Peatlands, Volume 9: Evolution and Records of Environmental and Climate Changes (Developments in Earth Surface Processes)

DEVELOPMENTS IN EARTH SURFACE PROCESSES, 9 PEATLANDS: EVOLUTION AND RECORDS OF ENVIRONMENTAL AND CLIMATE CHANGES DE...

Author: I.P. Martini | A. Martinez Cortizas | W. Chesworth

56 downloads 1215 Views 20MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

DEVELOPMENTS IN EARTH SURFACE PROCESSES,

9

PEATLANDS: EVOLUTION AND RECORDS OF ENVIRONMENTAL AND CLIMATE CHANGES

DEVELOPMENTS IN EARTH SURFACE PROCESSES

Volumes 1 and 3 are out of print

2.

WEATHERING, SOILS & PALEOSOLS I.P. MARTINI and W. CHESWORTH (Editors)

4.

ENVIRONMENTAL GEOMORPHOLOGY M. PANIZZA

5.

GEOMORPHOLOGICAL HAZARDS OF EUROPE C. EMBLETON and C. EMBLETON-HAMANN (Editors)

6.

ROCK COATINGS R.I. DORN

7.

CATCHMENT DYNAMICS AND RIVER PROCESSES C. GARCIA and R.J. BATALLA (Editors)

8.

CLIMATIC GEOMORPHOLOGY M. GUTIE´RREZ

Cover illustration: Mires...mires! Gennadii Petrovich Sapozhnikov painted this watercolor (28 x 23 cm) in 1975. He was born in 1930 and became a State Factory worker in Tomsk, Siberia. His paintings bear witness to the beauty of the landscape but also to the harsh natural, historical and economic conditions of that region. [Donated by his daughter Lida (Lidiya Gennad’evna Shirokikh) and grandson Casha (Aleksandr Evgen’evich Shirokikh), through the auspices of a friend/colleague, Tania (Tatiana A. Bliakhartchouk)]

Developments in Earth Surface Processes, 9

PEATLANDS: EVOLUTION AND RECORDS OF ENVIRONMENTAL AND CLIMATE CHANGES

Edited by I.P. Martini Department of Land Resource Science, University of Guelph, Guelph, Ontario, Canada

A. Martı´ nez Cortizas Department of Soil Science and Agricultural Chemistry, Faculty of Biology, University of Santiago, Santiago de Compostela, Spain

W. Chesworth Department of Land Resource Science, University of Guelph, Guelph, Ontario, Canada

Amsterdam – Boston – Heidelberg – London – New York – Oxford Paris – San Diego – San Francisco – Singapore – Sydney – Tokyo

Elsevier Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK

First edition 2006 Copyright r 2006 Elsevier B.V. All rights reserved No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the publisher Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333; email: [email protected]. Alternatively you can submit your request online by visiting the Elsevier web site at http://elsevier.com/locate/permissions, and selecting Obtaining permission to use Elsevier material Notice No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Library of Congress Cataloguing in Publication Data A catalog record for this book is available from the Library of Congress British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN-13: 978-0-444-52883-4 ISBN-10: 0-444-52883-0 ISSN: 0928-2025

For information on all Elsevier publications visit our website at books.elsevier.com

Printed and bound in The Netherlands 06 07 08 09 10 10 9 8 7 6 5 4 3 2 1

Foreword This new work of collected papers on, Peatlands: Evolution and Records of Environmental and Climate Changes, edited by I.P. Martini, A. Martı´ nez Cortizas, and W. Chesworth, represents an important and up-to-date aggregation of material useful to geoscientists and bioscientists who are interested in the basic nature of these unique plant–landform assemblages. The causes of peat accumulations, associated landforms, botanical associations, inherent instabilities, and the environmental damage, as well as the records of past elemental and particulate inclusions buried in the organic accumulations are discussed at length by the 54 different authors of the 23 chapters. The book is divided into four main sections: the ﬁrst deals with the evolution and structure of peatland basins, the second with selected characteristics of peat and peatland environments, the third with peatlands as multi-signal archives of environmental changes, and the fourth with direct human impact on peatlands. Peatlands occur from the circum-polar regions to the tropics and from marine coastlines to high-mountain basins, although their greatest extent is in northern, boreal latitudes. The geochemistry, microbiology, and hydrology of peatlands, together with their physical instabilities, comprise a diversity and complexity of conditions that leads to a great variation for the authors to assess. Deposition of mercury and lead in peatland environments is an important geochemical-archival complement to other sediment and ice-core data records. Finally, the drainage and utilization of peatland resources is discussed in terms of the possibility of economic gain or of additional environmental problems. Accumulations of organic matter in lacustrine and palustrine settings in late Quaternary time have long been recognized as important sources of information on local vegetation through palynological assessments temporally controlled through radiocarbon dating, tephrochronology, and other means of dating. The assessment of peat in such settings has been a source of much useful understanding of past environmental conditions, although their study in many places has been hampered by inaccessibility, hazardous insect populations as disease vectors, and by degradation of peatlands through human manipulation or removal. Nonetheless, the authors and editors of this book have persevered and given us the detailed analysis that we need the better to understand many of the questions concerning these phenomena. Of particular note in this volume is the establishment of research agendas related to peatlands. In this time of global warming, where humankind’s role in modifying the composition of atmospheric gases is so central to understanding the root causes of global change, the assessment of carbon sources and sinks is critical. For example, more carbon is stored in the world’s soils – including peatlands, wetlands, and

vi

Foreword

permafrost – than the one exists in the atmosphere. Recent estimates of potential loss of carbon from peatlands alone due to global warming indicate that as much as a quarter to a ﬁfth of it may degrade by the year 2100, but considerable uncertainty exists about the precise temperature sensitivity of decomposition. A high-research priority should be assigned to constrain the sensitivity of peatlands to such decomposition in the coming years of accelerated climate change. Phytogeomorphology, the speciﬁc plant-landform portion of biogeomorphology, constitutes the study of essential land-cover factors. Such factors collectively form the basis of a powerful tool for the survey, assessment, history, management, and planning of our environment. Phytogeomorphologic assessments of peatlands are uncommon but when they are done, new understandings of often-overlooked phenomena can emerge. This book on peatlands offers important new observations at the intersection of four ‘spheres’ – the lithosphere, hydrosphere, atmosphere, and phytosphere – that have not been much discussed in this interacting context heretofore. Places of such intersection, like so many other interdisciplinary ventures, tend to be scientifically fertile ground to plow. John F. Shroder Jr. Editor-in-Chief Developments in Earth Surface Processes

Preface A relatively large literature exists on peatlands, congresses are regularly scheduled on the subject and several societies maintain excellent websites and issue special publications. Although much is known, surprising lacunas still exist. For example, several of the largest unconﬁned boreal to subarctic peatlands are only superﬁcially studied: their size and the high cost of ﬁeldwork are limiting factors. The attitude and approach to the study of peatlands have changed over the past two to three decades. Formerly, peatlands were simply a barrier to settlement, or a resource to be exploited. Now, we are aware of their ecological importance. Even though they cover no more than about 3% of the continental area they are critical components of the environment at both the large and small scales. Locally, for example, they act as ﬁlters to increasing amounts of toxic substances and other pollutants, whereas globally they are now recognized as a major factor in inﬂuencing the gaseous composition of the atmosphere, and may therefore contribute signiﬁcantly to climate change. Peatlands are a product of particular climatic and geological settings. Consequently, they contain within them information of value in reconstructing past environments and conditions. In effect, peatlands are a historical archive, complementary to similar archives derived from the study of glacial ice and lake sediments. By studying such features of peatlands as trace element content, ﬂora and fauna, a reliable picture can be constructed of the Earth’s recent past, especially since the end of the last glaciation, including the rise to dominance in the biosphere of Homo sapiens. All of this is invaluable in our current attempts to make sense of global change and the role of human beings bringing about change. This book stresses the work done in the past few decades. The ﬁrst section reports some of the largest boreal peatlands in the world and other less well-known mires of hilly and mountainous settings, and, by inference, comparing them with tropical systems of South East Asia. The second section of the book focuses on selected properties of the peatlands reviewing basic information and adding new ﬁndings on geochemistry, microbiology, hydrology, and instability of peat deposits. The third section treats peatlands as archives of information, using trace elements, predominantly lead and mercury, to assess pollution and environmental changes that have occurred over the last few thousand years. The fourth and ﬁnal section reports on case histories of peatlands directly impacted by man by draining them and locally using them as agricultural lands. We thank the contributors for their excellent work and for enduring gracefully and responding appropriately to reviewers’ and editors’ comments and advice. A volume

viii

Preface

like this would not be possible without the help, unselfishly given, of many people (colleagues, students, technicians, family members) and supporting organizations, and we sincerely thank them all. In particular, we would like to thank the scientiﬁc reviewers who helped considerably in focusing the message of the various chapters. In addition we acknowledge the help of the editorial staff at Elsevier, and of the Series Editor Professor J. Shroder for his attentive, ﬁnal reading of the manuscripts. We also thank the publishers who allowed the reproduction of speciﬁc ﬁgures, and they are noted in the appropriate captions. It is probably superﬂuous to say it, but what we present here can only be part of the story. There is still much to be done, and that is excellent news for all those young researchers coming along with bright, new ideas. I. P. Martini, A. Martı´nez Cortizas and W. Chesworth Editors

Contents Foreword Preface List of Contributors List of Reviewers 1

Peatlands: a concise guide to the volume I.P. Martini, A. Martı´nez Cortizas and W. Chesworth

A. Peatland basin analysis: evolution and structure 2

3

4

5

6

7

Northern Peatlands: their characteristics, development and sensitivity to climate change C. Tarnocai and V. Stolbovoy The cold-climate peatlands of the Hudson Bay Lowland, Canada: brief overview of recent work I.P. Martini Mountain mires from Galicia (NW Spain) X. Pontevedra-Pombal, J.C. No´voa Mun˜oz, E. Garcı´a-Rodeja and A. Martı´nez Cortizas Geomorphologic emplacement and vegetation characteristics of Fuegian peatlands, southernmost Argentina, South America A. Coronato, C. Roig, L. Collado and F. Roig The peatlands of Argentine Tierra del Fuego as a source for paleoclimatic and paleoenvironmental information J. Rabassa, A. Coronato, C.J. Heusser, F. Roig Jun˜ent, A. Borromei, C. Roig and M. Quattrocchio Lowland tropical peatlands of Southeast Asia S.E. Page, J.O. Rieley and R. Wu¨st

B. Selected characteristics of peat and peatland environments 8

The redox–pH approach to the geochemistry of the Earth’s land surface, with application to peatlands W. Chesworth, A. Martı´nez Cortizas and E. Garcı´a-Rodeja

v vii xiii xvii 1

15

17

53 85

111

129

145

173

175

Contents

x 9 10

11

12

13 14 15 16

Weathering of inorganic matter in bogs G. Le Roux and W. Shotyk Molecular chemistry by pyrolysis–GC/MS of selected samples of the Penido Vello peat deposit, Galicia, NW Spain P. Buurman, K.G.J. Nierop, X. Pontevedra-Pombal and A. Martı´nez Cortizas Mineral matter, major elements, and trace elements in raised bog peat: a case study from southern Sweden, Ireland and Tierra del Fuego, south Argentina L.G. Franze´n Consequences of increasing levels of atmospheric nitrogen deposition on ombrotrophic peatlands: a plant-based perspective L. Bragazza Microbial diversity in Sphagnum peatlands D. Gilbert and E.A.D. Mitchell Peatland hydrology J. Holden Hydrogeology of major peat basins in North America P.H. Glaser, D.I. Siegel, A.S. Reeve and J.P. Chanton Slope instability and mass movements in peat deposits A.P. Dykes and K.J. Kirk

C. Peatlands as multi-signal archives of environmental changes 17

18

19 20 21

Using bog archives to reconstruct paleopollution and vegetation change during the late Holocene T.M. Mighall, S. Timberlake, D.A. Jenkins and J.P. Grattan Beyond the peat: synthesizing peat, lake sediments and soils in studies of the Swedish environment R. Bindler and J. Klaminder Occurrence and fate of halogens in mires H. Biester, A. Martı´nez Cortizas and F. Keppler Mercury in mires H. Biester, R. Bindler and A. Martı´nez Cortizas Archiving natural and anthropogenic lead deposition in peatlands M.E. Kylander, D.J. Weiss, E. Peiteado Varela, T. Taboada Rodriguez and A. Martı´nez Cortizas

D. Direct human impact on peatlands 22

Impacts of artiﬁcial drainage of peatlands on runoff production and water quality J. Holden, P.J. Chapman, S.N. Lane and C. Brookes

197

217

241

271 287 319 347 377

407

409

431 449 465 479

499

501

Contents 23

Peatland subsidence in the Venice watershed M. Camporese, G. Gambolati, M. Putti and P. Teatini

Glossary Subject Index

xi 529

551 573

This page intentionally left blank

List of Contributors H. Biester Institute of Environmental Geochemistry, University of Heidelberg, Im Neuenheimer Feld 236, 69120 Heidelberg, Germany [email protected] R. Bindler Department of Ecology and Environmental Science, Umea˚ University, SE-901 87 Umea˚, Sweden [email protected] A. Borromei Departamento de Geologı´ a, Universidad Nacional del Sur, San Juan 670, (8000) Bahı´ a Blanca, Argentina [email protected] L. Bragazza Department of Natural and Cultural Resources, University of Ferrara, Corso Ercole I d’Este 32, I-44100 Ferrara, Italy [email protected] C. Brookes School of Earth Sciences, Victoria University of Wellington, New Zealand [email protected] P. Buurman Laboratory of Soil Science and Geology, Department of Environmental Sciences, Wageningen University. P.O. Box 37, 6700 AA Wageningen, The Netherlands [email protected] M. Camporese Department of Hydraulic, Maritime, Environmental, and Geotechnical Engineering, University of Padova, Via Loredan 20, Padova 35131, Italy [email protected] J. Chanton Department of Oceanography, Florida State University, Tallahassee, FL 32306, USA [email protected] P.J. Chapman School of Geography, University of Leeds, Leeds LS2 9JT, UK [email protected] W. Chesworth Department of Land Resource Science, University of Guelph, Guelph, Ontario, Canada N1G 2W1 [email protected] L. Collado Subsecretarı´ a de Recursos Naturales y Ambiente Humano, San Martı´ n 1410, 9410 Ushuaia, Argentina [email protected]

List of Contributors

xiv

A. Coronato Centro Austral de Investigaciones Cientı´ ﬁcas (CADIC-CONICET), Bernardo Houssay 200, 9410 Ushuaia, Argentina; and Universidad Nacional de la Patagonia San Juan Bosco, Sede Ushuaia. Darwin y Canga, 9410 Ushuaia, Argentina [email protected] A.P. Dykes Limestone Research Group, University of Huddersﬁeld, Queensgate, Huddersﬁeld HD1 3DH, UK [email protected] L.G. Franze´n Physical Geography, Earth Sciences Centre, P.O. Box 460, SE-405 30 Go¨teborg, Sweden [email protected] G. Gambolati Department of Mathematical Methods and Models for Scientiﬁc Applications, University of Padova, Via Belzoni 7, Padova 35131, Italy [email protected] E. Garcı´a-Rodeja Departamento de Edafoloxı´ a e Quı´ mica Agrı´ cola. Facultade de Bioloxı´ a, Universidade de Santiago de Compostela, Campus Sur s/n. 15706 Santiago de Compostela, Galicia, Spain [email protected] D. Gilbert Laboratoire de Biologie environnementale, USC INRA-EA 3184, Universite´ de Franche-Comte´, Place Leclerc, 25030 BESANCON cedex, France [email protected] P. Glaser Department of Geology & Geophysics, University of Minnesota, Minneapolis, MN 55455, USA [email protected] J.P. Grattan Institute of Geography and Earth Science, University of Wales, Aberystwyth, Wales SY23 3DB, UK [email protected] C.J. Heusser

Clinton Woods, Tuxedo, New York 10987, USA

J. Holden Earth and Biosphere Institute, School of Geography, University of Leeds, Leeds LS2 9JT, UK [email protected] D.A. Jenkins School of Agricultural and Forest Science, University of Wales, Bangor, Gwynedd LL57 2UW, UK [email protected] F. Keppler Max-Planck-Institut fu¨r Kernphysik, Bereich Atmospha¨renphysik, Saupfercheckweg 1, 69117 Heidelberg, Germany [email protected] K.J. Kirk

39 Dalston Road, Carlisle CA2 5NN, UK

J. Klaminder Department of Ecology and Environmental Science, Umea˚ University SE-901 87 Umea˚, Sweden [email protected]

List of Contributors

xv

M. Kylander Department of Earth Science and Engineering, Imperial College London, London SW7 2AZ, UK [email protected] S.N. Lane Department [email protected]

of

Geography,

Durham

University,

UK

G. Le Roux Laboratory of Geochronology, Institute of Environmental Geochemistry, University of Heidelberg, Im Neuenheimer Feld 234, D-69120 Heidelberg, Germany [email protected] A. Martı´nez Cortizas Department of Soil Science and Agricultural Chemistry, University of Santiago, E-15782 Santiago de Compostela, Spain [email protected] I.P. Martini Department of Land Resource Science, University of Guelph, Guelph, Ontario, Canada N1G 2W1 [email protected] T.M. Mighall Department of Geography and Environment, University of Aberdeen, Elphinstone Road, Aberdeen AB24 3UF, UK [email protected] E.A.D. Mitchell WSL-AR & Swiss Federal Institute of Technology Lausanne (EPFL), ENAC,ISTE,ECOS Station 2, CH-1015 Lausanne-Ecublens, Switzerland [email protected] K.G.J. Nierop IBED-Physical Geography, Universiteit Nieuwe Achtergracht 166, 1918 WV Amsterdam, [email protected]

van The

Amsterdam, Netherlands

J.C. No´voa-Mun˜oz A´rea de Edafoloxı´ a, Departamento de Bioloxı´ a Vexetal e Ciencias do Sol, Facultade de Ciencias, Universidade de Vigo, As Lagoas s/n, 32004 Ourense, Galicia, Spain [email protected] S.E. Page Department of Geography, University of Leicester, University Road, Leicester LE1 7RH, UK [email protected] E. Peiteado Varela Departamento de Edafologia y Quimica Agricola, Campus Sur s/n, 15782 Santiago de Compostela, Spain [email protected] X. Pontevedra-Pombal Lab. Paleoambiente, Patrimonio e Paisaxe, Dpto. Edafoloxı´ a e Quı´ mica Agrı´ cola, Fac. Bioloxı´ a, USC Campus Sur s/n. 15706 Santiago de Compostela. Galiza, Spain [email protected] M. Putti Department of Mathematical Methods and Models for Scientiﬁc Applications, University of Padova, Via Belzoni 7, Padova 35131, Italy [email protected] M. Quattrocchio Departamento de Geologı´ a, Universidad Nacional del Sur, San Juan 670, (8000) Bahı´ a Blanca, Argentina [email protected]

List of Contributors

xvi

J. Rabassa CADIC-CONICET, C.C. 92, 9410 Ushuaia, Tierra del Fuego, Argentina; and Universidad Nacional de la Patagonia-San Juan Bosco, Sede Ushuaia, Darwin y Canga, 9410 Ushuaia, Tierra del Fuego, Argentina [email protected] A.S. Reeve Department of Earth Sciences, University of Maine, Orono, ME 04469, USA [email protected] J.O. Rieley School of Geography, University of Nottingham, University Park, Nottingham NG7 2RD, UK [email protected] C. Roig Universidad Nacional de la Patagonia San Juan Bosco, Sede Ushuaia, Darwin y Canga, 9410 Ushuaia, Argentina [email protected] F. Roig

IADIZA-CRICYT-CONICET. C.C. 300. 5000 Mendoza, Argentina

F. Roig Jun˜ent IANIGLA, CRICYT-CONICET, C.C.330, 550 Mendoza, Argentina [email protected] W. Shotyk Institute of Environmental Geochemistry, University of Heidelberg, Im Neuenheimer Feld 236, D-69120 Heidelberg, Germany [email protected] D.S. Siegel Department of Earth Sciences, Syracuse University, Syracuse, NY 13244, USA [email protected] V. Stolbovoy Senior Soil Expert, IES, Joint Research Centre EC, via Fermi, TP.280, I-21020 Ispra (VA), Italy [email protected] T. Taboada Rodriguez Departamento de Edafologia y Quimica Agricola, Facultad de Biologia, Campus Sur s/n, 15782 Santiago de Compostela, Spain [email protected] C. Tarnocai Agriculture [email protected]

and

Agri-Food

Canada,

Ottawa,

ON,

Canada

P. Teatini Department of Mathematical Methods and Models for Scientiﬁc Applications, University of Padova, Via Belzoni 7, Padova 35131, Italy [email protected] S. Timberlake Geography, School of Science and the Environment, Coventry University, Coventry CV1 5FB, UK [email protected] D. Weiss Department of Earth Science and Engineering, Imperial College London, London SW7 2AZ, UK [email protected] R. Wu¨st School of Earth Sciences, James Cook University, Townsville, 4811, Qld., Australia [email protected]

List of Reviewers J. R. Bacon A. Baird R. Bindler L. Bragazza W. Chesworth A. P. Dykes J. G. Farmer A-J Francez L. G. Franze´n P. Glaser J. Gunn M. Heijmans J. Holden G. Hope

R. Kelly H. Knicker G. Malatoni A. Martı´nez Cortizas I. P. Martini

The Macaulay Institute, Craigiebuckler, Aberdeen, UK Department of Geography, Queen Mary, University of London, Mile End, London, UK Department of Ecology and Environmental Science, Umea˚ University, SE-901 87 Umea˚, Sweden Department of Natural and Cultural Resources, University of Ferrara, Ferrara, Italy Department of Land Resource Science, University of Guelph, Guelph, Ontario, Canada Limestone Research Group, University of Huddersﬁeld, Queensgate, Huddersﬁeld HD1 3DH, UK School of GeoSciences, University of Edinburgh, Edinburgh, UK Sciences de laVie et l’Environnement, Univerite´ de Rennes, France Physical Geography, Earth Sciences Centre, Go¨teborg, Sweden Department of Geology & Geophysics, University of Minnesota, Minneapolis, MN, USA Department of Environmental and Geographical Sciences, University of Huddersﬁeld, Queensgate, Huddersﬁeld, UK Wageningen University, Wageningen, The Netherlands Earth and Biosphere Institute, School of Geography, University of Leeds, Leeds LS2 9JT, UK Division of Archaeology and Natural History, Research School of Paciﬁc and Asian Studies, Australian National University, Canberra, Australia Ontario Geological Survey, Sudbury, Ontario, Canada Technische Universitat Munchen, Lehrstuhl fu¨r Bodenkunde, 85350 Freising-Weihenstephan, Germany Depto. de Cs. Biologicas, Fac. Cs. Exactas y Naturales, Universitad de Bueanos Aires, Argentina Department of Soil Science and Agricultural Chemistry, University of Santiago, Santiago de Compostela, Spain Department of Land Resource Science, University of Guelph, Guelph, Ontario, Canada

xviii S. A. Norton H.P. Ritzema W. Shotyk P. J. Silk P. Steinmann S. Van Der Kaars J. Warburton

List of Reviewers Department of Geological Sciences, Bryand Global Sciences Center, University of Maine, Orono, ME, USA Wageningen University and Research Center, Wageningen, The Netherlands Institute of Environmental Geochemistry, University of Heidelberg, Heidelberg, Germany Canadian Forest Service, Atlantic Forestry Centre, Fredericton, New Brunswick, Canada Institut de Ge´ologie, Universite´ de Neuchaˆtel, Neuchaˆtel, Switzerland Geography & Environmental Science, Monash University, Australia Department of Geography, Durham DH1 3LE, UK

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

1

Chapter 1

Peatlands: a concise guide to the volume I.P. Martini, A. Martı´ nez Cortizas and W. Chesworth

In Chapter 1, the editors introduce the subject of peatlands by providing brief guides to the major points covered in each of the subsequent chapters. The views of the individual authors are summarized and followed by short editorial comments (italicized). Many systems of peatland classiﬁcation exist, with landform used as the basis for some, whereas others, for example, are based on types of vegetation, and yet others on chemical properties. In this volume, for the boreal and subarctic peatlands, we have adopted a general system that most people can agree with. A full explanation will be found in Chapter 2, and references to it occur throughout the book. In addition, the terms are deﬁned in the glossary at the end of the book. The following brief synopsis is intended to ease the reader into the subject. We use the terms ‘peatland’ and ‘mire’ synonymously, although the later normally refers to active peat-growing systems. All peatlands are wetlands. ‘Wetlands’ is a general term used to describe areas with the water table at or near the surface. Wetlands acquire the status of peatlands when at least 30 cm of peat has accumulated (according to the Canadian classiﬁcation) or 40 cm (in Russian usage). Peatlands develop either through paludiﬁcation (plant colonization of poorly drained lands) or terrestrialization (inﬁlling of ponds and shallow lakes mainly by plant remains), wherever the conditions of plant-matter accumulation exceed that of oxidation and consumption by organisms. In temperate and tropical areas, the preservation of organic matter is possible under wet conditions due to frequent rainfall; in coldclimate (from subarctic to boreal) areas both precipitation and frozen winter conditions favor preservation of part of the plant residue. Although active plant growth occurs in arctic regions, active peatlands (mires) may not develop extensively; peatlands currently present at high latitudes are frozen, and may have formed during more suitable climatic conditions of earlier times. Similarly, boreal ‘forests’ may be very productive, but they are not wetlands; they do not develop the necessary peat thickness and hence they are not considered peatlands. As indicated in Table 2.3 and in Figure 3.4, the basic classes of peatlands are bogs (obrotrophic peatlands that are acid and receive nutrients exclusively from the atmosphere – their development is tied to that of Sphagnum that allows maintenance of a perched water table); fens ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09001-8

2

I.P. Martini, A. Martı´nez Cortizas, W. Chesworth

(minerotrophic peatlands that are alkaline to slightly acidic and receive nutrients from groundwater as well as the atmosphere); swamps (minerotrophic, forested peatlands with about 25% of the area covered by trees); and marshes (minerotrophic wetlands with variable water cover and depth to water table – these commonly reach peatland status in cold-temperate and warmer climates, rarely in boreal and subarctic conditions where vegetation is also consumed by migratory fauna). Bogs, fens, swamps, and marshes are best considered as end members in a continuous peatland spectrum. Table 2.5 reports examples of other peatland terms used, based on various characteristics. Tropical peatlands stand apart from the rest with regard to the terminology used in their classiﬁcation, although the ombrotrophic and minerotrophic concepts are retained; relevant details are to be found in Chapter 7. The upper Quaternary dates in this book are reported in calendar years BC (before Christ) and AD (anno Domini), or in non-calibrated radiocarbon years indicated as 14 C yr BP (before present; that is, by convention, before 1950 AD), or calibrated radiocarbon years indicated as cal yr BP.

A.

Peatland basin analysis: evolution and structure

In Chapter 2, Tarnocai and Stolbovoy determine the extent of development and classiﬁcation of northern peatlands with particular attention to those of Canada and Western Siberia. They also offer an analysis of the effect that the ongoing episode of global warming has or could have on them. The oldest basal peats of the vast northernmost subarctic to boreal peatlands range in age from 8–10,000 yrs BP in Russia to 6–8000 years in Canada. Peatlands of the boreal and subarctic zones of Canada have been estimated to contain approximately 143 Gt (143 billion tons) of organic carbon, or 97% carbon occurring in all Canadian peatlands. Approximately 40% boreal and subarctic peatlands in Canada are perennially frozen. The Russian peatlands contain about 170 Gt (170 billion tons) of organic carbon. About twothirds of the Russian peatlands are perennially frozen. As the warming trend continues, the frozen peatlands are expected to thaw and increased oxidation should lead to an increase in the release of greenhouse gases (speciﬁcally methane and CO2) to the atmosphere from the present peatlands. However, the impact on the balance of greenhouse gases is problematical, in that the global increase in temperature will be accompanied by a general shift of ecozones to higher latitudes where carbon storage may take place in new peatlands. The determining factor will be the differential between the rate of decomposition of the present peat and the rate of accumulation of new organic matter in areas further north. Information available for northern peatlands has been used to estimate rates of change arising from variation in climatic conditions. Further research is needed, however, if reliable predictions are to be made that can serve as the basis for the implementation of peatland management policies on a global scale. In Chapter 3, Martini reports on work carried out in the latter part of the twentieth century on the Hudson Bay Lowland, Canada: the second largest (about 325,000 km2) unconﬁned peatland in the world after that of Western Siberia. During the last 50 years or so, the organic basins of the area have been partly studied,

Peatlands: a concise guide to the volume

3

together with their regional geological and geomorphologic settings, and a preliminary assessment has been made of the economic resources of the area. The Hudson Bay Lowland is a paludiﬁed coastal plain derived from an early postglacial sea. The land is still emerging at about 1 m/100 yr; it becomes progressively paludiﬁed, and new permafrost develops to the north. Extensive fens have developed in the coastal zone (40–50 km inshore from the present coastline) alternating with bogs on interﬂuves of slightly higher lands. Migratory birds and the sparse human population extensively use this relatively well-studied area. The vast inland bogs and fens are poorly known although a few, rapid, regional surveys have been made reporting on the plant species present there. A single study of greenhouse-gas emission from these peatlands during a single summer, revealed a much lower-than-expected value due, in part, to long, dry periods. Other studies of soils, local seasonal variation, and analysis of the peat as precursors of ancient cold-climate coals are of interest but limited in scope. Signiﬁcant reserves of diamonds have been discovered recently in the bedrock underlying part of these peatlands, and this may justify the establishment of a mine in the near future. The Hudson Bay Lowland can be considered the boreal – subarctic equivalent of the tropical forest of South America. The vastness of the territory and difficulty of access require extensive use of a helicopter, making study expensive. Consequently only the coastal area, which has continental significance for migratory species, has received any continued attention. Development associated with diamond extraction will open some areas to a greater human impact. Much must be learnt on the dynamics of these peatlands before such developments complicate the situation. Use of satellite data is particularly well suited to the task, provided it is accompanied by well-planned, longrange ground-truth studies. Monitoring the rate of expansion of newly formed peatlands and the disruption of mature ones is of particular importance. In Chapter 4, Pontevedra-Pombal et al. present the peatlands of the mountains of NW Spain, which have developed in a high-precipitation region. Basal peats of these mires are as old as 17,400 14C yr BP, indicating the onset of paludiﬁcation soon after the Last Glacial Maximum (20,000–18,000 14C yr BP). The main growth of peatlands occurred from about 11,000 to 2000 14C yr BP. Fens and bogs have developed, the latter being locally promoted by the granitic substrate. The bogs are good archives of environmental changes. The accumulation of Pb, for example, indicates that pollution may have started as early as 3000 years ago, associated with the importation of Spanish bronzes by the Phoenicians, but it became pervasive during the Iron Age (ca. 800 BC in this region) peaking during the Roman period when extensive mining occurred. In concert with surface deposits worldwide, the accumulation of Pb in the top peat layers is associated with the modern industrial period. Mercury shows similar trends, but is greatly affected by climatic conditions with higher accumulations of low thermal-stability Hg during cold periods and lower accumulations of higher thermal stability of Hg. It can thus be used as a paleothermometer for the Holocene. Despite the small area they cover (some 10,000 Ha) these peatlands are located at the southernmost limit for ombrogenous mires in Europe, so they represent a kind of bioclimatic frontier for bog development. In this sense they offer a great potential for new research since they represent a possible end term for comparison with peatlands from more northern locations. They also provide a continuous record of agricultural and industrial activity from the fourth millennium BC to the present.

4

I.P. Martini, A. Martı´nez Cortizas, W. Chesworth

Chapters 5 and 6 can be conveniently dealt with together. In Chapter 5, Coronato et al. provide information on the geomorphic units where various types of peatlands have formed in the southern part of Tierra del Fuego, Argentina. There is a close connection between development of peatlands and the glacial history of the area. The peatlands are mostly conﬁned to glacial and periglacial land features. Fens started developing from about 12,000 to 9000 14C yr BP upon deglaciation and were followed by the formation of Sphagnum bogs. The latter developed particularly from 9000 to about 3000 14C yr BP when humidity increased in the region. The geomorphic characteristics of the area have greatly inﬂuenced the thickness, type and extension of the various peatlands. In Chapter 6, Rabassa et al. provide a review of the major Quaternary paleoclimatic and paeleonviromental record obtained of the cold-temperate peatlands of Tierra del Fuego. They use the age of the basal peats to estimate the deglaciation stages of the various areas. The older peats (15,000 14C yr BP) occur to the east and become progressively younger to the west, reaching ages of 12,000–13,000 14C yr BP in the Ushuaia area, and less than 10,000 14C yr BP in the Andean valleys. Information on climatic change during the Holocene is provided by palynological proﬁles of various peatlands. Of particular interest is the pollen contribution to these proﬁles by the Nothofagus forest, typical of the area, which replaced earlier steppe vegetation. The analysis of peatlands and related features also shows that human occupation may have started as early as about 14,500 14C yr BP, and was certainly in full swing by 11,000 14C yr BP. A close association of the peatlands with geomorphic features is reported in both chapters, and allows the use of information obtained from peat profiles in the relatively detailed reconstruction of the changing environments of southernmost South America. These peats still exist under almost pristine conditions, a fact that may prove very useful in establishing reference levels against which to compare peatlands of other regions of the world may have been modified both naturally and by human activities. The detailed analysis of the landform – peatland relationship also provides a good initial estimate of available peat reserves in the region. In Chapter 7, Page et al. provide information on lowland peatlands of tropical Southeast Asia, including anthropologic impacts. Tropical peatlands constitute about 12% (about 40 million Ha) of the Earth’s total. They support a large diversity of plants and animals and are a signiﬁcant element in determining the carbon gas balance of the atmosphere. Yet these peatlands are under increasing pressure from human activities. The chapter reviews numerous studies and serves as a comparison and contrast to the mid-latitude mires dealt within the rest of the book. Of interest for the purposes of comparison with localities at higher latitudes, is the fact that most coastal basal peats worldwide range in age between 4000 and 5000 years BP. This reﬂects the time at which rising sea levels stabilized, and is in marked contrast to the age of basal peats inland that in the tropics are as old as 29,000 14C yr BP. Noteworthy in this chapter is a difference in classification and terminology used in the tropics compared to temperate and boreal regions. For example the term ‘ombrogenous peat swamp’ is used in the tropics whereas in boreal to subarctic settings ‘swamp’ is considered to be a heavily treed (425% tree cover) minerotrophic peatland. It is also worth noting that a kind of convergence takes place wherein tropical peatlands reach

Peatlands: a concise guide to the volume

5

similar states to those of other parts of the globe, in spite of the fact that the dominant plant species are different, as are the determining physical conditions. For example, by contrast with boreal to subarctic settings Sphagnum plays little or no role in the genesis of ombrogenous peat in the tropics. The reason is that low temperatures under relatively dry conditions allow the preservation of organic matter at high latitudes, whereas it is recurrent, frequent, high precipitation falling on vast, almost flat landscapes that bring this about in the lowland tropics.

B.

Selected characteristics of peat and peatland environments

In Chapter 8, Chesworth et al. develop a series of Pourbaix (Eh–pH) diagrams for the land-surface environment of the Earth. The chemical justiﬁcation arises from the fact that the geochemical evolution of Earth-surface deposits proceeds mainly through acid-base and redox reactions. The authors use their diagrams as a framework within which to deﬁne the general features of common environments of the weathering zone. The basic diagram allows the similarities and contrasts between ombrotrophic and minerotrophic mires and the common soils to be clearly displayed. Organic-rich deposits tend to be acid, at least in their superﬁcial parts, and bogs commonly retain this acidity to depth, unless the mineral load from the atmosphere is particularly high and reactive (calcite-rich for example). Conditions in minerotrophic mires appear to be controlled by ferrous–ferric equilibria at the lower limit of the pH range. This is a characteristic they share with mineral soils. The upper limit of the minerotrophic ﬁeld however, will depend on the nature and reactivity of extraneous additions of lithological materials, and/or the pH of invasive groundwater. The effective upper limit is marked by the boundary of the calcite predominance area. Histosols also lie within this extended ﬁeld of mires, and may be considered a kind of bridge between peatlands and mineral soils. Eh– pH diagrams are useful in defining the ambient conditions of natural environments, including peatlands. They provide information concerning the stability of inorganic components and by extension they suggest that the breakdown of mineralogical additions to peatlands may lead to the mobilization of elements assumed to be immobile for purposes of using bogs as archives of atmospheric change. This suggestion needs to be balanced, however, by consideration of the findings of Le Roux and Shotyk in the next chapter. In Chapter 9, LeRoux and Shotyk discuss the fate of inorganic solids, entering bogs. The principal materials added are clay minerals, quartz, feldspar and calcite, and the environment that they enter is overwhelmingly organic in nature. Laboratory experiments performed under the low pH conditions that would be encountered by the foregoing minerals as they become incorporated into the surface of a bog, and the low PO2 conditions characteristic of organic-rich environments that are water saturated at depth, indicate that many of them break down. In contrast, in the ﬁeld environment, mineral reactivity and breakdown are considerably diminished, the probable reason being that the simple organic compounds favored in laboratory studies are much more aggressive acids than the complex organic materials encountered in the real world of bogs. In addition, deeper in a given bog, acidity may

6

I.P. Martini, A. Martı´nez Cortizas, W. Chesworth

decrease and in fact reach neutrality, at least in part because of dissolution reactions. Mineral reactivity would then in most cases become insigniﬁcant. The authors emphasize the considerable importance of modern geochemical analytical techniques and instruments in deciphering the possibilities. Le Roux and Shotyk provide data with a bearing on the question of the reliability of inferences concerning environmental history drawn from elemental chemistry as revealed by the peat archive. If a given element is mobilized after deposition, use of that element in atmospheric reconstructions becomes problematical (Chapter 8). The authors provide reasons for doubting that this is a significant problem, though the question remains controversial and merits further study. In Chapter 10, Buurman et al. start with a brief review of studies linking the plant community in bogs, with their partially decomposed organic remains. To realize the potential of the organic chemistry of peat in environmental reconstructions, the authors set out to clarify the subject at the molecular level using pyrolysis GC/MS. A peat core from NW Spain is used, and extractable (more humiﬁed) and nonextractable (residual) fractions are made using NaOH extractions. Residues are characterized by a high abundance of aliphatics with chain lengths n417, all methyl ketones, fatty acids with n416, and lipids; whereas extracts have high lignin content, N-containing compounds, phenols, short-chain alkanes and alkenes (n ¼ 10–15), and C14 fatty acids. Polysaccharides are common in both fractions. There is a decay continuum of aliphatics between extracts and residues. Variations in the ﬁne details of the organic chemistry are interpreted as reﬂecting aerobic versus anaerobic decay and different plant inputs. A detailed study of lignin and polysaccharides, together with research on variations of superﬁcial bog wetness through time, suggest a link to climate change: peat formed during dry climatic periods has more decayed organic matter than peat formed during wet climatic periods. The authors recognize that more work is required to track further chemical changes in vegetation and its decay products, in order to check the utility and specificity of the various biomarkers that have been proposed in peat in the past. Obviously this is a fruitful area of research, and their opinion is that the initial step should be to establish correlations between the molecular chemistry of peat and the nature of the plant population. In Chapter 11, Franze´n deals with two major topics. (1) Using data principally from Southern Sweden, but supported also by observations in Norway, Ireland and Siberia, he advances the hypothesis that peats in the acrotelm of bogs are currently experiencing increasing decomposition. Among the evidence assembled are the widespread subsidence of the ombrotrophic peatland surfaces and a general loss of peat-forming vegetation. In addition, and recognizing that the data are sparse, measurements of CH4 and CO2 gas-ﬂuxes may be interpreted to suggest a net loss of carbon. In other words, bogs may be in the process of switching from being net carbon sinks to net carbon sources. If so, decomposing peatlands would be expected to have a signiﬁcant, reinforcing impact on current global temperature changes. As a possible reason for the putative decomposition of peat, Franze´n suggests that there has been a change in nutrient supply, which he believes to be related to anthropogenic activities. (2) He also analyzes the type of mineral matter (spherules) blown onto the bogs and tries to distinguish their source whether natural or anthropogenic,

Peatlands: a concise guide to the volume

7

for example marine aerosols, volcanic dust, agricultural soils, and extraterrestrial particles (micro-meteorites and micro-spherules). He presents methods for the morphological recognition of these by means of the scanning electron microscope and the distinction between ombrotrophic and minerotrophic peats through the relative concentrations of the rare earth elements in peat ash. (1) The decomposition of peat in northern peatlands represents the possibility of a significant addition of so-called greenhouse gases to the atmosphere, thereby reinforcing the global warming trend that can be expected to act as a positive feedback loop leading to the decomposition of ever more peat. Franze´n recognizes the possibility, but also recognizes that the evidence as yet, is equivocal, and that more observations in more localities across the globe are required (Chapter 2). (2) The recognition of type and amount of detrital particles (spherules) in bogs may prove to be a significant tool to reconstruct past environmental conditions and progressive changes, as it has been done in deep-sea sediments and in polar ice. A remaining problem for future research will be to study particle stabilities, particularly under acid and/or reducing conditions (Chapter 8). In bogs, particles that are present in the uppermost part of the peat profile, are absent or depleted in deeper layers, thereby reducing their usefulness to the relatively distant past. In Chapter 12, Bragazza demonstrates that one of the more obvious ways that human beings are inﬂuencing the chemistry of the planet is in terms of anthropogenic additions to the nitrogen cycle. In effect we are fertilizing the whole planet, in some cases to the extent that eutrophication develops. The unintended consequence of this forced fertilization may be to convert peatlands into net sources rather than sinks of carbon. The critical questions needing answers are: what is the feedback between increasing concentrations of N and CO2 and the decomposition rates of organic matter; how will future warming and increasing eutrophication affect ﬂuxes of CH4 and CO2; and what resilience does a bog have, should the anthropogenic N deposition cease. Answers to these questions are needed to elucidate and quantify the relationships between peatlands, human activities, and the overall picture of global change. We currently add N to the soil at a rate some one-and-a-half times faster than occurs as a result of natural soil processes. This is a massive intervention into the geochemical cycle, and the effects on the planet are seen all the way from the local contamination of rural ground and surface waters by excessive use of N-based fertilizers on farmers’ fields, to the development of hypoxia in marine environments at the mouths of major river systems such as the Mississippi and the Rhine. Bragazza’s contribution is a timely reminder that our disruption of the nitrogen cycle is bound to impact upon the carbon cycle, and that a feedback loop between the two exists. Distinguishing the effect of N additions, from the effect of increasing temperature as described by Franze´n in Chapter 11, will require a much more extensive database than we have at present. In Chapter 13, Gilbert and Mitchell synthesize what is known about some aspects of microbial ecology in peatlands with special focus on protists and micro-metazoa, and the value of these organisms, particularly the testate amoebae, in studies of present and past perturbations of peatlands. They show that to obtain reliable data much care needs to be taken in sampling and treating the material for analysis. The authors then report on the diversity, abundance, and biomass of the various groups

8

I.P. Martini, A. Martı´nez Cortizas, W. Chesworth

(heterotrophic and photosynthetic bacteria, protozoa, microalgae, and microscopic animals), and the vital functional role, they play mainly by microbial photosynthesis and by the recycling of nutrient through decomposition of organic matter. One group, the cyanobacteria, contributes nitrogen to peatlands by ﬁxing it from the atmosphere. Microorganisms can also provide vital information on the state of natural and impacted environments, in particular testate amoebae are also precious paleoecological indicators that can help in reconstructing past environmental conditions of growing peatlands. This chapter provides an extremely informative synthesis that should be understandable also by the general earth-science reader. It also provides the results of some interesting experiments on the effects of nutrient manipulation and elevated atmospheric CO2 concentration on microbial communities. The main achievements to date on this subject matter are the assessments of the diversity of microbial life in peatlands and their response to natural environmental gradients (moisture, pH) and temporal variations (seasonality). The most promising possible future research is on the response of microorganisms to environmental change and the implication of these responses on ecosystem functioning, including carbon sequestration by peatlands. In Chapter 14, Holden provides an overview on peat hydrology and focuses on the spatial heterogeneity of surface and subsurface ﬂows in blanket bogs. He proposes going beyond the acrotelm–catotelm model and analyzes the ecohydrological functioning of the peatlands also affected by internal macropores and pipes. He reports that up to 51–78% in fens and 35% in blanket peat of the groundwater ﬂux can occur through macropores. Flow through pipes has been found to be signiﬁcant as well, and piping can be exacerbated by droughts and artiﬁcial peat drainage. Under certain circumstances, nutrients can be provided to the mainly ombrotrophic blanketpeat systems through deep-seated macropore and pipe ﬂows, not just by rapid transfer through the overlying acrotelm. Water is the lifeblood of peatlands and affects their wellbeing, growth, disruption, redox status, nutrient availability, and emission of gasses. The hydrology of peatlands is complex as groundwater flows change dramatically in time and space depending on climatic conditions, topography and internal structure of the peat profile. Holden dares to break through the quasi-static, still valid but overemphasized concepts of the acrotelm-catotelm model, and suggests other approaches that may explain transition states and apparent anomalies in ombrotrophic peatland systems. In Chapter 15, Glaser et al. investigate the interactions between groundwater and peatlands in two of the largest northern peatlands of North America: Minnesota (USA) and Hudson Bay Lowland (Canada). They suggest that groundwater ﬂow affected by climate and geologic setting largely controls the distribution of bogs and fens in unconﬁned peatlands. Bogs, for example, are located over geologic sites where water-table mounds drive surface waters downward into the deeper part of the peat proﬁle or mineral substrate. Fens, in contrast are located over sites where dissolved solutes are transported upward from the underling deposits to the peat surface by regional discharge systems or along pathways for lateral ﬂow. In northwestern Minnesota the climatic controls on groundwater control and peatland development are particularly apparent. In this relatively dry climatic region, periodic droughts perturb groundwater ﬂow systems on raised bogs producing ﬂow reversals that inﬂuence the

Peatlands: a concise guide to the volume

9

subsurface transport of labile carbon and inorganic nutrients essential for anaerobic microbes. Gas bubbles produced by anaerobic decay in turn alter the local hydraulic properties of the peat creating an important feedback involving climate, hydrology, and carbon cycling in these large peatlands. In the Hudson Bay Lowland the geomorphic controls on groundwater are locally most apparent where raised bogs are forming within the interﬂuves between headwardly eroding streams or where rows of spring fen channels line the margins of a mounded plateau. The bogs most likely form in response to the creation of water-table mounds within the interﬂuves, whereas the spring fens developed over a regional seepage face for groundwater discharge. The authors conclude that the striking peat landform patterns in large peat basins are faithful indicators of the groundwater ﬂow systems and can be used to calibrate and verify groundwater ﬂow models. The peat distribution in the large, northern peatlands of North America is complex. The Hudson Bay Lowland (HBL) in particular has relatively young (less than 5000 years old), relatively thin (generally less than 5 m) peat accumulations. Although the interior parts of HBL is continuously covered by vegetation for hundreds of square miles, and it appears blanketed by peat, the mineral substrate and topography (bedrock and glacial features as well as coastal beach ridges formed during emersion of the land) strongly affect the peat distribution and thickness. These conditions are ideal for the development of the groundwater circulation cells leading to bog– fen associations analyzed in this chapter. In Chapter 16, Dykes and Kirk recognize and analyze several types of peat failure on unstable peatland hillslopes and in lowland-peat deposits. Overloading in situ peat with excavation spoil and undercutting natural slopes, for example, serve as anthropogenic triggers for failure. Natural triggers include heavy rainfalls with rapid transfer of rainwater to lower parts of a proﬁle through cracks, pipes, and macropores. The consequent increase in pore-water pressure leads to development of sliding surfaces and/or ﬂuidized basal peat. Within the peat deposits, failures occur at geotechnical discontinuities that are due to differences in peat composition, ﬁber content, humiﬁcation, and other characteristics. Speciﬁc examples of peat failures are described and analyzed in the context of the limited knowledge of peat mass properties and behavior in response to applied stresses. The authors of this chapter indicate that like any other Earth-surface deposit, peat is subject to deformation and collapse when subjected to particular combinations of intrinsic and external factors. Accordingly, conventional slope failure analysis can be applied, though this is a difficult medium from which to obtain the requisite accurate data. Failures of peat deposits and peat-covered slopes are most common, or more frequently observed, in Ireland and the UK, but they create damage and pose risk in many localities elsewhere, possibly increasingly so as a consequence of climate change effects.

C.

Peatlands as multi-signal archives of environmental changes

In Chapter 17, Mighall et al. review the use of the peatland archive in deciphering the impact of human activities on the landscape. Multi-proxy studies of the

10

I.P. Martini, A. Martı´nez Cortizas, W. Chesworth

high-resolution record using both biological (pollen, microscopic charcoal) and geochemical (trace elements, isotopes, magnetic properties) data, provide information on the history of mining and metallurgy, the scale of pollution, and vegetation changes. This complements the ﬁeld data of the archeologist, which in turn illuminates the peat record. In addition, synchronous changes in vegetation (mainly the forests) give insight into the role of mining and metallurgy on the creation of cultural landscapes. For example, temporary and small-scale changes in vegetation near metalworking sites could be deﬁnitively distinguished from the effects of low intensity agriculture by multi-proxy investigations combining geochemical research, archeological and palynological data. Archeological experiments using prehistoric methodologies provide a further means of checking the peat record. Until recently it was commonly accepted that mining and metallurgy did not cause widespread pollution before the industrial revolution. As Mighall et al. demonstrate this myth has now been refuted with evidence from peat archives (as well as from ice cores and lake sediments). In the British Isles there is a long history of mining and metallurgy extending back at least four millennia. This chapter also exemplifies the fact that multiproxy, multi-interdisciplinary research is a powerful approach in paleoenvironmental reconstructions, not the least because it creates positive feedbacks among the disciplines involved. In Chapter 18, Bindler and Klaminder emphasize that the peat archive is not the only one available to us in deciphering the past environmental history of the Earth. Lake sediments and ice cores provide comparable information, and in the case of lead, information from all three archives has been obtained. In fact, among the metals of concern in the biosphere, Pb is one of the most researched, and the connection between Pb contamination and soil geochemistry is relatively well known. The authors believe that our experience with lead should serve as a model for similar researches to elucidate the biogeochemistry of other metals of toxicological interest (cadmium and mercury for example). Bindler and Klaminder provide a roadmap of the way to go in deciphering the movement of metal contaminants in the biosphere. A concordance among environmental archives will have to be worked out in order to increase the reliability of our knowledge, but it will require an immense collective effort and a considerable capital investment. In Chapter 19, Biester et al. present the ﬁrst general treatment of halogens in mires. Halogens have received much less attention than other trace elements in peat, mainly because until recently they have been considered as conservative, non-reactive elements. This chapter reviews the sources, concentrations, and net accumulation rates of halogens in peat and their release into pore waters. They indicate that the chemical cycle of halogens is much more complex than previously thought, and suggest that peatlands constitute signiﬁcant reservoirs of halogens on a global scale. Since the oceans are the main source of halogens, maritime mires have higher concentrations than continental ones, but in both, the inventories are dominated by organo-halogen species (80–90% of total halogen content). Similarly, for bromine and iodine in pore waters (though not for chlorine) organo-halogens dominate. The concentration, accumulation, and release of halogens from peat were found to be related to the degree of peat decomposition, whereas release of halogens from peatlands is related to dissolved organic carbon release.

Peatlands: a concise guide to the volume

11

Ultimately the fate of halogens in mires seems to be linked to climate change and, particularly to variations in superficial wetness. The lowering of the water table during dry climatic periods enhances peat decomposition and accumulation of halogens, thus high concentrations of halogens in peat cannot directly be interpreted as the result of enhanced atmospheric fluxes. Dry climates may also result in a lower export of organohalogens (some of which are furan and dioxine-like substances) to the drainage waters, whereas warm and wet conditions may increase the release of organo-halogens. This is an important finding that must be put into context with the predictions of the effects of human induced climate change on peatlands (Chapters 11, 12). In Chapter 20, Biester et al. present a synthesis of current knowledge (sources, species, concentrations, accumulation rates, and related processes) bearing on the geochemical cycle of the toxicologically important element mercury, in peatlands. Based on the assumption that mercury is immobile in peatland environments, peat records have been used to trace the evolution of atmospheric mercury pollution. This chapter indicates that the assumption is unsafe for reasons similar to those presented for halogens in Chapter 19. In fact, a good correlation was found between the accumulation of mercury and that of halogens and other organically bound elements (such as Se). Thus mercury concentrations are likely to be modiﬁed as the organic matter undergoes diagenesis. Consequently, the large disruption of the Hg cycle by anthropic activity, that has been suggested by studies of recent Hg accumulations in the upper parts of peatlands normalized in terms of values of Hg in the older parts, need to be treated with skepticism. Support for a much smaller disruption of the natural cycle is provided by the lake sediment archive as well as the direct atmospheric measurements of wet deposition. The findings presented in this chapter challenge the traditional use of peat records as reliable archives of atmospheric Hg fluxes. Peatlands are open systems sensitive to climate changes. The integration of research on the carbon budget and processes related to organic matter transformations is needed if reliable models for the chemical cycles of elements in peatlands are to be developed. In Chapter 21, Kylander et al. take an in depth approach to the analysis of lead distribution in peatlands. As the authors indicate, most of what is known about heavy metal deposition in peatlands comes from the many accumulated studies of lead. Like mercury, lead is an element of environmental concern but has the advantage that its isotopic composition allows an accurate identiﬁcation of provenance. The chapter provides the basis of an understanding of the Pb cycle from a geological perspective, including the geochemistry of its isotopes and their application. Two concepts are stated as pivotal for this purpose, the use of three-isotope plots and the normalization to 204Pb. The chapter also reviews the applicability of the calculation of enrichment factors for estimating the intensity of lead contamination. Based on the extensive datasets available from Switzerland and NW Spain, the authors discuss the evolution of lead atmospheric pollution in terms of four main periods. These are: the pre-anthropogenic (or pre-pollution) period (43000 14C yr BP), the ancient period (3000–1600 14C yr BP), the pre-industrial period (1600–200 BP), and the industrial period (ca. 1800–1970 AD). Whereas lead variations and sources during the pre-anthropogenic period are coupled to climate-related processes (dust transport, soil weathering), as soon as mining and metallurgy commenced the lead

12

I.P. Martini, A. Martı´nez Cortizas, W. Chesworth

cycle is perturbed. Although effects are detectable since the Bronze and Iron Ages (Chapter 17), the Roman period represents the pre-industrial climax in atmospheric lead pollution. The pre-industrial period is characterized by a close link between lead pollution and the intricacies of socio-economic history, whereas during the industrial period, a greater variability in lead provenance results from a greater proportion of few sources, such as coal combustion, gasoline, and waste incineration. The last 30 years indicate a decrease in lead pollution, while keeping the heterogeneity in source. Since the general chronology of lead atmospheric pollution seems now to be well established, further investigations in more areas may contribute to clarification of the history of human activities, in the way suggested by Mighall et al. in Chapter 17. But the authors of this chapter have also identified other aspects of importance. In particular, there is a need to: (1) develop consistent methods to estimate the intensity of lead pollution from different peat records if we are to interpret the temporal and spatial scale of human perturbations; (2) shift the focus from single reference elements to a proper process-oriented multivariate approach (multi-proxy); (3) identify new lead sources; and (4) expand the geographic cover of peat records.

D.

Direct human impact on peatlands

In Chapter 22, Holden et al. analyze changes due to artiﬁcial drainage or other human activities. Such changes (the rapid emission of CO2 for example) are global and not just local in impact. Some of these changes, such as drainage to create farmland, are well known and examined elsewhere in this book (Camporese et al., 2006 – this book, Ch. 23). Many other examples exist that show how drainage leads to changes in the hydrological response of a catchment, changes in water quality, and other severe problems. This chapter reports on the various impacts of peatland drainage and how to ameliorate the hydrological problems that are generated. The properties of the remaining, drained peatlands are drastically changed as peat desiccates at the surface, shrinks throughout, and macropores and pipes form. Increased aeration increases microbial activity, peat oxidation, and mineralization of nutrient elements among other chemical changes. The waters removed from peatlands have also signiﬁcantly different properties from those of natural watercourses (much higher concentrations of SO4, Ca, Mg, and higher pH, for example). Full restoration is never possible, but under certain circumstances rehabilitization is advisable. This can be aided by taking account of the spatial and topographic context of the land management, rather than adopting blanket-management policies. Until recently, peatlands were considered a resource to be exploited. Peat was mined, forests were cut, and where forest could not regenerate fast enough, draining of the land was implemented to supposedly accelerate and increase tree growth. Many sites were rendered wastelands. Recently, the value of peatlands as a beneficial natural environment with a positive effect on atmospheric composition and water quality by filtering contaminants, has been recognized. In places peatlands are now protected, and, where possible and needed, programs of rehabilitation are implemented. This chapter provides us with an insight into what happens when peatlands are drained and how rehabilitation may be achieved. Much more work is needed on these important problems.

Peatlands: a concise guide to the volume

13

In Chapter 23, Camporese et al. examine subsidence in response to the artiﬁcial draining of wetlands in an area just south of Venice, Italy. The subsidence is basically related to oxidation of the marshy (mainly reeds) peat, enhanced by farming activities, and is for the most part irreversible. However, the peat surface experiences seasonal ﬂuctuations in ground elevation (mire breathing) associated with variation in climatic conditions and depth to the water table. Because of subsidence most of the study area now lies at an elevation lower than sea level. Ancient maps, air photographs, remote sensing, ground measurements, and observations on old constructions were used to estimate the amount (about 1.5–2 m in 70 years) and rate (1.5–2 cm/yr during the last two decades) of subsidence. The gaseous emission, of CO2 in particular, was also measured. All results were utilized in determining the accuracy of subsidence-predictive models. Subsidence of peaty and other organic-rich terrains as a result of drainage and agriculture is a common experience throughout the world. In the case near Venice, large portions of the original marshes in the area studied have been drained in the last century. Long-range historical information and modern techniques have allowed an accurate estimate of subsidence and CO2 emission due to oxidation of peat that has occurred and a prediction for what will happen in future. The authors have provided a further understanding of the physical and biochemical processes involved.

This page intentionally left blank

15

A. Peatland basin analysis: evolution and structure Section A deals with the evolution and structure of selected peatland basins, with emphasis on examples from medium to high latitude. The most recent work done on the vast boreal to subarctic peatlands of North America and Eurasia is summarized in two chapters (Chapters 2 and 3) and a ﬁrst approximation of the potential carbon release is made, for the eventuality that present global-climatic warming continues. Setting, structure, geochemistry and usefulness are examined for peatlands from less well-known localities [mires of the mountainous areas of NW Spain (Chapter 4) and Tierra del Fuego, Argentina (Chapters 5 and 6)] and are closely tied to geomorphologic features. Finally, the characteristics and plight of a tropical example is included from South East Asia (Chapter 7).

This page intentionally left blank

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

17

Chapter 2

Northern Peatlands: their characteristics, development and sensitivity to climate change C. Tarnocai and V. Stolbovoy

Introduction Globally, peatlands cover about 5 million square kilometers (IPCC, 2000) and are a unique ecosystem, containing one-third of the global soil carbon and 10% of the global freshwater (Bartalev et al., 2004a, b). In North America, peatlands are organic wetlands having 440 cm peat; mineral wetlands are those having o40 cm peat. In Russia, peatlands are organic wetlands having 430 cm peat; paludiﬁed lands are those having o30 cm peat. Peatlands are widespread in the landscape of the northern circumpolar area and are associated with the cool to very cool climates occurring in the Arctic, Subarctic, Boreal and northern Temperate regions of North America and the equivalent regions of Eurasia, the Polar Desert and Tundra, Forest–Tundra, Taiga, and Temperate Forest. The southern limits of these regions coincide approximately with 401N latitude in North America and 501N latitude in Eurasia. The physiography of this region varies from rugged mountains to well-worn-down hills and lowlands. Permafrost is common throughout this region, except for the northern Temperate and Temperate Forest regions, although not all northern peatlands are perennially frozen. Numerous publications concerning the various aspects of northern peatlands are available in North America and Eurasia. Zoltai and Pollett (1983), National Wetlands Working Group (1988), and Glooschenko et al. (1993) have provided detailed descriptions of the types of northern peatlands and wetlands in Canada and Greenland. Regional and site-speciﬁc peatland characterizations are found in Sjo¨rs (1961, 1963), Zoltai and Tarnocai (1971), Zoltai (1972), Vitt et al. (1975), Zoltai and Tarnocai (1975), Brown (1980), Seppa¨la¨ (1980), Washburn (1983), Vitt and Bayley (1984), Zoltai and Johnson (1985), and Halsey et al. (1997). The classiﬁcation of Canadian wetlands and peatlands is found in National Wetlands Working Group (1988, 1997), the classiﬁcation of Finnish peatlands is found in Ruuhija¨rvi (1960) and Eurola (1962), and the classiﬁcation of Russian peatlands is found in Tyuremnov (1976). Studies of the paleoecology of peatlands are given in Zoltai and Vitt (1990), ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09002-X

18

C. Tarnocai, V. Stolbovoy

Zoltai (1994), and Vitt et al. (2000). Carbon- and climate-change-related studies are included in Gorham (1988), Halsey and Vitt (1995), Kettles et al. (1997), and Yu et al. (2001). Peatland inventories, databases, and maps are included in Tarnocai et al. (2003) and Tarnocai et al. (2004a–e).

Area and extent The extent of northern circumpolar peatlands was determined by using the Northern and Mid Latitudes Soil Database (NMLSD) (Cryosol Working Group, 2003). In this database the histosol (organic soils) group, whose deﬁnition is based on approximately the same criteria as the North American deﬁnition of peatlands, was used to estimate the area of peatlands north of 401N latitude in North America and north of 501N latitude in Eurasia. Although this estimate includes organic soils developed on both peat and upland folic materials, the folic materials cover only very limited areas and are found mainly in oceanic peatland regions. The distribution of frozen and

Figure 2.1. Distribution of frozen and unfrozen peatlands in the northern circumpolar area.

Northern Peatlands: their characteristics, development and sensitivity

19

unfrozen peatlands in the circumpolar area, based on data in the NMLSD, is shown in Figure 2.1. These peatlands cover approximately 4543 103 km2, of which 66% are found in Eurasia and 34% in North America. Approximately 60% of these peatlands (2718 103 km2) are perennially frozen, with approximately 2152 103 km2 occurring in Eurasia and 565 103 km2 in North America. Russia has the largest area of peatlands of all the northern circumpolar countries with the world’s largest peatland being the West Siberian mire massif and the largest in Europe the Polistovo-Lovatsky mire in northern Russia (Bartalev et al., 2004a, b). The latest estimate, derived from the digital soil database of Russia at a geographical scale of 1:5 million (Stolbovoi and McCallum, 2002), indicates that the area of soils with a peat depth of more than 30 cm is nearly 2210 103 km2. Approximately 28% occurs in the zone of seasonally frozen soils, nearly 30% in the zone of sporadic and discontinuous permafrost, and 42% in the zone of continuous permafrost. Peat with a depth of more than 50 cm tends to be dominant in the Northern and Middle Taiga zones, but is uncommon in the Tundra zone. This distribution illustrates that deeppeat formation is limited by a shallow permafrost table (Stolbovoi, 2002). Peatlands are also common in Byeloruss and Ukraine, where they occupy approximately 497 103 km2. In addition, recent information indicates that, in the European Union (25 countries in Europe), peatlands cover approximately 291 103 km2, of which nearly 55% are in Finland and Sweden (Montanarella et al., 2006). In North America, Canada has the largest area of peatlands (1142 103 km2, or 87% of those occurring in North America) and the largest peatland, the Hudson Bay Lowland. Almost all of these peatlands (97%) occur in the Boreal (64%) and Subarctic (33%) peatland regions, and 67% of them are bogs, 32% are fens, and the remainders (o1%) are swamps and marshes. Perennially frozen peatlands cover approximately 36% of the peatland area of Canada.

Peatland regions Peatlands occur in broad climatic zones. Katz (1948) recognized eight broad, climatically controlled peatland zones in Russia. Botch and Masing (1979a, b) delineated seven unique peatland zones with numerous subzones in Siberia. In Finland, three main divisions were recognized, each dominated by particular mire forms—palsa mires, aapa mires, and raised bogs (Ruuhija¨rvi, 1960; Eurola, 1962). Moore and Bellamy (1974) identiﬁed nine mire zones in Europe. In Canada, 20 peatland regions were identiﬁed by the National Wetlands Working Group (1988). These systems differ on the basis of delineating criteria, terminology, and scale. Because of these differences, it is difﬁcult to present a framework for delineating the circumpolar peatland regions. Therefore, examples are presented here for Canada, Russia, and Finland. Canada The peatland regions are deﬁned as areas where characteristic peatlands develop within the given climatic limits through the interaction of components (topography,

20

C. Tarnocai, V. Stolbovoy

hydrology, climate, and nutrient regime) of the peatland ecosystem (National Wetlands Working Group, 1988). The location and orientation of the peatland region boundaries are similar to the broad vegetation and ecological regions, indicating that they are also climate controlled. The distribution of the various peatland regions in Canada is shown in Figure 2.2 and described in Table 2.1.

Russia The spatial delineation of peatlands in Russia follows the tradition zonal–provincial boundaries established by Botch and Masing (1979a, b). They delineated seven mire, or peatland, zones in Siberia on the basis of peatland types and then subdivided these zones into subzones based on geographic areas. These peatland, or mire, zones were modiﬁed for use in the former U.S.S.R., resulting in seven mire zones and three province groupings (Table 2.2). The mire zones were delineated on the basis of peat-forming vegetation complexes, which are climate controlled, whereas the province groupings were delineated on the basis of the continental and oceanic inﬂuences on regional climate.

Figure 2.2. Peatland regions and subregions of Canada (after National Wetlands Working Group, 1988).

Northern Peatlands: their characteristics, development and sensitivity

21

Table 2.1. A brief description of the Peatland Regions of Canada (information from National Wetlands Working Group, 1988; Glooschenko et al., 1993). Peatland region

Climate

Common peatlands

Peat development

Arctic

Cold to cool short summers, very cold winters and low precipitation. Continuous permafrosta

Active 1–2 m peat development in the Low Arctic; very little or no peat development in the Mid and High Arctic regions

Subarctic

Cool summers and cold winters with low precipitation. Discontinuous, with some continuous, permafrosta

Boreal

Cold winters and moderately cool to warm summers. Low precipitation in the west, increasing eastward. Discontinuous to no permafrosta

Prairie

Cold winters and hot summers. Moderate to high annual precipitation Mild winters and warm summers. Moderate to high annual precipitation

Low-center polygon bogs, high-center polygon bogs (Fig. 2.3), peat mound bogs, and basin fens. Peatlands are common in the Low Arctic, less common in the Mid and High Arctic regions Polygonal peat plateau bogs (Fig. 2.4), peat plateaus bogs (Fig. 2.5), palsa bogs, peat mound bogs, basin bogs, ribbed fens, and horizontal fens Peat plateau bogs, palsa bogs (Fig. 2.6), ribbed fens (Fig. 2.7), and horizontal fens in the High Boreal; ﬂat bogs, basin bogs, and horizontal fens in the Mid and Low Boreal regions Mainly fens and some bogs in the Intermountain Prairie Swamps (Fig. 2.8), bogs and fens

Temperate

Oceanic

Cold winters in Atlantic and mild winters in Paciﬁc. Moderate to high annual precipitation.

Flat bogs, basin bogs, slope bogs (Fig. 2.9), horizontal fens, and swamps.

Active peat development; 1–3 m in bogs and 1–2 m in fens

Active peat development; 2–4 m in bogs and up to 2 m in fens

Active peat development; less than 1 m peat Active peat development; 4–5 m peat in swamps, 2–3 m in bogs, and 1 m in fens Active peat development; peat thickness is o2 m in the Atlantic and 2–4 m in the Paciﬁc.

C. Tarnocai, V. Stolbovoy

22 Table 2.1 (continued ) Peatland region

Climate

Common peatlands

Peat development

Mountain

Cool to cold winters and cool summers. Low annual precipitation.

Flat, basin and slope bogs, horizontal fens and swamps.

Active peat development; peat thickness is 1–3 m.

a

Heginbottom et al. (1995); van Everdingen (1998, revised May 2005).

The distribution of these zones and province groupings in the former U.S.S.R. are shown in Figure 2.10 and given, with subzones and provinces, in Table 2.2.

Peatland classiﬁcation Over the years numerous peatland classiﬁcations have been developed based on different aspects of the peatland. The earliest peatland classiﬁcations were based on the shape and height of the peatland. As knowledge increased, further subdivisions were made based on vegetation cover, surface patterns, and regional variations. Still other classiﬁcations delineate peatland units according to the water chemistry, hydrology, associated vegetation, or soils. This section gives examples of some of the classiﬁcations used for northern peatlands. Canada In Canada, peatlands are deﬁned as wetlands that are saturated with water and contain more than 40 cm peat. The classiﬁcation of these peatlands is hierarchical, with three levels (National Wetlands Working Group, 1988, 1997). At the highest level, the peatland class, peatlands are classiﬁed according to their genesis. Four classes of peatlands are recognized: bogs, fens, swamps, and marshes. At the peatland form level they are classiﬁed according to their surface morphology, surface pattern, morphology of the underlying mineral terrain, and hydrology. At the lowest level, the peatland type is classiﬁed according to the general physiognomy of the vegetation cover and soils. Brief descriptions of the peatland classes are given in Table 2.3. Finland The Finnish peatland types are subdivided into three main ecological gradients: (1) Moist–dry conditions related to water level (2) Nutrient status (3) Peatland margin effect

Northern Peatlands: their characteristics, development and sensitivity

23

Figure 2.3. High-center polygon bogs, Low Arctic Peatland Region, Mackenzie River Delta area, Northwest Territories, Canada.

