components. Input–output N balances have been proven to be easy-to-understand management tools for farmers (Jarvis et╯al., 2011, Chapter 10 this volume), plant managers and policy managers (see Supplementary materials). Input–output balances and budgets are flexible tools, but require uniform definitions and conventions to circumvent bias (Oenema et╯al., 2003; de Vries et╯al., 2011, Leip et╯al., 2011, Chapters 15 and 16, this volume). Life Cycle Assessment (LCA) is an approach to account for emissions and resources during the entire life cycle of a product. It can be seen also as a tool for horizontal integration, similar to input–output budgets, but it integrates also over time. This type of tool is especially used by scientists, while also being relevant for use by practitioners. A fourth type of tool is integrated assessment modeling, including ecological food print analyses, cost–benefit analyses and target setting. These tools are indispensable for vertical integration, relating cause and effect to impact, and analyzing the responses by society (actors). The ‘DPSIR model’ is a conceptual tool for analyzing cause–effect relationships. It relates Driving forces of environmental change (population growth, economic growth, etc.), to Pressures on the environment (e.g., Nr emissions), to State of the environment (e.g., water quality), to Impacts on population, economy and ecosystems, and finally to the Response of the society (OECD, 1991; EEA, 1995). Integrated assessment modeling is the interdisciplinary process that quantifies and analyzes these cause–effect relationships in the current situation (using empirical data and information) and for future conditions (using scenario analyses), in order to facilitate the framing of strategies. Examples include reviews of the Gothenburg Protocol by the Taskforce on Integrated Assessment Modelling of the UNECE Convention on Longrange Transboundary Air Pollution (TFIAM/CIAM, 2007). Cost–Benefit Analysis (CBA) go a step further by expressing costs and benefits of policy measures in monetary terms. However, attaching financial values to for example improvement of human health and increased ecosystem protection is not without its challenges (Brink et╯al., 2011). This type of tool is generally applied at the science–policy interface. They are also used to assess uncertainties in the cause–effect relationships and in the effects of management measures. A fifth tool for integrated approaches to N management is ‘logistics and chain management’. This is the planning and management of activities, information and N sources in firms, installations and departments between the point of origin and the point of consumption. In essence, logistics and chain management integrate the supply and demand within and across companies. Logistics and chain management is especially important for N fertilizer producing companies, animal feed companies, transport and distribution sectors, processing industries, companies involved in recycling (sewage waste, composts, etc.), but also large farms. This type of tool is used especially by practitioners.

A sixth type of tool is stakeholder dialogue, including Multi Criteria Decision Analysis (MCDA), learning and participatory approaches. Evidently, this type of tool is indispensable for addressing the views of actors in N management issues (the 4th dimension of integration). The intention of stakeholder dialogue is to get people from different perspectives to enter a result-oriented conversation. Stakeholder dialogue is interaction between different stakeholders to address specific problems related to competing interests and competing views on how N and other resources should be used and managed. Rotmans (2003) describes the roles of stakeholders, networking, and self-governance in transition management. MCDA has been used in the water quality context and also in setting strategies for NH3 control in a wider context (including dietary change). It is a good way of involving different stakeholder interests and for dealing with uncertainties. Further, high-level meetings and resulting treaties are seen as a tool to achieve regional integration of N management measures. Regional integration is the most complex and encompassing way of integration. Also, there are many ways for and stages of regional integration, with not just one most superior outcome (in terms of ratification, exemptions, delayed implementation, etc.). This offers the opportunity of creating flexibility (Bull et╯al., 2011). Finally, integrated approaches to N management can be expected to have different policy targets than policies oriented toward single N sources and N species. Based in part on the critical-load concept and emission ceilings for N species developed under the CLRTAP Gothenburg Protocol, it is suggested that incentive-based N budgets and Nr ceilings per area, sector and or activity could be useful indicators, because they integrate multiple elements of N effects in the environment (see also Supplementary materials). The usefulness and analytical soundness of such indicators have to be further explored.

23.4╇ Developing integrated N management approaches further Integrated N management approaches have the potential of achieving various societal objectives and targets simultaneously more effectively and efficiently than disciplinary, single-issue approaches. The combination of achieving broader societal targets at a higher cost-effectiveness (without pollution swapping and/or other negative side-effects) is indeed a main driving force for developing integrated approaches. Such potential benefits may be achieved on the long term only, as there are possible disadvantages in the short term (see Supplementary materials). Like any other management approaches, integrated management approaches have to be also policy-relevant, �scientific-analytical sound, and political legitimate; i.e., acceptable and fair to users. The perceived greater effectiveness and efficiency of integrated approaches may also follow from a greater acceptability by and fairness to users. An integrated approach to N also holds the promise of decreasing the risks on inconsistency and pollution swapping. However, integrated approaches have higher demands as regards interdisciplinary cooperation and

545

Developing integrated approaches to nitrogen management

consensus building both in the domains of science, policy and practice (Tuinstra et╯al., 2006). As a consequence, integrated N management approaches may be more expensive, have lower initial adoption and have lower effectiveness, initially. By contrast, integrated approaches can be expected to lead to more efficient and cost-effective nitrogen management in the long term. There are various integrated N management policy measures existing in EU in practice already, though most of these may be considered as partial integration. The reform of the Common Agricultural Policy with cross compliance regulation, the National Emission Ceiling Directive (NECD) and the Water Framework Directive are examples of partial integrated approaches in current EU policy. It is therefore a remaining challenge to link and integrate the policies related to N in air (for example NECD) and those related to N in waters (for example Water Framework Directive and its related Directives) more intimately. The quantitative extent to which such increased linkage would yield a more optimal level of integration and increased cost-effectiveness needs to be further explored. Evidently, the challenge is to combine and coordinate those separate elements that provide the optimum level of integration, so as to ensure an organized and structured whole. Against this background, we first derive criteria (see below) and then consider actions for further integration of N management approaches (Section 23.5), while acknowledging that some integrated N management approaches are existing already (see Supplementary materials). The following criteria can be considered for identifying the most suitable ‘key actions’ (or intervention points) of an integrated approach to N management. • Any approach should consider that nitrogen is needed for food, feed and fiber production and that the production of food, feed and fiber is accompanied with diffuse losses of N to the environment. The beneficial effects of Nr use have to be weighted against the adverse effects to the wider environment. • Integrated approaches should lead to increased costeffectiveness of the policy measures from a societal point of view, and to less pollution swapping when compared to more disciplinary, single-issue approaches. • Strategies that decrease more than one form of Nr pollution are considered especially beneficial, such as those improving N use efficiency in food, feed and fiber production. • ‘The polluter should pay’ is a common principle in environmental protection and should hold also for integrated approaches to N management. Yet, the addressees of integrated approaches to N management may be different from those of more disciplinary, singleissue approaches. Addressing consumers’ behavior can be equally important as addressing producers’ behavior (see also Supplementary materials). • The optimal mix of instruments will depend on the objectives of the integrated management approach, the region-specific conditions and the rationale of the

546

addressees; Incentive based strategies appear more attractive to practitioners than regulatory instruments, as they provide greater flexibility to the practitioners (OECD, 2007). • The full suite of key actions must consider all major sources of reactive nitrogen and lead to appropriate sharing of efforts and benefits between the actors. • Any strategy should be subject to comprehensive stakeholder discussion to ensure buy in.

23.5╇ Key actions for integrated nitrogen management 23.5.1╇ Key actions for the global level Galloway et╯al. (2008) discussed four key actions for reducing Nr release to the environment at global level. They suggested that the following package of four measures could reduce total global Nr emissions by ~50 Tg per year, which is about onethird of the total estimated anthropogenic Nr release into the wider environment as of 2005. (1) Using best available techniques, Nr release from combustion may be decreased by ~18 Tg per year. (2) Increasing fertilizer N use efficiency may reduce fertilizer N use by 15 Tg per year. (3) Improving animal management strategies may decrease Nr release by ~15 Tg per year. (4) Improving conventional sewage treatment (especially in developing countries) may decrease the Nr release into surface waters by ~5 Tg per year, by converting it to N2. These suggested measures have great potential, but require also considerable efforts worldwide to implement these in practice. The measures may not threaten food and fuel securities, should not harm rural livelihoods and should not create a disproportionate large burden for a sector, group and/or area. This list only addresses part of the challenge in relation to emissions of Nr to the environment in Europe.

23.5.2╇ Key actions for major European sectors The European Union (EU) has various Directives and Regulations addressing N emissions from combustion, agriculture and households, and N deposition in terrestrial biosphere, air and waters (Oenema et╯al., 2011; Chapter 4, this volume). In reflecting on these, and the need to further develop a package for integrated management of N across sectors and actors in Europe, we identify the following seven key actions. These actions not only include technical and managerial measures for specific sectors, but also key actions related to patterns of societal consumption.

Agricultural sector Agriculture is by far the largest user of nitrogen, needed for the production of food, feed and fibre (Jensen et╯al., 2011, Chapter 3 this volume), and also the largest emitter of NH3 and N2O to air and of NO3 to groundwater and surface waters in EU

Oene Oenema

(de Vries et╯al., 2011; Leip et╯al., 2011, Chapters 15 and 16 this volume). It is also the sector that faces many different measures aimed at decreasing the losses of these N species (Oenema et╯al., 2011; Jarvis et╯al., 2011, Chapters 4 and 10 this volume). Given these arguments and the criteria above, it will be no surprise that agriculture is a central target for developing more integrated nitrogen management strategies in Europe. The following three key actions are suggested as the priority for European efforts to reduce N losses from agriculture. • Action 1:€Improving nitrogen use efficiency in crop production. Increasing crop yields through improving the genetic potential of crop varieties and improving soil, crop and N management, at similar Nr inputs, increases nitrogen use efficiency. Lowering Nr input through improved management and decreasing N losses, while maintaining crop yields has a similar effect. Strategies aimed at increasing nitrogen use efficiency decrease Nr losses per unit of produce, with minimal risk of pollution swapping (Tilman et╯al., 2002). Currently, there are no specific requirements, incentives or targets in EU agriculture to increase N use efficiency, though some policies contribute (Oenema et╯al., 2009). Where possible, such interventions should be combined with improving use efficiencies of phosphorus, water, pesticides, etc., in crop production. • Action 2:€Improving nitrogen use efficiency in animal production. Increasing animal productivity through improving the genetic potential of the animals, decreasing maintenance costs, and improving feed quality increases feed conversion efficiency and nitrogen use efficiency. Again, strategies aimed at increasing nitrogen use efficiency decrease Nr losses per unit of produce (Steinfeld et╯al., 2010). Similarly, there are currently no specific requirements or targets to improve animal nitrogen use efficiency in Europe. Where possible, such interventions should be combined with improving use efficiencies of phosphorus, micronutrients, antibiotics, etc., in animal production. • Action 3:€Increasing the fertilizer N equivalence value of animal manure. Farm animals excrete via dung and urine 60%–90% of the Nr (and other nutrients) in the animal feed. A large proportion of this Nr (up to 90%) can be used again for nourishing crops, but the reality is that on most farms only a relatively small fraction is re-utilized again. Increasing the fertilizer equivalence values requires conserving the N in the manure during storage and land application (especially requiring techniques to reduce NH3 emissions where a large fraction of the Nr is lost), while optimizing the rate and time of application to crop demand. Currently, there are no specific targets for the fertilizer N equivalence values in EU agriculture, apart from some Member States that have formulated target values in Nitrate Action Plans within the framework of the EU Nitrates Directive. Also, there are few requirements to use manure application techniques with low Nr emissions, although the Nitrates Directive and CLRTAP Gothenburg Protocol do recognize the need to use manure Nr efficiently. Further, phosphorus and other nutrients, including micro-nutrients like copper and zinc in manures, should be used efficiently too.

Energy, industry and transport sectors The energy, industry and transport sectors are by far the largest users of fossil energy and the largest emitters of NOx to the atmosphere. Significant reductions in NOx have been achieved during the last two decades (Erisman et╯al., 2011; Moldanová et╯al., 2011, Chapters 2 and 18 this volume), but the health impacts of NOx emissions in EU are still large (Brink et╯al., 2011, Chapter 22 this volume), and there is scope for further mitigation. • Action 4:€Low-emission combustion and energyefficient systems. There are many methods available to reduce NOx emissions from both stationary combustion sources and vehicles (Erisman et╯al., 2011, Chapter 2 this volume). At the same time, there are possibilities for increasing energy-efficiency and for alternative energy sources with less emissions. Although some of the emission abatement methods can increase emissions of NH3 and N2O, technological advances are reducing these tradeoffs. Overall, these advances represent an important technical success of existing policies that should be continued in the future. In addition, NOx emissions from shipping (currently ~ 20% of total emissions in EU) need to be addressed (Hertel et╯al., 2011, Simpson et╯al., 2011, Chapters 9 and 14 this volume). At the same time, there is a great need to increase the energy-use efficiency and to develop alternative energy sources. It is suggested to strengthen and to further integrate abatement strategies for NOx emissoins with strategies that increase energy use efficiency and develop clean energy sources. Steps are already being made under the ongoing revisions of the CLRTAP Gothenburg Protocol, the EU National Emissions Ceilings (NEC) Directive and the EU Integrated Pollution Prevention and Control (IPPC) Directive (Oenema et╯al., 2011).

Sewage treatment sector Sewage from households and industry is a major source of pollution of surface waters from Nr and phosphorus, in part because current sewage treatment is far from optimal. As noted above, Galloway et╯al. (2008) highlighted the global potential for increased waste water treatment, encouraging the conversion of anthropogenically produced Nr back to N2. Such treatment systems are already in widespread use across Europe, being based on removing Nr from the sewage through sequential nitrification and denitrification with di-nitrogen (N2) as main end product. However, this approach has the risk of significant N2O release, and also represents a waste of the energy used to produce Nr. Novel techniques are in development based in part on old practices. • Action 5:€Recycling nitrogen (and phosphorus) from waste water systems. There is potential for new sewage systems that recycle the nitrogen contained in the wastes, for use in crop production. These systems have to be further developed and tested (Sviriejeva-Hopkins et╯al., 2011, Chapter 12 this volume; see also Supplementary materials). Such approaches have the potential co-benefit of recycling phosphorus and can generate bio-energy at the same time

547

Developing integrated approaches to nitrogen management

(Magid et╯al., 2006). Over the next decades, where new sewage systems are planned (such as for new towns and cities) or major reconstructions of existing sewage systems become necessary, it should become a target to recycle sewage Nr rather than denitrify it back to N2.

Societal consumption patterns Society at large is ultimately responsible for the anthropogenic Nr emissions and can have a considerable influence on these emissions. Alteration of societal consumption patterns is also a most integral management measure as it harbours synergistic environmental effects (See Supplementary materials). Two key options emerge. • Action 6:€Energy and transport saving. Against the success of technical measures to reduce NOx emissions per unit consumption, both vehicle miles and energy use have increased substantially over past decades. Dissuasion of polluting cars and far-distance holidays, and stimulation of energy-saving houses and consumption patterns can greatly contribute to decreasing NOx emissions. The approach is fully aligned with efforts to reduce per capita energy use and CO2 emissions. Over the last decades, emissions per km have decreased (due to existing policy measures and advancements in technology), but the number of drivers and the number of km per driver have increased. • Action 7:€L owering animal protein consumption. Meat, milk and eggs contain high quality protein (amino acids), vitamins and micronutrients for humans. However, the current human consumption of protein from meat, milk, and eggs is far above the recommended per capita consumption in many parts of Europe. Lowering the per capita consumption to the recommended level will decrease the Nr release associated with the production of meat, milk, and eggs and will have positive human health effects where the current consumption is over the optimum (see also Winiwarter et╯al., 2011; Reay et╯al., 2011, Chapters 24 and 26 this volume). Moreover, the N cost of producing protein in milk, egg and poultry is much less than that in pork and especially beef, indicating that a shift from beef to pork to poultry and milk would also decrease Nr use (Steinfeld et╯al., 2010; see also Jarvis et╯al., 2011, Chapter 10 this volume). This list of seven key actions together provides a package that addresses the major sources of Nr to the environment. These are naturally not the only measures that should be considered. For example, better spatial and temporal planning approaches to landscape management, watershed management and airshed management (Cellier et╯al., 2011; Billen et╯al., 2011; Simpson et╯al., 2011, Chapters 11, 13 and 14 this volume) are needed and can contribute significantly. Such further actions will need to take account of patterns in each of the key societal threats of excess Nr:€ water quality, air quality, greenhouse gas balance, ecosystems and biodiversity, and soil quality (Grizzetti et╯al., 2011; Moldanová et╯al., 2011; Butterbach-Bahl et╯al., 2011; Dise et╯al., 2011; Velthof et╯al., 2011, Chapters 17–21 this volume). They will also need to consider the relative societal

548

costs of these threats and the future outlook (Brink et╯al., 2011, Winiwarter et╯al., 2011; Chapters 22 and 23 this volume). However, by focusing on the source of emissions, the seven key actions listed here provide the basic foundation for integrated Nr management.

23.6╇ Concluding remarks Integrated N management approaches have existed largely unconsciously since human beings have started to cultivate land and found out that the land does not provide food for free. Nevertheless, the specific request for developing integrated N management approaches emerged only two decades ago, following the increased awareness of the large emissions, its huge impacts, and the complexity and partial effectiveness of emission abatement measures. Indeed, the promise of integrated N management approaches lies in the increased effectiveness and efficiency of emissions abatement measures. An integrated approach to N also holds the promise of decreasing the risks on pollution swapping, but puts higher demands on interdisciplinary cooperation and consensus building both in the domains of science and policy. However, there is as yet little quantitative empirical evidence for these promises. Discussions about integrated approaches have often been confusing, in part because of a lack of clear concepts and definitions. This chapter provides a framework for developing and analyzing integrated approaches. It distinguishes five dimensions for developing integrated N management approaches. It builds on existing concepts and approaches. There is a need for improving and testing the framework further so as to achieve a better understanding of the effects of ‘integration’ on the effectiveness and efficiency of N management, and possibly a more insightful framework. Integrated approaches often have two opposite images. The one refers to a compact and fully integrated and smart approach, like a chip in electronics. The other refers to a complex approach, with many add-ons, that evokes the image of unclearness. Both images are indeed possible. The difference lies in the coalescence and organization, which makes the first effective as a functional whole and the latter potentially less effective. Clearly, integrated management approaches are demanding in terms of knowledge, organization, logistics, chain management and optimization. Given these issues, it is essential that approaches to integrated Nr policies take a stepwise approach, focusing on clearly identified priorities and an achievable level of integration at each stage. In this respect, the seven key actions identified here provide an important focus for the next stage of the societal and policy dialogue. Together they make the links between the five societal threats of excess Nr in the environment, and can be used to guide future mitigation and adaptation strategies.

Acknowledgements This chapter was prepared with the support of the NinE Programme of the European Science Foundation, the Nitro� Europe IP (funded by the European Commission) and the COST Action 729. The UK Department for Environment, Food

Oene Oenema

and Rural Affairs (Defra) and the Netherlands Ministry for Agriculture, Nature and Food Quality (LNV) are also thanked for their support of the UNECE Task Force on Reactive Nitrogen to which this work contributes.

Supplementary materials Supplementary materials (as referenced in the chapter) are available online through both Cambridge University Press: www.cambridge.org/ena and the Nitrogen in Europe website: www.nine-esf.org/ena.

References Billen, G., Silvestre, M., Grizzetti, B. et╯al. (2011). Nitrogen flows from European watersheds to coastal marine waters. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Brink, C., van Grinsven, H., Jacobsen, B. H. et╯al. (2011). Costs and benefits of nitrogen in the environment. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Bull, K., Hoft, R. and Sutton, M. A. (2011). Co-ordinating European nitrogen policies between directives and international conventions. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Butterbach-Bahl, K., Nemitz, E., Zaehle, S. et╯al. (2011). Nitrogen as a threat to the European greenhouse balance. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Cellier, P., Durand, P., Hutchings, N. et╯al. (2011). Nitrogen flows and fate in rural landscapes. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. de Vries, W., Leip, A. Reinds, G. J. et╯al. (2011). Geographic variation in terrestrial nitrogen budgets across Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University press. Dise, N. B., Ashmore, M., Belyazid, S. et╯al. (2011). Nitrogen as a threat to European terrestrial biodiversity. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. EC (2010). Water Framework Directive (WFD; 2000/60/EC). http://ec.europa.eu/environment/water/water-framework/ index_en.html EEA (1995). Europe’s Environment:€the Dobris Assessment. European Environment Agency, Copenhagen Erisman, J. W., van Grinsven, H., Grizzetti, B. et╯al. (2011). The European nitrogen problem in a global perspective. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Galloway, J. N. and Cowling, E. B. (2002). Reactive nitrogen and the world:€200 years of change. Ambio, 31, 64–71. Galloway, J. N., Aber, J. D. Erisman, J. W. et╯al. (2003). The nitrogen cascade. Bioscience, 53, 341–356. Galloway, J. N., Townsend, A. R. Erisman, J. W. et╯al. (2008). Transformation of the nitrogen cycle:€recent trends, questions and potential solutions. Science, 320, 889–892. Grizzetti, B., Bouraoui, F., Billen, G. et╯al. (2011). Nitrogen as a threat to European water quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press.

Hall, C. A. S., Lindenberger, D., Kummell, R., Kroeger T. and W. Eichorn, (2001). The need to reintegrate the natural sciences with economics. Bioscience, 51, 663–673. Hertel, O., Reis, S., Ambelas Skjøth, C. et╯al. (2011) Nitrogen turnover processes in the atmosphere. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Jarvis, S., Hutchings, N., Brentrup, F., Olesen, J. and van der Hoek, K. (2011). Nitrogen flows in farming systems across Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Jensen, L. S., Schjoerring, J. K., van der Hoek, K. et╯al. (2011) Benefits of nitrogen for food fibre and industrial production. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Kahn, P. H. (2001). The Human Relationship with Nature: Development and Culture, MIT Press, Cambridge, Ma. Leip, A., Achermann, B., Billen, G. et╯al. (2011). Integrating nitrogen fluxes at the European scale. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Magid, J., Eilersen, A. M., Wrisberg, S. and Henze, M., (2006). Possibilities and barriers for recirculation of nutrients and organic matter from urban to rural areas:€a technical theoretical framework applied to the medium-sized town Hillerod, Denmark. Ecological Engineering, 28, 44–54. Moldanová, J., Grennfelt, P., Jonsson, Å. et╯al. (2011). Nitrogen as a threat to European air quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Odum, E. P. (1996). Ecology: A Bridge between Science and Society, Sinauer Associates, Sunderland, Ma. OECD (1991). Environmental indicators: a preliminary set, OECD, Paris. OECD (2007). Instrument Mixes for Environmental Policy, OECD, Paris. Oenema, O., Kros, H. and de Vries, W. (2003). Approaches and uncertainties in nutrient budgets: implications for nutrient management and environmental policies. European Journal of Agronomy, 20, 3–16. Oenema, O., Witzke, H. P., Klimont, Z. Lesschen, J. P. and Velthof, G. L. (2009). Integrated assessment of promising measures to decrease nitrogen losses from agriculture in EU-27. Agriculture, Ecosystems and Environment, 133, 280–288. Oenema, O., Bleeker, A., Braathen, N. A. et╯al. (2011). Nitrogen in current European policies. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Reay, D. S., Howard, C. M., Bleeker, A. et╯al. (2011). Societal choice and communicating the European nitrogen challenge. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Rotmans, J. (2003). Transition Management: Key to a Sustainable Society (in Dutch), Koninklijke Van Gorcum, Assen, The Netherlands. Simpson, D., Aas, W., Bartnicki, J. et╯al. (2011). Atmospheric transport and deposition of nitrogen in Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Smil, V. (2001). Enriching the Earth, MIT Press, Cambridge, Ma. Steinfeld, H., Mooney, H., Schneider, F. and Neville, L. E. (eds.) (2010). In:€Livestock in a changing landscape:€Drivers, Consequences and Responses, Island Press, Washington, DC.

549

Developing integrated approaches to nitrogen management Sutton, M. A., Howard, C. M., Erisman, J. W. et╯al. (2011). The need to integrate nitrogen science and policies. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Svirejeva-Hopkins, A., Reis, S., Magid, J. et╯al. (2011). Nitrogen flows and fate in urban landscapes. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Tilman, D., Cassman, K. G., Matson, P. A., Naylor, R. and Polasky, S. (2002). Agricultural sustainability and intensive production practices. Nature, 418, 671–677. TFIAM/CIAM (2007). Reviews of the Gothenburg Protocol, MNP, Bilthoven, and IIASA, Laxenburg, Austria, The Netherlands. UNECE (1999). Protocol to the 1979 Convention on Long-range Transboundary Air Pollution to Abate Acidification, Eutrophication

550

and Ground-level Ozone. http://www.unece.org/env/lrtap/ multi_h1.htm Tuinstra, W., Hordijk, L. and Kroeze, C. (2006). Moving boundaries in transboundary air pollution:€co-production of science and policy under the Convention of Long-range Transboundary Air Pollution. Global Environmental Change, 16, 349–363. Velthof, G., Barot, S., Bloem, J. et╯al. (2011). Nitrogen as a threat to European soil quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press. Winiwarter, W., Hettelingh, J. P., Bouwman, L. et╯al. (2011). Future scenarios of nitrogen in Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al., Cambridge University Press.

Chapter

24

Future scenarios of nitrogen in Europe Lead authors: Wilfried Winiwarter and Jean-Paul Hettelingh Contributing authors: Alex F. Bouwman, Wim de Vries, Jan Willem Erisman, James Galloway, Zbigniew Klimont, Allison Leach, Adrian Leip, Christian Pallière, Uwe A. Schneider, Till Spranger, Mark A. Sutton, Anastasia Svirejeva-Hopkins, Klaas W. van der Hoek and Peter Witzke

Executive summary Nature of the problem • The future effects of nitrogen in the environment will depend on the extent of nitrogen use and the practical application techniques of nitrogen in a similar way as in the past. Projections and scenarios are appropriate tools for extrapolating current knowledge into the future. However, these tools will not allow future system turnovers to be predicted.

Approaches • In principle, scenarios of nitrogen use follow the approaches currently used for air pollution, climate, or ecosystem projections. Shortterm projections (to 2030) are developed using a ‘baseline’ path of development, which considers abatement options that are consistent with European policy. For medium-term projections (to 2050) and long-term projections, the European Nitrogen Assessment (ENA) applies a ‘storyline’ approach similar to that used in the IPCC SRES scenarios. Beyond 2050 in particular, such storylines also take into account technological and behavioral shifts.

Key findings/state of knowledge • The ENA distinguishes between driver-oriented and effect-oriented factors determining nitrogen use. Parameters that cause changes in nitrogen fixation or application are called drivers. In a driver-based approach, it is assumed that any variation of these parameters will also trigger a change in nitrogen pollution. In an effect-based approach, as the adverse effects of nitrogen become evident in the environment, introduction of nitrogen abatement legislation requiring the application of more efficient abatement measures is expected. This approach needs to rely on a target that is likely to be maintained in the future (e.g. human health). Nitrogen abatement legislation based on such targets will aim to counter any growth in adverse environmental effects that occur as a result of increased nitrogen application. • For combustion and industry, technical fixes for abatement are available. All scenarios agree in projecting a decrease in NOx emissions. Yet agricultural nitrogen use is expected to remain the leading cause of nitrogen release to the environment, as options to reduce emissions are limited. Thus, major changes will occur only if the extent of agricultural production changes, which may possibly be triggered by decreasing population numbers in Europe. The scenarios presented here project modest changes in NH3 and N2O emissions, or nitrate leaching, but do not agree on the direction of these changes. • Agricultural activity (and thus nitrogen loads to the environment) may decrease strongly if the European population adopts a healthier ‘low meat’ diet leading to lower nitrogen losses related to animal husbandry. Change to a ‘healthy diet’ across the EU, which consists of 63% less meat and eggs, would reduce ammonia emissions from animal production by 48%. However, if an agricultural area previously used for animal feed production is utilized for biofuel crops, additional nitrogen fertilizer may be required, which will partially offset reductions of nitrogen leakage to the environment.

Major uncertainties/challenges • International trade in nitrogen-containing goods (agricultural as well as industrial) represents a key uncertainty and is difficult to project. Estimating the demand for such goods for Europe alone may not at all reflect European production and related environmental effects. The industrial use of nitrogen is also very poorly understood, but it is expected to continue to grow considerably. The respective environmental impacts of such products cannot be clearly discerned from statistical information.

Recommendations • Scenarios need to be continuously updated in terms of economic, technical, and societal trends to reflect improved understanding of these factors. Using nitrogen budgets as tools could improve the consistency of scenarios.

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

551

Future scenarios of nitrogen in Europe

24.1╇ Introduction Scenarios developed for use in policy processes are intended to provide policymakers with insights into various possible futures based on a current understanding of the issues involved. Scenarios do not predict the future, but are made to provide decision support. Environmental scenario analysis typically addresses questions such as ‘What will be the impact on the environment or human health in, say, 2050, of policy alternatives implemented in a base year (e.g. 2020)?’. In this chapter scenarios referring to external trends are called ‘driver-oriented’. Conversely, scenarios can also be designed to identify the policies required to achieve predefined targets in the field of human or environmental health, referred to here as ‘effect-oriented scenarios’. Economic and technological considerations form the basis of driver-oriented scenarios, while effect-orientated scenarios use an environmental target to arrive at a particular future situation. Scenarios may or may not capture developments which, with hindsight, seem evident. However, as demonstrated in integrated assessment modeling (Hettelingh et╯al., 2009), they provide the only systematic approach to considering the possible consequences of policies for future events. In Europe, a few scenario approaches have been developed that deal with the future fate of nitrogen (N) in the environment. We introduce and discuss several of these scenarios here. Obviously, the rate of change, and especially the probability, of ‘technology leaps’ (i.e., drastic changes beyond an extrapolation of the current development) increase when observed over an extended time range. The strategy for creating such a scenario may thus also change. This chapter attempts to identify and quantify the most important influences on future developments in relation to application and release of N into the environment. It also touches on the uncertainty involved in any such expectation, recognizing that the lack of knowledge regarding uncertainties is even greater than that related to the scenarios themselves. Comparison of the different sets of available data indicates the range needing to be considered. However, as abrupt and unexpected changes are not covered at all, they obviously cannot show up in an assessment of uncertainty. Here, rather than developing and inventing new approaches or scenarios, we will review the material for Europe that already exists. Figure 24.1 illustrates the multiple interactions between economic activities, the release of trace compounds and their subsequent fate in the environment, the adverse effects of the compounds, and the emission/release control targeted to resolve the adverse situations. Boxes shaded green refer to processes directly linked to N; arrows indicate only the most important pathways within the ‘nitrogen cascade’ (Galloway et╯al., 2003). The ‘driver-oriented’ scenario uses information on economic activities and attempts to predict the environmental effects; the ‘effect-oriented’ scenario defines targets for environmental protection and then identifies emission control and activity numbers compatible with that target level. Based on these interactions, we will first describe the conÂ� cept€ of driver-oriented scenarios (Section 24.2) and effectÂ�oriented scenarios (Section 24.3). Their potential use is disÂ�cussed

552

for short- and medium-term scenarios (Section 24.4), as are their implications for long-term scenarios (Section 24.5). Section 24.6 concludes with the results from a number of scenario activities available for the European Union (EU).

24.2╇ Driver orientation 24.2.1╇ The concept of â•›‘drivers’ The extent of an economic activity and the changes in this activity over time can often be characterized by the parameters underlying that activity. If those parameters also describe a causal relationship, they are called ‘drivers’. Ideally, statistical information on drivers would be easily accessible, as would information regarding their future development; thus, a change in their scope would lead to a proportional or at least predictable change in the economic activity in question. Drivers of environmentally relevant activities may be classified according to their proximity to describe the respective economic activity. For example, Nowicki et╯al. (2006) distinguish exogenous drivers (demography, macroeconomic growth, consumer preferences, and technological development) from policy-related drivers (agricultural, trade, or climate policies). Likewise, one can distinguish primary drivers from secondary drivers, with the primary drivers (e.g. population numbers) influencing secondary drivers (e.g. food production). For the purposes of this chapter, it may be useful to describe drivers with regard to specific topics (Table 24.1). These topics, each covering several drivers, will be covered in more detail in the sections below. For the scenarios, we distinguish strictly between drivers and policy action, with policy action seen as affecting the release of adverse substances over and above the developments initiated by the drivers.

24.2.2╇ Population Food protein is essential to human metabolism. One of the constituents of protein is nitrogen, which is found in agriculturally produced food or in feed crops that become partially converted into animal products. The number of people requiring food determines agricultural demand. Population numbers for European countries can be predicted with quite high confidence. Statistical lifetimes are stable and, compared with the timescales of the scenarios in question, are long. Reproduction rates, which are dependent on fertility rates and the size of the female population in certain age groups, are also well known. Even fertility rates, which affect population numbers over a longer timescale, change only gradually in stable societies. Population forecasts, a key element in any public and governmental planning, are the subject of detailed study. For Europe, UN (2004) statistics predict a population of 630 million in 2050 and 540 million in 2100, down from 730 million in 2000. These numbers include assumptions regarding migration, which is much more difficult to estimate, as demonstrated by the example of Spain where the population grew unexpectedly from 40 million in 2000 to 46 million in 2008 (EUROSTAT, 2009), mainly as a result of migration.