Figure 2.4. Lichen-covered polygonal peat-plateau bogs, Subarctic Peatland Region, Horne Plateau, Northwest Territories, Canada.

24

C. Tarnocai, V. Stolbovoy

Figure 2.5. Open black-spruce- and lichen-covered peat plateau bogs, Subarctic Peatland Region, Mackenzie River Valley, Northwest Territories, Canada.

Figure 2.6. Palsa bog situated in a fen peatland, Boreal Peatland Region, Mackenzie River Valley, Northwest Territories, Canada.

Northern Peatlands: their characteristics, development and sensitivity

25

Figure 2.7. Ribbed fen with ridges covered by shrubs and stunted black spruce, Boreal Peatland Region, Fort Simpson area, Northwest Territories, Canada.

Figure 2.8. Swamp, Temperate Peatland Region, Ottawa, Ontario area, Canada.

26

C. Tarnocai, V. Stolbovoy

Figure 2.9. Slope bog with stunted pine, Oceanic Peatland Region, Queen Charlotte Island, British Columbia, Canada.

The Finnish peatland classiﬁcation is described in Table 2.4. Russia The widely used peatland classiﬁcation developed by Tyuremnov (1976) for the former U.S.S.R. is hierarchical, with four levels. Peat type, the highest taxon, is based on the composition of the peat-forming plants. Three peat types are deﬁned: lowmoor peat (eutrophic), transitional peat (mesotrophic), and high-moor peat (oligotrophic). Each type is divided into three peat subtypes, depending on the degree of wetness and the extent of the forest: Forest, Forest–swampy, and Swampy. Peat subtypes are further divided into peat groups, which are based on the living vegetation types that contribute to peat formation: woody, woody–herbaceous, and herbaceous–mossy. Lastly, peat groups are divided into 40 peat species based on the dominant peat-forming vegetation communities or plants such as pine, birch, alder, sedge, sphagnum, moss, horsetail, etc. (Table 2.5). This classiﬁcation has been widely accepted for peat in the Boreal zone of Russia. In the Arctic and eastern regions, however, there are some problems because of changes in the ecological ranges of species in the different zones (Botch and Masing, 1983).

Peat materials Peat materials associated with peatlands are separated according to their botanical composition. Thus, the name of the peat material indicates the most common plant

Northern Peatlands: their characteristics, development and sensitivity

27

Figure 2.10. Mire zones and provinces in the former U.S.S.R. (after Botch and Masing, 1979a, b). 1, Zone of polygon mires; 2, Zone of palsa mires; 3, Discontinuous zone of aapa mires; 4, Zone of raised string bogs; 5, Zone of pine bogs and fens; 6, Zone of reed and sedge fens; 7, Zone of fresh and salt-water marshes; 8, Continental provinces of Siberia and Far East; 9, Maritime provinces of the Far East; 10, High Mountain province and the Central Asian Mountain province.

material(s) associated with that speciﬁc peat. Some properties of the most common peat materials found in Canadian peatlands are given in Table 2.6. Table 2.7 provides the basic characteristics of the different peat types in Russia. Low-moor peat is characterized by a relatively high degree of decomposition, averaging 32%. For peat originating from woody remains, it can be as much as 45%. Low-moor peat is rich in ash elements (7.4%), and for woody peat the ash content can be as much as 18%. This peat has a pH of approximately 5.0. The mean caloriﬁc value of low-moor peat reaches 5542 kcal/kg. The characteristics of high-moor peat are much different from those of low-moor peat, being less decomposed, poor in ash elements (2.9%), and very low in acidity (pH ¼ 3.4). There is little difference in caloriﬁc value between low- and high-moor peats. The characteristics of transitional peat fall between those of the low- and high-moor peats.

Peatland development In northern regions, peatland development is generally initiated on lands associated with poorly drained soils. Wet meadows ringed with a shrubby swamp type of mineral wetland develop in such areas. As peat accumulates in these depressions, an open

Table 2.2.

Mire zones and provinces in the former U.S.S.R. (after Botch and Masing, 1979a, b). Province

Zone

Province

1. Polygon miresa

a,b 1.1. Arctic a,b 1.2. Northern Subarctic a,b 1.3. Southern Subarctic

7. Fresh and salt-water marshes

7.1. South Russian marsh 7.2. Central Asian salt-marsh

2. Palsa miresa

2.1. 2.2. 2.3. 2.4. 2.5.

8. Continental provinces of Siberia and Far Easta

8.1. 8.2. 8.3. 8.4. 8.5.

3. Aapa miresc

3.1. Kola 3.2. North Karelian 3.3. East European

4. Raised string bogsa,d

4.1. 4.2. 4.3. 4.4. 4.5.

5. Pine bogs and fensa

5.1. Pole’ye 5.2. East European pine bog a 5.3. West Siberian pine bog

6. Reed and sedge fensa

6.1. Central Russian fen a 6.2. West Siberian fen

b c d

White Sea Coast Northeast European South Karelian Baltic Coast East Baltic

Botch and Masing (1979a, b) Zones and subzones. Subzone, not province. Discontinuous zone. Botch and Masing (1979a, b) Eastern Siberia subzone is missing here.

Central Siberian Plateaua Northeast Siberian South Yakutiana Angaraa Dahurian

8.6. North Mongolian 8.7. Amur lowland

9. Maritime provinces of the Far East

9.1. 9.2. 9.3. 9.4. 9.5.

10. High Mountain province

10.1. The Carpathians 10.2. The humid province of the Caucasus 10.3. The arid province of the Caucasus 10.4. 10.5. 10.6. 10.7.

West Kamchatka East Kamchatka Sikhote–Alin North Sakhalin South Kuril

The Urals The Altaya The Sayan Mountainsa The Central Asian Mountain province

C. Tarnocai, V. Stolbovoy

a

North Kola Low Pechora North European West Siberiana Middle Siberiana

28

Zone

Northern Peatlands: their characteristics, development and sensitivity

29

Table 2.3. Peatland class descriptions used in the Peatlands of Canada Database (after National Wetlands Working Group, 1988). Peatland class

Descriptions

Bogs

Bogs have the water table at or near the surface. The bog surface, which may be raised or level with the surrounding terrain, is virtually unaffected by nutrient-rich ground waters from mineral soils (ombrotrophic). The dominant peat material is acidic, weakly to moderately decomposed, moss and woody peat Fens have the water table at or just below the surface. The waters originate from mineral soils and are nutrient-rich (minerotrophic). The dominant peat materials are slightly acidic to neutral, moderately to well decomposed, sedge and/or brown moss peat Swamps have standing water or water gently ﬂowing through pools and channels. The waters originate from mineral soils and are nutrient-rich (minerotrophic). The dominant peat material is strongly to slightly acidic, well-decomposed woody peat Marshes are periodically inundated by standing or slowly-moving, nutrient-rich water (minerotrophic). Surface water levels may ﬂuctuate seasonally, with declining levels exposing drawdown zones. The dominant peat material is slightly acidic to neutral, well-decomposed reeds and aquatic vegetation, with various amounts of mineral materials being present

Fens

Swamps

Marshes

fen develops, and with further peat accumulation the fen extends into the swamp fringes. As peat depth continues to increase, a peat mat gradually covers remnants of small ponds and a fen type of peatland covers the entire depression. At this stage the peat thickness generally exceeds 40 cm. If the fen has a slight slope, patterns of ridges and ﬂarks can develop on the peatland surface. Shrubs or stunted coniferous trees with fast-growing mosses can establish themselves where the peat accumulation is greatest, thus providing better drainage conditions. As a result of this process, small islands of treed fens may be established on the open fen, which becomes increasingly ombrotrophic and isolated from minerotrophic waters. If this process continues, mosses, primarily Sphagnum sp., are established and bog conditions develop since most of the water this peatland receives is from precipitation. Thus, ombrotrophic conditions prevail. At this stage, if permafrost is absent, the bog could develop into either ﬂat or raised bogs. In peatlands associated with a continental type of climate (continental peatlands), peat accumulation is generally uniform and a level surface is maintained (ﬂat bogs). In areas with higher rainfall, however, a raised surface develops (plateau bogs and dome bogs) as a result of the greater peat accumulation. When similar processes take place in the Discontinuous Permafrost zone (van Everdingen, 1998, revised May 2005), however, permafrost preferentially develops in those areas with Sphagnum moss cover (Fig. 2.11) because this cover provides good insulation, allowing frost to persist throughout the year. The peat material becomes perennially frozen ﬁrst, and then the permafrost extends into the underlying mineral

C. Tarnocai, V. Stolbovoy

30 Table 2.4.

The Finnish peatland classiﬁcation system (after Ruuhija¨rvi, 1960; Eurola, 1962).

Peatland types Ombrotrophic

Minerotrophic

Combination type with and without trees

Treeless fens

Dwarf shrub–pine bogs (forested) Sphagnum bogs (forested) Cottongrass–pine bogs Hummock and hollow bogs (sparsely forested) Sedge bogs (treeless)

Paludiﬁed pine forest Spruce–pine swamp

Sedge–birch swamp

Sphagnum fen

Sedge–pine swamp

Sedge fen

Carex–spruce–pine swamp Paludiﬁed spruce forest

Eutrophic spruce swamp Eutrophic pine swamp

Flark fen

Herb-rich paludiﬁed spruce forest Vaccinium–spruce swamp Rubus–spruce swamp Equisetum–spruce swamp Herb-rich spruce swamp Eutrophic paludiﬁed hardwood forest

Eutrophic birch swamp

Hollow bogs (treeless)

Meso-eutrophic sedge fen

Eutrophic Sphagnum fen Eutrophic Campylium fen Eutrophic ﬂark fen Eutrophic spring fen

layer, leading to the development of a perennially frozen peatland. At this point ice lenses form, raising the peat surface, and hence palsas and peat plateaus develop. In Subarctic regions, peat plateaus are associated with polygonal cracks resulting from thermal cracking. These thermal cracks develop in the polygonal peat plateaus because of low winter temperatures and are usually associated with ice wedges (Fig. 2.12). The continuous build-up of ice in the frozen core eventually breaks the surface of the palsas and peat plateaus. When breaks occur in the insulating peat layer, thermal erosion (melting) takes place in the frozen core. This process can also be initiated by wildﬁres. As a result, the perennially frozen peatland collapses, forming collapse scars. These wet, collapsed areas are invaded by mosses and sedges, resulting in high peat accumulation, better drainage conditions, and a raised peatland surface, which again leads to permafrost development and, thus, the cycle starts again. In the Arctic region (continuous permafrost zone; van Everdingen, 1998, revised May 2005), peatland development usually starts in depressions or drained lakes (Fig. 2.13). Drained lakes develop when permafrost establishes itself in the lake-bottom sediments, and, as a result of ice build-up, the lake bottom rises and ﬁnally drains all

Northern Peatlands: their characteristics, development and sensitivity Table 2.5.

31

The Russian peatland classiﬁcation system (after Tyuremnov, 1976).

Type

Subtype

Group

Species

Low-moor peat

Forests

Woody

Forest–swampy

Woody–herbaceous

Alder Birch Spruce Pine Willow Woody–sedge Woody–reed Woody–Bryales Woody–Sphagnum Horsetail Scheuchzeria low Sedge Bryales Sedge–reed Reed Horsetail Sedge–Sphagnum Hypnum Sphagnum Hypnum Woody Woody–sedge Woody–Sphagnum Sedge Scheuchzeria Sedge–Sphagnum Sphagnum Hypnum Pine–shrubs Pine–cotton grass Pine–Sphagnum Scheuchzeria Cotton grass Cotton grass–Sphagnum Scheuchzeria–Sphagnum Sphagnum fuscum Angustifolium Sphagnum magellanicum Complex peat Hollow peat

Woody–moss Swampy

Herbaceous

Herbaceous–moss Moss Transitional

Forest Forest–swampy Swampy

Woody Woody–herbaceous Woody–moss Herbaceous Herbaceous–moss Moss

High-moor peat

Forest Forest–swampy Swampy

Woody Woody–herbaceous Woody–moss Herbaceous Herbaceous–moss Moss

of the water. Polygonal cracks usually develop in these drained lake bottoms and ice wedges begin to form in the cracks. The rapid growth of these ice wedges pushes the mineral materials up on both sides of the polygon trench, causing the edges of the polygon to be higher than the central portion (low-center lowland polygons). The

32

Table 2.6.

Characteristics of common peat materials.

Peat type Sphagnum Sedge Brown moss sedge Woody sedge Woody Feather moss Sedimentary Amorphous a b c d e f g

i

c

Sphagnum mosses Sedgesd Brown mossese and sedges Sedges and woodf Wood Feather mossesg Aquatic plantsh Not recognizablei

Degree of decompositiona

van Post value

Fibre contentb (%)

Bulk density (g/cm3)

pH

Undecomposed Moderate Moderate–undecomposed Moderate Moderate–well Moderate Well Well

H1–3 H5–7 H5–7 H4–6 H5–8 H4–7 H8–10 H6–10

60–90 8–30 8–30 10–40 5–30 10–60 2–8 2–8

0.07 0.11 0.11 0.18 0.15 0.12 0.13 0.13

o3 4.5–7.0 5.0–7.0 4.5–7.0 5.0–6.5 4.5–6.5 4.5–6.5 4.5–6.5

Soil Classiﬁcation Working Group (1998). Percent unrubbed ﬁber. Sphagnum spp. Carex spp. Drepanocladus spp., Calliergon spp., Aulacomium spp. derived from coniferous and deciduous tree species. Hypnum spp., Hylocomium spp., and Pleurozium spp. Algae, diatoms, aquatic mosses, and other aquatic organisms. Plant materials that are unidentiﬁable by the naked eye.

C. Tarnocai, V. Stolbovoy

h

Dominant plant material

Northern Peatlands: their characteristics, development and sensitivity

33

Table 2.7. Characteristics of peat in the European part of the former U.S.S.R. (after Lishtvan and Korol’, 1975). Peat classes

Decomposition (%)

Ash (%)

pHKCl

Caloriﬁc value (Kcal/kg)

Mean Range

Mean Range

Mean Range

Mean Range

5.3 5.0 5.0 5.0 4.8 4.9 5.0

4.0–6.5 3.2–6.5 3.8–5.8 2.8–7.3 3.2–7.4 2.8–6.5 3.3–6.7

5540 5630 5570 5580 5510 5420 5542

4930–6230 5160–6140 5210–5930 4980–6130 4930–5880 5000–5790 5035–6016

4.6 4.1 4.6 3.9 4.0 4.1 4.2

3.2–5.9 2.8–5.2 3.2–5.4 3.0–5.5 3.2–5.9 3.2–5.6 3.1–5.6

5790 5870 5750 5690 5590 5400 5682

5380–6390 5430–6400 5370–6490 4510–6410 4860–6090 4910–5730 5077–6252

3.6 3.2 3.5 3.5 3.3 3.2 3.4

2.8–5.5 2.4–4.4 2.7–5.2 2.6–4.8 2.7–5.0 2.7–5.8 2.7–5.1

5940 6010 5760 5700 5560 5100 5678

5480–6130 5630–6510 5310–6480 5140–6310 4760–6480 4500–5990 5137–6317

Woody Woody–herbaceous Woody–moss Herbaceous Herbaceous–moss Moss Average

45 39 35 29 24 21 32

27–65 23–60 18–50 11–59 6–40 3–45 15–53

Woody Woody–herbaceous Woody–moss Herbaceous Herbaceous–moss Moss Average

45 39 38 29 27 22 33

35–65 25–60 25–55 10–60 10–54 3–40 18–56

Woody Woody–herbaceous Woody–moss Herbaceous Herbaceous–moss Moss Average

55 57 38 37 30 23 40

38–65 30–74 18–60 15–60 5–57 1–44 18–60

Low-moor peat 9.6 4.0–18.0 8.0 3.1–17.8 7.8 3.8–15.8 6.7 1.3–17.8 5.6 2.3–12.4 6.5 1.0–15.0 7.4 2.6–16.1 Transitional peat 6.8 2.4–12.8 4.8 2.1–11.4 7.1 2.2–13.7 3.5 1.3–14.2 4.9 1.2–15.9 4.7 1.6–9.1 5.3 1.8–12.9 High-moor peat 3.8 2.4–6.9 2.8 1.6–6.0 3.5 1.4–7.0 2.6 1.1–8.3 2.3 0.8–13.4 2.3 0.7–10.7 2.9 1.3–8.7

low center of the polygon is usually very wet, providing a favorable environment for peat development, which eventually leads to the development of a high-center lowland polygon. These high-center lowland polygons are relatively stable in the southern part of the Arctic. Degradation, if present, is found mainly along the seacoast, lakeshores, and riverbanks. In the High Arctic, however, little or no peat deposition is now occurring and these high-center lowland polygons are in an eroding stage. Basal peat dates indicate that peat deposition associated with the development of high-center lowland polygons in the High Arctic began about 7000 14C yr BP (dates are not calibrated in this chapter), shortly after deglaciation (early Holocene epoch) and, as was suggested by Tarnocai (1978), probably resulted from the favorable (Hypsithermal interval) climatic conditions existing there at that time. As the climate became colder, however, peat development virtually ceased in the High Arctic and these high-center lowland polygons are now in an eroding stage.

34

C. Tarnocai, V. Stolbovoy

Figure 2.11. Schematic diagram showing the current development and cyclic nature of perennially frozen peatlands in the Boreal and Subarctic peatland regions (after Zoltai et al., 1988a, b).

Figure 2.12. Cross section of an ice wedge in an eroding high-center lowland polygon, Low Arctic Peatland Region, Beaufort Sea coast, Northwest Territories, Canada.

Northern Peatlands: their characteristics, development and sensitivity

35

Figure 2.13. Schematic diagram showing the current development and cyclic nature of lowland polygons in the southern part of the Arctic Peatland Region.

Age of peat deposits Most of the northern part of the North American continent, except for the northwestern Yukon Territory and Alaska, was glaciated during the Late Glacial Maximum glaciation (Wisconsinan glaciation). Eventually, the glaciers retreated, exposing the land at about 13,000 14C yr BP in the south and disappearing from the north between about 7000 14C yr BP and 4500 14C yr BP (Prest, 1970). The age of the basal peat indicates the beginning of peat development (Tables 2.8, 2.9, 2.10). Average basal peat dates for the Arctic, Subarctic, and Boreal peatland regions indicate that peatland development was not uniform. The average age of basal peat, calculated from data derived from several sources, was 7200 14C yr BP in the Arctic (23 samples), 6200 14C yr BP in the subarctic (41 samples), and 6000 14C yr BP in the boreal (47 samples) areas. These basal peat dates indicate that peatland development began shortly after deglaciation in Arctic Canada and about 1000–2000 yr after deglaciation in the Subarctic and Boreal portions of the continent. There is evidence that there was a relatively warm and dry period lasting several thousands of years after the retreat of glacial ice from the North American continent (Ritchie and Hare, 1971; Terasmae, 1972). It is possible that, during this period, the climate was generally too warm and dry for optimum peat development over the continent. In High Arctic areas, however, the climate was probably cooler and moister (although warmer than at present) and, consequently, after the retreat of glacial ice conditions were favorable for peat development. As the climate became colder, peat development ceased in the High Arctic, and the Subarctic and Boreal regions became established as the areas of optimum peat development (Tarnocai, 1978). In Russia, the age of Holocene peatlands with the thickest peat is no more than 10–12 thousand years (Table 2.11; Neishtadt, 1985). At some locations in central European Russia, however, there is evidence for peat with ages greater than 45–54 thousand years (Arslanov, 1966). The geographical distribution of the ages of peat in Russia (Table 2.12) shows that younger peat occurs in the tundra (2000–3000 14C yr BP) and in desert and semidesert conditions (1000–3000 14C yr BP). The age of peat in the Taiga zone varies

Radiocarbon dates and rates of peat accumulation for peat deposits in the Arctic Peatland Region of Canada. Peatland type

Sample depth (cm)

81124’N, 76125’W Ellesmere Island, NWTb 81124’N, 76134’W Ellesmere Island, NWTb 75150.5’N, 9812.5’W Bathurst Island, NWTc 75150.5’N, 9812.5’W Bathurst Island, NWTb 75140’N, 97140’W Bathurst Island, NWTb

High-center lowland polygon high-center lowland polygon High-center lowland polygon High-center lowland polygon High-center lowland polygon

200 (basal)

74105’N. 96109’W Cornwallis Island, NWTb 71156’N, 123117’W Banks Island, NWTd 69115’N, 138102’W Phillips Bay, Yukone 69109’N, 134117’W Mackenzie Delta, NWTb 69107’N, 132156’W Eskimo Lakes, NWTe

High-center lowland polygon

a b c d e

High-center lowland polygon High-center lowland polygon High-center lowland polygon High-center lowland polygon

Peat accumulation based on intermittent dates. Tarnocai (unpublished data). Lowdon and Blake (1970). Dyck and Fyles (1963). Zoltai and Tarnocai (1975).

Age (14C yr BP)

Lab no.

Rate of peat accumulationa (cm/ 100 yr)

67807390

AECV+1065C

–

150 (basal)

57807120

AECV+1066C

–

(basal)

9040780

GSC–1887

–

(basal)

92107170

GSC–180

–

25 78 130 75 (basal)

5070760 5830770 6160790 65907100

GSC–2326 GSC–2355 GSC–2317 GSC–2532

– 6.97 15.75 –

GSC–10 GSC–197 BGS–196 BGS–197 Beta–11562 Beta–11564 Beta–11565 BGS–216 BGS–217

– 6.35 – 14.13 – 8.63 10.52 – 10.31

61 244 40 300 58 310 735 22 318

(basal) (basal)

(basal) (basal)

69407110 98207220 82607110 10 1007130 1890760 4810760 8850790 3150790 60207100

C. Tarnocai, V. Stolbovoy

Latitude, longitude, location, and source of dates

36

Table 2.8.

Radiocarbon dates and rates of peat accumulation for peat deposits in the Subarctic Peatland Region of Canada.

Latitude, longitude, location, and source of dates 67149’N, 139150’W Yukonb 67141’N, 132105’W NWTb 67106’N, 125147’W NWTc 63100’N, 129105’W NWTd 61110’N, 100155’W NWTe 58113’N, 71159’W Quebecf 55134’N, 84130’W Ontarioe a b c d e f

Peatland type

Sample depth (cm)

Age (14C yr BP)

Lab no.

Polygonal peat plateau Polygonal peat plateau Peat plateau

23 175 35 229 34 206 (basal) 30 230 (basal) 4 150 7 95 (basal) 5 89 (basal)

3025785 10 0807340 2710760 7200760

S–1865 S–1971 BGS–147 BGS–149

6790775

WIS–275

86407160

GSC–3097

57807110

WIS–67

3640780

QU–977

1897763

BGS–6

Peat plateau Peat plateau Peat plateau Palsa

Peat accumulation based on intermittent dates. Zoltai and Tarnocai (1975). Nichols (1974). MacDonald (1983). Nichols (1967). Couillard and Payette (1985).

Rate of peat accumulationa (cm/ 100 yr) – 2.15 – 4.32 – 3.45 – 3.55 – 2.83 – 3.38 – 5.08

Northern Peatlands: their characteristics, development and sensitivity

Table 2.9.

37

38

Table 2.10.

Radiocarbon dates and rates of peat accumulation for peat deposits in the Boreal Peatland Region of Canada. Peatland type

Sample depth (cm)

Age (14C yr BP)

Lab No.

Rate of peat accumulationa (cm/100 yr)

51105’N, 121159’W British Columbiab 54142’N, 116100’W Albertac 54134’N, 116148’W Albertab 55154’N, 108135’W Saskatchewand 53158’N, 104152’W Saskatchewand 52131’N, 101115’W Porcupine Mountain, Manitobae

Unknown Unknown Unknown Unknown Unknown Bog

630 410 320 420 239 50 80 100 145 170 459 350 290 300 275 300

92107150 83207260 41507140 68557160 37577120 1170760 2000755 2270760 4180775 5140775 43407155 36857240 55807150 57807100 68907120 64607140

GSC511 GSC500 GSC674 S2582 S2573 WIS287 WIS289 WIS303 WIS286 WIS308 S2473 S2468 GSC247 GSC15 QU499 GSC788

6.8 4.9 7.7 6.1 6.4 4.3 3.6 7.4 2.3 2.6 10.6 9.5 5.2 5.2 4.0 4.6

51125’N, 49124’N, 54134’N, 47134’N, 50118’N, 49101’N, a b c d e f g h i

96153’W 95122’W 84140’W 79145’W 77124’W 79105’W

Manitobad Manitobad Ontario, Canadaf Ontario, Canadag Quebec, Canadah Quebec, Canadai

Unknown Unknown Unknown Unknown Unknown Unknown

Peat accumulation based on intermittent dates. Lowdon and Blake (1968). Lowdon et al. (1967). Zoltai et al. (1988b). Nichols (1969). Dyck et al. (1965). Dyck and Fyles (1963). Dionne (1979). Lowdon et al. (1971).

C. Tarnocai, V. Stolbovoy

Latitude, longitude, location, and source of dates

Northern Peatlands: their characteristics, development and sensitivity

39

from 5000 to 8000 14C yr BP. This pattern of age zonation is the result of the different factors involved in peat formation in the north and south. In the north, peat formation began later because of the delayed retreat of glaciers, resulting in younger peat. In the south, the decomposition rate of the organic matter is too high to allow long-term conservation of peat. To a certain extent, this is reﬂected in the distribution of peat thickness, with shallow peat occurring in the north and medium-thick peat in the south.

Peat accumulation Peat is composed of the remains of plant materials deposited by the vegetation growing on the surface of the peatland. As peat deposition continues, the deposited organic material becomes the growing medium for the vegetation. The near-surface peat layer (rooting zone) is subject to decomposition, but organic material is also added because of the roots of the plants growing on it (acrotelm layer). As more peat is added to the surface, the underlying peat layers are submerged below the water table and the peat material is no longer subject to signiﬁcant decomposition (catotelm layer). Long-term average rates of peat accumulation can be calculated using the radiocarbon dates of the basal peat and the thickness of the peat deposit (assuming that the surface of the deposit represents the current date). When this method was used, the average rate of peat accumulation for the Subarctic was found to be 3.75 cm/ 100 yr (25 samples) and for the Boreal 6.35 cm/100 yr (35 samples). Zoltai et al. (1988b) reported an average peat accumulation of 6.4 cm/100 yr for Boreal peatlands. The problem with using this method to determine the rate of peat accumulation is, as indicated above, that the surface of the deposit (0 cm) is assumed to be the current date. During their long history, however, some deposits have been affected by wildﬁres that burn a certain amount of peat. This would affect the accuracy of this method. Another problem is that, although the rate of peat accumulation for the Arctic was found to be 10.49 cm/100 yr (8 samples), this was based on intermittent dates since the surfaces of these peatlands are badly eroded, especially in the Mid and High Arctic. This value, therefore, represents the rate of accumulation during the ﬁrst few thousand years of deposition, not the present rate of peat accumulation, which is very low to nonexistent. The rate of peat accumulation is not uniform over time and can vary greatly within the peat deposit. For example, the peat accumulations for the ﬁve Canadian peat deposits in Figure 2.14 show variations of several centimeters per 100 yr between the peat layers in the deposits. The average rates of peat accumulation for the deposits, calculated from the intermittent dates presented in Figure 2.14, were 6.76 cm/100 yr for Nichols (1974), 2.07 cm/100 yr for Ovenden (1982), 3.22 cm/100 yr for Nichols (1967), 5.18 cm/100 yr for MacDonald (1983), and 4.04 cm/100 yr for Nichols (1969).

40

Table 2.11.

Radiocarbon dates and rates of peat accumulation for peat deposits in the Boreal Peatlands of Russia.

Location and source of dates

Peatland type

Sample depth (cm)

Age (14C yr BP)

Lab no.

Rate of peat accumulation (cm/ 100 yr)

West Siberia, Russiaa

Organic-rich mineral layer underlying peat Herbaceous peat Carex-moss peat Birch–herbaceous Unknown Unknown Unknown Unknown Forest–steppe zone

875–880

96257100

TA–1137

9.1

680–700 280–300 400 275 375 Very deep peat Very deep peat 310 0–4

8040780 3550760 57697130 45707170 51507120 6120780 7260760 4350770 350

TA–1138 TA–113 Mo–434 Mo–433 Mo–467 TA–618 TA–666 TA–596

8.6 8.3 6.9 6.0 7.3

West Siberia, Russiab

a b c d

Piavchenko (1983). Khotinski (1977). Liss and Berezina (1978). Chichagova (1986).

7.1 1.14

C. Tarnocai, V. Stolbovoy

West Siberia, Russiac Siberia, Russiac Arctic, European Russiad

Northern Peatlands: their characteristics, development and sensitivity

41

Table 2.12. Approximate rates of peat accumulation in vegetation zones of Russia determined using age and depth of peat (after Botch et al., 1995). Peatland type and vegetation zone

Polygonal mires; Tundra and forest–Tundra Palsa; Tundra and Forest–Tundra Aapa; northern Taiga Raised string, sphagnum and blanket bogs; northern, middle and southern Taiga Pine bogs, black alder swamps and fens; mixed and broadleaf forest Reed and sedge fens; forest steppe and steppe Marshes; desert and semidesert

Period (14C yr BP)

Mean depth of peat (cm)

Rate of peat accumulation (cm/100 yr) Min

Max

2000–3000

20–70

0.7–1.0

2.3–3.5

5000–8000

100–300

1.3–2.0

3.7–6.0

7500 7000

200–400 200–500

2.3 2.8

5.3 7.1

2000–7500

100–300

1.4–5.0

4.0–15

5000–7000

100–300

1.4–2.0

4.3–6.0

1000–3000

100–200

3.3–6.6

10–20

Another method for determining peat accumulation is to use volcanic tephra of known age as markers. In Table 2.13 the rates of peat accumulation for fens and bogs in western and northwestern Canada were calculated using this method. Calculating peat accumulations using tephra of known age is relatively inexpensive and provides information on the peat accumulation for the given time period. The rate of peat accumulation in Russia is similar to that in Canada, excluding the Tundra zone where there is the lowest rate of peat accumulation (0.7–1.0 cm/100 yr; Table 2.12). The rate of peat formation is approximately 2–7 cm/100 yr in the boreal Taiga zones, which are similar to the boreal forest in Canada. In these zones, the highest rate of peat accumulation (15 cm/100 yr) occurs on nutrient-rich sites, such as pine bogs, black alder swamps and fens, and mixed and broadleaf forests. The most intensive accumulation of peat (10–20 cm/100 yr), however, is found in marshes in the desert and semi-desert vegetation zones because the highest biomass production occurs in wet ecosystems in arid and hot climates. Such marshes are common in the steppe and desert zones from the Dnepr River delta up to the Ural River, including the lower reaches of the Don and Volga rivers. In the Astrakhan State Reserve in the Volga Delta, the marsh successions are initiated by the regression of the Caspian Sea. It has been reported that the thickness of peat in marshes can reach as much as 4 m (Bilyk, 1973).

C. Tarnocai, V. Stolbovoy

42 a

b

0

Depth (cm)

Depth (cm)

100 150

100 150 200

200

250

250 0

c

0 50

50

5 10 15 20 Peat Accumulation (cm/100 yr)

25

0

2

1

3

4

5

6

7

Peat Accumulation (cm/100 yr)

d

0

0 50

40 Depth (cm)

Depth (cm)

20

60 80 100 120

100 150 200

140 160

250 0

e

1

2 3 4 5 6 7 Peat Accumulation (cm/100 yr)

8

9

2 3 4 5 6 Peat Accumulation (cm/100 yr)

7

8

0

2

6 8 4 10 Peat Accumulation (cm/100 yr)

12

0 20

Depth (cm)

40 60 80 100 120 140 160 180

0

1

Figure 2.14. The rates of peat accumulation in various peat layers in the peat deposits. The original data are from (a) Nichols (1974). (b) Ovenden (1982). (c) Nichols (1967). (d) MacDonald (1983). (e) Nichols (1969).

Table 2.13. Rates of peat accumulation in fen and bog peatlands determined using a volcanic ash (tephra) marker horizon. Name of tephra and source

Bridge Rivera St. Helens ‘‘y’’a Mazamaa White Riverb White Riverb a b

Zoltai and Johnson (1985). Tarnocai (1973).

Period (14C yr BP)

0–2350 2350–3500 3500–6600 0–1500 0–1500

Rate of peat accumulation (cm/100 yr) Fen

Bog

4.3 5.2 5.4 1.5 –

4.9 4.3 5.8 2.9 6.9

Northern Peatlands: their characteristics, development and sensitivity

43

Carbon stocks The carbon stocks in northern peatlands were determined by using the NMLSD. The organic carbon mass was determined for the 0–100 cm depth in both Eurasia and North America. Using this database, the organic carbon mass of the northern peatlands in these regions was calculated to be 257 Gt, with the carbon mass for Eurasia being 163 Gt and for North America 94 Gt. Approximately 49% of this carbon (127 Gt) occurs in perennially frozen peatlands, with approximately 95 Gt in Eurasia and 32 Gt in North America. Calculating the organic carbon mass for the 0–100 cm depth provides limited information about the carbon stored in peatlands since most of the peat deposits are more than 100 cm thick. The carbon mass for Canadian peatlands was calculated for the total depth of the peat deposit as well as for depths of 0–30 and 0–100 cm (Table 2.14). These data indicate that the organic carbon mass in the 0–100 cm depth accounts for only 34% of the organic carbon stored in Canadian peatlands. Using 76 peat samples, Zoltai and Johnson (1985) calculated an annual organic carbon accumulation rate of 18.9 g/m2 for Boreal peatlands in Canada. Based on these data and the area of Boreal peatlands, they estimated that the Boreal peatland region of Canada annually sequestered 9.8 million tonnes of organic carbon. The carbon stocks for Russia, based on the most recent estimate, are 194 Gt (Stolbovoi, 2002). This estimate was obtained using the digital soil map of Russia at a scale 1:5 million (Stolbovoi and McCallum, 2002). The calculation was done separately for soils with peat depths of 30–50 cm (swamps) and 450 cm (bogs). The assumption was made that the thickness of peat in soils with a peat depth of 450 cm is 2 m. This assumption is likely valid for the country as a whole since Vomperski et al. (1994) reported that the average peat depth for Russia is 2.11 m. Most of this carbon occurs in the permafrost regions of Russia (Table 2.15). Botch et al. (1995) reported carbon stocks in peatlands of the former U.S.S.R to be about 214 Gt C. Since Russia has more than 80% of the former U.S.S.R. peatlands, calculations indicate that Russia’s share of the carbon pool in peat would be approximately 170 Gt C. This is close to the estimate of Stolbovoi (2002). Approximately 32% of the organic carbon mass occurs in the unfrozen (seasonally frozen) regions of Russia, 27% in the zone of continuous permafrost, and 40% in the zones of sporadic and discontinuous permafrost (Table 2.15).

Table 2.14.

Organic carbon mass in Canadian peatlands.

Type of peatland

Unfrozen Frozen Total a

Carbon mass (Gt) 0–30 cm

0–100 cm

Totala

9.5 5.1 14.6

36.8 14.2 51.0

103.9 45.2 149.1

Calculated for the total depth of the peat deposit plus 10 cm of the underlying organic-rich mineral layer.

C. Tarnocai, V. Stolbovoy

44 Table 2.15.

Organic carbon mass in Russian peatlands.

Zone

Seasonally frozen Sporadic permafrost Discontinuous permafrost Continuous permafrost Total

Peatland carbon mass (Gt) 0–30 cm

0–100 cm

Total (0–200 cm)

12.8 11.3 5.0 27.7 56.8

40.9 38.0 12.3 40.7 131.9

62.9 58.1 20.2 52.9 194.1

Sensitivity to climate change The Subarctic and Boreal zones of North America and the equivalent zones in Northern Eurasia, the Forest–Tundra Transitional zone and Taiga zone, are similar in many respects. These are the areas where most of the world’s peatlands occur and where large portions of the peatlands are underlain by permafrost. In addition, predictions made using global circulation models indicate that these northern areas will be those most affected by climate warming. Based on recent estimates (Kallen et al., 2001), the increase in average annual air temperature for northern areas could be as high as 6 1C under a 2CO2 environment. It is predicted that climate change will not only increase the temperature, but also change the precipitation patterns and, thus, the hydrology of these areas. Because of the large areas of peatlands in these northern regions, the presence of permafrost under much of the area, and the possibility of such a large temperature increase, these peatlands are considered to be very vulnerable to the effects of climate change and are expected to experience similar problems.

Canada In Canada, where most of North America’s peatlands occur, approximately 61% of the area of peatlands in the Boreal and Subarctic peatland regions and 59% of the area of all peatlands will be severely to extremely severely affected by climate change (Tarnocai, 2006). The effect of climate change on the other peatland regions is very small to nonexistent. The impact of climate change will probably be most severe on perennially frozen peatlands, which cover about 36% of the peatland area of Canada. Approximately 97% of the organic carbon occurring in all Canadian peatlands occurs in the Boreal and Subarctic peatland regions. The peatland sensitivity model indicates that approximately 52% of the total organic carbon mass in these two regions (51% in all Canadian peatlands) will be severely to extremely severely affected by climate change (Tarnocai, 2006). Perennially frozen peatlands contain approximately 30% of the organic carbon occurring in Canadian peatlands. The peatland sensitivity model suggests that approximately 86% of this carbon will be severely to extremely severely affected by climate change.

Northern Peatlands: their characteristics, development and sensitivity

45

In the Subarctic Peatland region and the northern part of the Boreal Peatland region, where most of the perennially frozen peatlands occur, the increased temperatures are expected to cause increased thawing of the perennially frozen peat. Thawing of the ice-rich peat and the underlying mineral soil will result in watersaturated conditions similar to those associated with the collapse scar type of peatlands today. These water-saturated conditions, together with the higher temperatures, will probably result in anaerobic decomposition, thus, leading to the production of CH4. In the southern part of the Boreal peatland region, where the peatlands are generally unfrozen, the main impact is expected to be drought conditions resulting from higher summer temperatures and higher evapotranspiration. Under such conditions, peatlands become a net source of CO2 because the oxygenated conditions lead to aerobic decomposition. These dry conditions will likely also increase wildﬁres and, eventually, burning of peat, leading to the production of CO2. Similar drying conditions and wildﬁres probably will also affect the Mountain and Intermountain Prairie peatland regions. Climate warming is also expected to cause a rise in sea levels (Hungate et al., 2003). Flooding is expected to occur on low-lying coastal peatlands such as those in the Hudson Bay Lowland in central Canada and in the Paciﬁc, Atlantic, and Arctic coastal areas.

Northern Eurasia In Northern Eurasia, peatlands generally occur in the Subarctic and Boreal zones, and cover nearly 25% of the Tundra zone, 30% of the Forest–Tundra (the transitional zone between the Tundra and Taiga zones) and Northern Taiga zones, and 25% of the Middle and Southern Taiga zones (Stolbovoi, 2002). It is estimated that approximately 70% of Boreal Forest, or Taiga zone, the largest ecosystem in northern Eurasia, is underlain by permafrost. In the Asian part of Russia, where the transitional Forest–Tundra and Northern Taiga zones cover a substantial area, permafrost occur in more than 80% of the territory. Nearly all of the organic carbon in Eurasian peatlands occurs in the Subarctic and Boreal regions, with the total organic carbon mass in peatlands exceeding 30% of the total organic carbon mass in all soils in the Middle Taiga zone, and approaching 30% in the Forest–Tundra and Northern Taiga zones. The latest predictions indicate that most of these peatlands would be severely affected by climate change (IPCC, 2001). In the forest zones of European Russia and the relatively humid West Siberia, a drying climate is expected to cause increased evaporation in the bog ecosystems with a resultant drop in their water table. Drying of these bogs, especially in West Siberia, where the major areas of bogs occur, would cause an increase in albedo and total ecosystem respiration. Under this scenario, bogs might change from a CO2 sink to a CO2 source, thereby decreasing the peat accumulation and releasing additional CO2 and CH4 to the atmosphere (Vygodskaya et al., 2004).

46

C. Tarnocai, V. Stolbovoy

In the forests of the drier areas of central Siberia and European Russia, the drying climate would increase the possibility of ﬁres, especially if the expected increase in summer temperature is not accompanied by an adequate increase in precipitation. In areas where moisture content is relatively higher, however, the water table would rise, leading to the formation of peatlands in large areas of boreal forests on heavy clay soils or permafrost (Vygodskaya et al., 2004). A general circulation model for central and eastern Siberia (Vygodskaya et al., 2004) predicts that, by 2090, the Tundra (Arctic) and Forest–Tundra (Subarctic) zones, which now cover approximately 33% of the Siberian area, will nearly disappear and the Taiga (Boreal) zone, which now covers approximately 67% of the area, will shift northward, but will reduce in size to cover only 40% of the area. As a result, the Tundra, Forest–Tundra, and Taiga zones would have lost approximately 60% of their areas, having a major effect on the large areas of peatlands in these zones and the permafrost that underlies much of the area. The conditions of the Forest–Steppe (a transitional zone), Steppe, Semi-desert (or Steppe-Desert, a transitional zone), and Desert zones, which are now practically absent, are predicted to appear with the Forest-Steppe zone occupying up to 45% of the area and the Steppe, Semi-desert, and Desert zones occupying up to 15%. Thus, all the boundaries of the ecosystems in the area would be subject to spatial perturbations with the Taiga shifting northward into the Tundra and Forest–Tundra zones, but losing 40% of its area, whereas the Forest–Steppe zone would replace it in the south, with smaller amounts of the Steppe, Semi-desert, and Desert zones also appearing (ACIA, 2004; Volgodskaya, 2004). The predicted broad-scale change in climate and the consequent alteration of ecological conditions will affect the carbon cycle in the Subarctic and Boreal ecosystems. There are, however, two principal questions: what effect will climate change have on the net carbon balance of northern ecosystems, and will they be a CO2 sink or source? The uncertainty results from the fact that the terrestrial carbon balance will be driven by two conﬂicting processes: an increase in plant productivity (CO2 uptake from the atmosphere) and an intensiﬁcation of the decomposition of soil organic carbon (CO2 release into the atmosphere). Since there is a 5–10 times higher concentration of organic carbon in soils than in vegetation in these ecosystems, it is expected that the net effect of these processes would be for the northern ecosystems to become a CO2 source because of the relatively higher impact of decomposition (IPCC, 2001). It is worth noting, however, that the actual trajectory of the processes will depend to a great extent on the different response times of the various ecosystems in the affected zones. In addition, the observed climate changes in Russia form a mosaic that is largely dependent on regional physiographic features (IPCC, 2001). For example, changes in precipitation have differed dramatically from one region to another during the past 100 years (Hulme, 1995) and differences also have been reported in the global warming experienced from 1976 to 1999 in the Boreal zones of Eastern and Western Siberia and European Russia. Thus, the overall effect of climate change on ecosystem succession is very uncertain and difﬁcult to predict from a decadal perspective. In fact, studies do not always support the expected intensiﬁcation of decomposition and a shift in carbon balance

Northern Peatlands: their characteristics, development and sensitivity

47

towards a CO2 source. For example, Zamolodchikov and Karelin found the Tundra zone in Russia to be a CO2 sink (Zamolodchikov and Karelin, 1998). Field observations of the carbon balance show that recent climate warming has increased carbon accumulation in peatlands (Aurela et al., 2004; Domisch et al., 2004; Hartman and Pietila¨inen, 2004). This is the result of the thawing of permafrost, increase in depth of the active soil layer, and improvement of the thermal conditions, favoring peat growth. A comparison of the input of plant remains and heterotrophic respiration for peatlands in various regions of Russia shows that shallow peat is thickening while deep peat is beginning to degrade (Stolbovoi et al., 2001; Stolbovoi, 2005). Analysis of the organic carbon content in different Russian ecosystems indicates that zonal shifts will probably lead to the accumulation of organic carbon in soils (Stolbovoi, 2002). The available information on the effect of climate change on northern peatlands in both North America and Northern Eurasia indicates that these regions will similarly be severely affected. In both regions there will be major shifts northward in peatland, vegetation, and permafrost boundaries, with some of these regions being reduced signiﬁcantly in area. Since little is known about the northern circumpolar peatlands (extent, depth of peat, characteristics, etc.), additional information is needed to make more reliable predictions as to how these peatlands will react to climate change, both initially and as a result of feedbacks.

References ACIA, 2004. Impacts of a Warming Arctic: Arctic Climate Impact Assessment. Cambridge University Press, Cambridge, 139pp. (Available from http://www.acia.uaf.edu). Arslanov, A.A., 1966. Correction of the age of Upper Pleistocene deposits in Yaroslav Povolzhje. In the Russian journal: ‘Upper Pleistocene. Stratigraphy and absolute chronology’. Nauka, Moscow, Russia, pp. 133–140. (In Russian) Aurela, M., Laurila, T., and Tuovinen, J.-P., 2004. Interannual variation of the CO2 balance in a subarctic fen. In: Pa¨lva¨nen, J. (Ed.), Proceedings of the 12th International Peat Congress: Wise use of peatlands. Oral presentations, Tempere, Finland, Vol. 1, pp. 103–108. Bartalev, S.A., Isaev, A.S., Shugart, H.H., et al., 2004a. Terrestrial ecosystem dynamics. In: Groisman, P.Y. and Bartalev, S.A. (Eds.), Northern Eurasia Earth Science Partnership Initiative: Science Plan, pp. 18–28. (Available from http://www.neespi.org/science/). Bartalev, S.A., Isaev, A.S., Shugart, H.H., et al., 2004b. Scientiﬁc Background Appendix Chapter 3. In: Groisman, P.Y. and Bartalev, S.A. (Eds), Northern Eurasia Earth Science Partnership Initiative: Science Plan, pp. 168–217. (Available from http://www.neespi.org/science/). Bilyk, G.I. (Ed.), 1973, Torfovo-bolotnii fond U.S.S.R., iogo raionuvannya ta vikoristannya. Naukova dumka, Kiev, 236pp. (In Russian). Botch, M.S., Kobak, K.I., Vinson, T.S., and Kolchugina, T.P., 1995. Carbon pools and accumulation in peatlands of the former Soviet Union. Global Biochem. Cycles. 9, 37–46. Botch, M.S. and Masing, V.V., 1979a. Regionality in mire typology in the USSR. In: Classiﬁcation of mires and peats. Proceedings of International Symposium, Hyytia¨la¨, pp. 1–11. Botch, M.S. and Masing, V.V., 1979b. Ecosystemy bolot SSSR (Bogs Ecosystems in the USSR). Nauka, Leningrad, 188pp. (In Russian). Brown, G., 1980. Palsas and other permafrost features in the lower Rock Creek valley, west-central Alberta. Arctic Alpine Res. 12, 31–40. Chichagova, O.A., 1986. Radiocarbon dating of soils: methods, interpretation and application. In: Proceedings: Evolution and Age of the U.S.S.R. Soils. Puschino, pp. 75–93. (In Russian).

48

C. Tarnocai, V. Stolbovoy

Couillard, L. and Payette, S., 1985. Evolution holoce`ne d’une tourbie`re a` pe´rge´lisol (Que´bec nordique). Can. J. Bot. 63, 1104–1121. Cryosol Working Group, 2003. Northern and Mid Latitudes Soil Database, Version 1. National Soil Database, Research Branch, Agriculture and Agri-Food Canada, Ottawa. Dionne, J.C., 1979. Radiocarbon dates of peat and tree remains from James Bay area, subarctic Quebec. Can. J. Forestry Res. 9, 125–129. Domisch, T., Finer, L., Laiho, R., and Laine, J., 2004. Temperature effect on litter decomposition in peat soil. Interannual variation of the CO2 balance in a subarctic fen. In: Pa¨lva¨nen, J. (Ed.), Proceedings of the 12th International Peat Congress: Wise Use of Peatlands. Oral presentations, Tempere, Finland, Vol. 1, pp. 124–132. Dyck, W. and Fyles, J.G., 1963. Geological Survey of Canada radiocarbon dates, I and II. Geological Survey of Canada, Paper 63–21, Ottawa, 31pp. Dyck, W., Fyles, J.G., and Blake, W. Jr., 1965. Geological Survey of Canada radiocarbon dates, IV. Geological Survey of Canada, Paper 65–4, Ottawa, 23pp. Eurola, 1962. U¨ber die Regionale Einteilung der Su¨dﬁnnischen Moore. Ann. Bot. Soc. Zoolog. Bot. Fenn. Vanamo 33, 1–243. Glooschenko, W.A., Tarnocai, C., Zoltai, S., and Glooschenko, V., 1993. Wetlands of Canada and Greenland. In: Whigham, D.F., Dykyjova´, D., and Hejny´, S. (Eds), Wetlands of the World: Inventory, Ecology, and Management. Kluwer Academic Press, Netherlands, Vol. 1, pp. 415–514. Gorham, E., 1988. Canada’s peatlands: their importance for global carbon cycle and possible effect of ‘‘greenhouse’’ climate warming. Trans. Roy. Soc. Canada, Ser. V 3, 21–23. Halsey, L.A. and Vitt, D.H., 1995. Disequilibrium response of permafrost in boreal continental western Canada to climate change. Clim. Change 30, 57–73. Halsey, L., Vitt, D., and Zoltai, S., 1997. Climatic and physiographic controls on wetland type and distribution in Manitoba, Canada. Wetlands 17, 243–262. Hartman, M. and Pietila¨inen, P., 2004. Decomposition of surface peat on mires drained for forestry along throphic and climate gradients. In: Pa¨lva¨nen, J. (Ed.), Proceedings of the 12th International Peat Congress: Wise Use of Peatlands. Poster presentations, Tempere, Finland, Vol. 2, pp. 1215–1218. Heginbottom, J.A., Dubreuil, M.-A., and Parker, P.A., 1995. Canada–Permafrost. National Atlas of Canada, Natural Resources of Canada, Ottawa (MCR 4177). Hulme, M., 1995. Estimating global changes in precipitation. Weather 50, 2. Hungate, B.A., Dukes, J.S., Rebecca, M.R., et al., 2003. Nitrogen and climate change. Science 302, 1512–1513. IPCC (Irish Peatland Conservation Council), 2000. Peatlands around the world. (Available from http:// www.ipcc.ie/wptourhome1.html). IPCC (Irish Peatland Conservation Council), 2001. Climate change 2001: The scientiﬁc basis. In: Houghton, J.T., Ding, Y., Griggs, D.J., and Noguer, V. (Eds), Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, 881pp. Kallen, E., Katsov, V., Walsh, J., and Weatherhead, E., 2001. Report from the Arctic climate impact assessment modelling and scenarios workshop, January 29–31, 2001, Stockholm, Sweden. (Available from http://www.acia.uaf.edu). Katz, N.Y., 1948. Tipy bolot SSSR i Zapadnoi Yevropy i ikh geographicheskoe rusprostranenie (Bogs types in the USSR and Western Europe and their geographical distribution). Geografgiz, Moscow, 320pp. (In Russian). Kettles, I.M., Tarnocai, C., and Bauke, S.D., 1997. Predicted permafrost distribution in Canada under a climate warming scenario. In: Current Research 1997–E, Geological Survey of Canada, pp. 109–115. Khotinski, N.A., 1977. Holocene of the Northern Eurasia. Nauka, Moscow, (In Russian). Lishtvan, I.I. and Korol’, N.T., 1975. Osnovnye svoistva torfa I metody ikh opredeleniya (Main Peat Characteristics and Methods of Their Determination). Nauka i technuka, Minsk, Byelorussia, 318pp. (In Russian). Liss, O.L. and Berezina, N.A., 1978. Age of bogs and the intensity of peat forming in the central part of West Siberia. In: Liss, O.L. (Ed.), Genesis and Dynamics of Bogs. Moscow State University, Moscow, pp. 12–19. (In Russian).

Northern Peatlands: their characteristics, development and sensitivity

49

Lowdon, J.A. and Blake, W., Jr., 1968. Geological Survey of Canada radiocarbon dates, VII. Geological Survey of Canada, Paper 68-2, Part B, Ottawa, pp. 207–245. Lowdon, J.A. and Blake, W., Jr., 1970. Geological Survey of Canada radiocarbon dates, IX. Geological Survey of Canada, Paper 70-2, Ottawa, pp. 46–86. Lowdon, J.A., Fyles, J.G., and Blake, W., Jr., 1967. Geological Survey of Canada radiocarbon dates, VI. Geological Survey of Canada, Paper 67-2, Part B, Ottawa, 42pp. Lowdon, J.A., Robertson, I.M., and Blake, W., Jr., 1971. Geological Survey of Canada radiocarbon dates, XI. Geological Survey of Canada, Paper 71-1, Ottawa, pp. 255–324. MacDonald, G.M., 1983. Holocene vegetation history of the upper Natla River area, Northwest Territories, Canada. Arctic Alpine Res. 15, 169–180. Montanarella, L., Jones, R.J.A., and Hiederer, R., 2006. The distribution of peatland in Europe. Mires and Peat 1: Art. 1. (Online: http://www.mires-and-peat.net/map01/map_1_1.htm). Moore, P.D. and Bellamy, D.J., 1974. Peatlands. Elek, London, UK, 221pp. National Wetlands Working Group, 1988. Wetlands of Canada. Ecological Land Classiﬁcation Series, No. 24. Sustainable Development Branch, Environment Canada, Ottawa and Polyscience Publications Inc., Montreal, 452pp. National Wetlands Working Group, 1997. The Canadian Wetland Classiﬁcation System2nd Edn. Wetland Research Centre, University of Waterloo, Waterloo, Ontario, Canada, 68pp. Neishtadt, M.I., 1985. Bog-forming processes in Holocene. Reports of Academy of Sciences, Geography Vol 1, 39–48 (In Russian). Nichols, H., 1967. The post-glacial history of vegetation and climate at Ennadai Lake, Keewatin, and Lynn Lake, Manitoba, Canada. Eiszeitalter und Gegenwart, 18. 176–197. Nichols, H., 1969. The late Quaternary history of vegetation and climate at Porcupine Mountain and Clearwater Bog, Manitoba. Arctic Alpine Res. 1, 155–167. Nichols, H., 1974. Arctic North American palaeoecology: the recent history of vegetation and climate deduced from pollen analysis. In: Ives, J.D. and Barry, R.G. (Eds), Arctic and Alpine Environments. Methuen and Company Ltd, London, pp. 637–667. Ovenden, L., 1982. Vegetation history of a polygonal peatland, northern Yukon (Canada). Boreas 11, 209–224. Piavchenko, N.I., 1983. On the age of peats and the change in vegetation in the south of West Siberia in Holocene. Bulletin of the commission on the study of Quaternary period. Nauka, Moscow, Vol. 52, pp. 164–170. (In Russian). Prest, V.K., 1970. Quaternary geology of Canada. In: Douglas, R.J.W. (Ed.), Geology and Economic Minerals of Canada. Geological Survey of Canada, Economic Geology Report No. 1, Ottawa, pp. 675–764. Ritchie, J.C. and Hare, F.K., 1971. Late Quaternary vegetation and climate near the arctic tree line of northwestern North America. Quatern. Res. 1, 331–342. Ruuhija¨rvi, R., 1960. U¨ber die regionale Einteilung der Norﬁnnischen Moore. Ann. Bot. Soc. Zoolog. Bot. Fenn.Vanamo 31, 1–360. Seppa¨la¨, M., 1980. Stratigraphy of a silt–cored palsa, Atlin region, British Columbia, Canada. Arctic 33 357–365. Sjo¨rs, H., 1961. Forest and peatland at Hawley Lake, northern Ontario. National Museums of Canada, Bulletin No. 171, Ottawa, pp. 1–31. Sjo¨rs, H., 1963. Bogs and fens on Attawapiskat River, northern Ontario. National Museums of Canada, Bulletin No. 186, Ottawa, pp. 45–133. Soil Classiﬁcation Working Group, 1998. The Canadian system of soil classiﬁcation, 3rd edn. Agriculture and Agri-Food Canada Publication No. 1646, NRC Research Press, Ottawa, 187pp. Stolbovoi, V., 2002. Carbon in Russian soils. Clim. Change 55(1–2) 131–156. Kluwer Academic Publishers, the Netherlands. Stolbovoi, V., 2005. Carbon cycle in the regions with different extent of frozen soils in Russia. In: Cryosols: Genesis, Ecology and Management, Materials of the IV International Conference on Cryopedology, Arkhangelsk-Pinega, Russia, August 1–8, 2005. Institute of Geography, Russian Academy of Sciences, Moscow, pp. 11–12.

50

C. Tarnocai, V. Stolbovoy

Stolbovoi, V. and McCallum, I., 2002. Land resources of Russia. International Institute for Applied Systems Analysis and the Russian Academy of Science, Laxenburg, Austria. CD-ROM. (Available from http://www.iiasa.ac.at/Research/FOR/russia_cd/lcov_des.htm). Stolbovoi, V., Nilsson, S., and Shvidenko, A., 2001. Effect of climate warming on carbon balance of cold ecosystems of Russia. Transactions of the Third International Conference on Dynamics and Challenges of Cryosols. Institute of Geography, Copenhagen, Denmark. Tarnocai, C., 1973. Soils of the Mackenzie River area. Environmental–Social Committee. Northern Pipelines, Task Force on Northern Oil Development, Report No. 73-26, 136pp. Tarnocai, C., 1978. Genesis of organic soils in Manitoba and the Northwest Territories. Proceedings of the Third York Quaternary Symposium, pp. 453–470. Tarnocai, C., 2006. The effect of climate change on carbon in Canadian peatlands. Glob. Planet. Change 53, 222–232. Tarnocai, C., Hohban, L., and Lacelle, B., 2004a. Peatlands of coastal British Columbia. Agriculture and Agri-Food Canada, Research Branch, Ottawa. (Database and two 1:600 000 scale maps). Tarnocai, C., Hohban, L., and Lacelle, B., 2004b. Peatlands of Southern Ontario. Agriculture and AgriFood Canada, Research Branch, Ottawa. (Database and four 1:350 000 scale maps). Tarnocai, C., Kettles, I.M., and Lacelle, B., 2003. Peatlands of the Mackenzie River Valley. Geological Survey of Canada, Ottawa, Open File Report 4413. (Database and map). Tarnocai, C., Kettles, I.M., and Lacelle, B., 2004c. Peatlands of Canada. Agriculture and Agri-Food Canada, Research Branch, Ottawa. (Database and 1:6 500 000 scale map). Tarnocai, C., Kettles, I.M., and Lacelle, B., 2004d. Soil organic carbon content of Canadian peatlands. Agriculture and Agri-Food Canada, Research Branch, Ottawa. (1:7 500 000 scale map). Tarnocai, C., Kettles, I.M., and Lacelle, B., 2004e. Soil organic carbon mass of Canadian peatlands. Agriculture and Agri-Food Canada, Research Branch, Ottawa. (1:7 500 000 scale map). Terasmae, J., 1972. Muskeg as a climate–controlled ecosystem. Proceedings of the 14th Muskeg Research Conference, Kingston, N.R.C. Tech. Memo No. 102, pp. 147–158. Tyuremnov, S.N., 1976. Torfyanye mestorozhdeniya (peat deposits). Nedra, Moscow, p.464. (In Russian). van Everdingen, R. (Ed.), 1998, revised May 2005. Multi-language Glossary of permafrost and related ground-ice terms. National Snow and Ice Data Center/World Data Center for Glaciology. Boulder, CO. 90pp. (Available from http://nsidc.org/fgdc/glossary). Vitt, D.A., Achuff, P., and Andrus, R.E., 1975. The vegetation and chemical properties of patterned fens in the Swan Hills, north central Alberta. Can. J. Bot. 53, 2776–2795. Vitt, D.A. and Bayley, S., 1984. The vegetation and water chemistry of four oligotrophic basin mires in northwestern Ontario. Can. J. Bot. 62, 1485–1500. Vitt, D.A., Halsey, L.A., Bauer, I.E., and Campbell, C., 2000. Spatial and temporal trends in carbon storage of peatlands of continental western Canada through the Holocene. Can. J. Earth Sci. 37, 683–693. Vomperski, S.E., Ivanov, A.I., Ciganova, O.P., et al., 1994. Wetland soils and mires of Russia and carbon pool of their peat. Pochvovedenie 12, 17–25 (In Russian). Vygodskaya, N.N., Groisman, P.Ya., Tchebakova, N.M., et al., 2004. Ecosystems and climate interactions. In: Groisman, P.Y. and Bartalev, S.A. (Eds.), Northern Eurasia Earth Science Partnership Initiative: Science Plan, Chapter 3.5, pp. 62–78. (Available from http://www.neespi.org/science/). Washburn, A.L., 1983. Palsas and continuous permafrost. Proceedings, 4th International Permafrost Conference, National Academy Press, Washington, DC, pp. 1372–1377. Yu, Z.C., Bhatti, J.S., and Apps, M.J., 2001. Long-term dynamics and contemporary carbon budget of northern peatlands. Information Report NOR–X-383, Northern Forestry Centre, Edmonton, Canada, 86pp. Zamolodchikov, D.G. and Karelin, D.V., 1998. Biogenic carbon ﬂuxes in Russia tundra. In: Zavarzin, G.A. (Ed.), Carbon Turnover on Russia Territory. SSRC WGD Ministry of Education of Russia, Moscow, pp. 146–163 (In Russian). Zoltai, S.C., 1972. Palsas and peat plateaus in central Manitoba and Saskatchewan, Can. J. Forestry Res.. 2, 291–302. Zoltai, S.C., 1994. Permafrost distribution in peatlands of west–central Canada during the Holocene warm period 6000 years BP. Ge´ogr. phys. Quatern. 49, 45–54.

Northern Peatlands: their characteristics, development and sensitivity

51

Zoltai, S.C. and Johnson, J.D., 1985. Development of a treed bog island in a minerotrophic fen. Can. J. Bot. 63, 1076–1085. Zoltai, S.C. and Pollett, F.C., 1983. Wetlands in Canada: their classiﬁcation, distribution, and use. In: A.J.P. Gorem (Ed.), Mires: Swamp, Bog, Fen, and Moor. Regional Studies. Elsevier, Amsterdam, Vol. B, pp. 245–268. Zoltai, S.C. and Tarnocai, C., 1971. Properties of a wooded palsa in northern Manitoba. Arctic Alpine Res. 3, 115–129. Zoltai, S.C. and Tarnocai, C., 1975. Perennially frozen peatlands in the western Arctic and Subarctic of Canada. Can. J. Earth Sci. 12, 28–43. Zoltai, S.C., Tarnocai, C., Mills, G.F., and Veldhuis, H., 1988a. Wetlands of subarctic Canada. In: National Wetlands Working Group, Wetlands of Canada. Ecological Land Classiﬁcation Series, No. 24, Sustainable Development Branch, Environment Canada, Ottawa and Polyscience Publications Inc., Montreal, pp. 54–96. Zoltai, S.C., Taylor, S., Jeglum, J.K., et al., 1988b. Wetlands of Boreal Canada. In: National Wetlands Working Group, Wetlands of Canada. Ecological Land Classiﬁcation Series, No. 24, Sustainable Development Branch, Environment Canada, Ottawa and Polyscience Publications Inc., Montreal, pp. 97–154. Zoltai, S.C. and Vitt, D.H., 1990. Holocene climatic changes and the distribution of peatlands in western interior Canada. Quatern. Res. 33, 231–240.

This page intentionally left blank

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

53

Chapter 3

The cold-climate peatlands of the Hudson Bay Lowland, Canada: brief overview of recent work I.P. Martini

Introduction This chapter reports selected basic information about the peatlands of the Hudson Bay Lowland (HBL) gathered in the past four decades or so. (1) The geographic, geological and climatic settings are presented ﬁrst. (2) Various aspects of the evolution of the peatlands and peat will follow. (3) Next is a brief analysis of the peat stratigraphy reporting the few studies made on the palynology and the organic mineralogy (macerals). (4) The chemistry and carbon emission from the vast peatlands are then considered, the latter having been analyzed during the 1990 summer by American and Canadian multidisciplinary agencies. (5) A brief note on the main resources of the area ends the chapter. The HBL is a vast, 325,000 km2, poorly drained plain located southwest of James Bay and Hudson Bay, central Canada. More than 90% of the land is covered by one of the largest, unconﬁned, cold peatlands in the world, second only to the 540,000 km2 West Siberian Plain (Sjo¨rs, 1959, 1963; Neishstadt, 1977; Martini and Glooschenko, 1985; National Wetlands Working Group, 1988). It extends for about 1400 km in length and about 540 km at its widest, from about latitude 581 400 to 501N latitude and from longitude 931 to 701W longitude (Fig. 3.1a). The area is mostly underlain by Paleozoic carbonates, clastics and minor evaporites, and locally, in the southeastern corner, poorly cemented Mesozoic clastic rocks (Norris, 1986; Shilts, 1986). Its surface was molded by Pleistocene ice sheets that developed from centers just to the north (Keewatin area, K) and the northeast (Labrador trough, Q) (Fig. 3.1b). In the Upper Pleistocene, Laurentide Ice Sheet was the last one to cover the area and left a legacy of sediments (till, diamicton), landforms (such as eskers, moraines) and glacial and glaciomarine deposits. One effect of the continental glaciers was to force a signiﬁcant (more than 200 m) subsidence of the land under their weight. Upon ice retreat, the land was inundated ﬁrst by glacial lakes at the southern end, and soon after, about 7500 14C yr BP, seawater entered through the deglaciated Hudson Strait (HS), and an early glacial sea (Tyrrell Sea; Shilts, 1986) ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09003-1

54

I.P. Martini

The cold-climate peatlands of the Hudson Bay Lowland, Canada

55

covered the entire HBL depositing a blanket of marine silty clays, particularly in substrate swales. From early Holocene the HBL history has been one of postglacial isostatic rebound and emergence from the sea at rates ranging from about 3 million/ century during early deglaciation times to 0.7–1 million/century nowadays (Webber et al., 1970). The HBL is, therefore, a recently emerged, coastal plain characterized by extensive tidal ﬂat areas and series of beach ridges, everything locally dissected by slightly entrenching shallow rivers and creeks. The present shores of the James Bay and Hudson Bay represent the modern stage of development of such a regressive system. The average slope of the HBL plain is approximately 0.5 m/km and this has contributed to its poor drainage and the development of the unconﬁned peatlands. The HBL is now subjected to a humid microthermal–Arctic climate (Dcf under the Ko¨ppen system) (Chapman and Thomas, 1968). Mean annual temperature is about 11C in the south (between 521 and 501N latitude) and 41C in the north (between 581 40’ and 521N latitude). The winters are cold with January temperatures ranging between mean daily maxima of 151C in the south and 181C in the north, and respective mean daily minima of 271C and 291C. Summers can be warm with July average temperatures between 12 and 18 (Fig. 3.2) and daily maxima of 221C in the south and 201C in the north. Annual snow precipitation varies between 241 and 203 cm from south to north and total precipitation between 660 and 610 mm. Winds are usually consistent and strong. The climate of the region is inﬂuenced not only by the Arctic air masses that can move south unimpeded by mountain ranges or other obstacles but also by the movement of sea and ﬂuvial waters. Cold Arctic waters enter the inland seas from the Fury Hecla Strait (FH) at the northwest corner of Foxe Basin (FB) and from the HS, and move along a geostrophic anticlockwise current (Figs. 3.1b, 3.3). The effect of this ﬂow is to refrigerate the western coasts of Hudson Bay and James Bay. In the southern part of HBL, northward ﬂowing rivers carry warm, freshwater into the shallow James Bay. The result is a freshening and warming up the water of the bay, which is carried by the geostrophic current northward along the eastern shores warming them up. All this leads to a dramatic shift of the discontinuous and continuous permafrost boundary from lower latitudes in the west to higher latitudes in the east (Fig. 3.3). This permafrost is not a residue of Pleistocene glacial times, rather it is a new, progressive development as the land has emerged and is still emerging from the sea. Peatland evolution During the last 5–6 thousand years unconﬁned peatlands have developed on the enlarging, poorly drained strandplain ranging from marsh wetlands (that is, the Figure 3.1. North America Holocene and upper Pleistocene settings. (a) Wetland regions of Canada and location of the Hudson Bay Lowland (AH ¼ High Arctic; Am ¼ Mid-Arctic; Al ¼ Low Arctic; Sh ¼ High subarctic; Sl ¼ Low subarctic; Sa ¼ Atlantic subarctic; Bh ¼ High boreal; Bm ¼ Mid-boreal; Ba ¼ Atlantic boreal; Bl ¼ Low boreal; P ¼ Prairie; Pi ¼ Intermontane prairies; TE ¼ Eastern temperate; Mx ¼ Mountain compex (after Zoltai, 1980). (b) Upper Pleistocene–Holocene glacier retreat from North America (8 ¼ isochrone of glacier retreat; FB ¼ Foxe Basin; FH ¼ Fury Hecla Strait; HBL ¼ Hudson Bay Lowland; HS ¼ Hudson Strait; JB ¼ James Bay; K ¼ Keewatin center of glaciation; Q ¼ Labrador trough center of glaciation) (after Prest, 1970).

56

I.P. Martini

Figure 3.2. July mean air temperature and annual precipitation in Canada (after Atlas of climatic maps, Canada, 1967).

northern marshes are not true peatlands in this area, having less than 30 cm of peat on them), to coastal and inland fens, to bogs farther inland with peat thickness ranging upto 4–6 m (Fig. 3.4; Glooschenko and Martini, 1983; Klinger and Short, 1996). Fens rim the freshwater marshes in the coastal zone, and cover vast, slightly depressed inland area that is slowly drained by major rivers and their tributaries (Figs. 3.5a, b). Bogs occur preferentially in interﬂuvial, slightly higher areas that are generally poorly drained. Swamps are not common in HBL except in parts of some treed fen zones and, primarily, along streams at the transition between forests of the better drained part of the banks and the fens that are farther inland (Fig. 3.6a). The peat deposit has developed primarily through a process of progressive vegetation, colonization and paludiﬁcation of ﬂat, emergent tidal ﬂat and inter-ridge areas. Relatively few, local terrestrialization successions develop in ponds formed in stream-reaches abandoned by avulsion and in inter-ridge swales, and in disseminated shallow inland lakes. The rivers affect the development of the peatland in several ways. During incipient stages, near river mouths, peat formation is retarded on freshwater marshes due to recurring spring ﬂoods that carry suspended and ice-rafted sediment on the ﬂat lands (Tarnocai, 1982). Part of the river-born sediment is transported offshore during

The cold-climate peatlands of the Hudson Bay Lowland, Canada

57

Figure 3.3. Permafrost, mean annual temperature of Canada and marine currents in the Canadian inlands seas (after Atlas of climatic maps, Canada, 1967).

ﬂoods, where the geostrophic marine currents, tides and waves, redistribute it. The other part is deposited on shoals within the main channel, which eventually evolve into islands generating a typical anastomosing shallow estuarine reaches (Fig. 3.6b). As the land rises, the rivers entrench and progressively change their morphology to single, slightly meandering channels, by capturing and abandoning secondary channels of the original anastomosing system (Figs. 3.6a, b). Dry banks develop along the main streams sustaining mixed forest dominated by white spruce (Picea mariana) and white birch (Betula papyrifera). Inland, the main effect of the rivers and their tributaries on the surrounding peatlands is to provide ribbon-like, drainage pathways. Away from the direct effect of rivers, above an Upper Pleistocene till substrate, the incipient regressive sequence that develops on emerging tidal ﬂats is characterized mainly by a patchy, thin blanket (erosional remnant) of the Tyrrell Sea silty clays, capped by a veneer of tidal-ﬂat deposits of silt, sand with disseminated pebbles, and of salt marsh deposits of ﬁnely bioturbated, organic-rich, silty laminae alternated with slightly thicker storm laminae of silts and ﬁne- to medium-grained sand (Figs. 3.7a, b). These northern salt marshes are colonized primarily by Puccinellia phryganodes. The colonization of the marshes in places is patchy due to sea-ice

58

I.P. Martini

Figure 3.4. Generalized key for wetland classes (according the Canadian classiﬁcation peatlands are wetlands with at least 30 cm of peat; accordingly, the salt marshes of the forests developed on dry places of the Hudson Bay Lowland cannot be considered peatlands) (after Martini and Glooschenko, 1985).

scours. Numerous ponds develop which are enlarged on some coasts and kept open by hundreds of thousands of grazing geese (Kotanen and Jefferies, 1997). As the land rises further, freshwater upper marshes develop, which are characterized by numerous ponds partially colonized by plants. The upper marshes grade inland into graminoid to treed fens, usually sustaining stunted tamaracks (Larix laricina) (Fig. 3.7c). The fens eventually acquire various organic landforms among such as string fens characterized by alternating ﬂark and ridges (Fig. 3.7d). In the discontinuous permafrost zone, treed palsa often develop in the middle of shallow ponds. The coasts with beach ridges have a substrate composed of Upper Pleistocene till or locally calcareous bedrock. In most cases, the Tyrrell Sea silty clays have been eroded during emergence. The ridges start developing in the upper part of the intertidal area and are actively growing at the shoreline (Figs. 3.8a, b). They are composed of cross-bedded sand and gravel with some capping plane beds. Eolian sand dunes are rare and where present they are generally small (order of a few meters

The cold-climate peatlands of the Hudson Bay Lowland, Canada

59

Figure 3.5. Satellite images of he southern pat of Hudson Bay Lowland. (a) Fens developed along the coast and in depressions occupied by rivers, and vast bogs developed in the interﬂuvial areas. (b) Winter scene of the peatlnds dissected by a few large rivers and numerous creeks. James Bay (JB) and Hudson Bay (HB) are ice covered except for local polynias.

Figure 3.6. Peatland features of Hudson Bay Lowland. (a) Inland stream bounded by spruce forest along the dry bank, local narrow swamps and vast fens farther away. (b) Line diagram of an air photograph’s compilation showing coastal areas with raised beach ridges, spits and a small stream with typical anastomosing structure near the coast grading into a meandering form inland (black areas ¼ water).

in height). The plant colonization of the ridges is slow and progressively consists of grasses, shrubs and then trees. Except for the most inland parts of the peatlands, hundred of kilometers from the coast where bogs cover the landscape, little to no peat forms on most of the high-beach ridges. They mostly remain forested with undergrowth of lichen (Fig. 3.8c). Typical mineral soils that form on them are regosols near the coast and progressively more mature podzols inland (Fig. 3.8d; Protz, 1982a, b; Protz et al., 1984; Cowell et al., 1991). Extensive bogs develop and cover areas farther inland. They are either open bogs with Sphagnum and scrub cover and have thick Sphagnum peat, or have some sparse stunted tree cover (Figs. 3.9a, b). An all-encompassing survey of the ﬂora of the HBL has been done by Riley (2003). In the southern part of HBL an overall progressive development of plants occurs,

60

I.P. Martini

Figure 3.7. Flats of the Hudson Bay Lowland. (a) Tidal ﬂats transitional landward into salt marshes and freshwater marshes. (b) Schematic diagram showing the internal structure and progressive plant colonization and paludiﬁcation of tidal ﬂats. (c) Wet interior fen with disseminated pool and local tree growths (tent for scale). (d) String fen.

Figure 3.8. Beach ridges of southern Hudson Bay Lowland. (a) Beach ridges developed in the upper part of intertidal zone and progressively colonized by vegetation as the land uplifts. (b) Schematic diagram showing costal development and internal structure of beach ridges; peat develops ﬁrst in swales. (c) Raised beach ridges covered by a spruce-lichen forest alternating with bogs and fens in the inter-ridge areas. (d) Progressive podzol formation on raised beach ridges with distance from shore (after Protz, 1982b).

The cold-climate peatlands of the Hudson Bay Lowland, Canada

61

Figure 3.9. Peatlands of southern Hudson Bay Lowland. (a) Raised, low shrub bog with local stunted spruce growth in the foreground. (b) Sphagnum peat proﬁle in a bog.

interpreted slightly differently by various authors (Fig. 3.10; Glooschenko and Martini, 1983; Cowell et al., 1991; Klinger and Short, 1996). The principal types of inorganic and organic soils that develop sequentially in the HBL from the coast inland vary depending on whether on they form in swales or in ridges. Gleysols develop primarily along the low-lying coasts and in freshwater marshes and on low-lying riverbanks. regosols, brunisols and, farther inland, podzols develop on drier, more elevated areas such as on beach ridges. The organic soils that form consist of mesisols and ﬁbrisols (Fig. 3.11a; Wickware et al., 1980; Protz, 1982a, b; Protz et al., 1984, 1988; Cowell et al., 1991). In the northern, continuous permafrost area of HBL the soil and peatland development trends are similar to those observed in the southern part of HBL that has discontinuous to sporadic permafrost (Fig. 3.11a; Tarnocai, 1982; Cowell et al., 1991). In the north, however, as the land emerges, the developing continuous permafrost fosters development of cryosols, peat plateaus (Figs. 3.11a, b) and palsa. Accordingly, Riley (1982) found that the same type of peatlands occur in the northern part as in the southern part of HBL, except that most of them become frozen inland (Fig. 3.11c). Only a slight increase in areal cover of treed bogs occurs in the subarctic zone. Because of postglacial rebound, the distance from the coast on the whole represents both a spatial as well as temporal gradient (Sims et al., 1982). When similar geomorphic fen settings are considered in the ﬁrst 40 km from the coast, about 2 m of peat deposits are formed in the southern part of HBL, and about 0.8 m in the northern part (Figs. 3.12a, b). Considering both fens and bogs of the southern part of HBL further, in the ﬁrst 40–70 km from shore, the measured peat thickness increases at an average rate of 3.5–4.0 cm km1 (Sims et al., 1982). The measured rate of

62

I.P. Martini

Figure 3.10. Possible successional pathways along a coast–inland transect in southern HBL from the shoe of James Bay to Kinosheo Lake inland (see Fig. 3.16a for location). Alternative pathways are indicated by dashed lines (from Klinger and Short, 1996; used with the permission of Arctic, Antarctic Alpine Research, University of Colorado).

increase of peat thickness is reduced between 40 and about 150 km from shore (age of basal peat about 4000–4500 14C yr BP), possibly partly due to increased decomposition and compaction of the peat (Fig. 3.12c). However, a site farther inland at about 200 km from the coast shows peat thickness of about 4.5 m with a basal age of about 5900 14C yr BP, suggesting an increase of measured peat thickness inland from the 150 km site. However, this may be a spurious trend because there is no assurance of having consistently sampled peats developed on a similar geomorphic position of the substrate. A slightly better correlation exists between peat thickness and age, with an overall average measured rate of about 0.6–0.65 mm yr1 (Fig. 3.12d). Considering the HBL as a whole, the growth-rate of peats varies between 0.1 mm yr1 in the north and 0.6–0.65 mm yr1 in the south, which compares favorably with rates of other similar subarctic to high boreal (such as in Sweden, Abisko: 0.1 mm yr1) and boreal to cold temperate zones (such as in Minnesota: 0.95 mm yr1), but it has orders of magnitude lower than those of warmer maritime and tropical areas (such as in Wales: 1–3 mm yr1, and in Borneo: 3–4 mm yr1). At a continental scale, thick peats in Canada are found in the Maritime Provinces such as Newfoundland and British Columbia. In central Canada, the thickest peats occur in inland parts of the high boreal to low subarctic zones of the unconﬁned HBL peatlands, and in conﬁned peatlands of central–northern Ontario just south of it (Figs. 3.13a, b). This is due to morphology of the substrate, rainfall (and equivalent Figure 3.11. Peatland features of HBL. (a) Schematic changes in wetland types and associated soils along coast–inland transects in southern and northern HBL. In the northern HBL transect the vertical lined pattern indicates permafrost and the mounds consist of either frozen beach ridges (those with indication of clasts) or newly formed features by frost (after Wickware et al., 1980; Cowell et al., 1991; see Fig. 3.12a for area of their study). (b) Peat mound/plateau in northern part of HBL. (c) Variation in peatland types along a north–south transect in HBL (from Riley, 1982; used with the permission of Le Naturaliste Canadien).

The cold-climate peatlands of the Hudson Bay Lowland, Canada

63

64

I.P. Martini

Figure 3.11. continued.

snowfall) and temperature. By contrast with tropical settings, rainfall does not need to be continuous in cold peatlands because temperature determines plant production during thawed annual periods and preservation of organic matter during freeze-up periods. Peat stratigraphy Palynology The overall lateral variations in plants and peat development in the HBL can also be observed in the stratigraphy of mature sites (Klinger and Short, 1996). For instance, an overall paludiﬁcation peat sequence has developed in the Kinosheo Lake area (Figs. 3.14a, b). This site is about 100 km inland; it has a 2.8 m thick peat with basal layers 4000 14C year old. The basal peats have developed in minerotrophic freshwater marsh–fen conditions with maximum herb percentage, low values of shrubs, and low but signiﬁcant Salix contributions; tree pollen is present particularly of Larix, Picea and Pinus. A middle part is relatable to forested interior fen transitional to a forested bog with a high percentage of arboreal pollen (Picea, Pinus and some Larix). The upper part is relatable to development of Sphagnum – Ericaceae bog, with low concentrations of arboreal pollen (Betula and Picea) (Klinger and Short, 1996; Kettles et al., 2000). Variations that have occurred in the bog are attributed by Kettles et al. (2000) primarily to variation in climate, although changes due to other factors, including variation in geochemical conditions due to changing plant communities may have had an inﬂuence. A few other palynological proﬁles have been analyzed from higher altitude lands from the Precambrian Sutton Ridges inliers within HBL, and from Detour Lake on the Precambrian Shield just south of HBL. These lands were ice- and water-free before the lowland, and their basal peat is about 3000 years older (Figs. 3.14a, 3.15a, b). The site on the Sutton Ridges is now at 145 m asl, surrounded by Sphagnum bog with stunted

The cold-climate peatlands of the Hudson Bay Lowland, Canada

65

Figure 3.12. Peat development. (a) Location map (villages on similarly named rivers AL: Fort Albany; AT: Attawapiskat; CH: Churchill; FS: Fort Severn; KA: Kaschechewan; MF: Moose Factory; MO: Moosonee; PE: Peawanuck (former Winisk, built after this was destroyed by a large spring ﬂood in 1986). (b) Distance from the coast versus peat depth measured in fens of the coastal zone (see locations of study in Fig. 3.12a) (after Wickware et al., 1980). (c) Distance from coast against known peat thickness and maximum ages measured in both fen and bogs (squares ¼ age; rhombi ¼ thickness). (d) Maximum age versus distance for the coast and thickness of peat (squares ¼ thickness; rhombi ¼ distance).

spruce and tamarack trees, and small fen areas in the hollows (McAndrews et al., 1982). The pollen stratigraphy most likely indicates ‘‘y that an initial spruce–pine forest in subzone 1a was succeeded by a birch–willow–sedge tundra in subzone 1b, followed by the modern spruce woodland zone in zone 2 – a classic warm-cold–warm climatic ﬂuctuation’’ (McAndrews et al., 1982, p. 607). A similar indication of an early, relatively high concentration of Picea and Pinus pollen has been observed in the Detour Lake bog proﬁle to the south, suggesting that early warm climate conditions (Hypsithermal interval) were followed by a middle to late Holocene colder ones when the peatlands expanded, and, as at Kinosheo Lake, evolved from fen into bog in the last 4000 years (Kettles et al., 2000). Organic mineralogy A single study has been made treating the peats of HBL and of southern parts of Ontario as possible precursors to coal (Hawke et al., 1999). The macerals (organic

66

I.P. Martini

The cold-climate peatlands of the Hudson Bay Lowland, Canada

67

compounds) have been determined in peat proﬁles of a bog in the Kinosheo Lake area and in an interior fen 25 km inland from the coast in southern HBL (Fig. 3.16a). Both proﬁles show a high-organic carbon content, very low sulfur, practically no ash in the bogs and variable but fairly high ash content in the fen, and low pH in the upper 1 m thick ombrotrophic portion of the Kinosheo Lake proﬁle (Fig. 3.16b). The bog proﬁle is 2.80 m thick and consists mainly of ﬁbric Sphagnum peat in the upper part grading downward into hemic woody-herbaceous peat with occasional wood (Figs. 3.16c, d). During the 1990 summer when the sampling was done, the surﬁcial Sphagnum peat was very wet, whereas the underlying part of the proﬁle was rather dry indicating the existence of a perched water table. The interior fen is 1.36 m thick and has a surﬁcial layer of ﬁbric Sphagnum grading downward into a predominant hemic woody-herbaceous peat. In both sites, the vegetal composition of the peats, its waterlogged conditions, and chemistry have led to a distinct stratigraphy and maceral distribution (Figs. 3.17a, b). Textinite is the most abundant maceral (expressed as volume percent) in both proﬁles, followed by huminite macerals, variable amount of liptinite, and minor inertinite (Figs. 3.17c, d, e, f). The concentration of the various macerals reﬂects the vegetal composition of the peat and its degree of decomposition. Although similar environmental setting may have existed in ancient peatlands that led to the development of cold-climate coals, such as the Permo–Carboniferous ones of Brazil, South Africa and Australia, the recent Boreal peats are not good analogs for two reasons. First and foremost the Permo–Carboniferous fern-dominated vegetation was very different from the angiosperm-dominated one of today. The latter was not yet fully evolved at those times. The second difference is related to the relatively high content of funginite and low content in inertinite in the modern cold peats. The former may be related to the low-accumulation rates that allow long persistence of peat at the surface and extensive fungi colonization (Hawke et al., 1999). The latter has to do with the low degree of decomposition (transformation) of the Boreal/Subarctic peats, particularly the Sphagnum peats. Indeed, the surﬁcial layers of the cold HBL peats are much less decomposed than those of the coldtemperate peats of southern Ontario where relatively recent increases in evapotranspiration have led to oxidation of the surﬁcial layers rather than deposition of new peat (Hawke et al., 1999). One possibility is that the HBL peats have not yet signiﬁcantly felt the effect of the global warming trend. This has led to low content of inertinite, contrary to what is found in ancient cold-climate coals. A possibility is that if the cold-climate Holocene peatlands and their peats could be used as analogs to understand the ancient coals, inertinite did not develop contemporaneously with the peat, but was perhaps over-imprinted at a later time. Figure 3.13. Organic deposits of Canada. (a) Map of organic soils and location of the Province of Ontario (after Glooschenko et al., 1993). (b) Generalized peat thickness in various parts of Canada related to air temperature and annual precipitation (1–3 ¼ peat thickness; labels such AlW indicate wetland region as in Fig. 3.1a, with the addition of the superscripts: W ¼ west, C ¼ central, E ¼ east. The north–south trend in Ontario is shown by a line. Peat thickness is also indicated of Newfoundland and British Columbia, respectively the westernmost and easternmost provinces of Canada) (from Martini and Glooschenko, 1985, based on information from Zoltai and Pollett, 1983). (Used with the permission of Earth-Science Reviews, Elsevier).

68

I.P. Martini

Figure 3.14. Kinosheo Lake. (a) Location on a schematic geological map (after Kettles et al., 2000). (b) Pollen diagram (from Kettles et al., 2000; Government of Canada publication).

The cold-climate peatlands of the Hudson Bay Lowland, Canada

69

Figure 3.15. Pollen diagrams from the areas adjacent to southern HBL. (a) Sutton Ridges inliers (from McAndrews et al., 1982; used with the permission of Le Naturaliste Canadien). (b) Detour Lake (from Kettles et al., 2000; Government of Canada publication) (see Fig. 3.14a for locations).

Pb Chemistry in peat Numerous studies have been done on the chemistry of peats and lake deposits south of HBL, particularly as they related to the acidiﬁcation of the environment. By contrast, few studies have been done in the HBL itself and near its borders, such as in the Detour Lake area (Fig. 3.14a). In the Detour Lake area, a 13 m deep core (to substratum) from a hollow, and a shallow (35 cm) core from a nearby Sphagnum hummock were recently examined by Bell and Kettles (2003) (Fig. 3.18a). In the

70

I.P. Martini

The cold-climate peatlands of the Hudson Bay Lowland, Canada

71

Figure 3.16. continued

former core, wood at the base of the peat was dated at about 7280 14C yr BP. Peat between 92 and 100 cm depth was dated at about 6800 14C yr BP, and peat between 30 and 40 cm at about 3200 14C yr BP. In the hummock core, peat from 33–35 cm depth was dated at about 180 14C yr BP, and that from the surface to about 22 cm depth was dated by 210Pb to have formed during 100 years before 1993 (Fig. 3.18a; Bell and Kettles, 2003). The vertical concentrations of selected element indicate their increase in the lower layers associated with minerotrophic (fen) peat, and, except for Ca, in the surﬁcial layers of the bog (Fig. 3.18b). The Pb is rather uniform throughout the long-hollow core with an increase in the top 20 cm. The 206 PB/204Pb and 206Pb/207Pb, however, show high values in the basal bog layers (ca. 60 cm depth), just above the fen peat, and a decreasing trend to the surface (Fig. 3.18c). In the shallow, Sphagnum hummock-core, Pb has a high value at a depth of about 10 cm corresponding to about the 1960s CE and a decrease to the surface. The 206Pb/204Pb:206Pb/207Pb show a persistent decreasing trend in the top 20 cm of the hummock. The systematic decrease in 206Pb/207Pb values since the late 1800s at Detour Lake closely matches that observed in the more populated Europe (Mackenzie et al., 1998; Shotyk et al., 1998; Dunlap et al., 1999; Weiss et al., 2002; Bell and Kettles, 2003). This suggests that the remote Canadian site has been affected by anthropogenic activities in part related to past lead–gasoline use and in part due to airborne contaminations from distant smelters. Variation in ancient times, both in Pb and Pbisotopes ratios and Sc (widely concentrated in soils) values (Fig. 3.18c), may have been associated with increased soil-derived aerosols during warmer, drier periods prone to increased forest ﬁre activities (Bell and Kettles, 2003). Figure 3.16. Organic mineralogy of southern HBL peat. (a) Location of cores collected along a coast–inland transect during the 1990 Northern Peatlands Study (from Klinger et al., 1994; used with the permission of American Geophysical Union, J. Geophys. Research, Atmosphere). (b) Variation with depth of selected chemical parameters (triangles: Kinsheo Lake; squares: interior fen) (after Hawke et al., 1999). (c) Kinosheo Lake bog. 0–23 cm depth: Sphagnum with intact structure (scale bar ¼ 50 mm). (d) Kinosheo Lake peat core, 220–235 cm depth: rootlet layer (scale bar ¼ 50 mm) (from Hawke et al., 1999; used with the permission of International Journal of Coal Geology, Elsevier).

72

I.P. Martini

Figure 3.17. Organic mineralogy of southern HBL peat as seen in reﬂected ﬂuorescent light; scale bar ¼ mm (after Hawke et al., 1999). (a,b) Macroscopic stratigraphy (right) and maceral distribution (left): Kinosheo Lake core (a) and Interior fen core (b). (c,d,e,f) Petrographic micrographs from Kinosheo Lake core: Sphagnum textinite, showing autoﬂuorescence between tissue types; 40–55 cm depth (c); Bright autoﬂuorescent textinite with corpohuminite; 85–100 cm depth (d); Texto-ulminite; 155–160 cm depth (e); Decay-resistant cutinite in a leaf tissue cross-section; 175–190 cm depth (f).

The cold-climate peatlands of the Hudson Bay Lowland, Canada

73

Figure 3.18. Detour Lake bog (from Bell and Kettles, 2003; Government of Canada publication). (a) Location map and cores taken (500-times vertically exaggerated cross-section). (b) Age and distribution of selected trace elements in peat cores. (c) Pb concentrations and ratios in peat cores. Calendar year for Hummock core are 210Pb dates. ‘‘Abundance of Pb determined on the o2 and o63 mm fractions and that of Sc on the o63 mm fraction’’ (from Bell and Kettles, 2003, p. 5; Government of Canada publication).

Peatlands and the carbon cycle The vast HBL peatland is now an enormous carbon sink that may become an increasingly signiﬁcant contributor of greenhouse gasses (CO2, CH4) to the atmosphere as global warming continues (Gorham, 1991; Tarnocai et al., 2005; Tarnocai, 2006). During 1990, a one summer-long multi-agency, multidisciplinary Northern

74

I.P. Martini

The cold-climate peatlands of the Hudson Bay Lowland, Canada

75

Figure 3.19. continued

Wetlands Study (NOWES) was conducted on the HBL, involving high- and lowaltitude ﬂights, high tower and chamber (Fig. 3.19a) measurements, in addition to studies of the peat deposits (Glooschenko et al., 1994). In southern HBL, detailed measurements were made along a 100 km-long transect from costal marshes to coastal and inland fens, to inland bogs in the Carling and Kinosheo lakes area (Figs. 3.16a, 3.19b). A gradational increase in peat thickness and age occurs along the transect. Considering chamber measurements on vegetated plots, Klinger et al. (1994) found a good correlation between CH4 ﬂux and aboveground plant productivity with average maximum values in the coastal fen (Fig. 3.19c). They also found that there is net intake of CO2 in the coastal fen, whereas a net loss was measured in the interior fen and the inland bogs. A quasi-regular increase in Figure 3.19. Carbon emission from southern HBL. (a) Proposed vegetation succession in swales along a coast–inland transect (see Fig. 3.16a) (from Klinger et al., 1994; used with the permission of American Geophysical Union, J. Geophys. Research, Atmosphere). (b) Chamber measurements of gas emission from bog at Kinosheo Lake. (c) Estimated, 1990 summer CH4 emission along the coast–inland transect, in relation to other environmental variable (from Klinger et al., 1994; used with the permission of American Geophysical Union, J. Geophys. Research, Atmosphere). (d) CH4 ﬂux along a local transect in the Kinosheo lake area during summer 1990, in relation to water table depth and soil temperature (from Klinger et al., 1994; used with the permission of American Geophysical Union, J. Geophys. Research, Atmosphere).

76

I.P. Martini

non-methane hydrocarbon (NMHC: isoprene, terpene and total HCs) ﬂux was instead found with distance from the coast. However, the actual ﬂux estimates are highly uncertain as they are differently affected by the various environmental parameters. For instance within the Kenosheo Lake area, 100 km from the coast, a 400 m-long transect show no peat development on the forested beach ridge that bounds the lake, to a 2.8 m thick peat succession developed during the last 4000 years (Fig. 3.19d). Along this transect the CH4 ﬂux is affected by vegetation cover, peat thickness and associated temperature and depth to water table. The NOWES study estimated the overall CH4 ﬂux to between 1.5 and 3 g m2 season1 for HBL. This was lower than expectation (Roulet et al., 1994). Various factors contributed to the result, among them, the low net primary productivity (NPP) of the area, the above ground one having been estimated between 100 and 400 g dry weight (gdw) m2 yr1, and total NPP ranging between 150 and 600 gdw m1 yr1, which is several times smaller than that found in most other productive systems in the World (Bolin et al., 1979; Klinger et al., 1994). It should further be considered that the cold-climate peatlands such as HBL can exchange gasses with the atmosphere only during the summer thawed season, and that emission or uptake (of CO2) strongly varies from year to year and within the same season due to change in depth of water table, extension of surface open water, and plant growth. Even if the information obtained from a single year is extensive, it must be considered with caution. For CH4, for instance, a ﬂux burst may occur during early stages of the spring–summer thaw as trapped gasses are released. Furthermore, changes in the water table may drastically affect the CH4 ﬂux. Such a ﬂux is greater when the water table is at or near the surface, and drops drastically with increase water depth to become locally negative when the water depth is greater than 0.5 m (Klinger et al., 1994). The overall precipitation in southern HBL during the summer of 1990 was similar to that of previous years, but rain was concentrated near the start of the season and was followed by rather dry conditions (Mortsch, 1994). This may have led to a drop in water table level, and thus to a slightly lower-than-normal estimated value of the measured low CH4 ﬂux for the season (Klinger et al., 1994; Moore et al., 1994; Valentine, 1994). Furthermore, peatland ponds have been demonstrated to be a major regulator for the CH4 and CO2 exchange with the atmosphere. Their CH4 ﬂux, for instance, is about 10 times greater than the surrounding vegetated peatland in fens, and 3 times greater in bogs. The ﬂux of the fen ponds has also been shown to be 2–20 times greater than that of the bog ponds (Hamilton et al., 1994). In 1990, the CO2 ﬂuxes were greatest from costal ponds and progressively lower in interior fens and lowest in bog areas. Furthermore, whereas in ponds the CO2 ﬂux was mostly toward the atmosphere, in vegetated areas it was mostly toward the ground (Hamilton et al., 1994; Roulet et al., 1994). Variation in ﬂuxes from the ponds may vary through a season as well because of the partial to total drying out and/or to growth of plants. Changes in peatlands wetness, and snow and vegetation cover through the 1995 spring–summer–fall season has been demonstrated by the change in backscatter measurements on RADARSAT satellites images in southmost HBL (Figs. 3.16a, 3.20a; Murphy et al., 2001). Measured backscatter has generally low values and allows a distinct separation between wetlands in May due to sharper variations in slow cover and ﬂattening of the

The cold-climate peatlands of the Hudson Bay Lowland, Canada

77

Figure 3.20. Peatlands of southmost HBL (from Murphy et al., 2001; used with the permission of the Canadian Aeronautics and Space Institute, Canadian Journal of Remote Sensing). (a) Satellite image with distribution of peatlands and other environments (A ¼ tidal ﬂats; B ¼ coastal marshes; C ¼ raised beach ridges; D ¼ bedrock promontory; E ¼ bog; F ¼ bog–fen transition; G ¼ shrub and treed fen; Landsat image: E-1374-15496, August 1, 1973). (b) Backscatter values of wetlands types for standard-mode 1997 RADARSAT images (S1–S7: incidence angles).

78

I.P. Martini

previous year’s vegetation (Fig. 3.20b). The backscatter values increase in June due to increased wetness related to melting of the active layer, to rain, and to new plant growths. Backscatter values decrease toward the end of the summer and in the fall due to increasing dryness and the ﬁrst snowfalls in November. The relationships between such observations and the CO2 and CH4 ﬂuxes has not been studied in detail. However, the ponds and shallow lakes cover about 8–12% of the total HBL, and, therefore, exert a disproportionate inﬂuence on overall CH4 ﬂux, releasing about 30% of the total (Hamilton et al., 1994). In any case, considering CO2 ﬂuxes, the extent of pond water being much smaller than that of vegetative areas, there is a net accumulation of C in the HBL peatlands. Kettles and Tarnocai (1999), Tarnocai (1998, 2006) Tarnocai and Stolbovoy (2006 – this book, Ch. 2) have estimated that the Canadian peatlands have an extension of about 1142.103 km2 containing a total organic carbon mass of about 150 Gt. They also estimated that a continuation of the present global warming trend could lead to melting part of the permafrost and a release about 40 Gt of carbon into the atmosphere from the Boreal–Subarctic peatlands, mostly from HBL. It is uncertain, however, how all this will affect the rates of change and overall atmospheric balance. As pointed out by Gorham (1991) and Janssens et al. (1992), as present peatlands dry out as the global warming proceeds, the amount of CO2 release to the atmosphere will increase, but that of CH4 will proportionally decrease. Furthermore, the possibility exists that the warming trends will simply shift the peatlands northward and some sort of balance could be achieved. In any case, the HBL has experienced extreme environmental changes throughout the geological ages. During the Pleistocene icehouse period, glaciers have repeatedly covered it, and it has experienced interglacial periods warmer than the present Holocene. It will continue to do so for the foreseeable future until the distribution patterns of continents and oceans, and the associated atmospheric and marine currents change sufﬁciently to force a return to a warm conditions of the early Pliocene–Eocene times (Fig. 3.21) and perhaps to a hothouse period similar to what existed on Earth during the Mesozoic era.

Resources A few thousand people, mostly Cree and Metis permanently inhabit HBL in ﬁve villages (less than 1000 inhabitants each) and three small towns one (Churchill) at the northern end and the other two (Moosonee and Moose Factory) located across the Moose River at the southern end (Figs. 3.12a, 3.22). There are no roads to these centers. They can be reached by air, and Churchill and Moosonee can be reached by rail as well. Because James Bay is very shallow, Churchill has the only harbor for seagoing ships in the area. The harbor is open only during the short ice-free summer period, and has been used to export mainly cereals from the Canada interior. Most villages have a subsistence economy. Hunting and ﬁshing is still practiced. Tourism is a developing industry particularly in Churchill where polar bears and beluga whales can be seen during certain times of the year, and in Moosonee and Moose Factory, rich in history. European explorers made contact with natives in the

The cold-climate peatlands of the Hudson Bay Lowland, Canada

79

Figure 3.21. Generalized shift of biozones associated with climatic changes and glaciations of North America (after Hare, 1976).

early 1600s, and early in the third quarter of that century, the Hudson Bay Company established trading posts in the areas of Moose Factory, Fort Albany, Fort Severn and at the mouth of the Nelson River, south of Churchill (Fig. 3.12a). The HBL provided furs to export. At present, natives still hunt animals for fur to a limited extent, and also take caribou in the northern part of the lowland, and a large number of migratory geese are still hunted by natives and others everywhere along the coast (Thomas and Prevett, 1982). Hundreds of thousands of migratory geese, mostly Lesser Snow Geese (Chen caerulescens caerulescens) nest or stage along the coastal marshes of Hudson Bay and James Bay. Two major colonies are located just east of Churchill and near the promontory between the two bays, but smaller colonies occur

80

I.P. Martini

Figure 3.22. Localities and people of the Hudson Bay Lowland. (a) Touristy Churchill. (b) Cree dance, Moose Factory. (c) Fort Albany of 1886 (Photo by J.B. Tyrrell, 1886, Geological Survey of Canada. http:// atlas.gc.ca/site/english/maps/historical/territorialevolution/photo1886.jpg/image_view). (d) Modern village of Fort Albany seen during a winter ice jam.

all along the coast including Akimiski Island (Fig. 3.12a; Thomas and Prevett, 1982; Abraham et al., 1999). The large number of geese in these colonies, and their habit of eating roots in the late summer in preparation for the southward migration, lead to modiﬁcation of some marshes and enlargement of pools (Kotanen and Jefferies, 1997). The large Canada Geese (Branta canadensis) utilize instead the interior ponds of fens and, to a lesser extent, of bogs (Lumsden, 1957). The coasts, particularly where beach ridges are well developed such as in the Polar Bear Provincial park (Fig. 3.12a), are used during the summer by polar bears from Churchill down to Akimiski Island, and interior peatlands areas south of Churchill as winter maternity dens (Prevett and Kolenosky, 1982; Ramsay and Stirling, 1982). This is the southernmost polar bear population in the world, and recently it has started to show some stress (lower body weight of individuals) associated to reduced ability to feed because ice in the bays melts earlier in the summer and reforms later in the fall due to increased global warming. Polar bears need ice cover and ice ﬂoes in the bays to hunt seals, their staple food. As for other natural resources, HBL is not suitable for exploitation of peat or forestry because it is too wet and distant from commercial centers. Tests have been conducted for petroleum and show little promise due to lack of adequate traps in the

The cold-climate peatlands of the Hudson Bay Lowland, Canada

81

Paleozoic substrate (Johnson et al., 1986). Limited geological resources not yet exploited, such as kaolinite, quartz sand, evaporites and lignite, occur in the bedrock of the southernmost part of HBL. Recently, diamond-bearing kimberlite pipes have been discovered and evaluated about 100 km inland from Attawapiskat in the rock substrate of the peatlands (Fig. 3.12a; Telford and Long, 1986; Stott, 2002; Belleﬂeur et al., 2005). One of the most promising pipes, owned by DeBeers, is estimated to have 29-million tonnes of ore containing 6.3-million carats of high-quality diamonds. Production from this pipe will start within the decade. From all this it follows that the vast peatlands of the HBL do not provide great direct-economic beneﬁt to modern society. As a consequence little or just token attention is paid to it by government and research-granting agencies and much remains unknown about the area. Yet the HBL is located in a critical border area that is potentially greatly affected by climatic change and has an enormous regulatory potential for the gaseous composition of the atmosphere. Furthermore, the interior and the coastal areas of the HBL are of a global signiﬁcance as a major staging and nesting area for birds (geese, ducks, shorebirds) that migrate over North and South America. HBL still has a thriving population of polar bears, beluga whales, and some rivers and lakes are still rich in ﬁsh. The environment is largely pristine and for this reason is potentially a good, if expensive, ecotourism destination. If properly managed with, the few resources and tourism may beneﬁt the local inhabitants by providing novel opportunities to young and future generations.

References Abraham, K.F., Leaﬂoor, J.O., and Lumsden, H. G., 1999. Establishment and growth of the Lesser Snow Goose, Chen caerulescens caerulescens, nesting colony on Akimiski Island, James Bay, Northwest Territories. Can. Field Naturalist, 113, 245–250. Atlas of climatic maps, Canada, 1967. Map Series 1–10. Department of Transport, Meteorological Branch, Canada. Bell, K., and Kettles, I.M., 2003. Lead-isotope ratio measurements on hummock and hollow peat from Detour Lake area, Ontario. Geological Survey of Canada, Current Research 2003-C3, 12pp. Belleﬂeur, G., Matthews, L., Toberts, B., et al., 2005. Downhole seismic imaging of the Victor kimberlite, James Bay lowlands, Ontario: a feasibility study. Curent Research 1005-C1. Geological Survey of Canada, 7pp. Bolin, B., Degens, E.T., Duvigneaud, P., and Kempe, S., 1979. The global biochemical carbon cycle. In: Bolin, B., Degens, E.T., Kempe, S., and Ketner, P. (Eds), The Global Carbon Cycle. Wiley, New York. Chapman, L.T. and Thomas, M.K., 1968. The climate of northern Ontario. Met. Branch, Dept. Transport, Climatological Studies, Toronto, 6, 58pp. Cowell, D.W., Wickware, G.M., and Sims, R.A., 1991. Organic and mineral soils of the southwestern James Bay coastal zone in relation to landform and vegetation physiognomy. Canada–Ontario Forest Resources Development Agreement Report 3308, 40pp. (Obtainable form Forestry Canada, Ontario region, Sault Ste. Marie, Ontario, Canada). Dunlap, C.E., Steinnes, E., and Flegal, A.R., 1999. A synthesis of lead isotopes in two millenia of European air. Earth Planet. Sci. Lett. 167, 81–88. Glooschenko, W.A. and Martini, I.P., 1983. Wetlands of the Attawapiskat River mouth, James Bay, Ontario, Canada. Wetlands 3, 64–76. Glooschenko, W.A., Roulet, N.T., Barrie, L.A., et al., 1994. The northern wetlands study (NOWES): An overview. J. Geophys. Res. 99, 1423–1428.

82

I.P. Martini

Glooschenko, W.A., Tarnocai, C., Zoltai, S., and Glooschenko, V., 1993. Wetlands of Canada and Greenland. In: Whigham, D.F., Dykyjova, D., and Hejny, S. (Eds), Wetlands of the World I. Kluwer Academic Publishers, Amsterdam, pp. 415–514. Gorham, E., 1991. Northern peatlands: role in the carbon cycle and probable responses to climatic warming. Ecol. Appl. 1, 182–195. Hamilton, J.D., Kelly, C.A., Rudd, J.W.M., et al., 1994. Flux to the atmosphere of CH4 and CO2 from wetland ponds on eh Hudson Bay lowlands (HBLs). J. Geophys. Res. 99, 1495–1510. Hare, F.K., 1976. Late Pleistocene and Holocene climates: some persistent problems. Quatern. Res. 6, 507–517. Hawke, M.I., Martini, I.P., and Stasiuk, L.D., 1999. A comparison of temperate and boreal peats from Ontario, Canada: possible modern analogues for coals. Int. J. Coal Geol. 41, 213–238. Janssens, J.A., Hansen, B.C.S., Glaser, P.H., and Whitlock, C., 1992. Development of a raised-bog complexes in northern Minnesota. In: Wright, H.E. Jr, Cofﬁn, B., and Aaseng, N.E. (Eds), The patterned peatlands of Minnesota. University of Minnesota Press, Minneapolis, pp. 189–204. Johnson, R.D., Joubin, F.R., Nelson, S.J., and Olsen, E., 1986. Mineral resources. In: Martini, I.P. (Ed.), Canadian Inland Seas. Elsevier, Amsterdam, pp. 387–402. Kettles, I.M., Garneau, M., and Jette´, H., 2000. Macrofossil, pollen, and geochemical records of peatlands in the Kinosheo Lake and Detour Lake areas, Northern Ontario. Geological Survey of Canada, Bulletin 545, 24pp. Kettles, I.M. and Tarnocai, C., 1999. Development of a model for estimating the sensitivity of Canadian peatlands to climate warming. Ge´ographie physique et Quaternaire 53, 323–338. Klinger, L.F. and Short, S.K., 1996. Succession in the Hudson Bay Lowland, Northern Ontario, Canada. Arctic Alpine Res. 28, 172–183. Klinger, L.F., Zimmerman, P.R., Greenberg, J.P., et al., 1994. Carbon trace gas ﬂuxes along a successional gradient in the Hudson Bay lowland. J. Geophys. Res. 99, 1469–1494. Kotanen, P.M. and Jefferies, R.L., 1997. Long-term destruction of wetland vegetation by Lesser Snow Geese. E´coscience 4, 1895–1898. Lumsden, H.G., 1957. A snow goose breeding colony in Ontario. Can. Field Naturalist 71, 153–154. Mackenzie, A.B., Logan, E.M., Cool, G.T., et al., 1998. Distribution, inventories and isotopic composition of lead in 210Pb-dated peat cores from contrasting biochemical environments: implications for lead mobility. Sci. Tot. Environ. 223, 25–35. Martini, I.P. and Glooschenko, W., 1985. Cold climate peat formation in Canada and its relevance to Lower Permian coal measures of Australia. Earth-Sci. Rev. 22, 107–140. McAndrews, J.H., Riley, J.L., and Davis, A.M., 1982. Vegetation history of he Hudson Bay Lowland: a postglacial pollen diagram from the Sutton Ridge. Natural. Canad. 109, 597–608. Moore, T.R., Heyes, A., and Roulet, N.T., 1994. Methane emission from wetlands, southern Hudson Bay lowland. J. Geophys. Res. 99, 1455–1467. Mortsch, L.D., 1994. Assessment of the temperature and precipitation of 1990 during the Northern Wetlands Study (NOWES). J. Geophys. Res. 99, 1420–1438. Murphy, M.A., Martini, I.P., and Protz, R., 2001. Seasonal changes in subarctic wetlands and river ice breakup detectable on RADARSAT images, southern Hudson Bay Lowland, Ontario, Canada. Can. J. Remote Sensing 27, 143–158. National Wetlands Working Group, 1988. Wetlands of Canada. Ecological Land Classiﬁcation Series No. 24. Sustainable Development Branch. Environment Canada, Ottawa, Ontario, and Polyscience Publications Inc., Montreal, Quebec. 452pp. Neishstadt, M.I., 1977. The world’s largest peat basin, its commercial potentialities and protection. Int. Peat Soc. Bull. 8, 37–43. Norris, A.W., 1986. Review of Hudson Platform Paleozoic stratigraphy and biostratigraphy. In: Martini, I.P. (Ed.), Canadian Inland Seas. Elsevier, Amsterdam, pp. 17–42. Prest, V.K., 1970. Quaternary geology of Canada. In: Douglas, R.J.W. (Ed.), Geology and economic minerals of Canada. Geological Survey of Canada, Economic Geology, Report 1, pp. 675–764. Prevett, J.P. and Kolenosky, G.B., 1982. The status of polar bears in Ontario. Natural. Canad. 109, 933–939. Protz, R., 1982a. Development of gleysolic soils in the Hudson and James bay coastal zone, Ontario. Natural. Canad. 109, 491–500.

The cold-climate peatlands of the Hudson Bay Lowland, Canada

83

Protz, R., 1982b. Development of podzolic soils in the Hudson and James bay lowlands, Ontario. Natural. Canad. 109, 501–510. Protz, R., Ross, G.J., Martini, I.P., and Terasme, J., 1984. Rate of podzolic soil formation near Hudson Bay, Ontario. Am. J. Soil Sci. 64, 31–49. Protz, R., Ross, G.J., Shipitalo, M.J., and Terasme, J., 1988. Podzolic soil development in the southern James Bay Lowlands, Ontario. Can. J. Soil Sci. 68, 287–305. Ramsay, M.A. and Stirling, I., 1982. Reproductive biology and ecology of female polar bears in western Hudson Bay. Naturaliste Canadien 109, 941–946. Riley, J.L., 1982. Hudson Bay Lowland ﬂoristic inventory: wetlands catalogue and conservation strategy. Natural. Canad. 109, 543–555. Riley, J.L., 2003. Flora of the Hudson Bay Lowland and its Postglacial Origin. National Research Council of Vanada Press, Ottawa, 236pp. Roulet, N.T., Jano, A., Kelly, C.A., et al., 1994. Role of the Hudson Bay lowland as a source of atmospheric methane. J. Geophys. Res. 99, 1439–1454. Shilts, W.W., 1986. Glaciation of the Hudson Bay region. In: Martini, I.P. (Ed.), Canadian Inland Sseas. Elsevier, Amsterdam, pp. 54–78. Shotyk, W., Weiss, D., Appleby, P.G., et al., 1998. History of atmospheric lead deposition since 12,370 14 C yr BP from a peat bog, Jura Mountains, Switzerland. Science 281, 1635–1640. Sims, R.A., Cowell, D.W., and Wickware, G.M., 1982. Classiﬁcation of fens near southern James Bay, Ontario, using vegetational physiognomy. J. Bot. 60, 2608–2623. Sjo¨rs, H., 1959. Bogs and fens in the Hudson Bay lowlands. Arctic 12, 3–19. Sjo¨rs, H., 1963. Bogs and fens on Attawapiskat River, northern Ontario. Nat. Mus. Can. Bull. 186, 45–133. Stott, G.M., 2002. Diabase dyke swarms as structural controls for kimberlite pipes under the James Bay and Hudson Bay lowlands, Ontario. Precambrian Geoscience Section, Ontario Geological Survey. Sudbury, Ontario, Canada (also seen during 2005 In: http://64.233.161.104/search?q= cache:vRVMhO13q98J:www.mndm.gov.on.ca/mndm/mines/ogs/posters/OEGS_2003/ Stott_opa_2003_panel1.pdf+kimberlite+pipes+under+the+James+Bay&hl=en). Tarnocai, C., 2006. The effect of climate change on carbon in Canadian peatlands. Global Planet. Change 53, 222–232. Tarnocai, C., 1982. Soil and terrain development in the York Factory peninsula, Hudson Bay lowland. Natural. Canad. 109, 511–522. Tarnocai, C., 1998. The amount of organic carbon in various soil orders and ecological provinces in Canada. In: Lal, R., Kimble, J.M., Follett, R.L.F., and Stewart, B.A. (Eds), Soil processes and the carbon cycle. Advances in Soil Science, CRC Press, New York, pp. 81–92. Tarnocai, C., Kettles, I.M., Lacelle, B., 2005. Peatlands of Canada Database. Research Branch, Agriculture and Agri-Food Canada, Ottawa, Ontario, Canada (digital database). Tarnocai, C. and Stolbovoy, V., 2006 (this book, Chapter 2). Northern peatlands: their characteristics, development and sensitivity to climate change. In: Martini, I.P., Matı´ nez Cortizas, A., and Chesworth, W. (Eds.), Peatlands: Evolution and Records of Environmental and Climatic Changes. Elsevier, Amsterdam. Telford, P.G. and Long, D.G.F., 1986. Mesozoic geology of the Hudson Platform. In: Martini, I.P. (Ed.), Canadian Inland Seas. Elsevier, Amsterdam, pp. 43–53. Thomas, V.G. and Prevett, J.P., 1982. The roles of the James and Hudson Bay Lowland in the annual cycle of geese. Natural. Canad. 109, 913–925. Valentine, D.W., 1994. Ecosystem and physiological controls over methane production in northern wetlands. J. Geophys. Res. 99, 1563–1571. Webber, P.J., Richardson, J.W., and Andrews, J.T., 1970. Post-glacial uplift and substrate age at Cape Henrietta Maria, southeastern Hudson Bay, Canada. Can. J. Earth Sci. 7, 317–325. Weiss, D., Shotyk, W., Boyle, E.A., et al., 2002. Comparative study of the temporal evolution of atmospheric lead deposition in Scotland and central Canada using blanket peat bogs. Sci. Tot. Environ. 292, 7–18. Wickware, G.M., Cowell, D.W., and Sims, R.A., 1980. Peat resources in he Hudson Bay Lowland coastal zone. In: Proceedings of the 6th International Peat Congress, Duluth, MN, August 18–21, 1980, pp. 138–143.

84

I.P. Martini

Zoltai, S.C., 1980. An outline of the wetland regions of Canada. In: Rubec, C.D.A., and Pollett, F.C. (Eds), Proceedings of a workshop on Canadian wetlands. Environment Canada, Lands Directorate, Ecological land Classiﬁcation, Series, No. 12, pp. 1–8. [Trace: Minister of Supply and Services Canada, 1980. Cat. No. En 73-3/12; ISBN 0-662-50919-6]. Zoltai, S.C. and Pollett, F.C., 1983. Wetlands in Canada: Their classiﬁcation, distribution, and use. In: Gore, A.J.P. (Ed.), Mires: Swamp, Bog, Fen, and Moor. Regional studies. Elsevier, Amsterdam, Vol. B, pp. 245–268.

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

85

Chapter 4

Mountain mires from Galicia (NW Spain) X. Pontevedra-Pombal, J.C. No´voa Mun˜oz, E. Garcı´ a-Rodeja and A. Martı´ nez Cortizas

Introduction Galicia, on northwestern Iberian Peninsula (411 480 –431 480 N latitude, 61 420 –91 180 W longitude), is a coastal area in a transition zone between temperate and subtropical climates, and where mires are relatively abundant (Fig. 4.1). The most recent estimates indicate that Galicia has some 10,000 ha of mountain mires, which represents a 0.4% of the territory (Pontevedra-Pombal and Martı´ nez Cortizas, 2001; Pontevedra-Pombal, 2002). The Galician mires have been studied since the ﬁrst decades of the 20th century, mostly for palynology (Casares Gil, 1920; Jato, 1974; Ramil, 1992; Mun˜oz Sobrino, 1996). Recently, research has been focused on a more general approach to understand the ecosystems. In the last 10 years, a large number of mires were systematically characterized for their physicochemical properties (Pontevedra-Pombal et al., 1996, 2001b), organic chemistry and dynamics of the peat composition (Pontevedra-Pombal et al., 2001a, in press), understanding the whole ecosystem development (Martı´ nez Cortizas et al., 2001; Pontevedra-Pombal, 2002), and interpreting the environmental signals preserved in the peat, which provide information on the composition of the atmosphere as well as about climatic and human interference during the Holocene (Martı´ nez Cortizas et al., 1999, 2002, 2004). Mapping of the mires has also constituted an essential tool for the evaluation of their scientiﬁc potential, as well as a basic element for decision making related to protection issues (Pontevedra-Pombal, 2002; Pontevedra-Pombal et al., 2003).

Mires development Distribution and biogeographical conditions The location of the principal mountain mires of Galicia (NW Spain) is shown in Figure 4.2. These mires occur in three major areas: northern, coastal and pre-coastal ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09004-3

86

X. Pontevedra-Pombal et al.

Figure 4.1. Simpliﬁed map of Galicia, NW Spain.

Figure 4.2. Galician mountains: main areas and peatland coverage (gray areas).

Mountain mires from Galicia (NW Spain) Table 4.1. Galicia.

87

Environmental characteristics of the three main spatial units of the mires of

Environmental parameters

Peat mechanism

Coastal mountains and Dorsal Galega

Eastern mountains

Northern mountains

High precipitation Moderate average temperature Abundant ﬂuvial network Moderate freeze–thaw processes Moderate oceanity

High precipitation Low average temperature Abundant ﬂuvial network Strong glacial and periglacial paleoprocesses Continentality

High precipitation Low average temperature Very abundant ﬂuvial network Intensive glacial and periglacial paleoprocesses Strong oceanity

Paludiﬁcation

Terrestrialization

Paludiﬁcation; Terrestrialization

Table 4.2. Climatic conditions of the mires main areas of Galicia (data from Martı´ nez Cortizas and Pe´rez Alberti, 2000). Area

Precipitation

Temperature

Alt

Win

Spri

Sum

Aut

An

Win

Spri

Sum

Aut

An

North

600 800 1000

492 561 635

275 314 355

150 172 194

477 545 616

1394 1592 1800

6.3 5.1 3.9

9.9 8.4 6.9

14.6 13.2 11.7

9.1 7.9 6.7

10.0 8.7 7.4

East

1000 1200 1400

480 531 582

291 322 353

139 154 168

490 594 697

1400 1600 1695

4.2 2.3 0.3

10.3 8.2 6.1

16.0 14.2 12.4

6.8 4.9 3.1

9.4 7.4 5.5

Southeast

1000 1400 1800

428 536 644

256 320 384

120 151 181

438 547 657

1242 1554 1866

4.2 1.9 0.4

10.6 8.1 5.6

16.8 14.7 12.6

6.9 5.3 3.7

9.6 7.5 5.3

Notes: Alt, altitude; Win, winter; Spr, Spring; Sum, summer; Aut, autumn; An, annual. Precipitation in mm; temperature in 1C.

(mountains that separate the littoral area to the continental area), and eastern areas. Environmental and paleoenvironmental conditions in these three areas have conditioned the kind and intensity of processes associated to the formation of mires (Table 4.1). In peatland areas, annual precipitation ranges from 1200 mm in the southeastern areas (at 1000 m asl) to 1800–1900 mm in the most elevated ranges of the northern and eastern areas (Table 4.2). Mean annual temperatures vary between 5.3 and

X. Pontevedra-Pombal et al.

88

Table 4.3. Rainfall and temperature mean gradients and constants in the north and east of Galicia (data from Martı´ nez Cortizas and Pe´rez Alberti, 2000). Area

Temperature (oC)

Rainfall (mm)

Gradient (mm/100 m altitude) Constant Gradient (mm/100 m altitude) Constant North 100 East 75

795 647

0.67 0.50

14.1 14.2

10.3 1C. The oceanic character of the region is affected by the winter–summer equilibrium between temperate and subtropical components; the later is responsible for the rainfall seasonality (Table 4.3), which is more pronounced to the south (Martı´ nez Cortizas and Pe´rez Alberti, 2000). Present distribution of mires is the result of the interaction between several factors at different spatial and temporal scales, and their formation must be understood in paleoenvironmental terms. Upper Pleistocene–Holocene climate changes, a complex relief in terms of forms and spatial structure, and the predominance of acid rocks have generated diverse ecological conditions that favored the development of this kind of wetlands ecosystems. Geomorphologic criteria are essential to understand the development of Galician mires. Their location and typology is to a great extent linked to relief forms, generated under glacial and periglacial climatic conditions during the last phases of the Wu¨rm glaciation. In the northern coastal areas cold-climate processes regularized mountain summits, creating ﬂat surfaces where paludiﬁcation processes began as soon as the climate improved. In this area, due to the great quantity of rain, mires developed quickly and extended in all directions forming blanket mire macrotopes. On the other hand, in the old glaciated areas of the eastern and southeastern areas, erosive as well as sedimentary forms generated small basins (ponds) where terrestrialization prevailed in the development of mires. In terms of vegetation, Fraga Vila et al. (2001) elaborated a list of 182 species present in NW Spain mountain mires. Of these, 133 are vascular plants, 46 are bryophytes (11 of the genus Sphagnum) and 3 lichens. The greatest plant diversity was found in minerotrophic mires, whereas ombrotrophic mires showed a much lower biodiversity. Species with large ecological amplitudes, like Molinia caerulea and Festuca rubra, are represented in all types of mires, although with different abundance. On the contrary, some plants are speciﬁc of one area, as for example Carex duriaeui and Erica mackaiana, which are characteristic of the northern area, or Carex nigra and Erica tetralix in the eastern area. Typology Using a hierarchic classiﬁcation system of mires (Fig. 4.3), three main types of mires occur in these mountainous areas: blanket bogs (ombrogenic mires), fens (minerogenic mires) and raised bogs. These are comparable to those established for other Atlantic areas in Europe (Moore and Bellamy, 1974; Davis and Anderson, 1991; Lindsay, 1995).

Mountain mires from Galicia (NW Spain)

89

Figure 4.3. Hierarchic classiﬁcation system of Galicia mountain mires.

Ombrogenic mires are limited to the northern area, where they occur directly over acid rocks, periglacial stony deposits, weakly evolved soils or rarely over podzolic soils. Fens are well represented in all mountainous areas. The oldest fens are located in the eastern areas on small glacially excavated areas, and some date back to 17,000 14 C yr BP. In the northern area, fens occupy large depressions, up to a meter deep and 55 ha wide, formed by weathering and erosion of granites. Locally some fens have evolved into raised bogs. Owing to their high environmental value, the European Union (EU) included mire ecosystems in the Habitats directive and in the Natura 2000 framework, as ecosystems of preferential conservation (European Commission, 1996). Peatlands from Galicia can be included in the group RAISED BOGS AND MIRES AND FENS, Sphagnum acid bogs, and speciﬁcally in the habitats deﬁned as: 7110-Active raised bogs; 7120-Degraded raised bogs still capable of natural regeneration; 7130-Blanket bog, active only; and 7140-Transition mires and quaking bogs.

90

X. Pontevedra-Pombal et al.

Figure 4.4. Location of blanket bogs in Europe (black areas; after Raeymaekers, 1999).

Figure 4.5. Active blanket mire complex of Barreiras do Lago (Serras Septentrionais, Galicia).

Raeymaekers (1996) limited the distribution of blanket mires in Europe to western Ireland, Scotland, central Sweden and French Bretagne (Fig. 4.4). Our work has revealed the presence of this type of mire in Galicia as well (Figs. 4.4, 4.5; Pontevedra-Pombal, 2002). Furthermore, Raeymaekers (1996) reported the presence of raised bogs from the British Isles, Sweden, Finland, from elevated mountains of France, Switzerland and Italy, and in The Netherlands (Fig. 4.6). This type also

Mountain mires from Galicia (NW Spain)

91

Figure 4.6. Location of raised bogs in Europe (black areas; after Raeymaekers, 1999).

occurs in Galicia, although it is not the most abundant peatland, and shows a characteristic dome at the center.The last type, transition mires and quaking bogs, is the most abundant. We have identiﬁed both subtypes: those developed on slopes (transition mires) and those developed on depressions (quaking bogs) such as glacial or ﬂuvial valleys, weathering basins, over-excavated areas and depressions between moraines. From a geochemical and hydrological point of view, the ﬁrst subtype represents a gradation between minerogenic and ombrogenic conditions. The second subtype is usually hydrologically connected to the slope areas, so it grades into transition mires or aerobic wet soils rich in organic matter. This spatial connection is responsible for both the morphology and stratigraphy of these mires.

Chronology Data from 30 representative mires indicate that peat accumulation in fens of the eastern and southeastern areas began at least 11,000–10,000 14C yr BP by terrestrialization (Fig. 4.7). These are the areas most intensely glaciated during the latest Pleistocene (Valca´rcel, 1999). Radiocarbon dates of 17,300–17,400 14C yr BP of the basal peats at Lagoa de Lucenza–Serra do Caurel indicate a rapid onset of peat formation after the Last Glacial Maximum (20,000–18,000 14C yr BP). However, the most intensive fen formation occurred during two main episodes, one around 5000–4000 14C yr BP, and the other between 3000 and 2000 14C yr BP. The formation of blanket mires started 9000–8000 14C yr BP, and an intense period of bog formation occurred between 6000 and 5000 14C yr BP.

92

X. Pontevedra-Pombal et al.

Figure 4.7. Initial phases of peat formation in peatlands (mires) of different areas of Galicia (age in 14C yr BP).

The formation and expansion of Galician mires is related to climatic and soil evolution as well to human activities since prehistoric times. Many authors have indicated the importance of global climatic oscillations on the development of mires in Europe (Ratcliffe, 1977; Solem, 1986; Averdieck et al., 1993). Malmer (1975) reported that a vast region of United Kingdom is constituted by blanket bogs developed by paludiﬁcation after the sub-Boreal period, some 3000–2500 14C yr BP, probably as a consequence of the climate degradation during the sub-Boreal/sub-Atlantic transition, which resulted in an elevation of the water table in most mineral soils of western Europe. On a larger scale, this degradation also resulted in a generalized paludiﬁcation of the highlands with the development of iron pans and the accumulation of organic matter. In pedogenetic terms, blanket mires are considered as the ﬁnal stage of soil maturation in areas with high precipitation, thus their formation is essentially pedogenetic, and associated with the intense leaching of redox-dependent elements. In Great Britain, blanket mires are related to podzolization processes responsible for the formation of iron crusts and impeded water movement (Godwin, 1946). Although Dimbleby (1985) indicated that mires rarely develop over podzolic soils, some Galician mires have an accumulation of iron oxy-hydroxides at their base, at the contact with the sediment or rock. The studies about polycyclic soils near peatlands area have identiﬁed important extensions of accumulation of iron oxy-hydroxides layers. In some areas, these layers have generated an intense paludiﬁcation and stimulated the formation of peatlands. Pollen records obtained in Galician mires show a forest regression phase, coincident with the accumulation of inorganic layers in some fens (To¨rnqvist and

Mountain mires from Galicia (NW Spain)

93

Joosten, 1988; Ramil, 1992). Some studies on soil erosion, archeological records and landscape evolution during the Holocene also provide good evidence that these erosion episodes are coeval with the development of different human cultures in the area (Epipaleolithic, Neolithic, Bronze Age, Iron Age) (Benito et al., 1991; Martı´ nez Cortizas and Moares, 1995). So it is possible that human-induced erosion may have promoted increased water runoff in slope areas and elevated water tables in the lowlands, as also proposed in other areas (Moore, 1975; Chambers, 1988). The high charcoal content in the basal soils of some mire supports this interpretation. Growth and accumulation rates Average maximum depth of minerotrophic mires in Galicia is 2–3 m, although in the eastern and southeastern areas it may reach up to 7 m, but in the latter not all layers are organic. Ombrotrophic mires have depths of up to more than 5 m, with 3 m being a representative depth. This depth is greater than the 2.0–2.5 m reported for boreal and subarctic European and North America mires (Riley, 1987; Gorham, 1991). Age–depth relationships are shown in Figure 4.8. Most ages of Galician mires ﬁt to a linear function with a slope of 21 (n ¼ 55, correlation coefﬁcient r ¼ 0.98). This

Figure 4.8. Peat age–depth relationship of Galician mires. The points represent age of basal peats from different mires, and the black and gray lines indicate linear functions through two groupings of mires (age in 14C yr BP).

94

X. Pontevedra-Pombal et al.

indicates that average peat accumulation during the Holocene was 1 cm every 20–25 years. Some radiocarbon ages of different mires also ﬁt to a linear function with a similar slope (b ¼ 23; n ¼ 15, r ¼ 0.88), but shifted some 4000 14C yr BP indicating the presence of a discontinuity or a change in peat accumulation. The almost identical slope indicates that peat accumulation rates were the same for all mires, but the shift (different ages for the same depth) suggests the second set of mires were affected by erosion processes or strong change of decomposition of the organic matter, which eliminated younger peat sections, or there was a stop or dramatic reduction in peat accumulation at some time. The latter is unlikely because a gap of 4000 years seems too large to be explained by changes in peat accumulation. Thus, this may correspond to a global episode since it is observed in mires from different areas, and, in many cases, coincides with erosive features, decay change of the peat and strong peat compaction. It also is similar to the Grenzhorizont or Boundary Horizon reported by different authors, a highly decomposed black peat layer below very low decomposed brown peat (Weber, 1908; Godwin, 1981). This boundary horizon has been identiﬁed in many areas in continental Europe and in UK, and it was interpreted as a reaction of mire ecosystems to a global climate change. Granlund (1932) and Ma¨kila¨ (1997) reported at least ﬁve such events/horizons with ages of 4300, 3200, 1600 and 800 14C yr BP. Peat growth rates in Galician mires vary between 0.2 and 0.7 mm yr1, with average values of 0.45–0.47 mm yr1 (Fig. 4.9). These rates are comparable to those reported for Europe and North America. Gorham (1991) suggested a value of 0.5 mm yr1 as a reasonable conservative estimation for mires at a global scale, whereas the average rate for mires from northern Europe is 0.60–0.75 mm yr1 (Aaby, 1986). Growth rates in NW Spain seem to be mostly related to time periods and mire type. Numerous investigations in other parts of the world too have demonstrated that growth rates are as variable as typology, latitudinal location and microhabitats

Figure 4.9. Average accumulation rates of peat of studied Galician mires in different areas.

Mountain mires from Galicia (NW Spain)

95

of mires. Rates vary between 0.84 mm yr1 for the Florida Everglades (McDowell et al., 1969) and 0.025 mm yr1 for certain Canadian ombrotrophic bogs (Boville et al., 1983). For concentric raised bogs from SE Finland, Ma¨kila¨ (1997) obtained rates of 0.76 mm yr1 for Acutifolia, 0.67 mm yr1 for Sphagnum, 0.56 mm yr1 for Carex and 0.32 mm yr1 for Scheuchzeria–Cuspidata communities. Silvola (1986) reported a range of 0.5–1.5 mm yr1 for subarctic and boreal mires. The highest growth rates (0.6–0.7 mm yr1) occur in mires that began formation 3000–2000 years ago, coinciding with the generalized growth increase indicated by Aaby and Tauber (1974) for northern Europe since 3000 14C yr BP, with a maximum in the last 1000 years. The growth models indicate that ombrogenic mires whose formation began at the onset on the Holocene show a higher growth rate than the minerogenic mires. When multiple radiocarbon dates from each proﬁle of various mires are available, the calculated growth rates for ombrogenic mires are almost constant or show small variations. In Galicia, like other oceanic mires of North Hemisphere, this may be related to the relative greater effective precipitation in the oceanic areas where they are located. Mires can be considered sinks of a large number of chemical elements, and particularly carbon. Recent estimations of the total carbon reservoir in mires provided an average accumulation of 90–96 106 t yr1 (Gorham, 1991), although Silvola (1986) has suggested a somewhat greater value (110 106 t yr1). Carbon accumulation can be modeled by a non-linear function, if multiple ages are available for one mire. In this way Clymo (1991) calculated an average rate of C accumulation of 23 g m2 yr1 for 38 boreal mires. But, stratigraphic analyses of mires from different geographic areas have shown that accumulation rates have varied during their development, and that these variations correlate with local and global changes (Aaby and Tauber, 1974; Svensson, 1988). The variation in accumulation rates makes it difﬁcult to assign overall average values for all mires. The basal age, the stratigraphy, C content and peat density have to be known for a proper estimation of the longterm rate of carbon accumulation (LORCA; Tolonen et al., 1992). This complete set of data is available only for some 200 mires from Europe and North America (Gorham, 1991; Zoltai, 1991; Tolonen et al., 1992; Korhola et al., 1995). In NW Iberian Peninsula, we have studied 10 mires in this way, and the resulting accumulation rates (Table 4.4) vary between 48 and 112 g m2 yr1 of dry mass (82.0724.2 g m2 yr1), and 18.7–48.9 g C m2 yr1 (31.1711.0 g C m2 yr1). When this information is Table 4.4. mires. Mire type

All Ombrotr. Minerotr.

Statistics of mass and carbon accumulation rates in different types of Galician Carbon accumulation rates (g m2 yr1)

Mass accumulation rates (g m2 yr1) Aver.

SD

Min.

Max.

Aver.

SD

Min.

Max.

95.1 82.0 95.8

33.9 21.0 32.0

48.1 48.1 53.5

168.9 108.6 146.5

37.5 39.8 29.3

14.6 9.0 9.8

18.7 25.0 18.7

72.6 48.9 41.4

Note: SD, standard deviation.

96

X. Pontevedra-Pombal et al.

combined with that available in the literature for other mires of Galicia (Molinero et al., 1984; Aira and Guitia´n, 1986a, b; Ramil et al., 1994), we obtain, for the Holocene period, an average dry mass accumulation of 95.1733.9 g m2 yr1 and an average C accumulation of 37.5714.6 g C m2 yr1. Minerogenic mires have a higher mass accumulation (96 g m2 yr1) than ombrogenic mires (82 g m2 yr1), but the latter accumulated more carbon (39.8 versus 29.3 g C m2 yr1). This fact is related to the effect of the greater degree of decomposition and mineral matter incorporation in the density measurements in minerogenic mires. The average C accumulation rate in Galician mires is greater (37.5714.6 g C m2 yr1) than the global average value (20 g C m2 yr1) for mires given by Armentano and Menges (1986), as well as that reported for boreal Canadian mires (20 g C m2 yr1, Armentano and Menges, 1986; 29 g C m2 yr1, Gorham, 1991), Finland (o19 g C m2 yr1, Tolonen and Vasander, 1992; 25 g C m2 yr1, Silvola, 1986), Russia (20 g C m2 yr1, Armentano and Menges, 1986) and USA (19.9 g C m2 yr1, Korhola et al., 1995). But it is slightly lower to rates reported for European mires (48 g C m2 yr1, Armentano and Menges, 1986), W-NE USA (48 g C m2 yr1, Armentano and Menges, 1986) and the British Isles (450 g C m2 yr1, Ovenden, 1990). The C average accumulation rates of Galician ombrogenic mires (39.8 g C m2 yr1; range 25–49 g C m2 yr1) is higher than those average of ombrogenic mires from Scandinavia (9–34 g C m2 yr1; 20–30 g C m2 yr1, Tolonen, 1979), Estonia (22.5 g C m2 yr1) and USA (20–26 g C m2 yr1), but similar to those of raised bogs from Siberia and Finland (20–40 g C m2 yr1, Botch and Masing, 1983; 13–41 g C m2 yr1, Tolonen et al., 1992) or fall within the wide range reported for boreal mires (6.6–85.5 g C m2 yr1, Korhola et al., 1995). With regard to boreal minerogenic mires, Korhola et al. (1995) indicated C accumulation rates of 4.6–46.1 g C m2 yr1, which includes the range observed in Galicia (18.7–41.4 g C m2 yr1). But the average value for the later (29.3 g C m2 yr1) is greater than that of minerogenic mires from Finland (15.1 g C m2 yr1, Tolonen et al., 1994) and slightly higher than those of USA boreal minerogenic mires (27 g C m2 yr1, Tolonen et al., 1992) and France (28 g C m2 yr1, Francez and Vasander, 1995). Despite the high variability of C accumulation rates, coherent with differences in latitudinal distribution, mire type and local environmental factors, the results indicate that there is a qualitative global pattern in C accumulation between ombrogenic and minerogenic mires. So the greater C accumulation in ombrogenic as compared to minerogenic mires in Galicia is consistent with the process of formation of both types of mires in other distant geographic regions, as already pointed out by Korhola et al. (1995). In a research of more than 1300 mires from Finland, Estonia and USA, they observed that ombrogenic mires have accumulated more C (n ¼ 548, average ¼ 22.5 g C m2 yr1) than minerogenic ones (n ¼ 373, average ¼ 15.1 g C m2 yr1).