Wilfried Winiwarter and Jean-Paul Hettelingh

Driver orientation Economic activities

Emission control policies

Targets

Agriculture

NH3 control (& costs)

NH3 emissions

NH3 dispersion

Critical levels (ammonia)

Energy use

SO2 control (& costs)

SO2 emissions

S dispersion

Critical loads (acidification)

NOx control (& costs)

NOx emissions

NOx dispersion

Critical loads (eutrophication)

VOC control (& costs)

VOC emissions

O3 formation

Critical levels for ozone

PM control (& costs)

Primary PM emissions

GHG control (& costs)

GHG emissions

Transport

Solvents, fuels, industry Other activities / households

N removal: sewage treatment fertilizer balance (& costs)

Secondary PM formation Primary PM dispersion

Ozone exposure of population PM exposure of population

GHG dispersion Global warming Runoff/leaching -- dissolved N -- particulate N

Emission control costs

Denitrification, transport along waterbodies

Excess soil N Nitrate in waters

Effects Effect orientation

Figure 24.1 A schematic overview of the links between drivers and effects of environmental trace constituents in connection with nitrogen (boxes in green shading are directly linked to N). Table 24.1 Drivers€– important parameters for determining environmental nitrogen load

Topic

Driver

Population

Population number/change (birth rate, migration) Food choice (driven by income, health, dietary preference specifically regarding meat)

Land use

Cropland area Land loss to urbanization Loss to desertification, erosion, climate change Biofuel production

Fertilizer application

Agricultural demands (production) Agricultural technology/management Energy price and fertilizer price

N in combustion and industry

Industrial uses of Haber–Bosch nitrogen Reactive nitrogen in combustion

Markets

Trade liberalization (import/export) Global trade (‘nitrogen leaking’)

The nutritional habits of the population are also relevant. Europeans in general receive sufficient nutrition; food is available in abundance and, in favorable economic circumstances, frequently wasted. For Toronto, Canada, it has been estimated

that about 30% of all food available at retail level is not consumed (Forkes, 2007), which also means that roughly a third of all agriculture, and thus a third of the nitrogen applied to agricultural soils, is wasted (assuming no difference in protein share

553

Future scenarios of nitrogen in Europe Table 24.2 Land use impacts on the nitrogen cycle

Land is required for:

Relevance for the N cycle:

Urban areas:€housing, living, working

Release of reactive N in industrial and domestic effluents (air pollutants, sewage)

Water for recreation, fishing and water supply

Receptor of long-range transport N (deposition) and/or from run-off of NO3

Nature, recreation, biodiversity

Receptor of long-range transport N (deposition)

Carbon sequestration

Influenced by nitrogen deposition and agricultural management (fertilization)

Agriculture:€crops, livestock

Multiple use and losses of reactive N to different compartments

Transport of water, food, goods, people

Emission and transport of reactive N

Resources:€energy, raw material, fiber

Uses and emissions of reactive N

between waste and used product). Similar but somewhat lower figures are available for the United States (27%:€Kantor et╯al., 1997) and Sweden (20% in food service institutions:€Engström and Carlsson-Kanyama, 2004). Food wastage in Europe has been discussed in more detail by Resy et al. (2011, Chapter 26 this volume) as a ‘societal choice’. These authors arrive at very similar conclusions with food wastage at a levels of 20%–30% of purchased food. The choice of human diet plays an important role. The question of the conversion efficiency of meat production in terms of nitrogen has been covered in great detail by Smil (2002). Animal protein is available in the form of meat, milk, and eggs. Conversion efficiency varies strongly among different meat types. In the food chain, losses of protein are highest for beef (4% efficiency, or 96% loss), and lowest for milk (60% loss). Although some amino acids essential for human metabolism are not available from plant material, plant proteins offer considerable efficiency advantages. In terms of nitrogen losses, a vegetarian diet that includes milk and eggs is more efficient than a meat protein-based diet.

24.2.3╇ Land use and urbanization Land use determines the activities that result in losses of nitrogen to the environment. In principle, multiple land use (i.e., using the same piece of land for more than one purpose) is possible. Examples are the use of forest land for recreation and timber production, or of water bodies for fishing and water supply. However, in some cases, multiple use is excluded. For example, carbon sequestration cannot usually take place in the same area as intensive agriculture; the same piece of land cannot be used for food production/grazing of livestock and biofuel plantations, and this leads to ‘competition for land’. This competition needs to be recognized and can be used as a constraint in projections. Table 24.2 lists important categories of land use and their implications with regard to nitrogen. Land use itself is influenced by a multitude of drivers. Parameters like population numbers, macroeconomic growth, urbanization, globalization, resources (biodiversity, energy, water, soil and air) determine how people use the land, how it is managed, and in consequence how this relates to N. For instance,

554

the global increase in population density from 2.2 persons per ha of agricultural area in 1960 to 4.1 in 2006 increased urbanization, agricultural intensity and fertilizer application (Erisman et╯al., 2008). Moreover, higher meat consumption necessarily means an increase in the production of animal feed, resulting in changÂ� ing land use and N cycles. Both factors are of higher importance globally than for Europe. Most recently, biofuel production has been recognized as an important parameter for change in land use and N-cycling (Erisman et╯al., 2010; Howarth et╯al., 2009). Several studies have been conducted to determine how much land will be available for agriculture and bioenergy production, and how much for other purposes (Fischer et╯al., 2010b; Lysen et╯al., 2008; Londo et╯al., 2010). Such data are important for estimating N losses, storage, and recycling. Figure 24.2 presents results obtained from the REFUEL project (Londo et╯al., 2010; Fischer et╯al., 2010a). The increase in urban area resulting from urbanization (larger number of people in large cities) and urban sprawl (increased area requirement per urban inhabitant) compete against other land uses and affect the nitrogen cycle in various ways. As many human activities are focused in cities, cities will generate a considerable share of anthropogenic pollution (see SvirejevaHopkins et╯al., 2011, Chapter 12, this volume). City systems can evolve as a result of population growth, by increasing in either size (occupied land) or complexity (density). Each type of evolution has its own implications for what type of pollution dominates and for the fluxes of reactive nitrogen as they affect transport, heating/cooling requirements, or soil interactions. Modeling population density distributions (Svirejeva-Hopkins and Schellnhuber, 2008) helps potential trends to be understood. Total city areas to 2050 will have grown by only a few percent compared to the current city areas; especially affected will be cities with a population of 1–2€ million (Svirejeva-Hopkins et╯al., 2011, Chapter 12, this volume). Significant differences remain between different parts of Europe, especially regarding size (larger in Western Europe) and core densities (significantly higher in Eastern Europe). The most important changes with regard to nitrogen should be expected from changes in the extent of built-up area and associated urban sprawl (see Fischer et╯al., 2010b).

555

ENERGY YIELDS GJ ha–1 biofuel equiv. 9 - 40 41 - 60 61 - 80 81 - 100 101 - 120 121 - 140 141 - 160 161 - 180 181 - 200 201 - 242

Figure 24.2 Yield expectations for specific crops (here:€energy crops). Left panel shows the attainable energy yields for first- generation biofuels (starch crops, sugar crops and oil crops), right panel for second-generation biofuels (woody and herbaceous lignocellulosic feedstocks). Data from the EU REFUEL project, www.refuel.eu (Fischer et╯al., 2010a, reprinted from Biomass and Bioenergy, with permission from Elsevier).

ENERGY YIELDS GJ ha–1 biofuel equiv. 9 - 40 41 - 60 61 - 80 81 - 100 101 - 120 121 - 140 141 - 160 161 - 180 181 - 200 201 - 242

Future scenarios of nitrogen in Europe

Urban areas are particularly relevant in terms of assessing population exposure to pollution. As urban areas are both emission hotspots and areas of high population density, a considerable part of the overall exposure depends on urban development. Larger population densities at typically higher pollution levels appear in urban areas of Eastern Europe. Scenarios on human exposure need to include consideration of spatially diverse pollution as well as respective population density.

24.2.4╇ Fertilizer application The application of mineral fertilizer or manure to fields represents a direct release of N into the environment. While, under ideal situations, a crop can access a high proportion of the available nutrients, in certain circumstances nutrient dispersion to the environment cannot be fully constrained. OECD (2008) data show that the fraction of nitrogen captured in crops increased between 1990–1992 and 2002–2004 from 51% to 59% for the EU (year 2000 borders). Factors influencing nitrogen application range from energy prices (affecting also fertilizer prices) and crop prices to policy (adoption of nutrient management plans). As the amount of agricultural produce is determined by demand, the extent to which soil nitrogen needs to be replenished by fertilizer is similarly affected. For its members, the European Fertilizer Manufacturers Association (EFMA) develops a fertilizer use projection, which is also made available to the policy process (EFMA, 2007). This work is based on the development of scenarios for crop demand and the respective fertilizer application rate as a top-down activity starting from the total EU (EU scenario) broken down to countries. The information is then complemented by detailed national information collected by designated EFMA national experts. The national dataset contains information on national agricultural and environmental policy, specific national/regional limitations or programs, and information from agriculturespecific media. National information is used to feed back to the European level in terms of nutrient demand. Data collection applies a time perspective of up to 10 years into the future. For 2017 a slight increase in mineral fertilizer use of 3.6% to 11 Tg N is expected in the EU27; increases will mainly affect new EU member states. EFMA not only uses this data as guidance for manufacturers but also exchanges and discusses results with a number of European and global agencies and institutions to further increase the robustness of the data. These estimates take into account, but do not include, nutrients from manure application, which provides roughly two-thirds of N from mineral fertilizer (Leip et╯al., 2011, Chapter 16, this volume). Future application of manure is usually estimated from the total nitrogen excretion of farm animals, whose numbers are determined by human population numbers and diet choices.

24.2.5╇ Reactive nitrogen in combustion and industry Ammonia from the Haber–Bosch process is the starting point for production of the N used in most industrial processes (Domene and Ayres, 2001). Ammonium salts such as

556

ammonium chloride and ammonium sulfate are used mainly as fertilizers. Nitric acid and nitro chemicals are used in the making of dyes, rubber, herbicides, pesticides, plastics like urethanes, and explosives like nitroglycerine and TNT. Urea is mainly used as a fertilizer and as a component of animal feed, but it is also used in resins like melamine, a thermosetting plastic used in laminates, cooking utensils, electrical appliances, and insulators. Hydrogen cyanide and its derivatives are used to make nylon, thermosetting plastic resins, synthetic rubber resins, acrylic fibers, and epoxy resins. Further products include dyes and drugs. The industrial uses of N continue to expand. Recent global trends in the industrial uses of N are important for determining a mass balance of N, also because the uses of industrial N are growing. The total worldwide demand is expected to increase from 125 Tg N in 2007 to 142 Tg N in 2013 (IFA-PITCom, 2009). Global industrial (non-fertilizer) demand is projected to increase from 23 Tg N in 2007 to 28 Tg N in 2013, underscoring the importance of understanding the uses of industrial N. About 4.5 Tg of industrial N will be used in Western Europe in 2009, 0.7 Tg N in Central Europe, and 1.5 Tg N in the East Europe and Central Asia region, totalling about 6.7 Tg N for industrial processes in Europe. It should be noted that industrial N uses will account for a growing share of total N demand in Europe. In 2007 industrial N usages accounted for 30% of all N consumed in Western and Central Europe. This share is projected to expand to 34% in 2013 (Prud’homme, 2009). The fate of the N used in industrial processes is not well understood. As more industrial uses for N are discovered, its importance and prevalence continue to grow worldwide. Mass balances for industrial N uses, however, are deficient because of the limited data available and the lack of powerful process simulation tools (Domene and Ayres, 2001). Additional lifecycle analysis of N use in industry is needed to better understand its contribution to N emissions and its ultimate fate in the environment. Only on such a basis the parameters describing changes in N use in industrial processes over time can be assessed. Reactive nitrogen compounds are also created as Â�by-products of combustion. Formation of NOx, a common combustion byproduct, depends on the N content of fuel and on combustion temperatures. The high temperature processes that create reactive nitrogen compounds are found primarily in power plants for electricity generation and industrial use, and internal combustion engines in the transport sector. Global formation of reactive nitrogen over this pathway has been estimated at 13% of anthropogenically created reactive N (Galloway et╯al., 2008) immediately released to the environment. For Europe, excluding Russia, a similar share has been estimated (3.3 Tg NOx-N or 13% of the 25 Tg N total:€van Egmond et╯al., 2002; 5 Tg industrial N production were also considered in estimating the total), but is expected to be somewhat larger for the EU27 (3.7 Tg NOx-N or 17% of the 21 Tg total:€Leip et╯al., 2011, chapter 16, this volume). Differences in European figures, especially regarding the larger NOx emissions from the smaller group of countries, also derive from different base years. More energyefficient (high temperature) installations, a higher share of

Wilfried Winiwarter and Jean-Paul Hettelingh

diesel cars, and transport expansion in general are drivers of a notable increase in NOx emissions. For all these sources, abatement technology has become available and is being applied successfully, such that future emissions are expected to decrease despite increased activity (see Section 24.6).

24.2.6╇ Nitrogen and international trade The demand for nitrogen products derived from population and dietary trends, energy production, and industrial products can be satisfied by production within Europe as much as by imports. Moreover, Europe exports goods containing nitrogen and also goods whose production chain requires nitrogen but contain no nitrogen themselves. Understanding trade and its function in replacing or enhancing European production may become very important in terms of assessing its nitrogen�related impacts on the environment. Global trade, especially of agricultural goods, is strongly related not only to economic but also political conditions. Trade in the global nitrogen market is considerable already under present conditions (Galloway et╯al., 2008). Trade is expected to be further liberalized, which will affect its quality and extent in the future. As trade liberalization may also affect critical infrastructure, a considerable amount of work has been done to understand the future situation (see e.g., the status of the Global Trade Analysis Project (GTAP, 2010). GTAP also liaises with IPCC work).

24.3╇ Effect orientation 24.3.1╇ Basic concepts Oxidized and reduced nitrogen compounds transported through the atmosphere cause a broad sweep of adverse effects (the ‘Key societal threats of nitrogen’ described by the ENA). These effects inevitably give rise to remediation efforts. An effect-oriented scenario starts from an expected or desired future situation (target), which keeps adverse effects to an acceptable limit. Interpolation and/or expectations on economic and technological development may serve to identify the pathway to this expected future (‘backcasting’). Effect orientation is important, as it is an independent approach complementary to driver orientation. It assumes policy responses to resolve adverse effects. If the action of a driver has been underestimated, environmental policy will have to react with even stricter abatement measures to achieve environmental targets. A critical element in developing effect-oriented approaches is the setting of targets that are likely kept in the distant future (e.g. as mentioned above, human health). The targets need to be established such that an adverse effect that is known now€– and will be considered sufficiently adverse and important to be maintained in the future€– can be remedied. In the sections below we present targets that we consider to be stable even over long time periods and thus to be amenable to consistent environmental policy interventions. Furthermore, as a part of scenario development, some concept regarding feasibility is needed. This requires the release

(emissions) of environmentally relevant compounds first to be identified, and then for these emissions to be coupled to the adverse effects they cause. Such a task can be performed using models that capture the entire cycle between release and impact of compounds€ – see Figure 24.1 and De Vries et╯al. (2011, Chapter 15, this volume) for more information about ‘integrated assessment models’. specific technology measures need to be identified that are actually able to achieve the abatement required. Finally, when investigating interrelated environmental effects that are typical for nitrogen, cross-influences also need to be considered. Environmental policies and abatement technologies that address nitrogen should not be looked at in isolation (Oenema et╯al., 2011, Chapter 4, this volume). In the practical development of effect-based scenarios, it is quite common for a set target not to be fully achieved, even where policy development timescales are long. The basic concept of including abatement can, however, continue to be maintained; in this case, the limiting factor is not the target but rather the extent of available (known) technical measures for reduction. Terms used to describe such scenarios are ‘environmentally considerate’ or, more specifically, ‘maximum feasible reduction’. In addition to setting the conditions at a given point in time, some scenarios also attempt to provide the pathway leading to such a target. This procedure, starting from an effect-based target towards the current situation, is usually called backcasting (Carlsson-Kanyama et╯al., 2008).

24.3.2╇ Setting targets for future development Targets set for scenarios need to remain stable over the timescale of the scenario (see Table 24.3 for targets related to nitrogen). Obviously, targets for human health are the most easy to define and maintain, as human health is generally regarded as having a high value, a status which it will plausibly retain in the long term. This includes any direct nitrogen-based effects on human health. Further risks to human wellbeing that are indirectly related to nitrogen are becoming increasingly important, partly because of improvements in scientific knowledge. Several of these indirect risks affect biodiversity and interact with other environmental issues. The WHO prepared a report as a contribution to the Millennium Ecosystem Assessment on the issue of ecosystems and human wellbeing (Corvalan et╯al., 2005). Following the concepts of the Millennium Ecosystem Assessment, a first inventory of the effect of nitrogen on ecosystems services in Europe was produced (see Appendix C of Hettelingh et╯al., 2008). Ecosystem services describe the benefits to human wellbeing delivered by functioning ecosystems. The ability of ecosystems to perform such services can also be used as a robust target. We may specifically differentiate subtypes of ecosystem services. Provisioning services refer to ecosystems that help provide food, fiber, and fuel to humans. Regulating services refer to the potential of ecosystems to maintain the quality of air, water, or soil, and also the climate. Cultural services relate to recreational and aesthetic aspects of

557

Future scenarios of nitrogen in Europe Table 24.3 Quantifiable targets to avoid environmental deterioration due to nitrogen

Environmental problem

Nitrogen connection

Target

Loss of ecosystems services due to: •  acidification •  eutrophication •  biodiversity loss

Atmospheric deposition of nitrate and/or ammonia

Critical loads (exceedance of soil specific values)

Adverse impact on plant species composition

Atmospheric concentration of ammonia

Critical levels (exceedance of critical NH3 concentrations)

Crop damage

O3 formation (NOx chemistry)

Accumulated exposure over threshold (AOT)

Human health

NOx concentration O3 formation (NOx chemistry) PM formation (NOx chemistry and ammonia) Nitrates in drinking water

Years of life lost (YOLLs) Disablement Adjusted Life Years (DALYs) Quality standard (concentrations)

Climate change

N2O emissions CO2 emissions/uptake, as the natural C cycle is influenced by N in the environment O3 formation (NOx chemistry)

Global mean temperature (limit:€2 °C increase above pre-industrial)

ecosystems. ‘Critical loads’ describe a threshold of pollution beyond which the functioning of ecosystem services may be compromised. An important tool for addressing future targets is dynamic modeling of geochemical and biological processes that drive changes in vegetation (De Vries et╯al., 2010a). Dynamic modeling is especially important, as it is something of a tall order to avoid critical load exceedance. When the critical load targets cannot be met, dynamic modeling can be used in the context of integrated assessment to analyze the consequences for soil chemistry and vegetation in the future (see e.g. Posch et╯al., 2008). Thus, effects-based targets can serve as ex post constraints of driver-based scenarios.

24.3.3╇ Critical loads for eutrophication€– sample application The critical load for eutrophication is the maximum allowable input of nitrogen below which adverse effects to the structure and functioning of natural ecosystems do not occur, according to current knowledge. According to current computations, the percentage of the European natural area at risk of eutrophication in 2000 was 77% (as demonstrated by Dise et╯al., 2011, Chapter 20 this volume). Driver-oriented scenario analyses (Hettelingh et╯al., 2008) show that current policies (current legislation€ – CLE) would reduce the area exposed to risk to 67% of the natural area in 2020. This percentage could be further reduced to 31% if the best available control technology (maximum feasible reduction€– MFR) were applied in the agriculture and energy combustion sector to reduce emissions of ammonia and nitrogen oxides. While MFR is still not able to remove all risks, it comes closest to an effect-based approach that would be able to do so. Figure 24.3 illustrates how these policies affect the magnitude of critical load exceedance and the areas at risk of eutrophication in 2020. Hotspots of exceedance are the border between the Netherlands and Germany as

558

well as Belgium, north-western France, and the Po Valley in Italy. All these hotspots show significant improvements with the application of appropriate measures.

24.3.4╇ Abatement costs If we expect the environmental effects to determine the future release of nitrogen, then we must also expect to initiate specific abatement measures and accept to bear the costs. Integrated assessment models help prioritize measures so that cost-efficient remedies can be implemented. Many of the ecosystem service functions outlined above may also be considered in terms of their financial benefits. Comparing these benefits with the costs of remediation measures may be an efficient way of formulating agricultural policies on the management of the nitrogen cycle. Welfare optimization (see Brink et╯al., 2011, Chapter 22 this volume) is one modeling approach to provide maximum benefit to society and also to demonstrate that the cost of measures is turned into a net value for a community. In principle, considering the negative effects on the environment (and their cleanup costs) will make it easier to justify abatement costs. In view of today’s problems with N in the environment and the current exceedance of thresholds, from an effect perspective, it can be argued that a further increase in N exposure in Europe over large areas should not be expected, as long as environmental benefits maintain their high value in the eyes of the public.

24.4╇ Short- and medium-term scenarios 24.4.1╇ Timescale for scenarios Both driver-based and effect-based approaches are used to develop scenarios. For emissions of air pollutants, agricultural and industrial production, land use and nitrogen inputs like

Wilfried Winiwarter and Jean-Paul Hettelingh Exceedance of nutrient CLs eq

CLE 2020

ha–1a–1

Exceedance of nutrient CLs eq

no exceedance < 200 200 - 400 400 - 700 700 - 1200 > 1200

MFR 2020

ha–1a–1

no exceedance < 200 200 - 400 400 - 700 700 - 1200 > 1200

CCE Dep-data: EMEP/MSC-W

CCE Dep-data: EMEP/MSC-W

Figure 24.3 Exceedance of critical loads for eutrophication expected in 2020. The map on the left presents a driver-based projection of national emissions based on ‘Current Legislation’; at the right the results of an effect oriented approach is shown, assuming Maximum Feasible Reductions as an option of getting closest to the target of no exceedance (source:€Hettelingh et╯al., 2008).

fertilizers, scenarios are often limited to 2020, and have only recently started to be extended to 2030. For this time period, any emission abatement technology implemented is expected to be already available, at least at an experimental stage. Projections thus need not consider new unknown technologies. The time horizon of 10–20 years is considered short-term. Moreover, the most recent agricultural projection of the Food and Agriculture Organization of the United Nations (FAO, 2003) covers the period 2000–2030; a more recent outlook for 2050 is available, although with less geographic detail than the 2030 projection. Agricultural systems will show dramatic changes in these five decades, according to FAO. The changes are related to various developments, including population, changing markets and trade, technology, and diets. In scenarios for analyzing the acceleration of the nitrogen cycle, agricultural developments are the major drivers. There is a need to go beyond this scale. Most scenarios constructed for analysis of drivers of climate change consider a 100-year time period. To exhibit the potential co-benefits of expected measures with respect to air pollution, air pollutants have also started to be looked into. A recent workshop organized under the UNECE Convention on Long-range Transboundary Air Pollution (‘Workshop on non-binding aspirational targets for air pollution for the year 2050’ held in Utrecht, March 2009) called for an extension to 2050. This 40-year timescale will be covered in medium-term scenarios. The IPCC ‘Second Report on Emission Scenarios’ (SRES:€Nakicenovic et╯al., 2000) for projecting CO2 emissions uses consistent storylines to illustrate possible future situations. The scenarios of the Millennium Ecosystem Assessment (MA, 2005) were built based on the SRES concepts. For the Millennium Ecosystem Assessment, information on nitrogen

does play an important role, and nitrate leaching and transport in watersheds is also covered. Because of large uncertainties in non-energy-related fields the temporal extension is limited to 2050. Scientific evaluation is still going on, and important results have recently been published (Bouwman et╯al., 2009; Van Drecht et╯al. 2009; Seitzinger et╯al., 2010). Any scenario beyond the year 2050 is then considered a long-term scenario. Such scenarios typically are being prepared in the context of climate change considerations. Important examples of scenarios are presented in Table 24.4. Sets of scen� arios contain elements of both driver and effect orientation.

24.4.2╇ Basic approach€– Short-term scenarios Over a timescale of 10–20 years, a generally good understanding of the systematic relationship and interaction between economic development and environmental effects can be expected. Thus for short-term scenarios it is useful to prepare a ‘baseline’ path of development, which covers the most probable expectations at the time of inception of the scenario. This is based on a consistent set of assumptions used in a similar manner for different applications. Typically, any foreseen future changes are already considered. This includes, for example, emission abatement legislation already in place but effective in the future (scenario ‘with measures’). Despite the fact that such a path may need to be adapted to a changing reality (e.g. economic/technological development), relying on a single scenario is advantageous for comparing both measures and consequences. To understand the system response to certain, fixed interventions, the ‘baseline’ scenarios are often extended by certain, given measures (scenario ‘with additional measures’) which are not yet part of mainstream expectation (legislation or regulation). It is the common understanding that, even if development

559

Future scenarios of nitrogen in Europe Table 24.4 Overview on important scenario activities related to the future release of nitrogen to the European environment. See text for more detailed description

Activity

Reference

Remark/description

European Consortium for Modelling of Air Pollution and Climate Strategies EC4MACS1

EC4MACS (2010)

Short term:€‘current legislation’ vs. ‘maximum reduction’ approach, limited to the atmosphere as carrier.

IPCC second report on emission scenarios SRES2

Nakicenovic et╯al. (2000)

Long term:€four main storylines. Limited to energy and the carbon cycle

Millennium ecosystem assessment3

MA (2005)

Medium term, four storylines similar to SRES; extends to nitrogen, covers water and ecosystems

Eururalis

Eururalis (2010) Rienks (2008)

Short term, focus on land use, agriculture and related activities; an activity on mid-term scenarios is under way; four storylines

OECD environmental outlook

Bakkes et╯al. (2008)

Short-term perspective, many results extended to medium term; covers air, water pollution and agriculture; scenarios are driveroriented vs. climate- (effect-) oriented

Representative Concentration Pathways (RCP)

Representative Concentration Pathways (2010)

Long-term, with an ambition to 2300; extends SRES to more compounds (including nitrogen compounds), but excludes water, ecosystems

(1) This scenario approach has been used for European air quality policy and is reflected in ENA by Moldanová et╯al. (2011, Chapter 18, this volume) and Butterbach-Bahl et╯al. (2011, Chapter 19, this volume). (2) Emission scenarios have been used as input to climate models for climate projections; the impact of climate change on aspects of the nitrogen “key threats” is investigated by Grizzetti et╯al. (2011, Chapter 17, this volume) and by Butterbach-Bahl et╯al. (2011, Chapter 19, this volume). (3) Used also by Grizzetti et╯al. (2011, Chapter 17, this volume).

does not follow the baseline projection, at least the difference between the baseline and the ‘with additional measures’ scenario will be maintained. This will be the case when all changes considered are merely incremental to the major drivers of change, and the majority of uncertainty in future development is due to these major drivers rather than to the increment. The approach has been extensively applied in modeling European energy policy and the environmental applications derived from such energy models. Specifically, the PRIMES energy model works on that basis, providing a baseline of energy consumption and some industrial activities. These results are used as input by the GAINS model (http:www.gains. iiasa.ac.at) which links it to emission abatement measures, their abatement potential and applicability, and their costs, as well as to environmental effects. Coupling of PRIMES and GAINS is one essential part of the EC4MACS project (European Consortium for Modelling of Air Pollution and Climate Change, EC4MACS, 2010). This project prepares and provides consistent scenarios on energy demand/consumption, agricultural production, and emission abatement measures directed toward reducing air pollution and greenhouse gas emissions, and their costs. In this modeling chain, the CAPRI agricultural model is also of considerable interest to the nitrogen cycle, as it models the agricultural production supply as a result of demand and policy interventions. EC4MACS covers the EU member countries, but supplementary data are provided in GAINS to cover the whole of Europe.

560

In this chapter, we will use results of EC4MACS scenarios to illustrate nitrogen pathways in the short term. We will focus on the C&E package, which reflects the Climate and Energy package adopted by the European Commission.

24.4.3╇ Basic approach€– medium-term scenarios Some of the simplifications applicable to the short-term scenÂ� arios will no longer be plausible with respect to the 2050 timescale. For ‘medium-term’ scenarios there will be a need to account for substantial technological change beyond what currently exists as pilot plants. The idea of introducing ‘storyÂ� lines’, originally developed for climate research (Nakicenovic et╯al., 2000), allows several strands of activities to be consistently tracked into the future, based on the scopes of matching parameters. Scenarios have been grouped along two sets of key antagonisms (convergent world versus heterogeneous world; highly material intensive versus sustainable, service- and information-oriented economy), yielding four ‘families’ of scenarios. Even this concept is not only based on drivers; it also considers the effect-based approach in the ‘sustainable economy’ type of scenarios which aim to maintain environmental quality. This idea has been used widely. In addition to its integration into the Millennium Ecosystem Assessment (see above), it has been applied in connection with nitrogen pollution in Europe in the Eururalis project (Rienks, 2008; see also Verburg et╯al., 2006, for more information on land use scenarios).

Wilfried Winiwarter and Jean-Paul Hettelingh Figure 24.4 The four Eururalis scenarios, based on the concept of the IPCC SRES scenarios.

Global Economy

Global Co-operation

Market based solutions are most efficient to achieve strong economic growth and optimize demand and supply of goods, services and environmental quality.

Market based solutions to exploit comparative advantages, but strong internationally co-ordinated efforts are needed to address wealth distribution, social justice and the environment.

Continental Markets

Regional Communities

Market based solutions among like-minded countries, but shielded from countries with different values and standards. Cultural identity, strongly anchored in the countryside, must be preserved.

Self-reliance, environmental stewardship and equity are the key to sustainable development. Local communities are the cornerstone of society.

high regulation

low regulation

global

regional

Some results will be shown in Section 24.6. An overview of the Eururalis definition of storylines is presented in Figure 24.4. Again, these storylines are being built up from pairs of opposing parameters. Eururalis scenarios in combination with the IMAGE model provide detailed information on the agricultural sector, including changes in livestock, N fertilizer use, and land cover. Use of integrated models such as INTEGRATOR (see below) allows assessment of the impact of those changes on N (NH3, N2O, and NOx) emissions to the atmosphere, and N (nitrate and ammonium) leaching to groundwater and surface water. As this model does not include other anthropogenic emissions, only a very small part of NOx emissions to the atmosphere (emissions from soil) is considered. In Eururalis, the remaining NOx emissions from combustion processes can be taken from the IMAGE model on a coarser scale. While scenarios currently go to the year 2030, Eururalis works on an extension to 2050. Here we select scenarios A1 (Global Economy) and B2 (Regional Communities) to highlight the extremes of future possibilities:€ full globalization with a focus on economic growth versus regional protection with a focus on local cultural values. The Eururalis scenarios have also been chosen to represent the future situation in the NitroEurope integrated project (NEU, 2010, Sutton et╯al., 2007) and are processed using the dynamic nitrogen flow model INTEGRATOR (De Vries et╯al., 2010b). An important aspect of Eururalis is that within the scenarios different policy options can be evaluated. The interference

of different policy options–combinations of several individual measures – and other driving forces can be assessed. The Â�following policy options relevant in the EU have been considered in INTEGRATOR. • Policy on the Common Agricultural Policy (will affect both land use and livestock numbers and allocation). • Policy to stimulate the use of bio-energy (will affect land use). • Policy on Less Favoured Areas (will affect land use).

24.5╇ Developments toward long-term scenarios 24.5.1╇ Approach Extending a scenario beyond 2050 requires either a highly aggregated approach to be taken or the storylines to be defined in greater detail. Erisman et╯al. (2008) use a highly aggregated approach to estimate the global release of nitrogen to the environment based on the SRES storylines up to 2100. The authors do not distinguish N species, regions, or release conditions, and base their estimates on just five driving assumptions (size of population, food quality, food equity, efficiency gains and biofuel production). To obtain more specific projections, considerably more detail is required. This detail needs to be derived on a technology level. Any ideas on such future technologies must necessarily be

561

Future scenarios of nitrogen in Europe

speculative€– projecting how the technologies of 2100 (90 years hence) will look is like predicting the current situation from the perspective of 1920. At that time fossil fuels were in use, ammonia production from the elements had been invented, but the implication of each of these technologies, or the changes they would bring in terms of everyday life all over the globe could not be foreseen. We will now look at individual ‘emerging issues’ to describe possible future developments toward 2100, focusing especially on those that have not been already discussed as drivers. We will try as much as possible to link these issues with information available from other sources so as to draw plausible conclusions on the overall implications of these developments. A quantification of these implications has rarely been presented, and thus only few quantitative analyses can be provided.

24.5.2╇ Emerging agricultural systems Improvements in agricultural systems beyond 2050 may range from (1) increases in the efficiency of nitrogen uptake of moderately changed systems to (2) nitrogen emission reductions due to drastic changes in agricultural technology. Uptake efficiencies may be increased by improving (1) the spatial distribution of fertilizer via precision agriculture placing nitrogen at optimal distances from seeds or root zones with rates adjusted to current soil and crop conditions (Pagola et╯al., 2009) and (2) the timing of fertilization via controlled-release fertilizers (Chen et╯al., 2008; Yu et╯al., 2006) and application splitting (Olfs et╯al., 2005). Such improvements to current systems are to some extent already available, although not widely implemented. Drastic changes in agricultural technology are more difficult to predict, but several general arguments can be made. First, agricultural commodities may be produced in a more controlled environment in the future. Such systems could include soilless (hydroponic) agricultural systems (Jensen, 1997) in a closed environment with possibly zero nitrogen emissions. Second, future machinery for soil-based agriculture is likely to be more receptive and responsive to physical and biochemical signals and more versatile. The likely state of the art of such machinery would greatly depend on general progress in artificial intelligence (Murase, 2000). Agricultural robotic tools could be used for all agricultural management tasks including weed and pest control (Slaughter et╯al., 2008), irrigation (Phene, 1989), harvesting (Tanigaki et╯al., 2008; Van Henten et╯al., 2009) and determination of soil properties (Hemmat and Adamchuk, 2008; Liu et╯al., 2008). Robotic tools may also permit the use of multiple crop and companion planting systems as opposed to the crop rotation systems currently used with a single crop per field. Such systems may increase overall yields through decreased pest pressure and pesticide requirements and improved light, land, water, and nutrient efficiencies, as well as reducing soil erosion. Agricultural nitrogen emissions in a robotic agricultural system could substantially decrease because of higher fertilizer use efficiencies and reduced fertilizer requirements, higher yields and lower harvest losses, and reduced losses of nitrogen from other sources (soil, fuel).

562

24.5.3╇ Cereal crops with N2-fixing capacity

Cereal crops with N2-fixing capacity have long been a dream of scientists. If N losses from soil–plant systems are to be eliminated, a regulated N supply through a plant fixing its own N2 would be ideal (Ladha et╯al., 2005). Two different ways of engineering cereal crops have been explored since the early 1980s and again in the late 1990s (Giller and Merckx, 2003). The first approach is the introduction of the enzyme nitrogenase into the cereal plant so that it can fix atmospheric N2 directly. The unicellular cyanobacterium Gloeothece synthesizes nitrogenase in the dark and can fix atmospheric N2 when returned to the light. It has been proposed to engineer this ability into plants. However, such a plant system must supply energy for nitrogenase synthesis during the night and ensure protection of nitrogenase from molecular oxygen during photosynthesis in the daytime (Giller, 2001). Lack of funding for this research means that there has been little advance since the early 1980s. An alternative may be oxygen-tolerant nitrogenase. However, irrespective of the approach used, the functioning of nitrogenÂ� ase will require the supply of reductant and electrons, and this may be much more problematic than engineering plants to generate nitrogenase. The second approach is to manipulate the cereal plant so that it can nodulate with rhizobia, bacteria that live in symbiosis with the plant and that can fix atmospheric N2 in exchange for energy delivered by the plant. Although some exciting progress has been made with rice, the prospect of N2-fixing cereal crops remains distant (Ladha and Reddy, 2003). There are also disadvantages. Crop yields may be lower, as part of the energy from photosynthesis has to be invested in the process of breaking the triple bond of the N2 molecule. This energy cost may be 33% of plant photosynthate (Minchin et╯al., 1981). If we assume that crop yields would be 30% lower than the current European mean of ~3400 kg per hectare for the ~120 million hectares of harvested cereal area (FAO, 2009), an additional 35 million hectares would be needed to achieve the same total production. Although energy costs of N2 fixation are high, they may not be higher than those of nitrate assimilation (Pate and Layzell, 1981). It is thus uncertain what the net impact of the N2-fixation process on yields would be. A further disadvantage of N2-fixing nonlegumes is their high phosphorus requirement compared to crops that do not fix N2 (Giller, 2001). Large-scale production of N2-fixing cerÂ� eals would thus accelerate the depletion of global phosphaterock resources (Cordell et╯al., 2009).