Composition and properties Most mires have superﬁcial layers with low or very low decomposed ﬁbric material, no deeper than 25 cm (Table 4.5). Below hemic peat extends for almost the whole

Physico-chemical range values of the different areas of peatlands on the northwest Iberian Peninsula.

Mire

Area

Organic horizon

Depth (cm)

Bulk density

Particle density

Porosity (%)

Ash (%)

PI

C

N

S

AGN˜1 BAG BDX1 BLA BMC1 BPA BRN2 BUI3 CAD CDL4 CPD LUZ5 MII6 MIM6 PDC PNV4 PVO PZC QXI7 SUA

N E SE E C E C N N N E E SE SE N N N E SE E

Oi-Oe-Oa Oi-Oa Oi-Oe Oi-Oe-Oa Oi-Oa Oi-Oe Oi-Oa Oe Oi-Oe-Oa Oa Oi-Oa Oi-Oe Oi Oi-Oe Oi Oi-Oe-Oa Oi-Oe Oi-Oe-Oa Oi Oi-Oe-Oa

130 135 40 145 70 190 110 400 124 400 55 540 140 100 184 250 300 265 105 115

0.04–0.20 0.06–0.60 – 0.11–0.26 – 0.1–0.34 – 0.11–0.19 0.13–0.98 0.15–0.23 0.14–0.17 – – – 0.09–0.16 0.07–0.32 0.09–0.24 0.16–0.60 – 0.10–0.23

– 1.40–2.20 – 1.53–1.88 – 1.50–1.86 – 1.90–2.20 1.34–2.06 – 1.54–1.66 – – – 1.36–1.51 – 1.39–1.53 1.72–2.22 – 1.45–1.83

– 55–85 – 86–93 – 81–93 – 87–92 60–91 – 89–90 – – – 88–94 – 83–94 74–91 – 77–93

– 4–63 – 10–25 – 21–57 – 4–19 2.7–79 – 19–34 – – – 1.3–7.3 – 1–12 54–76 – 5–50

2–5 0–7 – 2–7 – 3–5 – – 1–6 – 1–5 – – – 4–7 – 4–5 2–7 – 2–5

21–43 29–49 14–27 29–44 30–32 23–46 15–32 48–60 17–55 39–60 31–37 14–35 38–40 26–34 47–57 35–51 44–57 14–25 22–45 24–50

– 0.2–2.3 0.1–1.0 1.8–2.4 0.5–0.7 1.1–1.7 0.7–2.0 – 0.5–2.0 – 2.0–2.4 0.8–1.8 1.8 1.6–2.1 1.5–2.2 – 1.2–2.3 0.3–1.6 0.7–2.0 1.1–2.7

– 0.20–2.60 – 0.73–0.98 – 0.50–1.10 – – 0.06–0.30 – 0.79–0.97 – – – 0.19–0.39 – 0.61–0.77 0.30–1.10 – 0.54–0.95

pH H2O

pH KCl

PH CaCl2

BAG BDX BLA BMC BPA BRN

3.5–4.3 4.3–4.6 5.9–6.0 3.9–4.0 3.5–4.3 4.2–4.7

2.3–4.3 3.4–3.8 5.1–5.3 3.5–3.9 – 3.8–4.2

2.1–4.1 – 4.7–4.8 – – –

Ca 0.8–17.8 0.40–0.70 1.80–5.91 5.60–7.50 1.0–7.1 0.20–0.50

Mg

Na

K

Al

CEC

0.2–12.1 0.08–0.20 0.31–1.00 1.90–2.20 0.2–0.9 0.10–0.50

0.03–0.9 0.01–0.08 0.73–2.60 0.40–0.70 0.02–0.9 0.10–1.70

0.01–1.3 0.06–0.10 o0.01–2.32 o0.01–0.06 0.02–1.0 0.05–0.20

2.9–155 8.00–10.80 o0.01–0.09 7.10–15.90 2.5–13.4 5.00–10.00

11.3–221.3 44.7–78.8* 3.79–12.31 99.5–126.9* 8.3–16.2 60.0–165.5*

97

Mire

Mountain mires from Galicia (NW Spain)

Table 4.5.

98

Table 4.5 (continued ) BUI CAD CPD LUZ MII MIM PDC PVO PZC QXI SUA

3.2–3.8 2.5–3.9 4.3–4.7 3.4–4.8 4.6–4.9 4.7–4.9 3.6–4.4 3.6–4.6 3.4–4.8 3.8–4.2 4.5–5.6

– 2.5–3.0 3.9–4.3 3.1–3.8 4.4–4.6 4.4–4.6 2.9–3.3 2.3–3.3 3.4–4.2 2.8–3.6 3.6–4.2

– 2.0–2.4 3.2–3.7 – – – 2.9–3.2 2.3–3.1 3.3–4.1 – 3.0–3.9

o0.01 0.1–8.1 0.60–3.59 3.00–4.30 – 0.40–0.60 0.2–4.2 0.08–6.69 0.6–11.0 0.50–0.70 1.91–10.02

1.40–3.00 0.5–9.3 0.18–2.47 0.30–1.20 – 0.10–0.30 4.7–13.2 4.33–9.36 0.9–2.6 0.40–1.20 0.28–1.29

1.00–2.40 0.03–1.98 0.14–0.97 0.20–0.40 – 0.30–0.70 0.5–1.2 0.46–1.39 0.08–1.5 0.30–0.40 1.49–3.37

o0.01 o0.01–1.1 0.25–0.71 0.03–0.20 – 0.05–0.10 o0.01–0.5 0.02–2.93 0.09–4.2 0.07–0.30 o0.01–4.02

– 2.7–14 6.77–10.38 3.10–4.70 – 0.90–7.10 1.0–3.4 0.31–3.76 0.5–5.1 3.20–5.3 0.61–3.99

74.9–95.5* 7.1–23.5 8.77–17.35 42.2–56.0* – 23.8–30.7* 9.8–21.1 6.98–20.1 5.6–19.4 12.0–16.0 8.18–20.56

X. Pontevedra-Pombal et al.

Note: Area: N, North; E, East; SE, Southeast; S, South; C, Center. Horizon: present organic horizons; bulk and particle density in Mg m3; ash, total carbon (C), nitrogen (N) and sulfur (S) in percentage; PI, pyrophosphate index; cations and CEC, cationic exchange complex in cmolc kg1, where the values with asterisk have analyzed the CEC in ammonic acetate pH 7, and the rest have analyzed the CEC in ammonic clorure to ﬁeld pH. The superscript numbers in the mire column reference to the source of the information: 1 Ramil and Aira Rodriguez (1993). 2 Leiro´s and Gutia´n Ojea (1983). 3 Molinero et al. (1984). 4 Ramil et al. (1994). 5 Aira and Guitia´n (1986a, b). 6 Torras (1982). 7 Sanmamed (1979).

Mountain mires from Galicia (NW Spain)

99

Figure 4.10. Bulk (BD) and particle (PD) density differences between an ombrogenic and a minerogenic mire. The proﬁles indicate the horizon types according to soil taxonomy (Soil Survey Staff, 1998).

depth of the mire. Only a few minerogenic mires have a more or less developed basal layer of sapric peat. Peat bulk density (BD) ranges between 0.06 and 0.6 Mg m3 (mostly between 0.1 and 0.3 Mg m3), and particle density (PD) between 1.39 and 2.22 Mg m3; being 0.2 Mg m3 and 1.5 Mg m3 the bibliography reference values (Fig. 4.10). The highest values correspond to sapric peats, as already indicated by Lynn et al. (1974), to transitions between the peat and the mineral soil, and to minerogenic peats with a high proportion of mineral components, though in this case the change in particle density is greater than the change in bulk density (Van Lierop, 1981; PontevedraPombal, 1995). Organic carbon content ranges between 15% and 57%, nitrogen between 0.1% and 1.7%, and sulfur between 0.2% and 2.6% (Fig. 4.10). The lowest concentrations occur in layers with higher particle density. Although the evolution of organic matter in these mires is very slow due to effects of the low oxygen availability and oligotrophy on microbial activity (Gorham, 1995), a minimum level is maintained through time (Damman, 1988), which, in the ombrotrophic mires, results in an accumulation of C with depth/age. Although the degree of humiﬁcation varies between mires, a greater degree of decomposition of organic matter occurs in the basal transition to the more mineralized layers where high ratios have been measured between the carbon extracted in pyrophosphate to total carbon (Cp/Ct), whereas the overlying peat samples show very low Cp/Ct ratios (Fig. 4.11). In obrotrophic mires, pH values range between 3.2 and 4.9, indicative of very acid to acid conditions (Fig. 4.12), a fact coherent with elevated rainfall, acid substrata, the acidifying effect of mosses (Motzkin, 1994) and organic acids formed during the decomposition of the organic matter (autoacidiﬁcation). Exceptionally, pH is slightly

100

X. Pontevedra-Pombal et al.

Figure 4.11. Distribution of total carbon (left) and pyrophosphate carbon–total carbon ratios (Cp/Ct) (right) in ombrogenic and minerogenic mires.

Figure 4.12. pH values in representative ombrogenic and minerogenic mires.

Mountain mires from Galicia (NW Spain)

101

Figure 4.13. Effective cationic exchange complex (eCEC), basic cations sum (S) and aluminum (Al) differences between representative ombrogenic (left) and a minerogenic (central and right) mires.

acid to neutral in mires fed by waters draining carbonate rocks (Bran˜a de Lamelas (BLA); Table 4.5). Effective cation exchange capacity (eCEC) varies between 4 and 20 cmolc kg1. The sequence of exchangeable cations is Mg2+,Ca2+4Al3+4Na+4K+ in the upper peat layers of ombrotrophic mires and Al3+4Mg2+, Na+4Ca2+4K+ in the deeper ones; in minerotrophic mires the most frequent sequence is Al3+4Ca2+4Mg2+4Na+4K+ (in BLA Ca is more abundant than Al). The abundance of Mg in the ombrotrophic, coastal, mires is due to the effect of marine aerosols, whereas the dominance of Al in the minerogenic ones is a consequence of the composition of the waters draining acid soils. Despite these generalizations, many mires show changes in the vertical proﬁles of the exchangeable cations (Fig. 4.13). Cation exchange capacity (CEC) measured at pH 7 is, as expected, much higher (24–165 cmolc kg1); the largest differences with eCEC at soil pH occur in the most acid and decomposed peats.

Classification The classiﬁcation of mires is particularly difﬁcult due to several issues. These include the subjectivity of the methods used to determine the degree of peat decomposition, ﬁber content, the difﬁculty to estimate clay content in a material dominated by organic matter and the unavoidable heterogeneity of the methodology applied to study peat properties. With these limitations in mind and based on the World Reference Base for Soil Resources (FAO-WRB, 1998), mires from Galicia can be classiﬁed as histosols (Table 4.6), mainly as ﬁbric, sapric and terric histosols; although thionic histosols are also represented. Using the USDA Soil Taxonomy (Soil Survey Staff, 1998), most of these soils are classiﬁed as ﬁbrist (boro-, sphagno- and medi-ﬁbrists), but also as hemists (boro-, sulﬁ- and medi-hemists) and saprists (boro, sulﬁ- and medi-saprists).

X. Pontevedra-Pombal et al.

102

Table 4.6. World reference base for soil resources classiﬁcation (FAO–WRB, 1998) and soil taxonomy, Soil Survey Staff (1998), of representative Galician mires. Mire

Location

Tremoal da Gan˜idoira Bran˜a de Agolada

Serra do Xistral Serra dos Ancares Serra dos Ancares ManzanedaQueixa Serra dos Ancares Montes d Buio

Bran˜a de Lamelas Bran˜a dos Xuncos Bran˜a de Porto Ancares Bran˜a do Buio

Tremoal Cadramo´n Chao de Lamoso Campa da Cespedosa Lagoa de Lucenza Manzaneda II Manzaneda I Tremoal Pena da Cadela Veiga do Tremoal Tremoal Penido Vello Poza da Lagoa Maior Bran˜a de Queixa Borralleiras da Cal Grande Bran˜a de Sua´rbol

Altitude (m asl) 720

6895750 (130 cm)

1230

3390740 (215 cm)

1280

3090735 (165 cm)

1580

–

1580

10,6507170 (195 cm)

620

7725750 (315 cm)

Serra do Xistral Serra do Xistral Serra dos Ancares Serra do Caurel ManzanedaQueixa ManzanedaQueixa Serra do Xistral A Toxiza

1040

Serra do Xistral Serra dos Ancares Sierra de Queixa Montes do Buio Serra dos Ancares

Age (14C yr BP)

1039

8785730 (415 cm)

1415

2070725 (95 cm)

1440

17,390790 (700 cm)

1700

–

1630

–

900

4600780 (185 cm)

700

5080740 (220 cm)

790

4070750 (245 cm)

1330

10,3707210 (265 cm)

1600

–

620

4660770 (230 cm)

1080

1250725 (70 cm)

Classiﬁcation Terric Histosol Medisaprist Thionic Histosol Sulﬁhemist Sapric Histosol Boroﬁbrist Terric Histosol Mediﬁbrist Saprinc Histosol Sulﬁhemist Terric/ﬁbric Histosol Medihemist/ Mediﬁbrist Fibric Histosol Sphagnoﬁbrist Fibric Histosol Sphagnoﬁbrist Sapric Histosol Borosaprist Terric Histosol Borohemist Fibric Histosol Sphagnoﬁbrist Fibirc Histosol Sphagnoﬁbrist Fibric Histosol Sphagnoﬁbrist Terric Histosol Sphagnoﬁbrist Fibric Histosol Sphagnoﬁbrist Sapric Histosol Boroﬁbrist Fibric Histosol Boroﬁbrist Fibric Histosol Medihemist/ Mediﬁbrist Sapric Histosol Sulﬁsaprist

Galician mires: geochemical archives of environmental changes The ombrotrophic mires of NW Spain have been successfully used as archives of environmental changes. The relationship between the accumulation of heavy metals and human activities was studied in detail for Pb and Hg. For Pb, for example, it has been found that anthropogenic pollution may have begun some 2800–3000 years ago,

Mountain mires from Galicia (NW Spain)

103

Figure 4.14. Chronology of Pb enrichments of Tremoal do Penido Vello mire (PVO, Serras Septentrionais). The EF Pb (enrichment factor) is normalized to the average Pb/Ti ratio of pre-anthropogenic peat samples (ﬁlled area); dashed line illustrates variation through time of the Pb isotopic ratio (206P/ 207Pb) (age in 14C yr BP).

as a result of the production and trade of ternary bronzes with the Phoenicians in southern Spain (Fig. 4.14; Martı´ nez Cortizas et al., 1997, 2002; Pontevedra-Pombal, 2002). The beginning of a pervasive atmospheric Pb pollution is dated to the Iron Age and attained its maximum during the Roman period, with intensities on the order of ﬁve times the pre-pollution times ﬂuxes. The records also show local pollution episodes during medieval times, coeval with forest decline and soil erosion, suggestive of intensive human impacts on the landscape (Martı´ nez Cortizas et al., 2005). The largest Pb enrichments occur near the surface of the proﬁles and are related to the industrial period. The analysis of the isotopic composition of the Pb in Galicia also supports the chronology of Pb ﬂuxes (Martı´ nez Cortizas et al., 2002). Lead from pre-pollution times (43000 years) shows high values of the 206Pb/207Pb ratios (1.275), whereas the Roman period and the uppermost samples of the proﬁles show much lower ratios (206Pb/207Pb, respectively of 1.182 and 1.157), indicating anthropogenic pollution by mining/smelting activities and, in more recent times, the combustion of leaded gasoline. This trend is consistent with that described for other areas of the world (Settle and Patterson, 1980; Shotyk et al., 1988; Norton et al., 1990; Nriagu, 1996; Steinnes et al., 1997; Dunlap et al., 1999; Renberg et al., 2000; Weiss et al., 2002). Mercury anthropogenic pollution was also traced back to 2400–2500 14C yr BP, and the chronology of the changes have been found to be consistent with the history of mining and metallurgy of this element in the Iberian Peninsula (Fig. 4.15; Martı´ nez Cortizas et al., 1999). But Hg accumulation and the thermal-lability of the accumulated Hg were also found to depend on climatic conditions at the time of deposition. In short, cold climates promoted an enhanced accumulation of low thermal-lability Hg, whereas warm climates favored a lower accumulation and higher thermal-stability

104

X. Pontevedra-Pombal et al.

Figure 4.15. Average anthropogenic/natural Hg ratio values through time. (a) Total Hg (HgT) and the natural component of Hg accumulation (HgNAT, gray area). (b) Anthropogenic Hg (HgANT) at Tremoal do Penido Vello mire (PVO, Serras Septentrionais) and the main prehistoric and historic phases of the Hg exploitation in Spain (age in 14C yr BP).

Hg. Based on this behavior it was possible to reconstruct the changes in paleotemperatures during the last 4000 years, which show prominent cold (Neoglaciation and Little Ice Age) and warm periods (warm Roman period, medieval warm period). The record also shows an increase of temperature during the 20th century. The records of other elements where also investigated. Table 4.7 shows the accumulation of several elements at the Pena da Cadela (PDC) bog during the last 4000, 500 and 300 years (Martı´ nez Cortizas et al., 2002). For the last 4000 years, Fe is the element showing a higher accumulation (some 3000 kg ha1) followed by Br and Ti (300–430 kg ha1). Zinc, Pb, Sr, Zr and Cr have net accumulations of 10–40 kg ha1, whereas the rest of the elements show values between 2 and 8 kg ha1. Thus, at this coastal mire the long-term accumulation is dominated by the most abundant lithogenic elements that arrived to the mire as dust, and by those of marine origin (like Br). Nevertheless, on short times scales, Zn, Mn, Pb, As, Cr, Ni and Cu show larger proportions of the total accumulation, suggesting a relationship with the increasing impact of industrialization.

Conclusions Strong oceanic inﬂuence, high precipitation and wide geomorphologic heterogeneity, linked to Quaternary evolution, allowed the development of more than 10,000 ha of

Mountain mires from Galicia (NW Spain)

105

Table 4.7. Accumulation of several elements (kg ha1) at the Pena da Cadela (PDC) bog (Serras Septentrionais, Lugo) during the last 4000, 500, and 300 years. The bold numbers indicate the relative proportion of the total accumulation. Element

4000 years

500 years

Fe Br Ti Zn Pb Sr Zr Cr Mn Cu Ni As Rb Se Ga Y

2874.0 431.8 329.0 36.0 22.6 15.1 13.4 9.8 7.5 6.9 5.9 5.8 4.1 2.9 2.6 2.4

908.0 64.1 82.3 28.2 16.5 4.6 3.5 3.2 5.7 2.4 1.9 4.1 1.5 0.6 1.1 0.8

300 years 0.32 0.15 0.25 0.78 0.73 0.30 0.26 0.33 0.76 0.35 0.32 0.71 0.36 0.21 0.42 0.33

548.0 32.3 41.1 20.1 12.2 2.5 1.6 2.5 4.9 1.6 1.2 2.6 1.9 0.3 0.7 0.5

0.19 0.07 0.12 0.55 0.54 0.16 0.12 0.26 0.65 0.23 0.20 0.45 0.22 0.11 0.28 0.19

mires ecosystems in Galicia (NW Spain). The peculiarities associated to their formation and evolution make these wetlands important in Europe, some having developed at extreme geographic locations. For example, blanket bogs of Galicia have developed at the southwestern limit of this type of mire in Europe. Galician mires have formed during the Late Quaternary in three main phases. During the Preboreal and the Boreal (11,000–8500 years ago) minerogenic mires were formed, in the mid Atlantic (7800–7100 years ago) both minerogenic and ombrogenic mires developed and between the mid Atlantic and the beginning of the sub-Boreal (6000–2000 years ago) an expansion of ombrogenic mires and a stabilization of ecologic conditions of the minerogenic ones took place. Average peat thickness is 2 m for minerogenic and 3 m for ombrogenic mires, with growth rates between 0.2 and 0.7 mm yr1; the higher values occur in ombrogenic mires. Mean dry mass accumulation is 95 g m2 yr1 and carbon accumulation averages 37 g C m2 yr1. Minerotrophic mires accumulate more mass but less carbon than ombrotrophic mires. Bulk density decreases from the basal layers to the surface of the mire; the highest values are found in sapric organic materials, at transitions to the mineral sediment, or in layers of minerogenic mires with high ash content. The mires of Galicia are mainly acid mires (pH 2.5–5.0), with the lowest pH in the superﬁcial peat layers. Cation exchange capacity ranges between 5 and 18 cmol(+) kg1, with Mg2+ dominating in ombrotrophic mires and Al3+ in the minerotrophic ones. In ombrotrophic mires C increases with depth/age, whereas in the minerotrophic C content decreases with depth as the degree of evolution of the organic matter. Ombrotrophic mires of Galicia are true archives of chemical proxies of environmental evolution, and are involved in the cycles of many trace elements (like As, Hg,

106

X. Pontevedra-Pombal et al.

Pb, Zn) of environmental concern due to their potential toxicity. These archives reveal that anthropogenic atmospheric pollution in Spain dates back to at least 2500–2800 years ago, the Roman period representing a pollution climax in preindustrial times. During the last 500 years atmospheric pollution has increased abruptly, although some elements are showing symptoms of moderate decreases in recent atmospheric ﬂuxes (as for example Pb).

References Aaby, B., 1986. Paleoecological studies of mires. In: Berglund, B.E. (Ed.), Handbook of Holocene Palaeoecology and Paleohydrology. Wiley, New York, pp. 145–164. Aaby, B. and Tauber, H., 1974. Rates of peat formation in relation to degree of humiﬁcation and local environment, as shown by studies of a raised bog in Denmark. Boreas 4, 1–17. Aira, M.J. and Guitia´n Ojea, F., 1986a. Contribucio´n al estudio de los suelos y sedimentos de montan˜a de Galicia y su cronologı´ a por ana´lisis polı´ nico. II. Perﬁles de la penillanura de cumbres de la sierra de Queixa (Orense). Anal. Edafol. y Agrobiol. 45, 1203–1218. Aira, M.J. and Guitia´n Ojea, F., 1986b. Contribucio´n al estudio de los suelos y sedimentos de montan˜a de Galicia y su cronologı´ a por ana´lisis polı´ nico. I. Sierra del Caurel (Lugo). Anal. Edafol. y Agrobiol. 45, 1189–1202. Armentano, T.V. and Menges, E.S., 1986. Patterns of change in the carbon balance of organic soilwetlands of the temperature zone. J. Ecol. 74, 755–774. Averdieck, F.R., Hayen, H., Heathwaite, A.L., and Willkomm, H., 1993. The chronology of mire development. In: Heathwaite, A.L. and Go¨ttlich, K. (Eds), Mires: Process, Exploitation and Conservation. Wiley, Chichester, pp. 123–170. Benito, E., Soto, B., and Dı´ az-Fierros, F., 1991. Soil erosion studies in NW Spain. In: Sala, M., Rubio, J.L., and Garcı´ a-Ruiz, J.M. (Eds), Soil Erosion Studies in Spain. Geoforma, Logron˜o, pp. 55–74. Botch, M.S. and Masing, V.V., 1983. Mire ecosystems in the USSR. In: Gore, A.J.P. (Ed.), Mires: Swamp, Bog, Fen and Moor. General Studies. Elsevier, Amsterdam, Vol. A, pp. 95–152. Boville, B.W., Jun, R.E., and Hare, F.K., 1983. Final report: the storage of non-living organic carbon in boreal and arctic zones-Canada. DE-AS01-81EV10688. Institute of Environmental Studies, University of Toronto, Toronto, Canada. Casares Gil, A., 1920. Sphagnum Pylaiei Brid. en el N. W. de la Penı´ nsula Ibe´rica. Boletı´ n de la Sociedad Espan˜ola de Historia Natural. Tomo XX, 17pp. Chambers, F., 1988. Archaeology and the ﬂora of the British Isles: The Moorland experience. In: Jones, M. (Ed.), Archaeology and the Flora of the British Isles. Oxford University of Committee for Archaeology, Monograph 14, pp. 107–115. Clymo, R.S., 1991. Peat growth. In: Shane, L.C.K. and Cushing, E.J. (Eds), Quaternary Landscapes. Belhaven Press, London, pp. 76–112. Damman, A.W.H., 1988. Regulation of nitrogen removal and retention in Sphagnum bogs and other peatlands. Oikos 51, 291–305. Davis, R.B. and Anderson, D.S., 1991. The eccentric bogs of Maine: a rare wetland type in the United States. Maine State Planning Ofﬁce, Critical Areas Programme, Planning Report 93, University of Maine. Dimbleby, G.W., 1985. The Palynology of Archaeological Sites. Academic Press, London. Dunlap, C.E., Steinnes, E., and Flegal, A.R., 1999. A synthesis of lead isotopes in two millenia of European air. Earth Planet. Sci. Lett. 167, 81–88. European Commission, 1996. Interpretation manual of European Union habitats. DG XI – Environment, Nuclear Saftey and Civil Protection. FAO-WRB, 1998. Base referencial mundial del recurso suelo. FAO/ISRIC/SICS. Roma, Italia. Fraga Vila, I., Sahuquillo Balbuena, E., and Garcı´ a Tasende, M., 2001. Vegetacio´n caracterı´ stica de las turberas de Galicia. In: Martı´ nez Cortizas, A. and Garcı´ a-Rodeja Gayoso, E. (Eds), Turberas de montan˜a de Galicia. Coleccio´n Te´cnica – Medio Ambiente. Consellerı´ a de Medio Ambiente. Xunta de Galicia, Santiago de Compostela, Galicia. Cap. 6, pp. 79–97.

Mountain mires from Galicia (NW Spain)

107

Francez, A.-J. and Vasander, H., 1995. Peat accumulation and peat decomposition after human disturbance in French and Finnish mires. Acta Oecol. 16, 599–608. Godwin, H., 1946. The relationship of bog stratigraphy to climatic change and archaeology. Proc. Prehistoric Soc. 1, 1–11. Godwin, H., 1981. The Archives of the Peat Bogs. Cambridge University Press, London. Gorham, E., 1991. Northern peatlands: role in the carbon cycle and probable responses to climatic warming. Ecol. Appl. 2, 182–195. Gorham, E., 1995. The biogeochemistry of northern peatlands and its possible responses to global warning. In: Gorham, E., Woodwell, G.M., and Mackenzie, F.T. (Eds), Biotic Feed Backs in the Global Climatic System. Oxford University Press, Oxford, pp. 169–186. Granlund, E., 1932. De svenska ho¨gmossarnas geologi. Sveriges Geologiska Underso¨kning C 373, 1–193. Jato, M.V., 1974. Contribucio´n a la cronologı´ a de suelos por ana´lisis de polen. T.D. Fac. Ciencias, Univ. Santiago de Compostela. Korhola, A., Tolonen, K., Turunen, J., and Jungner, H., 1995. Estimating long-term carbon accumulation rates in Boreal peatlands by radiocarbon dating. Radiocarbon 37, 575–584. Leiro´s, M.C. and Guitia´n Ojea, F., 1983. Suelos de la zona hu´meda espan˜ola. XI. Contribucio´n al estudio de los suelos hidromorfos de Galicia. 2. Suelos continentales. Anales de Edafologı´ a y Agrobiologı´ a 42, 427–461. Lindsay, R.A., 1995. Bogs: The Ecology, Classiﬁcation and Conservation of Ombrotrophic Mires. Scottish Natural Heritage, Perth. Lynn, W.C., McKinzie, W.E., and Grossman, R.B., 1974. Field laboratory tests for characterization of Histosols. In: Aandahl, A.R., Buol, S.W., Hill, D.E., and Bailey, H.H. (Eds.), Histosols: Their Characteristics, Classiﬁcation, and Use. Soil Sci. Soc. Am., Inc, Madison, WI, USA, pp. 11–20. Ma¨kila¨, M., 1997. Holocene lateral expansion, peat growth and carbon accumulation on Haukkasuo, a raised bog in southeastern Finland. Boreas 26, 1–14. Malmer, N., 1975. Development of bog mires. In: Hassler, A.D. (Ed.), Coupling of Land and Water Systems. Ecological studies 10. Springer, New York, pp. 85–92. Martı´ nez Cortizas, A., Garcı´ a-Rodeja, E., Pontevedra-Pombal, X., et al., 2002. Atmospheric Pb deposition in Spain during the last 4,600 years recorded by two ombrotrophic peat bogs and implications for use of peat as archive. Sci. Total Environ. 292, 33–44. Martı´ nez Cortizas, A., Mighall, T., Pontevedra-Pombal, X., et al., 2005. Linking changes in atmospheric dust deposition, vegetation evolution and human activities in NW Spain during the last 5300 years. The Holocene 15, 698–706. Martı´ nez Cortizas, A. and Moares Domı´ nguez, C., 1995. Edafologı´ a y Arqueologı´ a: estudio de yacimientos arqueolo´gicos al aire libre en Galicia. Consellerı´ a de CulturaXunta de Galicia, 199pp. Martı´ nez Cortizas, A. and Pe´rez Alberti, A., 2000. Atlas clima´tico de Galicia. Consellerı´ a de Medio AmbienteXunta de Galicia. Martı´ nez Cortizas, A., Pontevedra-Pombal, X., Garcı´ a-Rodeja, E., et al., 1999. Mercury in a Spanish peat bog: archive of climate change and atmospheric metal deposition. Science 284, 939–942. Martı´ nez Cortizas, A., Pontevedra-Pombal, X., No´voa Mun˜oz, J.C., et al., 1997. Four thousand years of atmospheric Pb, Cd and Zn deposition recorded by the ombrotrophic peat bog of Penido Vello (northwerstern Spain). Water Air and Soil Pollut. 100, 387–403. Martı´ nez Cortizas, A., Pontevedra-Pombal, X., No´voa Mun˜oz, J.C., et al., 2001. Distribucio´n geogra´ﬁca y cronologı´ a de las turberas de Galicia. In: Martı´ nez Cortizas, A. and Garcı´ a-Rodeja, E. (Eds), Turberas de Montan˜a de Galicia. Coleccio´n Te´cnica – Medio Ambiente, Xunta de Galicia, Cap. 2, pp. 33–37. McDowell, L.L., Stephens, J.C., and Stewart, E.H., 1969. Radiocarbon chronology of the Florida everglades peat. Soil Sci. Am. Proc. 33, 743–745. Molinero, A., Polo, A., and Dorado, E., 1984. Caracterı´ sticas fı´ sico-quı´ micas de la turbera de Vivero (Lugo). Anal. Edaf. y Agrobiol. 43, 1107–1122. Moore, P.D., 1975. Origin of blanket mires. Nature 256, 267–269. Moore, P.D. and Bellamy, D.J., 1974. Peatlands. Elek Sci, London, 221pp. Motzkin, G., 1994. Calcareous fens of western New England and adjacent New York State. Rhodora 96, 44–68. Mun˜oz Sobrino, C., 1996. Aportacio´n a la historia de la vegetacio´n de la sierra de Ancares y montes do Cebreiro (Lugo-Leo´n) a partir del ana´lisis polı´ nico de sedimentos turbosos. Tesis de Licenciatura, Depto. Bioloxı´ a Vexetal, Univ. Santiago de Compostela, 109pp.

108

X. Pontevedra-Pombal et al.

Norton, S.A., Dillon, P.J., Evans, R.D., et al., 1990. The history of atmospheric deposition of Cd, Hg, and Pb in North America: evidence from lake and peat bog sediments. In: Lindberg, S.E., Page, A.L., and Norton, S.A. (Eds), Sources, Deposition and Canopy Interactions Acidic Precipitation. Springer, Berlin, pp. 73–102. Nriagu, J.O., 1996. A history of global metal pollution. Science 272, 223–224. Ovenden, L., 1990. Peat accumulation in northern wetlands. Quatern. Res. 33, 377–386. Pontevedra-Pombal, X., 1995. Histosoles de la ‘Serra dos Ancares’ (Galicia). Tesis de Licenciatura. Depto. Edafoloxı´ a e Quı´ mica Agrı´ cola, Fac. Bioloxı´ a, Univ. Santiago de Compostela. Pontevedra-Pombal, X., 2002. Turberas de Montan˜a de Galicia: Ge´nesis, Propiedades y su Aplicacio´n como Registros Ambientales Geoquı´ micos. Tesis Doctoral. Depto. Edafologı´ a y Quı´ mica Agrı´ cola, Fac. Biologı´ a, Univ. Santiago de Compostela. Ed. Universidade de Santiago de Compostela, 489pp. Pontevedra-Pombal, X. and Martı´ nez Cortizas, A., 2001. Evolucio´n y estado actual de las turberas en Galicia. In: Martı´ nez Cortizas, A. and Garcı´ a-Rodeja, E. (Eds), Turberas de Montan˜a de Galicia. Coleccio´n Te´cnica – Medio Ambiente, Xunta de Galicia, Cap. 11, pp. 163–176. Pontevedra-Pombal, X., Martı´ nez Cortizas, A., and Buurman, P. (in press). Las turberas de montan˜a de Galicia como sumideros de carbono. Edafologı´ a. Pontevedra-Pombal, X., Martı´ nez Cortizas, A., and Garcı´ a-Rodeja, E., 1996. Caracterizacio´n fı´ sicoquı´ mica de depo´sitos orga´nicos en la ‘Serra dos Ancares’ (NW ibe´rico): origen y evolucio´n. In: Pe´rezAlberti, A. and Martı´ nez Cortizas, A. (Eds), Avances en la reconstruccio´n paleoambiental de las a´reas de montan˜a lucenses. Diputacio´n Provincial de Lugo, Lugo, pp. 91–119. Pontevedra-Pombal, X., Martı´ nez Cortizas, A., Garcı´ a-Rodeja, E., et al., 2001a. Composicio´n y transformacio´n de la materia orga´nica en un histosol de la Serra do Xistral (norte de Galicia) mediante CPMAS 13C-NMR. Edafologı´ a 8, 67–79. Pontevedra-Pombal, X., No´voa Mun˜oz, J.C., Garcı´ a-Rodeja, E., et al., 2001b. Composicio´n y propiedades de las turberas de Galicia. In: Martı´ nez Cortizas, A. and Garcı´ a-Rodeja, E. (Eds), Turberas de Montan˜a de Galicia. Coleccio´n Te´cnica – Medio Ambiente, Xunta de Galicia, Cap. 8, pp. 129–139. Pontevedra-Pombal, X., No´voa Mun˜oz, J.C., Garcı´ a-Rodeja, E., et al., 2003. Mires of Galicia (Spain). IMCG Newslett 2, 13. Raeymaekers, G., 1999. Conserving mires in the European Union. Actions co-ﬁnanced by life-nature. In: Nuclear European Commission DG XI, Environment, Nuclear Safety and Civil Protection. Ofﬁcial publications of the European Communities. Ecosystems LTD, Luxembourg, 90pp. Ramil, P., 1992. La vegetacio´n cuaternaria de las sierras septentrionales de Lugo a trave´s del ana´lisis polı´ nico. PhD Report, Universidad de Santiago de Compostela, 256pp. Ramil, P., Aira, M.J., and Taboada, M.T., 1994. Ana´lisis polı´ nico y sedimentolo´gico de dos turberas en la sierras septentrionales de Galicia (N.O. de Espan˜a). Rev. Pale´obiol. 13, 9–28. Ramil, P. and Aira Rodrı´ guez, M.J., 1993. Estudio de la turbera de Pena Veira (Lugo). Ana. Jardı´ n Bot. Madrid 51, 111–122. Ratcliffe, D.A., 1977. A Nature Conservation Review 1. Cambridge, England, 401pp. Renberg, I., Bra¨nnvall, M.L., Bindler, R., and Etmeryd, O., 2000. Atmospheric lead pollution history during four millennia (2000 BC to 2000 AD) in Sweden. AMBIO 29, 150–156. Riley, J.L., 1987. The Ontario peatland inventory, 1982–86. Proceedings of the symposium ‘87 Wetlands/ Peatlands. Edmonton, Alberta, pp. 373–382. Sanmamed, A., 1979. Contribucio´n al estudio de los suelos de montan˜a gallegos. Acta Cientı´ ﬁca Compostelana 16, 219–246. Settle, D.M. and Patterson, C., 1980. Lead in Albacore: guide to lead pollution in Americans. Science 207, 1167–1176. Shotyk, W., Weiss, D., Appleby, P.G., et al., 1998. History of atmospheric lead deposition since 12,370 14 C yr BP from a peat bog, Jura Mountains, Switzerland. Science 281, 1635–1640. Silvola, U., 1986. Carbon dioxide dynamics in mires reclaimed for forestry in eastern Finland. Ann. Bot. Fennici 23, 59–67. Soil Survey Staff, 1998. Keys to Soil Taxonomy 8th edn. USDA/NRCS, Washington, USA, 328pp. Solem, T., 1986. Age, origin and development of blanket mires in Sør-Trøndelag, Central Norway. Boreas 15, 101–115.

Mountain mires from Galicia (NW Spain)

109

Steinnes, E., Allen, R.O., Petersen, H.M., et al., 1997. Evidence of large scale heavy-metal contamination of natural surface soils in Norway from long-range atmospheric transport. Sci. Total Environ. 205, 255–266. Svensson, G., 1988. Bog development and environmental conditions as shown by the stratigraphy of store mosse mire in southern Sweden. Boreas 17, 89–111. Tolonen, K., 1979. Peat as a renewable source: long-term accumulation rates in North European mires. Proceedings of the International Symposium, Classiﬁcation peat and peatlands, Hyytla¨la¨, Finland, pp. 282–296. Tolonen, K., Turunen, J., Vasander, H., et al., 1994. Rate of carbon accumulation in boreal mires. In: Kanninen, M. and Heikinheimo, P. (Eds), SILMU, The Finnish Research Programme on Climate Change. Publications Academy, Finland 1/94, pp. 297–302. Tolonen, K. and Vasander, H., 1992. Rate of net accumulation of carbon in peat layers, with special emphasis on the past few hundred years. In: Kanninen, M. and Anttila, P. (Eds), The Finnish Research Programme on Climate Change: Progress Report. VAPK, Helsinki, Finland, pp. 219–225. Tolonen, K., Vasander, H., Damman, A.W.H., et al., 1992. Preliminary estimate of long-term carbon accumulation and loss in 25 boreal peatlands. Suo 43, 277–280. To¨rnqvist, T.E. and Joosten, J.H.J., 1988. On the origin and development of Subatlantic ‘‘man-made’’ mire in Galicia (northwest Spain). Proceedings of the 8th International Peat Congress, Leningrad, Russia, pp. 214–224. Torras, M.A., 1982. Aplicacio´n del ana´lisis polı´ nico a la datacio´n de paleosuelos en Galicia. PhD. Report, Universidad de Santiago de Compostela, 250pp. Valca´rcel, M., 1999. Evolucio´n geomorfolo´gica y dina´mica de las vertientes en el noroeste de Galicia: importancia de los procesos de origen frı´ o en un sector de las montan˜as lucenses. T.D., Departamento de Geografı´ a, Universidad de Santiago. Van Lierop, W., 1981. Laboratory determination of ﬁeld bulk density for improving fertilizer recommendations of organic soils. Can. J. Soil Sci. 61, 475–482. Weber, C.A., 1908. Aufbau und Vegetation der Moore Norddeutschlands. Englers Bot. Jahrb 90 . Weiss, D., Shotyk, W., Boyle, E.A., et al., 2002. Comparative study of the temporal evolution of atmospheric lead deposition in Scotland and eastern Canada using blanket peat bogs. Science of the Total Environment 292, 7–18. Zoltai, S.C., 1991. Estimating the age of peat samples from their weight: a study from west-central Canada. Holocene 1, 68–73.

This page intentionally left blank

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

111

Chapter 5

Geomorphologic emplacement and vegetation characteristics of Fuegian peatlands, southernmost Argentina, South America A. Coronato, C. Roig, L. Collado and F. Roig

Introduction Tierra del Fuego is an insular region of Patagonia in which several types of peatlands are very well represented, covering a vast portion of the landscape of the central and southern areas of the island (Fig. 5.1). The maritime, cold-temperate climate of the region provides the necessary conditions for the preservation of peatlands developed during the last ca. 15,000 years, and also for the present-day formation of peatlands. The landscape evolution of Tierra del Fuego during the Last Glacial Maximum (ca. 24,000 cal yr BP) and the following times has contributed to peatland formation because of abundant meltwater and closed or depressed landforms. The peatland formation started in Late Glacial times and, more frequently, in the early Holocene. The peatlands of Tierra del Fuego are of great scientiﬁc interest due to the complexity of the ecosystems and because they provide extensive data banks concerning paleoenvironmental conditions of the region (Rabassa et al., 2006 – this book, Ch. 6). They are also of tourist interest due to their singularity in the Argentine and South American landscapes, and of direct economic value considering the quality and quantity of mining the peat resources. In spite of that, very little is known of their ecosystem, their role in the regional hydrological cycle, and the effective amount of mining resources. The relationships between the different peatland types and the geomorphologic units in which they are emplaced are presented in this chapter.

Physical setting The studied peatlands are located in the eastern-central portion of Tierra del Fuego (Fig. 5.1). This region is composed of Tertiary marine sedimentary rocks, exposed on NW–SE oriented, hilly ranges (600 m asl), which progressively diminish in elevation towards the north and east. South of the Lake Fagnano depression, upper ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09005-5

112

A. Coronato et al.

Figure 5.1. Topographic and peatland distribution map of the studied area.

Jurassic–lower Cretaceous metamorphic marine sedimentary rocks and acid volcanics form the mountain ranges with heights up to 1200 m asl. The regional climate is determined by the high latitude of the region, its insularity and its closeness to the Antarctic continent. The mean temperature of the coldest month (July) is –11C and that of the warmest month (January) is 101C (Tuhkanen, 1992). Frosting and snowfall are present from June to September, but they can happen even during the summer. The topography provides an altitudinal thermal gradient, with frequent thermal inversions during winter months, and generates an orographic effect on precipitation, when intercepting the wet air masses coming from the Southern Paciﬁc cyclonic systems, and from the Polar Front. The annual precipitation is estimated in 400–500 mm/yr. The Nothofagus sp. subantarctic forest occurs in the lowlands and mountain slopes up to the elevation of 650 m asl (Moore, 1983). Nothofagus pumilio extends over the slopes and the plains whereas Nothofagus antarctica occupies the poorly drained soils, close to the peatlands (Roig, 1998).

Previous works Studies on the Fuegian peatlands started at the beginnings of the 20th century. Bonarelli (1917) described in detail the peat-forming, main plant species, and he recognized different peatland-forming environments, according to their marginal, superﬁcial and riverside ﬂora. He considered also the substratum characteristics, peat chemical composition, and environmental factors, such as temperature and humidity. He classiﬁed peatlands as those pertaining to (1) mountain slopes, (2) plains,

Geomorphologic emplacement of Fuegian peatlands

113

(3) valleys, (4) intermountain valleys, and (5) littoral environments. Guin˜azu´ (1934) distinguished Sphagnum peatlands as (1) non-humiﬁed, (2) semi-humiﬁed, (3) mature, (4) of ﬂat surface, (5) with moss hummocks, and (6) with blackish and reddish water circulation. He observed that young peatlands show a convex surface in their center. By means of proﬁle studies, he noted their compositional variation with a gradation of Carex peat at depth and of Sphagnum type at the surface, assigning this zonation to past climate changes. Roivainen (1954) classiﬁed different peatland types and subtypes, according to the plant communities present in each case. Based on vegetation census, he established 18 subtypes, grouped in three regional types: (1) prairie peatlands, (2) Sphagnum peatlands, and (c) pluvial peatlands. The most extensive work on the Patagonian and Fuegian peatlands was done by Auer (1965). The sections studied in those peatlands were distributed across Tierra del Fuego and provided the bases for a description of the component materials and their tephra layers. The latter could be used to correlate events in Patagonia with those in Scandinavia. These tephra layers, easily recognized in the sections, allowed Auer (1965) to establish different stages in the vertical and horizontal development of peatlands and to make comparisons among them. He deﬁned ﬁve peatland categories for Tierra del Fuego, following a north–south latitudinal distribution: (1) steppe, (2) transitional, (3) Sphagnum sp., (4) transitional again, and (5) of pluvial region. In the mid-20th century, the Fuegian peatlands became of interest for power production and alternative purposes; isopach maps and peat volume estimation were performed and recorded in several unpublished reports. The paleoenvironmental and paleoclimatic aspects, as they are reconstructed from the fossil pollen record, have been thoroughly studied after 1986 (Heusser, 1989a,b, 1998, 2003; Borromei, 1995; Heusser and Rabassa, 1995, 1987; Mauquoy et al., 2004). Rabassa et al. (1996) offered a summary on the state of the knowledge on the peatlands of Tierra del Fuego. Roig et al. (2001, 2004) and Roig and Collado (2004) presented peatland surveys based on remote-sensing analysis and a classiﬁcation according to the type of plant cover.

Methodology The geomorphologic analysis was performed by means of air photo interpretation (scale 1:40,000) and ﬁeld surveying, using state-of-the-art techniques for the preparation of geomorphologic maps. Data were spatially processed by means of GIS (Geographic information systems), on the basis of a cartography prepared with aerophotogrammetric mosaics and Landsat 7 satellite images. Twenty geomorphologic units were classiﬁed according to their morphogenesis. Based on the radiometric homogeneity from a non-supervised classiﬁcation on satellite images, sampling sites were deﬁned to establish the different types of vegetation cover. The ﬂora surveying of the selected sites was made by the BraunBlanquet (1979) phytosociological method, estimating abundance and dominance, ecological conditions, and mesologic observations. Not considering the forest ecosystem, 74 plant species were identiﬁed, grouped in three major plant communities: Sphagnum sp. Peatlands, Carex curta over-ﬂooded peaty meadows, and

114

A. Coronato et al.

Carex gayana wet peaty meadows. These two last communities were grouped in the remote sensing analysis as the Carex peatlands, due to the similarity of the hydrological conditions and the spectral behavior of the dominant species. Once the peatland and geomorphologic unit distributional maps were obtained, several derived aspects were analyzed, such as superﬁcial development, depth, and stratigraphic proﬁle.

Results As much as 19% of the surveyed region is occupied by 874 peatlands, whose individual surface varies between 0.05 and 949 ha. The average peatland surface size is 7.15 ha. The peatland types of units larger than 20 ha, which represent more than 75% of the total, were analyzed in detail (Table 5.1). Geomorphologic aspects Different geomorphologic agents modeled the local relief, but Pleistocene glaciers were the most powerful ones. Peatlands are widespread, but they are most commonly located on glacial geomorphologic units. In fact, a large outlet glacier occupied the entire Lake Fagnano region during the Last Glaciation Maximum (ca. 24,000 cal yr BP; Coronato et al., 2004; Rabassa et al., 2006 – this book, Ch. 6), which modeled the rocky hills and the lowlands. The postglacial drainage system was installed mainly following the ancient meltwater streams. Table 5.2 summarizes the present geomorphologic units, whereas its spatial distribution is shown in Figure 5.2. The geomorphologic units are grouped according to their morphogenesis: (1) Structural origin: rocky hills composed of marine sandstone ridges in the north (600 m asl) and slaty graywacke ranges in the south (700–1200 m asl). The Pleistocene glaciers generating many closed erosional features, such as small basins, have overridden the northern rocky hills. (2) Alluvial origin: including colluvium mantles in the low slopes and alluvial fans with mineral and nutrients supplied by sheet runoff or poorly-drained areas. (3) Littoral origin: including a lacustrine gravel bar that partially dammed the land drainage, fostering formation of coastal peatlands. (4) Fluvial origin: this includes valley bottoms with narrow, meandering, gravelbottom channels that develop coarse sand and gravel bars, wide ﬂood plains, ancient channels and non-functional meanders, terraces, and a delta. In all of them, major changes in basal ﬂow and water table conditions have had a strong inﬂuence in peatland formation. (5) Glacial origin: this is the most relevant morphogenetic process and includes many geomorphic units, described as follows: (a) frontal moraine, a convex arc-shaped hill which encloses the head of the lake with a gentle slope, (b) basal moraines, forming ﬂat-topped hills, strongly dissected by ancient meltwater streams, (c) glaciolacustrine plain composed of massive clayey silts, (d) glacioﬂuvial cone

Geomorphologic emplacement of Fuegian peatlands

115

Table 5.1. Surface occupied by peatlands, in hectares and percentage (calculated on the basis of the amount of peatlands larger than 20 ha). Types

Surface in Ha

%

Raised bog Middle bog Impoverished bog Carex fens Total

1803.9 1205.5 177.3 1546.4 4733.1

38.12% 25.46% 3.74% 32.68% 100%

Table 5.2.

Geomorphologic units recognized in the studied area.

Morphogenesis

Geomorphic units

Structural

Sandstone rocky hills Slaty graywacke rocky hills

Alluvial

Alluvial belts in low slopes Alluvial fan

Littoral

Lacustrine beach

Fluvial

Valley bottom Terrace Delta

Glacial

Palaeochannel Basal moraine Frontal moraine Glaciolacustrine plain Western kame terrace Eastern kame terrace I Eastern kame terrace II Kettle holes Glaciofuvial plain Glacioﬂuvial cone

composed of silt-gravelly terraces, kettle holes, and non-integrated channels, (e) glacioﬂuvial plain of almost ﬂat topography and high permeability gravels, (f) kettle-holes forming closed depressions without a drainage network, (g) kameterraces formed by varied-size gravel strata, medium to coarse sands with deformational structures and till pockets, and (h) paleochannels, deﬁned as elongated depressed zones that dissect a set of basal moraines. Vegetation and floristic diversity aspects The applied phytosociological method for type recognition has deﬁned Sphagnum ombrotrophic peatlands, and Cyperaceas (Carex curta and C. gayana) minerotrophic peatlands.

116

A. Coronato et al.

Figure 5.2. Geomorphologic map of the studied area.

(1) The Sphagnum sp. peatlands occur in open, relatively ﬂat, slightly raised areas, normally surrounded by forest. Sphagnum magellanicum is dominant, together with a rich ﬂora of other mosses, lichens, liverworts, fungi, and upper plants (Fig. 5.3). The Sphagnum peatlands are considered a part of the Myrteolo-Sphagnetea class (Roig et al., 1985), which includes the following typical species: S. magellanicum, Carex magellanica, Oreobolus obtusangulus, Tetroncium magellanicum, Rostkovia magellanica, and Marsippospermum grandiflorum. Among superior plants, Empetrum rubrum and Nothofagus antarctica are always present as accompanying species together with many cryptogams. The dominance of several of these accompanying plants, sometimes with high cover values, is a signal of ecosystem oldness. R. magellanica may form minerotrophic communities, together with S. fimbriatum and Juncus scheuzerioides (Greene, 1964). Considering the differences in ﬂoristic composition and ecological conditions, three types of Sphagnum peatlands have been identiﬁed as follows (a) Raised bogs, with a strong dominance of S. magellanicum. They present very few ponds, either with water on the surface and colonized by Rostkovia, Tetroncium, or S. fimbriatum, or more or less drier, with dead Sphagnum. There are a few or no lichens. They have a marginal hummocky topography, and a ﬂat, central surface, with reddish colors throughout (Fig. 5.4). They may present high cover values of M. grandiflorum. The water table is located at about 20 cm depth. (b) Middle bogs, where the S. magellanicum cover is as low as 50–75% of the total area, or they can be high, achieving the appearance of grassland where

Geomorphologic emplacement of Fuegian peatlands

117

Figure 5.3. Raised bog with Empetrum rubrum as an accompanying species. It is surrounded by Nothofagus pumilio forest. The sampler shown is the one used for sample coring and depth survey.

Figure 5.4. Bog with hummocky topography.

118

A. Coronato et al.

Figure 5.5. Middle bog invaded by Marssipospermum grandiflorum. Its high cover value provides a grassland aspect to the bog. Nothofagus antarctica trees are also well developed.

they are invaded by M. grandiflorum (Fig. 5.5). There is a strong dominance of E. rubrum. Numerous hollows are present, either water holes rich in Rostkovia magellanica or dry hollows with dead Sphagnum. There are lichens in the drier parts. At the edges of the peatlands, N. antarctica reach high abundance values and develop dwarf individuals. This type of peatland has a hummocky relief at the edges but a horizontal and ﬂat surface in the central areas (Fig. 5.5). (c) Impoverished bogs, in which S. magellanicum shows the lowest cover percentages (25–50% of the total). Conversely, there is an increase of the cover of E. rubrum that may become dominant (75–100%). Abundant Rostkovia and lichens are present. There are sectors of the peatland that present a high cover of dead Sphagnum. The topography is mostly ﬂat. (2) The Cyperacea peatland group includes C. curta and C. gayana as dominant species. The phytosociological survey shows that Carex is the most abundant species, but it is not the major peat-forming plant. Peat in Carex grassland is mainly formed by a group of mosses that could cover up to 100% of the surface in the lower vegetation strata, hidden under Graminaceae. (a) C. curta occurs on ﬂooded peaty meadows, where they are affected by slow sheet runoff. Water may occur free at the surface. Hummocks named ‘tussocks’ are common, and they are considered as the result of seasonal frozen ground (Corte, 1996). In these peaty meadows, C. curta is dominant, accompanied by Alopecurus magellanicus, Agrostis meyenii, Festuca contracta,

Geomorphologic emplacement of Fuegian peatlands Table 5.3.

119

Cover diversity (H) and speciﬁc richness (Re).

Community

H

Re

Sphagnum magellanicum Carex curta Carex gayana

2.044 2.65 2.80

37 43 52

and C. magellanica, which give them a grassland appearance. Maximum humidity conditions are shown by the occurrence of C. decidua, C. microglochin, Caltha sagitata, and Triglochin palustris. Hippuris vulgaris or S. fimbriatum may occupy the water holes. On the contrary, drier conditions are exposed by Acaena magellanica and Poa pratensis. (b) Carex gayana wet peaty meadows are dense, herbal communities, developed on saturated soils. They may occasionally have individuals of N. antarctica better developed than those of the Sphagnum bogs. No water comes from the soil when stepped on, even though it may be very humid. C. gayana dominates with covers up to 75–100% and it is accompanied by Gentianella magellanica, Trifolium spicatum, Phleum alpinum, Poa pratensis, and Scirpus nevadensis. According to Roivainen (1954), the presence of mosses of the genera Tortula and Brachytecium is frequent. As it happens in the Carex curta fens, tussocks are formed by plants with drier requirements like N. antarctica, Chiliotrichum diffusum, and Primula magellanica. In the water hollows H. vulgaris or S. fimbriatum occur. The previously deﬁned groups of peatlands were corroborated by factor analyses, which also incorporates a new plant community (Bolax gummifera). The latter is considered a cushioned meadow, not a peatland. In the S. magellanicum group three subgroups were recognized statistically: (a) commonly hummocks accompanied by M. grandiflorum, being equivalent to raised bogs; (b) dominant S. magellanicum in saturated soils, being equivalent to middle bogs; and (c) dominant S. magellanicum with high percentage of dead plants and intercommunicated ponds of various sizes, being equivalent to impoverished bogs. The C. gayana group shows the most homogeneous plant community. The C. curta group is ﬂoristically related with B. gummifera meadows due to the presence of tussock vegetation, particularly the lichens. Cover diversity and speciﬁc richness were also analyzed by statistics. Table 5.3 indicates that the S. magellanicum community peatlands are the poorest in both aspects, C. curta has intermediate values and C. gayana have the highest values of plant diversity and richness.

Relationships between geomorphology and vegetation Geomorphic units are the physical support where runoff or dammed water, sediments, and nutrients, among others factors, interact with climatic conditions

120

A. Coronato et al.

promoting the development of different types of peatlands. These key elements must be also considered along the evolutionary stages of the landscape because in time they could generate substantial changes in plant communities. Peatlands are more widely distributed over the glacial geomorphologic units than any other types (Fig. 5.6). This may be related to the major extent of glacial landforms over the landscape, as seen in the map (Fig. 5.2), or due to the existence of particularly favorable conditions for peatland development on them. Analyzing peatland extent, Figure 5.7 shows that the ‘paleochannels’ (9.8% of the total surveyed region) have an extent of 38% of the total peatlands, the ‘kettle holes’ (9.7% of the total surveyed region) have 9%, and the ‘eastern kame terraces II’ (11.3% of the

Figure 5.6. Peatland development in relation to geomorphologic environments.

Figure 5.7. Mires and bogs extent in the various geomorphologic units.

Geomorphologic emplacement of Fuegian peatlands

121

total surveyed region) has 7.7%. These data show that 54.8% of the peatland extent is developed only over 30.8% of the total surveyed surface, and it explains that favorable conditions for peatlands development exist in any of the glacial geomorphologic units. The other glacial units, are mostly occupied by deciduous forest or cushioned meadows, and only less than 20% of their extent is covered by peatlands. Fluvial and alluvial geomorphologic units as delta and valley bottom, fan and lower slopes, are occupied by peatlands in more than 50% of the cases, but they represent only a small percentage when the total extent of the region considered (Fig. 5.6). Only the valley bottom units are more widely extended (17% of the total extent) than the others. Figure 5.7 shows intermediate values of peatlands occupation of the lake beach, ﬂuvial terraces, glacioﬂuvial plane, and glacioﬂuvial cone, whereas those in rocky hills have the lowest values. In the alluvial-low-slope, geomorphologic unit where nutrients are supplied by sheet runoff, the Cyperaceae fens and the raised bogs are developed in similar proportions (Fig. 5.8). This situation might be unexpected because of the different

Figure 5.8. Percent distribution of peatlands in alluvial and ﬂuvial geomorphic units.

122

A. Coronato et al.

nutrient-requirement of these types of peatlands. The fact is that bogs are located far from the slopes and the strong inﬂuence of running water or in poorly drained areas. Wherever this happens, the change of fen to bog occurs by a gradual invasion of Sphagnum that generates hydraulic barriers isolating the bog from the groundwater nutrient supply. Success in this process is pointed out by the presence of raised Sphagnum-bearing peatlands. The development of impoverished bogs is the consequence of the nutrient supply from occasional runoff or by eolian action during dry summers. This process often occurs in those peatlands located in grazing areas or along dirty roads. Under the subsurﬁcial layers of humiﬁed mosses, thick layers of Carex peat is present in both types of Sphagnum bogs, in the impoverished and in the raised ones. In the low-gradient alluvial fans units, there is an input of water, sediment, and nutrient from shallow channels and snowmelt runoff. The availability of water on less permeable silt-clay sediments generates almost permanently ﬂooded areas that favor the Cyperaceae fens development instead of bogs (Fig. 5.8). In this geomorphic unit, the bogs are located in ponds formed between the fan and basal moraines, isolated rocky hills, or glacioﬂuvial plains units. Although other plant communities are better developed than peatlands in the ﬂuvial terraces, similar proportions between Cyperaceae fens and raised bogs were recorded (Fig. 5.8). The latter are developed in terraces located between the rivers and the valleys slopes (ﬂanks), away from the river ﬂooding and sheet runoff. Floods supply sediments and nutrients not favorable for Sphagnum colonization. Peatlands development on ﬂuvial terraces is related to the inﬂuence of river ﬂooding. Peatlands can start their development as a minerotrophic Cyperaceae fen and later become a bog when the stream excavates deep enough into its bed and ﬂooding does no longer bring nutrients over the banks. In any case, peatlands located along high riverbanks are subject to signiﬁcant ﬂuvial erosion. Plant communities developed in the valley bottom geomorphologic units are represented by bogs, Cyperaceae fens, and other types (Fig. 5.8). Fens are located along the lower riverbanks and are most abundant due to the mineral content of ﬂoodplain soils renewed annually by ﬂooding. Raised bogs are located away from the streams, near the terraces, where the input of water is mostly provided by precipitation instead of runoff or ﬂuvial water. When this situation is modiﬁed, raised bogs become impoverished, a process that often happens in those bogs affected by variations in the water-table level due to seasonal changes in the ﬂuvial basal ﬂow. Deltas only support fens (Fig. 5.8) due to their permanent water, sediments, and nutrients availability provided by several ﬂuvial streams. Peatlands and other plant communities developed on glacial geomorphologic units are shown in Figure 5.9. (1) The glaciolacustrine plain hosts a very small proportion of peatlands: the Cyperaceae fen is the more representative, occupying the 7% of the total surface, and only very few raised bogs are present. This fen, named Vega Varela (Fig. 5.1), is composed of Marsippospermum grandiflorum in the ﬁrst 0.5 m and by 2 m of continuous mixed Carex sp. and Bryales peat (Fig. 5.10).

Geomorphologic emplacement of Fuegian peatlands

Figure 5.9. Percent distribution of peatlands in glacial geomorphic units.

123

124

A. Coronato et al.

Figure 5.10. Stratigraphic proﬁles in two types of peatlands.

(2) In the glacioﬂuvial plain and cone, the Cyperaceae fens are dominant; they extend over vast portions of these ﬂat, elongated, and open units. Isolated bogs are also developed in central areas of these units; Sphagnum hummocks growing over the surﬁcial Graminaceae could be seen over the big fens, in those sectors poorly affected by water containing sediments and nutrients and when mosses develop hydraulic barriers. The stratigraphic proﬁle of these bogs shows less than 20% of surﬁcial Sphagnum peat. (3) In the glacioﬂuvial cone, bogs are located in quasi-circular depressions, originally small ponds created by ice-disintegration processes that took place during deglaciation. In the kettle holes units, the Sphagnum becomes impoverished due to excessive water supply, and lichens and shrubs appear. In those kettle holes that are still partially occupied by ponds or not fully closed by knolls, the dominant peatland is raised bogs. These bogs become impoverished when sediments deﬂated from adjacent dirty roads affect them. Cyperaceae are also developed in this unit, not in the holes, but in the almost ﬂat areas, close to the glacioﬂuvial plain. (4) The kame-terrace units have a small surface occupied by peatlands (Fig. 5.9). Isolated raised bogs are most abundant, occupying ancient ponds or ﬂooded areas. Cyperaceae fens are accompanied by raised and middle Spahgnum hummocks in the central sectors, indicating the transition toward ombrotrophic conditions. (5) Raised bogs are dominant in the paleochannel geomorphic units (Fig. 5.9), followed by impoverished Sphagnum and Cyperaceae. Peatlands of these units

Geomorphologic emplacement of Fuegian peatlands

125

Figure 5.11. Depth of peatlands in the various geomorphologic units.

are extensive, narrow, and very irregular in shape; they follow an inactive drainage pattern, inherited from the meltwater stream system during deglaciation times. In the areas where the paleochannels become narrow and minerals and nutrients are supplied by runoff waters coming from the surrounding moraines, Graminaceae fens are present. The stratigraphic proﬁle of La Correntina 2 bog (Fig. 5.1) represents the typical transition from fens to bogs, which occurs in environments where annual precipitation is higher than 400 mm. Figure 5.10 shows that Carex sp. was present since the beginning of the fen development, occupying the 80% of the stratigraphic proﬁle. The transition between Graminaceae and mosses takes place within a 0.5 m thick layer, and only the upper 6% of the proﬁle corresponds to S. magellanicum, with low humiﬁcation and similar proportion between dead and living ﬁbers (Fig. 5.10). The stratigraphic composition of La Correntina 2 bog is representative of those peatlands located on geomorphologic units in which the availability of running water, sediments, and nutrients changed with landscape evolution. This proﬁle also demonstrates the evolution from minerotrophic to ombrotrophic conditions, when Sphagnum peat is isolated from groundwater nutrient supply sources in environments with sufﬁcient precipitation. The thinner and poorly humiﬁed layers of Sphagnum suggest that the present ombrotrophic conditions have been reached recently. The average peatlands depth is about 2.86 m (Fig. 5.11). The peatlands low to medium depth (2.1–4 m) includes are the most common ones and occur on glacial geomorphologic units, such as paleochannels, kames, kettle-holes, and basal moraines. The deepest peatlands reach 6–8 m in depth, but they are scarce. The deepest fens have developed in the valley-bottom units because they are favored by water oversupply during ﬂooding or basal ﬂow, and water-table ﬂuctuations along the streams. The deep

A. Coronato et al.

126

Table 5.4. Late-Glacial and Holocene basal radiocarbon ages of selected bogs of the studied area. The 14C analysis were made by QUADRU, CSIR Environmentek (Pretoria, South Africa). Basal peat samples

Depth (m)

Radiocarbon age (14C yr. B.P.)

Laboratory code

Lago Fagnano 1 Las Lengas La Correntina 1 San Pablo

4.57 3 4.5 3.76

8920770 10030760 11830780 103207100

8515 8518 8521 8516

Carex fen of the delta unit is geomorphologically controlled by tectonic subsidence because it is located on top of an active fault. Subsidence favors the damming of the ﬂuvial water, shallow pond formations, accumulation of sediments, and continuous colonization of peat-forming plants. The deepest bogs (7 m) have developed in ponds formed in glacially eroded depressions of the rocky hills units.

Conclusions The geomorphologic distribution of Cyperaceae fens and Sphagnum bogs indicates that extensive and open fens are widely distributed over present and past ﬂuvial environments, whereas isolated, closed, and small bogs are located on basal moraines, kettle holes, and glacioﬂuvial plains. In geomorphologic units, such as paleochannels and kames, bogs have grown over the Cyperaceae fens due to isolation from nutrients, especially by their own development of a hydraulic barrier. When this important change took place is not yet unknown, but according to the limited thickness of the Sphagnum layers it should have happened in the recent past, under climatic conditions similar to the present ones. This means that, although paleochannels and kames were no longer occupied by running meltwater since deglaciation times (ca. 16,000 14C yr BP), for several thousands of years they remained environments more favorable for fen rather than for bog development because of persistence of warmer and drier climate conditions than the present one. The development of bogs can be associated with the expansion of Nothofagus sp. forest. This occurred in central Tierra del Fuego between 900 and 300 14C yr BP, when an increase in humidity is indicated by the San Pablo palynological proﬁle (Heusser and Rabassa, 1995; Rabassa et al., 2006 – this book, Ch. 6). The glacial and postglacial (ﬂuvial, alluvial, and littoral processes) history of the landscape contributed to peatlands development: their type, extension, and depth. Abandoned river channels, shallow ponds, or poorly drained areas were colonized by peat forming plants since Late Glacial (16,000–12,000 14C yr BP) to early Holocene times. Peatlands are still actively forming today. The basal age of the still active peatlands of La Correntina 1, Las Lengas, San Pablo, and Fagnano 1 ranges from about 12,000 to 9000 14C yr BP (Fig. 5.1; Table 5.4). Closed peatlands in glacial landscape are useful for the geomorphologic understanding of the Late Pleistocene

Geomorphologic emplacement of Fuegian peatlands

127

glacier recession, vegetation colonization, and landscape evolution (Rabassa et al., 2006 – this book, Ch. 6). The contemporaneous analysis of geomorphologic settings and the botanic classiﬁcation of peatlands provide adequate criteria to establish peatlands characteristics, such as topographic proﬁles, distribution of ponds, plant-cover type, and the depth, volume, and stratigraphic composition of peat. This is useful knowledge for planners for a wise-use strategy design and for ﬁrst estimations of peat resources.

References Auer, V., 1965. The Pleistocene of Fuego-Patagonia. Part IV: bog proﬁles. Ann. Acad. Sci. Fennicae. Series A III. Geol.-Geogr. (Helsinki) 80, 1–165. Bonarelli, G., 1917. Tierra del Fuego y sus Turberas. Ministerio de Agricultura de la Nacio´n Seccio´n Geologı´ a, Mineral. Miner. (Buenos Aires) XII (3), 119pp. Borromei, A.M., 1995. Ana´lisis polı´ nico de una turbera holoce´nica en el Valle de Andorra, Tierra del Fuego, Argentina. Rev. Chilena de Historia Natural 68, 311–319. Braun-Blanquet, J., 1979. Fitosociologı´ a. Bases para el estudio de las comunidades vegetales. Editorial Blume, Barcelona, 802pp. Coronato, A., Meglioli, A., and Rabassa, J., 2004. Glaciations in the Magellan Straits and Tierra del Fuego, southernmost South America. In: Ehlers, J. and Gibbard, P. (Eds), Quaternary Glaciations: Extent and Chronology. Part III. Quaternary Book Series. Elsevier Publishers, Amsterdam, pp. 45–48. Corte, A., 1996. Geocriologı´ a. El Frı´ o en la Tierra. Ediciones Culturales de Mendoza, Mendoza, pp. 344–346. Greene, S., 1964. The vascular ﬂora of South Georgia. British Antarctic Survey Scientiﬁc Reports 45, pp. 1–57. Guin˜azu´, J.R., 1934. Los depo´sitos de turba de Tierra del Fuego. Direccio´n de Minas y Geol. (Buenos Aires) 103, 119pp. Heusser, C.J., 1989. Late Quaternary vegetation and climate of Tierra del Fuego. Quatern. Res. 31, 396–406. Heusser, C.J., 1989. Climate and chronology of Antarctica and adjacent South America over the past 30,000 yr. Palaeogeogr. Palaeoclim. Palaeoecol. 76, 31–37. Heusser, C.J., 1998. Deglacial paleoclimate of the American sector of the Southern Ocean: Late Glacial–Holocene records from the latitude of Canal Beagle (551S), Argentine Tierra del Fuego. Palaeogeogr. Palaeoclim. Palaeoecol. 141, 277–301. Heusser, C.J., 2003. Ice Age Southern Andes. A Chronicle of Palaeoecological Events. Elsevier, London, 240pp. Heusser, C.J. and Rabassa, J., 1987. Cold climatic episode of Younger Dryas age in Tierra del Fuego. Nature 328, 609–611. Heusser, C.J. and Rabassa, J., 1995. Late Holocene forest-steppe interaction at Cabo San Pablo, Isla Grande de Tierra del Fuego, Argentina. Quatern. S. Am. Antarctic Peninsula 9, 179–188. Mauquoy, D., Blaauw, M., van Geel, B., et al., 2004. Late Holocene climatic changes in Tierra del Fuego based on multiproxy analyses of peat deposits. Quatern. Res. 61, 148–158. Moore, D., 1983. Flora of Tierra del Fuego. Anthony Nelson Publisher, London, 396pp. Rabassa, J., Coronato, A., Heusser, C., et al., 2006 (this book, Chapter 5). The peatlands of Argentine Tierradel Fuego as a source for paleoclimatic and paleoenvironmental information. In: Martini, I.P., Matı´ nez Cortizas, A., and Chesworth, W. (Eds), Peatlands: Evolution and Records of Environmental and Climatic Changes. Elsevier, Amsterdam. Rabassa, J., Coronato, A., and Roig, C., 1996. The peat-bogs of Tierra del Fuego, Argentina. In: Lappalainen, E. (Ed.), Global Peat Resources. International Peat Society Publisher, Jyska¨, Finland, pp. 261–266.

128

A. Coronato et al.

Roig, C. and Collado, L., 2004. Los turbales patago´nicos, ventanas N1 6 y 7. In: Blanco, D. and de la Balze, V. (Eds), Los turbales de la Patagonia. Bases para su inventario y la conservacio´n de su biodiversidad. Wetlands International, Buenos Aires, 19, pp. 62–71. Roig, C., Coronato, A., Collado, L., et al., 2001. Peat classiﬁcation in Central Tierra del Fuego, Argentina, South America. Irish Raised Bogs Conservation, Utilization and After-Use Conference, Potlaoise, Ireland, 10pp. Roig, C., Roig, F., and Martı´ nez Carretero, E., 2004. Los turbales patago´nicos, ventana N1 5. In: Blanco, D. and de la Balze, V. (Eds), Los turbales de la Patagonia. Bases para su inventario y la conservacio´n de su biodiversidad. Wetlands International, Buenos Aires, 19, pp. 55–61. Roig, F., Anchorena, J., Dollenz, O., et al., 1985. Las comunidades vegetales de la Transecta Bota´nica de la Patagonia Austral. In: Boelcke, O., Moore, D., and Roig, F. (Eds), Transecta bota´nica de la Patagonia Austral. CONICET – Instituto de la Patagonia-Roy. Soc, London, Buenos Aires, pp. 350–519. Roig, F.A., 1998. La vegetacio´n de la Patagonia. In: Correa, M. (Ed.), Flora Patago´nica, VIII, Parte 1. Instituto Nacional de Tecnologı´ a Agropecuaria (INTA), Buenos Aires, pp. 48–166. Roivainen, H., 1954. Studien U¨ber Die Moore Feuerlands. Soc. Bot. Fennicae ‘‘Vanamo’’ (Helsinki) 28, 1–205. Tuhkanen, S., 1992. The climate of Tierra del Fuego from a vegetation geographical point of view and its ecoclimatic counterparts elsewhere. Acta Bot. Fennica (Helsinki) 145, 64.

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

129

Chapter 6

The peatlands of Argentine Tierra del Fuego as a source for paleoclimatic and paleoenvironmental information J. Rabassa, A. Coronato, C.J. Heusser, F. Roig Jun˜ent, A. Borromei, C. Roig and M. Quattrocchio

Introduction Tierra del Fuego is the southernmost inhabited region in the world, closest to the Antarctic Peninsula and the Circum-Polar Current, the oceanic factor that controls most of the climate around our planet. Peatlands have developed in the central and southern part of Argentine Tierra del Fuego, which allow reconstruction of uppermost Pleistocene–Holocene history of the land. These Fuegian (meaning ‘of Tierra del Fuego’) peatlands are unique environments because of their geographical location, ﬂoristic and hydrological nature, ecosystem signiﬁcance, and paleoenvironmental and paleoclimatic records (Bonarelli, 1917). Indeed, these peatlands can be considered as environmental data banks, since the sediments accumulated in their layers contain information related to the environmental, ecological and climatic conditions existing in the surrounding region. Owing to their continued sedimentation pattern, a continuous paleoclimatic record can commonly be obtained, which may be constrained with great precision and reliability by means of radiocarbon dating of wood fragments or of the peat itself. Fuegian peatlands are true cathedrals of the paleoenvironmental and paleoclimatic knowledge of southernmost South America. Therefore, a great effort is needed to preserve them since these ecosystems will work as continuous data archives in the future. Here we present a short review of what is known about the Fuegian peatlands, their nature, distribution, paleoclimatic and paleoenvironmental proxy content, and their relationship with the human settlement of Tierra del Fuego.

Geographic setting The Isla Grande de Tierra del Fuego is the largest of the Fuegian Archipelago islands, located at the southernmost end of South America, between 531–551 S ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09006-7

130

J. Rabassa et al.

latitude and 661–741 W longitude (Fig. 6.1). Argentine Tierra del Fuego occupies the eastern half of this island; the other half belongs to Chile. Geologically, two crustal plates separated by the tectonically active Magellan transcurrent fault form the

Figure 6.1. Location map of Tierra del Fuego, southernmost Patagonia and Fuegian channels.

The peatlands of Argentine Tierra del Fuego

131

island. Tertiary marine sedimentary rocks that form high plains, isolated hills and low ranges overlie the South American Plate. On the Scotia Plate, the Fuegian Andes are characterized by upper Paleozoic greenschists, a Cretaceous batholith and Mesozoic (early–middle Cretaceous) acid lavas and low-grade, metamorphosed marine sedimentary rocks. Upper Cenozoic glaciers have molded the landscape, with vast plains in the northeastern part of Argentine Tierra del Fuego and rugged mountainous topography to the south. During the Late Pleistocene, only the central (Lake Fagnano area) and southern parts were glaciated (Coronato et al., 2004). Remnant glaciers still exist in Tierra del Fuego: larger ones, an extensive mountain ice sheet with large outlet glaciers that reach down to sea level, in the western (Chilean) portion of the Fuegian Andes, and small alpine glaciers on the Argentine mountains (Planas et al., 2002). The regional climate of Argentine Tierra del Fuego is cold, temperate in the south and temperate-oceanic in the north, with an overall mean annual temperature of ca. 51C. Moisture is derived primarily from the south and southwest, and the precipitation distribution is affected by the mountain ranges. Rainfall shows a strong gradient from the southwest, with 600 mm/yr, to the northeast, with less than 300 mm/yr. An east–west precipitation gradient is also evident in the southern part; that is, precipitation increases from the eastern mouth of the Beagle Channel toward the eastern end of the Isla Grande and Staaten Island (Isla de los Estados; up to 1000 mm/yr), due to the penetration of wet, oceanic air masses not obstructed here by the mountain ranges. Maximum precipitation in Argentine Tierra del Fuego occurs in the higher part of the Fuegian Andes, near the border with Chile and at the southernmost end (Fig. 6.1). There, the maximum development of peatland occurs. In general, the present vegetation of the central and southern part of Argentine Tierra del Fuego belongs to the Sub-Antarctic Deciduous Forest dominion from sea level to 700 m asl, to the High Andean Alpine Tundra above this altitude.

Methodology for peatland studies The Fuegian peatlands have been studied so far from the standpoint of pollen, spore, total organic matter, volcanic ash and charcoal particle records. More recent, ongoing research has been oriented toward geochemical studies and analysis of peat ash containing microparticles (mostly volcanic ash and dust particles). For these studies, peatlands are cored to substrate, generally with a Russian-type sampler. The cores are sub-sampled every 10- or 5-cm depth and the samples are later processed in the laboratory, according to the objective of the study (Heusser, 2003). The absolute age of the peat is determined by radiocarbon dating, and the age of the basal peat is taken to indicate the approximate time of initiation of the peatland development. Pollen ﬂux studies are based in estimating the actual amount of pollen rain falling on the surface of the bog (Heusser, 2003). This is done by means of the artiﬁcial incorporation of a known amount of exotic pollen (that is, pollen from a foreign species, not existing in the area today or in the geological past) during the lab processing of each sample. This allows calculation of the actual rate of pollen rain on the bog in the past, since the relationship of the known amount of exotic pollen to the

132

J. Rabassa et al.

original pollen grains, found in each sample, provides a fair approximation to the environmental conditions at such time. A weak absolute pollen rain of all species represented in the record indicates harsh climatic conditions. Considering an average peat accumulation rate of about 0.1–0.025 cm/yr, the 5-cm samples provide a time-resolution between 50 and 200 years, depending on the characteristics of the bog, but most likely close to 100 year periods. Knowing the type and distribution of the present vegetation, their ecological and reproductive characteristics, and the mode of dispersal of pollen and spores, it is possible to use them as analogs to reconstruct ancient plant communities and thus, ancient paleoenvironmental and paleoclimatic conditions for speciﬁc time periods.

Glacial history of Tierra del Fuego and basal peat chronostratigraphic data Current conditions in Tierra del Fuego are characterized by the existence of a mountain ice-sheet in the Darwin Cordillera (2500 m asl; 551 S latitude, 691 W longitude; Western Tierra del Fuego, Chile), one of the largest ice bodies in the Southern Hemisphere outside of Antarctica. Large ice tongues (outlet glaciers) discharge from it, reaching or getting close to sea level. In the eastern (Argentine) sector of the island, the ice bodies are restricted to cirque glaciers, a few small, high altitude valley glaciers and limited snowﬁelds in the higher zones. Repeated glaciations took place in Patagonia and Tierra del Fuego since the latest Miocene (Mercer, 1976; Clapperton, 1993). During the Quaternary, several large glaciers descended a number of times from the Darwin Cordillera to cover all or part of Tierra del Fuego. Those which are most important for the landscape evolution of Eastern Tierra del Fuego were the Magellan Glacier, the largest glacier that ever existed in the Southern Hemisphere during Cenozoic times outside of Antarctica, the Fagnano Glacier and the Beagle Glacier (Fig. 6.1; Porter, 1990; Meglioli, 1992; Rabassa et al., 1992, 2000; Clapperton, 1993; Isla and Schnack, 1995). Caldenius (1932) mapped the distribution of Fuegian and Patagonian glaciations. The glacial features indicate that, probably, the entire island was ice-covered before 1 million years ago. The Great Patagonian Glaciation (GPG), the maximum expansion in Extra-Andean Patagonia, occurred between 1.15 and 1.01 million years ago, during the oceanic oxygen isotope stages (OIS) 34–30 (Ton-That et al., 1999; Rabassa and Coronato, 2002). All subsequent glaciations were less extensive. The number of glacial advances that took place in Tierra del Fuego is still a matter of debate (Rabassa et al., 1990b; Meglioli, 1992). In any case, during the glacial maxima, including the Last Glacial Maximum (LGM, ca. 24,000–25,000 cal yr BP; Kaplan et al., 2004), the exposed surface of Tierra del Fuego was at least twice the present one due to lowering of sea level, allowing for a much larger degree of continentality than today (Rabassa et al., 1992, 2000). The marine coast displacement was at least 150 km eastwards. For this reason, the periglacial environments were characterized by a tundra environment and, farther eastwards, by an extensive herbaceous and shrubby steppe, similar to that existing today in continental Patagonia. Finally, somewhere in easternmost Tierra del Fuego, refugia of the Fuegian forest may have persisted,

The peatlands of Argentine Tierra del Fuego

133

which allowed recolonization during interglacial times (Fig. 6.1; Coronato et al., 1999). The Upper Pleistocene Darwin Cordillera ice cap and the associated Beagle Channel and Lake Fagnano lobes/outlet glaciers are most important in shaping the landscape over which the most important latest Pleistocene/Holocene peatlands of Argentine Tierra del Fuego formed (Figs. 6.1, 6.2). Conversely, the basal peat dates of the various peatlands allow an accurate reconstruction of the deglaciation stages of these areas, and of the Beagle Channel in particular. The glacial and periglacial processes left a legacy of steeply eroded mountain ﬂanks, deep entrenched valleys, a local drumlin ﬁeld (Harberton area; Rabassa et al., 1990c, 1992, 2000), several end moraines that locally dammed lacustrine settings and a series of alluvial fans (Coronato and Roig, 1999; Coronato et al., 2006 – this book, Ch. 5). The peatlands development was intimately associated with such landscape. They developed preferentially by the damming of the surﬁcial runoff in the deglaciated terrains, forming small lakes and ponds, which were later occupied by aquatic vegetation, mostly mosses and grasses, from which peat accumulation started. During the Late Pleistocene, the glaciers started to recede signiﬁcantly from the southern part of Argentine Tierra del Fuego around 15,000 or 16,000 14C yr BP. The dates obtained from the basal peat layers provide an approximate time for the

Figure 6.2. Glacial map of Tierra del Fuego (after Coronato et al., 1999) and location of sites where basal radiocarbon data were obtained from peatlands for minimum ice recession age (ka ¼ year 103 BP; Ma ¼ year 106 ago).

134

J. Rabassa et al.

progression of the deglaciation and for postglacial emersion of the coastal areas (Fig. 6.2). Peatlands started forming in the eastern area at least about 15,000 14C yr BP (Heusser, 1989a, 2003), but most in the Ushuaia area are only as old as 12,000–13,000 14C yr BP, whereas in the Fuegian Andean valleys they are less than 10,000 14C yr BP (Coronato, 1990, 1993, 1995). The complex deglaciation of the Beagle Channel valley could be deciphered because of the information provided by the peatlands basal ages. For instance, a ﬁrst event of deglaciation is shown by the Harberton Bog basal radiocarbon age of 14,6807100 14 C yr BP (Rabassa et al., 1990a), somewhat 50 km west of the terminal, LGM Moat moraines. Probably then the ice front reached the Isla Gable rise, which it later abandoned in a second deglaciation event, as suggested by the basal 14C age of the Caleta Ro´balo bog (12,700790 14C yr BP; Heusser, 1989b). Then, the main Beagle Glacier receded from the cirques, allowing their glaciers to expand down slope. Radiocarbon dating of basal peat of 12,060760 14C yr BP at the Pista de Ski moraine (Ushuaia, at ca. 300 m asl) suggests that this retreat phase probably peaked ca. 12,000 14 C yr BP, when a relative maximum of arboreal pollen was reached eastwards, at Puerto Harberton (Rabassa et al., 1990a), at 11,7807110 14C yr BP. In spite of the formation of the Late Glacial lateral moraines, the basal 14C ages of the Ushuaia 2 (80 m asl; 12,430780 yr BP) and Ushuaia 3 (10 m asl; 12,100750 yr BP) bogs show that the ice would have already disappeared from these sites allowing the formation of ice-marginal, lacustrine environments (Heusser, 1998). The similarity of the bog basal ages between 300 and 10 m asl in Ushuaia suggests that the ice recession had taken place in a single phase. Although the pollen proﬁles in these peatlands show evidence of cooling between 11,000 and 10,000 14C yr and subsequent vegetation changes (Heusser and Rabassa, 1987), perhaps the climatic conditions had not been harsh enough so as to alter the Beagle Glacier dynamics and to allow the ice stabilization and interrupt the general headward recession. The ca. 10,000 14C yr BP glacial retreat was deﬁnitive: basal peat layers of Punta Pingu¨inos in Ushuaia (20 m asl) and Bahı´ a Lapataia (20 km westwards, 18 m asl) show 14C ages of 10,080 yr BP (Rabassa et al., 1990a), a condition observed also for the glaciers that were tributaries to the glaciation axis located in the easternmost end of the island – 661 W longitude, Bahı´ a Aguirre – where the basal age of fossil peat reached 10,920770 14C yr BP (Rabassa et al., 2000). The Late Glacial–lower Holocene peat depositional sequence has been recently studied concerning its palynological and paleoclimatic aspects (Borromei et al., in press), conﬁrming the existence of a cold period around 10,000 14C yr BP. The geomorphologic evidence found in the Fuegian Andes indicates that the deﬁnitive deglaciation process would have started after 10,000 14C yr BP. Recession followed the Late-Glacial maxima and evidence for several Neoglacial readvances are observed in the cirques, but no absolute chronology has been established yet.

Peat accumulation rates Rabassa et al. (1989) calculated the relationship of age/depth at several active peatlands of Patagonia and Tierra del Fuego. Proﬁles of biogenic deposits formed in peatlands and alluvial and lacustrine deposits were analyzed to ascertain their

The peatlands of Argentine Tierra del Fuego

135

accumulation rate. The maximum accumulation rate for a given period has been established in 3.0 mm/yr, and the mean accumulation rate for one single site has been estimated between 0.51 and 0.93 mm/yr, at least during the Late Quaternary (see diagrams in Rabassa et al., 1989). It reﬂects among other factors, climatic changes and the compaction processes that the sediments have undergone. Statistically, most of the best-ﬁtted regression equations are of the type y ¼ axb and the determination coefﬁcients obtained are very large (40.84, most of them 40.9), indicating that almost the entire variance may be explained by these functions. The b coefﬁcient, always different from 1, indicates that the accumulation rate is not linear. These coefﬁcients also suggest that, at least in the studied sequences, no signiﬁcant hiatus in the sedimentary process has ever taken place. These data also suggest that it is possible to extrapolate with some conﬁdence the age of a peat layer intermediate between dated horizons. The rate of growth of the bog may vary with time, and it is probably related to environmental and climatic conditions. Two episodes of high accumulation rate have occurred in this region, during the latest Pleistocene (ca. 10,000 14C yr BP) and the Late Holocene (between 2500 and 3000 14 C yr BP). These high accumulation rates correspond to climatic conditions that favored peat accumulation, most likely due to higher annual precipitation. Similarly, a lower accumulation rate took place around ca. 6000 14C yr BP, perhaps due to lower primary productivity during Hypsithermal times. Probably, mean annual precipitation has been the dominant factor controlling growth and evolution of peatlands, and therefore, the accumulation rate of the organic sediments in them. Heusser (2000) applied the same criteria to the Puerto del Hambre (Magellan Strait, Chile) mire, obtaining a similar, overall sedimentation rate of 0.61 mm/yr, for the last 14,700 14C yr BP.

The occurrence of Holocene tephras in Fuegian peatlands Volcanic ashes accumulate and they are well preserved in peatlands, particularly in bogs. These ashes have been blown from distant sources in continental Patagonia, since no tephra-producing volcanoes are known in the Fuegian archipelago. Up to four different ash layers were already described by Auer (1956, 1958, 1965), and correlated to various tephra beds he had recognized in Patagonia. The oldest one was of Late Glacial age, and the other three were deposited during the Holocene. This stratigraphic framework is today under revision, since the pyroclastic activity on the continent seems to have been much more complex than previously thought. A sequence of ash events was presented by Stern (1990), who described and geochemically identiﬁed ash falls as pertaining to pyroclastic eruptions coming from the Andean volcanoes Reclus, Aguilera, Lautaro, Hudson and Barney, in continental South America. Many different tephra layers have been recorded in Fuegian bogs between 14,150 and 2700 14C yr BP (Heusser and Rabassa, 1995; Heusser, 2003), but the volcanic activity of the Patagonian volcanoes has continued until present times,

136

J. Rabassa et al.

as the large eruption of Mt. Hudson in 1991, whose wind-blown ashes occur at or very near the surface of the Fuegian peatlands.

Coastal wetlands The coastal wetlands of Tierra del Fuego have not been as thoroughly investigated as the inland peatlands. The only major exception is the La Misio´n Bog, north of Rı´ o Grande, along the Holocene marine terrace. This wetland has been studied by Auer (1959, 1974) and Markgraf (1980). At this site, wetland stratigraphy displays an intercalation of marine sediments corresponding to the mid-Holocene transgression (Porter et al., 1984). A Holocene lake (at a present level of 5.7 m below high tide level) was ﬂooded by the marine transgression about 9000 14C yr BP and a tidal ﬂat developed at an altitude of 0.9 m above high tide level at 4000–2000 14C yr BP. Other important sites related to coastal wetlands are the alluvial–littoral deposits described at Bahı´ a Lapataia (Rabassa et al., 1986; Gordillo et al., 1993), where a thin peat layer is covering metamorphic rocks and is overlain by lower Holocene marine deposits, whose shells gave a radiocarbon age of 8200 yr BP. The underlying peat bed has yielded a radiocarbon age of 7700 yr BP at its top layers. Thus, a minimum 500-year reservoir effect can be established for radiocarbon dating of marine shells in this portion of the Beagle Channel, since the radiocarbon age of the underlying peat layer may be taken as much closer to the true age. At Rı´ o Varela, just east of Estancia Harberton, mid-Holocene peat and alluvial deposits, presently under high-tide sea level (Grill et al., 2002), have proven that different parts of the Beagle Channel have had different tectonic behavior during the Late Holocene (Bujalesky et al., 2004). This fact is also indicated by alluvial and organic deposits in former lagoon deposits, including remnants of a submerged forest at Bahı´ a Sloggett, eastern Tierra del Fuego (Rabassa et al., 2004).

Climatic variability as demonstrated by proxy elements Pleistocene pollen records The Upper Pleistocene record at Lake Fagnano has been described by C. J. Heusser (in Bujalesky et al., 1997; Heusser, 2003). It corresponds to fossil peat beds, intercalated in lacustrine/diatomite deposits, marginal to a major lowland glacier, perhaps during the termination of a glacial stage, most likely OIS six or four. These fossil peat beds have been radiocarbon dated at 38,000 yr BP and greater than 51,000 yr BP, suggesting a Late Pleistocene age (Fig. 6.3). The pollen record shows, particularly when compared to the present pollen rain data, that the forest was absent from this region during such period. The Nothofagus pollen content is almost absent, even though tree pollen represents today almost 100% of the sample. This means that the forest had been wiped out from the central portion of Tierra del Fuego during glacial times, retiring eastwards to a still uncertain refuge. The forest had been then replaced by a tundra–shrubby steppe association (Heusser, 2003).

The peatlands of Argentine Tierra del Fuego

137

Figure 6.3. The pollen record at Lake Fagnano fossil peat beds (after Bujalesky et al., 1997) (in this ﬁgures, ka ¼ 14C yr 103 BP).

Late Glacial climate, according to the pollen stratigraphy at several studied sites, along the Beagle Channel, was apparently warmer between 14,600 and 13,000 14C yr and between 11,700 and 11,160 14C yr BP. Climate became cooler earlier than 13,000 until 12,000 14C yr BP, and from around 11,160 until 10,200–10,000 14C yr BP. This last cooling episode suggests the development of a Younger Dryas event in this part of the Southern Hemisphere, where the estimated summer temperature was o31C lower than the present at Ushuaia (Heusser and Rabassa, 1987; Heusser, 1998). The Late Glacial pollen signal is characterized by a fast recovery of the Nothofagus forest almost immediately after deglaciation, that is, sometimes before 16,000 14C yr BP. The basal deposits at the Harberton Bog are showing an unusually large content of tree pollen between 14,600 and 14,000 14C yr BP (Rabassa et al., 1990a; Fig. 6.4). This initial pollen content decreases later to two minima, the ﬁrst one between 13,000 and 12,000 14C yr BP and the second between 11,000 and 10,000 14C yr BP, with an intermediate rise. This is also shown by pollen inﬂux diagrams (Heusser, 2003). These two episodes have been interpreted as relatively cold and dry climatic events, with displacement of the Nothofagus forest outside of the region, and its replacement by an impoverished tundra environment (Heusser, 1989a). Similar conditions have been identiﬁed in the Ushuaia 1, Ushuaia 2 and Caleta Ro´balo sections (Heusser, 2003). Whether these Late Glacial cold and dry events can be timely correlated to the North Atlantic Older Dryas and Younger Dryas events is still debated (Heusser and Rabassa, 1987; Heusser, 1989a, b, 2003; Markgraf, 1991).

Holocene pollen records Numerous Holocene peatlands occur all over the island of Tierra del Fuego. The pollen records show that, after the Late Glacial ﬂuctuations already mentioned, milder Holocene climate was established between 10,000 and 9000 14C yr BP, allowing the progressive occupation of the land by the forest. First an open forest condition and later to the present, closed forest environment were established, as

138

J. Rabassa et al.

Figure 6.4. The pollen record at Harberton Bog (after Heusser, 1989b).

indicated by the increase of tree pollen over the steppe shrub and herb species. Several climatic oscillations have been recorded for the Holocene of the island, with a deﬁnitive installation of the forest around 6000 14C yr BP, achieving present climatic conditions. The existence of Neoglacial oscillations is not yet clearly identiﬁable in the pollen record, though detailed studies in mires located in higher valleys of the Fuegian Andes are presently being carried on. The formation of the Nothofagus woodland under the postglacial climatic amelioration started toward 10,000 14C yr BP in southern Tierra del Fuego, but it was not until 5000 14C yr BP that the closed forest developed, under wetter and cooler conditions (Heusser, 1989b), when the peatland rate of sedimentation increased as well (Rabassa et al., 1989). In northcentral Tierra del Fuego, in the forest/steppe ecotone area, a pollen record at Cabo San Pablo (Heusser and Rabassa, 1995) suggests that the behavior of the forest in those areas is different from what has been observed along the Beagle Channel (Fig. 6.5). On the northern slopes of the Fuegian Andes, the woodland stabilization took place around 8000 14C yr BP, but only after 5000 14C yr BP the forest established with open spaces until it reached a more recent, deﬁnitive expansion, as shown in the pollen proﬁle of Cabo San Pablo. In this San Pablo contact zone between the deciduous forest and the steppe, the open forest developed toward ca. 3000 14C yr BP, although with presence of steppe pastures. As indicated by the Nothofagus pollen content, the deﬁnitive expansion of the forest occurred toward 900 14C yr BP, becoming more intense after 300 14C yr BP due to the increasing humidity coming from the Southern Paciﬁc and the weakening of the subtropical anticyclonic cells (Heusser and Rabassa, 1995; Heusser, 2003). This delay in forest recovery shows continentality conditions, which has also been observed in other sections around Lake Fagnano. Recently, Mauquoy et al. (2004) have identiﬁed the existence of a warm period in the Late Holocene of Tierra del Fuego. Changes in temperature and/or precipitation were inferred from different proxy sources, such as plant macrofossils,

The peatlands of Argentine Tierra del Fuego

139

Figure 6.5. The pollen record at Cabo San Pablo (after Heusser and Rabassa, 1995).

pollen, fungal spores, testate amebae and peat humiﬁcation in peat monoliths collected from the Andorra Valley. Mauquoy et al. (2004) report evidence for a period of warming-induced drier conditions from 960 to 1020 AD, which may correspond to the Medieval Warm period in Europe (950–1045 AD) suggesting that this was a global warm period, synchronous in both hemispheres (Borromei et al., in press). In summary, the pollen data suggest that the steppe was dominant in northern Tierra del Fuego until the Nothofagus forest colonization. The paleoenvironment from middle to Late Holocene was characterized by a shrubby steppe with drier and probably colder climatic conditions than the present ones in the area, while the forest would have migrated to the east and south alternatively and irregularly as can be checked from the Lake Fagnano proﬁle, ca. 5000 years BP (Heusser, 1994a); thus, the forest was completely settled close to Onamonte (central Tierra del Fuego, Chile) ca. 1500 years BP and in Cabo San Pablo toward 1000 years BP (see Fig. 6.4; Heusser, 1993; Heusser and Rabassa, 1995). Dendrochronology Numerous tree trunks (many of them still rooted) in excellent degree of preservation were found in two bogs near Ushuaia (Carbajal and Monte Gallinero), among other sites. They have allowed a detailed dendrochronological study by Roig Jun˜ent et al. (1996), supported by several radiocarbon dates and cross-dating techniques, covering at least for the last 3770 years, of which over the last 1500 years consist of a continuous record, one of the oldest in the Southern Hemisphere (Fig. 6.6). These analyses allow reﬁnement of the worldwide tree-ring chronologies that have been used to determine the seasonal variation of air temperature over terrestrial and marine surfaces, precipitation and indexes of atmospheric pressure at sea level during the last

140

J. Rabassa et al.

Figure 6.6. Floating index tree-ring width chronology of the Carbajal Valley (541 440 S latitude, 681 130 W longitude) taken from Nothofagus wood and trunks contained in peatlands layers (after Roig Jun˜ent et al., 1996). The relative position of each record is based on radiocarbon age. The solid underlines correspond to 14C age (years BP7sigma) and the broken lines correspond to the range of age according to the Northern Hemisphere calibration curve. The AD years of horizontal scale are derived directly from living trees or from radiocarbon ages of dead wood calibrated following the Northern Hemisphere curve because there are not yet calibration curves for the Southern Hemisphere. (Diagram (a) continues into diagram (b) for older times).

centuries (Boninsegna et al., 1990; Boninsegna, 1992). The chronologies, obtained in the Southern Patagonian forests, rarely extend over 300 years; nevertheless, it is possible to obtain millenary chronologies by means of wood preserved in peatlands, as previously indicated by Bonarelli (1917) and Auer (1965). These ﬁndings corroborate the potential of this region to develop longer worldwide chronologies by using these subfossil woods (Roig Jun˜ent et al., 1996; Villalba et al., 1997). Bogs and archeology Human colonization in Tierra del Fuego dates from ca. 11,000 14C yr BP, when a group of hunters occupied the NW portion of the island, as shown by cultural and faunal remains found in the Tres Arroyos site (Massone, 1987; Borrero, 1999). At

The peatlands of Argentine Tierra del Fuego

141

that time, the Isla Grande of Tierra del Fuego was still part of the continent (McCulloch et al., 1997). These hunters reached the island from southernmost Patagonia, before the deﬁnitive retreat of the glaciers, and became isolated from their original populations in the continent after the opening of the Magellan Straits. An open steppe-environment is indicated by the faunal species recorded as well as the Bahı´ a Inu´til pollen proﬁle that shows a high proportion of steppe elements (Heusser, 1994a); this could have been the space where these colonizers moved. However, early settlement on the Isla Grande of Tierra del Fuego is suggested by charcoal particles detected in the aforementioned pollen proﬁles (Heusser, 1987, 1994b). Fire in the landscape of this region, where no volcanoes are present and lightning storms are very unusual, is mostly attributed to Paleoindian activity (Heuseer, 1994b, 2003 p. 74). If this were correct, the presence of charcoal in the Upper Pleistocene sections of central Chile (Heusser, 2003, p. 110) may suggest that humans would have arrived to these regions about the mid-portion of OIS 3, between 30,000 and 443,000 14C yr BP (Laguna de Tagua-Tagua). A hiatus in charcoal ﬁndings during the OIS 2 occurs in the record (Heusser, 2003), suggesting that very cold climate at the peak of LGM would have forced the aborigines to retreat equatorwards. They would have returned to Patagonia and Tierra del Fuego later in Late Glacial times (Heusser, 2003). Early human occupation at Monte Verde, Chile (Dillehay and Pino, 1989; dated at 12,500 14 C yr BP) and Piedra Museo, Argentine Patagonia (Coronato et al., 1999; dated at 12,890 14C yr BP) supports the use of charcoal evidence to follow the migration of Paleoindian hunters. The Bahı´ a Inu´til proﬁle shows a high content of charcoal particles as early as 13,300 14C yr BP, and, moreover, the Puerto del Hambre section (Heusser, 2003) provides a very high charcoal content between 12,300 and 414,500 14 C yr BP, which may indicate a human presence in the region, much earlier than previously known. In the Beagle Channel, peatland proﬁles at Lapataia, Caleta Ro´balo and Puerto Harberton show a sudden appearance of charcoal particles around ca. 10,000 14C yr BP. The Ushuaia sections indicate, however, the occurrence of charcoal particles as early as 12,100 14C yr BP (Heusser, 2003). If this charcoal were actually of anthropogenic origin, the presence of humans in the Beagle Channel would be more than 5000 years older than indicated by the now existing archeological record.

Final comments The Fuegian peatlands have greatly contributed to the scientiﬁc interpretation of the upper Quaternary environments of southernmost South America. Nevertheless, much more information remains to be obtained from them concerning the environmental changes in the Southern Hemisphere during the last 15,000 radiocarbon years. The low environmental and atmospheric pollution of Tierra del Fuego provides appropriate conditions to establish comparison parameters on the variability of atmospheric composition by means of stable isotopes. From a geological point of view, the record of ancient volcanic eruptions allows analysis of the recent history of the crustal dynamics in a region that is tectonically unstable and of high scientiﬁc interest. Moreover, the progressive knowledge of the peatlands stratigraphy would

142

J. Rabassa et al.

allow us to relate the changes in lithological composition to the environmental changes, as suggested by geomorphologic and palynological studies. The peatlands of Tierra del Fuego may contribute to the understanding of the natural history and the early peopling of this region of the Southern Hemisphere and its inter-hemispheric relationships. Our present and forthcoming studies point toward deglaciation chronologies, detailed studies of the Late Glacial pollen sections, particularly the Antarctic Cold Reversal and the Younger Dryas events, and its relationship with geomorphologic evidence, the Neoglacial and Little Ice Age chronology, and the preparation of a long-term dendrochonological curve that could be used in the calibration of a 14C curve for the Southern Hemisphere. These are some of the many fascinating scientiﬁc problems of the Quaternary of Tierra del Fuego in which peatlands are expected to provide us with reliable and detailed information. References Auer, V., 1956. The Pleistocene of Fuego-Patagonia. Part 1: The ice and interglacial ages. Annales Acad. Sci. Fenn., Ser.A. III Geol.-Geogr. 45, 1–226. Auer, V., 1958. The Pleistocene of Fuego-Patagonia. Part 2: The history of the ﬂora and vegetation. Annales Acad. Sci. Fenn., Ser. A. III Geol.-Geogr. 50, 1–239. Auer, V., 1959. The Pleistocene of Fuego-Patagonia. Part 3: Shoreline displacement. Annales Acad. Sci. Fenn., Ser. A. III Geol.-Geogr. 60, 1–247. Auer, V., 1965. The Pleistocene of Fuego-Patagonia. Part 4: Bog proﬁles. Ann. Acad. Sci. Fenn., Ser. A. III Geol.-Geogr. 80, 1–160. Auer, V., 1974. The isorhythmicity subsequent to the Fuego-Patagonian and Fennoscandian ocean level transgressions and regressions of the last glaciation. Ann. Acad. Sci. Fenn., Ser. A., III Geol.-Geogr. 115, 1–88. Bonarelli, G., 1917. Tierra del Fuego y sus turberas. An. Ministerio Agric. Sec. Geol. (Buenos Aires) 12 (3) . Boninsegna, J., 1992. South American dendrochronological records. In: Bradley, R.S. and Jones, P.D. (Eds), Climate Since A.D. 1500. Routlegde, London, pp. 446–462. Boninsegna, J., Keegan, G.C., Jacoby, R., et al., 1990. Dendrochronological studies in Tierra del Fuego. Quatern. South America Antarctic Peninsula 7, 305–326. Borrero, L., 1999. Human dispersal and climatic conditions during Late Pleistocene times in FuegoPatagonia. Quatern. Int. 53/54, 93–99. Borromei, A., Coronato, A.M.J., Quattrocchio, M., et al. (in press). Late Pleistocene–Holocene environments in Valle Carbajal, Fuegian Andes, Southern South America. J. South American Earth Sciences. Bujalesky, G., Coronato, A.M.J., Roig, C., et al., 2004. Holocene differential tectonic movements along the Argentine sector of the Beagle Channel (Tierra del Fuego) inferred from marine paleoenvironments. GEOSUR 2004. International Symposium on the Geology and Geophysics Southernmost Andes, Buenos Aires, November 2004 (extended abstract). Bujalesky, G., Heusser, C.J., Coronato, A.M.J., et al., 1997. Pleistocene glaciolacustrine sedimentation at Lago Fagnano, Andes of Tierra del Fuego, southernmost South America. Quatern. Sci. Rev. 16, 767–778. Caldenius, C., 1932. Las glaciaciones cuaternarias en la Patagonia y Tierra del Fuego. Geograﬁska Annaler 14, 1–164. Clapperton, C.M., 1993. Quaternary Geology and Geomorphology of South America. Elsevier, Amsterdam, 779pp. Coronato, A.M.J., 1990. Deﬁnicio´n y alcance de la U´ltima Glaciacio´n Pleistocena – Glaciacio´n Mota- en el valle de Andorra, Tierra del Fuego, Argentina. XI Congreso Geolo´gico Argentino, San Juan, Argentina, Actas I, 286–289.

The peatlands of Argentine Tierra del Fuego

143

Coronato, A.M.J., 1993. La Glaciacio´n Moat (Pleistoceno Superior) en los valles Pipo y Can˜ado´n del Toro, Andes Fueguinos. XII Congreso Geolo´gico Argentino, Mendoza, Argentina, Actas VI, 40–47. Coronato, A.M.J., 1995. The last Pleistocene glaciation in tributary valleys of the Beagle Channel, southernmost South America. Quatern. South America Antarctic Peninsula 9, 173–182. Coronato, A.M.J., Meglioli, A., and Rabassa, J., 2004. Glaciations in the Magellan Straits and Tierra del Fuego, Southernmost South America. In: Ehlers, J. and Gibbard, P. (Eds), Quaternary Glaciations: Extent and Chronology. Part III: South America, Asia, Africa, Australia and Antarctica. Developments in Quaternary Science. Elsevier, Amsterdam. Coronato, A.M.J. and Roig, C., 1999. Peligro geomorfolo´gico en ambientes de ge´nesis reciente. Valles de Tierra Mayor y Rı´ o Olivia. I Congreso Argentino de Geomorfologı´ a y Cuaternario, Santa Rosa, Argentina. Actas 1, 123–132. Coronato, A.M.J., Roig, C., Roig Jun˜ent, F., et al. 2006 (this book, Chapter 5). Geomorphologic emplacement and vegetation characteristics of Fuegian peatlands, southernmost Argentina, South America. In: Martini, I.P., Matı´ nez Cortizas, A., and Chesworth, W. (Eds.), Peatlands: Evolution and Records of Environmental and Climatic Changes. Elsevier, Amsterdam. Coronato, A.M.J., Salemme, M., and Rabassa, J., 1999. Paleoenvironmental conditions during the early peopling of southernmost South America (Late Glacial–Early Holocene, 14–9 ka B.P.). Quatern. Int. 53/54, 77–92. Dillehay, T. and Pino, M., 1989. Stratigraphy and chronology. In: Dillehay, T.D. (Ed.), Monte Verde. A Late Pleistocene Settlement in Chile. Paleoenvironmental and Site Context. Smithsonian Institution Press, Washington, DC, USA, Vol. 1, pp. 133–145. Gordillo, S., Coronato, A.M.J., and Rabassa, J., 1993. Late Quaternary evolution of a subantarctic paleofjord, Tierra del Fuego. Quatern. Sci. Rev. 12, 889–897. Grill, S., Borromei, A., Quattrocchio, M., et al., 2002. Palynological and sedimentological analysis of recent sediments from Rı´ o Varela, Beagle Channel, Tierra del Fuego, Argentina. Rev. Espan˜ola de Micropaleontol. 34, 145–161. Heusser, C.J., 1987. Fire history of Fuego-Patagonia. Quatern. South America Antarctic Peninsula 5, 93–109. Heusser, C.J., 1989a. Climate and chronology of Antarctica and adjacent South America over the past 30,000 yr. Palaeogeogr. Palaeoclim. Palaeoecol. 76, 31–37. Heusser, C.J., 1989b. Late Quaternary vegetation and climate of southern Tierra del Fuego. Quatern. Res. 31, 396–406. Heusser, C.J., 1993. Late Quaternary forest-steppe contact zone, Isla Grande de Tierra del Fuego, subantarctic South America. Quatern. Sci. Rev. 12, 169–177. Heusser, C.J., 1994a. Three Late Quaternary pollen diagrams from Southern Patagonia and their palaeoecological implications. Palaeogeogr. Palaeoclim. Palaeoecol. 118, 1–24. Heusser, C.J., 1994b. Paleoindians and ﬁre during the late Quaternary in southern South America. Rev. Chilena de Historia Natural. 67, 435–443. Heusser, C.J., 1998. Deglacial paleoclimate of the American sector of the Southern Ocean: Late Glacial–Holocene records from the latitude of Canal Beagle (551C), Argentine Tierra del Fuego. Palaeogeogr. Palaeoclim. Palaeoecol. 141, 277–301. Heusser, C.J., 2000. Deglacial palaeoclimate at Puerto del Hambre, subantarctic Patagonia, Chile. J. Quatern. Sci. 15, 101–114. Heusser, C.J. 2003. Ice Age Southern Andes: A Chronicle of Palaeoecological Events. Elsevier, Amsterdam, 40pp. Heusser, C.J. and Rabassa, J., 1987. Cold climate episode of younger Dryas age in Tierra del Fuego. Nature 328, 609–611. Heusser, C.J. and Rabassa, J., 1995. Late Holocene forest steppe interaction at Cabo San Pablo, Isla Grande de Tierra del Fuego, Argentina. Quatern. South America Antarctic Peninsula 9, 179–188. Isla, F. and Schnack, E., 1995. Submerged moraines offshore northern Tierra del Fuego, Argentina. Quatern. South America Antarctic Peninsula 9, 205–222. Kaplan, M., Ackert, R. Jr., Singer, B., et al., 2004. Cosmogenic nuclide chronology of millennial-scale glacial advances during O-isotope stage 2 in Patagonia. Geol. Soc. Am. Bull. 116, 308–321.

144

J. Rabassa et al.

Markgraf, V., 1980. New data on the late and postglacial vegetational history of ‘‘La Misio´n’’, Tierra del Fuego, Argentina. Proceedings of the IV International Palynological Congress, Lucknow, India, Vol. 3, pp. 68–74. Markgraf, V., 1991. Younger Dryas in southernmost South America? Boreas 20, 63–69. Massone, M., 1987. Los cazadores paleoindios de Tres Arroyos (Tierra del Fuego). An. Ins. de la Patagonia 17, 47–60. Mauquoy, D., Blaauw, M., van Geel, B., et al., 2004. Late Holocene climatic changes in Tierra del Fuego based on multiproxy analyses of peat deposits. Quatern. Res. 61, 148–158. McCulloch, R.D., Clapperton, C.M., Rabassa, J., et al., 1997. The natural setting. The glacial and postglacial environmental history of Fuego-Patagonia. In: Mc Ewan, C., Borrero, L.A., and Prieto, A. (Eds), Patagonia. Natural History, Prehistory and Ethnography at the Uttermost End of the Earth. British Museum, London, pp. 12–31. Meglioli, A., 1992. Glacial geology and chronology of Southernmost Patagonia and Tierra del Fuego, Argentina and Chile. Unpublished PhD Dissertation, Lehigh University, Bethlehem, Pennsylvania, 216pp. Mercer, J., 1976. Glacial history of southernmost South America. Quatern. Res. 6, 125–166. Planas, X., Ponsa, A., Coronato, A.M.J., et al., 2002. Geomorphological evidence of different glacial stages in the Martial Cirque, Fuegian Andes, southernmost South America. Quatern. Int. 87, 19–27. Porter, S., 1990. Character and ages of Pleistocene drifts in a transect across the Strait of Magellan. Quatern. of South America Antarctic Peninsula 7, 35–50. Porter, S., Stuiver, M., and Heusser, C.J., 1984. Holocene sea level changes along the Strait of Magellan and Beagle Channel, southernmost South America. Quatern. Res. 22, 59–67. Rabassa, J., Bujalesky, G., Meglioli, A., et al., 1992. The Quaternary of Tierra del Fuego, Argentina: the status of our knowledge. Sveriges Geol. Underso¨kning, Ser. Ca 81, 249–256. Rabassa, J. and Coronato, A.M.J., 2002. Glaciaciones del Cenozoico tardı´ o. In: Haller, M. (Ed.), Geologı´ a y recursos naturales de Santa Cruz, Relatorio del XV Congreso Geolo´gico Argentino, El Calafate 1 (19), 303–315. Rabassa, J., Coronato, A.M.J., Bujalesky, G., et al., 2000. Quaternary of Tierra del Fuego, southernmost South America: an updated review. Quatern. Int. 68–71, 217–240. Rabassa, J., Coronato, A.M.J., Roig, C., et al., 2004. Un bosque sumergido en Bahı´ a Sloggett, Tierra del Fuego, Argentina: evidencia de actividad neotecto´nica diferencial en el Holoceno tardı´ o. In: Blanco Chao, R., Lo´pez-Bedoya, J., and Pe´rez Alberti, A. (Eds), Procesos geomorfolo´gicos y evolucio´n costera. Cursos e Congresos N1 145. Universidad de Santiago de Compostela, Spain, pp. 333–346. Rabassa, J., Heusser, C.J., and Coronato, A.M.J., 1989. Peat-bog accumulation rate in the Andes of Tierra del Fuego and Patagonia (Argentina and Chile) during the last 43,000 years. Pirineos 133, 113–122. Rabassa, J., Heusser, C.J., and Rutter, N., 1990a. Late-Glacial and Holocene of Tierra del Fuego. Quatern. South America Antarctic Peninsula 7, 327–351. Rabassa, J., Heusser, C.J., and Stuckenrath, R., 1986. New data on Holocene sea transgression in the Beagle Channel, Tierra del Fuego. Quatern. South America Antarctic Peninsula 4, 291–309. Rabassa, J., Serrat, D., Martı´ , C., et al., 1990b. El Tardiglacial en el Canal Beagle, Tierra del Fuego, Argentina. Actas XI Congreso Geolo´gico Argentino 1, 290–293. Rabassa, J., Serrat, D., Martı´ , C., et al., 1990c. Internal structure of drumlins in Gable Island, Beagle Channel, Tierra del Fuego, Argentina. LUNDQUA Report, Lund, 32, pp. 3–6. Roig Jun˜ent, F., Roig, C., Rabassa, J., et al., 1996. Fuegian ﬂoating tree-ring chronology from subfossil Nothofagus wood. Holocene 6, 469–476. Stern, C.R., 1990. Tephrochronology of southernmost Patagonia. Natl. Geogr. Res. 6, 110–126. Ton-That, T., Singer, B., Mo¨rner, N., et al., 1999. Datacio´n de lavas basa´lticas por 40Ar/39Ar y geologı´ a glacial de la regio´n del Lago Buenos Aires, provincia de Santa Cruz, Argentina. Rev. Asoc. Geol. Argentina 54, 333–352. Villalba, R., Boninsegna, J.A., Veblen, T., et al., 1997. Recent trends in tree-ring records from high elevation sites in the Andes of Northern Patagonia. Clim. Change 36, 425–454.

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

145

Chapter 7

Lowland tropical peatlands of Southeast Asia S.E. Page, J.O. Rieley and R. Wu¨st

Introduction Peatlands are terrestrial wetland ecosystems in which the production of organic matter exceeds its decomposition and a net accumulation results. Several factors inﬂuence peat formation and preservation, including a positive climatic moisture balance (precipitation minus evaporation), high-relative humidity, topographic and geological conditions that favor water retention, and low substrate pH and nutrient availability. The majority of the world’s peatlands occur in boreal and temperate zones where they have formed under high-precipitation, low-temperature climatic regimes. In the humid tropics, however, regional environmental and topographic conditions have enabled peat to form under a high-precipitation, high-temperature regime (Andriesse, 1988) and, as a consequence, extensive peatlands occur in southeast Asia, mainland east Asia, the Caribbean and Central America, South America and southern Africa. Most of these are located at low altitudes where rain forest vegetation grows on a thick mass of organic matter accumulated over thousands or tens of thousands of years, to form deposits up to 20 m thick (Anderson, 1983). In the tropics, these lowland peatlands are almost exclusively ombrogenous (the peat surface only receives water from precipitation), whereas geogenous peatlands, that are fed additionally by water that has been in contact with the mineral bedrock and soils, are of more limited distribution, being conﬁned to the edges of coastal lagoons, the banks and ﬂood zones of rivers, and the margins of upland lakes. Undisturbed, lowland ombrogenous peatlands support peat swamp forest; freshwater swamp forests are associated with geogenous peatlands. The total area of undeveloped tropical peatland is in the range 31–46 million hectares, which is approximately 10–12 per cent of global peatland (Table 7.1; Immirzi and Maltby, 1992; Rieley et al., 1996). The large range of values for the extent of the resource is indicative of the problems associated with making accurate assessments, including (1) a lack of agreement on the deﬁnition of tropical peat soils, (2) the difﬁculty of accessing remote regions to carry out ground surveys, (3) application of different techniques used in surveying and mapping, especially for ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09007-9

S.E. Page, J.O. Rieley, R. Wu¨st

146

Table 7.1. Summary statistics for the area of tropical peatlands (based on Immirzi and Maltby, 1992). Region

Area (mean) (ha)

Area (range) (ha)

Central America South America Africa Asia (mainland) Asia (southeast) The Paciﬁc Total

2,437,000 4,037,000 2,995,000 2,351,000 26,435,000 40,000 38,295,000

2,276,000–2,599,000 4,037,000 2,995,000 1,351,000–3,351,000 19,932,000–32,938,000 36,000–45,000 30,627,000–45,965,000

determining accurately boundaries between mineral and peat soils on remote sensed images, since both can support forest of similar structure, and (4) the rapid decrease in peatland area as a result of oxidation of peat soils, following forest removal, drainage, agricultural utilization and ﬁre, which renders survey data obsolete. Tropical peatlands provide a range of valuable ecological functions and environmental services (Maltby et al., 1996; Page and Rieley, 1998). For example, they support a large diversity of plant and animal species, some of which are endemic or endangered (Page et al., 1997). Lowland tropical peatlands are important catchment and control systems that provide water for drinking and irrigation (Boelter, 1964) and, in coastal areas, they are buffers between salt and freshwater hydrological systems. Where the underlying mineral substratum is sulphidic, the peat layer acts as a protective wet sponge that keeps the mineral subsoil in an anaerobic condition, thus preventing the formation of highly toxic acid sulphate soils (Ritzema and Tuong, 1994). During the last 15 years, there has been increasing interest in peatlands globally because of their important role in the carbon cycle, which has raised the proﬁle of tropical peatlands and stimulated scientiﬁc investigation of their carbon dynamics (Sorensen, 1993; Neuzil, 1997; Brady, 2002; Jauhiainen et al., 2003; Page et al., 2004; Wu¨st and Bustin, 2004). Tropical peat deposits store very large amounts of carbon and are also repositories of important geochemical and paleoenvironmental information. In addition, lowland tropical peatlands have contributed to the way of life and economy of indigenous people for centuries through the provision of resources for food, shelter, medicine and cultural well-being. They may continue to provide long-term support to the socio-economy of local communities but only if their characteristics are understood and they are managed in a sustainable manner. In recent decades, tropical peatlands have come under increasing pressure from human-mediated disturbances, in particular logging, drainage and conversion to agricultural land. The associated changes in forest and land management practices have impaired the natural resource functions of the peatland ecosystem and, notably, increased greatly its susceptibility to ﬁre. In southeast Asia during 1994, 1997 and 2002, there were severe episodic droughts associated with the El Nin˜o-Southern Oscillation (ENSO). In combination with forest degradation and landuse change activities, these dry conditions triggered widespread peatland ﬁres (Page et al., 2002), leading to high levels of air pollution with serious consequences for human welfare

Lowland tropical peatlands of Southeast Asia

147

and regional economies (Schweithelm, 1999). This chapter provides information on the location, extent and nature of tropical peatlands, with a particular focus on southeast Asia where most tropical peatland occurs (Table 7.1). It also includes a review of some recent areas of investigation, with a particular focus on geochemistry, nutrient cycling, and rates of peat and carbon accumulation. The role of lowland tropical peatlands in the carbon cycle is discussed in some detail together with the impacts of ﬁre and land use change.

History, location and extent of lowland tropical peatlands in Southeast Asia The earliest description of peatland in Southeast Asia was provided more than 300 years ago by John Andersen (cited in Wichmann, 1910) who, in 1794, described the presence of peat deposits in the Riau region of Sumatra. Explorers to this region in the 19th century reported extensive peatlands in other parts of Sumatra, as well as in Sarawak and Kalimantan (Soepraptohardjo and Driessen, 1976). One of the earliest botanical accounts of peat swamp vegetation was provided by the German botanist Franz Wilhelm Junghuhn in the middle of the 19th century. Since then, numerous studies have described and characterized tropical peatlands and their biodiversity (Gates, 1915; Polak, 1933; Anderson, 1961; Maltby et al., 1996; Rieley and Page, 1997). Yet, for the most part, the existence of these vast peatland systems remains little known among the wider scientiﬁc community and international recognition of their biological, environmental and economic importance has been a slow process. It is estimated that more than 23 million hectares (62%) of the global area of tropical peatland occur in Southeast Asia (Table 7.1; Fig. 7.1). They occupy mostly low altitude, coastal and sub-coastal settings (from sea level to about 50 m asl) and can extend inland for distances of more than 200 km along river valleys and across watersheds (Rieley et al., 1996). They are most fully developed near the coasts of East Sumatra, Kalimantan, West Papua, Papua New Guinea, Brunei, Peninsular Malaysia, Sabah, Sarawak and Southeast Thailand (Table 7.2). There are also some peatlands in mountainous regions of the tropics (Maloney, 1983/1984; Wayi and Freyne, 1992; Dam et al., 2001), although these are of limited extent, and will not be considered in this review. The peat deposits of Southeast Asia lie within the inter-tropical convergence zone that experiences a wet tropical climate with annual rainfall generally in excess of 2500 mm. Seasonality of rainfall is not usually marked, but there may be either a long, wet season of 9–10 months alternating with a shorter dry season of two- or three-months duration, or two monsoon seasons (October–March and April– August) interspersed by two short dry periods. Daily and annual variation in the amount of rainfall can be considerable, and long periods of extremely low precipitation, when evaporation exceeds rainfall, can lead to temporary drought (Takahashi and Yonetani, 1997). For peatland maintenance and growth, precipitation rates must exceed evaporation rates. In coastal regions of Sarawak, for example, where there are extensive peat deposits, the annual rainfall varies from 2800 to 4700 mm with an overall average of around 3600 mm. The average evaporation is fairly constant, with a total of around 1500 mm yr 1. During the wettest months (December–February), rainfall may exceed 600 mm per month but, even during the

S.E. Page, J.O. Rieley, R. Wu¨st

148

Figure 7.1. The distribution of peatlands in Southeast Asia. The map indicates that most peatlands occur on the islands of Sumatra and Borneo (Kalimantan, Sarawak and Brunei) and in Peninsular Malaysia. The true extent and thickness of these deposits is poorly documented (see text for further details) (after Maltby and Immirzi, 1996).

Table 7.2. Summary statistics for tropical peatlands of Southeast Asia (based on Immirzi and Maltby, 1992; Rieley et al., 1996). Region

Area (mean) (ha)

Area (range) (ha)

Indonesia Malaysia Papua New Guinea Thailand Brunei Vietnam Philippines Totals

18,963,000 2,730,000 1,695,000 64,000 110,000 24,000 10,700 23,596,700

17,853,000–20,073,000a 2,730,000b 500,000–2,890,000c 64,000d 110,000e 24,000f 10,700g 21,291,700–25,901,700

a

The ﬁgures for Indonesia show the greatest variation in estimates for the tropical region as a whole. The range is based on information in RePPProt (1988, 1990). b Data are derived from several sources, Mutalib et al. (1992). c Shier (1985) and FAO (1974) provide the lower estimate, Wayi and Freyne (1992) who included upland peat are responsible for the higher. d Lappaleinen (1996). e Hunting Technical Services Ltd (1969); Anderson and Marsden, 1984. f Institute of Soil and Agriculture, Ministry of Agriculture and Rural Development. g Inventory data are for wetlands and it is difﬁcult to determine the area of peatland, much of which has been drained and converted to agriculture and aquaculture (Oraveinen et al., 1992).

dry season (March–November), the monthly average rainfall is 200–300 mm, which still exceeds the rate of evaporation (DID, 1962/1997). Structure and hydrology of lowland tropical peatlands The structure of tropical peat deposits has been studied by means of surface leveling combined with peat coring from which cross-sections have been prepared (Anderson,

Lowland tropical peatlands of Southeast Asia

149

1983; Tie, 1990; Tie and Esterle, 1992; Page et al., 1999). The majority of the peatlands in Southeast Asia are domed, ombrogenous systems, in which water and nutrient supplies to the surface are derived entirely from precipitation (rainfall, dusts and aerosols) (Rieley et al., 1996). The convex shape of the surface causes radial drainage and prevents ﬂooding from adjacent freshwater or marine systems (Neuzil et al., 1993). Many ombrogenous peat swamps are bordered either by the coast or by rivers or both. On the seaward side, they adjoin mudﬂats or sandy beaches whereas inland they may be delimited by levees of mineral soils along the banks of major rivers (Phillips and Bustin, 1996a). The gradient of the marginal zone of the peat domes of coastal peat swamps in Sarawak and Sumatra is relatively steep (6.85 m in 604 m in Sarawak, equal to 1 in 88) (Anderson, 1983). The center of these peatlands is almost ﬂat with a rise of only about 1 m km 1. This steep marginal zone is not evident on the inland peat swamps of Central Kalimantan where the gradients are much less (7.6 m in 5500 m, equivalent to 1 in 724) and the landscape appears virtually ﬂat. The gradient on the surface of these peatlands changes across the watershed but, as in the coastal swamps of Sarawak, is least in the center (Table 7.3; Page et al., 1999). In Sarawak, the subsoil beneath the peat is mainly below normal river level but only rarely below sea level, decreasing slightly as the surface level rises with distance from the edge, giving a biconvex appearance to the peat deposit cross-section (Anderson, 1961). The ﬂuctuation of the peatland water table depends mainly on rainfall because evaporation and outﬂow are fairly constant. During the wet season, rainfall always exceeds the combination of evaporation and surface run-off. In this period, the water table rises and may come to or rise above the surface, creating conditions that are favorable for peat accumulation. During the drier months of the year when, in some years, the rain-free period may last for several weeks, the water level drops below the soil surface often to depths of 0.2–0.6 m or even 1.0 m, creating conditions that can promote the decomposition of organic matter. Water table ﬂuctuations vary across the peat dome by up to 0.60 m near the edge and 0.45 m near the center (Ong and Yogeswaran, 1992; Takahashi and Yonetani, 1997; Page et al., 1999; Takahashi et al., 2002). Table 7.3. Changes in peat thickness, peat surface elevation, peat surface gradient and forest types along a 25 km transect from Sg. Sebangau to its watershed in Central Kalimantan, Indonesia (Page et al., 1999). Distance from Sg. Sebangau

Mean peat thickness (m)

Cumulative change in surface elevation from Sg. Sebangau (m)

Gradient

Forest type

0–1.5 1.5–5.5 5.5–12.5 12.5–16 16–20 20–25 Overall

1.2 3.7 8.4 8.3 10.5 9.5 7.8

+1.7 +7.6 +11.9 +17.5 +20.2 +19.1 +20.2

1 : 882 1 : 678 1 : 1628 1 : 625 1 : 1482 1 : 4091 1 : 990

Riverine Mixed swamp Low pole Tall interior Tall interior Tall interior –

150

S.E. Page, J.O. Rieley, R. Wu¨st

Classiﬁcation and genesis In Southeast Asia, three major categories of lowland, ombrotrophic peatland have been proposed, based upon their location, mode of formation and the maximum age of the peat deposits: coastal peatlands; basin or valley peatlands; and high, interior or watershed peatlands (Rieley et al., 1996; Page et al., 1999). Coastal peatlands Coastal peatlands occur along the maritime fringe and in deltaic areas where they have developed over marine sediments of clay and silt at, or only slightly above, sea level (1–2 m asl). They are situated inland of accreting mangrove and Nipa palm swamps, which they replace, and where the accumulation of organic deposits eventually excludes inundation by brackish waters. The abundance of toxic sulﬁdes in the waterlogged, brackish mangrove muds restricts bacterial activity enabling the initiation of peat formation under conditions of high rainfall and restricted drainage. The mangrove vegetation is replaced by peat swamp forest and, as organic material continues to accumulate, these forests become increasingly ombrogenous, forming a domed mound of peat (Anderson, 1983). Peat core data presented by Staub and Esterle (1994) for the Rajang delta in Sarawak indicate that this model of coastal vegetation succession can be circumvented along sand-dominated shorelines, where the build up of mineral sediments behind coastal beach ridges rapidly brings the vegetation above the zone of tidal inﬂuence, and the mangrove/Nipa phase is absent. Likewise, in Peninsular Malaysia, particularly on the east coast, the initial development of peat may have occurred in isolated lagoons formed by sand bars and spits over subsoil composed of coarse sediments. Basin or valley peatlands Basin or valley peatlands occur inland in sub-coastal locations along river valleys at slightly higher altitudes than coastal peatlands (5–15 m asl), with which they may be contiguous. Peat formation in basin or valley peatlands appears to have been initiated at an earlier period than in the coastal peatlands, as a result of rising ground water levels, linked to changes in sea level. Restriction of drainage led to permanent waterlogging and the establishment of freshwater herbaceous vegetation that, under high rainfall conditions, was followed by a transition to swamp forest and subsequent accumulation of ombrogenous peat. These basin peatlands are often located along rivers in backswamp situations behind alluvial levees. They can achieve a peat thickness of up to 20 m (Anderson, 1983). High, interior, or watershed peatlands High (sensu Sieffermann et al., 1988), interior (sensu Page et al., 1999) or watershed peatlands (sensu Morley, 2000) have only been described from Central Kalimantan

Lowland tropical peatlands of Southeast Asia

151

where they cover low-altitude watershed positions (10–30 m asl) between major rivers. These peatlands extend up to 200 km or more inland from the coast and occupy thousands of square kilometers, covering the gently sloping landscape in a manner analogous to temperate zone blanket peat. Peat formation commenced on top of the upper-coarse sand layer of the tropical podzol soil formation that extends across the middle of this province (Sieffermann et al., 1988, 1992). The creation of an impervious hard pan within the podzol, at a depth of several meters below the original mineral surface, gradually impeded vertical drainage and led to the waterlogging that was a prerequisite for peat initiation and accumulation. These watershed deposits are only slightly dome-shaped and their maximum-recorded thickness is 13 m (Page et al., 1999). On shallower ombrotrophic peats, Brunig (1990) described somewhat analogous kerapah vegetation from inland Sarawak. Kerapah is a type of heath forest (kerangas) that develops under waterlogged conditions. In addition to the above categories, some isolated peatland deposits have formed further inland in and around lake basins, such as Lake Sentarum in West Kalimantan (Anshari et al., 2001, 2004), Tasek Bera in central Peninsular Malaysia (Wu¨st and Bustin, 2004), in Java and Sulawesi (Dam et al., 2001; Hope, 2001; van der Kaars et al., 2001) in northeastern Thailand (Penny, 2001) and Cambodia (Maxwell, 2001; Maxwell and Lui, 2002).

Physical and chemical characteristics of lowland tropical peat Lowland tropical peat is relatively homogeneous consisting of slightly or partially decomposed debris of the former forest vegetation, mostly trees. Well preserved tree trunks, branches, twigs and coarse roots are found within a matrix of dark brown, amorphous organic material in various stages of humiﬁcation. The physical and chemical properties of tropical peat are a result of many factors, including wood content, microbial activity, degree of decomposition (oxidation of organic matter), mineral matter inﬂux, stratiﬁcation, compaction, and land use practices, which determine bulk density, hydraulic conductivity, water-holding capacity, ash content, acidity and chemical composition (Rieley et al., 1996). There are numerous accounts of the physical and chemical properties of tropical peat (Coulter, 1950; Anderson, 1961, 1964, 1983; Tay, 1969; Andriesse, 1974, 1988; Polak, 1975; Driessen, 1978; Esterle et al., 1992; Brady, 1997; Radjagukguk, 1997; Kurnain et al., 2001, 2002; Sajarwan et al., 2002; Wu¨st et al., 2002). A few of these present information for peat in its natural, forested condition, but most are concerned with the properties of tropical peat following deforestation, drainage and cultivation. Thick peats (42 m) have different physical and chemical properties compared to thin peats (o2 m). The former are characterized by lower bulk density, lower degree of decomposition, higher acidity, lower nutrient content, higher porosity and lower weight bearing capacity than the latter (Radjagukguk, 1992). The surface layer (0–50 cm) of thick peats is poorer in plant nutrient elements than the surface of thin peats (Wu¨st and Bustin, 2003), which may be inﬂuenced by the plants growing upon them, especially if roots penetrate to the underlying mineral layer. Leaf fall from

152

S.E. Page, J.O. Rieley, R. Wu¨st

vegetation growing on thin peat results in a higher nutrient return to the peat surface than on thick peat (Sulistiyanto, 2004). In addition, peat developed over ‘‘poor’’ mineral substrates (quartz sand) has a lower nutrient content compared to peat formed on top of clay (Widjaja-Adhi, 1988). Bulk density of tropical peat is generally low (ranging from 0.1 to 0.32 g cm 3) and decreases with depth (Driessen and Rochimah, 1976; Brady, 1997). Bulk density of the upper 30 cm layer varies between 0.12 and 0.17 g cm 3 under pristine peat swamp forest (Lambert and Staelens, 1993; Sajarwan et al., 2002) whereas, in cultivated and ﬁre-damaged peatlands, it ranges from 0.17 to 0.31 g cm 3 (Kurnain et al., 2001, 2002). Bulk density increases following land reclamation as peat starts to decompose and compact. Hydraulic conductivity in peat soils is inﬂuenced by several factors, including total porosity, bulk density and degree of decomposition and it varies greatly from 0.001 to 0.032 cm s 1 owing to the heterogeneous nature of the peat soil matrix. Hydraulic conductivity is very rapid in the surface peat layer (acrotelm) (Takahashi and Yonetani, 1997; Kurnain et al., 2002; Sajarwan et al., 2002). The ash, or inorganic, content of tropical peats varies considerably, from less than 1% to more than 65% (Rieley et al., 1996; Wu¨st et al., 2003). In ombrogenous deposits ash contents are mostly o1% although higher values (410–15%) can result from past mineral deposition (such as volcanic ash) or, in shallow deposits, from inwash (such as river ﬂooding). Several peat classiﬁcation schemes use the ash content to categorize peats and organic soils (Wu¨st et al., 2003 for a summary). Undeveloped tropical peat soils are characterized by high acidity and a deﬁciency of mineral nutrients (Table 7.4) but comparative data on their chemical composition are scarce (Coulter, 1957; Andriesse, 1974; Polak, 1975; Driessen, 1978; Anderson, 1983; Page et al., 1999). Under natural conditions the pH of ombrogenous peat usually ranges from 3.0 to 4.0 (Kurnain et al., 2002). The electrical conductivity is usually much lower than that of mineral soils, ranging from 40 to 100 mS cm 1 (Kurnain et al., 2001) for inland peat in Central Kalimantan, to 140–320 mS cm 1 for coastal peat in West Kalimantan (Suryanto, 1994), the latter probably reﬂecting the input of marine aerosols to the peat surface. The organic carbon content of tropical peat usually exceeds 50% dry weight, whereas total nitrogen content is up to about 2% (Kurnain et al., 2001; Wu¨st et al., 2003). The C:N ratio of these peats has a wide range and generally increases with increasing depth. In Kalimantan and Malaysia, maximum C:N ratios exceed 50 (Kurnain et al., 2001; Wu¨st et al., 2003). Tropical peat is rich in lignin and its derivatives, originating from trees, but low in cellulose and hemicellulose (Andriesse, 1988).

Vegetation and biodiversity The vegetation of natural lowland tropical peat swamp forest is dominated by trees (Fig. 7.2), in contrast to most temperate and boreal peatlands, which are usually characterized by bryophytes, grasses, sedges and shrubs (Wyatt-Smith, 1959, 1964; Anderson, 1963, 1964, 1976, 1983; Rieley and Ahmad-Shah, 1996; Page et al., 1999). Many of the trees have buttress or stilt roots that provide improved stability on the

pH and nutrient concentrations (mg kg 1) in the upper 50 cm of peat soils in peat swamp forests in Central Kalimantan and Riau,

Location

pH

N

P

K

Ca

Mg

Na

Fe

Mn

Reference

Sg. Sebangau, Central Kalimantan (MSF) Sg. Sebangau, Central Kalimantan (LPF) Sg Sebangau, Central Kalimantan (LPF)a Sg. Enok, Riau, Sumatra Sg. Siak, Riau, Sumatra Sg. Rokan, Riau, Sumatra

3.3 3.5 2.9 4.2 3.6 3.8

28,637 21,901 14,000 16,300 19,800 21,300

183 139 194 500 500 900

288 220 677 800 600 800

941 889 709 2800 1700 –

709 669 532 1600 800 1200

311 140 –

469 271 –

4 4 –

– –

– –

– –

Sulistiyanto (2004) Sulistiyanto (2004) Page et al. (1999) Suhardjo and Widjaja-Adhi (1976) Suhardjo and Widjaja-Adhi (1977) Suhardjo and Widjaja-Adhi (1977)

Lowland tropical peatlands of Southeast Asia

Table 7.4. Sumatra.

Note: LPF, low pole forest; MSF, mixed swamp forest. a Data for top 30 cm of peat.