24.5.4╇ Collection and recycling of human excreta in agriculture. While fertilizer nitrogen can be created from the atmosphere, this is not the case for phosphate. Depletion of global rock phosphate reserves is expected to become critical in the coming five to ten decades (Herring and Fantel, 1993). This will possibly lead to price increases for phosphate fertilizers, with important consequences for agricultural production systems beyond the time horizon of 2050.

Wilfried Winiwarter and Jean-Paul Hettelingh Table 24.5 Meat/egg production and consumption, 2007, in the EU countries

Activity

Meat/egg production [106 ton/yr]

Share of domestic production

2007 diet [kg/ cap/yr]

Healthy diet [kg/cap/yr]

Reduction in intake

Ruminantsa

9.3

96%

19.8

6.24

68%

Pigs

22.9

108%

44.8

5.66

87%

Poultry

11.4

103%

23.1

25.3

32%

Eggs

7.0

14.1

Total meat + eggs

50.6

101.8

37.2

63%

Ruminants are cattle, sheep, and goats Sources:€FEFAC, 2008; healthy diet:€Stehfest et╯al., 2009

a

Considerable amounts of phosphorous are available in human excreta. As rock phosphate reserves decrease, this other source is likely to be tapped. Closure of the P cycle also contributes to nitrogen recycling, while currently both compounds are disposed of in wastewater treatment plants. Both P and N are predominantly excreted in liquid form (urine). Sanitation systems for utilizing human excreta will likely provide separate collection for urine and feces. Magid et╯al. (2006) describe how such a sanitation system might look in practice. Different handling is suggested according to population density (e.g. urban centers versus rural areas). A decisive element in such a system is a strict separation of human excreta from industrial waste, as the latter may be contaminated with xenobiotic compounds. According to the authors, the additional costs of the alternative system would be fairly modest. Using Magid’s estimate of N excretion in urine (4 kg per person and year) yields roughly 3 Tg for Europe, about one-quarter of the N amount currently applied as mineral fertilizer. Although applying such systems would relieve the pressure to fix nitrogen from the atmosphere (and to mine for phosphate), this measure alone would not change conventional application practices. Despite closing the cycles, emission reductions would not occur in agriculture but in wastewater treatment and in industrial fertilizer production. The major release pathways of nitrogen into the environment would remain intact.

24.5.5╇ Effects of a low-meat scenario on nitrogen emissions It is well known that the quantity and quality of the human diet has an impact on the environment. Preventing food wastage and overconsumption of food automatically decreases the amount of food necessary to feed the entire global population. The actual composition of the human diet is also relevant because of the different environmental footprint of its components. Carlsson-Kanyama and González (2009) compared three meal options from Sweden each with an edible weight of about 0.5 kg. The meals had similar energy and protein contents but the greenhouse gas emissions varied from 0.42 (soybeans) through 1.3 (pork meat) to 4.7 (beef) kg CO2 equivalents.

In a Dutch study, four low-meat diets were compared with respect to their effect on global greenhouse gas emissions (Stehfest et╯al., 2009). The diets ranged from (1) no animal products in the diet, (2) no meat, (3) no ruminant meat to (4) a ‘healthy diet’. The healthy diet is based on sparing consumption of ruminant meat and pork (Willett, 2001). The healthy diet has a daily intake of 10 g beef, 10 g pork, 46.6 g chicken meat and eggs, and 23.5 g fish per capita, based on which some calculations on the effect on nitrogen emissions have been made for the whole of the EU. There are no changes assumed in the milk consumption per capita. Table 24.5 presents some data on actual meat production for the total EU (FEFAC, 2008). Using annual meat production and shares of domestic production we estimate the average intake of meat products and eggs per capita in the EU and compare this to the consumption in the case of the healthy diet. The total potential reduction in meat and egg consumption is 63%. As the milk demand is kept stable, there is also no reduction in the number of dairy cattle and cattle for replacement. The number of beef cattle, pigs, and poultry can be reduced because of the decreased demand for their products. Simple calculations show that nitrogen excretion in the EU could decrease by about 44% and the ammonia emissions by about 48%. The potential decrease in ammonia emissions is somewhat higher because the number of housed animals (which contribute more strongly to NH3 emissions) decreases while the number of grazing animals remains constant. In essence, low-meat diets may result in lower greenhouse gas emissions and in lower ammonia emissions.

24.5.6╇ Biofuels and nitrogen in the environment While the REFUEL project (Londo et╯al., 2010; Fischer et╯al., 2010a,b) focuses on bioenergy potential in Europe by 2030, it also provides relevant information for the longer time scales. According to their ‘Land use-energy scenario’, which we regard as the upper margin of potential, about 50 Mha cultivated land and 19 Mha pasture land could be made available for biofuels in a region covering the whole of the EU and Ukraine (plus Switzerland and Norway), while fully securing food production. This is a third of the region’s cultivated land and about 20% of total pasture. Creation of new agricultural area is not

563

Future scenarios of nitrogen in Europe Table 24.6 Models contributing to the ‘Representative Concentration Pathways’

model

Publication

description

RCP8.5

MESSAGE

Riahi et╯al. (2007)

Rising radiative forcing to 8.5 W/m2 in 2100

RCP6

AIM

Fujino et╯al. (2006)

Stabilization at 6 W/m2

RCP4.5

MiniCAM

Clarke et╯al. (2007)

Stabilization at 4.5 W/m2

RCP3-PD(2.6)

IMAGE

Van Vuuren et╯al. (2007)

Peak at 3 W/m2 before 2100 and decline thereafter

assumed, partly because there is not much unused land in Europe. Diet changes, as described in the section above, are not considered and may even enhance this bioenergy potential. In their reports, authors do not cover nitrogen requirements. As biofuel production adds to existing food/feed production, which will be intensified but not replaced, we may assume that the nitrogen requirements for food/feed production will in principle remain, save for some efficiency increases. Thus possible fertilizer needs for biofuels would occur on top of existing or expected needs for food/feed production, irrespective of the fact that the same area is used. In contrast to food and feed production, which also aim to provide protein to animals and humans, organic nitrogen is an undesired component in biofuel crops. Second-generation biofuels can thus, in principle, be grown on relatively small amounts of fertilizers, or even without addition of nitrogen under certain circumstances (Tilman et╯al., 2006). The other end of the scale, focusing on quick growth of biofuels without considering nitrogen demand, could easily require a fertilization rate of 100 kg N/ha, more closely resembling the current agricultural situation (but not nearly the level needed for firstgeneration biofuels). Using the land potentials of REFUEL, this would add about a third to the current fertilizer N input. Depending on the assumptions made (and how practical implementation occurs) it is clear that biofuel production could become a marginal factor, or a major player, in future nitrogen load to the environment in the long term. An increase in nitrogen input to soils could easily translate to similar increases in emissions to the atmosphere and to watersheds.

24.5.7╇ Long-term greenhouse gas scenarios Scenarios on future greenhouse gas emissions were initially limited to CO2 (Nakicenovic et╯al., 2000). As climate models have been further developed and are now also able to account for atmospheric chemistry and for the effects of tropospheric ozone or particulate matter, emission data must also be adapted accordingly. The first results of the ongoing work to create new IPCC scenarios have recently been made available. Data of the ‘Representative Concentration Pathways’ (RCPs) are hosted by among others IIASA (Representative Concentration Pathways, 2010). Through this site, more extensive documentation is available, although still in draft form, the ‘RCP handshake document’. The RCPs too consist of elements of both driver orientation and effect orientation. Here it is the respective pathways themselves that cover certain expectations on future abatement (radiative forcing target). The respective storylines leading to

564

the individual RCP derive from relatively independent exercises. Table 24.6 shows that each of the storylines is represented by just one model. Because of lack of knowledge regarding future technologies, a detailed accounting of abatement options is not possible. The storylines carry just a general understanding of long-term development, not only covering CO2 but also including gases relevant for the N cycle (NH3, NOx, N2O).

24.6╇ Available nitrogen scenarios for Europe 24.6.1╇ Concept The concepts for creating scenarios and projections as described above have been consistently applied to specific situations in Europe. Here we provide a compilation of existing and publicly available estimates of nitrogen release. Certain limitations make it difficult to compare data, similar to comparing the current or the past situation. • Depending on the purpose for which data have originally been collected, only some of the nitrogen release sources are covered in the respective studies. • Likewise, some of the emission fluxes (pathway of reactive N) may be missing. Nitrogen release through the aqueous pathway (surface water and groundwater) is seldom dealt with in scenarios covering emissions to the air. Moreover, modeling of water pollution is mostly based on watersheds, while modeling of airborne pollution (and greenhouse gases) usually follows administrative boundaries. • Because of the significant efforts on the part of the European Union to collect and evaluate environmentally relevant information, there are large differences in data availability between the EU member countries and the rest of Europe. While the EU comprises two-thirds of the total European population, it covers under half of its geographical area. While the EU does not represent Europe as a whole, it is often very useful to perform reliable comparisons on available information rather than attempt complete coverage. In Figures 24.5–24.7, we present projections of nitrogen release to the atmosphere (and leaching to groundwater, in one case). The frameworks of the respective scenarios were introduced in the previous sections. Results from the GAINS model represent a short-term scenario, comparing current legislation and control options. The Eururalis scenarios, as

Wilfried Winiwarter and Jean-Paul Hettelingh 4000 3500 GAINS NOx OPT MRR GAINS NH3 OPT MRR GAINS N2O OPT MRR

3000

Gg N/yr

2500 2000 1500 1000

Figure 24.5 Total release of nitrogen compounds (NOx, NH3,and N2O) to the atmosphere from EU27 from the GAINS model (scenario ‘C&E package’; current legislation plus two levels of further action for the year 2020 only).

500 0 2000

2005

2010

2015

2020

2025

2030

used by the INTEGRATOR model, represent a mid-term scenario interpreting the IPCC storylines, but nevertheless not extending beyond 2030 in the current phase of development. The RCP work reflects long-term climate scenarios. As all these approaches are work in progress, we can merely present results for the time of writing (end 2009). As generally the work focuses on elements other than nitrogen, further improvements seem possible and may be needed in order to understand the future role of nitrogen in the environment. Further to the European projections presented, it seems useful to refer to the global scenarios published by Erisman et╯al. (2008). As mentioned before (Section 24.5), global totals of nitrogen release only are available from this study, based on the impacts of a few key drivers. Despite the diversity of storylines chosen, all scenarios point toward increased pollution, constrained by a factor of roughly two, mainly as a result of the still strongly increasing world population. In contrast, European population is expected to decrease (UN, 2004), which should also drive a decline in nitrogen pollution.

24.6.2╇ Short-term scenarios:€GAINS The results from GAINS (Figure 24.5; the model itself is accessible at GAINS, 2010) display important changes€– reductions in NOx emissions€– already in the current legislation scenario. Still further reductions in NOx emissions, and also considerable reductions in NH3 emissions seem feasible should appropriate measures be taken. It is particularly interesting to note that applying additional measures leads to increases in the case of N2O emissions. This is quite typical for nitrogen because of the effects of pollution swapping. The GAINS model scenario here aims to reduce emissions of air pollutants and ignores the adverse effects of N2O emissions.

24.6.3╇ Medium-term scenarios:€INTEGRATOR The INTEGRATOR model was used in Eururalis to predict the N input and output fluxes of agricultural areas in response to the A1 and B2 default scenarios, as explained in Section 24.4. Here we present the results of the main N output fluxes (Figure

24.6) for the period 2000–2030 (see also De Vries et╯al., 2011, Chapter 15 this volume, who describe developments since 1970). While this period is only short-term, here we treat it as an example of a medium-term scenario because of the methods it uses. Differences in N fluxes are fairly small between the two scenarios. The A1 scenario assumes much larger crop production than B2, while under A1 livestock production is only slightly higher. The A1 scenario also provides clearly higher N fertilizer use and plant uptake, which leads to about 10% higher gaseous emissions (N2O, NOx, NH3), but only about a 7% increase in leaching compared to the more moderate B2 scenario. INTEGRATOR shows only half the N2O emissions of GAINS (Figure 24.5), which is mainly because of the much higher indirect emissions included in GAINS (N2O formation attributable to agricultural N release, but not occurring directly at the plot), namely, 45% of total emissions. In particular, indirect emissions from leached N are considered to be much smaller by INTEGRATOR because of a lower leaching fraction and a lower emission factor (see De Vries et╯al., 2011, Chapter 15 this volume). Moreover, INTEGRATOR, unlike GAINS, does not consider NOx emissions from combustion processes, which are responsible for the main share of emissions, as these come from non-agricultural sources. More information on the INTEGRATOR approach and the results obtained is given by De Vries et╯al. (2010b,c). In general, the changes in the coming decades are expected to be smaller than those of the past (see De Vries et╯al., 2011, Chapter 15 this volume), and the scenarios are also expected to differ less than those described in GAINS. While GAINS specifically focuses on abating emissions, in Eururalis the alternative scenario provides some elements of environmental consideration, but this is meant to be consistent rather than specifically targeted; this, in turn, could indicate that the maximum reductions outlined in GAINS are difficult to achieve, as they are not focused on being fully consistent. Moreover, Eururalis is not able to capture the decrease in NOx emissions shown so clearly by GAINS, as it is limited to the agricultural system and does not cover combustion emissions and their reductions.

565

Future scenarios of nitrogen in Europe

4000 3500 3000

NH3 A1 NH3 B2 NOx A1 NOx B2 N2O A1 N2O B2 Nleaching A1 Nleaching B2

Gg N/yr

2500 2000 1500 1000

Figure 24.6 Release of nitrogen from agriculture and animal husbandry from EU27 according to the INTEGRATOR model using Eururalis (version 2.0) scenarios. A1 scenario stands for a global, low regulation scenario, B2 for a regional and highly regulated, environmentally considerate scenario. Note that only agricultural emissions are included, in contrast to other studies that provide a complete inventory of emissions from all sources.

500 0 2000

2005

2010

2015

2020

2025

2030

6000 5000 NH3_R26 NH3_R45 NH3_R85 NOX_R26 NOX_R45 NOX_R85 N2O_R26

Gg N/yr

4000 3000 2000

Figure 24.7 Nitrogen emissions from EU27 as of RCP (v.0.9.9rc11), for three different storylines on radiative forcing. Only the model running the R26 storyline (IMAGE) provides spatially distributed N2O emissions. Storyline names indicate the radiative forcing exerted in 2100, between 2.6 and 8.5 W/m2.

1000 0 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100

An estimation of the future development of nitrate in river discharges is available from the Millennium Ecosystem Assessment (Seitzinger et╯al., 2010), while detailed data is also presented by Grizzetti et╯al. (2011, Chapter 17 this volume). The Millennium Ecosystem Assessment, whose concepts are very similar to those of Eururalis, provides a mid-term scenario based on an interpretation of the original IPCC storylines, and extends to 2050. The total release of nitrate from European catchments is estimated to change from 4.04 Tg N in 2000 to 3.99 Tg N in 2050 (or 3.5 Tg N, for an environmentally sensitive scenario), which is close to the agricultural N leaching reported just for the EU in the model results above (Figure 24.6). The slightly decreasing trend over time is also interesting. Data on the EU are not available and would not really be useful, as the flux into the marine environment is critical, especially with respect to the seas that have limited exchange with the world’s oceans:€the Baltic Sea, Black Sea, and Mediterranean.

24.6.4╇ Long-term scenarios:€RCP As the long-term greenhouse gas scenarios now also include the release of reactive trace constituents, it is useful to include them in this comparison. The RCP results (Figure 24.7), like GAINS,

566

cover all anthropogenic release to the atmosphere. The results presented for the EU thus generally resemble those of GAINS. While NOx declines strongly (and most clearly until 2040), ammonia and nitrous oxide emissions display an unclear trend. The most environmentally considerate storyline (R26) suggests that both nitrous oxide and ammonia will decrease consistently until 2100. This is because it was set up to keep greenhouse gas emissions to the minimum (which was not an aim in the GAINS scenarios used). Consequently, ammonia emissions are reduced much less in the RCPs than in the GAINS scenario optimized for maximum reduction.

24.6.5╇ Outlook Even across different timescales, comparison of existing scenarios on future nitrogen emissions provides interesting insights that are a valuable basis for further work. In general, trends are comparable, with emissions of NOx decreasing and N2O and NH3 largely remaining at a constant level over time. Where differences occur, these can be explained by recognizing the different ambition levels of the respective scenarios or simply by understanding the different system boundaries involved.

Wilfried Winiwarter and Jean-Paul Hettelingh

While scenarios are available that cover all media, there is not one single approach that consistently addresses nitrogen fluxes to the air, to groundwater, and to surface water. Scenarios have been built with respect to environmental problem areas, the key topics being ecosystem protection, air pollution, and greenhouse gas emissions. This is also true for the spatial resol� ution, with riverine pollution being defined along watersheds, air pollution according to country boundaries, and greenhouse gas emissions according to still larger units, possibly with some downscaling included. Extensions into the neighboring environmental media are being performed slowly. There may be some risk that storylines become inconsistent in the course of such an extension. The efforts of Leip et╯al. (2011, Chapter 16 this volume) demonstrate how nitrogen budgets can be used to assess specific fluxes in more detail. In principle it should also be possible to provide the input data needed to derive such budgets for scen� arios. Some effort will be needed to clearly define such a task. Specifically, a clear separation is required to assess the extent to which a specific storyline is affected by drivers only, and how strongly effects are considered. The extent to which effects are relevant may be called the ambition level. Covering nitrogen in different environments and media will need tradeoffs to be considered. Constraining the respective nitrogen flows in a budget approach will clearly improve the consistency of scenarios.

Acknowledgements This chapter was prepared with the support of the NinE Programme of the European Science Foundation, the NitroEurope IP (funded by the European Commission), the COST Action 729 and the EC LIFE project EC4MACS.

References Bakkes, J. A., Bosch, P. R. Bouwman, A. F. et╯al. (eds.), (2008). Background report to the OECD Environmental Outlook to 2030: overviews, details, and methodology of model-based analysis, MNP Report 500113001. MNP Netherlands Environmental Assessment Agency, Bilthoven, The Netherlands. Bouwman, A. F., Beusen, A. H. W., Billen, G. (2009). Human alteration of the global nitrogen and phosphorus soil balances for the period 1970–2050. Global Biogeochemical Cycles, 23, GB0A04. Brink, C., van Grinsven, H., Jacobsen, B. H. et╯al. (2011). Costs and benefits of nitrogen in the environment. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al. Cambridge University Press. Butterbach-Bahl, K., Nemitz, E., Zaehle, S. et╯al. (2011). Nitrogen as a threat to the European greenhouse balance. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al. Cambridge University Press. Carlsson-Kanyama, A. and González, A. D. (2009). Potential contributions of food consumption patterns to climate change. American Journal of Clinical Nutrition, 89 (supplement), 1740S–1709S. Carlsson-Kanyama, A., Dreborg, K. H., Moll H. C. and Padovan, D. (2008). Participative backcasting:€a tool for involving stakeholders in local sustainability planning. Futures, 40, 34–46. Chen, D., Suter, H., Islam, A. et╯al. (2008). Prospects of improving efficiency of fertiliser nitrogen in Australian agriculture:€a review of enhanced efficiency fertilisers. Australian Journal of Soil Research, 46, 289–301.

Clarke, L., Edmonds, J., Jacoby, H. et╯al. (2007). Scenarios of greenhouse gas emissions and atmospheric concentrations, Subreport 2.1A of Synthesis and Assessment Product 2.1 by the U.S. Climate Change Science Program and the Subcommittee on Global Change Research. Department of Energy, Office of Biological and Environmental Research, Washington, DC. Cordell, D., Drangert, J.-O. and White, S. (2009). The story of phosphorous:€global food security and food for thought. Global Environmental Change, 19, 292–305. Corvalan, C., Hales, S. and McMichael, A. (2005). Ecosystems and human well-being:€health synthesis, report prepared by WHO as contribution to the Millennium Ecosystem Assessment, WHO, Geneva. www.millenniumassessment.org. De Vries, W., Wamelink, W., Van Dobben, H. et╯al. (2010a). Use of dynamic soil vegetation models to assess impacts of nitrogen deposition on species composition and to estimate critical loads:€an overview. Ecological Applications (in press). De Vries, W., Kros, J., Reinds, G. J. et╯al. (2010b). INTEGRATOR:€a modelling tool for European-wide assessments of nitrogen and greenhouse gas fluxes in response to changes in land cover, land management and climate: calculation procedures, application methodology and examples of scenario results. Alterra Wageningen UR, Wageningen, the Netherlands. De Vries, W., Kros, J., Reinds, G. J. et╯al. (2010c). Assessment of nitrogen and green house gas fluxes at the European scale in response to land cover, livestock and land management change. Agriculture, Ecosystems and Environment (in preparation). De Vries, W., Leip, A., Reinds, G. J. et╯al. (2011). Geographic variation in terrestrial nitrogen budgets across Europe. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al. Cambridge University Press. Dise, N. B., Ashmore, M., Belyazid, S. et╯al. (2011). Nitrogen as a threat to European terrestrial biodiversity. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al. Cambridge University Press. Domene, L. and Ayres, R. U. (2001). Nitrogen’s role in industrial systems. Journal of Industrial Ecology, 5, 77–103. EC4MACS (2010). www.ec4macs.eu EFMA (2007). Forecast of food, farming and fertilizer use in the European Union 2007–2017. European Fertilizer Manufacturers Association, Brussels. Engström, R.J., and carlsson-Kanyama, A. (2004). Food losses in foos service institutions – examples from Sweden. Food Policy, 29, 203–213. Erisman, J. W., Sutton, M. A., Galloway, J., Klimont, Z. and Winiwarter, W. (2008). 100 years of ammonia synthesis:€how a single patent changed the world. Nature Geosciences, 1, 636–639. Erisman, J. W., van Grinsven, H., Lei, A., Mosier, A. and Bleeker, A. (2010). Nitrogen and biofuels; an overview of the current state of knowledge. Nutrient Cycling in Agroecosystems, 86, 211–223. EUROSTAT (2009). Statistical tables (population statistics). http://epp. eurostat.ec.europa.eu/portal/page/portal/statistics/themes Eururalis (2010). www.eururalis.eu FAO (2003). World agriculture:€towards 2015/2030: An FAO perspective, ed. J. Bruinma. Earthscan Publications, London. FAO (2009). FAOSTAT database collections. Food and Agriculture Organization of the United Nations, Rome. (http://faostat.fao.org/) FEFAC (2008). Feed and Food Statistical Yearbook 2007. FEFACEuropean feed manufacturers federation, Brussels. Fischer, G., Prieler, S., van Velthuizen, H. et╯al. (2010a). Biofuel production potentials in Europe:€sustainable use of cultivated land and pastures. Part I:€Land productivity potentials. Biomass and Bioenergy, 34, 159–172.

567

Future scenarios of nitrogen in Europe Fischer, G., Prieler, S., van Velthuizen, H. et╯al. (2010b). Biofuel production potentials in Europe:€sustainable use of cultivated land and pastures. Part II:€Land use scenarios. Biomass and Bioenergy, 34, 173–187. Forkes, J. (2007). Nitrogen balance for the urban food metabolism of Toronto, Canada, Resources Conservation and Recycling, 52, 74–94. Fujino, J., Nair, R., Kainuma, M., Masui, T. and Matsuoka, Y. (2006). Multi-gas mitigation analysis on stabilization scenarios using AIM global model. Multigas Mitigation and Climate Policy. Energy Journal, Special Issue 3, 343–354. GAINS (2010). http://gains.iiasa.ac.at Galloway, J. N., Aber, J. D., Erisman, J. W. et╯al. (2003). The nitrogen cascade. BioScience, 53, 341–356. Galloway, J., Townsend, A., Erisman, J. W. et╯al. (2008). Transformation of the nitrogen cycle:€recent trends, questions, and potential solution. Science, 320, 889–892. Giller, K. E. (2001). Nitrogen Fixation in Tropical Cropping Systems. CAB International, Wallingford,UK. Giller, K. E. and Merckx, R. (2003). Exploring the boundaries of N2-fixation in non-legumes:€an hypothetical and experimental framework. Symbiosis, 35, 3–17. Grizzetti, B., Bouraoui, F., Billen, G. et╯al. (2011). Nitrogen as a threat to European water quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al. Cambridge University Press. GTAP (2010). www.gtap.agecon.purdue.edu/ Hemmat, A. and Adamchuk, V. I. (2008). Sensor systems for measuring soil compaction:€review and analysis. Computers and Electronics in Agriculture, 63, 89–103. Herring, J. R. and Fantel, R. J. (1993). Phosphate rock demand into the next century:€impact on world food supply. Natural Resources Research, 2, 226–246. Hettelingh, J.-P., Posch, M. and Slootweg, J. (eds.) (2008). Status of the critical load database and impact assessment. In:€Critical load, dynamic modelling and impact assessment in Europe, CCE Status Report 2008. www.rivm.nl/cce Hettelingh, J.-P., De Vries, B. and Hordijk, L. (2009). Integrated assessment., In:€Principles of Environmental Sciences, J. Boersema, and L. Reijnders, ed., Springer, New york, pp. 385–417. Howarth, R., Bringezu, S., Bekunda, M. et╯al. (2009). Rapid assessment on biofuels and the environment:€overview and key findings. In:€Biofuels:€Environmental Consequences and Interactions with Changing Land Use, ed. R. W. Howarth and S. Bringezu, Cornell University, Ithaca, NY, pp. 1–13. International Fertilizer Industry Association:€Production and International Trade Committee (IFA-PITCom) (2009). World nitrogen industrial use in 2007. International Fertilizer Industry Association, Shanghai. Jensen, M. H. (1997). Hydroponics. Hortscience, 32, 1018–1021. Kantor, L. S., Lipton, K., Manchester, A. and Oliveira, V. (1997). Estimating and addressing America’s food losses. Food Review, 20, 2–12. Ladha, J. K. and Reddy, P. M. (2003). Nitrogen fixation in rice systems:€state of knowledge and future prospects. Plant and Soil, 262, 151–167. Ladha, J. K., Pathal, H., Krupnik, T. J., Six, J. and Van Kessel, C. (2005). Efficiency of fertilizer nitrogen in cereal production:€retrospects and prospects. Advances in Agronomy, 87, 85–156. Leip, A., Achermann, B., Billen, G. et╯al. (2011). Integrating nitrogen fluxes at the European scale. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al. Cambridge University Press.

568

Liu, H. B., Xie, D. and Wu, W. (2008). Soil water content forecasting by ANN and SVM hybrid architecture. Environmental Monitoring and Assessment, 143, 187–193. Llorens Abando, L. and Martinez Palou, A. M. (2006). Main stages in the meat food chain in Europe, Statistics in focus, Agriculture and Fisheries 6/2006. EUROSTAT, Luxembourg. Londo, M., Lensink, S., Wakker, A. et╯al. (2010). The REFUEL EU road map for biofuels in transport:€application of the project’s tools to some short-term policy issues. Biomass and Bioenergy, 34, 244–250. Lysen, E., Van Egmond, S., Dornburg, V. et╯al. (2008). Biomass assessment:€global biomass potentials and their links to food, water, biodiversity, energy modelling and economy,€Main report. Publisher:€Netherland Environmental Assessment Agency (MNP), Bilthoven, The Netherlands. MA (2005). Ecosystems and human well-being:€Scenarios, Volume 2, Findings of the Scenarios Working Group, Millennium Ecosystem Assessment. S. R. Carpenter, P. L. Pingali, E. M. Bennett and M. B. Zurek, ed., Island. press, Washington DC. Magid, J., Eilersen, A. M., Wrisberg, S. and Henze, M. (2006). Possibilities and barriers for recirculation of nutrients and organic matter from urban to rural areas:€a technical theoretical framework applied to the medium-sized town Hillerød, Denmark. Ecological Engineering, 28, 44–54. Minchin, F. R., Summerfield, R. J., Hadley, P., Roberts, E. H. and Rawsthorne, S. (1981). Carbon and nitrogen nutrition of nodulated roots of grain legumes. Plant, Cell and Environment, 4, 5–26. Moldanová, J., Grennfelt, P., Jonsson, Å. et╯al. (2011). Nitrogen as a threat to European air quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al. Cambridge University Press. Murase, H. (2000). Artificial intelligence in agriculture. Computers and Electronics in Agriculture, 29, 1–2. NEU (2010). www.nitroeurope.eu. Nowicki, P. L., van Meijl, J. C. M., Kneirim, A. et╯al. (2006). Scenar 2020:€scenario study on agriculture and the rural world EC. Directorate Gen. for Agriculture and Rural Development, Brussels. Nakicenovic, N., Alcamo, J., Daris, G., et al. (2006). Special report on Emissions Scenarios. Working group III of the Intergovernmental Panel on Climate Change, cambridge University Press, Cambridge, UK, 2000. OECD (2008). Environmental performance of agriculture in OECD countries since 1990. Organisation for Economic Co-operation and Development, Paris. Oenema, O., Bleeker, A., Braathen, N. A. et╯al. (2011). Nitrogen in current European policies. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al. Cambridge University Press. Olfs, H. W., Blankenau, K., Brentrup, F., et╯al. (2005). Soil- and plantbased nitrogen-fertilizer recommendations in arable farming. Journal of Plant Nutrition and Soil Science, 168, 414. Pagola, M., Ortiz, R., Irigoyen, I. et╯al. (2009). New method to assess barley nitrogen nutrition status based on image colour analysis:€comparison with SPAD-502. Computers and Electronics in Agriculture, 65, 213–218. Pate, J. S. and Layzell, D. B. (1981). Energetics and biological costs of nitrogen assimilation. In:€The Biochemistry of Plants, ed. Miflim, B. J. and Lea, P. J. Academic Press, San Diego, Ca, pp. 1–41. Phene, C. J. (1989). Techniques for computerized irrigation management. Computers and Electronics in Agriculture, 3, 189–208. Posch M., Hettelingh, J.-P., Slootweg, J. and Maas, R. (2008). Illustrative dynamic modeling applications for use in support

Wilfried Winiwarter and Jean-Paul Hettelingh of European air pollution abatement policies. In:€Critical load, dynamic modelling and impact assessment in Europe, J.-P.,€Hettelingh, M. Posch, and J., Slootweg, eds. CCE Status Report 2008. www.pbl.nl/cce Prud ’ homme, M. (2009). Global Fertilizers and Raw Materials Supply and Supply/Demand Balances. International Fertilizer Industry Association, Shanghai. Resy, D.S., Hasand, C.M., Bleeker, A. et al. (2004). Societal Choice and Communicating in the European Nitrogen Challenge. In: The European Nitrogen Assessment, ed. M.A. Sutton, C.M. Hasand, J.W.€Erisman et al., Cambridge University Press. Representative Concentration Pathways (2010). www.iiasa.ac.at/webapps/tnt/RcpDb Riahi, K., Gruebler, A. and Nakicenovic, N. (2007). Scenarios of longterm socio-economic and environmental development under climate stabilization. Technological Forecasting and Social Change, 74, 887–935. Rienks, W. (ed.) (2008). The Future of Rural Europe. Wageningen University Research and Netherlands Environmental Assessment Agency, Wageningen, The Netherlands. Seitzinger, S. P., Mayorga, E., Bouwman, A. F. et╯al. (2010). Global river nutrient export:€a scenario analysis of past and future trends. Global Biogeochemical. Cycles, 24, GB0A08. Slaughter, D. C., Giles, D. K. and Downey, D. (2008). Autonomous robotic weed control systems:€a review. Computers and Electronics in Agriculture, 61, 63–78. Smil, V. (2002). Eating meat:€evolution, patterns, and consequences. Population and Development Review, 28, 599–639. Stehfest, E., Bouwman, A. F., van Vuuren, D. P. et╯al. (2009). Climate benefits of changing diet. Climatic Change, 95, 83–102. Sutton, M. A., Nemitz, E., Erisman, J. W. et╯al. (2007). Challenges in quantifying biosphere–atmosphere exchange of nitrogen species. Environmental Pollution, 150, 125–139. Svirejeva-Hopkins, A. and Schellnhuber, H.-J. (2008). Urban expansion and its contribution to the regional carbon

emissions:€using the model based on the population density distribution. Ecological Modelling, 216, 208–216. Svirejeva-Hopkins, A., Reis, S., Magid, J. et╯al. (2011). Nitrogen flows and fate in urban landscapes. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et╯al. Cambridge University Press. Tanigaki, K., Fujiura, T., Akase, A. and Imagawa, J. (2008). Cherryharvesting robot. Computers and Electronics in Agriculture, 63, 65–72. Tilman, D., Hill, J. and Lehman, C. (2006). Carbon-negative biofuels from low-input high-diversity grassland biomass. Science, 314, 1598–1600. UN (2004). World population to 2300. United Nations Publications, New York. Van Drecht, G., Bouwman, A. F., Harrison, J. and Knoop, J. M. (2009). Global nitrogen and phosphate in urban wastewater for the period 1970 to 2050. Global Biogeochemical. Cycles, 23, GB0A03. Van Egmond, Bresser, T. and Bouwman, L. (2002). The European Nitrogen case. Ambio 31, 72–78. Van Henten, E. J., Van ’ t Slot, D. A., Hol, C. W. J. and Van Willigenburg, L. G. (2009). Optimal manipulator design for a cucumber harvesting robot. Computers and Electronics in Agriculture, 65, 247–257. van Vuuren, D., den Elzen, M., Lucas, P. et╯al. (2007). Stabilizing€greenhouse gas concentrations at low levels:€an assessment of reduction strategies and costs. Climatic Change, 81, 119–159. Verburg, P. H., Veldkamp, A. and Rounsevell, M. D. A. (2006). Scenario-based studies of future land use in Europe. Agriculture, Ecosystems and Environment, 114, 1–6. Willett, W. C. (2001). Eat, Drink, and be Healthy:€the Harvard Medical School Guide to Healthy Eating. Simon & Schuster, New York. Yu, L., Li, D., Yu, S. et╯al. (2006). Research advances in slow/controlled release fertilizers. Shengtaixue Zazhi, 25, 1559–1563.

569

Chapter

25

Coordinating European nitrogen policies between international conventions and intergovernmental organizations Lead author: Keith Bull Co-authors: Robert Hoft and Mark A. Sutton

Executive summary Nature of the problem • International treaties, such as multilateral environmental agreements (MEAs), have sought to protect the environment by intergovernmental action on many issues. • The MEAs and intergovernmental organizations (IGOs) have, between them, targeted most known environmental problems, but none has tackled nitrogen management holistically since the nitrogen issue is much broader than any of the individual interests. • Even so, several conventions have taken action to develop nitrogen policies in their specific areas of interest, but they are often limited in their options to increase their scope of action beyond their agreed mandates and may be reluctant to coordinate action with those of others. • As a result, there remains a need to develop an integrated, holistic approach for nitrogen management; an international treaty targeted explicitly on nitrogen would have the potential to bring the different elements of the nitrogen problem together.

Approaches • Some coordination between MEAs and IGOs already occurs with regard to different nitrogen threats, but the focus is inevitably on areas of overlapping interests. This chapter explores the potential for available mechanisms to be applied further across these institutions to harmonize work and to promote effective coordination on nitrogen-related threats and abatement options.

Key findings/state of knowledge • Coordination between MEAs and IGOs on various topics, often not related to nitrogen policy, has been achieved through, inter alia, formal agreements, joint participation in meetings or projects, and actions by convention secretariats. All these approaches, and others, have the potential to stimulate coordination of nitrogen issues, but perception of overlapping interests and recognition of the benefits of coordination are key to success. • The European Union (EU), itself established by international treaties, has a major role in Europe harmonizing policy in EU Member States and coordinating their actions regionally and globally. • Scientific knowledge and understanding is usually the pre-requisite to formal agreement between States for action on environmental issues. For nitrogen, the different measuring and modelling activities between air, land and water need to be brought together and harmonized. • Scientific and technical cooperation between MEAs has proved especially important in identifying the many links between reactive nitrogen threats, with the international scientific community able to provide a role in harmonizing information supply to different forums and promoting coordination. • Coordination of national policies by individual countries can ensure harmonized national action and help stimulate the work of MEAs and IGOs in coordinating nitrogen related policies; effective dialogue between national delegates to institutions with overlapping interests can ensure consistent and harmonized national and international action. • In the long term, a new international treaty on nitrogen could be a powerful mechanism for coordinated global or regional nitrogen management; but this could be complex to negotiate. Meanwhile, continued efforts using existing MEAs and IGOs to coordinate action shows success in some quarters.

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

570

Keith Bull

Major uncertainties/challenges • A new legal instrument, new convention or joint protocol, for nitrogen management would need to be harmonized with existing agreements. • Influencing decisions by MEAs, IGOs and the EU can be difficult if it involves new initiatives or changes to existing plans. Influencing coordinated decisions between MEAs and IGOs multiplies these difficulties. • Coordinating the work of the scientific community to address the needs of several forums requires effort and resources, especially in the case of nitrogen management, which faces considerable technical challenges in linking between environmental media on multiple scales. • International agreements are concerned with implementation of measures such as limits, regulations and guidelines, at the international and national levels, but at the local level action to comply with such measures may still present major challenges.

Recommendations • The immediate recommendation is to exploit established mechanisms and institutions to develop new coordinating links on nitrogen management between MEAs and IGOs. National coordination should be encouraged to harmonize action programmes at national and international levels. • In the longer term, possible options for a new framework convention or inter-convention joint protocol should be explored to assess the potential benefits of such an instrument. • The Global Partnership on Nutrient Management (GPNM), established under the United Nations Environment Programme (UNEP), and the International Nitrogen Initiative should be encouraged to develop policy and scientific cooperation, respectively, at a global level. • Regionally, the scientific bodies of the Convention on Long-range Transboundary Air Pollution and its Task Force on Reactive Nitrogen are in a good position to take a lead on some aspects of coordination; they should be encouraged to work to link atmospheric with other nitrogen threats. At another level, there is the opportunity for the United Nations Economic Commission for Europe (UNECE) Committee on Environmental Policy to develop the nitrogen management links between the UNECE Conventions. • There is a pressing need to coordinate different nitrogen measurement and modelling activities between air, land and water. Opportunities should be sought to bring together relevant, multi-media nitrogen science to provide cross-cutting information to underpin policy decisions. In the first instance, approaches should be made to other UNECE environmental conventions in order to explore possibilities for collaboration.

25.1╇ Introduction Multilateral environmental agreements (MEAs), such as conventions and protocols, have successfully tackled many of the known environmental problems and continue to address the many outstanding issues of international concern. They have done much to harmonize the efforts of governments and have provided important driving forces for international and national action on environmental matters including the management of nitrogen. The development and interests of several environmental conventions have already been outlined in relation to current European policies on nitrogen (see Oenema et╯al., 2011a, Chapter 4, this volume). There is no single MEA that covers nitrogen management holistically. The need to consider air, land and water as well as a wide range of sectors, including industry, agriculture and transport makes it difficult for any one convention to deal with all issues when their mandates for action are usually limited. Of further concern is that the lack of coordination of action can lead to simply shifting the environmental problem from one area to another, for example, nitrogen management problems can be shifted from water to air and vice versa (Bleeker et╯al., 2009; Spranger et╯al., 2009; Cellier et╯al., 2011 and SvirejevaHopkins et╯al., 2011, Oenema et╯al., 2011b€ – Chapters 11, 12 and 23, this volume). Where environmental problems are relatively simple and well-defined, some MEAs have had major successes. The Convention on Long-range Transboundary Air Pollution (CLRTAP) of the United Nations Economic Commission for

Europe (UNECE) has, through its protocols on sulphur, cut emissions of sulphur in Europe by more than 70% (55╛Tg to 15╛Tg from 1980 to 2004) (Vestreng et╯al., 2007). Unfortunately, nitrogen management is not such a simple, easily defined problem. There are several MEAs and intergovernmental organizations (IGOs, see Sections 25.2.1 and 25.2.2 for definitions) with vested interests in nitrogen management. Such organizations could encourage the international community to work together on environmental issues, and between them they have potential for coordinating further action on mitigating reactive nitrogen (Nr) emissions and effects. For this, they will need to collaborate and coordinate their individual activities in a way that has proved difficult in the past. Though some coordination already exists on nitrogen management, for example, through EU policy and legislation (Oenema et╯al., 2011a, Chapter 4, this volume) there is scope for increased links and coordination between institutions such as MEAs and IGOs. Cooperation tends to be focused on areas of overlapping interests, but more needs to be done with such existing links and mechanisms to broaden the collective approach. Links between MEAs and IGOs can take place at a range of different levels, from formal to informal, and from using high-level overarching bodies to individual governments and stakeholders, such as industry and environmental non�governmental organizations (NGOs). But each MEA or IGO has its own priorities and interests and it can be difficult to bring them together to focus on common goals. Some stakeholders

571

Coordinating European nitrogen policies

can play a useful role in linking MEAs and IGOs, since they attend meetings of several bodies under different agreements; but the nature of their special interests, e.g., motor manufacturer or chemical industry, means their real role in any one body is very limited. The need for interlinkages between MEAs to address many of the problems that we face today is widely recognized. The Interlinkages Initiative of the United Nations University (2002) was started in the 1990s to draw attention to the need to build links between MEAs. The Initiative drew attention to the lack of an over-arching, unitary structure for global environmental governance (United Nations, 1999). Existing over-arching bodies, such as the United Nations (UN) Economic and Social Council (ECOSOC) and the Commission on Sustainable Development (CSD) have proved incapable since their mandates are vague and UN States have been reluctant to invest the necessary power in these bodies. In addition, there is apparent weakness in international law and the ability of international institutions to create or enforce rules. The Initiative recorded that no amount of coordination of MEAs would overcome these fundamental shortcomings. One over-arching body, the United Nations Economic Commission for Europe (UNECE), through its Committee on Environmental Policy (CEP), has made several attempts to develop interlinkages between its five regional MEAs.1 In 2000, a review of synergies to be derived from closer cooperation (Economic Commission for Europe, 2000) was discussed by CEP and its recommendations endorsed. Subsequently, CEP agreed guidelines for strengthening compliance with and implementation of multilateral environmental agreements in the UNECE region (Economic Commission for Europe, 2003a) that were endorsed by the fifth Ministerial Conference ‘Environment for Europe’; more recently, CEP submitted information and recommendations on implementation of UNECE multilateral environmental agreements (Economic Commission for Europe, 2007a) to the sixth ‘Environment for Europe’ Ministerial Conference. Despite such continued action and improvements in implementation across all five conventions, there are few interlinkages of note. There remains good potential for interaction since: • the secretariats for all five conventions operate within a single UNECE division so convention meetings and activities can be coordinated and joint discussions arranged easily; • the conventions have only Parties from the UNECE region€– the 56 member States of Europe, North America and Central Asia€– so cultural, economic and geographic differences are not so great as with global agreements.

1

572

The 1979 Convention on Long-range Transboundary Air Pollution (CLRTAP), the 1991 Convention on Environmental Impact Assessment in a Transboundary Context, the 1992 Convention on the Protection and Use of Transboundary Watercourses and International Lakes (the Water Convention), the 1992 Convention on Industrial Accidents, and the 1998 Convention on Access to Information, Public Participation in Decision Making and Access to Justice in Environmental Matters (Aarhus Convention).

However, there is reluctance by UNECE convention bodies to spend time and effort where they believe there is little to be gained. The only real coordination has taken place where there are specific overlapping interests and UNECE continues to look at its MEAs collectively to see where overlaps and common challenges, e.g., implementation, might be exploited (Schrage et╯al., 2007). While it might be considered disappointing that interaction between institutions is generally lacking, some believe that such decentralized, fragmented governance is a good thing, since it avoids burdensome overarching bureaucracy and may encourage competition and opportunities for learning. However, if specific problem issues fall within the remit of many MEAs and IGOs, there are great risks of duplication of effort and inconsistencies in decision making. With increasing proliferation of MEAs in recent years these risks have much increased. The UNU (United Nations University) Interlinkages Initiative highlights that we have failed to prepare socioeconomic systems to deal with inter-linked problems. Due to institutional, historical, financial or capacity reasons, our laws, conventions, treaties, institutions, mechanisms and information have developed in isolation and focus on separate topics or themes. So although we know that we have to deal with the environment and development at the same time, most institutions still focus mainly on one or the other. More recently, the UN’s Joint Inspection Unit (JIU) has carried out a ‘Management review of environmental governance within the UN System’ (see Inomata, 2008). The review aimed ‘to strengthen the governance of and programmatic and administrative support for MEAs by UN organizations by identifying measures to promote enhanced coordination, coherence and synergies between MEAs and the UN system, thus increasing the UN system’s contribution towards a more integrated approach to international environmental governance and management at national, regional and international levels’. The review notes the institutional fragmentation and specialization and the lack of a holistic approach to environmental issues and sustainable development, as well as the lack of interaction of UN entities responsible for development with MEAs. The review makes a number of recommendations to the UN Secretary-General and the UN General Assembly regarding future action; these focus, in particular, on mechanisms to improve the functioning of MEA secretariats and intergovernmental bodies. Such high-level action is not easily influenced by individual MEAs or parties to such agreements. The next part of this chapter looks at the nature of MEAs and IGOs to consider coordination at a more general level; it then assesses the available options for interaction and coordination between institutions on common environmental issues, such as nitrogen. While there are reasons that international organizations and agreements often work in isolation or with limited interaction with others, there are quite a number of mechanisms that can overcome these barriers; these are explored through existing examples where they have proved effective.

Keith Bull Table 25.1 Overview of Multilateral Environmental Agreements (MEAs) and Intergovernmental Organizations (IGOs) with interests in nitrogen management

Main nitrogen management interest

MEA/IGO

Scope

United Nations Framework Convention on Climate Change (UNFCCC)a

Global

Nitrogen containing greenhouse gases and ozone. Carbon sequestration

Convention on Biological Diversity (CBD)a

Global

Nitrogen and ozone impacts on biodiversity

Convention on Long-range Transboundary Air Pollution (CLRTAP)a

Europe, Central Asia and North America

Nitrogen oxides, and ammonia emissions, ozone and impacts on human health and the environment

Water Conventiona

Europe, Central Asia and North America

Water quality and management

Baltic Sea States

Eutrophication of Baltic Sea

North Atlantic Sea States

Eutrophication of areas of the Atlantic Ocean and North Sea

United Nations Environment Programme (UNEP)c

Global

Broad interests in the environment. Global Partnership on Nutrient Management

World Health Organization (WHO)c

Global

Human health impacts of air and aquatic pollution. Food and nutrition

Global

Agriculture and forestry

Global

Weather and climate. Air pollution monitoring

Intergovernmental Panel on Climate Change (IPCC)c

Global

Scientific aspects of climate change. Ozone and N greenhouse gases. The nitrogen/carbon cycles

Arctic Councilc

Arctic region States

Effects of nitrogen in the Arctic region

Helsinki Commission (HELCOM)b Oslo Paris Commission (OSPARCOM)

b

Food and Agriculture Organization (FAO)c World Meteorological Organization (WMO)

c

Multilateral environmental agreement (MEA). Governing body of MEA. c Intergovernmental Organization (IGO). For further details see Oenema et╯al. (2011a, Chapter 4, this volume). a b

25.2╇ The functions and operations of multilateral environmental agreements and intergovernmental organizations The activities of many of the organizations and agreements that have interests in nitrogen management are described in Oenema et╯al., 2011a (Chapter 4, this volume) and summarized briefly here in Table 25.1. Note that several of these organizations are UN-based, e.g. open for membership by UN States. They are seen as part of the UN ‘family’, but they have their own, separate constitutions and independent governing bodies; the UN does not act as an over-arching body. While considering some of the current interests of MEAs and IGOs, it becomes clear that there are problems in developing coordinated approaches. Each institution or body has specific goals in line with their given mandates, and this inevitably limits their scope of work. Coordination is therefore essential to develop sound policy for nitrogen management, and here we look generally at the institutional set-up of MEAs and IGOs to see what basis there is for developing coordination between bodies. As noted previously, there is no single MEA specifically addressing nitrogen management issues. Yet, even if there was,

there would still need to be coordination between it and the various other institutions with overlapping interests in nitrogen. The absence of an over-arching institution and the wide range of activities under existing treaties and IGOs, means there is no holistic approach to nitrogen management and there is no obvious existing institution that can take the lead to coordinate the activities of others. For effective coordination, each and every MEA and IGO with nitrogen interests needs to play an active role. Here we consider the nature and mechanisms of international agreements and organizations in order to explore what options exist for promoting interlinkages in the future.

25.2.1╇ Treaties, conventions and protocols A treaty is an agreement under international law that may take the form of a convention, protocol, exchange of letters, etc. Such a treaty is established between ‘Sovereign States’ and possibly international organizations through their signing and agreeing to comply with the terms of a written agreement. Failure to comply has the consequence of being held liable under international law, though enforcement of international law for many issues, such as environmental ones, is not necessarily strong.

573

Coordinating European nitrogen policies Table 25.2 Institutions associated with a Multilateral Environmental Agreement (MEA)

Meeting/conference of the Parties Secretariat Bodies for scientific and technological advice Bodies for technical assessment of information Bodies for assessing compliance and reporting non-compliance Financial institutions Capacity-building institutions

MEAs are a special form of treaty that target environmental issues and are often looked upon as being separate from other types of treaty. For MEAs, the initial or framework treaty is usually called a convention. It is often the culmination of long negotiations between states, usually held within the framework of relevant political gathering of states, such as that provided by UNEP globally or UNECE regionally. Such negotiations may be initiated by an individual state or group of states with particular concerns on an environmental issue and that have sufficient persuasive arguments to convince the gathering of states of the importance of embarking on further investigation or negotiations. A protocol is a supplement to such a convention. A convention may have several protocols under it. In many instances these protocols detail specific actions to be taken by the parties which address the general aims of the convention. For example, the 1984 Protocol on Sulphur to CLRTAP requires its Parties to cut emissions of sulphur by 30% thus addressing article 2 of the Convention by reducing and preventing air pollution. Quite a number of MEAs have been established in recent decades by States coming together, either regionally or globally, to sign up to treaties aimed at specific environmental goals, such as biological diversity, climate change and air pollution. The texts of the agreements spell out what these goals are and how the parties to the agreement will reach them, including, for example, how a secretariat will be established and how a governing body will operate, as well as specifying which States might become parties and how they should do this, i.e., the procedures for becoming a signatory and for ratifying the agreement.2 One way to consider possible coordination mechanisms is to consider the types of institutions built up within, or sometimes external to, MEAs (Table 25.2). Coordination is possible at all these levels and options for some of these will be explored in detail in Section 25.3. Interlinkages between the various institutions of different MEAs can be both through formal and informal channels. Formal links are those recognized by the parties to an

2

574

Ratification of a treaty by a State, which often takes place months or years after the signing, indicates its agreement to be bound by its terms. Signing a treaty simply shows agreement with the principles of the treaty text.

MEA through collective decisions taken within its framework. Informal links might result, for example, from the memberships of a state to several MEAs; delegates of that state can develop national coordination through their awareness of the need for harmonization, exchange of information, etc. This ‘individual coordination’ might also take place fairly formally at the national level, through establishing national coordinating bodies, or even formally between groups of governments of several parties, e.g., the Nordic States. Overall, the interlinkages between an MEA and other institutions depend to a great extent on the views of the parties and how proactive they are in seeking links and coordination. Sometimes the adopted text of an MEA will include a requirement to collaborate with another body, e.g., the text of the Kyoto Protocol makes repeated reference to IPCC. Mostly though, interlinkages with other institutions are developed during the implementation of an MEA as parties become aware of the need to share knowledge and information, and to harmonize approaches to common issues. Such harmonization is often of major interest to states that are party to more than one agreement, which have similar requirements or obligations. A good technical example of this is the emission reporting requirements under the CLRTAP and UNFCCC protocols. These requirements were harmonized through cooperation between their respective technical bodies. In this way emission experts harmonize their reports on air pollutant and greenhouse gas emissions and avoid the unwanted task of calculating and reporting emissions using two different methods. Secretariats also have an important role to play. Most are aware of the activities of other MEAs and IGOs with similar interests and they may participate as observers at meetings of other bodies. They may share information on meetings and activities, and they are the usual focal point for sending official invitations to other MEAs and IGOs inviting them to participate or contribute to meetings or work. As noted in Oenema et╯al., 2011a (Chapter 4, this volume), several conventions have taken action individually to develop nitrogen policy in their specific areas of interest. However, again as previously noted, the scope of their individual actions is limited by the terms of their treaties. While governing bodies can act autonomously and decide upon a course of action to expand their scope of work, treaty texts may specify geographic limits (e.g., regional seas) or limits to the environmental interests (e.g., air pollution) and parties will be unable, or reluctant, to step outside the provisions of the original agreement even if they are aware of the broader issues. So issues such as nitrogen management are only tackled piecemeal because of the restricted mandates. One area where there is much scope for developing interlinkages and coordination is in the scientific and technical work of MEAs (an example on emissions reporting has been given above). Science is seen as a fundamental starting point for environmental law and policy, and various authors have drawn attention to the importance of strong links between science and policy. The global science development is coordinated under the flag of the International Council for Science (ICSU),

Keith Bull

currently developing a vision on Earth sciences. There are several Interdisciplinary Bodies and Joint Initiatives under ICSU that address nitrogen issues, such as the Global Environmental Change Programmes: • International Programme of Biodiversity Science (DIVERSITAS) • International Geosphere-Biosphere Programme (IGBP) • International Human Dimensions Programme on Global Environmental Change (IHDP) • WMO-ICSU-IOC World Climate Research Programme (WCRP) and the monitoring and observations programmes: • Global Climate Observing System (GCOS) • Global Ocean Observing System (GOOS) • Global Terrestrial Observing System (GTOS). In a similar way, the scientific links between the different interests of MEAs can be an important driving force for better policy coordination. Using the same science as a starting point for different interest areas can greatly improve the consistency of policy development and provide the basis for coordinating important policy-related activities, such as monitoring, reporting, and impacts assessment.

25.2.2╇ Intergovernmental organizations Intergovernmental organizations (IGOs) are also created between States by international agreement, treaty or charter, which provide the goals and scope of the interests of the organization. They are legal entities and important parts of public international law. They often have a much broader remit than MEAs and with less specific targets. While they might take decisions that have implications for international and national law, they are not law-making bodies and are generally open to participation by states, not just to parties that have signed up to the agreements. Even so, states might need to declare membership to indicate they wish to participate in IGOs activities, and IGOs often champion MEAs in order to establish international law in their areas of interest. For example, CLRTAP was negotiated under the auspices of UNECE. Similar to MEAs, IGOs also have governing bodies that, through collective decision by the participating states, determine how the organization will operate and agree its work. IGOs might too have subsidiary bodies and set up scientific or technical groups to carry out specific tasks. They also have secretariats whose range of activities can be quite extensive and who may carry out much of the work of the organization. Funding of IGOs is provided by its members and funding mechanisms can be strong enough to support significant institutional structures. For example, the UN has a system of Trust Funds, some voluntary but others that member States are obliged to contribute to according to an agreed sharing of costs. Such sharing is often based on the UN scale of assessment; this is set by the UN General Assembly and is broadly based upon States ability to pay. Some MEAs make use of a similar costsharing mechanism, though mostly they depend upon voluntary contributions.

25.3╇ Mechanisms for future coordination of action on nitrogen by MEAs and IGOs For any environmental issue, there are a variety of ways that MEAs and IGOs can coordinate and harmonize their actions. Coordination can operate at different levels of formality and be implemented by parties or states individually or collectively. But key to success is a perception of over-lapping interests and recognition of benefits from the coordination process. Cooperation and coordinated action will inevitably require resources and effort and these will need to be balanced against the benefits achieved. Over-arching institutional interlinkages are one way to bring individual MEAs and IGOs together. As noted in Section 25.1, the UNECE Committee on Environmental Policy, itself responsible for initiating negotiation of several MEAs, has tried to bring together the five UNECE MEAs to explore synergies and prompt interlinkages on implementation. Mostly, such efforts have achieved little, since the areas of common interest identified so far have been limited. However, all five conventions have interests in capacity building in certain countries, e.g., states in Central Asia, so there has been value in coordinating action to achieve UNECE and MEA goals. Broadly speaking, we can conclude from the above that there are three possible ways that nitrogen management might be addressed more holistically. (a) Start a new process to develop a new MEA or IGO for nitrogen; this would be a difficult option, but, if achievable, it has much potential for tackling the nitrogen management issue. (b) Work from an existing MEA/IGO and broaden its scope gradually to involve other bodies for addressing specific nitrogen issues. This would not be a fully integrated approach and would need changes to, or work around, existing institutional structures in order for it to be effective. (c) Use current MEAs and/or IGOs and put effort into establishing links and cooperation. Formally, for example through a joint protocol, this might be difficult to negotiate, but less formally, for example through exchanges of letters or joint technical bodies, this might have some potential. To see how institutions with interests in nitrogen management might harmonize their work and coordinate their activities with one another, we here explore existing mechanisms used. Various options are discussed below broadly based upon the ‘level of activity’ within a MEA or IGO, ranging from highlevel, formal and international down to low-level, informal and national. Even so, no matter which level is targeted, few approaches are simple and easy, and some would require considerable amounts of work and effort.

25.3.1╇ A new MEA or IGO for nitrogen management A new international treaty, such as a convention on nitrogen could be a powerful mechanism for nitrogen management at

575

Coordinating European nitrogen policies

the regional or even global level. As outlined in previous chapters of this assessment, current policies related to the mitigation of nitrogen effects have not been fully successful, typically addressing different aspects, with little overall coordination. For this reason much more effort is needed to develop more holistic, integrated approaches to nitrogen management (Oenema et╯al., 2011a,b, Chapters 4 and 23, this volume). Developing a coordinated strategy under an international treaty targeted explicitly on nitrogen has the potential to bring the different elements of the nitrogen problem together. Thus it could consider both the benefits of reactive nitrogen for food and energy security etc. (Jensen et╯al., 2011, Chapter 3, this volume) and the five key societal threats:€water quality, air quality, greenhouse gas balance, ecosystems / biodiversity, and soil quality (Grizzetti et╯al., 2011; Moldanová et╯al., 2011; Butterbach-Bahl et╯al., 2011; Dise et╯al., 2011; Velthof et╯al., 2011; Chapters 17–21, this volume). In the long term, a new international treaty on nitrogen could provide a solution for coordinated global or regional nitrogen management. Against the possible attractiveness of such a treaty must be balanced the overlap with existing MEAs and IGOs, which would be considerable. The requirement for coordination and formal interlinkages with existing bodies would make both negotiation and implementation difficult. Nitrogen involves many policy sectors and there would be strong pressures for any instrument to be harmonized with actions by existing MEAs and IGOs. Furthermore, many countries are concerned about the proliferation of MEAs and are reluctant to negotiate new instruments that consume national resources both in setting up and for implementing nationally. As well as this, some countries do not, as yet, perceive nitrogen to be a major environmental problem. The case for a new instrument would need to be very strong even to initiate discussions on possible negotiations. Even so, possibilities for a framework or coordinating convention could still be explored. This could lay down basic principles that might facilitate inter-MEA and IGO action coordination. Another alternative, also worthy of consideration, is a joint protocol between two or more conventions. Such an approach has the advantage of building on the work of existing MEAs and IGOs, while focusing on the common links between them specifically related to nitrogen. This option is explored further below, under formal high-level agreements between MEAs and IGOs (Section 25.3.2). A new international body for coordinating or promoting cooperation on nitrogen management is perhaps a more feasible option than a full MEA dedicated to nitrogen management. This could bring together the various institutions that have interests in nitrogen management and promote discussion and cooperative action. The UN has established such a body for water issues, ‘UN Water’, which brings together 26 bodies from the UN system, as well as external partners representing organizations and civil society (UN-WATER, 2010). It was established following the 2002 World Summit on Sustainable Development with the aim of supporting States to reach the water-related Millennium Development Goals.

576

Another UN coordinating body with a much broader remit is the Environmental Management Group (EMG). This is a UN system-wide body with a membership consisting of the specialist agencies, programmes and organs of the UN including the secretariats of the MEAs (Environmental Management Group, 2010). It is chaired by the Executive Director of UNEP and supported by a secretariat provided by UNEP. The EMG aims to further cooperation in support of the implementation of the international environmental and human settlements agenda. On specific issues it does this through Issues Management Groups (IMGs), for example, a report of an IMG on atmosphere and air pollution was submitted to the fourteenth session of the Commission on Sustainable Development (CSD) in 2006 (Environmental Management Group, 2006). One of the conclusions of this report was that organizations at the technical level lacked awareness of one another’s activities and programmes. While this was referring to air pollution, it also highlights the dilemma faced in dealing with the nitrogen problem. The report suggested a UN Technical Forum on Air Pollution Activities from Urban to Global Scales to be held every few years. A similar forum could be useful for nitrogen management. A recent initiative which could provide future potential for cooperation is the Global Partnership on Nutrient Management. It was launched at the time of the seventeenth session of the Commission on Sustainable Development in May 2009 with the support of the United States and Dutch governments and leading stakeholders, and with UNEP providing secretariat support. It is open to States and organizations with interests in nutrients, including nitrogen. The Partnership recognizes the need to optimize the use of nutrients to realize food security, while minimizing negative impacts on the environment and human health. It aims to raise awareness of these issues, to build political and stakeholder interest and impetus, to assist countries through exchange of knowledge and good practices, and to foster action (United Nations Environment Programme, 2009). At a more scientific level, the International Nitrogen Initiative (INI) has promoted interest in the global nitrogen problem (INI, 2010). Both of these institutions could provide the necessary stimulus for a more formal global agreement in the future.

25.3.2╇ Formal high-level agreements between MEAs/IGOs The highest level of agreement between MEAs, or between IGOs, is that taken by formal agreement between the governing bodies concerned. The resulting action might be high level or take place at a more practical level, but it is seen to have the backing of the governing bodies through the decisions of the parties. It is possible, for example, for two MEA governing bodies to decide upon action to develop a separate, new MEA that has particular relevance to them both. There is no example for this in nitrogen management, but an example of such possibilities is given with the case of the governing bodies of the UNECE Water Convention and the UNECE Industrial Accidents Convention. These negotiated jointly and adopted the 2003

Keith Bull

Protocol on Civil Liability and Compensation for Damage Caused by the Transboundary Effects of Industrial Accidents on Transboundary Waters (Economic Commission for Europe 2003b). The relationship between UNFCCC and IPCC is an example of an interlinkage between an MEA and an IGO which has some relevance to nitrogen management. While the Convention text merely indicates ‘The head of the interim secretariat… will cooperate closely with the Intergovernmental Panel on Climate Change to ensure that the Panel can respond to the need for objective scientific and technical advice’, some would believe the link is much stronger. The Convention’s Kyoto Protocol makes stronger reference to IPCC and also recognizes other MEAs (1987 Montreal Protocol) and IGOs (e.g., International Maritime Organization). While such links might be formally agreed, it does not mean that they will be effective. The relationship between UNFCCC and IPCC has been criticized since it affords poor links with natural, economic and social sciences and the assessment process of IPCC is slow and unable to respond quickly to the demands from UNFCCC (Raes and Swart, 2007). Formal links between an MEA and IGO can also be established through adoption of a specific MEA. For example, the UNECE Water Convention negotiated its 1999 Protocol on Water and Health in collaboration with WHO (Economic Commission for Europe, 1999a). Continued collaboration is ensured through a joint UNECE/WHO secretariat as well as meetings of delegates from both water and health sectors. Inter-MEA/IGO collaboration need not be through adoption of a protocol. From its early years, the CLRTAP governing body, the Executive Body, looked to WHO to provide information on the impacts of air pollution on human health, though without making any formal decision on this. However, in 1997, the Executive Body decided to establish, with WHO agreement, the Joint Task Force on the Health Effects of Air Pollution€– a task force of the Executive Body and WHO. There is a strong link between the UNECE Water Convention and FAO, since it is recognized that run-off from agriculture can strongly influence the quality of water in rivers and lakes. FAO as an organization could even be seen as a possible coordinator of a more integrated approach. However, while its scope is large, it is, by definition, limited to the agricultural sector, so while specific inter-institutional links might help coordinate particular aspects of nitrogen management, they are unlikely to offer the more holistic solutions of other approaches. Collaborative agreements between MEAs and/or IGOs, on, for example, exchange of data or joint action, often take the form of memorandums of understanding (MOUs) or even simple exchanges of letters. Such simple mechanisms could do much to promote collaborative action on nitrogen management between MEAs and IGOs. There are plenty of examples of these in other areas, for example, CLRTAP has collaborative agreements with the European Environment Agency and the Oslo and Paris Commission for the Protection of the NorthEast Atlantic (OSPAR, 2010). The Convention on Biological Diversity (CBD), which is also concerned about nitrogen impacts on biodiversity, is

particularly interested in interlinkages and cooperation. The Conference of the Parties (COP) has adopted numerous decisions directly pertaining to cooperation with other conventions, organizations and processes, and significant elements of cooperation are included in goal 1 of its Strategic Plan. So, for example, the Convention has signed Memorandums of Cooperation and Joint Work Programmes with many of its partners and it hosts a joint website of biodiversity-related conventions (Convention on Biological Diversity, 2010). CBD is also discussing a target for 2020, where Parties bring pollution from excess nutrients (nitrogen and phosphorus) below critical ecosystem loads; this initiative could be the basis for coordinating activities with other bodies. Under CLRTAP there have been discussions on the benefits of using biodiversity as an indicator of achievement, Â�particularly with respect to nitrogen impacts, as discussed in Chapter 20 (Dise et╯al., 2011, this volume). Until now, critical loads maps of Europe have been used to identify the benefits of emission control and these proved a persuasive argument for policy action (Bull, 1995). Using biodiversity indicators would link CLRTAP more closely to the efforts of the CBD. This would require more coordination between the two MEAs, which might be achieved through a simple collaborative agreement. CLRTAP has long had an awareness of nitrogen management problems and how they relate to air pollution. It established the Sofia Protocol on emissions of nitrogen oxides in 1988, and the multi-pollutant Gothenburg Protocol in 1999 which included both NOx and NH3 emissions (Bull and Sutton, 1998; Economic Commission for Europe, 1988, 1999b). The convention has since focused increasing attention on the aspects of nitrogen management that fall within its scope, and promoted scientific and technical work to support its decision making. In December 2007, the Executive Body for the Convention established its Task Force on Reactive Nitrogen (Economic Commission for Europe, 2007b; TFRN, 2010). The Task Force was given the long-term goal of ‘developing technical and scientific information, and options which can be used for strategy development across the UNECE, to encourage coordination of air pollution policies on nitrogen in the context of the nitrogen cycle and which may be used by other bodies outside the Convention in consideration of other control measures’. This is an example of option (a) referred to in Section 25.3 where an MEA is seeking to extend its scope to link with other MEAs/ IGOs, though still exercising due care to ensure it is not seen to attempt to encroach on the work of other bodies, but simply to make information available for use. The UNECE Water Convention actively collaborates with other institutions that have interests in the management and protection of water. For example, it participates in UN-Water through UNECE, which is a member organization, and has invited the Ramsar Convention on Wetlands to participate in its programme ‘Nature for Water’. Both activities have some relevance for nitrogen management albeit relatively minor. UN Water is particularly relevant as an example institution, however, since it demonstrates how it is possible to draw together a wide range of UN organizations to discuss common issues such as water (UN-WATER, 2010). UN-Water is not an

577

Coordinating European nitrogen policies

implementing body, its specific activities and programmes are hosted by individual member agencies on behalf of UN-Water, but it is a good example of a body promoting cooperation and coordination. A permanent Secretariat, hosted by the United Nations Department for Economic and Social Affairs (UNDESA) in New York, provides administrative, technical and logistical support. Finally, there is the case of institutions like the EU that are regional economic integration organizations, which have competence to act on behalf of their member States. Using its regional economic integration status the EU has been an active participant and a major driving force in regional and global policy-making through its membership of IGOs, and through being party to many regional and global MEAs. It is therefore in a key position to promote cooperation and coordinate action on nitrogen in Europe. Each of the above high-level approaches, including those mentioned in Section 25.3.1, has the potential to form strong coordination links between organizations and their institutions. While they do not automatically ensure that cooperation and coordination will take place, they do provide mechanisms to enable institutions, their various bodies and their member countries to develop practical links and to take the necessary action for coordination to be effective. Even so, persuading governing bodies of MEAs and IGOs to enter into such agreements is seldom easy, and the difficulties are compounded when more than one organization is involved in the decisionmaking process. New initiatives or changes to existing plans are often viewed with scepticism by many countries and there is always caution when such steps involve the need for additional resources. Usually it is a question of appropriate timing for action to be successful. Kingdon (1995) has drawn attention to the need for a ‘policy window’ or ‘window of opportunity’ where the problem stream (identified by the science), the policy stream (the policy action, e.g., legal instrument) and the political stream (political will) need to come together at the same time for highlevel action to be taken. This multiple stream theory, in contrast to a stage by stage concept, is widely applicable and goes a long way to explain why some items find their way on to the political agenda and others do not. Kingdon has noted that at critical points in time, the streams are coupled by ‘policy entrepreneurs’. He has suggested that the combination of all three streams into a single package enhances dramatically the chances that an issue will receive serious attention by policymakers. The key to understanding Kingdon’s argument is to see streams not as additive, but as interactive. Choice is determined not by the effects of each stream in isolation, but by the impact of one depending on critical values of the others. Kingdon’s theory has been applied widely, even outside environmental issues (e.g. Blackman (2005) examined tobacco control in California drawing upon Kingdon’s model). Of more relevance to the issue of nitrogen management, Brunner (2008) considered multiple streams in relation to emissions trading in Germany. In relation to MEAs, the signing of CLRTAP in 1979 demonstrated the need for political attention and awareness,

578

even when the scientific evidence for action was overwhelming and the policy action needed was clear. In conclusion, it is clear that there are many opportunities for high-level agreements to bring MEAs and IGOs together to address the nitrogen issue, but the arguments for such agreements must be persuasive and timely if they are to succeed.