153

154

S.E. Page, J.O. Rieley, R. Wu¨st

Figure 7.2. View inside a tropical peat swamp forest in Central Kalimantan (Indonesian Borneo).

waterlogged peat soils and breathing roots (pneumatophores) that protrude above the peat surface, enabling respiratory gas exchange to occur under anaerobic conditions. Most of the tree families of lowland dipterocarp forests in Southeast Asia are found in lowland peat swamp forests (Polak, 1975; Whitmore, 1984) with members of the Anacardiaceae, Annonaceae, Burseraceae, Clusiaceae, Dipterocarpaceae, Euphorbiaceae, Lauraceae, Leguminosae, Myristicaceae, Myrtaceae and Rubiaceae being well represented (Wyatt-Smith, 1959; Morley, 1981; Flenley, 1985, 1998; Brunig, 1990; Ibrahim and Hall, 1992; Shepherd et al., 1997). Members of the Pandanaceae often form a dense ground cover; ferns and insectivorous pitcher plants (Nepenthaceae) also occur. Bryophytes are abundant on the tops of hummocks and on tree bases, but Sphagnum spp. are present in marginal drainage areas only and are not associated with peat formation (Gates, 1915; Flenley, 1979). Differences in hydrology and nutrient availability exert strong inﬂuences on the composition and structure of the forest vegetation, with most large peatland domes exhibiting a concentric zonation of forest sub-types, ranging from tall, ﬂoristically diverse and structurally complex forest over shallow peat at the margins to less diverse, low-canopy pole (small diameter) forest over thicker peat toward the center although some exceptions to this general pattern have been described (Page et al., 1999). Six phasic communities have been described for the peatlands of Sarawak (Anderson, 1963, 1964), ranging from a structurally complex, species-rich community around the edge of the peat dome, to a stunted padang (open) community on deep peat. The intermediate communities are dominated by the dipterocarp tree Shorea albida. In the peat swamps of Central Kalimantan, ﬁve phasic communities

Lowland tropical peatlands of Southeast Asia

155

have been identiﬁed (Page et al., 1999), which differ in several respects from those described from Sarawak, principally the absence of S. albida, whereas the peatlands of the Malay Peninsula and Sumatra appear to support only two main forest types (Anderson, 1976; Morley, 1981, 2000). Pollen analyses from peat cores in Sarawak have shown that the zonation of forest types may represent succession over time (Anderson and Muller, 1975). Whereas the diversity associated with ombrotrophic lowland tropical peatlands is usually lower than adjacent rain forest ecosystems on mineral soils, many peatland plants are specialists, which are not found in other habitats. In Southeast Asia, the tree species Dactylocladus stenostachys, Gonystylus bancanus, Horsfieldia crassifolia, Shorea belangeran and S. teysmanniana are conﬁned almost exclusively to peat swamp forest. In general terms, tall peat swamp forest sub-types, which have the greatest tree species diversity and canopy stratiﬁcation, support the greatest faunal diversity (Page et al., 1997). Lower canopy sub-types are less species diverse. The latter coincide with zones of hydrological and nutrient stress (as a result of waterlogging and low nutrient availability) and resemble closely the depauperate plant and animal communities described in earlier accounts of the ecosystem (Merton, 1962; Janzen, 1974). These forests are not without interest, however, and several noteworthy species of mammal and bird have been recorded in low pole forest (Page et al., 1997), while many water-ﬁlled hollows on the forest ﬂoor support unusual species of blackwater ﬁsh (Ng et al., 1994). Accounts of the animal communities associated with peat swamp forest emphasize the signiﬁcant contribution that this habitat makes to the maintenance of regional and global biodiversity. The blackwater rivers that drain the peat swamps were once considered to have low ﬁsh species diversity and productivity, but this view has changed following the discovery of many new taxa associated with these unique habitats (Kottelat and Ng, 1994; Ng et al., 1994). The avian species diversity is also noteworthy. The peat swamp habitat provides a refuge for a number of rare and threatened species, including Storm’s stork (Ciconia stormi) and white-winged duck (Cairina scutulata) (Page et al., 1997; Wibowo et al., 2000) which, along with hookbilled bulbul (Setornis criniger) and gray-breasted babbler (Malacopteran albogulare), are considered to be peat swamp specialists within the Southeast Asian region (Sheldon, 1987). Other studies have highlighted the role these wetland forests play in providing a refuge for vulnerable species of reptiles, for example, the false gharial (Tomistoma schlegelii) (Bezuijn et al., 2004), and mammals (Prentice and Parish, 1992; Page et al., 1997). Of particular conservation importance are the relatively large populations of orangutan (Pongo pygmaeus) associated with peat swamp forests in Borneo, which are now one of the most important remaining habitats for this endangered primate (Meijaard, 1997; Morrogh-Bernard et al., 2003). The inaccessibility of tropical peatlands has drawn in animals that, although not conﬁned to this habitat, are dependent upon the shelter and food that peat swamp forests provide, compared to the adjacent intensively logged forests on mineral soils. As more information on the biodiversity of tropical peat swamp forests accumulates it is clear that this ecosystem has been undervalued as a habitat for rare and threatened species.

156

S.E. Page, J.O. Rieley, R. Wu¨st

Biomass and nutrient dynamics Recent studies of both above and belowground biomass demonstrate differences between peat swamp forest sub-types. The total above-ground forest biomass (woody vegetation plus Pandanus spp.) of peat swamp forest in Central Kalimantan varies from 314 t ha 1 for marginal mixed swamp forest on peat 2–3 m thick to 252 t ha 1 for low canopy pole forest on peat 47 m thick; most of the biomass in both vegetation types is contributed to by trees of diameter45 cm (Sulistiyanto, 2004). Above ground biomass of peat swamp forest vegetation in eastern Sumatra ranges from 395–641 t ha 1 for mixed swamp forest on peat of 3–6 m thickness, to 85–177 t ha 1 and 13–24 t ha 1 for low pole forest on peat of 9 and 12 m thickness, respectively (Brady, 1997). Comparisons with biomass of other lowland forests in the same region show that mixed swamp forest is similar to lowland dipterocarp forest (509 t ha 1; Yamakura et al., 1986), whereas low pole forest is comparable to heath forest (200–250 t ha 1; Miyamoto et al., 2000). Below-ground (root) biomass in Central Kalimantan varies from 26.5 t ha 1 for mixed swamp forest to 14.4 t ha 1 for low pole forest (Sulistiyanto, 2004), suggesting a trend of decreasing root biomass with increasing peat thickness. Based on studies in eastern Sumatra, however, Brady (1997) presented data that indicated a trend in the opposite direction. In his study, marginal mixed swamp forest had the lowest root biomass (2.8 t ha 1), and low pole forest vegetation on the thickest peat had the highest root biomass (9.0 t ha 1). Several reasons have been proposed to account for these discrepancies, including differences in tree species and tree densities between sites (Sulistiyanto, 2004). Since roots are believed to be a major source of organic matter in an accumulating tropical peat deposit (Brady, 1997), further studies of below-ground biomass and decomposition should contribute to a better understanding of their organic matter dynamics. Ombrotrophic tropical peat is deﬁcient in plant nutrients, as shown by the low elemental content of both peat and peat water, and vegetation growth is dependent therefore upon both the supply of nutrients from the atmosphere and efﬁcient recycling of the existing nutrient pool within the ecosystem (Page et al., 1999). In a study of the nutrient dynamics of the peat swamp forest ecosystem (Sulistiyanto, 2004) it was demonstrated that internal nutrient transfers were related to forest tree biomass through gravitational inﬂuence (throughﬂow, stemﬂow and litterfall) and recycling (organic matter decomposition and uptake by the trees), and to storage in and release from the surface peat (accumulation or degradation) (Weiss et al., 2002). The nutrient dynamics of two sub-types of tropical peat swamp forest (mixed swamp forest and low pole forest) in Central Kalimantan, over a one-year period, were compared by measuring the amounts of essential plant nutrient elements entering and leaving these systems and those contained in the various structural components (tree stems and branches, roots and the surface peat) together with the transfer of chemical elements in throughfall, litterfall and plant uptake (Sulistiyanto, 2004). Nutrient inputs (in precipitation and dry fall) were higher than nutrient losses (in runoff from the peat dome), with the greatest gain for calcium. Nutrient outputs and retentions were similar for both forest sub-types for every element studied, despite major structural differences between them (Sulistiyanto, 2004). The overall nutrient budgets indicate that chemical elements are being retained within both forest

Lowland tropical peatlands of Southeast Asia

157

sub-types and, since these are in dynamic equilibrium, it is possible that this surplus is being stored in currently accumulating peat. This study showed retention within the peat swamp forest ecosystem of calcium, magnesium, potassium, phosphorus, nitrogen, sodium, iron and manganese (Table 7.5). In contrast, studies on non-peat forming tropical forests show mostly annual net losses of most plant nutrient elements, especially calcium, magnesium and potassium (Lewis, 1986; Bruijnzeel, 1991).

The geochemical record Geochemical investigations of ombrotrophic peatlands from Indonesia and Malaysia have provided critical information on the aerial and vertical distribution of major and trace elements within tropical peat deposits (Neuzil et al., 1993; Weiss et al., 2002; Wu¨st and Bustin, 2003) and support the results of peat swamp forest nutrient cycling studies referred to above. The number of pathways for allogenic material inﬂux to ombrogenous deposits is limited, thus the inﬂux of plant-essential elements is constrained similarly. Airborne terrestrial dust and volcanic ash that reach the peat surface as dryfall or in rain are the most common sources of mineral inputs, whereas dissolved solids entering the deposit in rain, marine aerosols, and as a result of diffusion from substrate pore water, are very limited and often locally restricted. In addition, concentration proﬁles of Mn, Sr and Ca suggest only a very minor inﬂuence from dissolving sediments (Weiss et al., 2002). Autogenic processes within the deposit, however, can result in mobilization and stabilization of mineral matter by hydrological, chemical and biological processes (Neuzil et al., 1993). Geochemical studies of lowland peat deposits in South Sumatra, West Kalimantan (Neuzil et al., 1993), Central Kalimantan (Weiss et al., 2002), and Peninsular Malaysia (Wu¨st and Bustin, 2003) reveal several key characteristics of the inorganic chemistry of tropical, domed, ombrotrophic deposits. Ash yields are low (usually o2%), only rising to higher values (up to 10% or slightly higher) within the peat–sediment interface and the topsoil layer (top 20–50 cm). Peat pH also remains fairly constant throughout the peat proﬁle, averaging 3–4.5, rising slightly near the base of the deposit. Sulphur contents are low, averaging 0.14%, but rising slightly to 0.2–0.3% near the base of the deposit; in contrast, under the inﬂuence of marine aerosols, these values can be as high as 10% in coastal environments (Phillips and Bustin, 1996b). In a geochemical study of a 9.6-m thick ombrotrophic peat deposit in Central Kalimantan, Weiss et al. (2002) demonstrated that the upper 100–150 cm of the proﬁle was enriched in nutrients compared to the middle section. In fact, the elements P, K, Ca, Mg, Si and Na, were much higher within the top 50–80 cm, compared to the lower parts of the proﬁle, indicating a strong elemental recycling within the uppermost peat layer. The thickness of this biologically inﬂuenced zone in tropical peatlands is in sharp contrast to temperate bogs, in which only 10–20 cm of the uppermost peat proﬁle shows enrichment (Damman, 1978). This difference most likely reﬂects the presence of forest trees on tropical peatlands and their greater

158

Table 7.5. Nutrient balance (inputs and outputs in kg ha from Sulistiyanto, 2004). Water

Ca

MSF LPF MSF LPF

1

yr 1) for two peat swamp forest types in Central Kalimantan, Indonesia (derived

Mg

K

P

MSF LPF

MSF LPF

MSF

N LPF

MSF

LPF

Na

Fe

Mn

MSF LPF

MSF LPF

MSF

LPF

Input 2761 2761 15.7 15.7 5.8 5.8 9.6 9.6 4.6 4.6 0.5 0.5 5.5 5.5 3.2 3.2 0.22 0.22 Output 1523 1523 8.2 7.1 2.5 2.2 2.2 1.3 0.26 0.07 0.26 0.23 4.8 4.7 0.7 0.7 0.01 0.02 Difference 1238 1238 +7.5 +8.6 +3.3 +3.6 +7.4 +8.3 +4.34 +4.53 +0.24 +0.27 +0.7 +0.8 +2.5 +2.5 +0.21 +0.20

S.E. Page, J.O. Rieley, R. Wu¨st

Note: MSF, mixed swamp forest; LPF, low pole forest.

Lowland tropical peatlands of Southeast Asia

159

rooting depths compared to the shallow-rooting, low-growing vegetation of temperate peatlands, linked with the greater thickness of the acrotelm in the former.

Age and rates of peat and carbon accumulation Paleoenvironmental studies of lowland tropical coastal peat deposits in Southeast Asia have demonstrated that these are the youngest peatlands in the region. Peat accumulation commenced in most of these around 4000–5500 cal yr BP, following stabilization of rising sea levels (Anderson and Muller, 1975). In comparison, investigations of sub-coastal and inland peatlands, particularly in Borneo, have revealed much earlier initiation dates, ranging from Late Pleistocene (ca. 29,000 cal yr BP ) in the Danau Sentarum basin of West Kalimantan (Anshari et al., 2001, 2004) to ca. 26,000 cal yr BP in the Sebangau catchment, Central Kalimantan (Page et al., 2004) through to early Holocene (8000–9000 cal yr BP) for other high and basin/ valley peatlands within Borneo (Table 7.6; Sieffermann et al., 1988; Staub and Esterle, 1994; Neuzil, 1997). Whereas it had been assumed previously that tropical peat accumulation was primarily a feature of the Holocene, it is now clear that it was also a feature of the last glacial period. Since most peatland carbon globally is contained within postglacial boreal and temperate peatlands (Gorham, 1991) studies of the role played by peatlands in the carbon cycle have focused on a period of less than 10,000 years. The discovery of a much longer paleo-record in some tropical peat deposits, extending over 30,000–40,000 years, could therefore make an important new contribution to knowledge of the longer-term behavior of the peatland carbon reservoir (Large et al., 2004) and, in particular, where carbon was stored during glacial periods of low atmospheric carbon dioxide levels (Anshari et al., 2004).

The role of tropical peatlands in the global carbon cycle In an undisturbed condition, tropical peatlands form an efﬁcient terrestrial carbon sink that, at the present time, makes a signiﬁcant contribution to global terrestrial carbon storage. Tropical peatlands may account for only 10–12% of the global peatland resource by area but, owing to their considerable thickness and high carbon content, they contain between 50 and 70 Gt (16–21%) of the peat soil carbon store and 2–3% of the total soil carbon pool (Gorham, 1991; Immirzi and Maltby, 1992; Page et al., 2002). Carbon allocation rates in tropical peatlands in the past have been considerable. During the Late Pleistocene and early Holocene, the inland peatlands of Central Kalimantan had very high peat, and hence carbon, accumulation rates. Page et al. (2004), for example, reported an average peat accumulation rate of 2.55 mm yr 1 between 8540 and 7820 14C yr BP, with an average carbon accumulation rate over the 2 1 same period of 92 g C m yr . Other rates of early Holocene peat and carbon accu2 1 mulation in Central Kalimantan range from 0.3 to 2.4 mm yr 1 and 47–75 g C m yr , respectively (Sieffermann et al., 1988; Neuzil, 1997). These values are three to four

160

Table 7.6. Range of ages for the origin of tropical peat in Southeast Asia. Country

Location

Sample no. Peat age 7 years Sample depth (14C yr (cm) BP)

Thailand

To Daeng Swamp, Narathiwat

P3

6700

190

400

400

Bacho Swamp, Narathiwat

P2

3956

120

340

340

30

Nong Thale Song Hong, Trang

Beta106538

21,170

90

36–40

38

?

Pekan Nanas, Johore Tasek Bera, BetaPahang, B78 118948

4896

70

50–55

53

4560

70

428–439

434

23,050

330

190–203

196

Nee Soon

UWA-34

Sungei Nipah, ECON 2

Beta 78262

7790

60

1180–1190 1185

Marudi, Baram GRO 1963 EastRiver Malaysia, Sarawak

4270

70

1200

Singapore

1200

Depth Peat type, above sample dated base (cm) Freshwater peat

Freshwater peat

Source

m above MSL

15 Vijarnsorn and Liengsakul (1986) 10 Vijarnsorn and Liengsakul (1986) Maloney (1998)

2

1

Peat

66

Haseldonckx (1977) Limnic peat, Wu¨st and Bustin wood (1999) (B78)

km to coast

Clayey peat and plants Wood fragment 10

Peat

30

150

Taylor et al. (2001) Hesp et al. (1998) Anderson and Muller (1975)

28

S.E. Page, J.O. Rieley, R. Wu¨st

WestMalaysia

Avg. depth (cm)

Rajang River, Sibu Rajang River, Sibu Sumatra, Indonesia

Batang Hari River, Jambi Batang Hari River, Jambi Berbak, Jambi

4270

70

1535

1535

1950

70

394

394

RB-1

7340

220

R-090

7060

280

7580

340

JB-15

5

Basal peat 30

700–750

725

4250

0 5

4500

300

0

Nipa peat overlying white clay Basal, hemic peat Fine hemic peat, basal Basal peat Marginal mire

Wilford (1959) Wilford (1959) Staub and Esterle (1994) Staub and Esterle (1994) Cameron et al. (1989) Esterle and Ferm (1994) Silvius et al. (1984) Neuzil (1997)

7.6

39

Neuzil (1997)

0.5

0

Siak Kanan, Riau Bengkalis Island, Riau

SK5-T9A

5220

220

433–443

438

4

BK13WD1

5500

200

300

300

100

Java, Indonesia

Situ Bayongbong

GrN-11522 16,800

300

772–780

776

4

Kalimantan, Indonesia

Kapuas River, Pemerak, HN3 Setia Alam, S. Sebangau Palangka Raya, S. Sebangau Palangka Raya, S. Sebangau Palangka Raya, S. Sebangau

Wk-5779

28,780

100

104–124

114

?

Hemic peat

SA6.5-54

22,620

310

880–900

890

73

8260

380

500

500

Sapric peat, wood, herb High peat

PK2-C14

8140

180

214–223

218

23

PK6-C14

9070

200

700–710

705

16

Marginal mire Central mire Neuzil (1997)

Lowland tropical peatlands of Southeast Asia

Baram River, Bakong River Lawas River

Stuijts et al. (1988) Anshari et al. (2001) Weiss et al. (2002) Sieffermann et al. (1988) Neuzil (1997)

110

12

124

18

118

161

35

162

S.E. Page, J.O. Rieley, R. Wu¨st

times higher than accumulation rates reported for this same period in temperate and 2 1 boreal bogs, which are about 20–25 g C m yr (Turunen and Turunen, 2003). From 8590 14C yr BP, rates of peat and hence carbon accumulation within the inland 2 1 peatlands began to decline, decreasing to 0.23 mm yr 1 and 11.0 g C m yr by 6610 14 C yr BP (Page et al., 2004). This period of reduced deposition coincides, however, with the main middle to late Holocene peat initiation and accumulation phase of the majority of the coastal peatlands of Borneo, which commenced around 7000 14C yr BP (ca. 6500 14C yr BP) (Wilford, 1959; Staub and Esterle, 1994; Hesp et al., 1998) when rising sea levels stabilized, or dropped slightly, resulting in exposure of large, relatively ﬂat areas of marine sediments (Geyh et al., 1979; Tjia et al., 1984; Tjia, 1992). The combination of favorable topographic and climatic conditions at that time led to rapid peat accumulation throughout the coastal lowlands. In the Rajang delta of Sarawak, 4.45 m of peat accumulated between 5610 and 2070 14C yr BP (Staub and Esterle, 1994), equivalent to an average peat accumulation rate of 1.26 mm yr 1 (4.45 m in 3540 years) while, on the east coast of Sumatra, the peatlands of Riau province also underwent rapid accumulation with initial rates as high as 6–13 mm yr 1 between 4700 and 3900 14C yr BP, reducing subsequently to 0.6–2.7 mm yr 1 (Neuzil, 1997). In the absence of human intervention, many tropical deposits are currently either accumulating peat or are in a steady state (Brady, 2002), although hydrological conditions are no longer conducive to continued accumulation at all sites (Sieffermann et al., 1988; Page et al., 1999). The current average accumulation rate for Indonesian peatlands has been estimated to be between 1 and 2 mm yr 1 (Sorensen, 1993; Page et al., 2004), which is substantially higher than the range of 0.2–0.8 mm yr 1 obtained for boreal and subarctic peatlands (Gorham, 1991) and 0.2–1 mm yr 1 for temperate peatlands (Aaby and Tauber, 1975). There are few data available on long-term (apparent) rates of carbon accumulation (LORCA) in tropical peatlands, although knowledge of the rate of carbon accumulation and its change through time is necessary when investigating the carbon cycle in peatlands and its relationship to climate change. The average LORCA value for an inland peat deposit in Central Kalimantan is 56 g C m 2 yr 1 (Page et al., 2004), which is higher than most values for boreal and temperate peatlands that range between 15 and 26 g C m 2 yr 1 and 10–46 g C m 2 yr 1, respectively (Turunen et al., 2002). Continued peat accumulation and carbon storage are favored by high effective rainfall and waterlogged soils; any persistent environmental change, particularly a decrease in either quantity and/or frequency of rainfall, or a drop in the peatland water table, will deepen the oxic surface peat layer (the acrotelm), thereby increasing substrate availability for CO2-releasing decomposition processes, resulting in peat degradation and release of carbon from storage. In recent decades, an increasing proportion of the tropical peatland carbon store has been converted to a carbon source through deforestation, landuse change and ﬁre (Fig. 7.3). It is estimated that the widespread peatland ﬁres that occurred throughout Indonesia during the strong ENSO-related drought of 1997 resulted in the combustion of 0.87–2.57 Pg (1012 kg) of stored carbon that took between 1000 and 2000 years to accumulate, with up to 8% of the total carbon store within the peat being released in a few months (Page et al., 2002). At the current estimated rate of carbon accumulation in the peatlands of Central Kalimantan, this single ﬁre event represents an approximate loss of between

Lowland tropical peatlands of Southeast Asia

163

Figure 7.3. Fire-damaged peat swamp forest in Central Kalimantan. Uncontrolled forest ﬁres swept through these peat swamps in 1997 and 2002, consuming both vegetation and underlying peat, leading to signiﬁcant emissions of carbon into the atmosphere.

70 and 200 years of carbon sink function. The transfer of large amounts of carbon to the atmosphere has major implications for climate change processes (van der Werf et al., 2004). In addition to losses attributable to ﬁre, the tropical peatland carbon store is also being depleted by conversion to agricultural use (Fig. 7.4). This process requires both removal of the primary vegetation and soil drainage, the latter resulting in peat shrinkage through a combination of water loss, enhanced aerobic decomposition and compaction. It has been estimated that subsidence of drained and cultivated peat in Malaysia results in a loss of 2.00 kg C m 2 yr 1 through decomposition (CO2 production) of organic matter (Wo¨sten et al., 1997); losses to the atmosphere of a similar magnitude (up to 2.97 kg C m 2 yr 1) have also been recorded from peat soils in Sumatra (D. Taylor, personnal comm) and Kalimantan (Jauhiainen et al., 2003). Conversion to plantation crops appears to result in somewhat lower CO2 ﬂuxes, with reported values of 1.5 kg C m 2 yr 1 for oil palm and 1.1 kg C m 2 yr 1 for sago (Melling et al., 2005).

164

S.E. Page, J.O. Rieley, R. Wu¨st

Figure 7.4. Dalat sago plantation established on shallow peat at the edge of a coastal peatland near Muka, Sarawak. Compared with other types of crop, sago is relatively tolerant of a high water table (the optimum level is 0.2–0.4 m below the peat surface), however, it only grows successfully on peat less than 3 m thick.

The Southeast Asian region is currently subject to increasing climatic variability (Heaney, 1991; Easterling et al., 2000) and it is predicted that seasonal precipitation extremes associated with future ENSO-events are likely to become more pronounced (Meehl and Washington, 1996). This may lead to reduced water supply to and retention by peatlands, leading to a lowering of water tables. This will limit the rate of peat accumulation where it is still taking place, enhance degradation and oxidation on peatlands that are no longer actively forming peat, and greatly increase the likelihood of peatland ﬁres, with consequent rapid loss of stored carbon. Modern tropical peatlands and past coal deposits Several studies have focused on modern tropical peats as analogues to the precursors of low-ash, low-sulphur coal deposits in order to understand the changes in organic matter that occur during peat deposition through to eventual coal formation (Neuzil et al., 1993; Esterle and Ferm, 1994; Phillips and Bustin, 1996a,b, 1998; Wu¨st et al., 2002). Coal and peat studies have indicated compaction ratios from peat to bituminous coal of between 3:1 and 24:1 (Ryer and Langer, 1980; Esterle et al., 1989; Esterle and Ferm, 1994), with Southeast Asian tropical peats of 8 m thickness probably resulting in a 1 m thick coal seam (Esterle et al., 1989). Peat accumulation in the past (such as during the Carboniferous, Cretaceous, Tertiary and Miocene periods) often spanned several thousand to tens of thousands of years and is assumed to have

Lowland tropical peatlands of Southeast Asia

165

Figure 7.5. The one million hectare Mega Rice Project in Central Kalimantan was an attempt to convert a vast area of tropical peatland to agricultural use. It was brought to an early close in 1999, following a series of difﬁculties, not the least of which was the failure to grow rice on the acidic, nutrient-deﬁcient peat. This view shows one of the large channels intended to provide drainage and irrigation for rice cultivation during the wet and dry seasons. Over 4000 km of channel were constructed in 1996–1998, and abandoned subsequently. The legacy is a vast, over-drained, ﬁre-prone peatland landscape that is now the focus of investigations for ecological restoration.

taken place mostly under tropical ever-wet climatic conditions and with a constant subsidence rate (Large et al., 2004). These circumstances resulted in many coal seams exceeding 3 m thickness, which would be equivalent to approximately 25 m of peat. By contrast, most modern tropical peat deposits started accumulation only after the last glacial maximum (around 12,000 14C yr BP or later) and, as a result of the limited time available for accumulation of organic matter, they rarely exceed a thickness of 6–8 m. Despite the shorter time period over which postglacial peat formation has taken place, studies have shown that most modern coastal and inland tropical peats, such as those occurring in Sumatra, Kalimantan, Sarawak and Peninsular Malaysia, would result in very low-ash coal deposits and these Holocene deposits can, therefore, be used as templates for understanding coal composition and former depositional environments (Esterle et al., 1989; Esterle, 1990; Demchuk and Moore, 1993; Neuzil et al., 1993; Wu¨st et al., 2002).

Future prospects for tropical peatlands: the wise use approach The maintenance and status of the world’s peatlands is a matter of considerable concern since their degradation leads to release of stored carbon in the form of

166

S.E. Page, J.O. Rieley, R. Wu¨st

greenhouse gases that contribute to climate change. This is recognized by governments in most developed countries where the utilization of peat and peatlands is subject to rigorous control through land planning procedures and regulations. In the developing countries of the tropics, however, this control and regulation is largely lacking and major developments on peatland have and still are taking place without adequate assessment of their importance as carbon stores or the environmental, social and economic consequences of converting them to some other purpose (Fig. 7.5). Inevitably, every type of human intervention on peatland leads to impairment or even loss of natural resource functions (ecology, hydrology, biodiversity, carbon storage). The challenge facing those involved in the management of tropical peatlands in the 21st century is to develop integrated planning and management mechanisms that can balance the conﬂicting demands on the tropical peatland heritage and its environmental feedback mechanisms to ensure its continued survival to meet the future needs of humankind. This strategy of wise use of tropical peatlands involves evaluation of their functions and uses, impacts caused by and constraints to development so that, by assessment, reasoning and consensus, it should be possible to highlight priorities for their management and use, including mitigation of past and future damage. There is, however, still very little comprehensive knowledge of the socio-economic, socio-cultural or socio-ecological aspects of tropical peatland ecosystems. It is only once this is understood better and combined with biogeophysical data, that a comprehensive wise use approach can be formulated.

References Aaby, B. and Tauber, H., 1975. Rates of peat formation in relation to local environment as shown by studies of a raised bog in Denmark. Boreas 4, 1–17. Anderson, J.A.R., 1961. The ecology and forest types of the peat swamp forests of Sarawak and Brunei in relation to their sylviculture. Ph.D. Thesis, University of Edinburgh, UK. Anderson, J.A.R., 1963. The ﬂora of the peat swamp forest of Sarawak, and Brunei, including a catalogue of all recorded species of ﬂowering plants, ferns and fern allies. Singapore: Garden Bull 20, 131–228. Anderson, J.A.R., 1964. The structure and development of peat swamps of Sarawak and Brunei. J. Trop. Geog. 18, 7–16. Anderson, J.A.R., 1976. Observations on the ecology of ﬁve peat swamp forests in Sarawak and Kalimantan. Soil Research Institute Bogor, Indonesia. Bulletin No. 3, pp. 45–55. Anderson, J.A.R., 1983. The tropical peat swamps of western Malesia. In: Gore, A.J.P. (Ed.), Ecosystems of the World: Mires: Swamp, Bog, Fen and Moor. 4B, Regional Studies. Elsevier, New York, pp. 181–199. Anderson, J.A.R. and Marsden, J.D., 1984. Brunei forest resources and strategic planning study. Final report submitted to the Government of His Majesty the Sultan and Yang Di Pertuan of Negara Brunei Darussalam, vol. 1 and 2. Anderson & Marsden Forestry Consultants Ltd. Anderson, J.A.R. and Muller, J., 1975. Palynological study of a Holocene peat and a Miocene coal deposit from NW Borneo. Rev. Palaeobot. Palynol. 19, 291–351. Andriesse, J.P., 1974. Tropical Lowland Peats in Southeast Asia. Comm. 63, Roy. Tropical Institute, Department of Agricultural Research, Amsterdam, The Netherlands. Andriesse, J.P., 1988. Nature and Management of Tropical Peat Soils. FAO Bulletin 59. Food and Agriculture Organisation of the United Nations, Rome. Anshari, G., Kershaw, A.P., Kaars, S.V.D., and Jacobsen, G., 2004. Environmental change and peatland forest dynamics in the Lake Sentarum area, West Kalimantan, Indonesia. J. Quatern. Sci. 19, 637–655.

Lowland tropical peatlands of Southeast Asia

167

Anshari, G., Kershaw, A.P., and van der Kaars, S., 2001. A Late Pleistocene and Holocene pollen and charcoal record from peat swamp forest, Lake Sentarum Wildlife Reserve, West Kalimantan, Indonesia. Palaeogeogr. Palaeoclim. Palaeoecol. 171, 213–228. Bezuijn, M.R., Webb, G.J.W., Hartoyo, P., and Samedi, 2004. The conservation value of peat swamp forest for the False Gharial (Tomistoma schlegellii) with reference to the Merang river in eastern Sumatra, Indonesia. In: Mansor, M., Ahyaudin, A., and Rieley, J., et al. (Eds), Tropical Peat Swamps: Safe-Guarding a Global Natural Resource. Universiti Sains Malaysia, Penang, pp. 31–38. Boelter, D.H., 1964. Water storage characteristics of several peats in situ. Soil Sci. Soc. Am. Proc. 28, 433–435. Brady, M.A., 1997. Organic matter dynamics of coastal peat deposits in Sumatra. Ph.D. Thesis, University of British Columbia, Vancouver, Canada. Brady, M.A., 2002. Peat accumulation in Sumatra: does Clymo’s model apply?. In: Rieley, J.O., Page, S.E., and Setiadi, B. (Eds), Peatlands for People, Natural Resources Function and Sustainable Management. Proceedings International Symposium on Tropical Peatland, Jakarta, Indonesia. BPPT (Badan Pengkajian dan Penerapan Teknologi) and Indonesian Peat Association, Jakarta, pp. 50–51. Bruijnzeel, L.A., 1991. Nutrient input-output budgets of tropical forest ecosystems: a review. J. Trop. Ecol. 7, 1–24. Brunig, E.F., 1990. Oligotrophic forested wetlands in Borneo. In: Lugo, A.E., Brinson, M., and Brown, S. (Eds), Forested Wetlands. Elsevier, Amsterdam, pp. 299–344. Cameron, C.C., Esterle, J.S., and Palmer, C.A., 1989. The geology, botany and chemistry of selected peatforming environments from temperate and tropical latitudes. Int. J. Coal Geol. 12, 105–156. Coulter, J.K., 1950. Peat formations in Malaya. Malay. Agric. J. 33, 63–81. Coulter, J.K., 1957. Development of the peat soils of Malaya. Malay. Agric. J. 60, 188–199. Dam, R.A.C., Fluin, J., Suparan, P., and van der Kaars, S., 2001. Paleoenvironmental developments in the Lake Tondano area (N-Sulawesi, Indonesia) since 33,000 yr B.P.. Paleogeog. Paleoclim. Paleoecol. 171, 147–183. Damman, A.W.H., 1978. Distribution and movement of elements in ombrotrophic peat bogs. Oikos 40, 490–495. Demchuk, T. and Moore, T., 1993. Palynoﬂoral and organic characteristics of a Miocene bog-forest, Kalimantan, Indonesia. Org. Geochem. 20, 119–134. DID, 1962/1997. Hydrological Yearbooks: Edition 1962–1997. Department of Irrigation and Drainage, Sarawak. Driessen, P.M., 1978. Peat soils. In: IRRI, International Rice Research Institute, Soils and Rice, Los Banos, Philippines, pp. 763–779. Driessen, P.M. and Rochimah, L., 1976. The physical properties of lowland peats from Kalimantan. In: Peat and podzolic soils and their potential for agriculture in Indonesia. Proceedings ATA 106 Midterm Seminar, Soil Research Institute, Bogor, pp. 56–73. Easterling, D.R., Meehl, G.A., Parmesan, C., et al., 2000. Climate extremes: observations, modelling, and impacts. Science 289, 2068–2074. Esterle, J.S., 1990. Trends in petrographic and chemical characteristics of tropical domed peats in Indonesia and Malaysia as analogues for coal formation. Ph.D. Thesis, University of Kentucky, Lexington, KY, USA. Esterle, J.S., Calvert, G., Durig, D. et al., 1992. Characterization and classiﬁcation of tropical woody peats from Baram River, Serawak and Jambi, Sumatra. In: Aminuddin, B.Y. (Ed.), Proceedings International Symposium on Tropical Peatland, Kuching, Sarawak. MARDI, Kuala Lumpur, pp. 33–48. Esterle, J.S. and Ferm, J.C., 1994. Spatial variability in modern tropical peat deposits from Sarawak, Malaysia and Sumatra, Indonesia: analogues for coal. Int. J. Coal Geol. 26, 1–41. Esterle, J.S., Ferm, J.C., and Tie, Y.L., 1989. A test for the analogy of tropical domed peat deposits to ‘‘dulling up’’ sequences in coal beds – preliminary results. Org. Geochem. 14, 333–342. FAO, 1974. The peat swamp forests of Sarawak and their potential for development. United Nations Food and Agriculture Organization Forestry and Forest Industries Development, Malaysia. FO: DP/MAL/ 72/009 Technical Report No. 3 Kuala Lumpur, Malaysia. Flenley, J.R., 1979. The Equatorial Rain Forest: A Geological History. Butterworths, London.

168

S.E. Page, J.O. Rieley, R. Wu¨st

Flenley, J.R., 1985. Quaternary vegetational and climatic history of island SouthEast Asia. Mod. Quatern. Res. in SE Asia, 9, 55–63. Flenley, J.R., 1998. Tropical forests under the climates of the last 30,000 years. Clim. Change 39, 177–197. Gates, F.C., 1915. A Sphagnum bog in the tropics. J. Ecol. III, 24–30. Geyh, M.A., Kudrass, H.R., and Streif, H., 1979. Sea-level change during the late Pleistocene and Holocene in the Strait of Malacca. Nature 278, 441–443. Gorham, E., 1991. Northern peatlands: role in the carbon cycle and probable responses to climatic warming. Ecol. Applications, 1, 182–195. Haseldonckx, P., 1977. The geology of Pulau Tekong, Singapore. Geol. Soc. Malaysia 8, 75–87. Heaney, L.R., 1991. A synopsis of climatic and vegetational change in Southeast Asia. Clim. Change 19, 53–61. Hesp, P.A., Cheng, C.H., Hilton, M., et al., 1998. A ﬁrst tentative Holocene sea level curve for Singapore. J. Coast. Res. 14, 308–314. Hope, G.S., 2001. Environmental change in the Late Pleistocene and later Holocene at Wanda site, Soroako, South Sulawesi, Indonesia. Palaerogeogr. Palaeoclimatol. Palaeocol. 171, 129–145. Hunting Technical Services, 1969. Land capability study: physical resources and development priorities. Report submitted to the Government of State of Brunei. Junting Technical Services Ltd. Ibrahim, S. and Hall, J.B., 1992. A phytosociological survey of peat swamp forest at Pekan, Pahang, Peninsular Malaysia. In: Aminuddin, B.Y. (Ed.), Proceedings International Symposium on Tropical Peatland, Kuching, Sarawak, Malaysia, pp. 277–285. Immirzi, C.P. and Maltby, E., 1992. The global status of peatlands and their role in carbon cycling. A report for Friends of the Earth by the Wetland Ecosystems Research Group. Report 11, Department of Geography, University of Exeter, Exeter, UK. Friends of the Earth, London. Janzen, D.H., 1974. Tropical blackwater rivers, animals, and mast fruiting by the Dipterocarpaceae. Biotropica 6, 69–103. Jauhiainen, J., Takahasi, H., Jaya, A., et al., 2003, Tropical peat soil carbon emissions. In: Osaki, M. (Ed.), Proceedings Int. Symposium on Land Management and Biodiversity in SE Asia, Bali, Indonesia. Hokkaido University, Japan and Indonesian Institute of Sciences, pp. 107–114. Kottelat, M. and Ng, P.K.L., 1994. Diagnoses of ﬁve new species of ﬁghting ﬁshes from Banka and Borneo (Teleostei: Belontiidae). Ichthyol. Explor. Freshwaters 5, 65–78. Kurnain, A., Notohadikusumo, T., Radjagukguk, B., and Sri Hastuti, 2001. Peat soil properties related to degree of decomposition under different land use systems. Int. Peat J. 11, 67–77. Kurnain, A., Notohadikusumo, T., Radjagukguk, B., and Sri Hasuti, S., 2002. The state of decomposition of tropical peat soil under cultivated and ﬁre damaged peatland. In: Rieley, J.O., Page, S.E., and Setiadi, B. (Eds), Peatlands for People: Natural Resource Functions and Sustainable Management. Proceedings Int. Symposium on Tropical Peatland, Jakarta, Indonesia. BPPT (Badan Pengkajian dan Penerapan Teknologi) and Indonesian Peat Association, Jakarta, pp. 168–178. Lambert, K. and Staelens, N., 1993. The agricultural and horticultural value of tropical peatland, a case study. Int. Peat J. 5, 41–57. Lappaleinen, E., 1996. Global Peat Resources. International Peat Society, Jyvaskyla, Finland. Large, D.J., Jones, T.F., Briggs, J., et al., 2004. Orbital tuning and correlation of 1.7 m.y. of continuous carbon storage in an early Miocene peatland. Geology 32, 873–876. Lewis, W.M. Jr., 1986. Nitrogen and phosphorus runoff losses from a nutrient-poor tropical moist forest. Ecology 67, 1275–1282. Maloney, B.K., 1983/1984. A preliminary pollen diagram from Pea Sijajap, North Sumatra. Mod. Quatern. Res. SE Asia, 8, 35–42. Maloney, B.K., 1998. A 10,600-year pollen record from Nong Thale Song. Hong TrangProvince, South Thailand. J. Siam Soc. 86, 201–217. Maltby, E., C.P. Immirzi and R.J. Safford (Eds), 1996, Tropical lowland peatlands of Southeast Asia. IUCN, Gland, Switzerland, . Maxwell, A.L., 2001. Holocene monsoon changes inferred from lake sediment pollen and carbonate records, northeastern Cambodia. Quatern. Res. 56, 390–400. Maxwell, A.L. and Liu, K-B., 2002. Late Quaternary pollen and associated records from the monsoonal areas of continental South and SE Asia. In: Kershaw, A.P., Tapper, N.J., and David, B., et al. (Eds), Bridging Wallace’s Line. Advances in GeoEcology 34. Catena Verlag, Reiskirchen, pp. 189–228.

Lowland tropical peatlands of Southeast Asia

169

Meehl, G.A. and Washington, W.M., 1996. El Nin˜o-like climate change in a model with increased atmospheric CO2 concentrations. Nature 382, 56–60. Meijaard, E., 1997. The importance of swamp forest for the conservation of the orang utan (Pongo pygmaeus) in Kalimantan, Indonesia. In: Rieley, J.O. and Page, S.E. (Eds), Biodiversity and Sustainability of Tropical Peatlands. Samara Publications, Cardigan, UK, pp. 243–254. Melling, L., Hatano, R., and Goh, K.J., 2005. Soil CO2 ﬂux from three ecosystems in tropical peatland of Sarawak, Malaysia. Tellus 57B, 1–11. Merton, F., 1962. A visit to the Tasek Bera. Malay. Nat. J. 16, 103–110. Miyamoto, K., Kohyama, T., Suzuki, E., and Simbolon, H., 2000. Primary production of a heath (kerangas) forest in Lahei, Central Kalimantan. In: Iwakuma, T. (Ed.), Peatlands for People: Natural Resource Functions and Sustainable Management. Proceedings of the International Symposium on Tropical Peatland, Bogor, Indonesia. Hokkaido University, Japan and Indonesian Institute of Sciences, pp. 283–287. Morley, R.J., 1981. Development and vegetation dynamics of a lowland ombrogenous swamp in Kalimantan Tengah, Indonesia. J. Biogeog. 8, 383–404. Morley, R.J., 2000. Origin and Evolution of Tropical Rain Forests. Wiley, Chichester, Sussex. Morrogh-Bernard, H., Husson, S., Page, S.E., and Rieley, J.O., 2003. Status of the Bornean orang-utan (Pongo pygmaeus) in the Sebangau peat swamp forest, Central Kalimantan, Indonesia. Biol. Cons. 110, 141–152. Mutalib, A.A., Lim, J.S., Wong, M.H., and Koonvai, L., 1992. Characterization, distribution and utilization of peat in Malaysia. In: Aminuddin, B.Y. (Ed.), Proceedings of the International Symposium on Tropical Peatland, Kuching, Sarawak. MARDI, Kuala Lumpur, pp. 7–16. Neuzil, S.G., 1997. Onset and rate of peat and carbon accumulation in four domed ombrogenous peat deposits, Indonesia. In: Rieley, J.O. and Page, S.E. (Eds), Biodiversity and Sustainable Management of Tropical Peatlands. Samara, Cardigan, UK, pp. 55–72. Neuzil, S.G., Supardi, Cecil, C.B. et al., 1993. Inorganic geochemistry of domed peat in Indonesia and its implication for the origin of mineral matter in coal. In: Cobb, J.C. and Cecil, C.B. (Eds), Modern and Ancient Coal-Forming Environments. Geological Society of America Special Paper, Boulder, CO, pp. 23–44. Ng, P.K.L., Tay, J.B., and Lim, K.K.P., 1994. Diversity and conservation of blackwater ﬁshes in Peninsular Malaysia, particularly in the north Selangor peat swamp forest. Hydrobiologia 285, 203–218. Ong, B.Y. and Yogeswaran, M. (1992). Peatland as a resource for water supply in Sarawak. In: Aminuddin, B.Y. (Ed.), Proceedings of the International Symposium on Tropical Peatland, Kuching, Sarawak. MARDI, Malaysia, pp. 225–268. Oraveinen, H., Crisologo, E.J., Abando, R. et al. 1992. Pre-feasibility study of the Philippines peat resources. In: Aminuddin, B.Y. (Ed.), Proceedings of the International Symposium on Tropical Peatland, Kuching, Sarawak. MARDI, Kuala Lumpur, pp. 49–56. Page, S.E. and Rieley, J.O., 1998. Tropical peatlands: a review of their natural resource functions, with particular reference to Southeast Asia. Quatern. Res. 8, 95–106. Page, S.E., Rieley, J.O., Doody, K. et al., 1997. Biodiversity of tropical peat swamp forest: a case study of animal diversity in the Sungai Sebangau catchment of Central Kalimantan, Indonesia. In: Rieley, J.O. and Page, S.E. (Eds), Biodiversity and Sustainable Management of Tropical Peatlands. Samara, Cardigan, UK, pp. 231–242. Page, S.E., Rieley, J.O., Shotyk, W., and Weiss, D., 1999. Interdependence of peat and vegetation in a tropical swamp forest. Philos. Trans. Roy. Soc. B 354, 1885–1897. Page, S.E., Siegert, F., Rieley, J.O., et al., 2002. The amount of carbon released from peat and forest ﬁres in Indonesia during 1997. Nature 420, 61–65. Page, S.E., Wu¨st, R.A.J., Weiss, D., et al., 2004. A record of Late Pleistocene and Holocene carbon accumulation and climate change from an equatorial peat bog (Kalimantan, Indonesia): implications for past, present and future carbon dynamics. J. Quatern. Sci. 19, 625–635. Penny, D., 2001. A 40,000-year palynological record from north-east Thailand; implications for biogeography and paleo-environmental reconstruction. Palaeogeogr. Palaeoclim. Palaeoecol. 171, 97–128. Phillips, S. and Bustin, R.M., 1996a. Sedimentology of the Changuinola peat deposit: organic and clastic sedimentary response to punctuated coastal subsidence. Geol. Soc. Am. Bull. 108, 794–814.

170

S.E. Page, J.O. Rieley, R. Wu¨st

Phillips, S. and Bustin, R.M., 1996b. Sulfur in the Changuinola peat deposit, Panama, as an indicator of the environments of deposition of peat and coal. J. Sed. Res. 66, 184–196. Phillips, S. and Bustin, R.M., 1998. Accumulation of organic rich sediments in a dendritic ﬂuvial/lacustrine mire system at Tasik Bera, Malaysia: implications for coal formation. Int. J. Coal Geol. 36, 31–61. Polak, B., 1975. Character and occurrence of peat deposits in the Malaysian tropics. Mod. Quatern. Res. SE Asia, 2, 71–81. Polak, E., 1933. Ueber Torf und Moor in Niederla¨ndisch Indien. Verhandelingen der koninklijke Akademie van Wetenschappen te Amsterdam, No. 3. Van de N.V. Noord-Hollandische UitgeversMaatschappi, Amsterdam, Vol. XXX, 84pp. Prentice, C. and Parish, D., 1992. Conservation of peat swamp forest: a forgotten ecosystem. Proceedings of the International Conference on Tropical Biodiversity. Malaysian Nature Society, Kuala Lumpur, Malaysia, pp. 128–144. Radjagukguk, B., 1992. Utilization and management of peatlands in Indonesia for agriculture and forestry. In: Aminuddin, B.Y. (Ed.), Proceedings of the International Symposium on Tropical Peatland, Kuching, Sarawak, Malaysia. Malaysian Agriculture Research Development Institute & Department of Agriculture, Sarawak, Kuching, Malaysia, pp. 21–27. Radjagukguk, B., 1997. Peat soils of Indonesia: location, classiﬁcation and problems for sustainability. In: Rieley, J.O. and Page, S.E. (Eds), Biodiversity and Sustainability of Tropical Peatlands. Samara, Cardigan, UK, pp. 45–54. RePPProt, 1988. A National Overview from the Regional Physical Planning Programme for Transmigration. UK Overseas Development Administration and Directorate Bina Program. Ministry of Transmigration, Jakarta. RePPProt, 1990. A National Overview from the Regional Physical Planning Programme for Transmigration. UK Overseas Development Administration and Directorate Bina Program. Ministry of Transmigration, Jakarta. Rieley, J.O. and Ahmad-Shah, A.A., 1996. The vegetation of tropical peat swamp forests. In: Maltby, E., Immirzi, C.P., and Safford, R.J. (Eds), Tropical Lowland Peatlands of Southeast Asia. IUCN, Gland, Switzerland, pp. 55–73. Rieley, J.O., Ahmad-Shah, A-A., and Brady, M.A., 1996. The extent and nature of tropical peat swamps. In: Maltby, E., Immirzi, C.P., and Safford, R.J. (Eds), Tropical Lowland Peatlands of Southeast Asia. IUCN, Gland, Switzerland, pp. 17–53. Rieley, J.O. and Page, S.E. (Eds), 1997. Biodiversity and Sustainability of Tropical Peatlands. Samara, Cardigan, UK. Ritzema, H.P. and Tuong, T.P. (1994). Water-management strategies as a tool for the sustainable use of acid sulphate soils. Regional workshop on the sustainable use of coastal land in south-east Asia. Asian Institute of Technology, Bangkok, 34pp. Ryer, T.A. and Langer, A.W., 1980. Thickness change involved in the peat-to-coal transformation for a bituminous coal of Cretaceous age in central Utah. J. Sed. Petrol. 50, 987–992. Sajarwan, A., Notohadiprawiro, T., Radjagukguk, B., and Hastuti, S., 2002. Diversity of tropical peat characteristics in intact peatland forest, under the inﬂuence of forest type, peat thickness, and position of peat deposit. In: Rieley, J.O., Page, S.E., and Setiadi, B. (Eds), Peatlands for People: Natural Resource Functions and Sustainable Management. Proceedings of the International Symposium on Tropical Peatland, Jakarta, Indonesia. BPPT (Badan Pengkajian dan Penerapan Teknologi) and Indonesian Peat Association, Jakarta, pp. 119–124. Schweithelm, J., 1999. The ﬁre this time. An overview of Indonesia’s forest ﬁres in 1997/98. Discussion Paper, Jakarta, WWF Indonesia, 51pp. Sheldon, F., 1987. Habitat preferences of the Hook-billed Bulbul Setornis criniger and the White-throated Babbler Malacopteron albogulare in Borneo. Forktail 3, 17–25. Shepherd, P.A., Rieley, J.O., and Page, S.E., 1997. The relationship between forest vegetation and peat characteristics in the upper catchment of Sungai Sebangau, Central Kalimantan. In: Rieley, J.O. and Page, S.E. (Eds), Biodiversity and Sustainability of Tropical Peatlands. Samara, Cardigan, UK, pp. 191–210. Shier, C.W., 1985. Tropical peat resources – an overview. Proceedings Symposium on Tropical Peat Resources – Prospects and Potential. Kingston, Jamaica, International Peat Society, Helsinki, pp. 29–46.

Lowland tropical peatlands of Southeast Asia

171

Sieffermann, R.G., Fournier, M., Truitomo, S. et al., 1988. Velocity of tropical forest peat accumulation in Central Kalimantan province, Indonesia (Borneo). In: Proceedings 8th Internatinal Peat Congress, Leningrad, USSR. International Peat Society, Jyva¨skyla¨, Finland, Vol. 1, pp. 90–98. Sieffermann, R.G., Rieley, J.O., and Fournier, M., 1992. The lowland peat swamps of Central Kalimantan (Borneo): a complex and vulnerable ecosystem. In: Proceedings of the International Conference Geography in the Asean Reg., Yogyakarta, Indonesia, 22pp. Silvius, M.J., Simons, H.W., and Verheught, W.J.M., 1984. Soils, Vegetation and Nature Conservation of the Berbak Game Reserve, Sumatra, Indonesia. Arnhem Research Institute for Nature Management, Arnhem, Netherlands. Soepraptohardjo, M. and Driessen, P.M., 1976. The lowland peats of Indonesia, a challenge for the future. Soil Res. Inst., Bogor Bull. 3, 11–19. Sorensen, K.W., 1993. Indonesian peat swamp forests and their role as a carbon sink. Chemosphere 27, 1065–1082. Staub, J.R. and Esterle, J.S., 1994. Peat-accumulating depositional systems of Sarawak, East Malaysia. Sediment. Geol. 89, 91–106. Stuijts, I., Newsome, J.C., and Flenley, J.R., 1988. Evidence for late Quaternary vegetational change in the Sumatran and Javan Highlands. Rev. Palaeobot. Palynol. 55, 207–216. Suhardjo, H. and Widjaja-Adhi, I.P.G., 1976. Chemical characteristics of the upper 30 cm of peat soils from Riau. Proc. ATA 106 Midterm Seminar, Peat and podzolic soils and their potential for agriculture in Indonesia, Tugu, Indonesia. Suhardjo, H. and Widjaja-Adhi, I.P.G., 1977. Chemical characteristics of the upper 30 cm of peat soils from Riau, Sumatra (Indonesia). In: Final Report, Agricultural Technical Assistance Programme (Indonesia – The Netherlands) 1974–1977. Lembaga Penelitian Tanah, Bogor, Indonesia, pp. 74–92. Sulistiyanto, Y., 2004. Nutrient dynamics in different sub-types of peat swamp forest in Central Kalimantan, Indonesia. Ph.D. Thesis, Department of Geography, University of Nottingham, UK. Suryanto, 1994. Improvement of the P nutrient status of tropical ombrogenous peat soils from Pontianak, West Kalimantan, Indonesia. Ph.D. Thesis, Faculty of Agricultural and Applied Biological Sciences, RUG, Gent. Takahashi, H., Shimada, S., Ibie, B.I. et al., 2002. Annual changes of water balance and a drought index in a tropical peat swamp forest of Central Kalimantan, Indonesia. In: Rieley, J.O., Page, S.E. and Setiadi, B. (Eds), Peatlands for People: Natural Resource Functions and Sustainable Management, Proceedings of the International Symposium on Tropical Peatland, Jakarta, Indonesia. BPPT (Badan Pengkajian dan Penerapan Teknologi) and Indonesian Peat Association, Jakarta, pp. 63–67. Takahashi, H. and Yonetani, Y., 1997. Studies on microclimate and hydrology of peat swamp forest in Central Kalimantan, Indonesia. In: Rieley, J.O. and Page, S.E. (Eds), Biodiversity and Sustainability of Tropical Peatlands. Samara, Cardigan, UK, pp. 179–187. Tay, T.H., 1969. The distribution, characteristics, uses and potential of peat in West Malaysia. J. Trop. Geog. 29, 57–63. Taylor, D., Yean, O.W., Sanderson, P.G., and Dodson, J., 2001. Late Quaternary peat formation and vegetation dynamics in a lowland tropical swamp; Nee Soon, Singapore. Palaerogeogr. Palaeoclimatol. Palaeocol. 171, 269–287. Tie, Y.L., 1990. Studies of peat swamps in Sarawak with particular reference to soil–forest relationships and development of dome-shaped structures. Ph.D. Thesis, Department of Geography, Polytechnic of North London, UK. Tie, Y.L. and Esterle, J.S., 1992. Formation of lowland peat domes in Sarawak, Malaysia. In: Aminuddin, B.Y. (Ed.), Proceedings of the International Symposium on Tropical Peatland, Kuching. MARDI, Kuala Lumpur, Malaysia, pp. 81–89. Tjia, H.D., 1992. Holocene sea-level changes in the Malay-Thai Peninsula, a tectonically stable environment. Bull. Geol. Soc. Malaysia 31, 157–176. Tjia, H.D., Sujitno, S., Suklija, Y., et al., 1984. Holocene shorelines in the Indonesian Tin Islands. Mod. Quat. Res. SE Asia, 8, 103–117. Turunen, C. and Turunen, J., 2003. Development history and carbon accumulation of a slope bog in oceanic British Columbia, Canada. Holocene 13, 225–238.

172

S.E. Page, J.O. Rieley, R. Wu¨st

Turunen, J., Tomppo, E., Tolonen, K., and Reinikainen, E., 2002. Estimating carbon accumulation rates of undrained mires in Finland – application to boreal and subarctic regions. Holocene 12, 79–90. van der Kaars, W.A., Penny, D., Tibby, J., et al., 2001. Late Quaternary palaeoecology, palynology and palaeolimnology of a tropical lowland swamp: Rawa Danau, West Java, Indonesia. Palaerogeogr. Palaeoclimatol. Palaeocol. 171, 129–145. van der Werf, G.R., Randerson, J.T., Collatz, G.J., et al., 2004. Continental-scale partitioning of ﬁre emissions during the 1997 to 2001 El Nin˜o/La Nin˜a period. Science 303, 73–76. Vijarnsorn, P. and Liengsakul, M., 1986. Formation and characterization of the peat swamps in Narathiwat Province, Peninsular Thailand. In: Narong, T. (Ed.), Workshop on Economic Geology, Tectonics, Sedimentary Processes and Environment of the Quarternary in Southeast Asia. Haadyai, Thailand, pp. 87–100. Wayi, B.M. and Freyne, D.F., 1992. The distribution, characterisation, utilisation and management of peat soils in Papua New Guinea. In: Aminuddin, B.Y. (Ed.), Proceedings International Symposium on Tropical Peatland. Kuching, Sarawak, Malaysia, pp. 28–32. Weiss, D., Shotyk, W., Rieley, J., et al., 2002. The geochemistry of major and selected trace elements in a forested peat bog, Kalimantan, SE Asia, and its implications for past atmospheric dust deposition. Geochim. Cosmochim. Acta 66, 2307–2323. Whitmore, T.C., 1984. Tropical Rain Forests of the Far East. Clarendon Press, Oxford, UK. Wibowo, P., Hasudungan, F., Syah, R., and Noor, Y.R., 2000. White-winged Duck Surveys in North Sumatra. Threatened Waterfowl Specialist Group News 12, 11. Wichmann, C.E.A. (1910). De Veenen in den Indischen Archipel. Verslag Gew. Verg. Wis- en Natuurk. Afd. Der. Kon. Akad. Wetensch., Amsterdam. Widjaja-Adhi, I.P.G., 1988. Physical and chemical characteristics of peat soils of Indonesia. I.A.R.D. Journal 3, 59–64. Wilford, G.E., 1959. Radiocarbon age determination of Quaternary sediments in Brunei and northeast Sarawak. British Borneo Geological Survey, Annual Report 1959, pp.16–20. Wo¨sten, J.H.M., Ismail, A.B., and van Wijk, A.L.M., 1997. Peat subsidence and its practical implications: a case study in Malaysia. Geoderma 78, 25–36. Wu¨st, R.A.J. and Bustin, R.M., 1999. Geological and Ecological Evolution of the Tasek Bera (Peninsular Malaysia) Wetland bBasin since the Holocene: Evidences of a Dynamic System from Siliciclastic and Organic Sediments, Wetlands International-Asia Paciﬁc, Kuala Lumpur, Malaysia. Wu¨st, R.A.J. and Bustin, R.M., 2003. Opaline and Al–Si phytoliths from a tropical mire system of West Malaysia: abundance, habit, elemental composition, preservation and signiﬁcance. Chem. Geol. 200, 267–292. Wu¨st, R.A.J. and Bustin, R.M., 2004. Late Pleistocene and Holocene development of the interior peataccumulating basin of tropical Tasek Bera, Peninsular Malaysia. Paleogeog. Paleoclim. Paleoecol. 211, 241–270. Wu¨st, R.A.J., Bustin, R.M., and Lavkulich, L., 2003. New classiﬁcation systems for tropical organic-rich deposits based on studies of the Tasek Bera Basin, Malaysia. Catena 53, 133–163. Wu¨st, R.A.J., Ward, C.R., Bustin, R.M., and Hawke, M.I., 2002. Characterization and quantiﬁcation of inorganic constituents of tropical peats and organic-rich deposits from Tasek Bera (Peninsular Malaysia): implications for coals. Int. J. Coal Geol. 49, 215–249. Wyatt-Smith, J., 1959. Peat swamp forest in Malaya. The Malayan Forester 22, 5–32. Wyatt-Smith, J., 1964. A preliminary vegetation map of Malaysia with descriptions of the vegetation types. J. Trop. Geog. 18, 200–213. Yamakura, T., Hagihara, A., Sukardjo, S., and Ogawa, H., 1986. Aboveground biomass of tropical rainforest stands in Indonesian Borneo. Vegetatio 66, 71–82.

173

B. Selected characteristics of peat and peatland environments Section B deals principally with the geochemistry, microbiology, and hydrology of peatlands and the local instability of their peat deposits. A basic geochemical framework in terms of the environmental variables pH and Eh is constructed in Chapter 8. Chapter 9 examines the behavior of inorganic materials and the distribution and movement of trace elements in bogs. The organic chemistry of the peatlands of Galicia (northwestern Spain) is described in Chapter 10. Chapter 11 studies elemental distributions resulting from the incorporation of microscopic-terrigenous particles, including cosmogenic ones. The important nutrient element nitrogen is dealt within Chapter 12, in terms of its accumulation in bogs, and the consequent response of plants. Chapter 13 illustrates the great microbiological diversity of mires, and both this and the previous chapter, reports on the effect of changing conditions on the diversity of species in the examples studied. Hydrological study shows that complex water movements through micro- and macro-pores (Chapter 14) may produce considerable diversity in peatland types. This can lead, even in unconﬁned lowland settings, to the juxtaposition of different mires on a local scale (Chapter 15). Chapter 16 considers failure and slumping in peat deposits as a result of shear stresses produced under high water pressure.

This page intentionally left blank

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

175

Chapter 8

The redox–pH approach to the geochemistry of the Earth’s land surface, with application to peatlands W. Chesworth, A. Martı´ nez Cortizas and E. Garcı´ a-Rodeja

Introduction To a ﬁrst approximation, any scheme that seeks to deﬁne the broad geochemical evolution of the land surface of the Earth needs to take the following features into account. The ubiquitous presence and importance of water. Geochemical change in materials in soils and other weathering systems depends primarily on water: its supply, abundance and residence time in the system. The water acts physically as an agent of transport, as a medium through which reactants diffuse to and from sites of reaction, and in freezing exerts a pressure capable of disintegrating rocks along discontinuities in their structure from the scale of a grain boundary to that of the jointing system. In addition it exerts a partial pressure that ranges from the saturation value to lower, but never completely zero, values in the arid regions. Chemically, water acts as a solvent, as a reactant in all important reactions in the weathering zone, a chemical constituent of many secondary minerals, especially clays and hydroxides, and as a chemical buffer ﬁxing the thermodynamic activity of soil solution at one, or close to it, except in arid systems and episodes such as in the formation of sal/sodic soils. The ubiquitous presence and importance of organic matter. Organic materials may range from total dominance in the system as in ombrotrophic peatlands to a minor or insigniﬁcant role in very young or incipient soils such as lithosols. In the A horizon of a soil, organic matter undergoes decay, and organic acids and complexants are produced. These promote the weathering of inorganic constituents and the mobilization of inorganic ions in solution. The concentration of organic material attenuates with depth, and is generally taken to be negligible, or at least not obvious in the C horizon of a soil. The nature of the inorganic substrate. The dominant minerals in virtually all soils are silicate or aluminosilicate in composition. Over time they tend to be stripped of the so called basic cations, and to be converted into some combination of quartz, 1:1 sheet silicates, and hydroxides of Al and Fe, thereby producing acid soils of low ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09008-0

176

W. Chesworth, A. Martı´nez Cortizas, E. Garcı´a-Rodeja

cation exchange capacity. If calcite is present, it reacts more rapidly than the (alumino)silicates and a pH between 7 and 8 is established. A rarer occurrence is the presence of sulﬁdes, pyrite being the most likely. It too weathers more rapidly than the (alumino)silicates, and gives rise to waters with a pH of 3 or less. Of the elemental components in the inorganic substrate, Fe plays an extremely important role. It is the most abundant element with more than one valence, and takes part in reactions between the soil solution and ferric-hydroxy phases that relatively quickly set the overall redox state of the soil. The physical nature of the substrate is also of importance, particularly its hydrodynamic properties. The latter determine rates of drainage of the material and hence the degree of saturation with respect to water. The partial pressure of O2. Initial values (20 kPa) are set by atmospheric O2. In the vadose zone of the weathering system it will be essentially the same near the surface, but with depth and the increased biological oxygen demand of aerobic decay of organic matter the value will be less, and as microbial organisms proliferate and use up all the available resource of uncombined oxygen, will tend to reach the limit where PO2 is equal to PH 2 , and the system as a whole changes from aerobic to anaerobic. This limit will normally coincide with the water table, below which any pore space will be saturated with water to the exclusion of all oxygen except the small amounts actually dissolved in the water right at the water table. In some cases, lateral inﬂow of oxygenated water, may result in anomalously high levels of oxygen within the saturated zone. The partial pressure of CO2. Again the initial value is set by the atmospheric partial pressure: 101.5 kPa (103.5 atm). Respiration in the root zone, and decay of organic matter may raise this up to a maximum of about 100 times the atmospheric value. PCO2 determines the pH of the H2O–CO2 (carbonic acid) system, with 5.7 as the common value in unpolluted rainwater. In the absence of organic acids, carbonic acid is the most important one in acid–base reactions in the weathering mantle. This means that its chemical importance in soil tends to increase with depth as the importance of the organic component diminishes. Clearly there is ample opportunity for feedback among these factors, and taken together, they determine that the commonest types of chemical reaction within a weathering system are of acid–base (including cation-exchange) and redox type. Consequently, serviceable choices of master variables for the soil environment are those that are related to the thermodynamic activities of the proton and the electron. Regarding the activity of protons in a system, the universal variable of choice is pH. Regarding electrons, the equivalent parameter pe may be chosen, but the classic variable employed is Eh, which has the advantage of being measurable both in laboratory and ﬁeld. As shown in Figure 8.1, there is a simple relationship between pe and pH. The application of this type of diagram to weathering systems generally and to peatlands in particular will now be described.

Physico-chemical background Figure 8.1 is variously referred to as a redox–pH diagram, an Eh–pH diagram, a pe–pH diagram, or simply a Pourbaix diagram, after the metallurgist who devised this

The redox– pH approach to the geochemistry of the Earth’s land surface

177

Figure 8.1. Pourbaix diagram. The sloping, dashed line separating oxidizing from reduced conditions is for PH 2 ¼ PO2 in the equilibrium H2O ¼ H2+0.5 O2.

type of Cartesian graph for the purposes of examining corrosion (Pourbaix, 1974). In earth science their use was pioneered by Krumbein and Garrels (1952), and a good introduction is provided by Garrels and Christ (1965). A compilation of basic diagrams useful to the geochemist may be found in Brookins (1988). Truesdell (1968) suggested the adoption of pe rather than Eh as the redox parameter as a means of simplifying the calculation of redox–pH diagrams from fundamental thermodynamic data. Berner (1970) details a number of practical limitations related to the construction and interpretation of Eh–pH diagrams. At the most fundamental level, there is the problem of the quality of the thermodynamic data upon which the diagrams are based. In this regard, Woods and Garrels (1987, p. ix) refer to ‘‘the perils of indiscriminate selection of values from various sources’’. A major difﬁculty, and one which clearly has a determining inﬂuence on any ultimate conclusions to be drawn from the diagrams, lies in choosing what minerals to show on them, whether to consider metastable phases, whether to consider solid solution in condensed phases, and what to do about bacterial reductions. Furthermore, it is assumed that equilibrium (stable or metastable) holds for a particular diagram constructed. Finally, there is the major difﬁculty of obtaining good measurements of Eh in the ﬁeld (Shotyk, 1988). In spite of all this, a consensus exists that the diagrams are useful in indicating general conditions and general tendencies in a system. Even for more parochial considerations, redox–pH diagrams may prove useful provided that matters of scale, and the possibilities of local reactions are kept in mind.

178

W. Chesworth, A. Martı´nez Cortizas, E. Garcı´a-Rodeja

Figure 8.2. Predominance ﬁelds in the system Fe–Ca–K–S–CO2–H2O. Conditions: 100 kPa total pressure, 251C, activities of all components except S, 106. S is undeﬁned in terms of a speciﬁc activity in order to show the sulﬁde ﬁeld at its maximum extent.

Predominance ﬁelds Figure 8.2 is a redox–pH diagram of the type used in the geochemical interpretation of low-pressure/temperature environments. Areas called predominance ﬁelds are labeled in terms of the minerals that would be expected to form under the Eh and pH combinations of those ﬁelds. The actual positions of the boundaries between ﬁelds depend on the choice of ionic activities made in calculating the appropriate mineralogical reactions for the system. A total pressure of 100 kPa and a standard temperature of 25 oC are usually chosen as representative ambient conditions. This diagram is of special interest in the context of the geochemistry of peatlands, as will be made clear later.

Geochemical fences A notable contribution of Krumbein and Garrels (1952) to the usefulness of the Eh–pH approach in earth science is the concept of the geochemical fence: a relatively narrow, linear zone in redox–pH space deﬁned by a speciﬁc chemical reaction or by a related group of reactions. Their original diagram, developed to help in the

The redox– pH approach to the geochemistry of the Earth’s land surface

179

Figure 8.3. The original system of geochemical fences devised by Krumbein and Garrels (1952).

interpretation of sedimentary environments, is shown in Figure 8.3. It requires modiﬁcation if the full scope of soil conditions are to be covered, especially below pH 7. In practical terms this amounts to changing existing geochemical fences, and adding new ones. The modiﬁed fences, shown in Figure 8.4, are: Organic fence. Within the upper organic-rich part of the solum, microbially mediated breakdown of organic matter will give a range of possible Eh values for a given pH, up to the lower limit of oxidation by O2 in soil. The latter will likely coincide with the water table in the soil system as a whole, with the exclusion of oxygen to microsystems being controlled by details of micromorphology, for example by the presence of clay cutans protecting a mineral surface from oxidation. Above the water table the microbial biomass, using dead organic matter as its carbon source, will generally experience exponential growth up to the point where free O2 is fully utilized. Again, this will obviously coincide with the lower limit of oxidation by O2. It should therefore be placed at the Eh–pH contour along which PO2 is equal to PH 2 in the gaseous dissociation reaction of H2O, rather than where Eh is zero, the value chosen by Garrels and Christ (1965). In theory, the upper limit would be marked by the partial pressure of oxygen in the atmosphere. In practice, the biological oxygen demand in virtually all soils depletes oxygen to lower levels than this.