25.3.3╇ Links through participation in meetings or joint actions While joint meetings between MEA and/or IGO bodies are unusual they can occur and they could provide a mechanism for discussing and agreeing action on issues of common interest. Moreover, they have possibilities for developing mutual understanding of common issues between delegations that are usually focused upon the particular interests of just one MEA or IGO. Such joint meetings, because of their unusual nature, are more likely to lead to further joint agreements or agreed joint action, and hence promote much better cooperation and coordination of activities. More commonly, an MEA or IGO will invite participants from other MEAs and IGOs to attend its meetings that it believes are of common interest. The other MEAs and IGOs might be represented by one or more of its officers, delegations or experts charged with representing the MEA/IGO, or by a member of its secretariat. Similarly, non-governmental organizations (NGOs) may also be invited to attend MEA or IGO meetings, and these too may play an important role in linking MEAs and IGOs through participation in the bodies of several institutions. NGO networks can be particularly effective in drawing attention to common interests, synergies or trade-offs between institutions. For example, the Global Atmospheric Pollution (GAP) Forum, registered as an NGO with CLRTAP, promotes inter-regional collaboration between the various regional air pollution networks, including CLRTAP, in the absence of any other forum for bringing all the regions together. Such MEA/IGO links as described above can be particularly effective mechanisms for exchanging views and information, and a significant help to coordination between institutions. The greater degree of informality in such arrangements makes them more acceptable, or even welcomed, by participating countries. However, the effectiveness of such mechanisms once again relies upon the availability of resources and in particular the willingness of individuals from one institution to commit time, not only to attend meetings of others, but also to make appropriate contributions to their work. In addition, it is important that the institution or body inviting participation ensures that it gives due consideration to the information presented by those invited.

25.3.4╇ Links through scientific and technical bodies Because of the importance of the scientific underpinning of MEAs, science can provide a persuasive mechanism for cooperation between bodies and for coordination of subsequent action.

Keith Bull

The role of science in environmental regimes has been a topic for a number of studies (Lidskog and Sundqvist, 2002; Farrell et╯al., 2001; Tuinstra et╯al., 2006) and the links between the science and the policy within an institution have much bearing on the contribution science can make to policy development. Lidskog and Sundqvist even suggest that co-production of science and policy is a prerequisite for creating effective environmental regimes. How scientific assessments learn and develop also influences how the links between science to policy are forged and how influential science can be to policy decisions. Comparison between CLRTAP and IPCC in the past has highlighted the differences between the two insitutions and the importance of issues such as storage of knowledge, dissatisfaction and conflict, media coverage, formal and informal communication, as well as the state of the art in science (Siebenhüner, 2002). Science usually provides the underpinning for the development of MEAs and IGOs, indeed scientific knowledge and understanding is normally the prerequisite to get any formal agreement between States to take action on environmental issues. CLRTAP and UNFCCC both evolved after a great deal of scientific work and debate, though the creation of the MEAs provided a much needed focus for further scientific studies to give underpinning to subsequent action in the form of protocols. Partly as a result of the decisions taken by parties to an MEA, scientific and technical work will evolve in a way that provides the required evidence of an environmental problem and provides underpinning for further decision making. Recognizing the disparity of scientific development under different institutions, the options below have the potential for creating interlinkages to address nitrogen management. Scientific and technical links between MEAs/IGOs may occur through either formal and/or informal mechanisms. Parties to MEAs will sometimes encourage formal cooperation between their scientific and technical bodies and those of other organizations. This is especially the case when they have an awareness of possible common interests or where there might be practical benefits to the parties, e.g., the harmonization of reporting to avoid duplication of effort and to maximize available national resources. Such cooperation at a formal level is only likely to be successful where there are clear common interests and scientists are willing to participate, since parties will need to be convinced that resources should be used in this way. For example, OSPAR encourages cooperation with many international and regional organizations on science and research, monitoring and assessment, as well as the promotion of actions where the competence for such actions is vested with other organizations or is most efficiently taken in their frameworks. Even so, the science of the nitrogen cycle remains a challenge, so scientific interest continues. This ensures that nitrogen, at least in scientific circles, has a high profile and stands a better chance of finding its way onto the political agenda. Even so, there are barriers to overcome in particular those related to a lack of coordination. The different nitrogen measurement and modelling activities between air, land and water need to be brought together and harmonized. The relevant, multi-media nitrogen science is needed to provide cross-cutting information to underpin policy decisions.

In Europe, there has been major progress through two coordination programmes, COST 729 and ESF NinE, which are both created from scientific research to work towards integrated nitrogen approaches and policies at the European level. COST (European Cooperation in Science and Technology) is an intergovernmental framework that complements the activities of the EU framework programme, and the COST 729 project on assessing and managing nitrogen fluxes in the atmospherebiosphere in Europe has provided an excellent mechanism for bringing scientists together from across Europe to tackle the common issues related to nitrogen management (COST 729, 2009). The European Science Foundation (ESF) Nitrogen in Europe (NinE) project is a research networking programme that addresses nine interacting problems affected by excess nitrogen in the environment (e.g., aquatic, coastal, terrestrial, ozone; see Sutton et╯al., 2011, Chapter 5, this volume). Again this has provided an important networking facility for scientists involved in the nitrogen problem in Europe (NinE, 2010). One key to coordination is to make effective use of the ‘programme centres’ established by MEAs. These generally work under scientific or technical bodies to provide additional, centralized resources for their scientific and technical activities, and can provide a good cooperation mechanism. Such centres might hold the parties’ common databases or might provide expert guidance to the parties’ scientific community, e.g., for monitoring. They are often an important resource for implementing an MEA and the parties may support funding mechanisms for a centre’s activities. However, a centre might also provide, with the parties’ approval, important data and information to other institutions where there are overlapping scientific interests. CLRTAP centres have shared information with other regions and have shared data with other regional MEAs such as the regional sea commissions and the Nordic Council’s Arctic Monitoring and Assessment Programme (AMAP) (e.g. AMAP, 2006; OSPAR, 2007; Bartnicki et al., 2008). This not only adds to the credibility of a programme centre, but ensures that harmonized data and information are used by the different institutions€– an obvious benefit to developing harmonized national and international policy. Centres, or individuals from them, might even publish reports drawing attention to linkages and synergies between different MEAs to stimulate cooperation (Amann, 2003). Scientific bodies of international institutions involve national expert scientists, who usually have much broader scientific interests than those of a single institution. There is often a wealth of scientific knowledge in a scientific body that broadens the base of understanding of the science of that body and increases the possibilities for exchange of information between institutions. Overlapping or similar interests are readily identified, and recommendations for collaboration and coordination with other institutions can be forwarded from the scientific body to an institution’s decision-making body. Some scientific organizations have been set up specifically to bring together and coordinate the work of other institutions; such a coordinating institution might be seen as an attempt to create an over-arching coordinating body. Reference was made above to UN Water, but there are other examples. The Group for Earth Observations (GEO) was set up at the first

579

Coordinating European nitrogen policies Figure 25.1 The Global Earth Observation System of Systems (GEOSS). (Image courtesy of ‘The Group on Earth Observations’.)

Earth Observation Summit in 2003 with a view to establishing a comprehensive and sustained earth observation system or systems (GEO, 2009). GEO is coordinating international efforts to build a Global Earth Observation System of Systems (GEOSS) and develop instruments and systems for monitoring and forecasting changes in the global environment (Figure 25.1). It currently has 79 member countries (and the EU) as well as 56 participating organizations, including UNEP, CLRTAP and WMO. In a similar way, a new international coordinating body for nitrogen management could bring together many of the MEAs, IGOs and other stakeholders with an interest in the topic; such a body could be scientifically or policy focused. As noted earlier, the Global Partnership on Nutrient Management may have potential to develop into a global coordinating body for nitrogen management issues. Some organizations that bring scientific networks together do so through funding projects that have common interests. As noted above, the ESF NinE programme and COST 729 have played key roles in linking networks in Europe. Globally, the International Council for Science (ICSU), the Scientific Committee on Problems of the Environment (SCOPE) and the International Geosphere-Biosphere Programme (IGBP) play important roles in developing scientific networks in a variety of scientific areas, with the International Nitrogen Initiative being a joint project of these organizations (SCOPE, 2010; IGBP, 2010; ICSU, 2010). The international scientific community at large and the extensive numbers of formal and informal scientific networks are a valuable resource to all MEA/IGO scientific bodies. Sometimes, when an MEA establishes a scientific body, it recognizes that there needs to be interaction with other communities and networks with similar interests and common goals. The CLRTAP Task Force on Reactive Nitrogen includes in its mandate the need to coordinate with other bodies under the Convention. But the mandate also recognizes that ‘different aspects of the nitrogen cycle are considered separately under different regulatory frameworks’ and that ‘the nitrogen cycle is

580

multimedia in nature, and that it may be beneficial to have fully informed or coordinated regulatory frameworks to address various aspects and issues’ (Economic Council for Europe, 2007b). While CLRTAP parties make reference to specific inter-institutional links in the Task Force mandate, it is implicit that such interlinkages need to be explored and developed where needed. If an MEA is to ‘step outside’ its mandate it may be able to explore ‘external’ opportunities for cooperation. Parties are likely to find this more acceptable if it is done at the scientific level. The CLRTAP Task Force on Hemispheric Transport of Air Pollution (TFHTAP) is a good example of this. Established under one of the Convention’s main scientific bodies, the Steering Body of EMEP3, the Task Force seeks to understand the movement of air pollution at the hemispheric scale. While the Convention has limited its policy interests to transboundary air pollution between countries in the UNECE region, the Task Force has successfully engaged with national experts and international institutions from outside the region to help it understand the hemispheric movement of pollutant emissions (see TFHTAP, 2007). The Parties to the Convention agreed to this as they were persuaded that such understanding was needed in order to explain pollution levels within the UNECE region. Science has played an important role, not just in the development and implementation of many MEAs, but also in promoting coordination between them and other institutions. It is easy to underestimate the effort and resources needed for scientists and scientific organizations to play effective roles across more than one scientific forum. It is possible to forget the importance of scientific literature and scientific conferences, seminars and meetings; these may not appear to be directly linked to the activities of MEAs/IGOs yet, nevertheless, they enable scientists to share information and forge links that have relevance to more than one institution. Nitrogen management is a particular challenge to science and, because of the Co-operative programme for monitoring and evaluation of the long range transmission of air pollutants in Europe.

3

Keith Bull

complexity of its biogeochemistry and the need to consider science across several policy sectors, it is especially demanding on scientific resources and effort.

25.3.5╇ Common parties or membership There are many scientific networks, some not even specifically linked to MEAs and IGOs, which may promote effective sharing of scientific knowledge between MEAs and IGOs. However, analogous policy networks are far fewer and less able to offer links between different institutions. Nevertheless, policy coordination is essential if such institutions are to make sound decisions that are complementary, consistent and nonconflicting. An apparently obvious way to achieve links between policies and institutions is through national delegations where a country belongs to more than one institution. However, delegations to different institutions are seldom the same individuals and they may even be drawn from different government departments or divisions. But it is important, for both national and international policy development, that there is effective dialogue between delegates to institutions with overlapping interests. While this is seemingly obvious, in practice it requires effort and planning, especially when government structures do not lend themselves to national coordination. For example, with the increasingly perceived importance of climate change, governments are setting up new ministries or departments to tackle climate issues; this is likely to separate those working with UNFCCC and the Kyoto Protocol from those working in areas which have possible synergies or interlinkages with climate change, such as air pollution, forestry and biodiversity€– and all these are relevant to nitrogen management. It is in a country’s own interests that its national policies are coordinated and harmonized with respect to the international agreements by which it is bound. Conflicting obligations can result in major problems in implementation and it is better to deal with such conflicts at the institutional level while agreements are being drawn up. National delegations are key to ensuring consistent and harmonized action. Non-governmental organizations (NGOs) also have an important role in linking MEAs and IGOs through their participation in the bodies of several institutions where they can highlight common interests between institutions. NGOs and other stakeholders, such as industry, may have much to offer in helping to address nitrogen management issues.

25.3.6╇ Links through secretariats The secretariats of MEAs and IGOs are often a key to successful cooperation and coordination. As well as organizing the day-to-day operations of governing and subsidiary bodies, secretariats are in a good position to have an overview of the work of their institution and identify how it relates to the work of others. They are also in an excellent position to inform the appropriate body of the institution of possible action needed and to initiate such action when appropriate. Using the mechanisms and bodies of institutions, secretariats can, provided that they have the necessary resources and appropriate expertise,

stimulate and encourage cooperation at all levels and keep the parties and member states informed of action taken. For example, OSPAR has established a direct institutional link through its Secretariat between the OSPAR Commission (OSPAR, 2010) and the North Sea Conferences (NSC, 2010). Through this, OSPAR has taken on the follow-up to the last Gothenburg Conference in 2006; the North Sea Network of Investigators and Prosecutors (NSN) established under the North Sea Conferences, which works to protect the marine environment from pollution by shipping; and the 1983 Bonn Agreement (1998) for cooperation in dealing with pollution of the North Sea by oil and other harmful substances. Similarly, CLRTAP has charged its secretariat with ‘outreach’ activities, and in particular to ‘coordinate the dissemination of information and take an active part in raising awareness in other regions’ (Economic Commission for Europe, 1999c). Through this mechanism the knowledge and experience of the Convention is being shared with regions outside Europe and North America.

25.4╇ Conclusions and recommendations (a) International treaties, such as conventions and their protocols, and especially MEAs, have done much to protect the global environment through promoting intergovernmental action on many environmental issues. MEAs and IGOs between them have, in recent years, targeted most of the known environmental problems, but none has targeted nitrogen management policy holistically. (b) There is no formal over-arching body to coordinate global action on nitrogen management between the various MEAs and IGOs that have interests in the matter, though the Global Partnership on Nutrient Management (GPNM) and the International Nitrogen Initiative (INI) might provide a stimulus for this in the future from policy and scientific perspectives, respectively. A new international treaty on nitrogen might seem an effective solution for global or regional nitrogen management, but this would be complex to negotiate. It would involve many policy sectors and it would need to be harmonized with ongoing and planned work by existing MEAs and IGOs. Even so, a simple framework agreement or joint protocol between two or more MEAs could offer a way ahead and should be further explored. (c) The development of a new treaty starts with the scientific underpinning of the issues and approaches. Many successful MEAs have developed from scientific concerns on perceived environmental problems. For nitrogen, this European Nitrogen Assessment helps stimulate the process and it is worthwhile extending this to more regions and the global level. In Europe, COST 729 and the ESF NinE programme have initiated integrated nitrogen approaches by bringing the science together, which helped stimulate the establishment of the TFRN. Globally the same approach was done by INI which helped establish the GPNM.

581

Coordinating European nitrogen policies

(d) While several conventions have taken action individually to develop nitrogen policy in their specific areas of interest, they have been unable or unwilling to increase their scope of action beyond their agreed mandates. Even so, there are many established mechanisms for developing inter-MEA/ IGO links (see points (e) to (j) below) and these should be further explored to help develop new linkages and provide more harmonized and coordinated approaches to nitrogen management. (e) Coordination between MEAs and IGOs already occurs in a variety of ways and at different levels, but the focus is inevitably on areas of overlapping interests. Existing mechanisms for coordination, for example through formal agreements, joint participation in meetings or projects, and actions by convention secretariats, might be applied in relevant bodies with nitrogen management interests. They have the potential to improve harmonization and promote effective coordination. However, it can be difficult to influence action taken by conventions, IGOs and the EU, especially if it involves new initiatives or changes to existing plans. Political willingness is an important factor and this can be influenced by timely scientific and public pressure. (f) Scientific and technical cooperation between MEAs has proved especially important; the international scientific community is able to provide harmonized information to different forums and thus promote coordination. Even so, there remain problems in coordinating different nitrogen measurement and modelling activities between air, land and water. There is a need to bring together relevant, multi-media nitrogen science to provide cross-cutting information to underpin policy decisions. Coordinating the work of the scientific community to address the needs of several forums requires effort and resources and is not achieved easily, especially since there are still scientific challenges for effective nitrogen management. In addition, the links between science and policy need to be effective if science is to be truly influential. (g) For the UNECE region (Europe, North America and Central Asia), the scientific bodies of the Convention on Long-range Transboundary Air Pollution (CLRTAP) and its Task Force on Reactive Nitrogen (TFRN) provide some coordination of efforts for addressing nitrogen management. They are in a good position to promote more cooperation and take further steps on some aspects of coordination to link atmospheric with other nitrogen threats. (h) Considering a wider view of the UNECE, five MEAs have been established on different aspects of the environment, and there remains the potential for nitrogen management to be taken up as an opportunity for linking these issues. The overarching Committee on Environmental Policy (CEP) and the linked series of ministerial conferences ‘Environment for Europe’, could play an important role in developing the momentum for establishing more joined-up approaches.

582

(i) The EU, itself established by international treaties, has a major role in Europe to harmonize policy in EU Member States and coordinate their actions regionally and globally. By being proactive in its interlinkages role, it can utilize the resources of other institutions and help coordinate the development of policy with existing MEAs/IGOs. (j) National policy coordination, already developed by some countries, can not only ensure that national policies are developed consistently, but can also play an important role, through national delegations, in coordinating international action between MEAs and IGOs.

Acknowledgements The authors are especially grateful to Andrea Weiss for providing information on OSPAR, to Jan Willem Erisman for many useful suggestions on the manuscript, and to various stakeholders for contributions during the 5th ENA workshop on ‘European Nitrogen Policies and Future Challenges’. This work was in part supported by the Nitrogen in Europe (NinE) programme of the European Science Foundation, COST Action 729, NitroEurope IP (funded by the European Commission), and the Task Force on Reactive Nitrogen, with the latter supported by the UK Department for Environment Food and Rural Affairs (Defra) and the UK Centre for Ecology and Hydrology.

References AMAP (2006). Arctic Pollution 2006 (Acidification and Arctic Haze). Arctic Monitoring and Assessment Programme (AMAP), Oslo, Norway. Amman, M. (2003). Linkages and synergies of global, regional and local emission controls. In:€Air Quality: Assessment and Policy at Local, Regional and Global Scales, 14th International Conference of IUAPPA and CAPPA, Dubrovnik, April, 2003, pp. 55–60. Bartnicki, J., Gusev, A., Aas, W., Fagerli, H. and Valiyaveetil, S. (2008). Atmospheric Supply of Nitrogen, Lead, Cadmium, Mercury and Dioxines/Furanes to the Baltic Sea in 2006. EMEP Centres Joint Report for HELCOM. EMEP/MSC-W Technical Report 3/2008. HELCOM, Helsinki. Blackman, V. S. (2005). Putting policy theory to work:€tobacco control in California. Policy, Politics, and Nursing Practice, 6, 148–155. Bleeker A., Sutton M. A., Acherman B. et╯al. (2009). Linking ammonia emission trends to measured concentrations and deposition of reduced nitrogen at different scales. In:€Atmospheric Ammonia:€Detecting Emission Changes and Environmental Impacts, ed. M. A. Sutton, S. Reis and S. M. H. Baker, pp. 123–180, Springer, New York. Bonn Agreement (1998) www.bonnagreement.org (accessed July 2010). Brunner, S. (2008). Understanding policy change:€multiple streams and emissions trading in Germany. Global Environmental Change, 18, 501–507. Bull, K. R. (1995). Critical loads: possibilities and constraints. Water, Air and Soil Pollution, 85, 201–212. Bull, K. R. and Sutton, M. A. (1998). Critical loads and the relevance of ammonia to an effects based nitrogen protocol. Atmospheric Environment, 32, 565–572.

Keith Bull Butterbach-Bahl, K., Nemitz, E., Zaehle, S. et╯al. (2011). Nitrogen as a threat to the European greenhouse balance. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Cellier, P., Durand, P., Hutchings, N. et╯al. (2011). Nitrogen flows and fate in rural landscapes. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. COST 729 (2009). Assessing and Managing Nitrogen Fluxes in the Atmosphere–Biosphere System in Europe. http://cost729.ceh. ac.uk/ (accessed July 2010). Convention on Biological Diversity (2010). www.cbd.int/cooperation and www.cbd.int/blg (accessed July 2010). Dise, N. et╯al. (2011). Nitrogen as a threat to European terrestrial biodiversity. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Economic Commission for Europe (1988). The Sofia Protocol concerning the Control of Emissions of Nitrogen Oxides or their Transboundary Fluxes. www.unece.org/env/lrtap/nitr_h1.htm (accessed July 2010). Economic Commission for Europe (1999a). Protocol on Water and Health. www.unece.org/env/water/text/text_protocol.htm (accessed July 2010). Economic Commission for Europe (1999b). The Gothenburg Protocol to Abate Acidification, Eutrophication and GroundLevel Ozone. www.unece.org/env/lrtap/multi_h1.htm (accessed July 2010). Economic Commission for Europe (1999c). ECE/EB.AIR/68, Annex III (Decision 99/2) www.unece.org/env/documents/1999/eb/ece. eb.68.e.pdf (accessed July 2010). Economic Commission for Europe (2000). Review of Synergies to Be Derived from Closer Cooperation, Report to the Committee on Environmental Policy, CEP/2000/1. Economic Commission for Europe (2003a). Guidelines for Strengthening Compliance with and Implementation of Multilateral Environmental Agreements in the ECE Region, Document submitted to the 5th Ministerial Conference ‘Environment for Europe’. ECE/CEP/107. Kiev, Ukraine. Economic Commission for Europe (2003b). www.unece.org/env/civilliability/protocol.html (accessed July 2010). Economic Commission for Europe (2007a). Implementation of UNECE Multilateral Environmental Agreements, Document submitted to the 6th Ministerial Conference ‘Environment for Europe’. ECE/BELGRADE.CONF/2007/12. Belgrade, Serbia. Economic Commission for Europe (2007b). Decision 2007/1, Establishment of a Task Force on Reactive Nitrogen. ECE/ EB.AIR/91/Add.1. www.unece.org/env/lrtap/ExecutiveBody/ Eb_decision.htm (accessed July 2010). Environmental Management Group (2006). Atmosphere/Air Pollution:€Best Practices, Lessons Learnt and Case Studies by UN Cooperative Programmes and Activities, Report by the UN Issues Management Group of EMG. DESA/DSD/2006/5. Environmental Management Group (2010). www.unemg.org (accessed July 2010). Farrell, A., VanDeveer, S. D and Jäger, J. (2001). Environmental assessments:€four under-appreciated elements of design. Global Environmental Change, 11, 311–333. GEO (2009). Group on Earth Observations. www.earthobservations. org (accessed July 2010).

Grizzetti, B., Bouraoui, F., Billen, G. et╯al. (2011). Nitrogen as a threat to European water quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. ICSU (2010). International Council for Science. www.icsu.org (accessed July 2010). IGBP (2010). International Geosphere Biosphere Programme. www. igbp.net/ (accessed July 2010). INI (2010). International Nitrogen Initiative. www.initrogen.org (accessed July 2010). Inomata, T. (2008). Management Review of Environmental Governance within the UN System, JIU/REP/2008/3, United Nations, Geneva. www.unjiu.org/data/reports/2008/en2008_3.pdf (accessed July 2010). Jensen, L. S., Schjoerring, J. K., van der Hoek, K. et al. (2011). Benefits of nitrogen for food, fibre and industrial production. In: The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge Unversity Press. Kingdon, J. W. (1995). Agendas, Alternatives, and Public Policies, 2nd edn.,€Longman New York:. Lidskog, R. and Sundqvist, G. (2002). The role of science in environmental regimes:€the case of LRTAP. European Journal of International Relations, 8, 77–101. Moldanová, J., Grennfelt, P., Jonsson, Å. et╯al. (2011). Nitrogen as a threat to European air quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. NinE (2010) Nitrogen in Europe:€Current Problems and Future Solutions. www.nine-esf.org (accessed July 2010). NSC (2010). North Sea Commission. www.northseacommission.info (accessed July 2010). Oenema, O., Bleeker, A., Braathen, N. A. et╯al. (2011a). Nitrogen in current European policies. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Oenema, O., Salomez, J., Branquinho, C. et╯al. (2011b). Integrated approaches to nitrogen management. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. OSPAR Commission (2007). Atmospheric nitrogen in the OSPAR Convention Area in 1990–2004. OSPAR Publication 344/2007. OSPAR, London. OSPAR Commission (2010). www.ospar.org (accessed July 2010). Raes, F. and Swart, R. (2007). Climate assessment:€what’s next? Science, 318, 1386. Schrage, W, Bull, K. R. and Karadjova, A. (2007). Environmental legal instruments in the UNECE region. Yearbook of International Environmental Law, 18, 3–31. SCOPE (2010). Scientific Committee on Problems of the Environment. www.icsu-scope.org/ (accessed July 2010). Siebenhüner, B. (2002). How do scientific assessments learn? Part 2. Case study of the LRTAP assessments and comparative conclusions. Environmental Science and Policy, 5, 421–427. Spranger T., Klimont Z., Sponar M. et╯al. (2009). Ammonia policy context and future challenges. In:€Atmospheric Ammonia:€Detecting Emission Changes and Environmental Impacts, ed. M. A. Sutton, S. Reis and S. M. H. Baker, pp. 435–444, Springer, New York.

583

Coordinating European nitrogen policies Sutton, M. A., Howard, C. M., Erisman J. W. et╯al. (2011) The need to integrate nitrogen science and policies. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Svirejeva-Hopkins, A., Reis, S., Magid, J. et╯al. (2011). Nitrogen flows and fate in urban landscapes. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. TFRN (2010). Task Force on Reactive Nitrogen. www.clrtap-tfrn.org (accessed July 2010). TFHTAP (2007). Hemispheric Transport of Air Pollution 2007, Air Pollution Studies No. 16. United Nations,€New York. Tuinstra, W., Hordijk, L. and Kroeze, C. (2006). Moving boundaries in transboundary air pollution, co-production of science and policy under the Convention on Long-Range Transboundary Air Pollution. Global Environmental Change, 16, 349–363.

584

United Nations (1999). Inter-Linkages:€Synergies and Coordination between Multilateral Environmental Agreements. United Nations University, Tokyo. United Nations Environment Programme (2009). Global Partnership on Nutrient Management. www.gpa.unep.org/content. html?id=418&ln=6 (site accessed July 2010). United Nations University (2002). www.unu.edu/inter-linkages (accessed July 2010). UN-WATER (2010). United Nations Water Partnership. www. unwater.org (accessed July 2010). Velthof, G. et╯al. (2011). Nitrogen as a threat to European soil quality. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Vestreng, V., Myre, G., Fagerli, H. et╯al. (2007). Twenty-five years of continuous sulphur dioxide emission reduction in Europe. Atmospheric Chemistry and Physics, 7, 363–368.

Chapter

26

Societal choice and communicating the European nitrogen challenge Lead author: Dave S. Reay Contributing authors: Clare M. Howard, Albert Bleeker, Pete Higgins, Keith Smith, Henk Westhoek, Trudy Rood, Mark R. Theobald, Alberto Sanz Cobeña, Robert M. Rees, Dominic Moran and Stefan Reis

Executive summary Nature of the problem (science/management/policy) • Increased public and institutional awareness of both the benefits and threats of nitrogen has the potential to greatly increase the efficacy of nitrogen policy. • Insufficient recognition of the financial, behavioural and cultural barriers to achieving an optimal nitrogen future risks policy antagonisms and failure. • Here we examine some of the key societal levers for and barriers to achieving an optimal nitrogen future in Europe, drawing lessons from the more-developed societal and policy challenge of climate change mitigation.

Key findings/state of knowledge • There is currently a very low level of public and media awareness of nitrogen impacts and policies. However, awareness is high regarding the threats and benefits of ‘carbon’ to society (e.g. energy use and enhanced climate change). • Many national climate change mitigation policies now overtly recognize the importance of societal choice, and are increasingly utilizing behavioural change strategies to achieve greenhouse gas emission reduction targets. • In achieving an optimal nitrogen future, lessons can and should be learned from existing climate change-focused communication and behavioural science (e.g. use of a ‘segmented strategy’ to reach disparate groups of stakeholders). • Key sectors where societal choice has the potential to greatly influence nitrogen use efficiency include food production, consumption and waste.

Major uncertainties/challenges • Public confusion/despair/apathy in a world of ‘carbon calculators’, ‘doomsday environmental scenarios’ and ‘decision support tools’ must be avoided. • Awareness and use of proven nitrogen communication tools for policy makers, media and public should be increased. • Succinct messages that get across the complexity of the nitrogen challenge while properly representing the balance of the pros and cons are required.

Recommendations (research/policy) • Nitrogen policy development and implementation should more overtly include understanding of social science and societal choice.

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

585

Societal choice

26.1╇ Introduction Whereas previous chapters of the assessment have focused on policy instruments and how these have been and could be used to address the issues posed by the ‘Nitrogen Cascade’ (Galloway et al., 2004), we turn our attention now to societal choice and the communication of the European Nitrogen Challenge. What role could society and changing behaviour play in reducing the threat of reactive nitrogen (Nr) to our environment and how might this be better enabled? This chapter aims to address this question€ – the question of ‘societal choice’, by identifying the ways in which positive behaviour change can be engendered, and the effect alterations in societal choice could have on the nitrogen challenge. Such changes are only possible if society and the individuals, communities and stakeholder groups of which it is comprised are aware of the issues and the need to make changes€– therefore this chapter also addresses the issue of ‘communicating’ the nitrogen problem. It examines the lessons to be learned from climate change-related communication methods, their relevance to nitrogen, and the complications faced therein. Finally, we present some tools and examples of nitrogen knowledge exchange which have been used to inform the non-specialist to date and suggest strategies for enhancing nitrogen-relevant knowledge exchange and awareness-raising in the future.

26.1.1╇ Human fingerprints on the global nitrogen cycle Previous chapters have provided a wealth of information on the anthropogenic changes in reactive nitrogen fluxes and issues in Europe, but to underline just how great is the extent of human society’s impact on the global nitrogen cycle it is useful to revisit the past, current and projected estimates of anthropogenic nitrogen fluxes around the world. More reactive nitrogen is now created each year by human activities than all natural sources combined. Alongside industrialization and the associated increases in emissions of nitrogen oxides (NOx) from fossil fuel burning, the intensification of agriculture and associated ammonia (NH3), nitrous oxide (N2O) and NOx emissions has led to a three to five fold increase in reactive nitrogen emissions over the last century (Denman et al., 2007). Overall global emissions of oxidized nitrogen (NOy) and reduced nitrogen (NHx) are mainly terrestrial in origin and in 2000 stood at 52.1 and 64.6 Tg N y−1, respectively (Dentener et al., 2006).

Human society and projected changes in nitrogen flux Both NOy and NHx emissions are predicted to further increase in many regions during the twenty-first century (Galloway et al., 2004), with population growth and dietary changes likely to be a key driver of alterations in the magnitude and geographical distribution of these emissions. Emissions of NOy from fossil fuel combustion are intrinsically linked to societal choice, through changing demand for domestic electricity use, space heating and private transport. However, these emissions may be most influenced in coming years by top-down drivers such as the Gothenburg protocol, and by national and international efforts to reduce fossil fuel combustion as a means to cut

586

greenhouse gas emissions (e.g. UNFCCC). Though demandside management will be a key part of national and regional strategies to reduce these emissions, decarbonization (and simultaneous ‘denitrogenization’) of supply is the central plank of current policy (e.g. UK Climate Change Act 2008). Given the large increases in population and demand for animal protein projected in some regions for 2030, it is likely that NHx emissions will become an increasingly important source of Nr deposition globally. Indeed, a 20% decrease in NHx deposition over Europe, but a 40–100% increase in Central and South America, Africa, and parts of Asia by 2030 is predicted (Dentener et al., 2006). Looking further into the future, Nr deposition over land may increase by a factor of 2.5 by 2100 (Lamarque et al., 2005), with a concurrent increase in deposition to marine systems of around 50% (Krishnamurthy et al., 2007). As such, it is on agriculture and food consumption that much of this chapter focuses and, in particular, on the role that behaviour change may play in reducing nitrogen wastage€ – the implications of human dietary choice are considered in more detail in Section 26.3.2.

26.2╇ Societal choice The concept of ‘societal choice’ is complex and contested, but is central to any debate on human responses to an environmental issue. Its importance, and the role that levels of awareness and education play in informing such choice, cannot be overstated when it comes to successfully implementing any policy, environmentally focused or otherwise. Blennow and Persson (2009) provide the specific example of forest owners in Sweden and their adaptive response to climate change. They show that the level of ‘belief ’ in climate change held by the forest owners was a crucial factor in determining the adaptation choices made. Such a finding highlights the key role that engagement and awareness can play in realizing change and it is on these drivers of societal choice, in the context of the European nitrogen challenge that we concentrate. For the practical purposes of this chapter, society is defined as the group of individuals that live within ‘Europe’ (exactly which definition of Europe is not important at this point, more the notion of a geographic region with broadly similar problems, behaviours and choices). Many of the examples used hereafter to support our statements are drawn from Northern and Western Europe, owing to the geographical bias of the authors’ expertise and the greater preponderance of published studies for these areas than for Southern and Eastern Europe. However, we believe that the fundamental points we address in this chapter on the efficacy of communication and engagement strategies apply across the whole of Europe and, in most cases, right across the developed world. It is important to recognize that a ‘society’, certainly on a national or European scale cannot be expected to ‘make choices’. Whilst politicians may make choices on behalf of society, other forms of decision-making are made at the individual rather than societal level. Individuals may also choose to ally themselves with smaller groups which may, and frequently do, establish and promote differing stances on an issue. As alluded to above,

Dave S. Reay

such individual and collective decisions are based on a complex interplay between a wide range of factors such as knowledge and understanding, self-interest, altruism, social- or peer-pressure, etc. In the case of the environmental significance of nitrogen the lack of public awareness, certainly in comparison with CO2, is plain from the lack of media coverage which is both a reflection of and a contributor towards poor societal understanding of the issue (see Sampei and Aoyagi-Usui, 2009). Consequently, the portrayal of nitrogen as both an integral part of Earth systems and as an environmental issue, with its attendant benefits and more negative impacts at a local level is crucial to public understanding and societal (individual) choice (Bickerstaff and Walker, 2001). Such a portrayal must be scientifically accurate (as far as possible) but recognize that choices will be based on value judgements. Hence this chapter discusses a range of issues pertinent to nitrogen and human society, but also acknowledges that reactive nitrogen is a double-edged sword€– providing great benefits to so many people but posing a great threat to others. This duality is highlighted by the existence of this publication, the recent establishment of a United Nations Economic Commission for Europe (UNECE) Task Force on Reactive Nitrogen under the Convention on Long-Range Transboundary Air Pollution (UNECE, 2010), plus a range of policies and initiatives (at various scales, from regional to European) aimed at integrated (or sectoral) mitigation and management of the ‘Nitrogen Cascade’. In order to maximize the benefits of human society’s interaction with the global nitrogen cycle, and minimize the penalties, governments can employ several kinds of strategies, including improved communication and education, the incentivization of good practice, and the use of legal requirements and penalties to deter bad practice. Whilst communication and education are the focus of this chapter, it should be noted that public understanding of the issue is also a valuable prerequisite for the more prescriptive approaches open to government. Indeed, it is plainly hazardous to assume that a single top-down strategy will be successful without some degree of engagement with and awareness in the stakeholder community at which the strategy is aimed.

26.2.1╇ Behavioural change The cryptic, indirect and interlinked nature of many of the impacts of nitrogen on society inevitably means that both awareness of its importance and the responses to the causes can be low or non-existent. For instance, those affected by harmful algal blooms, either through lost revenue from fishkills or even by shellfish poisoning, are unlikely to connect such impacts with inland nitrate leaching or ammonium pollution from a distant sewage processing plant. Indeed, such connections can be almost impossible to prove even where they are specifically looked for, given the complex array of determinants that can combine with enhanced nitrogen availability to induce such blooms. For stakeholders to respond in the most appropriate way requires awareness of the complexity of the nitrogen issue, the transboundary nature of most nitrogen fluxes and, most importantly, its direct relevance to them. To date there is a dearth of research into how the nitrogenrelated behaviour of human society and its myriad stakeholder

groups is determined, and how such behaviour might be most effectively changed as part of addressing the overall nitrogen challenge. The inducement of pro-environmental behaviours is a fast developing area of study for behavioural scientists, but most relevant studies to date have focused on topics such as climate change adaptation or carbon management (Urwin and Jordan, 2008), rather than reactive nitrogen. While it is true that nitrogen plays an important role in global climate change, as well as in other high-profile environmental challenges€– such as acidification and eutrophication€ – the links between societal or individual nitrogen-relevant choices and their resulting impacts remain much more opaque than those for carbon. Given the many parallels and cross-cutting themes between the two topics, it therefore makes sense to examine the current evidence and views on eliciting carbon-related behavioural change and see how these might be applied in the context of nitrogen.

26.2.2╇ Segmented strategy As mentioned previously, a ‘one size fits all’ approach to achieving behaviour change is very likely to fail. There are myriad cultures, backgrounds, priorities and awareness levels across the range of stakeholder groups of importance to nitrogen management in Europe, and such a strategy will inevitably prove ineffective in eliciting widespread change across all of them. Instead, the concept of a segmented strategy has been developed for climate change mitigation whereby the stakeholder groups and their characteristics are defined and the methods then employed to achieve the desired change are tailored to reflect these different characteristics (DEFRA, 2008). We recommend a similar approach be applied to the European nitrogen issue, with due recognition of differing awareness levels and priorities. Figure 26.1 shows a segmented approach to eliciting behaviour change that could be applied to the European nitrogen issue. The key tools (shown in green) for engendering change are categorized as follows. • Encourage:€a ‘carrot and stick’ approach whereby undesirable practice is discouraged by taxation, fines or negative public exposure (e.g. performance league tables) and desirable practice is encouraged through grants, reward schemes and positive public exposure. • Enable:€the removal of technological or policy barriers to change, the provision of alternatives, and support of the education necessary to inform the desired behaviour change. • Engage:€provision of forums, support networks, awareness raising (e.g. media campaigns). • Exemplify:€use of consistent policies and exemplars within the target group (e.g. early adopters of a nitrogen reduction strategy) who can inform and support their peers. The seven categories of ‘stakeholder’ positions shown in Figure 26.1 and the associated approaches, though inevitably still broad in scope, can be usefully applied in the context of the European nitrogen issue, for example as follows: Categories 1, 3 and 4 ‘Positive Greens, Concerned Consumers, and Sideline Supporters’ are characterized by those who are

587

Societal choice Figure 26.1 Segmented strategy for eliciting behaviour change. Red boxes denote axes, green boxes the tools for engendering change, and yellow elipses the stakeholder categories (adapted from DEFRA, 2008).

High Enable

Potential to do more

Engage Encourage

1. Positive greens

3. Concerned consumers

Exemplify Enable

4. Sideline supporters

Low

2. Waste watchers

5. Cautious participants

Willing to act High

6. Stalled starters

7. Honestly disengaged

Encourage Enable Low

already willing to change behaviour and require enabling, for example through top-down policy, to take action. These groups have a very high capacity for change given the correct support, in particular the strategies of ‘Engage’ and ‘Enable’. In a nitrogen context such a grouping could include consumers who, when provided with information on the nitrogen impacts of their purchases (e.g. red meat) may reduce consumption of these products for less nitrogen-intensive alternatives. Enabling such change includes the provision of alternatives and, in this example, information on recommended protein intake and diet. Categories 2 and 5 ‘Cautious Participants, Waste Watchers’ are often very substantial groupings and can be exemplified by, for instance, the many farmers in Europe who have a reasonable level of nitrogen awareness and willingness to act, but are cautious of being an early adopter of new recommendations or are jaded by tides of other pressures, advice and policies. Here, a combination of ‘Encourage’, ‘Enable’ and ‘Exemplify’ has the potential to achieve very extensive behaviour changes, for example as below. • Encourage:€e.g. through financial rewards for improved nitrogen management. • Enable:€e.g. provision of tailored nitrogen management advice/model outputs. • Exemplify:€e.g. providing visits or for where farmers are able to see recommended nitrogen management being successfully used. Finally, categories 6 and 7 (Honestly Disengaged, Stalled Starters) represent groupings where ‘Engage’ and ‘Exemplify’

588

strategies are unlikely to yield much change. Increased awareness, by itself, is likely to have only a limited effect on behaviour for these groups, with negative impacts perceived as being spatially or temporally distant from the ‘polluter’ and any significant change being seen as a threat to competitiveness, profit and financial sustainability. An example of this in the context of nitrogen may be the simple linking in the media of adipic acid production (used in the production of nylon) with climate forcing by N2O, without any attempt by government or industry bodies to support change in industry behaviour through ‘Encourage’ and ‘Enable’ strategies. In such cases, it is likely that directed policy that either rewards the producer for reduced N2O losses or penalizes them for failure to act (‘Encourage’) will have much more effect than their simply being aware of the negative consequences of emission for people thousands of miles distant or indeed generations to come. Similarly, ensuring effective technology is available for reducing N2O emissions is vital if such ‘carrot and stick’ approaches are not to result in industry-wide penalties and leakage of production and emissions to non-European producers benefiting from the competitive advantage that would arise. A segmented approach to communication and to eliciting behaviour change can be further adapted for use within many discrete groups of nitrogen stakeholders. For policy makers, such an approach could allow differentiation of those with priorities in one geographical area from another, and so potentially improve the relevance of and engagement with the information provided. At the other extreme, a segmented approach applied

Dave S. Reay

to European householders could identify and make use of discrete groupings based on factors such as income, culture, location and climate to inform and tailor the strategy employed. In summary, the segmented approach to behaviour change championed for climate change mitigation would seem equally appropriate for engendering behaviour change with the aim of addressing the European nitrogen challenge. Clearly, the individuals and groups that this approach addresses may be very different in the context of nitrogen, but the core message€– that a tailored approach will be more successful than a generic one€– remains central to this chapter and its recommendations.

UK in terms of their cost and the amount of greenhouse gas (GHG) emission avoided (Figure 26.2). This exercise is helpful given that it can identify measures that both save money and reduce GHG emissions. As such, the MACC approach may also be applied in assessing the most cost-effective measures to enhance nitrogen use efficiency and reduce negative impacts in the European LULUCF sector and more widely. In reality though, such an approach would likely serve only as a qualitative guide to policy makers, the complexities and interactions inherent in nitrogen management making any robust assessment of opportunity and abatement costs very problematic. One of the major difficulties in an analysis of this sort is the estimation of current fluxes (whether GHGs or nitrogen) and the further problem of anticipating how these fluxes will change through the adoption of mitigation measures. Another difficulty, as mentioned above, is linked to the problem of anticipating how mitigation measures might interact. It is likely in many circumstances, that a number of mitigation measures would be used in combination in order to achieve a certain level of emission reduction or nitrogen use efficiency. However measures that are used in combination do not necessarily lead to effects that are simply additive. In many circumstances, the adoption of the first mitigation option (e.g. reduced nitrogen fertilizer applications) may lead to a reduced potential for additional mitigation measures (e.g. use of nitrification inhibitors). In devising such curves for GHG abatement or nitrogen management, assumptions must be made about future changes in key factors such as land use and farming practice, the wider policy and market context, and about how climate and technology may change. Such assumptions and uncertainties further limit the usefulness of these curves for decision makers, particularly when applied across national boundaries. In summary, though MACCs can provide a useful visualization of relative costs and benefits of a range of nitrogen management strategies available, decision makers will inevitably need to draw on the wider nitrogen management knowledge base to reduce the risk of unexpected and unwanted outcomes.

26.3╇ Realizing a low nitrogen economy and society 26.3.1╇ Marginal costs as a driver of change Economic sustainability is a critically important driver in human decision-making and this is especially true in relation to farmers€– a group of individuals that exert influence on the nitrogen cycle through their daily working practices. While farmers may show enthusiasm for environmental initiatives that contribute to a nitrogen-efficient economy, this can be constrained by their ability to pay, both financially and in terms of effort, for any required changes in farming practice. Understanding of these relative costs and benefits is therefore crucial to policymakers tasked with engendering behaviour change in the farming community and increasing nitrogen use efficiency across Europe. To help inform policymakers on the cost and benÂ�efit implications of achieving significant levels of change, Marginal Abatement Cost Curve (MACC) analyses have been applied to many sectors (most commonly in terms of reducing greenhouse gas emissions). The MACC identifies both the amount of ‘benefit’ (e.g. greenhouse gas savings) and its ‘cost’ through the introduction of different mitigation measures. For example, Moran et al. (2008) ranked mitigation options in the Land Use and Land Use Change and Forestry (LULUCF) sector in the Cost effectiveness £2006/tCO2e

Figure 26.2 A projected mitigation abatement cost curve for UK agriculture for 2022 assuming a discount rate of 3.5% (from Moran et al., 2008). Negative values represent values that save costs.

Livestock measures Crops/soils measures Forestry Anaerobic digestion

300 250 200 150 100 50 0 –50 0

1

2

3

4

5 Afforestation

–100 Improved timing (mineral N)

–3,650

Beef imp. genetics

Beef ionophones

6

7

8

9

Improved drainage OFAD and CAD

Avoid N Improved timing Red. till excess Dairy: (organic N) Dairy Dairy. imp. productivity ionophores Maize imp. fertility silage Using composts Full manure N Separating slurry Plant breeding and mineral N

OFAD

10 Nitrification inhibitors

DairybST

Species introduction

Abatement potential (MtCO2e/year)

589

Societal choice

26.3.2╇ Behaviour and culture Even where the MACC approach is able to identify least- or even negative-cost measures that can serve to improve nitrogen use efficiency and reduce nitrogen pollution, the simple demonstration of such abatement costs rarely, by itself, results in overtly fiscal-led behaviour change. Similar analyses for CO2 mitigation in the built environment sector show huge financial savings for strategies such as improved insulation, yet adoption of these strategies remains patchy. Individual behaviour often appears ‘irrational’ with respect to what the ‘market’ would indicate (Oikonomou et al., 2009), but then the true opportunity costs of a given strategy may be poorly represented in MACC analyses; for example, the disruption caused to a householder during installation of insulation, or the time cost to a farmer of collating and analysing farm-scale nitrogen budgets. As discussed later, the social behaviour and culture that underpin the actions of humans must be acknowledged as a potential barrier to addressing nitrogen through behaviour change, even where the economics would appear to make it an entirely rational transition. With nitrogen being so integral to so many facets of human life, its use and impacts have also become engrained in human behaviours and cultures. From the farmer who applies a few tens more kilograms of nitrogen per hectare to his fields each year ‘to be on the safe side’ to the desire for meat with every main meal so prevalent in the West, any balanced nitrogen policy must take some account of the social and cultural landscape in which it will operate€ – a segmented strategy once again. Nitrogen’s role in human society is poorly understood by the public in comparison to, for instance, carbon. Indeed, there is the risk that the limited awareness that does exist is based upon a poor evidence base or a long-standing misinterpretation. Behaviours based upon poor or out of date information€– for instance, the direct link drawn in the 1950s and 1960s between nitrate concentrations in drinking water and methemoglobinÂ� emia in babies (Smil, 2004)€– may hinder moves to increase the effectiveness and efficiency of nitrogen use by human society.

Livestock production as a barrier to change There are numerous examples of how behaviour and culture represent barriers to achieving a low nitrogen economy and society, and these range from contemporary society’s high dependence on fossil fuel, through to the strong political elements common to many trade agreements and subsidies. As one such example of the issues associated with human behaviour and culture, this section examines some of the potential barriers to achieving a reduction in livestock productionrelated nitrogen emissions (or at least a reduced growth rate) through a reduction in consumption of animal products. A more in depth analysis of human dietary choice and its impact on nitrogen fluxes is also provided in Section 26.3.5. Producers and suppliers The livestock sector is an important economic sector in Europe. The value of livestock production of the EU-27 in 2007 was €136 billion€– almost 40% of the total economic value of EU

590

agricultural production (EC, 2008). The economic interests of many actors and stakeholders in the livestock sector may lead to resistance to nitrogen-focused change, where this is seen as potentially damaging to livelihoods. Many livestock farmers are specialists who cannot switch easily to forms of nonlivestock production. Many of these farmers have also invested heavily in grazing land, animal housing, human resources, machinery and other infrastructure geared specifically to livestock or dairy production. Curtailing these activities across Europe could therefore mean a large loss of capital, turnover and jobs. In addition, there are many additional stakeholders in the livestock and dairy production chain outside of the production units themselves (Herrero et al., 2009). The European feed sector, for example, has an annual turnover of €36 billion (FEFAC, 2008), while the meat and dairy processing industry, transport companies and retailers also have significant stakes in the livestock industry. Finally, the livestock sector is strongly interlinked with other sectors across Europe and globally, many by-products of the food industry being used as feed, and the livestock industry itself providing many valuable by-products, such as leather and medicines. Consumers There exist major barriers to the successful implementation of any strategy aimed at reducing the consumption of meat and other animal products in Europe. Meat and other animal products are a very important (for some the most important) component of meals and play an important role in dietary traditions, social interaction and social norms (Kenyon and Barker, 1998). Portion size is also a consideration, as there is not always a choice of portion size (in relation to the animal protein content) and portion sizes have increased markedly in recent decades (Nielsen and Popkin, 2003), therefore the consumer must choose between being either meat-free, or to consume a portion of meat that may well be larger than the amount needed or desired (see, e.g., NinE, 2009).

Cultural values and co-benefits The extensive production of some livestock types (notably cattle and sheep) may lead to subjectively positive impacts on landscape aesthetics and on biodiversity (Luick, 1998). Reduced livestock demand and cessation of grazing in such areas€– with the consequent changes in appearance, vegetation and biodiversity€ – may therefore result in a negative response and resistance from some land users. Indeed, many so-called ‘High Nature Value (HNV) farmlands’ are grazed pastures (Paracchini et al., 2008) and as such the cultural, aesthetic and conservation values placed upon many grassland areas in Europe must also be considered as a potential barrier to strategies aimed at reducing livestock consumption. The concept of ‘ecosystem services’ is worth mentioning at this point, as it encompasses the cultural, aesthetic and conservation values that may be applied to different land-uses. Essentially, an economic value can be estimated for each ecosystem service (e.g. biodiversity value) with a view to marketbased initiatives then providing greater protection for types

Dave S. Reay

and uses of land that might otherwise be at risk of change to more overtly commercial uses. Much of the impetus for this approach has come from the carbon markets and the desire to improve their coverage of forestry by acknowledging the biodiversity value of forest ecosystems, as well as that of the carbon they store. The concept of placing a value (or cost) on nitrogen that will then enable market forces to drive management improvements has recently been proposed in the context of water quality trading (Gross et al., 2007) and is discussed in detail by Brink et al. (2011, Chapter 22, this volume). There are some important co-benefits to reduced livestock consumption in addition to reduced nitrogen losses and their impacts. These include potential human health improvements, through avoidance of excessive protein intake and reduced obesity rates (de Boer et al., 2006), and in particular the potential reduction of greenhouse gas (GHG) emissions and the associated anthropogenic enhancement of global climate change. For the latter, livestock production is estimated to result in some 18% of global anthropogenic GHG emissions, with approximately one third of these emissions arising as CO2 emissions from land use change, another third from livestockrelated N2O emissions, and most of the rest from ruminant and manure-derived methane emissions (Herrero et al., 2009). If significant per capita reductions in meat consumption are needed, it is unlikely that such changes could be effected through consumer choice alone. Policy makers would need to play a role, but the big questions are:€what reduction is necessary or feasible, and how can this reduction be achieved? McMichael et al. (2007) argue that from a climate change perspective, average per capita meat consumption will have to be reduced to 90 g per day globally in order to stabilize greenhouse gas emissions from livestock production. Deckers (2010) suggests that this would be a modest reduction compared with the ambitions to reduce greenhouse gas emissions from other sectors and considers the necessity of more extreme policy options from increasing meat prices to a qualified or complete ban on the consumption of meat products. As discussed previously, national and international climate change mitigation policy may therefore play an increasingly powerful role in driving change in livestock production practice and nitrogen management in future years and decades.

26.3.3╇ Budgets There are many potential strategies that may be employed in order to help realize a low nitrogen society. Here we examine the concept of ‘Budgets’, not with a view as to how they would bring about a low nitrogen society in themselves, but rather as to how such tools might raise awareness and inform decision makers as to how best to address the European nitrogen challenge. There are several ways to consider budgets in the context of nitrogen management, including the use of a sectoral approach, the examination of direct and indirect fluxes (e.g. end user versus supply chain), and separation into scope 1, 2 or 3 categories (see WRI/WBCSD, 2010). Here we examine budgets in the context of broad stakeholder groupings, using the three categories of ‘personal/household’, ‘business’, and ‘regional’ to highlight key considerations and issues.

Personal and household budgets In principle, these types of budgets are well known and utilized€ – there are numerous calculators now available for calculating a personal or household carbon budget for example. These individual and household budget calculators invariably show the consequences of personal choices with respect to energy consumption (Directgov UK, 2010), with some of the more sophisticated versions including analysis of and recommendations on water use and food consumption as part of the user’s so called ‘ecological footprint’ (EPA Victoria, 2010). These calculators and personal budgets are not directly intended for policymaking, but are focussed more on raising public awareness with a view to engendering behavioural change. In addition to carbon footprinting, personal ‘Nitrogen Footprint’ calculators have also been developed (e.g. ‘N-Print’ model under development by the International Nitrogen Initiative and CBF, 2010). These calculators demonstrate to a consumer the potential nitrogen use, and associated losses, that would occur as a consequence of their behaviour. Like the more established ecological footprint calculators, it is primarily related to food consumption, waste and energy use. In the case of nitrogen, it is the food consumption and waste of an individual that will often have both the major impact on their overall nitrogen use and be the area of behaviour where their choices and actions can make the greatest impact. Although personal nitrogen calculators are useful as a means of communication and awareness-raising, they do bring ‘yet more calculators’ into this already crowded public domain. There is a risk that uptake and resulting actions will be limited by public apathy towards nitrogen-related problems and overexposure to such footprint calculators. Therefore, for any calculator to be successful (i.e. to raise awareness of the nitrogen issue, leading to increased nitrogen use efficiency) it has to be well developed, with an interesting and clear story to communicate to the public. Consistency between different versions of a nitrogen calculator is also very important, to maintain credibility with users.

Business budgets Although nitrogen budgets for businesses may be of broadly the same type as those described above for personal or household use, the overall aim is often rather different. In carbon budgeting for example, many businesses have looked to provide information on the carbon footprint of their operations for reasons of corporate social responsibility and at the behest of the investment community in general (e.g. The Carbon Disclosure Project, 2010). Experience from carbon budget reporting has shown that the very process of calculating a budget can yield benefits in terms of identifying efficiency savings and improving public relations. Such nitrogen budgeting for businesses remains in its infancy, but the models that exist for calculating the carbon budgets of business and their supply chains provide an excellent framework for adaptation to nitrogen budget calculation. Recent evidence from the Carbon Disclosure Project (Nigel Topping, CDP, personal communication), for example, has indicated that, as large corporations increasingly quantify their carbon budgets (i.e. scope 1 and 2€– direct and indirect emissions from energy use respectively) and those of their

591

Societal choice

National and government budgets On a national or governmental scale, calculation of nitrogen budgets is again possible. Such budgets give a general view of the inputs, outputs, and losses of nitrogen nationally (see de Vries et al., 2011, and Leip et al., 2011, Chapters 15 and 16, this volume). As such, they can be used as a broad basis for policy making, allowing key areas of loss and inefficiency to be identified at the macro scale. The clear downside of such budgets is their lack of detail, with any regional or local scale anomalies likely to be missed with such an approach. Here, again, established carbon and GHG budget methods provide an example of how flux budgets at smaller scales can be nested within national budgets to allow policy makers to make �better-informed decisions and avoid antagonisms and undesirable outcomes from any strategy that is developed (compare, e.g., the Paris nitrogen budget, Svirejeva-Hopkins et al., 2011, Chapter 12, this volume). Finally, some governments (e.g. the Scottish Government) are now required to calculate the carbon budget and impact of their own spending decisions; this process allows them to scrutinize and compare the climate change implications of the options open to them. Yet again, a similar approach for nitrogen budgeting and auditing can be envisaged, whereby proposed investment in new transport networks or in farm subsidy provision can be assessed in terms of its impact on nitrogen fluxes and losses.

26.3.4╇ Choice in food production Previous chapters of this assessment have demonstrated the many choices that society has in determining nitrogen use efficiency, equity of access, wastage and negative impacts of reactive nitrogen. Of these, food production and consumption choices represent a key area where these choices may have major implications for nitrogen fluxes. Here we assess in more depth some of these choices and the extent to which they may affect nitrogen use efficiency and fluxes.

Market choice and nitrogen imports Despite their potential advantages in contributing to more efficient use of nitrogen in agricultural systems, legume crop and forage production has declined in the EU. Wider European (as defined by the FAO, including Ukraine, etc.) pulse crop production declined from 11.3 Mha in 1961 to about 3.4 Mha in 2005. Forage legume area also showed significant decline between 1980 and 2001 (Rochon et al., 2004). The area data for peas and beans are typical of a wide range of legume crops (Figure 26.3). Grain legumes occupy less than 2% of the agricultural land in Europe compared with 8% in Australia and western Canada. Furthermore the distribution of legume crops across Europe is uneven with 86% of the EU grain legume production in 2005

592

1200 Beans Peas

1000 800 1000s ha

suppliers (i.e. scope 3€ – indirect emissions not covered by scope 2), they are driving supply chain emissions cuts in addition to those under their direct control. Such ‘mainstreaming’ of carbon budget calculation and mitigation by the business community, if applied to nitrogen, has the potential to radically improve the awareness of and action on avoidable nitrogen losses.

600 400 200 0 1960

1970

1980

1990

2000

Figure 26.3 Changes in areas planted with bean and pea crops in the EU between 1961 and 2007 (FAO, 2009).

occurring in only five countries€– Spain, France, UK, Italy and Germany (see AEP, 2007). However, European agricultural systems rely heavily on legumes€ – particularly soybeans grown in South America. The global trade in soy equates to a movement of more than 10 million tonnes of reactive nitrogen (UNEP and WHRC, 2007). This, combined with the other plant nutrients, particularly phosphorus, represents a challenge to geochemical cycles on a global scale. The import of soybeans and soymeal into Europe increased from the equivalent of 3.4 million tonnes of soybean in 1961 to 55 million tonnes in 2006. EU consumption now accounts for about 25% of the world crop and an even larger proportion of the soybean production in Argentina and Brazil. Soy is directly or indirectly linked to deforestation, and in the Amazon in particular (Nepstad et al., 2006; Simon and Garagorry, 2005). Soy imports are central to a global flow of resources from South America, supporting intensive livestock production in Europe. The resulting separation of livestock from the natural resource base that they would naturally draw on is associated with high losses from the nitrogen cycle (Galloway et al., 2007). Europe’s dependence on imported soy is a major economic and environmental challenge for European livestock production which will intensify (Steinfeld et al., 2006) as the global demand for livestock products increases, as predicted by many sources. Per unit of protein, meat and dairy require more land, mainly because of losses caused by transformation in the production chain. Not all nutrients used to grow crops end up in the feed, not all feed nutrients will be converted into animal tissue and not all of an animal is marketable or edible. Mother animals are also needed to maintain the herd€ – this requires considerable quantities of feed especially for cattle and to a lesser extent for pigs. Measured over the entire chain, in beef (steers) just 4% of the protein from animal feed is converted into protein. In pigs this conversion is 8% and in chickens 23% (Sebek and Temme, 2009). The global movement of nutrients in animal feed (Naylor et al., 2005) may warrant international and global approaches (UNEP and WHRC, 2007) that actually limit the market-based choice of producers and drive a move towards reduced global exchange of nitrogen-rich products.

Dave S. Reay

same production, then the reductions will be cancelled out. More generally, the contribution of meat in food production is a key factor:€about 80%–95% of the nitrogen intake with feed is excreted as dung and urine, so crop production for human nutrition rather than animal is one of the most efficient measures for mitigating greenhouse gas emissions from agriculture.

26.3.5╇ Choice in food consumption

Figure 26.4 Current extent of self-sufficiency in nitrogen-rich products in the EU 25. Figure reproduced with permission of the Netherlands Environmental Assessment Agency (PBL).

Overall, European consumption of livestock feed, together with imported fish, meat and dairy products, puts pressure not only on the local environment, but also on that outside of Europe (Figure 26.4).

Cultivation choices The use of legumes in pasture presents special challenges and opportunities. Despite the low overall response of grass-clover pasture to synthetic nitrogen application (Bax and Schils, 1993), the use of high applications of synthetic fertilizer in pastures is common and is reducing the role of clover in forage production. In addition to nitrogen fixation and drought resistance, clover and other forage legumes offer opportunities to improve forage quality and end-product quality. In summary, the choice of many arable farmers in Europe to reduce their reliance on legume production, and for livestock producers to opt for an increasing proportion of imported feed legumes (especially soybeans from South America) has resulted in substantial inefficiencies in European and global nitrogen use. An international or global approach is beginning to emerge (see Legume Futures, 2010) to address such nitrogen and land-use change ‘leakage’ from European agriculture and provide alternative choices for producers. Several European studies (e.g. Flessa et al., 2002; Olesen et al., 2006; Petersen et al., 2006) have also compared the greenhouse gas emissions from conventional farming systems (i.e. those using synthetic fertilizers and pesticides) with those from organic systems that rely on the use of animal manures, crop residues and biological nitrogen fixation as nutrient sources. The study by Flessa et al. (2002) of two adjacent farms in south Germany, with arable cropping and beef steer production, showed the important contribution of nitrous oxide (N2O) emissions from soils to total greenhouse gas emissions (even though indirect sources of N2O emission were not included). As Figure 26.5 shows, the fraction of emissions coming from N2O was very similar in the conventional and organic farms. The conversion from conventional to organic farming resulted in reduced emissions per hectare, but yield-related emissions were not reduced. The authors conclude that conversion to organic farming may contribute to the reduction of greenhouse gas emissions from agriculture if policies seek to reduce the intensity of agricultural production. However, if lower intensity is compensated for by the use of larger land areas for the

As discussed previously, human dietary choices are a key determinant of anthropogenic nitrogen fluxes and impacts at scales from the individual to worldwide. Since 1960, global consumption of meat, fish and dairy products has risen sharply, with both positive and negative consequences for public health. In 2005, the average global consumption was 39 kg of meat per year per person. In Europe (EU-15), protein consumption was almost three times higher than this€– at 91 kg. This compares to 121 kg for North America, 54 kg in China and only 14 kg of meat per year per person in Africa (FAO, 2009). For fish, the average was 16 kg per person (included in live-weight) and for EU-27 it was 23 kg. The consumption of dairy products also varies widely, consumption rates for Europe (and North America) being 5–6 times higher than that in Asia and Africa. Protein is an important part of the human diet and meat, fish and dairy products are important sources of it. It is interesting, therefore, to look at consumption patterns of total protein from these different sources by geographical area. Such analysis shows that total protein consumption in Europe is more than 4 times that in Africa. This, along with high levels of consumption in North America and OECD-Asia means that 10% of the world population are consuming 25% of the animal protein (Figure 26.6).

Dietary evolution and human health Global consumption of animal proteins has more than doubled in comparison to 1970 (Figure 26.7). EU consumption of animal protein is 14 kg per person per year, which is higher than the world average of 11 kg. Globally, 15% of the animal protein originates from fish, in the EU this figure is 11%. Clearly, there are great differences in the consumption of proteins from meat, milk and fish between countries. Consumption in Europe and the United States together double the world average (FAO,€2006, 2009). It is expected that meat consumption in the most developed countries will grow by approximately 10% between 2005 and 2030. However, the biggest growth is expected in the rest of the world (Figure 26.6). Global demand for animal proteins is projected to increase some 60% (Figure 26.7), whereas in Europe the projected increase in demand is much smaller. If all countries are to obtain a diet equal in animal protein to that of the Western world then global meat production will need to triple. In the most developed countries in the period of 1960–1990 meat, milk and fish consumption increased rapidly€ – in the Netherlands, the consumption of meat and fish has doubled since 1960. As on the global scale, so the consumption of animal products in Europe differs between countries. Consumption is related to both economic productivity and nation-specific social-cultural aspects (De Boer et al., 2006).

593

Societal choice Figure 26.5 Total CO2, CH4 and N2O emissions (from Flessa et al., 2002) from the agricultural production of two farming systems (A:€conventional and B:€organic) in southern Germany and the contributuion of different sources to these emissions. The farming systems are described in more detail in Table 1 of Flessa et al., 2002).

Figure 26.6 Per capita consumption of animal protein in 2003 by geographical region and projection to 2030 (FAO, 2006a; FAO, 2009a,b). Figure reproduced with permission of the Netherlands Environmental Assessment Agency€(PBL).

Figure 26.7 Past, recent and projected global (billion kg y−1) and per capita (kg y−1) consumption of animal protein by geographic region. Figure reproduced with permission of the Netherlands Environmental Assessment Agency (PBL).

594

Dave S. Reay

Figure 26.8 Per capita protein consumption by source in the Netherlands between 1960 and 2007 (PBL-calculations based on FAO). Figure reproduced with permission of the Netherlands Environmental Assessment Agency (PBL).

Although protein consumption is necessary for humans, in Europe (and the USA) more protein is consumed than is required in EU-27 an average of around 31â•›kg protein (23â•›kg of animal origin) was consumed per person per year in 2007. European consumption of protein is 70% higher than the minimum amount which is recommended (see Fig. 26.8) and although this is not always harmful, European consumption is also 15% higher than the upper limit set by WHO (T Rood personal communication). Too much saturated fat€ – often associated with meat and dairy-rich diets€– also increases the risk of cardiovascular diseases (Lloyd-Williams, et al., 2008). With three thousand cases of heart disease and over 700 deaths every year in the Netherlands (from a population of 16 million) being a consequence of excessive intake of saturated fats from animal products (Büchner et al., 2007). Eating too much red meat (>160 grams per day) may also increase the risk of colorectal cancer (Norat et al., 2005). The promotion of healthier diets to reduce the burden to society of non-communicable diseases therefore holds huge potential for simultaneous indirect reductions in nitrogen wastage and increased efficiency of use across Europe and globally. Indeed, initiatives aimed at engendering behaviour change away from animal product-intensive diets€– by emphasizing the health benefits that may accrue to the individual consumer€ – may prove far more successful in addressing food-related nitrogen challenges than approaches based solely on communicating the nitrogen-related externalities of meat-rich diets to consumers.

Food wastage In addition to what is eaten, European consumers discard some 20%–30% of purchased food, representing an important component of overall nitrogen wastage centered on consumer behaviour. In the UK alone, an estimated £10.2 billion of food is purchased, but not eaten, each year. Discarded meat and fish meals amount to approximately 161 000 tonnes of waste each year, at a total cost to UK households of £602 million (Ventour, 2008). The main reasons given by consumers for this wastage include meat and fish being out of date (35% of UK household wastage) or left over after meals (25% of UK household wastage). Efforts to reduce such high levels of consumer-based wastage have centered on

highlighting the financial cost to individual households (e.g. £420 per UK household per year) and provision of consumer advice on food purchasing, storage and preparation (WRAP, 2008). As with dietary choice, addressing the negative externalities of household food wastage€– such as increased nitrogen pollution and environmental degradation€– may best be achieved by aligning new policy and behaviour change initiatives with the existing health and financial cost approaches.