180

W. Chesworth, A. Martı´nez Cortizas, E. Garcı´a-Rodeja

Figure 8.4. The pedogenic grid (after Chesworth 1992), an extension of the Krumbein and Garrells diagram (Fig. 8.3) to include new fences and fences modiﬁed from the diagram. The extended diagram is applicable to pH values lower than Krumbein and Garrells consider, including values found in bogs, many mires, and acid soils (see Figs 8.9, 8.11, 8.13).

The width of the organic fence shown in Figure 8.7 is consistent with measurements in such materials. Neutral fence. Garrels and Christ (1965) give no mineralogical justiﬁcation for a neutral fence, and in the ubiquitous presence of atmospheric and respiratory CO2 in a weathering system, it is unlikely that the ionic dissociation of water will buffer a natural system to a pH of 7. However, an average smectite (Weaver and Pollard, 1973), with so-called basic cations (Mg, Ca, Na, K) in exchange positions, marks a near-neutral fence according to the reaction: Mg0:2 ðSi3:81 Al1:71 Mg0:29 O10 ðOHÞ2 ðmontmorilloniteÞ þ 0:98 Hþ þ 0:22e þ 0:33H2 O ¼ 0:86 Al2 Si2 O5 ðOHÞ4 ðkaoliniteÞ þ 0:22 FeOOH ðgoethiteÞ þ 2:1SiO2 ðquartzÞ þ 0:49Mg2þ Because of minor substitution of Fe in the octahedral layer it will not be a vertical fence as in the Garrels and Christ diagram (Garrels and Christ, 1965). It should also be realized that this particular fence does not mark the lower pH limit of a smectite predominance ﬁeld, since there exists a spectrum of smectites down to beidellite that is stable at low pH. Fe hydroxide fence. Garrels and Christ (1965) based their Fe-oxide fence on the solubility of Fe2O3. In the weathering regime Fe-hydroxy phases are overwhelmingly

The redox– pH approach to the geochemistry of the Earth’s land surface

181

more important. Goethite is generally accepted as the stable hydroxide, though ferrihydrite is probably its precursor. The new fences in Figure 8.4 are : An aluminosilicate fence. This is deﬁned by reactions in the system SiO2– Al2O3–H2O. For example by the congruent dissolution of kaolinite: Al2 Si2 O5 ðOHÞ4 ðkaoliniteÞ þ 6Hþ ¼ 2Al3þ þ 2SiO2 ðquartzÞ þ 5H2 O ðpH4SiO4 o4:7Þ by the incongruent dissolution of kaolinite: Al2 Si2 O5 ðOHÞ4 ðkaoliniteÞ þ 5H2 O ¼ 2AlðOHÞ3 ðgibbsiteÞ þ 2H4 SiO4 ðquartzÞ ðpH4SiO4 o4:7Þ and by dissolution of gibbsite: AlðOHÞ3 ðgibbsiteÞ þ 3Hþ ¼ Al3þ þ 3H2 O A sodium carbonate fence. This deﬁnes the high-pH extreme of the soil environment, and lies in the pH range of feldspar dissolution, if the latter were to reach a saturated equilibrium, as well as the range in which zeolites form authigenically under earth-surface conditions. In addition, two fences on the original Krumbein and Garrels diagram are left essentially in their original form. These are: The limestone fence. Renamed the calcite fence, on the high pH side of which solid calcite persists in a soil system. The sulfate– sulfide fence. Name simpliﬁed to sulﬁde fence, marking the upper limit of Eh of the ﬁeld wherein sulﬁdes, and in particular pyrite, may form in a soil. Other geochemical fences may be useful in speciﬁc or local situations. For example, in ombrotrophic bogs with low-atmospheric input of particulate matter, the lowpH limit could be marked by a fence deﬁned by the dissociation of H+ from carboxylic functional groups in the organic component of the soil system. An even lower pH fence can be deﬁned for the interpretation of acid-sulfate soils by the breakdown products of pyrite. Where neither organic matter nor pyrite is common, the system H2O–CO2 will be useful in deﬁning a fence from about pH 5.7 to 5.5 depending upon the balance between atmospheric and respiratory inputs. Under extreme reduction a CH4 fence comes into play.

The pedogenic grid The resulting adaptation (Fig. 8.4) of the Krumbein and Garrels (1952) diagram is referred to as the pedogenic grid (Chesworth, 1992) and covers the broad range of conditions in common mineral soils. Certain organic soils and peatlands, and less common soils such as acid-sulfate soils, lie outside this range. Parallels exist between the pedogenic grid and the pressure–temperature (PT) diagram of the petrologist. The invariant point of a PT diagram, where pressure and temperature are held constant, is similar to those regions on the grid where two

182

W. Chesworth, A. Martı´nez Cortizas, E. Garcı´a-Rodeja

fences cross. Chesworth (1992) called such crossing points nodes. They act as quasiinvariant points where the pH and Eh of a system is buffered and poised, respectively. A further analogy arises from the fact that the essentially univariant (or pseudounivariant) fences of the grid delimit separate ranges of conditions with each area being characterized by its own individual assemblage of minerals. This is implicit in the diagram of Krumbein and Garrels (Fig. 8.3).

Geochemical trends in the weathering zone Three general trends are followed in the geochemical evolution of surface materials in the weathering zone (Fig. 8.5). Essentially they are the result of the pumping of protons and electrons between sources and sinks in the land environment. Proton and electron pumps A consequence of considering the activities of the proton (H+) and electron (e) as the major driving forces in effecting chemical change in soils (H+ in hydrolysis, and acid-base reactions generally, e in oxidation–reduction reactions) is that conceptually both can be considered as being pumped between a source (or sources) and sink (or sinks) in the weathering regime. First consider H+. On the largest scale, weathering may be described as the titration of acids from atmospheric and organic sources, against bases represented by the aluminosilicate, carbonate, and other mineralogical constituents of the earth’s surface. Figure 8.6 illustrates this within Eh–pH space as a proton pump between source and sink. In a humid climate, with adequate atmospheric precipitation, the land surface is ‘‘inevitably over-titrated, acting as a sink for protons’’ (Edmond et al., 1979, p. 21). Given enough time, all so-called base cations (Na, K, Ca, Mg) will be stripped from solid phases and removed by leaching. Starting from calcareous materials, e.g., solution of calcite, would buffer the system between pH 7 and 8 until it was completely dissolved. Then progressive overtitration would lead to hydrolysis and the formation of a residuum made up of such minerals as kaolinite, gibbsite, and goethite, together with resistant minerals, of which quartz is the commonest. Eventually (and with the proviso that no metal sulﬁdes were present) a steady (or quasi-steady) state would be set up, buffered by the aluminosilicate fence of the pedogenic grid. How close a given weathering system will approach this steady state will depend largely on the availability of water, and the hydrodynamics of the landscape. A close approach is favored by high rainfall in warm climates, and with a good drainage to carry away dissolved cations. In addition, the persistence of phases such as calcite and smectites saturated with the so-called basic cations Ca, Mg, Na, and K will buffer pH and retard the progress of acidiﬁcation. In humid conditions, and taking the long-term geological perspective, this can only be considered a transient state, the lifespan of which will depend on such factors as landscape position (shorter lifespan on slopes and topographic highs, a longer one in less well-drained topographic lows)

The redox– pH approach to the geochemistry of the Earth’s land surface

183

Figure 8.5. The three major geochemical trends in the weathering zone on the land surface of the Earth. The arrows indicate an acid trend characteristic of regions of humid climate, an alkaline trend characteristic of dry climates, and a reduced (or hydromorphic) trend found, for example, under waterlogged conditions in peatlands and gleysolic soils.

and texture of the weathering materials, coarse, well-drained deposits leading to a shorter persistence of the basic minerals, than ﬁner, less well-drained ones. Semi-arid and arid conditions will lead to an opposite trend; that is, a trend of undertitration wherein it is the basis in the system that predominates over the additions of H+, and the system is buffered at an alkaline pH. The development of sodicity, and in the limit the precipitation of salts of the alkali metals, is the likely result. Figure 8.7 is a similar interpretation regarding e. The most reduced material in the biosphere is the living biomass. When death and decay intervene, organic materials become a proliﬁc source of electrons in soil and in effect pump electrons to any available sink, the largest of which is atmospheric oxygen. For Malthusian reasons (the tendency of organisms to use resources to the maximum), any resource they depend upon will normally mean that the overall Eh of the soil will be poised along the contour in Figure 8.1 that represents this limit (PO2 ¼ PH 2 ). For example, if the ready diffusion of oxygen into the soil is impeded, for reasons of water logging, heavy texture, or compaction , microbial activity will be fuelled by

184

W. Chesworth, A. Martı´nez Cortizas, E. Garcı´a-Rodeja

Figure 8.6. The proton pump, showing the movement of hydrogen ions from sources such as dissociating carboxylic groups in decaying organic matter, and from carbonic acid from atmospheric precipitations, to sinks represented by the inorganic materials of the Earth surface, ranging from carbonate to silicate materials. In effect, this is a natural titration, which in humid climates leads to overtitration (and low pH values at the Earth surface) and in dry climates to undertitration (and higher values at the Earth surface).

the movement of electrons from the organic matter to some other available sink. The range of operation of these alternative electron acceptors is well known, and is shown in Figure 8.8. At ﬁner scales, particularly in the contact zones between phases, and on the reactive interface between solid and soil solution, an immense number of transfers of H+ and e are possible. On the whole, however, and on the macroscopic scale of the solum presented here, the general tendencies outlined are reasonable ones, at least semi-quantitatively. The peatland environment From the geochemical standpoint of this chapter, the most signiﬁcant differences within the peatland environment are shown within the framework of the system Fe–Ca–K–S–CO2–H2O (Fig. 8.2). Our interpretation is based on the following observations developed from the foregoing discussion. Presence of a water table divides any peatland into an oxidized and a reduced zone. This of course is recognized in the acrotelm/catotelm vertical stratiﬁcation in mires.

The redox– pH approach to the geochemistry of the Earth’s land surface

185

Figure 8.7. The electron pump. This represents the overall movement of electrons at the macroscale of a bog, mire, or soil landscape from the principal source in decaying organic matter, to the principal sink of atmospheric oxygen. In a decaying peat, the oxidation state would rise in the direction of the heavy arrows. Oxidation of a peat deposit in the water-unsaturated acrotelm, would lead to conditions above the PO2 ¼ PH 2 contour, and as oxidation proceeded PO2 ¼ 100 kPa contours. Conditions in the watersaturated catotelm would normally rise no further than the PO2 ¼ PH 2 contour, unless there was an inﬂux of oxygenated water. However, biological oxygen demand would determine that the oxygenated state did not persist.

The predominance of organic matter means that it is the principal source of H+ for acid–base reactions. The deﬁnitive distinction between ombrotrophic and minerotrophic peatlands is the addition of extraneous materials only via the atmosphere in the former case, and additionally (and more copiously) via ground and surface waters in the latter case. Ombrotrophic mires The geochemistry of the ombrotrophic environment is the simpler of the two types. Living and decaying vegetation (Sphagnum, Carex, and other genera) dominates the environment and pH is determined basically by the dissociation of carboxylic functional groups in the dead organic matter. Since the rate of addition of terrigenous minerals and the rate at which many of them dissolve in the acidic bog waters are too

186

W. Chesworth, A. Martı´nez Cortizas, E. Garcı´a-Rodeja

Figure 8.8. Several sources and sinks of electrons in the land surface. At a microscale in a bog, mire, or soil landscape, the movement of electrons is more complicated than the big picture shown in Figure 8.7. Many more sources (for example, ferrous iron, sulﬁdes, native metals ) and sinks (for example, manganous oxides and hydroxides, and nitrates) exist, and will inﬂuence the detailed geochemistry of a natural system, but in a peatland the overall geochemistry will be dominated by the reactions between organic matter and oxygen as shown in Figure 8.7.

slow to consume the protons generated by CO2 and organic acids the pH can be expected to remain constant with depth as shown in Figure 8.9. Eh will vary independently, and in the catotelm may reach levels of reduction low enough to allow the generation of CH4. It is possible that oxygenated groundwater may enter the system below the water table, and rise up in a plume, but in general oxygen is unlikely to be the terminal electron acceptor within the zone of water saturation. The fate of the scarce wind-borne additions to ombrotrophic bogs have been discussed by Martı´ nez Cortizas et al. (2001a, b) in the context of the predominantly quartzo-feldpathic additions to the wetlands of Galicia, northwestern Spain (Fig. 8.10). The material is assumed to be added in amounts too small to have an effect on the overall pH of the environment.

The redox– pH approach to the geochemistry of the Earth’s land surface

187

Figure 8.9. The ombrotrophic environment. The ﬁeld of ombrotrophic peatlands will be approximately as shown in the gray, cigar-shaped area, with the relatively low pH determined by the dissociation of carboxyllic acid groups in organic matter. Generally, eolian inputs of inorganic dust will be too feeble to move the pH to higher values.

Minerotrophic mires The increased importance of extraneous additions to minerotrophic as opposed to ombrotrophic peatlands shows up in two ways. First, the addition of Fe-containing minerals (ferromagnesians, oxides or hydroxides) and their reaction with water tends to produce a redox proﬁle that follows a contour parallel to the boundary of the ferric-hydroxide ﬁeld of predominance. Second, hydrolysis of added minerals will raise pH, most dramatically if calcite or dolomite is in the mix. The boundary of the predominance ﬁeld of calcite will provide an upper limit in this case, at least as long as carbonates persist. Where only primary silicates are added (and the most common will be feldspar) a raised pH will be more slowly established and will not normally reach the calcite ﬁeld. The possibilities are shown in Figure 8.11. Discussion Mires, mainly ombrotrophic ones, are extensively used as archives of compounds deposited by wet and dry deposition from the atmosphere, heavy metals in particular.

188

W. Chesworth, A. Martı´nez Cortizas, E. Garcı´a-Rodeja

Figure 8.10. Reaction products of terrigenous dust in the ombrotrophic environment of Northwestern Spain, showing the local suite of input minerals, the PH 2 and PO2 ranges of the local blanket bogs and the resulting reaction products in six segments of the PH 2 PO2 framework, shown as mineral facies diagrams based on a triangular projection of the system Si–Al–Fe–O–H–S.

The basic idea underlying this kind of research is that dust and other compounds deposited on the mire will eventually be buried as plant remains accumulate. Any subsequent core may then be sectioned and analyzed to give a high-resolution record of changes in atmospheric deposition for up to a few thousand years. Elemental mobilities in the peat are a potential complicating factor, though most studies assume that elements of interest are immobile. Few papers actually address the geochemical conditions of the mire in terms of their effect on the stability of the mineral phases hosting an element (Shotyk, 1995; Steinmann and Shotyk, 1997) and thereby on the possibility of that element being released in mobile form. Four aspects are of particular interest in this regard: (1) the mineralogical composition of the deposited dust, (2) the geochemical environments of the mire (the acidic, oxidized acrotelm and the slightly less acidic but anoxic catotelm), (3) the rate of peat accretion, and (4) the interference of human activities on the natural composition of the atmospheric dust and aerosols. Materials deposited from the atmosphere onto the surface of the mire are incorporated initially into the acid/oxidized environment of the acrotelm. Most primary minerals are unstable in this environment, and at the extreme (pH below 4), most

The redox– pH approach to the geochemistry of the Earth’s land surface

189

Figure 8.11. The minerotrophic environment. The lower pH values or the cigar-shaped ﬁeld of the minerotrophic mines are deﬁned by ferrous–ferric equilibria that partially delimit the predominance ﬁeld of ferrihydrite, a precursor of the stable-phase goethite, shown in Figure 8.10, and formed as an initial product of the breakdown of ferromagnesians or magnetite. Depending on the amount and type of inorganic crustal material present in a mire, the upper pH limit of the minerotrophic environment will extend in the direction of the heavy arrows, towards the calcite-present area.

common minerals, primary or secondary, are unstable. Only if a given element is hosted in stable minerals may its concentration in the peat be assumed to reﬂect that of the original dust (though correction for the effect of the organic substrate may be necessary). Problems arise when the inorganic phase breaks down and an element is released as an ion, into solution. Once in solution it may leach (essentially through lateral movement) or diffuse through the mire water column following concentration gradients. Figure 8.12 exempliﬁes this by showing the proﬁle of Fe concentrations in an ombrotrophic mire. The highest Fe concentrations occur in the aerated acrotelm of the upper 25 cm of the bog; below this, concentrations abruptly decrease to reach minimum values where the ﬂuctuation of the water table takes place. Below this level (and particularly below 110 cm in the example shown) values increase again. A similar pattern was obtained in an experiment where cores from uncontaminated sites were transplanted to contaminated sites, and vice versa. Iron concentrations in the peat core of the contaminated mire decreased after being transplanted to the uncontaminated site, whereas the opposite was found for the peat core of the uncontaminated mire after it was transplanted to the contaminated one (Martin Novak, personal communication). This indicates that Fe was mobilized in or out of the

190

W. Chesworth, A. Martı´nez Cortizas, E. Garcı´a-Rodeja

Figure 8.12. Proﬁle of Fe (ppm) concentrations in an ombrotrophic bog (Pena da Cadela, Xistral Mountains, NW Spain). The vertical scale is in centimeters, the horizontal scale of the Fe proﬁle is in parts per million. Such a proﬁle might be interpreted as being due to a greater inﬂux of Fe in more recent dusts than in older ones (represented by deeper parts of the bog). However, in the upper 50 cm in which Fe varies 2.3-fold, normalization against a known immobile element (Ti) shows a variation of 3.2-fold. The fact that the two ranges are different is prima facie evidence that Fe is mobile. Extrapolating to the total proﬁle suggests the possibility that Fe has also been mobilized at depth, a possibility supported by the transplant experiments described in the text.

transplanted cores, following concentration gradients. Using a chemical speciation approach, Shotyk (1995) also suggested that there is no obvious chemical mechanism to retard the possible migration of As and Sb. However in a recent investigation of a peat core from a blanket bog from the Faroe Islands it was found that Sb and Pb

The redox– pH approach to the geochemistry of the Earth’s land surface

191

records were highly correlated, and this correlation was taken as evidence of effective Sb immobilization (W. Shotyk, personal communication). Other authors have also indicated that changes in redox potential, dependent on water table ﬂuctuations, are the most important mechanisms in the distribution of heavy metals (Damman, 1978; Clymo and Hayward, 1982; Damman et al., 1992). Steinmann and Shotyk (1997), on the other hand, found relatively high Fe3+ to Fe2+ ratios in bog-pore waters, despite the anoxic conditions, which they explained in terms of the higher thermodynamic stability of the organic complexes of Fe3+ compared to those of Fe2+. Unlike the case of a mineral soil, in ombrotrophic mires released elements are unlikely to ﬁx by complexation to another inorganic phase, since no more than 2% of the total dry mass of the peat is inorganic matter. However, released elements may obviously become bound to organic matter, for example, Hg shows a marked afﬁnity for humiﬁed organic matter (Benoit et al., 1994), with binding to reduced sulfur groups being preferred (Skyllberg et al., 2003). Complexation is the most important mechanism for metal ﬁxation, as Shotyk (1995) has shown. He calculated that in bogs 99.9% of Cu, Pb, and Zn can be bound to an organic component, either in the solid phase or in solution (as dissolved organic carbon, DOC). So, organically bound elements can be retained by the peat or they can be leached from the system with the dissolved organic matter. In other words, the extent to which the original concentration of a given element will be changed subsequent to the addition of that element to a mire will depend on the degree to which the element is partitioned between the immobile solid phase and the mobile dissolved phase. This in turn will depend upon the kinetics of the reactions that produce the soluble chemical species. Finally, the hydrodynamics of the system plays an important role. If the mobile phase is deﬁnitively trapped within the mire, no compositional change will take place. If, as seems most likely, the mobile phase ﬁnds an exit route, then elements complexed to DOC will leave the system. In minerogenic mires the higher mineral content makes the role of the inorganic compounds more relevant in terms of the redistribution of metals. For example, Franze´n et al. (2004) analyzed a minerogenic mire and found enrichments in Hg associated to elevated Fe concentrations and the presence of goethite, suggesting Hg adsorption to Fe-oxyhydroxides probably as Hg-fulvo-acid complexes. One important aspect of the fate of organically bound elements is that their potential release is coupled to that of dissolved organic matter (DOM) and linked to climate changes. As indicated by Biester et al. (2006 – this book, Ch. 19) the release of organo-halogens is expected to increase during warm and wet periods and decrease during dry periods; this may also affect other organically bound elements (that is, metal-DOM complexes). Continuous growth of vegetation in the surface of the mire tends to keep the acrotelm at an almost constant thickness (Belyea and Clymo, 2001). By contrast, the catotelm grows upwards. This means that any given peat surface will reach the catotelm in a few hundreds to a thousand years (Martı´ nez Cortizas et al., 2001a, b). Thus minerals/elements progressively sink into a more anoxic environment as time goes by. Vertical water movement is largely impeded in the catotelm (Ingram, 1983)

192

W. Chesworth, A. Martı´nez Cortizas, E. Garcı´a-Rodeja

and the fate of an element in this section will mostly be governed by redox reactions and diffusion. Human activities can affect the amount and composition of the deposited dust in mires. Induced soil erosion due to forest clearance, ﬁres, and agriculture have increased the amount of atmospheric dust and changed its composition (by increasing the surface area and types of exposed sources: soils and rocks) especially since the so called Neolithic Revolution; that is, from 10,000 to 5,000 years ago (Martı´ nez Cortizas et al., 2005). The most prominent changes in atmospheric composition occurred once the smelting of metals began. In the last 250 years, the changes have accelerated as a consequence of the Agricultural and Industrial revolution. Metallurgy, mining, fuel burning, waste incineration, and other human activities have released to the atmosphere compounds that are markedly different from those originating from soils and rocks. As with dusts of natural origin, the anthropogenic material is also subject to possible modiﬁcation in the mire-environment. For example, Sonke et al. (2002) investigated metal pollution in a bog close to a Zn-smelter in the Kempen region (Belgium) and found that though oxidized smelter dusts were found in the nearby mineral soils, the bog showed extensive in situ FeS2 and ZnS, indicating that anoxic conditions were responsible for rapid diagenetic transformations of anthropogenic metal oxides into sulﬁdes. Rausch et al. (2005a, b) investigated solid-phase and pore-water concentrations of some trace metals in uncontaminated and contaminated sites in Finland and found that Cu was generally retained more effectively than Ni, Co, Zn, or Cd, but in highly contaminated sites even Cu was substantially transferred from the solid phase to the pore waters. These authors indicate that ombrotrophic bogs may or may not function as reliable archives of atmospheric trace metals, depending on the concentration and chemical properties of the elements considered, the mineralogical form at the time of deposition, as well as the pH of the bog waters. As a general rule, they hypothesize that sites that are remote from point-pollution sources and receive metals exclusively by atmospheric deposition are more conducive to preserving the chronology of deposition. Not only element concentrations can be affected by ambient geochemical conditions, isotope ratios may also be modiﬁed. The bulk isotopic composition of peat arises from a mixture of sources (for example, soil dust, anthropogenic pollutants), which can have different isotope ratios. This characteristic is used, for instance, in tracing the source of Pb pollution, since the isotopic signature of Pb from an anthropogenic source such as gasoline, differs from that of Pb in soil dust. However, there may be considerable heterogeneity in the Pb from different soil pools (Emmanuel and Erel, 2002; Wong and Li, 2004). Consequently the materials added to a mire, whether they are from natural or from anthropogenic sources, and whether they are from ﬁner or coarser fractions, from the primary mineral or secondary mineral, or other fractions of the soil, will show a range of geochemical stabilities in the peat environment. This will complicate attempts to decipher the geochemical history of a peatland. In addition, if the hydrodynamics of a given system is conducive to the leaching of the more labile weathering products, post-depositional changes of the original composition, including its bulk isotope ratio become possible, and use of the peat as an archive is compromised. The danger is that these changes may be interpreted as changes in Pb atmospheric composition through time.

The redox– pH approach to the geochemistry of the Earth’s land surface

193

Conclusions Redox–pH diagrams provide a concise graphical way in which to depict Earthsurface environments in terms of the activities of electrons, the operators involved in the two commonest types of reaction in the weathering zone on land, i.e., redox and acid–base reactions. Figure 8.13 illustrates the following points by way of summary. Peatlands tend to fall on the acid side of the diagrams with ombrotrophic bogs showing least overlap with common soil-forming environments. Minerotrophic mires overlap the ﬁeld of ombrotrophic bogs under relatively high redox conditions, though with depth they move along a line deﬁned by ferrous–ferric equilibria, to a near-neutral pH. The same kind of equilibria also plays a controlling inﬂuence on ambient conditions in mineral soils at the low limit of their pH range. The upper limit of the minerotrophic ﬁeld in peatlands, however, will depend on whether or not there are extraneous additions of carbonates. If there are ambient conditions, the peatland may extend to the calcite predominance ﬁeld, where a pH approaching 8 is possible. Within this extended ﬁeld lies the environment of histosol formation. Histosols may be looked upon as a kind of bridge between peatlands and mineral soils.

Figure 8.13. The geochemical environments of bogs and mires shown in the context of the Eh–pH range of common soils and of acid-sulfate soils.

194

W. Chesworth, A. Martı´nez Cortizas, E. Garcı´a-Rodeja

In the context of this volume, the most important conclusion is that the peatland environment covers a wide range of geochemical conditions and encompasses those under which most primary, and many secondary minerals would break down. If in fact inorganic additions to a peatland do break down, the possibility of the mobilization of the elemental components must be entertained. Consequently redox–pH diagrams offer a ﬁrst line of defense against the hasty use of any given proxy element assumed to be immobile, in the reconstruction of past environmental conditions.

References Belyea, L. and Climo, R.S., 2001. Feedback control on the rate of peat formation. Proc. Roy. Soc. Lon. 268, 1315–1321. Benoit, J.M., Fitzgerald, W.F., and Damman, A.W.H., 1994. Historical atmospheric mercury deposition in the mid-continental U.S. as recorded in an ombrogenic peat bog. In: Huckabee, J. and Watras, C. (Eds), Mercury Pollution: Integration and Synthesis. Lewis Publishers, Chelsea, MI, pp. 187–202. Berner, R.A., 1970. Behaviour of iron during weathering, sedimentation and diagenesis. In: Wedepohl, K.H. (Ed.), Handbook of Geochemistry, Section 26G. Springer, Berlin, pp. 1–8. Biester, H., Martı´ nez Cortizas, A., and Keppler, F., 2006. Occurrence and fate of halogens in mires. In: Martini, I.P., Matı´ nez Cortizas, A., and Chesworth, W. (Eds), Peatlands: Evolution and Records of Environmental and Climatic Changes, Ch. 19. Elsevier, Amsterdam. Brookins, D.G., 1988. Eh–pH Diagrams for Geochemistry. Springer, Berlin, 176pp. Chesworth, W., 1992. Weathering systems. In: Martini, I.P. and Chesworth, W. (Eds), Weathering, Soils and Paleosols. Elsevier, New York, pp. 19–40. Clymo, R.S. and Hayward, P.M., 1982. The ecology of Sphagnum. In: Smith, A.J.E. (Ed.), Bryophyte Ecology. Chapman & Hall, London, pp. 229–289. Damman, A.W.H., 1978. Hydrology, development, and biogeochemistry of ombrogeneous peat bogs with special reference to nutrient relocation in a western Newfoundland bog. Can. J. Bot. 64, 384–394. Damman, A.W.H., Tolonen, K., and Sallanus, T., 1992. Element retention and removal in the ombrotrophic peat of Hadekeidas, a boreal Finnish peat bog, Suo 43, 137–145. Edmond, J., Corliss, J.B., and Gordon, L.I., 1979. Ridge crest hydrothermal metamorphism in the Galapagos spreading center and reverse metamorphism. In: Talwani, M., Hay, W.W., Ryan, W.B.F. (Eds.), Deep Drilling Results in the Atlantic Ocean : Continental Margins and Paleoenvironment. Proceedings of the 2nd Maurice Ewing Symposium, American Geophysical Union, Washington, DC, pp. 383–390. Emmanuel, S. and Erel, Y., 2002. Implications from concentrations and isotopic data for Pb partitioning processes in soils. Geoch. Cosmochim. Acta 66, 2517–2527. Franze´n, C., Kilian, R., and Biester, H., 2004. Natural mercury enrichment in a minerogenic fenevaluation of sources and processes. J. Environ. Monit. 6, 466–472. Garrels, R.M. and Christ, C., 1965. Solutions, Minerals and Equilibria. Harper & Row, New York, 450pp. Ingram, H.A.P., 1983. Hydrology. In: Gore, A.J.P. (Ed.), Ecosystems of the World 4A: Moors: Swamp, Bog, Fen and Moor. Elsevier, Amsterdam, pp. 67–224. Krumbein, W.C. and Garrels, R.M., 1952. Origin and classiﬁcation of chemical elements in terms of pH and oxidation–reduction potentials. J. Geol. 60, 1–33. Martı´ nez Cortizas, A., Chesworth, W., and Garcı´ a-Rodeja, E., 2001b. Dina´mica geoquı´ mica de las turberas de Galicia. In: Martı´ nez Cortizas, A., Garcı´ a-Rodeja, E., (Eds.), Turberas de Montan˜a de Galicia, Coleccio´n Te´cnica de Medio Ambiente. Xunta de Galicia, pp. 141–148. Martı´ nez Cortizas, A., Garcı´ a-Rodeja, E., and Chesworth, W., 2001a. Geochemistry of Si–Al–Fe Minerals in Mire Environments of Northwestern Spain, ICOBTE (International Conference on the Biogeochemistry of Trace Elements), Proceedings, Guelph, Ontario, Canada. p. 491. Martı´ nez Cortizas, A., Mighall, T., Pontevedra-Pombal, X., et al., 2005. Linking changes in atmospheric dust deposition, vegetation change and human activities in northwest Spain during the last 5,300 years. The Holocene 15, 698–706.

The redox– pH approach to the geochemistry of the Earth’s land surface

195

Pourbaix, M., 1974. Atlas of electrochemical equilibria in aqueous solutions. Natatinal Association. Corrosion Engineering, Houston, 644pp. Rausch, N., Nieminen, T., and Ukomaanaho, L., 2005a. Comparison of atmospheric deposition of copper, nickel, cobalt, zinc, and cadmium recorded by Finnish peat cores with monitoring data and emission records. Environ. Sci. Technol. 39, 5989–5998. Rausch, N., Ukomaanaho, L., and Nieminen, T., 2005b. Porewater evidence of metal (Cu, Ni, Co, Zn, Cd) mobilization in an acidic, ombrotrophic bog impacted by a smelter, Harjavalta, Finland and comparison with reference sites. Environ. Sci. Technol. 39, 8207–8213. Shotyk, W., 1988. Review of the inorganic geochemistry of peats and peatland waters. Earth-Sci. Rev. 25, 95–176. Shotyk, W., 1995. Natural and anthropogenic enrichments of As, Cu, Pb, Sb, and Zn in ombrotrophic versus minerotrophic peat bog proﬁles, Jura Mountains, Switzerland. Water Air Soil Pollut. 90, 375–405. Skyllberg, U., Qian, J., Frech, W., et al., 2003. Distribution of mercury, methyl mercury and organic sulphur species in soil, soil solution and stream of a boreal forest catchment. Biogeochemistry 64, 53–76. Sonke, J.E., Hoogewerff, J.A., van der Lanna, S., and Vangronsveld, J., 2002. A chemical and mineralogical reconstruction of Zn-smelter emissions in the Kempen region (Belgium), based on organic pool sediment cores. Sci. Tot. Environ. 292, 101–119. Steinmann, P. and Shotyk, W., 1997. Chemical composition, pH, and redox state of sulfur and iron in complete vertical porewater proﬁles from two sphagnum peat bogs, Jura Mountains, Switzerland. Geoch. Cosmoch. Acta. 61, 1143–1163. Truesdell, A.H., 1968. The advantage of using pe rather than Eh in redox equilibrium calculations. J. Geol. Educ. 16, 17–20. Weaver, C.E. and Pollard, L.D., 1973. The Chemistry of Clay Minerals. Elsevier, New York, 213pp. Wong, C.S.C. and Li, WX.D., 2004. Pb contamination and isotopic composition of urban soils in Hong Kong. Sci. Total Environ. 319, 185–195. Woods, T.L. and Garrels, R.M., 1987. Thermodynamic Values at Low Temperature for Natural Inorganic Materials. Oxford University Press, New York, 266 pp.

This page intentionally left blank

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

197

Chapter 9

Weathering of inorganic matter in bogs G. Le Roux and W. Shotyk

Introduction Because of their unique geochemical properties, weathering associated with wetlands has been the subject of scientiﬁc investigations since at least the 18th century (Bischoff, 1854; Hunt, 1875; Endell, 1911; Humphreys and Julien, 1911; Blanck and Rieser, 1925; Blanck and Keese, 1928; Shotyk, 1992). Since these pioneering works where reductive conditions, abundance of organic acids and low pH have been cited to explain intense weathering of various classes of minerals (Endell, 1911; Humphrey and Julien, 1911), few studies have been made using modern geochemical tools to study the rates and mechanisms of weathering by peatlands. There is a real need to bridge this gap because peatlands may play a globally signiﬁcant role in the weathering cycle (Shotyk, 1992). (1) Wetlands presently cover more than 5% of the Earth’s total land area and their role in large scale weathering of the upper continental crust is poorly understood. (2) The unique geochemical properties of peatlands, namely an anaerobic water column, abundance of natural organic matter and low pH, make them an ideal milieu to study the action of organic acids on weathering of minerals. Moreover the possibility of age dating the peat column where weathering takes place may permit the kinetics of mineral dissolution to be quantiﬁed if the extent of weathering of minerals prior to their entry in the bog can be estimated. Study of weathering at a larger scale is of importance to understand soil and sediment formation as well as the chemical composition of surface waters. In addition to representing a signiﬁcant carbon pool, peatlands could have a large impact on the global weathering cycle and therefore also on the carbon cycle. This effect should not be underestimated because boreal peatlands cover large areas of Canada and Russia, which are situated mainly on previously glaciated, crystalline rocks (Canadian and Siberian cratons). Wetlands are known to function as source regions for metals, which are more soluble under anoxic conditions (Fe, Mn), but the ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09009-2

198

G. Le Roux, W. Shotyk

details of reductive dissolution remain obscure. On the other hand, for elements with reduced solubility under anoxic conditions (Cu, U, Hg, As) and for elements, which form sulphide minerals (Ni, Cu, Zn, As, Cd, Sb, Hg, Pb), wetlands are effective geochemical traps, removing these elements from surface waters (Shotyk, 1989; Shotyk et al., 1992). Because modern peatlands serve as analogues for coal formation (Martini and Glooschenko, 1985; Cross and Phillips, 1990), they could help to provide us with a better understanding of mineral matter and trace elements in coal. (3) Finally, reactions and processes in the peat column also inﬂuence the ability of bogs to preserve records of atmospheric dust, volcanic emissions and anthropogenic deposition. Our approach in this chapter is to focus on weathering reactions and processes in raised, ombrotrophic bogs. Our goals are to ﬁrst brieﬂy describe the physical and chemical characteristics of bogs, then summarize the predominant sources of mineral matter. The most common minerals occurring in bogs are identiﬁed and the main mechanisms of weathering discussed.

Characterization of the weathering milieu In this part, our aim is to illustrate the main characteristics of the bog, which could have an inﬂuence on the fate of the mineral matter in the bog. Botanical composition and climate The ﬁrst modern scientiﬁc studies of ombrotrophic bogs were undertaken in Europe, especially Germany, Switzerland, Sweden, Denmark and Scotland (Shotyk, 1988). As a consequence, our knowledge of the botanical composition of ombrotrophic bogs is largely biased toward European mires where peat formation is dominated by Sphagnum mosses. However, except for the arid and semi-arid regions, ombrotrophic bogs are found worldwide (Von Bu¨low, 1929). For example, in the tropics there are extensive formations of raised, ombrotrophic peat, which consist mainly of partly decomposed remains of deciduous trees (Page et al., 1999; Wu¨st et al., 2002). In the Falkland Islands, raised bogs are dominated by Poa flabellata, a tussock-forming grass (Lewis Smith and Clymo, 1984). In New Zealand, large, deep ombrotrophic bogs are dominated by Sporodanthus, a giant rush (Shearer, 1997). Extensive blanket bogs are found on the subantarctic islands such as Campbell Island, 600 km south of New Zealand (McGlone et al., 1997). Even in the High Arctic, in the Carey Islands of Greenland, there are peat hummocks formed by Aplodon wormskioldii, a moss whose growth is promoted by the constant supply of seabird excrement. As far as we know, studies of the retention of minerals and elements have focused on Sphagnum mosses (Malmer, 1988; Punning and Alliksaar, 1997) with very little known about retention of elements by the vascular plants forming bogs in the

Weathering of inorganic matter in bogs

199

Southern Hemisphere. In Sphagnum-dominated bogs, which are the dominant form of bogs especially in the Northern Hemisphere, the surface layers are often enriched in elements such as K, P or Zn due to the biological uptake especially in the capitulum of the Sphagnum mosses (Malmer, 1988; Shotyk, 1996). The climatic regimes of ombrotrophic peatlands range from warm and wet in the tropics, through cool and humid in the temperate maritime regions of the northern hemisphere, to below freezing in the High Arctic of Canada. These extremes of temperature and humidity play an overwhelming role in governing the botanical composition of the peat, the microorganisms, which are found there (Go¨ttlich, 1990; Dierssen and Dierssen, 2001), and the rates of all chemical reactions.

Bog water chemistry Bogs are fed only by rainwater. Therefore, concentrations of dissolved solids in the surface waters are very low, with concentrations varying from 10 to 100 mg L1 (Gorham et al., 1984; Go¨ttlich, 1990; Bennett et al., 1991). Consequently, the electrical conductivity is on the order 10–30 mS cm1 (Malmer, 1986; Go¨ttlich, 1990). A low pH (3.7–4.2) is characteristic of the surface layers of ombrotrophic bogs (Shotyk, 1989). Because of the low alkalinity of rainwater, the abundance of dissolved organic acids and the pKa of their carboxyl groups control the pH (Shotyk, 1989). Depending on the type of underlying rock and the historical development of the bog, pH generally increases with depth (Fig. 9.1). Whereas pools of water at the bog surface may be oxygen-saturated, the major part of water is below the surface and anoxic (Shotyk, 1989; Go¨ttlich, 1990; Shotyk, 1992). The range of Eh values in measured bog water is between 0.5 and 0.5 V. Martinez Cortizas et al. (2001) emphasized the distinction between the surface peat layer and the deepest layer and its importance for the behavior of minerals in peat. The surface layer (acrotelm) is an active oxic layer, where plant decay occurs rapidly whereas the deepest layer (catotelm) is an anoxic layer, characterized also by a low hydraulic conductivity. Particles might reside in the oxic acrotelm for decades to centuries, depending on the peat accumulation rate and climate. During water table drawdown, the pore spaces in the acrotelm may become unsaturated, allowing the movement of gases and particles.

Geological substrate The geological substrates, which function as templates of peat formation, are diverse (Fig. 9.1). Peatlands are generally found on impermeable layers such as clay or ﬁnegrained glacial tills, which impede the drainage of surface water. The distribution of peatlands in the Franches Montagnes of the Swiss Jura illustrates the degree to which peat formation depends on the availability of water-saturated substrates: in this highly fractured, well drained, karstic terrain, peat formation is typically found on outcrops of Oxfordian and Tertiary clays that became exposed at the Earth’s surface. Whereas peat might be most commonly found on impermeable substrates, peat also

200

G. Le Roux, W. Shotyk

Figure 9.1. pH proﬁles and stratigraphy of bogs on contrasting geological substrates: E´tang de la Grue`re in Switzerland (EGR) and Kolhu¨ttenmoor in Germany (BF), the former developed upon calcareous clay and the latter on clays derived from a granite. Note that the difference in geology and geochemistry had a profound inﬂuence on the botanical composition of the peat.

forms on highly permeable substrates such as coarse sands, in areas of groundwater discharge (Anderson, 1797). Distribution and supply of inorganic compounds in bog profile The abundance of inorganic material in peat is usually quantiﬁed by the ash content, which is the dry weight of material left after overnight combustion of the peat, at 5501C (Givelet et al., 2004). The ash content in ombrotrophic peats is typically on the order of 1–3% by weight. Lower ash contents have been reported in the tropical ombrotrophic peatlands of Kalimantan, Indonesia; here, many samples contain as

Weathering of inorganic matter in bogs

201

little as 0.5% ash (Weiss et al., 2002) but it is not yet clear whether this reﬂects lower rates of atmospheric deposition of mineral matter, higher rates of peat accumulation, accelerated rates of mineral weathering or elevated inputs of mineral phases that are more susceptible to chemical weathering. Greater concentrations of mineral matter have been reported at discrete depths in many ombrotrophic bogs, reﬂecting either changes in peat accumulation rate (Steinmann and Shotyk, 1997b), variations in the rate of deposition of atmospheric soil dust (Shotyk, 2001), or episodic inputs of volcanic-ash particles (Dugmore and Newton, 1992; Wastegard et al., 2003). Elevated concentrations of mineral matter are also found in the surface layers of all ombrotrophic bogs, as well as in the deepest peat layers, giving rise to C-shaped concentration proﬁles. At the bog surface, the elevated concentrations of mineral matter are often thought to be due to biological uptake and recycling of essential elements (especially K, Mg, Ca, but also P, S and perhaps Si) by the living plant layer. Sphagnum mosses, for example, are enriched in K, N and P in their upper part (Malmer, 1988). However, various human activities have contributed to elevated ﬂuxes of soil dust, namely deforestation, the increasing intensity of modern agricultural practices (Gorham and Tilton, 1978), as well as construction. Concentrations of refractory trace metals, with very low solubilities and no known biological functions (Sc, Ti, Zr), are also more abundant in the surface layers of ombrotrophic bogs, suggesting that human impacts on the dust ﬂuxes are leaving their mark in the bog archives (Gorham and Tilton, 1978; Shotyk, 2001).

Preparation of peat samples for mineral identification In ombrotrophic peat, the ash content is less than 5%; therefore it is difﬁcult to observe minerals in situ. The challenges posed for mineral identiﬁcation by the paucity of mineral particles in peat has been succinctly described by Raymond et al. (1983, p. 170): ‘‘Another problem of major consequence in studying mineral matter in peat is that some inorganic fractions may be so scarce that they will not be observable either in thin section or in the X-ray diffraction pattern of a low or high temperature. However, the peat contains some non-biogenic mineral particles of both siliceous and non-siliceous compositions that can be separated from the organic peat material for SEM analysis by ashing and acid bath techniques. Though the speciﬁc relationships between minerals and botanical constituents are lost using this technique, very minor mineralogical constituents that would otherwise be ignored are observable.’’ The most commonly used technique is a combination of high-temperature ashing and washing with diluted HCl (Raymond et al., 1983; Steinmann and Shotyk, 1997b). However one problem is the risk of change to the mineralogical composition due to processing. For example, expandable clays can be altered during ashing. Another possibility is low-temperature ashing (Ward, 2002) or wet-oxidation using oxidizing agents such as H2O2 to destroy the organic matter (Finney and Farnham, 1968; Givelet et al., 2004). Again the possible importance of mineralogical artifacts cannot be ignored and the speciﬁc relationships between minerals and botanical constituents are lost (Raymond et al., 1983).

202

G. Le Roux, W. Shotyk

Another technique discussed by Raymond et al. (1983) is impregnating the peat using a resin to slice it sufﬁciently thin to allow the optical microscope to be used. This approach is helpful to understand the positioning of minerals in the mosses, vascular plants and peat derived from them. This technique was used successfully for swamp peat containing abundant mineral matter, but for low ash peat many challenges remain. It should be noticed that similar problems are encountered for mineral observation in coal (Ward, 2002). Inorganic constituents supplied by atmospheric deposition Mineral matter in atmospheric precipitation originates from different sources (Fig. 9.2, Table 9.1; Mattson and Koutler Andersson, 1954): pedogenic resulting from soil erosion, oceanogenic from salt sprays, pyrogenic from smoke and ash supplied by ﬁres, volcanogenic particles from volcanic emanations, cosmogenic from meteorites and cosmic dust, anthropogenic from industrial, vehicle emissions and others. (1) Soil erosion is the major source of atmospheric particles to most bogs. Particles from local soils, but also from more remote areas, are constantly being deposited in the bog. Intense storms could considerably increase the mineral content of the peat by either local (Shotyk, 1997) or long-range (Shotyk et al., 1998) transport of particles. Most dust particles have a radius between 0.1 and 100 mm (Schu¨tz, 1989).

Figure 9.2. Supply of inorganic material to bog and shape of Ca distribution in pore water of the Kohlhu¨ttenmoor, southern Germany illustrating the two main reaction fronts: atmosphere – peat surface and peat – substrate. The two main reaction fronts are illustrated in the insert by an increase of the concentration of Ca in the pore waters.

Weathering of inorganic matter in bogs Table 9.1. aerosols.

203

Estimated ranges of yearly input ﬂuxes of particles that make up atmospheric

Aerosol

Natural (N) or anthropogenic (A)

Annual ﬂux 1012 g yr1

Sea spray (oceanogenic) Dust (pedogenic) Forest ﬁres (pyrogenic) Volcanic emissions (plutogenic) Meteors (cosmogenic) Anthropogenic combustion Condensation

N N, A N, A N N A N, A

1000–10,000 60–750 35–1500 6.5–150 1 50 1500

Source: Data from VanLoon and Duffy (2000) and Nriagu (1979, 1989).

(2)

(3)

(4)

(5)

Distances from the source of dust and wind speed are the dominant factors affecting the average particle size. The greater the transport distance and the lower the wind speed, the smaller is the average radius of the particle (Schu¨tz, 1989). The elemental composition of soil dust reﬂects the soil it is derived from (Schu¨tz, 1989). However there is some elemental and moreover mineral fractionation during erosion and wind transport (Schu¨tz, 1989; Martı´ nez Cortizas et al., 2002). Because bogs are situated in zones of humid climate, where a cover of native vegetation stabilizes the soil, wind erosion of the soil is reduced to a minimum (Mattson and Koutler Andersson, 1954). Therefore a large contribution of dust to the bog could be expected from more remote places such as deserts. Sahara dust, for example, is mainly composed of quartz, clays, carbonates and feldspars and is a signiﬁcant source of mineral dust to southern European bogs. As an example, Bu¨cher and Lucas (1984) studied the mineralogy of dust coming from Sahara and to the Pyrenees. They found that the main part of the dust has a radius between 2 and 30 mm and that the main minerals are quartz (50%), clays (25%) mainly kaolinite, carbonates (15%) and feldspars (o5%). Sea salts due to chemical conditions are not retained in bogs. Shotyk (1997) showed that less than 10% of Na, Mg, Ca and Sr supplied principally by sea salt inputs in oceanic bogs are preserved. Fires concentrate inorganic compounds from the soil and vegetation cover around the bog and also increase soil erosion. Higher ash content due to ﬁres is often indicated by the presence of pieces of charcoal in the peat (Ho¨lzer and Ho¨lzer, 1998). Some volcanic eruptions produce large amounts of minerals and glasses dispersed more or less far away from the crater. When layers of this type of material called tephra are found in bogs they can be useful in establishing a regional or hemispheric stratigraphy (tephrostratigraphy) and can function as absolute age markers. The material deposited varies according to the type of volcanoes, the time of eruption and the distance to the bog (Hodder et al., 1991). Silicate materials emitted by volcanic eruptions are mainly feldspars, glasses and ferromagnesian silicates. Cosmogenic material is rarely found in bogs but some traces of meteorite impacts have been found in bogs of Russia (Kolemikov and Shestakov, 1979) and Estonia (Veski et al., 2001).

204

G. Le Roux, W. Shotyk

(6) Material from anthropogenic sources is abundant in the surface peat layers. When the emissions are severe, for example in the vicinity of smelters, bog vegetation may be severely damaged (Glooschenko, 1986). Different particles of anthropogenic origin containing metals and sulphur are deposited due to fossil fuel burning, especially coal, and emissions from heavy industries (Fig. 9.3; Glooschenko et al., 1986; Nova´k et al., 2003).

Anthropogenic components such as spherical carbonaceous particles are very resistant to acids and are trapped by the Sphagnum mosses at the surface of their leaves (Punning and Alliksaar, 1997). In a 341-day experiment, ﬂy-ash particles (80% o5 mm diameter) were mostly trapped within the upper 3 cm of the peat column. The mechanism of the trapping appeared to be a sorption mechanism on the surface of the plants or localization of the particles within the pores of hyaline cells. The effective entrapment of atmospheric particles suggested that peat cores could be used as archives of atmospheric deposition. Magnetite spherules derived from coal burning have been found in bogs from urban areas in England (Williams, 1992; Franze´n, 2006 – this book, Ch. 11). Other exotic anthropogenic compounds occurring in peat are derived from smelting, fuel combustion and other human activities. Plants, invertebrates and microorganisms living on, in and around the bog can produce biogenic silicates in the form of phytoliths or diatom frustules (Andrejko et al., 1983). In the ombrotrophic part of a tropical peat deposit, Wu¨st et al. (2002) found that the main part (up to 475%) of ash content is made of biogenic inorganic material.

Figure 9.3. Anthropogenic compound found in a heavily polluted bog near Manchester in England from fossil fuel burning (ﬂy ash). The compound is mainly composed of Ca, Fe and S.

Weathering of inorganic matter in bogs

205

Mechanisms and rates of weathering To investigate the possible dissolution of minerals, pH and pE are fundamental parameters in the peat column, which characterize the stability of minerals (Garrels and Christ, 1965; Martı´ nez Cortizas et al., 2001). Mineral dissolution in bog waters, however, is far more complicated than in simple experimental lab systems.

Reductive dissolution Reductive dissolution of iron-bearing minerals is one of the most important weathering processes in anaerobic terrains. First, Fe is the second most abundant metal in the Earth’s upper crust (3.5%) (McLennan, 2001) after Al (8%). Second, the solubility of Fe is strongly dependant on its redox state (Fe II more soluble than Fe III) and breaking down orthosilicate and inosilicate minerals such as olivine and pyroxenes is achieved simply by attacking the relatively weak Fe–O bonds. In Table 9.2 is an example of a pioneering study by Endell (1911), which shows the differences in the chemical composition of weathered basalt underneath a bog (anoxic conditions) versus the basalt outside the peatland. In an oxidizing soil environment (Chesworth and Macias Vazquez, 1985), the mobility of Fe released by weathering is severely restricted by the insolubility of the hydroxide Fe(OH)3, which has a Ksp of approximately 1038.7. In contrast, the solubility of Fe(OH)2 is 1014.5, which means that in an anaerobic and acid environment, the dissolved Fe concentration is higher and the mobility of Fe released by weathering is much less restricted (Fig. 9.4a). For example, consider the anoxic conditions in pore waters at the bottom of the E´tang de la Grue`re (EGR) proﬁle: at equilibrium with FeCO3 (siderite) at pH 6.2, the waters contain 100 mM Fe (Fig. 9.4a). As shown in Table 9.2.

Chemical composition of weathered basalt under and near a peatland.

Oxide (%)

Weathered basalt near but outside the peatland

Weathered basalt underneath 125 cm of peat

SiO2 Al2O3 TiO2 FeO Fe2O3 CaO MgO K2O Na2O SO2 P2O5 LOI Total

45.21 7.82 1.69 8.08 3.41 12.31 8.43 2.94 6.64 0.56 0.52 1.82 99.43

59.98 11.50 1.92 1.86 2.42 2.80 0.75 1.48 1.18 0.21 trace 16.53 100.63

Source: Data from Endell (1911).

G. Le Roux, W. Shotyk

206

Figure 9.4. Eh–pH diagram for iron in bog environment. Iron activity is aFe3þ ¼ 106 mol L1 in both diagram. (a). Eh/pH diagram for Fe–C–O–H system. (b). Eh/pH diagram for Fe mica (annite) assuming that in the dissolution all the Al is used to form kaolinite (aKþ ¼ 10 mol L1 , aH4 SiO4 ¼ 102:7 mol L1 ) (after Hodder, 1991).

Figure 9.4b (Hodder et al., 1991), most of the peatland waters plot in Eh/pH plots far from the stability ﬁeld of ferromagnesian minerals. Hodder et al. (1991) also suggest that anoxic conditions present in the bog prevent the formation of Fe-hydroxide, which would have protected against further weathering. Anthropogenic magnetite is also subject to reductive dissolution, which might be further enhanced by dissolved sulphide (Williams, 1992). Indeed above the water table, ferromagnetic spherules are not weathered, whereas at a depth of 20 cm in anoxic conditions, they show a strong alteration and absence of oxide coating.

Proton-promoted dissolution The low pH of bog water promotes the dissolution of silicate minerals except for quartz whose solubility is independent of pH below 9. Feldspars are thermodynamically unstable in bogs, because of the low pH and paucity of dissolved major element cations. Figure 9.5 shows, for example, the instability of K-feldspar weathered by the incongruent reaction þ

þ

0

2KAlSi3 O8 þ 2 H þ 9H2 O 2 Al2 Si2 O5 ðOHÞ4 þ 2K þ 4H4 SiO4 Experimentally, at pH o6, the dissolution rate R of K-feldspars and albite is pHdependant (Rp[H+]0.5) (Blum, 1994).

Weathering of inorganic matter in bogs

207

Figure 9.5. K-Feldspar stability diagram (log aH4 SiO4 vs. log (aKþ =aHþ )) with ellipse showing normal concentrations in pores water of bog. The diagram indicates that all mineral phases shown, as well as amorphous silicate, except for gibbsite, are thermodynamically unstable in peat.

As shown in Figure 9.5, kaolinite also is not thermodynamically stable in peat waters. Feldspars, kaolinite and other clay minerals therefore, should all be subjected to proton-promoted dissolution reactions (Cama et al., 2002). Below a silica activity of 104.7 the end product of the incongruent weathering of K-feldspar is gibbsite. Gibbsite weathering is also promoted with increasing H+ AlðOHÞ3 þ 3Hþ 2 Al3þ þ 3H2 O; the solubility of this reaction in pure water at 25 1C and pH 4 being 104.26 mol L1, which corresponds to a pore water concentration of ca. 1500 mg L1 Al. Typical values of aluminum in bog waters are around 50 and 250 mg L1: peat waters are therefore under-saturated with respect to gibbsite. Organic acids-promoted dissolution The role of organic acids in mineral weathering processes has been reviewed elsewhere (Bennett and Casey, 1994; Drever and Vance, 1994). The reductive dissolution of Fe-bearing minerals is well known because of the elevated concentrations of Fe in anoxic peatland waters (Johnson, 1866; Hunt, 1875; Ramann, 1895; Endell, 1911; Humphreys and Julien, 1911; Niklas, 1912). Except for the effects on Fe-containing minerals (Stumm, 1992), however, the effects of organic ligands on the rates and mechanisms of aluminosilicate weathering are imperfectly understood. In peatland waters there is a wide range of organic ligands capable of sequestering Al. These ligands range in chemical complexity from simple, small molecular weight

208

G. Le Roux, W. Shotyk

organic acids such as oxalic acid (molecular weight: 90) to large molecular weight, polymeric, polyelectrolytic humic acids (MW 10,000 or more). These ligands could affect weathering in several ways: (1) they lower the pH as they dissociate: H2A 22H++A; (2) they form complexes with metals, especially Al, which not only affects the solubility of the metal, but also by decreasing the concentration of free metal ions, which further promotes mineral dissolution: Men++m L2 Me–Lmnm (aq); (3) they form surface complexes, which may accelerate the detachment of ions from the crystal, depending on the molecule size of the ligand; and (4) they form surface coatings, which may protect the surface against further weathering. Bennett et al. (1991) investigated the weathering of quartz and aluminosilicates in a bog from Minnesota. They proposed the following model based on the pore water chemistry and mineralogical investigations of the peat: (1) in the ﬁrst meter, which is characterized by a low pH (4.6), only aluminosilicates are weathered, apparently by dissolution reactions promoted by the formation of Al-organic-acid-complexes, (2) in deeper samples, where pH is neutral and Al-complexation less favorable, dissolution is promoted by the formation of Si–organic-acid complexes (Bennett et al., 1988); in this zone, both aluminosilicates and quartz are weathered due to ligand exchange-promoted pathways at the silicon containing groups. One of their main arguments is the observed difference between the morphological aspect of minerals at the surface of the bog and in deeper layers. Minerals in deeper layers indeed show chemical weathering features whereas at the surface minerals show only features of wind erosion. However, their evidence is restricted to only ﬁve samples within a 3 m peat proﬁle and variations in mineralogy could be due to other causes such as variations in the supply of mineral particles from the air. For example, the chemically weathered quartz samples were located in the minerotrophic layer where input is not exclusively atmospheric. However, their approach reintroduced the idea of peatlands as strong agents of weathering in and underneath them (Endell, 1911; Humphreys and Julien, 1911; Blanck and Rieser, 1925; Blanck and Keese, 1928) and it serves as a case study to be followed by future investigations. Moreover, weathering of quartz and amorphous silica in bogs is of importance to understand tephra stability. Silicic grains from Holocene tephra found in bogs from northern Europe show geochemical stability for at least the last 4000 years (Dugmore et al., 1992). In that study, though, no data on pH was given. Siever (1962) studied amorphous silica dissolution using pore waters at pH 5 for 2 years. He showed that dissolution is lower in peat waters than in distilled waters. To conﬁrm that silica dissolution is enhanced by organic acids at neutral pH and not at acid pH, one possible way would be to study in parallel a tephra layer in a fen (pH 7) and in a bog (pH 4).

Weathering of inorganic matter in bogs

209

Finally, the dissolution of clay minerals, which constitute a main part of the mineral dust deposited in the peat, is also enhanced experimentally by the presence of simple organic acids (Huang and Kelle, 1971; Ganor and Lasaga, 1994; Hamer et al., 2003). Rates of weathering Based on a mineralogical composition determined using optical microscopy combined with SEM-EDS observations and constrained by major elements analyses of the insoluble ash fraction of peat, Steinmann and Shotyk (1997b) argued that Fe-hydroxides dissolution was very slow, and slower than predictive models. They claimed that a possible coating by organic molecules could have protected the Fe-bearing particles from weathering. In a bog from New Zealand, Hodder et al. (1991) showed a complete dissolution of biotite and a strong depletion of other ferromagnesians from a tephra layer within 770 years. They also found consistencies between differences in rates of dissolution of amphibole and pyroxenes in bogs compared with kinetic studies, but no absolute comparison between experimental rates of dissolution and observed rates were given. Rates of dissolution of feldspars were estimated to be at least three times slower in a Swiss bog than in experimental studies at pH 4 in HCl (Steinmann and Shotyk, 1997b). Different hypotheses were proposed such as the lower temperature in bog compared to the experiments, or the development of an amorphous Si layer around the mineral layer controlling the rate of dissolution. In addition, Al and other conservative elements such as Sc, Ti and Zr are preserved in the peat column (Shotyk, 2001; Shotyk et al., 2002), which is consistent with their conservative behavior during chemical weathering. Whereas there is no doubt of the presence of weathered aluminosilicates in bogs (Fig. 9.6), it is not yet known whether there are enhanced rates of dissolution of aluminosilicates because of the abundance of organic acids in pore waters or if these minerals could have been weathered before their entry into the bog. In addition, the high molecular weight humic substances, by becoming strongly adsorbed, may act more as inhibitors of dissolution rather than catalysts (Stumm et al., 1987). This phenomenon was studied by Eggenberger (1995) who found that the dissolution rate

Figure 9.6. Altered surface of quartz macro-grain found in Kolhu¨ttenmoor peat bog (Southern Black Forest, Germany) at a depth of 450 cm. In the same layer are also found macro-grains of K-feldspar characteristic of the granite surrounding the bogs, as well as smaller plagioclase grains.

210

G. Le Roux, W. Shotyk

of plagioclase feldspar (bytownite) in aqueous extracts of leaf litter at pH 4.0 was not signiﬁcantly different compared to the rate in HCl at the same pH; whereas the leaf litter extract did contain higher Al concentrations, the rate-controlling step in the dissolution of plagioclase feldspar in acidic solutions is the destruction of the siliceous leached layer at the weathered mineral surface (Shotyk and Metson, 1994). The decomposition of this layer apparently is not affected by organic ligands in the low-pH environment characteristic of bogs. A study by Curran et al. (2002) investigated the burial of archeological stones in a bog environment for 2000 years. Examples of two rock types, quartz porphyry and porphyritic andesite, placed there by humans, occur both buried and unburied in the bog. A comparison showed a transformation of biotite and secondary chlorite to ﬁne-grained micaceous products and clay minerals, alteration of sericitized potassium feldspars to clay minerals and removal of alteration products in solution. However because of peat cutting, pre-weakening of stones before burial and episodes of freeze–thaw episodes, it is difﬁcult to understand the mechanisms and rates of these changes. Special case: carbonate dissolution At low pH, the dissolution rate of carbonate is proportional to the activity of H+ (Plummer et al., 1978; Chou et al., 1989). The experimental rate of dissolution for carbonates at pH 4–5 is around 108–109 mol cm2 s1. In the Black Forest bog (Fig. 9.1), analyses of snow samples reveal that the largest calcite minerals entering the bog are 40 mm across (G. Le Roux, unpublished data). Assuming the mineral has the form of a cube, the time needed for the minerals to be completely dissolved in a well-stirred beaker will be between 30 min and 5 h. Rapid dissolution of calcite is consistent with XRD and optical and scanning electron microscope examinations, which failed to reveal carbonate minerals in the peat. Calcite was found, however, in Sphagnum mosses of the living plant layer, possibly because they had not yet been submerged in the acidic bog waters. At the peat–substrate interface, the bog could also be subjected to intense weathering of carbonates if they are present in the rock (Steinmann and Shotyk, 1997a) (Fig. 9.1). However the mechanism of dissolution will evolve over time, as dissolution proceeds and pH increases, with the rate at equilibrium (pH 6–7) much less than the initial rate (at pH 4).

Perspectives Microorganisms Bennett et al. (1991, 1994) suggested that microorganisms enhance the dissolution rate of silicates in organic environments such as bogs. Indeed, microorganisms may create pronounced chemical gradients near colonies and also produce organic acids. Unfortunately, little is known about the microbiology of peatlands and the related weathering processes, which these organisms might regulate.

Weathering of inorganic matter in bogs

211

Weathering rate and mass balance By studying bogs with varying pH proﬁles and peat accumulation rates in areas of given atmospheric deposition, it should be possible to perform chemical and mineralogical mass balances over time, to identify and quantify bulk changes. Experimental studies of speciﬁc mineral phases in acid bog waters using micro-analytical techniques such as synchrotron XRF and XRD, PIXE, LA-ICP-MS, TOF-ICP-MS and electron microprobe are badly needed. Pb and Sr isotopes tracers of dust sources and mineral weathering Pb and Sr isotopes can be used to trace sources of atmospheric dust as well as to identify weathering reactions (Erel et al., 1994; Shotyk et al., 1998; Emmanuel and Erel, 2002; Harlavan and Erel, 2002; Klaminder et al., 2003). For example, the isotopic signature of Pb is dependent on the host mineral and its geological age. Composition of water leaching the rock will depend, therefore, on the minerals weathered. Because Pb is immobile in bogs, probably because of high afﬁnity with organic matter, we can anticipate a different signature for the mineral phases and the organic matter in the peat. The isotopic composition of the organic matter would reﬂect the state of weathering of the minerals, possibly increasing with age and depth of the peat. However, the surface layers of all peat deposits studied to date have been found to be contaminated by anthropogenic Pb, thereby masking natural inputs. Moreover in ombrotrophic parts of peat not affected by anthropogenic contamination, Pb concentrations are very low, which makes it difﬁcult to achieve the necessary precision to reconstruct dust sources and weathering intensity.

Conclusions (1) Inorganic material entering a bog, principally quartz, clays, feldspars and calcite, enters a reactive environment, that is mainly anoxic, acid and dominated by organic matter. (2) The supply of minerals is controlled by climate, wind strength, plant cover, with particles provided both by local and long-range transport sources. (3) Whereas theoretical and laboratory studies indicate that many minerals such as aluminosilicates are unstable in this type of acid environment, reactivity and dissolution appear to be slower than expected. (4) The formation of a siliceous layer during the early stages of dissolution and/or coating by humic acids can help to explain these discrepancies. (5) Despite differences between the laboratory and natural weathering, pH plays a crucial role in the dissolution of minerals in peat. Ombrotrophic bogs are acidic at the surface, but dissolution reactions can neutralize the pore waters in deeper layers. Any attempts to reconstruct dust deposition should take this into account and present pH data both for the surface and in the deeper parts of the proﬁle to understand possible differences between the predominant weathering processes.

212

G. Le Roux, W. Shotyk

(6) Modern geochemical tools such as micro-analytical and isotopic instruments should be employed to better understand weathering of minerals in bogs.

References Anderson, J., 1797. A Practical Treatise on Draining Bogs and Swampy Grounds, Illustrated by Figures; with Cursory Remarks upon the Originality of Mr. Elkington0 s Mode of Draining. Paternoster-Row, London, 308pp. Andrejko, M.J., Cohen, A.D., and Raymond, R.J., 1983. Origin of mineral matter in peat. In: Raymond, R.J. and Andrejko, M.J. (Eds), Mineral Matter in Peat, its Occurrence, Form and Distribution. Los Alamos National Laboratory, Los Alamos, pp. 3–24. Bennett, P.C. and Casey, W., 1994. Chemistry and mechanisms of low-temperature dissolution of silicates by organic acids. In: Pittman, E.D. and Lewan, M.D. (Eds), Organic Acids in Geological Processes. Springer, Heidelberg, pp. 162–200. Bennett, P.C., Melcer, M.E., Siegel, D.I., and Hasset, J.P., 1988. The dissolution of quartz in dilute aqueous solutions of organic acids at 251C. Geochim. Cosmochim. Acta 52, 1521–1530. Bennett, P.C., Siegel, D.I., and Glaser, P.H., 1991. Fate of silicate minerals in a peat bog. Geology 19, 328–331. Bischoff, G., 1854. Elements of Chemical and Physical Geology. Cavendish Society, London, 158pp. Blanck, E. and Keese, H., 1928. U¨ber sogenannte Kaolinitisierung eines Granits unter Rohhumusbedeckung im Schwarzwald. Chem. Erde 4, 33–41. Blanck, E. and Rieser, A., 1925. U¨ber die chemische Vera¨nderung des Granits unter Moorbedeckung. Chem. Erde 2, 15–48. Blum, A., 1994. Feldspars in weathering. In: Parson, I. (Ed.), Feldspars and Their Reactions. Kluwer Academic Publishers, Amsterdam, pp. 595–630. Bu¨cher, A. and Lucas, C., 1984. Se´dimentation e´olienne intercontinentale, poussie`res sahariennes et ge´ologie. Bull. centre de recherche, expl. prod. Elf-Aquitaine 8, 151–165. Cama, J., Metz, V., and Ganor, J., 2002. The effect of pH and temperature on kaolinite dissolution rate under acidic conditions. Geochim. Cosmochim. Acta 66, 3913–3927. Chesworth, W. and Macias Vasquez, F., 1985. pe, pH, and podzolization. Am. J. Sci. 285, 128–146. Chou, L., Garrels, R.M., and Wollast, R., 1989. Comparative study of the kinetics and mechanisms of dissolution of carbonates minerals. Chem. Geol. 78, 269–282. Cross, A.T. and Phillips, T.L., 1990. Coal-forming plants through time in North America. Int. J. Coal Geol. 16, 1–46. Curran, J., Smith, B., and Warke, P., 2002. Weathering of igneous rocks during shallow burial in an upland peat environment: observations from the Bronze Age Copney Stone Circle Complex, Northern Ireland. Catena 49, 139–155. Dierssen, K. and Dierssen, B., 2001. Moore. Ulmer, Stuttgart, 230pp. Drever, J.I. and Vance, G.F., 1994. Role of soil organic acids in mineral weathering processes. In: Pittman, E.D. and Lewan, M.D. (Eds), Organic Acids in Geological Processes. Springer, Heidelberg, pp. 138–161. Dugmore, A.J. and Newton, A.J., 1992. Thin tephra layers in peat revealed by X-radiography. J. Archeol. Sci. 1992, 163–170. Dugmore, A.J., Newton, A.J., Sugden, D.E., and Larsen, G., 1992. Geochemical stability of ﬁne-grained silicic Holocene tephra in Iceland and Scotland. J. Quatern. Sci. 7, 173–183. Eggenberger, U., 1995. Mineral weathering in soils: experiments, ﬁeld studies, and modeling. Ph.D. Thesis, University of Bern, 198pp. Emmanuel, S. and Erel, Y., 2002. Implications from concentrations and isotopic data for Pb partitioning processes in soils. Geochim. Cosmochim. Acta 66, 2517–2527. Endell, K., 1911. U¨ber die chemische und mineralogische Vera¨nderung basicher Eruptivgesteine bei der Zersetzung unter Mooren. Neues Jahrbuch Miner. Geol. Pal. 31, 1–54.