Nitrogen and animal welfare Animal welfare is an important issue which is addressed by policy and periodically receives media attention. It is an emotive subject which can influence consumer choices (Duffy and Fearne, 2009), depending on other issues such as cost, knowledge and understanding. However, modifying the living conditions of animals may cause changes in the ‘Nitrogen Cascade’ (e.g. it can be more difficult to manage manure if animals are allowed to roam outside rather than housed inside), leading to potential trade-offs, antagonisms and synergies€ – these need to be assessed and communicated to the industry and public, so as to move forward with animal welfare without moving backwards in farm nitrogen management. Similarly, consumer choice for organic rather than conventional food production has the potential to alter nitrogen fluxes and use efficiencies, but again with numerous trade-offs€– such as changes in nitrous oxide flux, as discussed previously€– being inherent in such alterations in production methods. Reduction in the consumption of animal products, as discussed above, is the most far-reaching but potentially the most effective option. The net impact on the environment depends not only on the extent to which the consumption declines, but also on the products which are then used to replace animal protein. There are also substantial leakage risks where reductions in consumption are confined to only one region€ – a drop in market price potentially enhancing consumption in other areas and negating any net reduction on a global scale. Overall, decreasing animal product consumption may create very substantial positive benefits for both human health and the environment (De Boer et al., 2006), however a large scale shift in European consumption would require careful

595

Societal choice

management of the industries that depend on the livestock sector and consideration of how policies and markets outside of Europe may negate some or all of the environmental benefits.

26.3.6╇ Prioritizing response and importance A clear task for future nitrogen policy in the EU, and globally, is to increase the awareness of the role of nitrogen in myriad proÂ� cesses and impacts, and whether they have a positive or negative net outcome for society. Central to achieving such increased awareness is the use of audience-tailored messages on the role of nitrogen. These can range from user-friendly farm nitrogen management models that provide individual farmers with recommendations on how to improve nitrogen use efficiency and the resulting fiscal benefits, to online ‘Nitrogen Policy Crossreferencing’ databases that allow potential issues of pollutionswapping and synergies to be flagged up at an early stage of policy development or revision. Importantly, the nitrogen policy agenda should avoid ‘re-inventing the wheel’. Where policies already exist that address nitrogen-related issues, such as on diet, transport or energy, then these should be assessed to identify what ‘win-wins’ or antagonisms exist and to identify any key nitrogen-related gaps that new or revised policy could meet. With such a plethora of information and advice directed at consumers on these issues, especially on carbon emissions, it is likely that nitrogen policy directed at the key supply-side sector (i.e. primary food production) will provide the greatest net impact. On the consumer behaviour side, aligning nitrogenÂ�related policy with that aimed at improving human dietary health would appear to hold the greatest potential for achieving significant and sustained success.

26.4╇ Communication As discussed earlier in this chapter, the communication of nitrogen as an environmental issue is problematic because of its perceived complexity. In addition, its profile in formal education and the media remains very low, with current public understanding likely to be based on school curricula which have traditionally focussed on the perpetual recycling of compounds through the nitrogen cycle. This is further compounded by frequent mass-media and public misunderstanding of ‘the scientific method’. The fact that scientists generally communicate their findings with caution and explicitly state their uncertainties may give their peers confidence in their work and their integrity, but this is often not seen as helpful by those looking for ‘answers’. Furthermore, dealing with nitrogen as an environmental issue is likely to be costly and inconvenient and hence a message that many will be reluctant to accept€ – again with parallels to climate change communication. The communication of consistent messages is important yet problematic. As curricula from school through to higher education are subject to only periodic validation procedures, they are necessarily slow to respond to changing scientific understanding, whilst media and political responses can be immediate and unpredictable for the reasons noted above. To be effective in increasing public understanding scientists will need to engage actively with educationists, journalists (in

596

general but also specifically those working with farmers, industrialists and environmental groups) and politicians. Informal electronic communication will also have an increasing role, especially in youth culture. Whilst messages need to be consistent, they also need to be audience-specific in level and tone. Below we provide some examples of past and current nitrogen communication activities to different stakeholder groups.

26.4.1╇ Current examples of communication (case studies) This section provides an opportunity to look at how the nitrogen issue has been communicated so far, where instances of its influence have reached the public domain, and how related environmental messages (especially those which have asked the public to make behavioural changes) have been successfully (or unsuccessfully) communicated. It also demonstrates the delicate interplay between the science providers (researchers), science users (policymakers) and the main communicators in this arena (the media). This provides a useful resource on which to develop our own communication resources and strategies (not ‘re-inventing the wheel’) and to think about how best (or how not) to manage the policy, science, media interface.

Changing behaviour with farmers As discussed earlier, an analysis of marginal Abatement Cost Curves has indicated that there are many actions that could be taken by farmers that would achieve GHG reductions at negative costs (in other words save money) (Moran et al., 2008), and it is therefore surprising to some that these opportunities have not been realized. The reasons for this may be complex, but it is likely that profit maximizing is not the only driver determining farmer behaviour, and that a range of other factors in addition to market conditions (i.e. input and output prices) are likely to be important (cf. Oenema et al., 2011, Chapter 4, this volume). Market conditions can be considered as a part of a wider set of considerations that influence farm decision making, which includes:€internal factors (e.g. cognition, habit and attitude); social factors (e.g. norms and roles); the policy environment; and other farm business constraints (Pike, 2008). It is also possible that farmers lack sufficient information to make rational decisions in the face of complex and multifaceted farm management decisions. A study carried out on a group of Danish livestock farms has shown that indicators of nutrient balance (for nitrogen and phosphorous) can prove to be useful tools for farmers on which to base management decisions in order to improve nutrient use efficiency (Halberg, 1999). A combination of modelling and measurement that has been introduced in the UK to improve fertilizer nitrogen efficiency also aims to provide farmers with a greater degree of understanding in order to help develop more efficient use of nutrient resources (Nicholson et al., 2000). It has been argued that we need to develop better feedback loops between practice and theory in order to share knowledge and develop policy (Deugd et al., 1998). This essentially involves a three-way dialogue between farmers or practitioners, researchers and policy makers.

Dave S. Reay

Biofuels and nitrous oxide Biofuels are fuels from plant derivatives that gain their carbon from the atmosphere as CO2, and when they are burnt no new CO2 is released, so they are said to be ‘carbon-neutral’. In Europe, concern about global warming has been the driving force behind efforts to partly replace fossil fuel with biofuels, whereas in the USA the main stimulus for biofuel production came originally from the Bush Administration’s desire to reduce dependence on imported oil. Both factors have led to rapid expansion of the biofuels industry, but the idea that this trend is environmentally desirable has been increasingly challenged, and social impacts such as threats to food supplies have also been highlighted. These offsets are encompassed in Life Cycle Analysis (LCA) models, which also include an allowance for the N2O emission associated with fertiliser use in the production of the biofuel crop. There is a strong case for reconsidering these N2O emissions, and their impacts on the LCA of biofuels. Crutzen et al. (2008) have estimated that, globally, 3%–5% of all new reactive nitrogen input into terrestrial systems is converted to N2O. They estimated that the consequent extra N2O emission from the production of ethanol from maize or wheat, and from the production of biodiesel from rapeseed, calculated in ‘CO2equivalent’ terms, can contribute as much or more to global warming than the quasi-cooling effect it achieves by ‘saving’ emissions of fossil-fuel-derived CO2. Thus, the way the benefits and problems associated with biofuel production (especially the nitrogen relevant elements) have been communicated to policy makers and the public have provided a prime example of the need for such communication and awareness to follow robust, peer-reviewed research, rather than to precede it. The challenge for research is to provide this communication before important policy decisions are made.

e-Nitrogen Recent years have seen an increase in the availability of nitrogenÂ�relevant information and advice to the public via websites and downloadable resources. As with climate change science and mitigation, the easy availability of such resources worldwide can increase awareness, debate and knowledge exchange on a scale that would have been impossible 20 years ago. The International Nitrogen Initiative (INI, 2010) is an example of an organization that has made increasing use of the internet for communication€– it aims to increase awareness of the nitrogen challenge and to enhance the integration of stakeholder expertise and activity. As well as organizing regular workshops for stakeholders from around the world to exchange views and knowledge, it has been active in developing nitrogen communication tools and resources, including its ‘Nitrogen Visualization Tool’ (see Figure 26.9). Such stakeholder-focussed resources are now complemented by websites such as Nitrogen News (www.nitrogennews.org) that provide online news and opinion articles, accessible background documents, and the chance for the public to interact with experts working on various aspects of nitrogen. In coming years, more e-resources focussed on the nitrogen challenge will be required, making use of emerging

technologies and delivery platforms (e.g. iPad applications, virtual world environments). The N-visualization tool is not a ‘policy-making’ instrument in itself, but rather it shows the complexities policy makers have to face, and that there are different ways to approach solving the nitrogen problem.

26.4.3╇ Useful messages and tools To effectively communicate the European nitrogen issue requires the targeted approach discussed throughout this chapter. This includes recognition of the issues that are most important to the target audience, their level of awareness, and the ultimate outcome that is desired as a result of the communication. For climate change, O’Neill and Hulme (2009) have suggested the use of an ‘iconic’ approach, whereby an iconic species, landscape or activity affected by climate change is employed to engage and inform a given community more successfully. In the case of reactive nitrogen, a similar approach could also prove successful where such local or regional ‘icons’ and their interaction with nitrogen can be identified. For example, the destruction of an iconic building due to nitric acid erosion, or the loss of keystone ombrotrophic wetland plant species due to reactive nitrogen deposition. Which icon is used will depend on its importance to the target community and how robustly the link between nitrogen and the highlighted consequences can be made. The nitrogen ‘issue’ as we have seen frequently in the preceding chapters is a complex one. Like climate change, such complexity necessitates well thought-through messages and tools to make the information accessible to a wide range of audiences. There is a requirement for useful analogies and metaphors to be developed that will better help communicators get their message across. Similarly, the further development of tools, such as N-cockpit, can help communicators and educators reach stakeholder groups (e.g. children) who would find the existing nitrogen-relevant knowledge exchange resources uninteresting and/or inaccessible (see Figure 26.10). Below we provide an example of this segmented communication approach in the context of raising nitrogen awareness in Europe.

Nitrogen Young Scientists In 2008 a Workshop was organized jointly by the ESF Nitrogen in Europe programme and the NitroEurope Integrated Project to encourage enthusiastic young scientists to develop novel ideas for engaging the public with research into nitrogen and its effects on the environment. One stage of this process was to produce a list of messages that could be discussed with different target audiences (Figure 26.11). In order to be focused, the group concentrated on three target audiences:€ interested adults; schools and policy makers/NGOs. During the course of the workshop, the young scientists worked in groups to develop prototype communication tools aimed at one or more of these target audiences. Examples included:€illustrated stories and role-plays introducing the different forms of nitrogen, their associated environmental impacts and possible solutions; suggested lesson plans for schoolchildren;

597

Societal choice Slash & Burn

Introduction of Transport

Invention of Art. Fertilizer

Intensification of crop production

Intensification of animal production

Globalisation

Biofuel production

Biofuel use for energy production

leaflets; short films; games and a website design. With additional development, these ideas will provide a resource for laboratory open-days, science fairs and scientist-teacher partnerships as well as an online source of information for the general public (for more information, see NitroEurope, 2010).

598

Figure 26.9 Graphics from N-Vis tool, showing steps where nitrogen plays a role, and with increasing nitrogen production/consumption.

26.5╇ Summary To summarize this chapter, we pose two key questions focused on the need for understanding of societal choice and communication in the context of the European nitrogen challenge.

Dave S. Reay Figure 26.10 Graphics from the N-cockpit vizualization and decision support tool.

Figure 26.11 The three selected target audiences, the key messages that could be used when engaging with these audiences, and the tools developed during the ‘Nitrogen Young Scientists’ Workshop’ to aid this communication. The colours indicate the different target audiences.

Do we need public awareness of nitrogen? Though public awareness of nitrogen remains very low across Europe and globally, this offers the chance for very significant improvements in nitrogen-related communication and knowledge exchange with the public. As we have discussed, increased awareness of the nitrogen issue may be able to change behaviour in certain groups of individuals (e.g. the ‘Positive Greens’), but such direct impacts of enhanced awareness are likely to be limited. More important are the wider indirect impacts that can be engendered and the way this can affect public acceptability of policy change. Climate change and carbon communication again serves as an exemplar of this, with the public acceptability of ‘carbon’ policies likely to be greater where levels of

understanding of the topic and the need for policy action are also high. The answer must therefore be that public awareness of nitrogen is needed, providing an essential foundation to develop solid actions.

Do we need more ‘policy’ awareness of nitrogen? Again the answer must be:€yes. However, with the myriad messages, competing priorities and mission briefs of policy making institutions in Europe, the application of an additional set of nitrogen assessment criteria to every decision may prove a challenge to be adopted by policy-makers. However, a very significant step change would be for the nitrogen-awareness level of policy-makers to at least reach a level where the basic

599

Societal choice

nitrogen implications of the policy are assessed. For some policy decisions nitrogen will have little real import and further consideration will be unnecessary. However, on many issues, such a ‘scope 1’ examination of nitrogen implications could serve to highlight potentially very important antagonisms, and indeed significant synergies that reinforce the putative policy’s aims. In the longer term a ‘mainstreaming’ of nitrogen issues in European policy making€– as already exists in many sectors for carbon€– could provide a strategy by which many of the key policy gaps, overlaps and antagonisms highlighted in earlier chapters could be effectively addressed. Finally, it is important to reiterate the central conclusion and recommendation of this chapter:€that a ‘segmented’ approach to engendering choice, behavioural change and communication is required. This approach is one that will need to evolve as attitudes, awareness and policy aims change over time. It has a well-established basis in the context of climate change mitigation and we believe that its adoption to address the European nitrogen challenge is one that can deliver far-reaching and lasting benefits for Europe and the world.

Acknowledgements The authors are grateful for financial support from the European Science Foundation programme Nitrogen in Europe (NinE), the COST Action 729, the UK Department for Environment, Food and Rural Affairs, and the European Commission Integrated Project NitroEurope.

References AEP (2007). European Association for Grain Legume Research. www. grainlegumes.com/aep/production/growing_gl_in_europe/grain_ legumes_crops_adapted_to_all_european_regions (Site accessed 1 August 2010). Bax, J. A. and Schils, R. L. M. (1993). Animal responses to white clover. In: White Clover in Europe:€State of the Art, pp. 7–16. FAO, Rome. Bickerstaff, K. and Walker G. (2001). Public understandings of air pollution:€the ‘localisation’ of environmental risk. Global Environmental Change, 11, 133–145. Blennow, K. and Persson, P. (2009). Climate change:€motivation for taking measure to adapt. Global Environmental Change, 19, 100–104. Brink, C., van Grinsven, H., Jacobsen, B. H. et al. (2011). Costs and benefits of nitrogen in the environment. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Büchner, F. L., Hoekstra, J., van den Berg, S. W., Wieleman, F. and van Rossum, C. T. M. (2007) Quantifying Health Effects of Nutrition, RIVM Report 350080001/2007. Carbon Disclosure Project (2010). www.cdproject.net (Site accessed 1 August 2010). CBF (2010). Chesapeake Bay Foundation. Your Bay Footprint. http:// www.cbf.org/yourbayfootprint/ (Site accessed 1 August 2010). Crutzen, P. J., Mosier, A. R., Smith, K. A. and Winiwarter, W. (2008). N2O release from agro-biofuel production negates global warming reduction by replacing fossil fuels, Atmospheric Chemistry Physics, 8, 389–395. De Boer, J., Helms, M. and Aiking, H. (2006). Protein consumption and sustainability:€diet diversity in EU-15. Ecological Economics, 59, 267–274.

600

Deckers, J. (2010) What policy should be adopted to curtail the negative global health impacts associated with the consumption of farmed animal products? Res Publica, 16, 57–72. DEFRA (2008). A Framework for Pro-Environmental Behaviours. www.defra.gov.uk/evidence/social/behaviour/documents/ behaviours-jan08-report.pdf Denman, K. L. et al. (2007) In: Climate Change 2007:€The Physical Science Basis, ed. S. Solomon, et al., pp. 499–587. Cambridge University Press. Dentener, F. et al. (2006). Nitrogen and sulfur deposition on regional and global scales:€a multimodel evaluation. Global Biogeochemical Cycles, 20, GB4003. Deugd, M., Roling, N. and Smaling, E. M. A. (1998). A new praxeology for integrated nutrient management, facilitating innovation with and by farmers. Agriculture, Ecosystems and Environment, 71, 269–283. EPA Victoria (2010). www.epa.vic.gov.au/ecologicalfootprint/ calculators/personal/results.asp (Site accessed 1 August 2010). de Vries, W., Leip, A., Reinds, G. J. et al. (2011). Geographic variation in terrestrial nitrogen budgets across Europe. In:€The European Nitrogen Assessment, eds. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Directgov UK (2010). UK Government Digital Service:€Carbon Calculator.€http://carboncalculator.direct.gov.uk/index.html (Site accessed 1 August 2010). Duffy, R. and Fearne, A. (2009). Value perceptions of farm assurance in the red meat supply chain. British Food Journal, 111, 669–685. EC (2008). Agriculture in the European Union: Statistical and Economic Information 2007. European Commission, Brussels. FAO (2006). Livestock Report 2006. Food and Agriculture Organization of the United Nations, Rome. FAO (2009). FAOSTAT Statistical Database. Food and Agriculture Organization of the United Nations, Rome. http://faostat.fao.org/ FEFAC (2008). Feed and Food Statistical Yearbook 2007. FEFACEuropean Feed Manufacturers Federation, Brussels. Flessa, H., Ruser, R., Dörsch, P. et al. (2002). Integrated evaluation of greenhouse gas emissions (CO2, CH4, N2O) from two farming systems in southern Germany. Agriculture, Ecosystems and Environment, 91, 175–189. Galloway, J. N., et al. (2004). Nitrogen cycles:€past, present, and future. Biogeochemistry, 70, 153–226. Galloway, J. N., Burke, M., Bradford, G. E. et al. (2007). International trade in meat:€the tip of the pork chop. Ambio, 36, 622–629. Gross, C., McKinney, S. and Harbans, L. (2007). Nitrogen credit trading tool to facilitate market based water quality trading. Proceedings of the Soil and Water Conservation Society, Florida, USA, July 2007. Halberg, N. (1999). Indicators of resource use and environmental impact for use in a decision aid for Danish livestock farmers. Agriculture, Ecosystems and Environment, 76, 17–30. Herrero, M., Thornton, P. K., Gerber, P. and Reid, R. S. (2009). Livestock, livelihoods and the environment:€understanding the trade-offs. Current Opinion in Environmental Sustainability, 1, 111–120. INI (2010). International Nitrogen Initiative. www.initrogen.org (Site accessed 1 August 2010). Kenyon, P. M. and Barker, M. E. (1998). Attitudes towards meat-eating in vegetarian and non-vegetarian teenage girls in England: an ethnographic approach. Appetite, 30, 185–198. Krishnamurthy, A., Moore, J. K., Zender, C. S. and Luo, C. (2007). Effects of atmospheric inorganic nitrogen deposition on ocean biogeochemistry. Journal of Geophysical Research, 112, G02019.

Dave S. Reay Lamarque, J.-F. et al. (2005). Assessing future nitrogen deposition and carbon cycle feedback using a multimodel approach:€analysis of nitrogen deposition. Journal of Geophysical Research, D19303. Legume Futures (2010). www.legumefutures.eu (Site accessed 1 August 2010). Leip, A., Achermann, B., Billen, G. et al. (2011). Integrating nitrogen fluxes at the European scale. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Lloyd-Williams, F., Mwatsama, M., Birt, C. et al. (2008). Estimating the cardiovascular mortality burden attributable to the Common Agricultural Policy on dietary saturated fats. Bulletin of the World Health Organization, 86, 535–545. Luick, R. (1998). Ecological and socio-economic implications of livestock-keeping systems on extensive systems in south-western Germany. Journal of Applied Ecology, 35, 979–982. McMichael, A. J., Powles, J. W., Butler, C. D. and Uauy, R. (2007). Food, livestock production, energy, climate change, and health. The Lancet, 370, 1253–1263. Moran, D., MacLeod, M., Wall, E. et al. (2008). UK marginal abatement cost curves for the agriculture and land use, land-use change and forestry sectors out to 2022, with qualitative analysis of options to 2050. In: 83rd Annual Conference, Agricultural Economics Society, March 30-April 1, 2009, Dublin, Ireland. Naylor, R., Steinfeid, H., Falcon, W. et al. (2005). Losing the links between livestock and land. Science, 310, 1621–1622. Nepstad, D. C., Stickler, C. M. and Almeida, O. T. (2006). Globalization of the Amazon soy and beef industries:€opportunities for conservation. Conservation Biology, 20, 1595–1603. Nicholson, F. A., Smith, K. A., Chambers, B. J. and Lord, E. I. (2000). Prediction of farm manure nitrogen availability and losses following land application (manner). Animal, Agricultural and Food Processing Wastes, 441–448. Nielsen, S. J. and Popkin, B. M. (2003). Patterns and trends in food portion sizes, 1977–1998. Journal of the American Medical Association, 289, 450–453. NinE (2009). The Barsac Declaration:€Environmental Sustainability and the Demitarian Diet, prepared under the lead of the Nitrogen in Europe (NinE) programme of the European Science Foundation. www.nine-esf.org/sites/nine-esf.org/files/Barsac%20 Declaration%20V3.pdf. NitroEurope IP (2010). NitroEurope Young Scientists’ Forum. www. nitroeurope.eu/ysf (Site accessed 1 August 2010). Norat, T., Bingham, S., Ferrari, P. et al. (2005). Meat, fish and colorectal cancer risk: The European Prospective Investigation into Cancer and Nutrition. Journal of the National Cancer Institute, 97, 906–916. Oenema, O., Bleeker, A., Braathen, N. A. et al. (2011). Nitrogen in current European policies. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Oikonomou, V., Becchis, F., Steg, L. and Russolillo, D. (2009). Energy saving and energy efficiency concepts for policy making. Energy Policy, 37, 4787–4796. Olesen, J. E., Schelde, K., Weiske, A. et al. (2006). Modelling greenhouse gas emissions from European conventional and organic dairy farms. Agriculture, Ecosystems and Environment, 112, 207–220.

O’Neill, S. J. and Hulme, M. (2009). An iconic approach for representing climate change. Global Environmental Change, 19, 402–410. Paracchini, M. L., Petersen, J.-E., Hoogeveen, Y. et al. (2008). High Nature Value Farmland in Europe: An Estimate of the Distribution Patterns on the Basis of Land Cover and Biodiversity Data. Joint Research Centre, Institute for Environment and Sustainability, Ispra, Italy. Petersen, S. O., Regina, K., Pöllinger, A. et al. (2006). Nitrous oxide emissions from organic and conventional crop rotations in five European countries. Agriculture, Ecosystems and Environment, 112, 200–206. Pike, T. (2008). Understanding Behaviours in a Farming Context:€Bringing Theoretical and Applied Evidence Together from across DEFRA and Highlighting Policy Relevance and Implications for Future Research. DEFRA, London. Rochon, J. J., Doyle, C. J., Greef, J. M. et al. (2004). Grazing legumes in Europe:€a review of their status, management, benefits, research needs and future prospects. Grass and Forage Science, 59, 197–214. Sampei, Y. and Aoyagi-Usui, M. (2009). Mass-media coverage, its influence on public awareness of climate-change issues, and implications for Japan’s national campaign to reduce greenhouse gas emissions. Global Environmental Change:€Human Policy Dimensions, 19, 203–212. Sebek, L. B. J. and Temme, E. H. M. (2009). Human Protein Requirements and Protein Intake and the Conversion of Vegetable Protein into Animal Protein, External Report. Animal Sciences Group, Wageningen UR, Wageningen, The Netherlands. Simon, M. F. and Garagorry, F. L. (2005) The expansion of agriculture in the Brazilian Amazon. Environmental Conservation, 32, 203–212. Smil, V. (2004) Enriching the Earth. MIT Press, Cambridge, MA. Steinfeld, H., Gerber, P., Wassenaar, T. et al. (2006). Livestock’s Long Shadow: Environmental Issues and Options. FAO, Rome. Svirejeva-Hopkins, A., Reis, S., Magid, J. et al. (2011). Nitrogen flows and fate in urban landscapes. In:€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. UNECE (2010). Options for Revising the 1999 Gothenburg Protocol to Abate Acidification, Eutrophication and Ground-Level Ozone: Reactive Nitrogen, Report by the co-chairs of the Task Force on Reactive Nitrogen, Presented to the Working Group on Strategies and Review 47th session. ECE/EB.AIR/WG.5/2010/13. www.unece. org/env/lrtap/WorkingGroups/wgs/docs47th%20session.htm UNEP and WHRC (2007). Reactive Nitrogen in the Environment:€Too Much or Too Little of a Good Thing. United Nations Environment Programme, Paris. Urwin, K. and Jordan, A. (2008). Does public policy support or undermine climate change adaptation? Exploring policy interplay across different scales of governance. Global Environmental Change, 18, 180–191. Ventour, L. (2008). The Food We Waste, Food Waste Report version 2, RBC405–0010. Waste and Resources Action Plan. WRAP (Waste and Resource Action Plan) (2008). The food we waste. http://www.wrap.org.uk. WRI/WBCSD (2010). The Greenhouse Gas Protocol Initiative. World Resources Institute and the World Business Council for Sustainable Development. www.ghgprotocol.org (Site accessed 1 August 2010).

601

Glossary

Glossary Item Acidification (with respect to soil) Aerosols Algal blooms

Ammonia

Ammonium AN Anamox

Anoxia Anthropogenic Anthroposphere Atmospheric deposition

Autotrophy

Benthos Biodiversity

Bioenergy

602

Explanation Acidification is used to describe the loss of nutrient bases (calcium, magnesium, potassium) in the soil, through leaching, and their replacement by acidic elements (hydrogen and aluminium). Pollutant deposition (e.g. NOx and Ammonia) enhances the rate of acidification. Solid or liquid particles suspended in the air. This includes dust, soot and sea-salt crystals, size 1 nm to 100 μm. Rapid increase or accumulation of algal biomass in a body of water, due to excessive nutrient loading (see also ‘Eutrophication’). Bloom forming algae can be toxic or inedible and their decomposition depletes oxygen in the water body, leading to ‘hypoxia’ and ‘anoxia’, with resultant lethal effects on fish and benthos. See Grizzetti et al. 2011, Chapter 17 this volume, and references therein. A reactive nitrogen form (NH3), which is colourless gas with a pungent odour when in high concentration. A product of ‘biological nitrogen fixation’ (see entry), it can also be synthesized using the ‘Haber–Bosch’ process (see entry). Most industrially synthesised ammonia is used in the manufacture of synthetic fertilizers, however some other industrial uses exist. A reactive nitrogen form (NH4+), closely linked to production and destruction of ‘ammonia’ (see entry). Constituent of many synthetic fertilizers, such as ammonium nitrate. A reactive nitrogen form, ammonium Nitrate (NH4NO3). A common synthetic nitrogen fertilizer in Europe, and present in atmospheric aerosol. ‘ANaerobic AMmonia OXidation’. An alternative pathway (to denitrification) for the generation of N2 from Nr. Bacteria of the group Planctomyces fix CO2 and use NH4+ to reduce NO2−, which results in the production of N2 (Voss et al. 2011, Chapter 8 this volume). The situation where the dissolved oxygen concentration of a water body is zero. See also ‘hypoxia’. Effects which relate specifically to human activities, i.e., anthropogenic reactive nitrogen production, through the ‘Haber–Bosch’ process. All parts of the planetary system which are affected by human activities. Removal of suspended material from the atmosphere, this can be classed as either ‘wet’ or ‘dry’. Wet deposition occurs when material is removed from the atmosphere by precipitation. In dry deposition, the material is removed from the atmosphere by contact with a surface. (For more detail see Hertel et al. 2011, Chapter 9 this volume.) When discussing agricultural production in regions or watersheds, autotrophy refers to areas where food for humans and feed for livestock is produced locally and any excess products are exported. See also Â�‘heterotrophy’ in this setting where the food and livestock feed is imported into the system. (See also Billen et al. 2011, Chapter 13 this volume.) Organisms that live on, in, or near the seabed, also known as the benthic zone. Biodiversity is the variability among living organisms, from genes to the biosphere. The value of biodiversity is multifold, from preserving the integrity of the biosphere as a whole, to providing food and medicine, to spiritual and aesthetic well-being (see Dise et al. 2011, Chapter 20 this volume). Energy derived from biofuel sources€– see ‘biofuels’.

Glossary

Glossary Item Biofuels

Biological Nitrogen Fixation BNF Bryophytes CAN CBD Carbon sequestration Catch crops Cloud condensation nuclei CLRTAP Common Agricultural Policy (CAP) Cost–benefit analysis Critical level Critical loads Cross-compliance

DALY Denitrification DIN Di-nitrogen

DON DPSIR framework Endogenous nitrosation Epidemiological studies Eutrophication

Explanation Biofuels are fuels from plant derivatives that gain their carbon from the atmosphere as CO2, and when they are burnt no new CO2 is released, with the aim of being ‘carbon-neutral’ (see also Reay et al. 2011, Chapter 26 this volume). ‘Fixing’ of unreactive di-nitrogen (N2) to reactive nitrogen species by microorganisms. Microorganisms which can fix nitrogen are called Diazotrophs. (See Erisman et al. 2011, Chapter 2 this volume.) Biological Nitrogen Fixation. The earliest land plants on earth, consisting of mosses, liverworts and hornworts. They are non-vascular land plants which are pioneer colonists of bare or disturbed ground. A reactive nitrogen form, Calcium Ammonium Nitrate. A common synthetic nitrogen fertilizer in Europe. Convention on Biological Diversity, under the United Nations Environment Programme (UNEP). The capture and removal of carbon dioxide from the atmosphere and storing it in an alternative carbon related reservoir, e.g. soil organic matter, charcoal, tree growth. Crops grown after the main crop has been harvested to retain nutrients (especially nitrogen) in order to prevent environmental losses and to promote recycling of nutrients. Very small particles in the atmosphere, which are required to physically mediate water vapour to coalesce to form clouds of liquid water. Convention on Long-range Transboundary Air Pollution, under the United Nations Economic Commission for Europe (UNECE). System of European Union agricultural subsidies and programmes, combining a direct subsidy payment for crops and land which may be cultivated with price support mechanisms, including guaranteed Â�minimum prices, import tariffs and quotas on certain goods from outside the EU. Economic tool to explicitly or implicitly, weigh the total expected costs against the total expected benefits of one or more actions in order to choose the best or most profitable option. Concentration or cumulative exposure of atmospheric pollutants above which direct adverse effects on sensitive vegetation may occur according to present knowledge. A quantitative estimate of an exposure to one or more pollutants below which significant harmful effects on specified sensitive elements of the environment do not occur according to present knowledge. Instrument of Common Agricultural Policy (CAP) where single payments to farmers are linked to Â�meeting environmental, public, animal and plant health and animal welfare standards and the need to keep land in good agricultural and environmental condition. Disability Adjusted Life Years:€a measure of overall disease burden, expressed as the number of years lost due to ill-health, disability or early death. The microbial regeneration of di-nitrogen (N2) or nitrous oxide (N2O) from nitrate (NO3−). N2O Â�represents an intermediary on the overall pathway of denitrification to form N2. Dissolved Inorganic Nitrogen N2, a colourless, odourless, and unreactive gas which makes up around 78% of the atmosphere. Di-nitrogen is the thermodynamically stable state (‘unreactive nitrogen’) to be contrasted with many Â�different reactive nitrogen forms. Dissolved Organic Nitrogen. Framework of causality and policy response:€Driver€– Pressure€– State€– Impact€– Response (see Oenema et al. 2011, Chapter 4 this volume, for further information). Process in the body of converting organic compounds into nitroso derivatives, e.g. N-nitrosamines, Â�including the carcinogenic variety. N-nitrosamines arise from the reaction of nitrite sources with amino compounds (see also nitrate and nitrite). Studies into health and disease within the population. Studies aim to relate the disease pattern to factors such as pollution or infectious agents. The enrichment of the nutrient load in ecosystems (terrestrial and aquatic), especially compounds of nitrogen and/or phosphorus. This leads to an undesirable disturbance to the balance of organisms in the ecosystem, affecting terrestrial and aquatic biodiversity and water quality. (See also Durand et al. 2011, Grizzetti et al. 2011 and Dise et al. 2011, Chapters 7, 17 and 20, this volume.)

603

Glossary

Glossary Item Exceedance GHG Gothenburg Protocol

Greenhouse gas balance GPNM Haber–Bosch process Heterotrophy (wrt nitrogen budgets and watersheds) Hyporheic zone Hypoxia Immobilization (in soil) Leaching Legumes

Lentic Lichen

Methane

Methemoglobinemia

Morbidity N fixing bacteria

N2 N2O Natura 2000

604

Explanation The amount of pollution above a ‘critical level’ or ‘critical load’, expressed in different ways, such as accumulated area of exceedance. Greenhouse Gas€– includes carbon dioxide (CO2), nitrous oxide (N2O), methane (CH4), ozone (O3), water vapour and various other gases. A multilateral environmental agreement signed under the UNECE Convention on Long-range Transboundary Air Pollution (CLRTAP), which sets maximum emissions to the atmosphere of SO2, NOx, NH3 and VOCs for national parties to the protocol to be achieved by 2010 (see Oenema et al. 2011, Chapter 4 this volume). The balance of greenhouse gases (‘GHG’) in the atmosphere. In addition to the contribution of greenhouse gases, the overall ‘radiative forcing’, is also affected by the contribution of aerosol components. Global Partnership on Nutrient Management (see Bull et al. 2011, Chapter 25 this volume). The high pressure chemical process which synthesizes reactive nitrogen as ammonia (NH3) from the reaction of N2 and H2. Fritz Haber was responsible for the discovery of the process (1908) and Carl Bosch later developed the technique on an industrial scale. When discussing agricultural production in regions or watersheds, heterotrophy refers to areas where either food for the human population is imported from elsewhere or feed for livestock production is imported from elsewhere. See also ‘autotrophy’ in this setting, where locally produced food and feed is used and any excess products are exported (Billen et al. 2011, Chapter 13 this volume). A region beneath and lateral to a stream bed, where there is mixing of shallow groundwater and surface water. The situation where the dissolved oxygen concentration of a water body is very low, e.g. <10 μmol of oxygen per litre of water (see Voss et al. 2011, Chapter 8 this volume). The incorporation of compounds (such as reactive nitrogen) into soil microbial biomass. The washing out of soluble ions and compounds by water draining through soil. Plants which are able to fix nitrogen from the atmosphere (see ‘Biological Nitrogen Fixation’), due to root nodules which contain rhizobia bacteria, which act with the plant in a symbiotic relationship. Legumes can be used by farmers to replenish the reactive nitrogen levels in the soil, in a crop rotation sequence. Ecosystem of standing or still water, e.g., a lake, pond or swamp. Lichen is a partnership between a fungus and an alga. The algae can photosynthesize, providing organic nutrients (in some cases it can also fix nitrogen) and the fungus provides water, nutrients and gases from the environment. The fungus also prevents the drying out of the algae and damage by excessive sunlight. A greenhouse gas (CH4), which is 21 times more effective at trapping heat in the atmosphere than carbon dioxide (CO2) over a 100 year period. Methane is the major constituent of natural gas, it is formed biogenically in anaerobic environments (those without oxygen). It is also a major product of enteric fermentation (the digestive process of cows, sheep, etc.). A disorder characterized by the presence of a higher than normal level of methemoglobin (metHb) in the blood. Methemoglobin is an oxidized form of hemoglobin that has almost no affinity for oxygen, resulting in almost no oxygen delivery to the tissues. Formation of metHb is promoted, e.g., by high levels of nitrite, often precursed by nitrate. Diseased state, disability, or poor health due to any cause. Often used in conjunction with mortality. A bacteria which is able to fix nitrogen from the N2 in the atmosphere (also known as a diazotroph, see also ‘biological nitrogen fixation’ and ‘legumes’). They can be free-living or symbiotic with other organisms. See ‘Di-nitrogen’. See ‘Nitrous oxide’. Ecological network of protected areas in the territory of the European Union, brought about by the implementation of the EU Habitats Directive (1992).

Glossary

Glossary Item National Emissions Ceilings Directive

NH3 NH4+ NHx Nitrate Vulnerable Zones Nitrate(s)

Nitrates Directive Nitric oxide Nitrification Nitrite

Nitrogen cascade Nitrogen dioxide Nitrogen oxides

Nitrogen use efficiency Nitrophilic Nitrous oxide

NO2 NO2− NO3− NOx NOy Nr O3 Oligotrophic OSPAR Oxidative stress

Explanation A European Commission Directive (1999), often referred to as the NEC Directive, which sets maximum emissions to the atmosphere of SO2, NOx, NH3 and VOCs for the European Union to be achieved by 2010. The NEC Directive is part of the EU transposition of the UNECE Gothenburg Protocol (see Oenema et al. 2011, Chapter 4 this volume). See ‘Ammonia’. See ‘Ammonium’. Sum of ammonia and ammonium. Designated areas of land which drain into areas of water which are polluted by nitrates from agriculture€– as defined by the ‘Nitrates Directive’. In these areas, mandatory rules exist for agricultural practices which can lead to nitrate loss into the environment. A reactive nitrogen form, (NO3−) a nitrate ion. Nitrate salts are soluble in water and very mobile. Nitrates are an important nutrient for crop plants, but in high concentrations can cause eutrophication, especially in semi-natural ecosystems and cause adverse health effects when present in excess in drinking water. A European Commission Directive (1991), which regulates agricultural practices which can lead to losses of nitrate to the environment. It is part of the Water Framework Directive. A reactive nitrogen form (NO), formed mainly in combustion processes and also emitted during nitrification and denitrification processes in soils. The microbial conversion of ammonium ions (NH4+) to nitrate (NO3−). A reactive nitrogen form (NO2−) a nitrite ion. It has a high oxidative potential and is therefore used in cured meats.€Nitrite is typically present in much less abundance than nitrate (NO3−) and in high concentration is toxic to humans. A term used to describe the passage of reactive nitrogen through the environment (see Sutton et al. 2011, Chapter 1 this volume). A reactive nitrogen form (NO2), is an oxide of nitrogen formed mainly by combustion processes where fuel N is oxidized or atmospheric N2 is oxidized at high temperatures. A reactive nitrogen form, refers specifically to the sum of NO (nitric oxide) and nitrogen dioxide (NO2), also known as NOx. As with other oxides of nitrogen these oxides are formed mainly by combustion processes where fuel N is oxidized or atmospheric N2 is oxidized at high temperatures. The ratio of nitrogen input and ouput of a system (e.g., soil, plant, farm animal, farm). Various techniques can be employed to increase nitrogen use efficiency in crop and livestock systems (see Jarvis et al. 2011 and Oenema et al. 2011, Chapters 10 and 23 this volume). Literally translates to ‘nitrogen-loving’, i.e., a species which may be nitrogen limited and therefore benefits from an increase in reactive nitrogen availability in the environment. A reactive nitrogen form (N2O), also known as laughing gas. This is an oxide of nitrogen formed mainly by microbial denitrification processes in soils and waters. It is also emitted by combustion and other industrial processes. N2O is also a greenhouse gas which is 310 times more effective at trapping heat in the atmosphere than carbon dioxide over a 100 year period. See ‘Nitrogen dioxide’. See ‘Nitrite’. See ‘Nitrate(s)’. Sum of NO and NO2. Various forms of oxidized nitrogen including, NOx and nitrates. Reactive nitrogen. See ‘Ozone’. ‘Poor in nutrient’. The opposite of eutrophic (see Eutrophication). Convention for the Protection of the Marine Environment of the North-East Atlantic established in Oslo€– Paris (OSPAR) 1992. Imbalance between the production of reactive oxygen species and a biological system’s ability to readily detoxify the reactive intermediates or to repair the resulting damage.

605

Glossary

Glossary Item Ozone

PAN

Pedosphere Pelagic fish PM2.5/PM10

Pollution swapping Radiative balance Reactive nitrogen

Redfield ratio Riparian Ruminants

Runoff Saprotrophic decomposers Sewage sludge Stratospheric ozone depletion

Sulphur dioxide Synthetic fertilizer TAN TFRN Throughfall Tillage practices Troposphere Urea

Volatile Organic Compounds (VOCs) Willingness to pay (WTP)

606

Explanation Powerful oxidizing gas formed by the reaction of sunlight on air which contains hydrocarbons and nitrogen oxides. These can react to form ozone directly at the source of the pollution or many kilometers down wind. A reactive nitrogen form, peroxyacytyl nitrate (C2H3O5N), is one constituent of photochemical (formed with the input of sunlight) smog air pollution, both VOCs and nitrogen dioxide (NO2) contribute towards its generation. The planetary sphere which contains soil. Fish that live in the water column of coastal, ocean and lake waters. PM2.5/PM10 :€Aerosol mass contained in particles with an aerodynamic diameter below 2.5 (10) micrometre, measured with a reference technique. Used as a metric to assess the human health impact of particle air pollution (see Moldanová et al. 2011, Chapter 18 this volume). Applying an environmental measure which results in one form of pollution being reduced, but which then results in the introduction or increase in another form of pollution. Balance of ingoing and outgoing thermal radiation of the planet. Collectively any chemical form of nitrogen other than di-nitrogen (N2). Reactive nitrogen (Nr) compounds include NH3, NOx, N2O, NO3− and many other chemical forms, and are involved in a wide range of chemical, biological and physical processes. Redfield ratio or Redfield stoichiometry is the molecular ratio of carbon, nitrogen and phosphorus in phytoplankton. Interface between land and a river or stream. Mammals with a four-chambered complex stomach, that digests plant-based food by initially softening it within the animal’s first stomach, then regurgitating the semi-digested mass, now known as cud, and chewing it again. The process of rechewing the cud to further break down plant matter and stimulate digestion is called ‘ruminating’. Waterflow over land€– which occurs when the soil is saturated with water. Organisms which can generate energy from dead or decaying matter from plants or animals. Residual, semi-solid material left from industrial wastewater, or sewage treatment processes. Depletion of ozone in the stratosphere (the second layer of the atmosphere, situated above the ‘troposphere’). This depletion allows increased levels of UVB (a harmful form of ultraviolet radiation) to reach Earth’s surface. When the depletion is strong in a specific area, this is commonly referred to as an ‘ozone hole’. SO2, atmospheric pollutant mainly emitted from fossil fuel combustion. It is also a precursor (aids in the creation of) particulate matter in the atmosphere (see PM10 and PM2.5) which can be harmful to health. Industrially produced fertilizer, using the ‘Haber–Bosch’ process. Total Ammoniacal Nitrogen. Task Force on Reactive Nitrogen. A task force under the Working Group on Strategies and Review of the CLRTAP (see Bull et al. 2011, Chapter 25 this volume). Rainwater (precipitation) falling through the canopy (foliage) of a forest or crop. The agricultural preparation of the soil by ploughing, ripping, or turning it. The lowest portion of the Earth’s atmosphere, the depth of which varies geographically, being deepest at the tropics and shallowest at the poles. A reactive nitrogen form, urea (or carbamide) is an organic compound with the chemical formula (NH2)2CO. Urea is widely used in fertilizers as a convenient source of nitrogen. Urea is also an important raw material for the chemical industry. Organic compounds that easily vaporize at room temperature, e.g., benzene. Maximum amount a person would be willing to pay, sacrifice or exchange for goods or services.

Chapter

Index

N labelling fertilizers, 38 NHx deposition, 305 acid deposition, 24 acid rain, 63 acidification, 38 acidifying, 38 costs, 38 effects and indicators, 38 soil, 38 terrestrial ecosystems, 10, 38 acidity, 468 aerosols, 435, 450, secondary, 17 Agricultural buildings, 179 Agriculture, 441 costs and benefits, 533 Decreasing Nr losses, 83 economic, 517 European Nitrogen Budget, 365, 370 and greenhouse gases, 25 intensification, 34 dairy, 50 mapping fluxes, 325 N losses and policy, 74 N use in, 10, 322 soils, 500, 502, 503 system, 320 agro-ecosystems, N budgets, 318 agronomic efficiency, 38 air pollution, 116, 406, 415, 480, 527 Air Quality Directive, 222 air scrubbing, 524 aircraft, 427, 449 albedo, 451, 452 algal growth, 22 alkalinization, 500 ammonia, 339, 367, 406, 410, 417, 425, 483 15

emissions, 2 industrial uses, 34, 54, 367 slip, 185 ammonification, 13, 148 ammonium, 12, 40, 102, 104, 106, 107, 113, 115, 129, 135, 138, 258, 273, 290, 423, 500, 556, 587 ammonium nitrate, 17, 35, 54, 451 ammonium sulphate, 257 ancillary impacts, 517 animal manure, 112 animal protein, 593 animal welfare, 76, 595 annamox, 14, 100, 110, 135, 149, 150, 156 anoxic conditions, 388 anthropogenic, 384, 435, 514 reactive N, 11 modifications of N transport, 237 AOT40, 414, 421 aquifer, 381, 384 arctic habitats, 466 assimilation, 14 atmospheric deposition, 5, 273, 300, 301, 464 contribution to streams, 23 European N budgets, 301, 351, 355 global scale, 306 marine, 24 methane, 444 modelling, 311, 312 observations, 299, 306 variability, 242 atmospheric transport/ transfer, 5, 237, 355 NOx, 19 Landscape scale, 236 autotrophy, 128, 274, 359

aviation, 370, 408 Baltic Sea Action Plan, 397 atmospheric deposition, 303 N input, 24 policy, 63, 396, 529 Barcelona, 396 barley, 35, 47 baseline, 559 benefits, 33, 514, 517 Best Available Technology, 66, 71 bioaccumulation, 389 biodiversity, 10, 115, 160, 464 evidence, 470 impacts and indicators, 87, 481 legislation, 481 recovery, 477 soil, 503, 506, 527 threats to, 465 valuation, 519 bioenergy (see also biofuels), 12, 13, 47, 70 bio-ethanol, 34 biofuels (see also bioenergy), 12, 57, 322, 554, 563, 597 biomass (see also bioenergy, biofuels), 17, 34, 41, 367, 449 Birds Directive, 243 black carbon, 452 Black Sea, 396 bloom-forming algae, 388, 393 BNF, 1, 101, 102, 351, 365 cultivated, 11 process, 12, 13 bogs, 475

boreal habitats, 466 bryophytes, 466, 467, 470 Bucharest convention, 396 budgets, 591 buffer zone, 163, 234, 398 buffering, 136, 497 butterflies, 476 Câ•›:â•›N ratio, 441, 450, 501 calcareous soils, 468 calcium ammonium nitrate, 35, 552 cancer, 385, 521 canopy, 236, 299 carbon, accumulation, 446 sequestration, 115, 370, 435, 446, 450, 503 storage, 450, 452 carbon–nitrogen interactions, 2 cardiovascular health, 385 carnivorous plants, 475 catalytic converters, 76, 178 catch crop, 396, 525, 526 caterpillars, 476 CH4, 435, 442, 444, 445, 454 Chemistry Transport Models (CTM), 301 Emissions input, 311 Performance vs. observations, 307 chlorophyll, 291 climate change, 26, 392, 394, 397, 406, 436, 456, 465, 468, 480, 517 cloud condensation nuclei, 451 clover, 36 CLRTAP, 4, 24, 63, 77, 391, 406, 407, 482, 504, 527, 571 CO2, 504

607

Index

coastal regims, 2, 5, 367, 381, 441, 466 co-benefits, 517 code of good agricultural practice, 215 combustion, 258, 259, 260, 264, 357, 406, 546, 556, 586 Common Agricultural Policy, 35, 64, 68, 395 communication, 596 communicative instruments, 64 compensation point, 195 consumption, 548, 587, 590, 593 contamination, 497 controlled release, 40 Convention on Biological Diversity, 4, 464, 481, 577 COPERT, 183 CORINAIR, 182 corridor, 234 corrosion, 415 cost, 515, 525 effectiveness, 543 benefit, 543 cost–benefit analysis (CBA), 396, 514, 516, 527 Cost-effectiveness analysis (CEA), 514 COTAG, 192 critical level, 24, 340, 414, 421, 481 critical load, 24, 326, 391, 408, 482, 483, 504 crops, 356 economics, 518 fertilizer use, 35 harvested product quality, 46 major crops in EU, 35 mosaic, 235 N imports to farm, 236 N uptake, 40 ozone effects, 421, 450 productivity and quality, 40, 41 residue, 220 rotation, 219 soil, 497 cross-compliance in EU policy, 69, 244 current legislation, 558

608

cyanobacteria, 129, 132, 139, 154, 165 dairy farming, 49, 532 Economic value, 52 Unit benefits, 519 ‘dead zones’, 2, 388 deep ocean, 356 denitrification, 2, 13, 25, 100, 106, 110, 131, 136, 138, 139, 149, 156, 159, 355, 365, 396, 398, 435, 439, 441, 442, 497, 499 aerobic, 218, 260, 261, 290 autotrophic, 135 heterotrophic, 135 microbial, 110 pyro-, 110 deposition velocity, 193 diesel, 427 diet human, 554, 563 livestock, 47 di-nitrogen, 1, 2, 337 disability-adjusted life years (DALY), 524 discounting, 515, 516 dissolution, 504 dissolved inorganic nitrogen (DIN), 129, 159, 275, 442 dissolved organic carbon (DOC), 131, 136 dissolved organic nitrogen (DON), 24, 105, 106, 113, 129, 130, 135, 148, 159, 190, 213, 223 ditch drainage system, 237, 396 DNRA, 108, 149, 150 dose–response, 410, 411, 415, 520 double mobile, 542 downwind gradient, 305 DPSIR framework, 65, 543, 545 drinking water, 381, 384, 385, 386, 525 pollution, 398 quality, 386 soils, 499, 500 sources, 385 Drinking Water Directive, 27, 385, 386 drivers, 64, 349

driving forces (see DPSIR), dry deposition, 299, 302, 310, 312 dynamic modelling, 558 EC Biodiversity Conservation Strategy (ECBS), 481 EC National Emissions Ceiling (NEC) Directive, 18, 66, 243, 391, 395, 407, 483, 527 EC4MACS, 560 Ecological Surveillance Networks, 470 economic impacts, 421 economic instruments, 64 economic optimal N application rate (EoNR), 42, 43, 45 milk, 52 economic returns on investment (ERON), 34, 56 economic value, dairy farming, 52 reactive N, 56 ECOSAN, 262 ecosystem, aquatic, 380, 386 atmospheric deposition, 302 biodiversity, 470 CO2 exchange, 446 costs and benefits, 532 effects, 22 health and functioning, 519 N budget, 338 N processes, 236 resilience, 384 sensitivity, 465 services, 384, 396, 398, 464, 506, 517, 519, 557, 590 ecotone, 234 EDGAR, 182, 312, 347 edge effect, 103 effects-based approach, 543 egalitarian principle, 529 EMEP model, 182, 301, 307, 481 emission ceilings (LRTAP Convention), 406 emission factors, 26, 437 emissions abatement, 75

emissions standards, 71 empirical modelling, 478 energy production (effects on N cycle), 17, 365 environmental effects of nitrogen, 10, 87 environmental Management, 393 environmental policy, 63 epiphytes, 103 equity, principle of, 529 erosion, 497 estuary, 157, 289 Eulerian models, 300, 306 European Fertilizer Manufacturers Association (EFMA), 556 Eutrophication, 10, 17, 22, 133, 138, 149, 155, 163, 164, 165, 168, 231, 284, 291, 340, 380, 381, 384, 386, 389, 393, 466, 529, 558 aquatic ecosystems, 467 Baltic Sea, 519, 520 biodiversity, 482 costs, 24 critical loads, 395 effects, 24 implementation measures, 392 marine ecosystems, 502, 506 policy considerations, 391 soil, 501, 505 terrestrial ecosystems ex-ante assessment, 71 exceedances, 340, 482 excretion rates, 52 exogenous species, 389 explosives, 34, 54, 56 ex-post assessment, 71 exposure, 265 farm nitrogen cycle, 213, 226 budget, 221 farming systems, dairy, 49 landscape design and structure, 230, 235 management, 233, 237, 244 organic, 503 feed, livestock, 445, 524, 525

Index

fens, 466 fertilizer, 1, 382, 436, 504, 517, 519, 524, 535 ammonium nitrate, 54, 450 coastal waters, 442 consumption in Europe, 17, 35, 338, 392 crops, 35 export, 20 manure, 1, 13, 40, 179, 212, 218, 339, 350, 525, 546, 556 acidification, 500 application methods, 40, 41 incineration, 524 low-emission storage, 41 management, 50 N2O, 439 N inputs, 36 soil biodiversity, 503 SOM, 34 mineral, 13, 33, 319, 350 soil, 502 synthetic, 10, 104, 112, 179, 217, 556 field Survey, 471 fish kills, 22 flux networks, 312 foliage enrichment, 466 impacts, 467 food-print, 250, 260 food wastage, 595 footprint, 250, 259, 563, 591 forbs, 471, 472 forests, 352, 370, 441, 444, 448, 466, 470, 501, 502, 505 atmospheric deposition in, 302, 446 fossil fuel, 10 Framework Directive on Ambient Air, 408 freeze–thaw, 109 freshwater, 127 GEIA, 182 GENEMIS, 182 genetic diversity, 464 GEO, 579 geo-engineering, 1, 5 global environmental change programmes, 575 global warming, 25

global Warming Potential (GWP), 436, 452 globalization, 10 Gothenburg Protocol, 4, 18, 391, 426, 504, 527, 543 governmental policies, 63, 64 grain, 46 graminoids, 471 grassland, 35, 466, 471, 472, 478, 505 grazing, 500 greenhouse gases, 3, 17, 435, 449, 505, 525 costs and benefits of N, 532 emissions and land use change, 33 peatlands, 439 ‘grey energy’, 250 groundwater, 135, 256, 381 nitrate concentrations, 338 nitrogen losses, 331, 336 pollution and leaching, 22 soil, 501 Groundwater Directive, 68, 243 GTAP, 557 H2SO4, 451 Haber–Bosch process, 1, 10, 54, 365, 367 Habitats Directive, 243 haptophytes, 168 harmful algal blooms (HABs), heathland, 446, 466, 475 heavy metals, 300 hedgerows, 234, 235, 237 Helophyte, 525 Helsinki Commission (HELCOM), 4, 63, 396 Henry’s law coefficient, 194 heterotrophy, 128, 157, 250, 274, 359 high-temperature combustion, 2 HNO3, 311, 415, 450 HONO, 189, 197, 311 human health, 2, 26, 63, 89, 340, 384, 385, 406, 410, 420, 520, 532 human welfare, 514 hydrological regime, 127

hydrology, 236, 237 hydrosystem, 277 hypertrophic streams, 130, 140 hypoxia, 155, 161, 389 ICEP, 284, 291 IGBP, 580 IGO, 571, 575 immobilisation, 497 impact assessment, economic, 514 impacts (see DPSIR), indicators, 89, 338, 339, 340, 348, 359, 396 industrial nitrogen fixation, 3 Industrial Revolution, 10 industry, 17, 258, 365, 367 Infection disease risk, 389 Inferential models, 300, 310 Integrated assessment modelling, 545, 552 cost, 514 landscape scale, 239 Integrated Nitrogen Budget (iNB), 360 integration, 3, 4, 87, 542 INTEGRATOR model, 561, 565 intensive animal husbandry, 532 intergovernmental organizations, 65 international conventions, 65 international treaties, 63 intervention points, 547 IPPC, 66, 243, 407 Lagrangian models, 302 lakes, 380, 391 land management, 501 land use Change, 33, 109, 393, 452 Heterogeneity, 231 land-use/land-use change and forestry (LULUCF), 370, 589 Management, 242, 394 N2O emissions, 439 productivity, 359 Soil, 444, 500 landscape assessment tools, 244 ecology, 230

description and function, 233 mitigation of N effects, 232, 524 modelling, 237 N cascade, 230 N transfer processes, 235 pollution transfer paths, 234 Large Combustion Plants Directive, 184 lateral flow, 236 leaching, 14, 22, 332, 337, 351, 393, 525 indicators, 340 modelling, 239 nitrate, 38 legislation, 407, 481 legumes, 36, 102, 180, 212, 214, 218, 220, 501, 592 lichens, 468, 470, 475, 545, 597 life cycle analysis, 534 lignification, 450 liming, 501, 504, 527 liquid manures, 40 livestock production, 13, 213, 223, 370, 554, 590, 592 intensive, 10 low-protein animal feeding, 504 management, 437 manure, 396 N2O Emissions, 47 Low NOx combustion technology, 18 macrophytes, 133, 161 manipulation experiments, 470 marginal abatement cost curve (MACC), 589 marine ecosystems, 24, 148 Marine Pollution Convention (MARPOL), 408 Marine Strategy Directive, 67, 395 Marine waters, 74 mass balance, 256 maximum feasible reduction, 558 Multi Criteria Decision Analysis (MCDA), 544, 545

609

Index

MEAs, 571 meat production, 10, 14 Mediterranean Sea, 304, 333, 396 mega-cities, 253 meteorology, 311 methaemoglobinaemia, 385, 521 metropolitan area, 253 microsite hypothesis, 107 Millennium Ecosystem Assessment, 384, 393, 519, 557, 565 mineral fertilizer equivalent (MFE), 40, 42, 52 mineralization, 105, 108, 114, 393, 497, 501 mitigation, of N at landscape scale, 242 modelling, 320 biodiversity, 478 comparison of budgets, 337 greenhouse gases, 239 inverse models, 337, 483 landscape N, 237 NH3, 239 nitrogen inputs, 327 regional landscape, 242 models, 299, 346 atmospheric, 299 DNDC, 320 EDGAR, 320 EMEP, 320, 347 EPIC, 320 ForSAFE, 480 GAINS, 320, 524, 529, 560, 565 GREEN, 320 IDEAg, 320, 346 IMAGE, 320 Initiator 2, 238 INTEGRATOR, 320, 347, 561, 565 LANAS integrated model, 239 MITERRA, 320 MOVE, 483 Nitroscape, 240 SMART2, 483 molasses licking blocks, 445 Monin–Obukhov length, 193 montane habitats, 466 Montreal Protocol, 522 moths, 476 motor Vehicles, 407

610

N2, 1, 13, 351, 365 global fixation, 36 industrial uses, 53, 54 N2O, 25, 365 agricultural, 236 emissions, 328, 334, 339, 352, 435, 437 from fertilizer manufacture, 17 indirect emissions, 231 wetland emissions, 231 European N budget, 337 exchange, 398 global cycle, 332 fluxes, 132, 150 livestock, 436 losses, 111 mitigation, 440, 525 model estimates, 334 sewage treatment, 439 soils, 498 stratospheric ozone depletion, 522 terrestrial, 441 trends, 440, 441 unit damage costs, 534 water quality, 356, 358, 366 National Emission Ceilings Directive (NECD), 183, 543, 546 National Integrated Nitrogen Budgets (NiNB), 360 Natura 2000, 244, 482 natural sources of reactive N, 11 net primary production, 100 nitrate, 2, 11, 115, 129, 133, 135, 338, 380, 451, 521 drinking water, 385, 525 exposure, 385 human health, effects on, 26 leaching, 38, 112, 114, 116, 133 organic manures, 40 particulate, 406, 407, 423 peroxy acetyl nitrate (PAN), 191, 198 pollution of groundwater, 22, 381 soils, 500 nitrate vulnerable zones, 24, 243, 397 Nitrates Directive, 22, 64, 67, 76, 77, 166, 212, 219,

243, 280, 380, 381, 395, 397, 504, 529, 531 nitric acid, 133, 186, 440 nitric oxide (NO), 406, 414 nitrification, 14, 100, 106, 108, 112, 131, 138, 149, 218, 261, 290, 435, 439, 497, 526 autotrophic, 106 heterotrophic, 107 bacterial, 150 nitrogen, 151, 183, 273 balance, 15, 318, 319 cascade, 2, 10, 360, 542, 552, 586 consequences of, 21 European vs. global perspective, 28 integrating, 85 quantifying, 346 rural landscapes, 230 budgets, 5, 318, 360, 545 global vs. Europe, 19, 21 heterogeneity, 230 terrestrial EU scale, 318, 322 farm N budget, 318, 320, 322 land, 319 data sets, 320, 346 country scale, 323, 334 emissions, 356 Ocascade cycle, 4, 104, 108, 109, 130 human intervention, 1, 12 main processes, 13, 14 deposition, 2, 100, 102, 110, 114, 115, 159, 339, 498 European totals, 20 fauna, 476 enrichment, 384 Fixation, 19, 36 crop, 273 bacteria, 13 biological, 274, 593 process, 1 Fluxes, 332, 347 industrial products, 34 inheritance, 1 losses, 331 management, 542, 571 plant uptake, 107 removal, 318

saturation, 24, 114 surplus (see also balance), 15, 318, 323, 339, 350, 382, 497, 504 throughput, 320 use efficiency (nitrogen efficiency), 1, 14, 20, 42, 215, 370, 394, 445, 525, 546 agro-ecosystems, 318 effect of animal breeding, 49, 50 NPK interactions, 34 optimization of, 33 pigs and poultry diet, 47 nitrogen visualization tool, 597 nitrogenase, 562 nitrous oxide, 2, 337, 450 NMVOC, 419 N-nitrosodimethylamine (NDMA), 385 NO, emissions, 112 non-agricultural systems, 333, 498 North Sea, 303, 396 NOx, 2, 3, 10, 17, 407, 414, 417, 426, 449, 450, 528 atmospheric deposition, 304 emissions, 338, 357, 367, 435 European Nitrogen Budget, 524 formation, 17–18, Energy, industry and transport, 419 human health, effects on, 26 ozone precursor, 411 particulate Matter, 336, 365 policy, 66 satellite observations, 311 soils, 440, 499 atmospheric, 20, 236, 406 from combustion, 73 European, 19 Standards for, 66 trends, 332 nutrient balances, animal production, 50 farming systems, 49 grassland, 52 river basin scale, 396 nutrient enrichment, 395

Index

nutrient loading, 388 nutrition, 552 nutritional nitrogen, 100 OH radical, 189 oilseed rape, 35, 45, 518, 534 oligotrophic streams, 130, 165 ombrotrophic bog, 446, 466 optimal fertilization, 396 Organic nitrogen, 11, 133 amines, 11 proteins, 11 Organisation for Economic Cooperation and Development (OECD), 318 orographic enhancement, 199 Oslo and Paris Commission (OSPARCOM), 63, 74, 396 OSPAR, 4, 169, 581 oxidation states of N, 2 Oxidized nitrogen, 2, 151, 183, 273 formation of, 12 NO, 11, 301, 334, 427 NO2, 11, 311, 406, 410, 414, 415, 417, 427, 467 NO3, 11, 236, 311, 340 oxygen deficiencies, 22 ozone, 2, 117, 264, 427 biodiversity, 469 concentrations, 419 destruction of stratospheric ozone, 10, 450, 522 effects and indicators, 89 and human health, 10, 411 Plant growth, 454 policy, 66 Secondary pollutants, 406 on vegetation, 414 on materials, 415 Trends, 415 tropospheric, 17, 26, 406, 435, 449, 469 Paris Metropolitan area, 256, 267 participatory approach, 544 particles, 2, 312, 407 particulates, 10, 26, 66, 264, 406, 407, 411, 415, 417, 421, 427

peanut, 36 peatland, 439, 446, 471, 475, 503 photochemical smog, 406 photolysis, 188 photosynthesis, 100 phytoplankton, 133, 141, 150, 152, 154, 167, 169, 284 pigs diet, 47 farming policy, 66 plant functional types, 474 growth, 449, 452 metabolites, 445 stress, 468 ploughing, 237 plume models, 300 PM10, 407, 408, 415, 421, 422, 428 PM2.5, 407, 421, 423, 428 Po Valley, 357, 359, 382, 425 policy assesment of, 70, 514 biodiversity, 66, 483 bio-energy, 70 compliance, 64, 76 cost–benefit analysis, 514 instruments, 64 landscape scale, 233, 243 N measures in EU, 57, 65, 66 side effects, 516 targets, 71 water quality, 69, 395, 524 pollution swapping, 2, 398 poultry diet, 47 farming policy, 66 pressure, 349 primary production, 132, 152, 156, 159, 161 PRIMES, 561 Privately Optimal Nitrogen Rates (PONR), 517, 519, 534 process-based modelling, 478 protein consumption, 15, 18 public health, 593 pyrites, 499 radiation balance, 452 radiative forcing (RF), 436, 449, 450, 452, 454 rainout, 199

reactive nitrogen aquatic systems, 435 availability and crop productivity, 358 CH4 exchange, 33 consumption, 360, 442 emissions to hydrosphere, 356 industrial uses of, 53, 56 policy, 63 sources, 11, 63 temporal changes in, 89 recovery, of species composition, 477 reduced nitrogen, 557, 586 deposition, 20, 231 emission, 236, 328, 337, 338, 435, 524 European budget, 356, 365, 366, 635 Modelling, 305 NH3, 11, 13, 112, 301, 311, 324, 450, 467 NH4+, 11, 467, 478 sources, 12, 232 vegetation effects, 414 regulatory instruments, 64, 71 Representative Concentration Pathways (RCPs), 10, 435, 564, 566 reservoirs, 384 residence time, 128, 161, 164, 165, 273, 284, 384 resistance, 193 respiratory disorder, 410 response curves, 41 responses (see DPSIR), re-surveys, 470, 473 retention, 276, 283, 289 rice paddies, 444 river nitrogen export, 25 rivers, 380, 441 road transport, 264 Rubisco, 41 ruminants, 52, 435, 442, 445 runoff, 14, 239, 337, 340, 392 rural development policy, 69 salinisation, 497 salinity, 148 satellite, 311 scale effects, 22 scenario, 552 driver oriented, 552 effect oriented, 552, 557

SCOPE, 580 seas, 380 European, 303 N budget, 34 N export, 380 secondary organic aerosol particles (SOA), 406 security, 33 sedimentation ponds, 396 sediments, 372 segmented strategy, 587 selective catalytic redactors (SCR)/ selective noncatalytic redactors (SNCR), 18, 524 sensitive species, 474 set-aside, 239 sewage systems, 359, 439 sheep, 475 ship traffic, 304, 370, 408 shrubland, 466, 471 silicate, 468 slow-release fertilisers, 40 slurry, 40, 524 SO2, 301, 451 social costs, 514 socially optimal nitrogen rates (SONR), 517, 534 societal visualisation of threats, 56, 91 societal values, 10, 89, 586 soil acidification, 468, 497, 499, 527 biodiversity, 502 clay, 531 compaction, 496, 498 emissions, 312, 435, 437 nitrogen budget, 101, 192, 193 Nr stocks, 358 pollution and leaching, 22 quality, 527 respiration, 446 sandy, 531 upland, 442 soil mineral N (SMN), 42 soil nutrient supply, 50 soil organic matter (SOM), 34, 40, 105, 106, 116, 217, 446, 497, 501, 502, 504, 505 SOx deposition, 468

611

Index

soybean, 36 spatial nutrient management, 396 species richness, 464 stabilised fertilisers, 40 stemflow, 103 stratification, 148 stream order, 128 Streamlining European 2010 Biodiversity Indicators (SEBI2010), 481 subarctic habitats, 466 sulphuric acid, 451 sum of means over 35 ppb (SOMO35), 411, 427 surface water, 331, 337, 338, 380 synergies, 3 system boundaries, 318 systems analysis, 544, 545 Task Force on Reactive Nitrogen, 4 TEP, 154 terrestrial ecosystems, 24 tertiary treatment, 260, 289 TFHTAP, 580 Thematic Air Pollution Strategy EU (TSAP), 406, 527

612

thematic policy for soil protection, 496 throughfall, 103, 114 TIN, 133 total ammoniacal N (TAN), 178 trade, 557 transport production, 17, 367 tree belt, 242 tropical habitats, 337 tundra, 466 UK National Emissions Inventory, 312 UNECE, 4, 524 UNFCCC–IPCC, 4, 320, 337, 455 unit crop benefits (euro) per kg of N (UBoN), 518, 519 United Nations Conference on the Environment, 63 urban, 395, 397, 406, 427 density, 255 landscape, 250 nitrogen budget, 258 systems, 250 Urban Wastewater Treatment Directive, 24, 67, 74, 243, 395, 525

urbanisation, 250, 554 urea, 35, 54, 367, 524 urease inhibitors, 40 urine, 40 value of a life year (VOLY), 520 value of statistical life (VSL), 520 vegetation, 406, 408, 413, 421, 464 vertical flow, 236 virtual land use, 34 VOC, 427 volatilization, 14, 40, 435 fertiliser use, 38 terrestrial, 40 urease inhibitors, 441 urea use, 35 washout, 199 Waste Incineration Directive, 407, 440 wastewater, 138, 254, 257, 259, 260, 267, 288, 371, 395, 546 discharge, 23, 396 policy, 74 water pollution, 2 quality, 386, 392

Water Framework Directive (WFD), 66, 165, 243, 280, 390, 395, 397, 497, 544, 546 watershed, 5, 24, 272, 394 weathering of N from rocks, 14 wet deposition, 198, 299, 303, 308 wetlands, 136, 237, 242, 396, 398, 442, 444, 448, 466, 525 wheat, 33, 35 biofuel, 34 EONR, 45 protein content, 46 economics, 518 wildfire, 110 willingness to pay (WTP), 515, 520, 534 woodland, 237 World Health Organisation (WHO), 385 years of life lost (YOLL), 408, 517 yield, 389 crop, 41 grassland, 52 ozone, 421 response curves, 519