Weathering of inorganic matter in bogs

213

Erel, Y., Harlavan, Y., and Blum, J.D., 1994. Lead isotopes systematics of granitoid weathering. Geochim. Cosmochim. Acta 58, 5299–5306. Finney, H.R. and Farnham, R.S., 1968. Mineralogy of the inorganic fraction of peat from two raised bogs in northern Minnesota. Third International Peat Congress, pp. 102–108. Franze´n, L.G., 2006 (this book, Chapter 11). Mineral matter and major and trace elements in raised bog peat. A case study from southern Sweden, Ireland and Tierra del Fuego, south Argentina. In: Martini, I.P., Matı´ nez Cortizas, A. and Chesworth, W. (Eds), Peatlands: Evolution and Records of Environmental and Climatic Changes. Elsevier, Amsterdam. Ganor, J. and Lasaga, A.C., 1994. The effects of oxalic acid on kaolinite dissolution rate. Mine. Mag. 58A, 315. Garrels, R.M. and Christ, C.L., 1965. Solutions, Minerals, and Equilibria. Harper & Row, New York, 460pp. Givelet, N., Le Roux, G., Cheburkin, A.K., et al., 2004. Suggested protocol for collecting, handling and preparing peat cores and peat samples for physical, chemical, mineralogical and isotopic analyses. J. Environ. Monitor. 6, 481–492. Glooschenko, W.A., 1986. Monitoring the atmospheric deposition of metals by use of bog vegetation and peat proﬁles. In: Nriagu, J.O. and Davidson, C.I. (Eds), Toxic Metals in the Atmosphere. Wiley, New York, pp. 507–533. Glooschenko, W.A., Holloway, L., and Arafat, N., 1986. The use of mires in monitoring the atmospheric deposition of heavy metals. Aquat. Bot. 25, 179–190. Gorham, E., Bayley, S.E., and Schindler, D.W., 1984. Ecological effects of acid deposition upon peatlands: a neglected ﬁeld in ‘‘acid rain’’ research. Can. J. Fish. Aquat. Sci. 41, 1256–1268. Gorham, E. and Tilton, D.L., 1978. The mineral content of Sphagnum fuscum as affected by human settlement. Can. J. Bot. 56, 2755–2759. Go¨ttlich, K., 1990. Moor-und Torf-Kunde. E.Schweizerbart0 sche Verlagsbuchhandlung, 529pp. Hamer, M., Graham, R.C., Amrhein, C., and Bozhilov, K.N., 2003. Dissolution of Ripidolite (Mg, Fe-Chlorite) in organic and inorganic acid solutions. Soil Sci. Soc. Am. J. 63, 815–822. Harlavan, Y. and Erel, Y., 2002. The release of Pb and REE from granitoids by the dissolution of accessory phases. Geochim. Cosmochim. Acta 66, 837–848. Hodder, A.P.W., De Lange, P.J., and Lowe, D.J., 1991. Dissolution and depletion of ferromagnesian minerals from Holocene tephra layers in an acid bog, New Zealand, and implications for tephra correlation. J. Quatern. Sci. 6, 195–208. Ho¨lzer, A. and Ho¨lzer, A., 1998. Silicon and titanium in peat proﬁles as indicators of human impact. Holocene 8, 685–696. Huang, W.H. and Kelle, W.D., 1971. Dissolution of clay minerals in dilute organic acids at room temperature. Am. Mineral. 56, 1082–1095. Humphreys, E.W. and Julien, A.A., 1911. Local decomposition of rock by the corrosive action of preglacial peat-bogs. J. Geol. 47–56. Hunt, T.S., 1875. Chemical and Geological Essays. Osgood, Boston, 489pp. Johnson, S.W., 1866. Peat and Its Uses, as Fertilizer and Fuel. O. Judd, New York, 168pp. Klaminder, J., Renberg, I., and Bindler, R., 2003. Isotopic trend and background ﬂuxes of atmospheric lead in northern Europe: analyses of three ombrotrophic bogs from south Sweden. Global Biogeochem. Cycles 17, 1019–1029 doi:10.1029/2002GB001921. Kolemikov, Y.M. and Shestakov, G.I., 1979. Lead isotope compositions for peat from the area of the Tungunska explosion of 1908. Geokhimyia 8, 1202–1211. Lewis Smith, R.I. and Clymo, R.S., 1984. An extraordinary peat-forming community on the Falkland Islands. Nature 309, 617–620. Malmer, N., 1986. Vegetational gradients in relation to environmental conditions in northwestern European mires. Can. J. Bot. 64, 375–383. Malmer, N., 1988. Patterns in the growth and the accumulation in inorganic constituents in the Sphagnum cover on ombrotrophic bogs in Scandinavia. Oikos 53, 105–120. Martı´ nez Cortizas, A., Chesworth, W., and Garcia Rodeja, E., 2001. In: Ma´rtinez Cortizas, A. and Garcia Rodeja, E. (Eds), Turberas de Montana de Galicia. Chapter 9: Dinamica geoquimica de las turberas de Galicia, Xunta de Galicia, pp. 141–148.

214

G. Le Roux, W. Shotyk

Martı´ nez Cortizas, A., Garcia Rodeja, E., Pontevedra-Pombal, X., et al., 2002. Atmospheric Pb deposition in Spain during the last 4600 years recorded by two ombrotrophic peat bogs and implications for the use of peat as archive. Sci. Tot. Environ. 292, 33–44. Martini, I.P. and Glooschenko, W.A., 1985. Cold climate peat formation in Canada, and its relevance to lower Permian coal measures of Australia. Earth-Sci. Rev. 22, 107–140. Mattson, S. and Koutler Andersson, E., 1954. Geochemistry of a raised bog. Kungl. Lantbrusho¨gskolans Annaler. 21, 321–366. McGlone, M.S., Moar, N.T., Wardle, P., and Meurk, C.D., 1997. Late-Glacial and Holocene vegetation and environment of Campbell Island, far southern New Zealand. Holocene 7, 1–12. McLennan, S.M., 2001. Relationships between the trace element composition of sedimentary rocks and upper continental crust. Geochem. Geophys. Geosyst. 2, 2001GC000172. Niklas, H., 1912. Untersuchungen u¨ber den Einﬂuss von Humustoffen auf der Verwiterrung der Silikate. Int. Mitt. Bodenk. 2, 214–244. Nova´k, M., Adamova, M., and Milicic, J., 2003. Sulfur metabolism in polluted Sphagnum peat bogs: a combined 34S–35S–210Pb study. Water Air Soil Pollut. 3, 181–200. Nriagu, J.O., 1979. Global inventory of natural and anthropogenic emissions of trace metals to the atmosphere. Nature 279, 409–411. Nriagu, J.O., 1989. A global assessment of natural sources of atmospheric trace metals. Nature 338, 47–48. Page, S.E., Rieley, J., Shotyk, W., and Weiss, D., 1999. Interdependence of peat and vegetation in a tropical peat swamp forest. Philos. Trans. Roy. Soc. B 354, 1885–1897. Plummer, L.N., Wigley, T.M.L., and Parkhurst, D.L., 1978. The kinetics of calcite dissolution in CO2water system at 5–601C and 0–1.0 atm CO2. Am. J. Sci. 278, 179–216. Punning, J.-M. and Alliksaar, T., 1997. The trapping of ﬂy-ash particles in the surface layers of Sphagnumdominated peat. Water Air Soil Pollut. 94, 59–69. Ramann, E., 1895. Organogene Ablagerungen der Jetztzeit. Neues Jahrb. Mineral. Geol. Pala¨o. 10, 119–166. Raymond, R.,Jr., Andrejko, M.J., and Bardin, S.W., 1983, Techniques for applying scanning electron microscopy to the study of mineral matter. In: Raymond, R., Jr. and Andrejko, M.J. (Eds), Proceedings or meeting on Mineral matter in peat; its occurrence, form, and distribution. Los Alamos, Los Alamos National Laboratory Report LA–9907–OBES, September 26–30, pp. 169–179. Schu¨tz, L., 1989. Atmospheric mineral dust: properties and source markers. In: Leinen, M. and Sarnthein, M. (Eds), Paleoclimatology and Paleometeorology: Modern and Past Patterns of Global Atmospheric Transport. NATO Series C: Mathematical and Physical Sciences. Kluwer Academic Publishers, London, Vol. 282, pp. 359–384. Shearer, J.C., 1997. Natural and anthropogenic inﬂuences on peat development in Waikato/Hauraki Plains restiad bogs. J. Roy. Soc. NZ 27, 295–313. Shotyk, W., 1988. Review of the inorganic geochemistry of peats and peatland waters. Earth-Sci. Rev. 25, 95–176. Shotyk, W., 1989. The geochemistry of peatland waters. Water Quality Bull. 14, 47–53+108. Shotyk, W., 1992. Organic soils. In: Martini, I.P. and Chesworth, W. (Eds), Weathering, Soils and Paleosols. Elsevier, Amsterdam, pp. 203–224. Shotyk, W., 1996. Natural and anthropogenic enrichments of As, Cu, Pb, Sb, and Zn in ombrotrophic versus minerotrophic peat bog proﬁles, Jura Mountains, Switzerland. Water Air Soil Pollut. 90, 375–405. Shotyk, W., 1997. Atmospheric deposition and mass balance of major and trace elements in two oceanic peat bogs proﬁles, northern Scotland and the Shetland Islands. Chem. Geol. 138, 55–72. Shotyk, W., 2001. Geochemistry of the peat bog at Etang de la Grue`re, Jura Mountains, Switzerland, and its record of atmospheric Pb and lithogenic trace metals (Sc, Ti, Y, Zr, and REE) since 12,370 14C BP. Geochim. Cosmochim. Acta 65, 2337–2360. Shotyk, W., Krachler, M., Martı´ nez Cortizas, A., et al., 2002. A peat bog record of natural, preanthropogenic enrichments of trace elements in atmospheric aerosols since 12370 14C yr, and their variation with Holocene climate change. Earth Planet. Sci. Lett. 199, 21–37. Shotyk, W. and Metson, J.B., 1994. Secondary ion mass spectrometry (SIMS) and its application to chemical weathering. Rev. Geophys. 32, 197–220.

Weathering of inorganic matter in bogs

215

Shotyk, W., Nesbitt, H.W., and Fyfe, W.S., 1992. Natural and anthropogenic enrichments of trace metals in peat proﬁles. Int. J. Coal Geol. 20, 49–84. Shotyk, W., Weiss, D., Appleby, P.G., et al., 1998. History of atmospheric lead deposition since 12,730 14C yr BP from a peat bog, Jura Moutains, Switzerland. Science 281, 1635–1640. Siever, R., 1962. Silica solubility, 0–2001C, and the diagenesis of siliceous sediments. J. Geol. 70, 127–150. Steinmann, P. and Shotyk, W., 1997a. Chemical composition, pH, and redox state of sulfur and iron in complete vertical porewater proﬁles from two Sphagnum peat bogs, Jura Mountains, Switzerland. Geochim. Cosmochim. Acta 616, 1143–1163. Steinmann, P. and Shotyk, W., 1997b. Geochemistry, mineralogy, and geochemical mass balance on major elements in two peat bog proﬁles (Jura Mountains, Switzerland). Chem. Geol. 138, 25–53. Stumm, W., 1992. Chemistry of the Solid–Water Interface: Processes at the Mineral–Water and Particle–Water Interface in Natural Systems. Wiley, New York, 428pp. Stumm, W., Wehrli, B., and Wieland, E., 1987. Surface complexation and its impact on geochemical kinetics. Croatia Chem. Acta 60, 429–456. VanLoon, G.W. and Duffy, S.J., 2000. Environmental Chemistry. Oxford University Press, Oxford, 492pp. Veski, S., Heinsalu, A., Kirsima¨e, K., et al., 2001. Ecological catastrophe in connection with the impact of the Kaali meteorite about 800–400 B.C. on the island of Saaremaa, Estonia. Meteorit. Planet. Sci. 36, 1367–1375. Von Bu¨low, K., 1929. Allgemeine Moorgeologie: Einfu¨hrung in das Gesamtgebiet der MoorkundeGerbruder Borntraeger, 308pp. Ward, C.R., 2002. Analysis and signiﬁcance of mineral matter in coal seams. Int. J. Coal Geol. 50, 135–168. Wastegard, S., Hall, V.A., Hannon, G.E., et al., 2003. Rhyolitic tephra horizons in northwestern Europe and Iceland from the A.D. 700s–800s: a potential alternative for dating ﬁrst human impact. Holocene 13, 277–283. Weiss, D., Shotyk, W., Rieley, J., et al., 2002. The geochemistry of major and selected trace elements in a forested peat bog, Kalimantan SE Asia, and its implications for past atmospheric dust deposition. Geochim. Cosmochim. Acta 66, 2307–2323. Williams, M., 1992. Evidence for the dissolution of magnetite in recent Scottish peats. Quatern. Res. 37, 171–182. Wu¨st, R.A.J., Ward, C.R., Bustin, M.R., and Hawke, M.I., 2002. Characterization and quantiﬁcation of inorganic constituents of tropical peats and organic-rich deposits from Task Brea (Peninsular Malaysia): implications for coals. Int. J. Coal Geol. 49, 215–249.

This page intentionally left blank

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

217

Chapter 10

Molecular chemistry by pyrolysis–GC/MS of selected samples of the Penido Vello peat deposit, Galicia, NW Spain P. Buurman, K.G.J. Nierop, X. Pontevedra-Pombal and A. Martı´ nez Cortizas

Introduction The organic matter of peat deposits has been analyzed by many authors, using different techniques for different purposes. In peats, each vegetation leaves its remains on top of those of the previous vegetation phase, the mixing effect is much less severe than in soils, and each layer may represent a speciﬁc vegetation community that moreover can be identiﬁed microscopically by its preserved remains. Therefore, the link between the vegetation community and its (partially decomposed) organic remains can be studied. Peats of ombrogenic (rain-fed) mires, which do not have sedimentary addition of organic and mineral matter and lack speciﬁc protection due to mineral–organic bonding, are well suited to such analysis. Various researchers have recognized these possibilities and have tried to ﬁnd links between vegetation and peat chemistry. Pancost et al. (2002) wrote a general plea for the use of biomarkers to characterize plant inputs in peat. Biomarkers are chemical components that are speciﬁc for species or groups of plants. Many biomarkers and groups of biomarkers have been suggested: lignin, lipids such as n-alkanes, sterols and hopanoids, speciﬁc organic acids, triterpenoids, resorcinols and others. Because of the low rate of Sphagnum decomposition, much work has been dedicated to the composition of Sphagnum peats and their constituents. Van Smeerdijk and Boon (1987) investigated Sphagnum and Ericaceae by pyrolysis–gas chromatography–mass spectrometry (py–GC/MS). Lehtonen et al. (1991) investigated lipids in dissolved organic material of Carex and Sphagnum peats. Van der Heijden and Boon (1994) investigated decomposition of Calluna wood in Sphagnum peat, and Van der Heijden (1994) investigated the chemistry of a number of peatiﬁed plant tissues. Verhoeven and Liefveld (1997) made a detailed investigation of Sphagnum chemistry, whereas Gleixner and Kracht (2001) investigated the inﬂuence of humiﬁcation on Sphagnum chemistry. More generally, the relation between n-alkanes and vegetation was investigated by Nott et al. (2000). Avsejs et al. (2002) used resorcinols as biomarkers of ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09010-9

218

P. Buurman et al.

sedges. Kuder and Kruge (1998) and Kuder et al. (1998) searched for biomarkers as paleoenvironmental proxies. Farrimond and Flanagan (1996) compared lipid stratigraphy of Holocene deposits in the UK with the pollen record. Karunen and Ekman (1982) found changing lipid contents with peat age. Some authors looked at decomposition of speciﬁc chemical compounds, such as lignin (Young and Frazer, 1987; Stout et al., 1988). The diagenetic trend in chemical composition of peat refers to anaerobic decomposition that causes a gradual chemical change in buried peat deposits. In low-resolution studies (large intervals between samples) using 13C-cross polarization-magic angle spinning (CPMAS) NMR, Hammond et al. (1985) and Fox et al. (1994) in Canada and Beyer et al. (1997) in Antarctica found a general increase of alkyl C and a decrease of O-alkyl C with depth. In a high-resolution study (5 cm sample intervals) of the Penido Vello (PVO) peat in NW Spain, using the same technique, Pontevedra-Pombal (2002) and Pontevedra-Pombal et al. (2001) found a general loss of oxygenated groups and a relative increase of non-oxygenated aliphatics and aromatics with depth. Consequently, the C content of the peat organic matter increased from about 42% in the top vegetation layer to 64% at 2.5 m depth. In addition to the normal anaerobic decomposition trend, there is enhanced decomposition brought about by artiﬁcial drainage and resulting entry of oxygen. Preston et al. (1987, 1989) used 13C-CPMAS-NMR to characterize decomposition in peat in cultivated and virgin sites. Krosshavn et al. (1992) used the same technique to compare chemical trends in peat and other soils in relation to vegetation and decomposition. Other authors investigated the effect of decomposition on speciﬁc chemical compounds, such as degradation of glucose in Sphagnum (Bergman et al., 2000). Kondo and Tsutsuki (2001) found increase in phenolics and long-chain fatty acids upon decomposition of a high moor. The technique of py–GC/MS was recognized early as a suitable tool for the study of whole peat samples and several authors have used this technique for in-depth characterization. Early work by Halma et al. (1984) and Boon et al. (1985, 1986) established that the technique could be used to characterize peats, and to separate peats with different characteristics using factor analysis on pyrolysis results. Notwithstanding the clear potential of peat for understanding chemical differences and differentiation, peats have not been analyzed systematically at the molecular level. The PVO peat deposit in the Serra do Xistral (Galicia, NW Spain, Fig. 10.1) is the only peat of which samples for 13C-CPMAS-NMR were taken at short intervals (Pontevedra-Pombal et al., 2001; Pontevedra-Pombal, 2002). This peat deposit was previously studied for its long-term documentation of atmospheric pollution (Martı´ nez Cortizas et al., 1997, 2002). The 80 samples of the 2.5-m thick deposit show, in accordance with other studies, that alkyl C generally increased with depth in the deposit, whereas O-alkyl C decreased. This trend is illustrated in Figure 10.2 (other functional groups omitted). The general trend, however, is not smooth and shows a large number of deviations. Whereas the general trend represents the gradual anaerobic decay of a homogeneous peat deposit, deviations may have several causes. Peat vegetation changes with time, and therefore the chemistry of the peat also changes. In addition, periods of drought, which force vegetation change, also improve peat drainage that may onset aerobic decay of the peat and sometimes of peat

Molecular chemistry by pyrolysis– GC/MS

219

Figure 10.1. Location of the Penido Vello (PVO) peat deposit in NW Spain.

ﬁres. Because aerobic decay is more effective than anaerobic decay for chemical compounds such as lignin, it is expected to have a distinct effect on the chemical signature of the residue. The purpose of this investigation is to analyze the causes of the observed chemical ﬂuctuations and to obtain further detail at the molecular level of the general trend of decomposition with depth.

Materials and methods The PVO peat deposit and its composition based on 13C-CPMAS-NMR, was described in detail by Pontevedra-Pombal et al. (2001) and Pontevedra-Pombal (2002). The ﬁrst meter of this peat deposit was sampled at intervals of 2 cm; below 1-m depth the interval was 5 cm. Based on differences in quantiﬁed NMR data, 21 samples were selected for further analysis (Fig. 10.2). The samples were extracted with NaOH (see below) to separate extractable matter from the residue. We expect the residue to more closely mirror the chemistry of the original vegetation, whereas differences between extracts and residues illustrate effects of decomposition and of solubility. Extraction of humus Five-gram samples were shaken for 24 h under nitrogen with 50 ml of 0.1 M NaOH. The suspension was centrifuged for 1 h at 4000 rpm and decanted. The extracted matter was not separated into humic acid and fulvic acid fractions. The extract was acidiﬁed to pH 1–2 with concentrated HCl, after which 9 ml of concentrated HF was added to reach a ﬁnal concentration of 0.3 M. This solution was shaken for 24 h, dialyzed against distilled water to neutral pH, and freeze-dried. The residue was acidiﬁed, washed with distilled water, centrifuged and also freeze-dried.

220

P. Buurman et al.

Figure 10.2. Changes with depth of O-alkyl C and alkyl C in the Penido Vello peat deposit. Numbers correspond to the peat samples selected for this study.

Pyrolysis– gas chromatography– mass spectrometry Freeze-dried extracts and residues were pyrolyzed using a Curie-Point pyrolyser (Curie-temperature of wire ¼ 6001C) connected to a Carlo Erba GC 8000 gas chromatograph. The pyrolysis products were separated on a fused silica column (Chrompack 25 m, 0.25 mm i.d.) coated with CP-Sil 51b (ﬁlm thickness 0.40 mm), with helium as carrier gas. The initial oven temperature was 401C, the heating rate 71C min1. The ﬁnal temperature of 3201C was maintained for 20 min. The GC column was connected to a Fisons MD 800 mass spectrometer (mass range m/z 45–650, cycle time 1 s). The pyrograms of the extracts and residues of 12 samples were analyzed in detail. This resulted in more than 600 different pyrolysis fragments. After removal of minor compounds that occurred in few samples (unidentiﬁed compounds and silica-carbon compounds (silanes, siloxanes)), a list of 225 identiﬁed compounds remained. These

Molecular chemistry by pyrolysis– GC/MS

221

compounds are given in Table 10.1 (van Bergen et al., 1997). They were quantiﬁed for all 42 extracts and residues, using the two main ions of each compound (Table 10.1). The total ion current (TIC) for these 225 products was set as 100%, and relative amounts were calculated with respect to this sum. One should bear in mind that this quantiﬁcation is valid only for the pyrolyzed part of the organic matter that is amenable to GC, but may not reﬂect the composition of all the organic matter present. Furthermore, the quantiﬁcation indicates relative abundances of pyrolysis products and not weight percentages. Acetic acid has a molecular weight (MW) of 60, whereas an n-C33 alkane fragment has a MW of 464, so that with equal fragment abundance, the latter fragment represents a weight percentage that is more than seven times larger. Quantiﬁed abundance, however, does not need to be converted into weight percentages to give meaningful results. The compounds were grouped according to probable origin into the following groups (ﬂags): straight alkanes and alkenes, other alkanes, alkenes and alcohols, aromatics, cyclo-pentenes and -hexenes, polyaromatics, fatty (alkanoic) acids, lignin, lipids (steroids and terpenoids), esters, methyl ketones, nitrogen compounds, phenols and polysaccharides. Principal component analysis Principal component analysis (PCA), using Statistica (Statsoft) software, was carried out for the combination of 42 extracts and residues extracts and 225 chemical fragments. In a second approach, all fragments with a mean relative abundance of less than 0.1% were omitted, and PCA was carried out with the remaining 141 fragments (printed in bold in Table 10.1). The results obtained from the full set and from the reduced set were not essentially different, and therefore further analyses and discussion are based on the reduced set alone.

Results and discussion In the analyzed samples we distinguish between the extractable fraction and the nonextractable residue. Because the extractant is a polar liquid, non-polar organics tend to remain in the residue. Also, smaller molecules tend to be more soluble than larger, polymerized ones, and one should expect to ﬁnd this reﬂected in the differences between extracted fraction and residue. In Table 10.2, the mean composition of extracts and residues is given by chemical group. The differences between extracts and residues are especially clear in the summed (abundance of) lignin, phenols and polysaccharide fragments, but also aromatics, polyaromatics, and nitrogen-containing compounds are signiﬁcantly different. In the residues, polysaccharides decrease with depth (not shown), especially from sample 46 downwards. Aliphatics, alcohols, fatty acids and methyl ketones increase signiﬁcantly from sample 59 downwards. Methyl esters have highest contents in samples 05 and 08 (they are in a peat layer with the lowest content in aromatic and phenolic C determined by NMR; data not shown). Table 10.2 indicates that the analysis of bulk samples without separation into extractable and

P. Buurman et al.

222

Table 10.1. List of pyrolysis compounds, weights, typical masses, retention times, abbreviations (code) and scores on Factor 1. Score on Factor 1

Code

Name

10:0 11:0 11:1 12:0 12:1 13:0 13:1 14:0 14:1 15:0 15:1 16:0 16:1 17:0 17:1 18:0 18:1 19:0 19:1 20:0 20:1 21:0 21:1 22:0 22:1 23:0 23:1 24:0 24:1 25:0 25:1 26:0 26:1 27:0 27:1 28:0 28:1 29:0 29:1 30:0 30:1 31:0

n-C10:0 n-C11:0 n-C11:1 n-C12:0 n-C12:1 n-C13:0 n-C13:1 n-C14:0 n-C14:1 n-C15:0 n-C15:1 n-C16:0 n-C16:1 n-C17:0 n-C17:1 n-C18:0 n-C18:1 n-C19:0 n-C19:1 n-C20:0 n-C20:1 n-C21:0 n-C21:1 n-C22:0 n-C22:1 n-C23:0 n-C23:1 n-C24:0 n-C24:1 n-C25:0 n-C25:1 n-C26:0 n-C26:1 n-C27:0 n-C27:1 n-C28:0 n-C28:1 n-C29:0 n-C29:1 n-C30:0 n-C30:1 n-C31:0

M alkane alkane alkene alkane alkene alkane alkene alkane alkene alkane alkene alkane alkene alkane alkene alkane alkene alkane alkene alkane alkene alkane alkene alkane alkene alkane alkene alkane alkene alkane alkene alkane alkene alkane alkene alkane alkene alkane alkene alkane alkene alkane

142 156 154 170 168 184 182 198 196 212 210 226 224 240 238 254 252 268 266 282 280 296 294 310 308 324 322 338 336 352 350 366 364 380 378 394 392 408 406 422 420 436

Masses 57+71 57+71 55+69 57+71 55+69 57+71 55+69 57+71 55+69 57+71 55+69 57+71 55+69 57+71 55+69 57+71 55+69 57+71 55+69 57+71 55+69 57+71 55+69 57+71 55+69 57+71 55+69 57+71 55+69 57+71 55+69 57+71 55+69 57+71 55+69 57+71 55+69 57+71 55+69 57+71 55+69 57+71

Positive

Negative

RT (min)

Extract

Residue

8.996 11.170 10.932 13.249 13.039 15.236 15.102 17.154 17.012 18.975 18.751 20.625 20.453 22.210 22.070 23.747 23.597 25.175 25.044 26.537 26.426 27.853 27.747 29.114 29.022 30.316 30.225 31.473 31.383 32.587 32.497 33.658 33.587 34.672 34.565 35.663 35.611 36.620 36.561 37.542 37.490 38.445

0.73 0.84 0.60 0.68 0.68 0.52 0.47 0.45 0.42 x x 0.27 x 0.28 x 0.55 x 0.29 0.28 0.65 0.31 0.45 0.44 0.66 0.54 0.69 0.74 0.82 0.54 0.51 0.85 0.77 0.73 0.35 0.70 0.31 0.58

Molecular chemistry by pyrolysis– GC/MS

223

Table 10.1 (continued ) Score on Factor 1

Code

Name

M

33:0 Al01 Al03 Al04 Al05 Al07 Al08 Al18 Al20 Al21 Al22 Al23 Al24 Al26 Ar01 Ar02 Ar03 Ar04 Ar05 Ar06 Ar07 Ar08 Ar09 Ar10 Ar11

n-C33:0 alkane Alkene/alcohol Alkane Alkene Alkene Alkadiene/alkenol Alkane diol/alkadiene C18-alcohol C20-alcohol C21-alcohol C22-alcohol C23-alcohol C24-alcohol C26-alcohol Benzene Toluene C2-benzene 1,2/1,4 Dimethylbenzene C2-benzene C2-benzene Propylbenzene Butylbenzene Nonylbenzene Benzene, alkylAlkylbenzene compound (C12) Alkylbenzene (tridecyl) 135 Cyclopentene Cyclopropane Cyclohexene, 4ethenyl,1,4-dimethyl/ limonene Naphtalene, 1/2-methyl, compound 1H-Indene-1-one, compound x,x,xTrimethylnaphtalene Anthracene 3,7,11,15-Tetramethyl-2hexadec-1-ol 109

464

Ar12 C01 C02 C05

Ch01 Ch02 Ch03 Ch04 Cp1

Masses

RT (min)

Positive

Negative

Extract

Residue

204 164 246

57+71 55+69 57+71 55+69 55+69 55+67 55+69 55+69 55+57 55+69 55+57 55+69 55+69 55+59 77+78 91+92 91+106 91+106 78+104 91+106 91+120 91+134 91+92 107+108 91+92

40.137 21.692 24.292 24.734 25.717 28.093 29.405 27.521 30.061 31.299 32.407 33.504 34.553 36.206 2.889 4.182 5.853 6.033 6.416 6.550 7.783 10.036 20.019 21.438 24.911

260

91+92

26.368

0.810

53+67 56+70 68+93

3.860 6.575 9.557

0.870 x

115+142

14.972

188

173+188

16.169

0.960

170

155+170

20.381

0.960

176+178 57+71

23.371 23.158

x

326

78 92 106 106 104 120

0.63 x 0.780 0.53 0.40 0.78 x 0.39 0.53 0.81 0.72 0.66 0.70 0.890 0.970 0.960 0.660 0.960 0.920 0.930 0.940 0.900 0.860

0.80

0.62

P. Buurman et al.

224 Table 10.1 (continued )

Score on Factor 1

Extract

Residue

Name

M

Cp2

3,7,11,15-Tetramethyl-2hexadecene-ol (phytadiene) 2-Chlorethyl linoleate (phytadiene?) n-C8 alkanoic acid n-C14 alkanoic acid n-C15 alkanoic acid iso-C15 alkanoic acid n-C16 alkanoic acid n-C17 alkanoic acid n-C18 alkanoic acid n-C18:1 alkenoic acid n-C19 alkanoic acid n-C20 alkanoic acid n-C22 alkanoic acid n-C24 alkanoic acid 2-Methoxyphenol (guaiacol) 4-Methylguaiacol (2methoxy-4methylphenol) 4-Vinylphenol 4-Ethylguaiacol 4-Vinylguaiacol Substituted methoxyphenol Syringol (2,6dimethoxyphenol) 4-(-2-Propenyl)phenol 4-(1-Propenyl)guaiacol (eugenol) 3,5-Dimethoxyphenol 4-Propylguaiacol 4-Formylguaiacol, vanillin cis-4(2Propenyl)guaiacol 4-Methylsyringol trans-4-(2Propenyl)guaiacol 4-Acetylguaiacol Substituted guaiacol

296

57+81

24.894

x

342

54,67+81

25.493

0.67

F08 F14 F15 F15i F16 F17 F18 F18:1 F19 F20 F22 F24 Lg01 Lg02

Lg03 Lg04 Lg05 Lg06 Lg07 Lg08 Lg09 Lg10 Lg11 Lg12 Lg13 Lg14 Lg15 Lg16 Lg17

144 228

RT (min)

Negative

Code

Cp3

Masses

Positive

312 340 368 124

60+73 129+185 129+185 60+73 129+185 129+185 129+185 55+69 57+71 129+185 129+185 129+185 109+124

12.608 23.014 24.461 23.954 25.904 27.227 28.531 28.221 34.252 30.949 33.194 35.275 10.534

138

123+138

12.764

0.880

120 152 150 138

91+120 137+152 135+150 107+138

13.391 14.532 15.215 15.332

0.920 0.890 0.860 0.930

154

139+154

15.760

0.890

134 164

133+134 77+164

15.768 16.012

0.950 x

154 166 152

139+154 137+166 151+152

16.148 16.245 16.632

0.910 0.880 0.600

164

149+164

16.979

x

168 164

153+168 77+164

17.577 17.732

0.840

166 166

151+166 151+166

18.250 18.617

0.930 0.880

256 270 284

0.34 0.350

0.39 x 0.35 0.28 x 0.76 0.69 0.960

Molecular chemistry by pyrolysis– GC/MS

225

Table 10.1 (continued ) Score on Factor 1 Negative

RT (min)

Extract

Residue

Code

Name

M

Lg18 Lg19

Vanillic acid methyl ester Guaiacylacetone (4propan-2one)guaiacol 4-Ethylsyringol 4-Propan-2-one guaiacol 4-Hydroxy-3methoxy benzoic acid (vanillic acid) 4-(Prop-1-enyl)syringol 4-Formyl syringol 4(Prop-2-enyl) syringol, cis Substituted syringol 4-(2-Propenyl)syringol, trans 4-Acetylsyringol 4-(Propan-2one)syringol 4-(Propan-3one)syringol Syringic acid Substituted syringol Hopanoid Cholest-5-en-3-ol carbochloridate/ propanoate Hopanoid? Neooleana component Verticiol Cholest-ene compound urs-20-en-16-one (hopanoid?) Ergost-22-en-3-ol component Cholest-5-en-3-ol component Gamma tocopherol Stigmasta-5, -22dien-3ol, acetate Beta amyrin Chol-8-(14)-en-24-ol (5 beta)

182 180

151+182 137+180

18.827 18.980

0.920 0.950

182 180 168

167+182 151+180 153+168

19.025 19.906 19.962

0.890 0.910 0.810

194 182 194

91+194 181+182 91+194

20.261 21.100 21.081

0.740 0.670 0.790

194

167+182 91+194

21.690 21.862

0.670 0.760

196 210

181+196 167+210

22.297 22.808

0.970 0.890

210

181+210

23.672

0.920

23.922 24.544 36.438 36.948

0.920 0.330 0.300

448

183+198 179+194 123+191 81+147

191+231 365+366 121+133 81+145 149+191

37.227 37.332 37.332 37.585 37.654

400

257+298

37.871

448

147+367

38.005

0.45

151+416 255+394

38.030 38.237

0.58 0.40

203+218 215+344

38.306 38.406

0.53 x

Lg20 Lg21 Lg22

Lg23 Lg24 Lg25 Lg26 Lg27 Lg28 Lg29 Lg30 Lg31 Lg32 Lp01 Lp02

Lp03 Lp04 Lp05 Lp06 Lp07 Lp08 Lp09 Lp10 Lp11 Lp12 Lp13

408

Masses

Positive

x 0.300 0.62 x

P. Buurman et al.

226 Table 10.1 (continued )

Score on Factor 1

Code

Name

Lp14

Cholesta-4,6-dien-3-ol compound Stigmastan-3-ol compound Hopanoid I Lup-2(29)-en-28-ol D-friedoolean-14-en-3one (taraxerone) Lup-20(29)-ene-3, 21dione,28-hydroxy Pregnan-3-one 28-nor-18-alpha (Armanios et al., 1995) Stigmasta-3,5-dien-7-one Stigmast-4-en-3-one Friedoolean-7-ol compound Alkanoic acid methyl ester (C7?) Aatty acid, methyl ester, dioctylester Difatty acid dimethyl ester (C16?) n-C24-alkanoic acid, methyl ester n-C16-alkanoic acid octadecyl ester Methyl ketone (2-one) Methylketone Alkane-2,4-dione Alkane-2,4-dione C-33 methylketone n-C24:0 methylketone n-C25:0 methylketone n-C26:0 methylketone n-C27:0 methylketone n-C28:0 methylketone n-C29:0 methylketone n-C31:0 methylketone n-C33:0 methylketone Pyridine C1-Pyrrole

Lp15 Lp16 Lp17 Lp18 Lp19 Lp20 Lp21

Lp22 Lp23 Lp24 Me1 Me2 Me3 Me4 Me5 Mk01 Mk02 Mk03 Mk04 Mk05 Mk24 Mk25 Mk26 Mk27 Mk28 Mk29 Mk31 Mk33 N01 N02

M

Masses

RT (min)

Positive

Negative

Extract

Residue

135+143

38.601

x

147+396

38.809

0.48

424

69+191 191+207 204+300

39.699 40.274 40.720

454

95+205

41.133

0.49

302

231+302 135+245

41.200 41.236

x 0.57

410

174+410 124+229 69+95

41.649 42.080 43.227

0.36

74+87

16.651

57+129

31.257

72+98

31.671

x

74+87

34.867

0.520

313+564

39.962

58+59 58+59 85+100 85+100 58+59 58+59 58+59 58+59 58+59 58+59 58+59 58+59 58+59 52+79 80+81

24.217 27.762 37.687 39.440 42.236 33.667 34.725 35.734 36.631 37.647 38.549 40.274 33.667 3.758 5.209

130

318

394 408 422

79

x x 0.67

0.34 0.300 x

x 0.80 0.71 0.32 0.60 0.84 0.81 0.84 0.83 0.80 0.85 0.60 0.940 0.840

Molecular chemistry by pyrolysis– GC/MS

227

Table 10.1 (continued ) Score on Factor 1

Code

Name

N03 N04 N05 N06 N07

1H-Pyrrole, 3-methyl 3-Methylpyridine 2,4-Dimethylpyridine ?? Indole 3/5-Methyl-pyridinone2-(1 H) Benzene, 1-isocyanato-2methyl Diketodipyrrole Pyrrolydine compound x-Methylphenol Phenol 2-Methylphenol 3/4-Methylphenol Dimethylphenol 3/4-Ethylphenol Dihydroxybenzene (catechol) 3-Methoxyphenol 3-Methoxy-1,2benzenediol 3-Methyl-1,2benzenediol 4-Methyl-1,2benzenediol 4-Ethyl-1,2-benzenediol Acetylphenol Butanephenol compound 1,3-Benzenediol, C15sidechain Prist-1-ene 2-Methylfuran Acetic acid 2,5-Dimethylfuran (2H)-Furan-3-one(?) 3-Furaldehyde 2-Furaldehyde 2-Propylfuran +98, 110 2,3-Dihydro-5methylfuran-2-on 3-Methyl butanoic acid 5H-Furan-2-one

N08 N09 N10 Ph01 Ph02 Ph03 Ph04 Ph05 Ph06 Ph07 Ph08 Ph09 Ph10 Ph11 Ph12 Ph13 Ph14 Ph15 Pr Ps01 Ps02 Ps03 Ps04 Ps05 Ps06 Ps07 Ps08 Ps09 Ps10

Positive

Negative

RT (min)

Extract

Residue

80+81 66+93 106+107 90+117 80+109

5.486 5.724 7.198 14.708 16.437

0.910 0.890 0.820 0.940

104+133

18.077

0.840

93+186 98+113 107+108 66+94 107+108 107+108 107+122 107+122 64+110

21.956 22.900 5.209 8.378 9.937 10.396 11.918 12.312 13.274

0.720 0.860 0.920 0.960 0.950 0.980 0.930 0.950

94+124 125+140

13.511 14.238

0.950 0.960

78+124

14.375

0.940

78+124

15.079

0.640

123+138 121+136 131+192

16.787 17.634 21.553

0.890

320

124+137

25.823

0.720

266 82 60 96 84 96 96 98

56+57 53+82 60 53+96 54+84 95+96 95+96 53+81 55+98

22.695 2.384 2.465 3.317 4.323 4.644 4.989 5.591 5.591

102 84

60+87 55+84

5.651 6.235

M 81 93 117

186

94 108 122 122 110

136

Masses

x

0.930

0.59 0.850 0.920 0.800 x x 0.700 x 0.620 0.910 0.640

P. Buurman et al.

228 Table 10.1 (continued )

Score on Factor 1

Code

Name

Ps11

2,3-Dihydro-5methylfuran-2-one 2,3-Dimethyl cyclopent2-en-1-one 2-Pentenoic acid 5-Methyl-2-furaldehyde ? ‘‘Sugar’’ 4-Hydroxy-5,6-dihydro(2H)-pyran-2-one 3,4-Dihydropyran-2,5dione 2-Hydroxy-3-methyl-2cyclopenten-1-one Rhamnose marker (dianhydrorhamnose) 58 2-Propan-2-one tetrahydrofuran Levoglucosenone 3-Hydroxy-2-methyl4H-pyran-4-one 4H-Pyran-4-one,2,3dihydro-3,5dihydroxy-6-methyl 1-Deoxy-2,4-methylene3,5-anhydro-D-xylitol 3,5,Dihydroxy-2-methyl(4H)-pyran-4-one 1,4:3,6-Dianhydroalpha-Dglucopyranose 5-Hydroxymethyl-2furancarboxaldehyde Sugar 1 1,4-Dideoxy-D-glycerohex-1-enopyranose-3ulose Levogalactosan Levomannosan Butenoic acid, 2-methyl, propyl Levoglucosan

Ps12 Ps13 Ps14 Ps15 Ps16 Ps17 Ps18 Ps19

Ps20 Ps21 Ps22 Ps23

Ps24 Ps25 Ps26

Ps27 Ps28 Ps29

Ps30 Ps31 Ps32 Ps33

M 98

Masses

RT (min)

Positive

Negative

Extract

Residue

55+98

6.877

57+67

7.458

114

55+100 109+110 55+86 58+114

7.474 7.598 7.657 8.378

112

55+122

8.725

112

55+112

9.115

128

113+128

9.355

128

85+128

10.296

126 126

68+98 71+126

10.612 10.935

144

101+144

11.745

0.65

130

69+100

12.135

0.29

142

128+142

12.721

0.47

144

57+69

13.030

0.57

126

97+126

13.231

0.62

144

60+73 87+144

14.747 15.079

0.37 0.72

60+73 60+73 81+101

17.154 18.501 18.751

60+73

19.555

110

x

0.53 0.810 0.51

x x 0.680

0.45 0.520

0.480 0.67 0.95

Molecular chemistry by pyrolysis– GC/MS

229

Table 10.1 (continued ) Score on Factor 1

Code

Name

Ps34

1,6-Anhydro-b-Dglucofuranose Sugar 5 Triterpene

Ps35 Tp

M

Masses

RT (min)

Positive

Negative

Extract

Residue

73+85

21.324

0.68

60+73 69+81

21.463 35.839

0.95

Note: Bold type codes (mean ‘contents’ 40.1%) are included in the reduced data set. Scores on Factor 1 have been given numerically if higher than 0.25 or lower than –0.25. Scores between 0.10 and 0.25 are indicated with an (x), whereas lower scores have been omitted. Codes: Al ¼ alcohols; Ar ¼ aromatics; C ¼ cyclo-pentenes and -hexenes; Ch ¼ poly-aromatic charcoallike compounds; Cp ¼ chlorophyll-like products; F ¼ fatty acids; Lg ¼ lignin compounds; Lp ¼ lipids, sterols; Me ¼ esters; Mk ¼ methylketones; N ¼ nitrogen containing compounds; Ph ¼ phenols; Pr ¼ pristene; Ps ¼ polysaccharide compounds; Tp ¼ terpenes.

Table 10.2. Mean composition and standard deviations (n ¼ 21) for extracts and residues. Percentages are relative to total ion current. Group

Aliphatic biopolymers Alcohols Aromatics Cyclopentenes/hexenes Polyaromatics Aliphatic acids (fatty acids) Lignin Lipids Esters Methyl ketones N-compounds Phenols Polysaccharides

Extracts

Residues

Mean (%)

S.D.

Mean (%)

S.D.

9.0 0.8 4.6 0.8 0.3 1.5 27.0 0.7 1.0 1.3 2.5 20.2 30.2

3.5 0.3 0.9 0.2 0.05 0.8 7.0 0.3 0.6 0.7 0.7 5.4 11.9

11.6 1.3 1.2 0.5 0.1 1.9 8.4 1.0 0.5 3.3 0.5 3.2 66.3

6.5 0.6 0.4 0.2 0.04 1.0 2.5 0.5 0.4 1.7 0.3 1.5 14.4

non-extractable fractions, and with varying contributions of both, would allow a less detailed interpretation. In the extracts, polysaccharides decrease with depth whereas phenols and aliphatics show a slight increase. Samples 05 and 08 have a peak in polysaccharides (also coincident with the NMR results) and methyl esters, accompanied by a low in lignins, phenols and aromatics. Other trends are not clear. Differences between extracts and residues are due to two main factors: humiﬁcation and extractability. The extracted fraction is supposedly more humiﬁed, whereas the residue may also contain fractions

230

P. Buurman et al.

that are less easily extracted by polar liquids such as a NaOH solution. Of the fragments of non-hydrolyzable aliphatic biopolymers, chain lengths of n-C19–C25 dominate, whereas ‘‘free’’ n-alkanes (without the corresponding alkenes) are dominated by n-C31 and C33, the characteristic alkanes present in Calluna ﬂowers and leaves (Nierop et al., 2001). This type of chain-length distribution was also observed by Xie et al. (2003) in a 40 cm core from Bolton Fell Moss (UK), who attributed it to epicuticular waxes of peat-forming plants. Shorter chain n-alkanes homologues (C14–C20) were attributed to a bacterial and/or algae origin (Blumer et al., 1971; Otto et al., 1994). Of the linear alcohols, C20, C22 and C24 are the most abundant ones, typically derived from the suberin present in bark (of the stems) and roots (van Smeerdijk and Boon, 1987; Nierop et al., 2001). The most abundant fatty acids have chain lengths of C16 and C18. Of the iso and anteiso fatty acids, only the iso-C15 acid is present in identiﬁable amounts. Among the lignin fragments, guaiacol, 4-methylguaiacol, 4-ethylguaiacol, 4-vinylguaiacol, 4-acetylguaiacol, 4-vinylphenol, syringol and syringic acid dominate whereas the fragments with C3 side chains are relatively scarce and predominantly oxidized. Of the methyl ketones, C25, C27 and C29 chain lengths dominate. They are most likely derived from Calluna bark and roots (van Smeerdijk and Boon, 1987; Nierop et al., 2001). The most abundant nitrogen-containing compounds are pyridine and indole. Of the phenol group, phenol is especially abundant in the extracts (8.572.5%) versus 0.9870.4 in the residues, and also 3/4-methylphenol and 3-methoxy-1, 2-benzenediol are relatively abundant in the extracts. In the extracts, polysaccharide fragments are dominated by acetic acid, (2 H)-furan-3-one, 2-furaldehyde, 5-methyl-2-furaldehyde, 4-hydroxy-5, 6-dihydro(2 H)-pyranose, levogalactosan, levomannosan and levoglucosan. In the residues, levoglucosan fully dominates (43.675.9%) whereas levoglucosenone, 1,4-dideoxyD-glycero-hex-1-enopyranose-3-ulose and 1,6-anhydro-b-D-glucofuranose play a minor role. These differences suggest that residues are richer in Calluna wood remnants, whereas the extracts contain relatively more moss and/or microbial derived polysaccharides (van der Heijden, 1994; Nierop et al., 2001). In the principal component analysis, extracts and residues have ﬁrst been analyzed together. Four factors explained 66.7% of the variation in all 225 chemical fragments. The explained variation by four factors increased to 75.4% in the reduced data set of 141 compounds, but plots of factor loadings and sample scores hardly changed. We will use the reduced data set in the following discussions. Furthermore, we will limit the discussions to Factor1–Factor2 (F1–F2) space, which explained 65.2% of all variation in the reduced data set. When the samples are projected in F1–F2 space (Fig. 10.3), extracts and residues plot in two different areas. Residues have negative scores on F1, and extracts have positive scores. Factor loadings (Table 10.1) indicate that negative scores are related to a relatively high abundance of aliphatics with chain lengths nX17, all methyl ketones, fatty acids with n416 and lipids. Positive scores on F1 are associated with relatively high contents of lignin fragments, nitrogen-containing compounds, phenols, short-chain alkanes and alkenes (n ¼ 10–15) and C14 fatty acid. Factor 2 does not differentiate between extracts and residues.

Molecular chemistry by pyrolysis– GC/MS

231

Figure 10.3. Scores of extract and residue samples in F1–F2 factor space. The dashed lines connect upper (01) and lower (79) sample.

A detailed analysis of the factor loadings of the various groups gives further insight into the differences between residues and extracts and in the processes of decomposition. The n-alkanes, n-alkenes and n-2-methylketones (Fig. 10.4) are mostly distributed along the upper periphery of the F1–F2 space. The n-C28–n-C33 alkanes/ alkenes and the Cn2 methyl ketones are found in the lower left corner. These are compounds found in pyrolysates of (fresh) Calluna tissues, and therefore this corner represents relatively unaltered plant material. As shown in Figure 10.5, from this corner there is an almost regular decrease of alkane, alkene and methylketone chain length in the direction of the arrow and toward the right-hand part of the diagram, where the extracts plot in Figure 10.3. It is tempting to interpret this trend as a decay process, in which case shorter chain lengths need not be attributed to bacterial or algal sources. Lehtonen and Ketola (1990) also found an increase in the abundance of short chain methyl ketones homologues with increasing degree of humiﬁcation, whereas Xie et al. (2003) explained the lack of short chain homologues in the peat core they analyzed by a lower state of humiﬁcation due to shallow burial (o40 cm). Some components, notably n-C27 alkane and alkene and n-C12:1–n-C15:1 alkenes do not conform to the trend. The n-C27 alkanes have different possible origins, whereas the short-chain alkenes occur in very small amounts only. Also branched alcohols or alkenes and branched methyl ketones are found more in the periphery of the general trend. Apart from Calluna derived, methyl ketones are either oxidation products of free alkanes or secondary pyrolysis products of fatty acids in the presence of metal oxides (Raven et al., 1997). Because the samples have very low initial ash contents (o1%) and were further de-ashed by HF treatment, the second option is unlikely. Long-chain odd methyl ketones (C29 and C31) are sometimes associated with

232

P. Buurman et al.

Figure 10.4. Factor loadings (reduced set) in F1–F2 space. Chemical groups that are not indicated are more scattered.

Figure 10.5. Loadings of n-Cx alkanes and alkenes in F1–F2 factor space. Reduced data set, extracts and residues. The arrow indicates a regular decrease in chain length. At the top of the diagram, the components are closer together than the labels indicate.

Molecular chemistry by pyrolysis– GC/MS

233

microbial oxidation of n-alkanes (Amble`s et al., 1993; Xie et al., 2003). The latter authors also indicate that the oxidative decarboxylation of even n-fatty acids (C24 and C26) would yield C23 and C25 methyl ketones, which are not abundant in their peat proﬁle and thus they conclude that n-fatty acids are not main precursors of methyl ketones in Bolton Fell Moss. The fact that C31–C24 methylketones in the PVO project exactly in the trend of alkane/alkene chain lengths strongly suggests that they are mainly plant-derived, as observed for Calluna roots and bark (van Smeerdijk and Boon, 1987; Nierop et al., 2001). Lignin moieties are strongly clustered in the right-hand side of the diagram (Fig. 10.4), with the syringol fragments concentrated in the top of the cluster (not visible in the ﬁgure). Only the unoxidized guaiacols Lg9 (4-(1-propenyl) guaiacol), Lg13 (cis 4-(2-propenyl) guaiacol) and Lg15 (trans 4-(2-propenyl)guaiacol) plot outside the cluster, to the left. The separation of syringol and guaiacol moieties in factor space may indicate preferential decay of syringols, as is documented for anaerobic environments (of peatiﬁed Calluna woods and bark) (van Smeerdijk and Boon, 1987; Nierop et al., 2001). Van der Heijden and Boon (1994) also report a gradual decrease in the syringyl to guaiacyl ratio with peatiﬁcation, which they consider to be related to the removal of syringyl-rich secondary cell wall material and the retention of guaiacyl-rich compound middle lamellae during plant decay. All polysaccharide-derived pyrolysis fragments have negative scores on F2 (Fig. 10.4), whereas aliphatics have a predominantly positive score. The projection of a sample on F2 is therefore largely determined by relative amounts of aliphatics and polysaccharides, which reﬂects the general trend with depth. In both residues and extracts, top samples 01, 05 and 08 are situated in the lower part of the diagram, and the deeper samples 56, 77, 79 in the upper half (Fig. 10.3). The other samples do not show a consistent behavior, which implies that deviations from the depth trend are encountered. Polysaccharides have both positive and negative scores on F1. Strong positive scores are found for 11 of the compounds (Table 10.1), whereas 14 others have strong negative scores. Among the second group is levoglucosan, which represents fresh organic matter. Fragments with strong positive scores (and coinciding with the cluster of extracts in Fig. 10.3) are partially polysaccharide degradation products, partially compounds that should be attributed to microbial input. By analyzing the C isotope composition of carbohydrates other investigators also reached the conclusion that there is a production of microbial carbohydrates, since some of these (particularly xylose, glucose and galactose) become more depleted in 13 C during peat formation, whereas others (mannose, rhamnose, arabinose and cellulose) showed a fairly constant isotopic ratio (Macko et al., 1991; Kracht and Gleixner, 2000). Because the main aim of this research is not the distinction between extractable material and residue, but to ﬁnd out whether deviations on the main chemical trend can be explained by detailed chemistry, we will further consider the non-extractable fractions, as these might most closely reﬂect plant input chemistry. In principal component analysis of the selected set of 141 chemical fragments for the residue samples, four factors explain 71.5% of the variation of all fragments. Factors 1 and 2 together explain 53.2%. The projection of all residue samples in

234

P. Buurman et al.

Figure 10.6. Scores of residue samples in F1–F2 factor space. The arrow connecting samples 01 and 79 approximately represents the main decomposition trend. Overlapping sample numbers have been slightly displaced.

F1–F2 space is given in Figure 10.6, showing that sample 01 and 79 project at opposite sites in factor space. If the trajectory between these samples represents the general trend of peat decomposition, some samples show signiﬁcant deviation from this trend. To understand the differences between the samples, we need to analyze the plot of factor loadings. In Figure 10.7, the approximate positions of clusters of fragments with the same origin or signature have been indicated. The n-C27 alkane and alkene, and the n-C13:1 and n-C12:1 alkenes plot in the center of the diagram, outside the main aliphatic domains. Fatty acids plot in the lower half of the diagram, with n-C18, 20, 22, 24 (predominantly plant-derived) at the right-hand side and n-C14, 16, 18:1 (both plant and microbially derived) and a methyl ester on the left. Aromatics and phenols group largely with lignin. Nitrogen compounds are scattered over the center of the diagram. As it was expected, the main trend is determined by relative amounts of polysaccharides and non-hydrolyzable aliphatic biopolymers (indicated by straight alkanes and alkenes as pyrolysis products), whereas deviations from the trend appear to result from variations in lignin contents and long-chain aliphatics and methyl ketones. Compared to the other samples, the bottom samples 77 and 79 have more alkanes and alkenes (of all chain lengths; n-C31 and n-C33 alkanes excepted), more n-C25–29 methylketones, prist-1-ene, and various minor lipids that appear to increase with depth. The n-C18,20,22,24 fatty acids are relatively higher than in the other samples, whereas polysaccharide moieties are signiﬁcantly less abundant. The extreme deviations from the trend, as illustrated by samples 05, 08 and 39 in the lower left corner, are mainly caused by higher contents in 05, 08 and 39 of

Molecular chemistry by pyrolysis– GC/MS

235

Figure 10.7. Factor loadings in F1–F2 space of the same residue samples used in Figure 10.6. More scattered chemical groups have not been indicated.

n-C31 and n-C33 alkanes n-C22 alcohol n-C16 and n-C18:1 fatty acids n-C7 methyl ester n-C25,27,29 methyl ketones

(Calluna leaves and ﬂowers), (Calluna bark and roots), (various sources), (source unknown), (Calluna bark and roots)

and the polysaccharide-derived pyrolysis compounds levogalactosan, levomannosan, levoglucosan, levoglucosenone and 1.4:3.6-dianhydro-alpha-D-glucose (intact plant material, mosses, Calluna). This demonstrates that the deviating chemistry of samples 05, 08 and 39 is mainly due to relative abundance of Calluna components. Pancost et al. (2002) also suggest that the predominance of high molecular weight n-alkanes can serve as a qualitative record of the shift from Sphagnum-dominated peat to Ericaceae-dominanted (Erica and Calluna) peat. Nevertheless, samples 05, 08 and 39 have relatively low contents of lignin moieties, phenols and toluene; whereas samples 53 and 59 have lower contents of the listed compounds and higher contents of the aromatic ones. The question arises whether the deviating position of samples 53 and 59 is also due to speciﬁc vegetation. Using paleovegetation (pollen and non-pollen palynomorph records) and chemical data Martı´ nez Cortizas et al. (2004) have recently made a reconstruction of wet/dry climate cycles in a peat record from an ombrotrophic bog at Pena da Cadela (PDC), located less than 5 km away from the PVO. From this climatic record and the ages of the PVO samples (Martı´ nez Cortizas et al., 2002) analyzed here, we can infer that samples 02, 28, 32, 41, 53 and 59 belong to dry periods, whereas samples 05, 08 and

236

P. Buurman et al.

39 belong to wet periods. This suggests that the deviations from the main decomposition trend are related to changes in the hydrology of the bog as a result of variations in effective precipitation. The wet period appears to coincide with a relative abundance of Calluna markers. In the PDC record, wet phases are represented by a high abundance of sedges (Cyperaceae) whereas woody plants (Calluna and Erica) are more abundant during dry phases. This would imply that at the PVO, a higher lignin content during the dry phases should be due to woody plants other than Calluna, or that the higher lignin content is due to selective degradation of other compounds (peat degradation decreases during wet phases, so the increase in Calluna can also be the result of a greater preservation of plant remains due to reduced degradation). More information is obtained through a detailed look at the lignin moieties. In a principal components analysis of lignin fragments alone (not shown), F1 explains 66.3% of the variance and F1+F2 together 77.1%. On F1, all samples plot between 05 and 53–59. Factor loadings indicate that samples 53 and 59 plot in the corner with relatively large amounts of oxygenated lignin fragments and are more oxidized than the remainder. These samples are therefore set apart by both higher lignin contents and stronger decay of the lignin. The lignin factor loadings do not show a separation of guaiacol and syringol fragments as in the factor loadings of residues and extracts together, so selective decay of these groups is not found in the residues, which is in line with the idea that the residues represent relatively intact plant material. The higher intensity of lignin degradation in samples 53 and 59 is consistent with accelerated decomposition during dry conditions. Also the distribution of polysaccharide compounds in F1–F2 factor space may indicate that there is a difference in oxidation state between these samples. The polysaccharide-derived pyrolysis products levoglucosan, -mannosan, -galactosan and ‘sugar 5’ plot on the left of the factor loadings. These are components derived from rather undecomposed plant structures. On the other hand, the furans and acetic acid, which are smaller pyrolysis products of polysaccharides, plot at the top, where samples 53 and 59 are encountered. Because a relatively high amount of such small pyrolysis products appears to indicate more strongly decomposed polysaccharides (Buurman et al., 2005), decomposition in samples 53 and 59 appears to be stronger than in samples 05, 08 and 39. The present data do not allow more detailed conclusions. The combined evidence suggests that the samples that represent dry periods (53 and 59, and to a lesser extent also 32, 50, 28 and 62) have more decayed organic matter. Higher lignin contents might be due to tree vegetation instead of Ericaceae, but there is no evidence of trees in the records of blanket bogs from NW Spain. Kuder and Kluge (1998) reported a number of speciﬁc chemical changes upon oxidation of peat during a dry phase. Although their conclusions are based on remnants of the sedge (Eriophorum vaginatum), they corroborate our own ﬁndings. They found that shortening of alkyl side chains of methoxyphenols and loss of pentosan polysaccharides are among the main characteristics of aerobic degradation. Lehtonen and Ketola (1993) reported changes in lipid chemistry upon oxidation, but we have not obtained speciﬁc information on this fraction.

Molecular chemistry by pyrolysis– GC/MS

237

The n-alkane record of the PVO proﬁle does not provide clear evidence of the presence of Sphagnum moss. This is probably due to the fact that Sphagnum was never a dominant component of the local vegetation. Also at present, Ericaceae are among the main components.

Conclusions (1) Molecular chemistry reveals signiﬁcant differences between extracted and non-extractable (residue) peat. In the extracts, lignin and other aromatic fragments, nitrogen-containing moieties are more abundant whereas aliphatics, methyl ketones and lipids are more abundant in the non-extractable fraction. This is not astonishing, since partly oxidized lignin and proteins are hydrophilic and extractable with NaOH, whereas for the extraction of lipids one needs organic solvents. (2) Short-chain straight aliphatics are more abundant in the extractable fraction, and there appears to be a regular decay series of alkanes and alkenes from n-C28,28:1 to n-C10. One branched alkane that is predominantly present in the extractable fraction is possibly of microbial origin. Some polysaccharide moieties are speciﬁcally associated with the extractable fraction and some with the residues. This probably reﬂects a predominantly microbial origin for the ﬁrst and predominantly plantderived for the second, because the plant-derived polysaccharides may still be present in larger molecular units. (3) In the non-extractable fraction, the general trend of increasing alkyl-C contents with depth, as found in 13C-CPMAS-NMR, is encountered. The general increase in aliphatics is accompanied by an increase in lipids in the bottom part of the proﬁle. Long-chain alkenes or alkanes are often accumulating under anaerobic condition due to the fact that the ﬁrst step includes the attack of an oxidized group. Deviations from the general chemical trend indicate that a wet climate was accompanied by a relative abundance Calluna markers and relatively unoxidized peat. (4) Dry periods appear to be accompanied by a higher oxidation state of the lignin moieties. Because this higher oxidation state appears to be accompanied by relatively larger lignin contents, it is likely that such changes reﬂect both a more woody vegetation and a drier bog surface.

Future research The general chemical trends described here are insufﬁciently detailed to track further vegetation changes, but it is likely that a more detailed analysis of n-alkanes and other lipids such as terpenoids could offer better clues. Various authors have proposed a number of speciﬁc biomarkers to trace vegetation changes through time in mire ecosystems, although a large uncertainty has arisen due to the multiple origin and lack of speciﬁcity of most biomolecules in peat organic matter (Ficken et al., 1998). In our opinion, multivariate tests – such as PCA analysis with reverse modeling – applied to a large set of organic compounds offer a greater potential to discriminate between the different sources of variation accounting for peat organic

238

P. Buurman et al.

chemistry (vegetation changes, climate changes, decomposition trends, microbial modiﬁcations, differences in compounds resistance to degradation). Establishing correlations between peat molecular chemistry and botanical composition should be a ﬁrst step in this direction. References Amble`s, A., Jambu, P., Jacquesy, E., et al., 1993. Changes in the ketone portion of lipidic components during the decomposition of plant debris in a hydromorphic forest-podzol. Soil Sci. 156, 49–56. Armanios, C., Alexander, R., Kagi, R.I., Skelton, B.W., and White, A.H., 1995. Occurrence of 28-nor18alpha-oleanane in the hydrous pyrolysate of a lignite. Org. Geochem. 23, 21–27. Avsejs, L.A., Nott, C.J., Xie, S., et al., 2002. 5-n-alkylresorcinols as biomarkers of sedges in an ombrotrophic peat section. Org. Geochem. 33, 861–867. Bergman, I., Lundberg, P., Preston, C.M., and Nilsson, M., 2000. Degradation of 13C-U-glucose in Sphagnum majus litter: responses to redox, pH, and temperature. Soil Sci. Soc. Am. J. 64, 1368–1381. Beyer, L., Blume, H.P., Sorge, C., et al., 1997. Humus composition and transformations in a pergelic cryohemist of coastal Antarctica. Arctic Alpine Res. 29, 358–365. Blumer, M., Guillard, R.L., and Chase, T., 1971. Hydrocarbons of marine phytoplankton. Mar. Biol. 8, 183–189. Boon, J.J., Dupont, L., and de Leeuw, J.W., 1986. Characterization of a peat bog proﬁle by Curie point pyrolysis–mass spectrometry combined with multivariate analysis and by pyrolysis–gas chromatography–mass spectrometry. In: Fuchsman, C.H. (Ed.), Peat and Water. Elsevier, Amsterdam, pp. 215–239. Boon, J.J., Dupont, L., van der Hammen, T., and de Leeuw, J.W., 1985. Characterization of a peat bog proﬁle by Curie point pyrolysis–mass spectrometry and high-resolution gas chromatography mass spectrometry. Int. Peat Research Conference, Riviere du Loup, Canada, 16–19 June. Buurman, P., van Bergen, P.F., Jongmans, A.G., et al., 2005. Spatial and temporal variation in podzol organic matter – pyrolysis–GC/MS and micromorphology. Eur. J. Soil Sci. 56, 235–279. Farrimond, P. and Flanagan, R.L., 1996. Lipid stratigraphy of a Flandrian peat bed (Northumberland, UK): comparison with the pollen record. Holocene 6, 69–74. Ficken, K.J., Barber, K.E., and Eglinton, G., 1998. Lipid biomarker, d13C, and plant macrofosil stratigraphy of a Scottish montane peat bog over the last two millenia. Org. Geochem. 28, 217–237. Fox, C.A., Preston, C.M., and Fyfe, C.A., 1994. Micromorphological and 13C NMR characterization of a humi, lignic, and histic folisol from British Columbia. Can. J. Soil Sci. 74, 1–15. Gleixner, G.M. and Kracht, O., 2001. Molecular processes in the humiﬁcation of sphagnum moss in a peat proﬁle. In: Swift, R.S. and Spark, K.M. (Eds.), Understanding and Managing Organic Matter in Soils, Sediments and Waters. Proceedings of 9th International Conference of the International Humic Substances Society, Adelaide, September 1998. International Humic Substances Society, St. Paul, pp. 195–201. Halma, G., van Dam, D., Haverkamp, J., et al., 1984. Characterization of an oligotrophic–eutrophic peat sequence by pyrolysis–mass spectrometry and conventional analysis methods. J. Anal. Appl. Pyrol. 7, 167–183. Hammond, T.E., Cory, D.G., Ritchey, W.M., and Morita, H., 1985. High-resolution solid state 13C NMR of Canadian peats. Fuel 64, 1687–1695. Karunen, P. and Ekman, R., 1982. Age-dependent content of polymerized lipids in Sphagnum fuscum. Physiol. Plant. 54, 162–166. Kondo, R. and Tsutsuki, K., 2001. Organic indices for the decomposition of peat in a high-moor peatland under severe drying and vegetation change. In: Swift, R.S. and Spark, K.M. (Eds), Understanding and Managing Organic Matter in Soils, Sediments and Waters. Proceedings of 9th International Conference of the International Humic Substances Society, Adelaide, September 1998. International Humic Substances Society, St. Paul, pp. 215–220. Kracht, O. and Gleixner, G., 2000. 13C isotope analysis of pyrolisis products from Sphagnum peat and dissolved organic matter from bog water. Org. Geochem. 31, 645–654.

Molecular chemistry by pyrolysis– GC/MS

239

Krosshavn, M., Southon, T.E., and Steinnes, E., 1992. The inﬂuence of vegetational origin and degree of humiﬁcation of organic soils on their chemical composition, determined by solid-state 13C NMR. J. Soil Sci. 43, 485–493. Kuder, T. and Kruge, M.A., 1998. Preservation of biomolecules in sub-fossil plants from raised peat bogs – a potential paleoenvironmental proxy. Org. Geochem. 29, 1355–1368. Kuder, T., Kruge, M.A., Shearer, J.C., and Miller, S.L., 1998. Environmental and botanical controls on peatiﬁcation – a comparative study of two New Zealand bogs using py–GC/MS, petrography and fungal analysis. Int. J. Coal Geol. 37, 3–27. Lehtonen, K. and Ketola, M., 1990. Occurrence of long-chained acyclic methyl ketones in Spahgnum and Carex peats of various degrees of humiﬁcation. Org. Geochem. 15, 275–280. Lehtonen, K. and Ketola, M., 1993. Solvent-extractable lipids of Sphagnum, Carex, Bryales and CarexBryales peats: content and compositional features vs. peat humiﬁcation. Org. Geochem. 20, 363–380. Lehtonen, K., Ketola, M., and Pihlaja, K., 1991. Water-soluble lipids in Carex and Sphagnum peats. Int. J. Env. Anal. Chem. 43, 235–244. Macko, S.A., Engel, M.H., Hartley, H., et al., 1991. Isotopic compositions of individual carbohydrates as indicators of early diagenesis of organic matter in peat. Chem. Geol. 93, 147–161. Martı´ nez Cortizas, A., Garcı´ a-Rodeja, E., Pontevedra-Pombal, X., et al., 2002. Atmospheric Pb deposition in Spain during the last 4600 years recorded by two ombrotrophic peat bogs and implications for the use of peat as archive. Sci. Tot. Environ. 292, 19–32. Martı´ nez Cortizas, A., Mighall, T., and Biester, H., 2004. Reconstructing Holocene paleoclimate using pollen, non-pollen palynomorphs and geochemical data from the ombrotrophic peat bog of Pena da Cadela, NW Spain. Pollen 14, 274. Martı´ nez Cortizas, A., Pontevedra-Pombal, X., No´voa Mun˜oz, J.C., et al., 1997. Four thousand years of atmospheric Pb, Cd, and Zn deposition recorded by the ombrotrophic peat bog of Penido Vello (Northwestern Spain). Water Air Soil Pollut. 100, 387–403. Nierop, K.G.J., van Lagen, B., and Buurman, P., 2001. Composition of plant tissues and soil organic matter in the ﬁrst stages of a vegetation succession. Geoderma 100, 1–24. Nott, C.J., Xie, S., Avsejs, L.A., et al., 2000. n-Alkane distributions in ombrotrophic mires as indicators of vegetation change related to climatic variation. Org. Geochem. 31, 231–235. Otto, A., Walther, H., and Pu¨ttman, W., 1994. Molecular composition of a leaf- and root-bearing Oligocene oxbow lake clay in the Weisselster Basin, Germany. Org. Geochem. 22, 275–286. Pancost, R.D., Baas, M., van Geel, B., and Sinninghe Damste´, J.S., 2002. Biomarkers as proxies for plant inputs to peats: an example from a sub-boreal ombrotrophic bog. Org. Geochem. 33, 675–690. Pontevedra-Pombal, X., 2002. Turberas de Montan˜a de Galicia: Ge´nesis, Propiedades y su Aplicacio´n como Registros Ambientales Geoquı´ micos. Tesis Doctoral. Depto. Edafologı´ a y Quı´ mica Agrı´ cola, Fac. Biologı´ a, Univ. Santiago de Compostela. Ed. Universidade de Santiago de Compostela. Pontevedra-Pombal, X., Martı´ nez Cortizas, A., Garcı´ a-Rodeja, E., et al., 2001. Composicio´n y trasformacio´n de la material orga´nica en un histosol de la Serra do Xistral (norte de Galicia) mediante CPMAS-13C-NMR. Edafologı´ a 8, 67–79. Preston, C.M., Axelson, D.E., Levesque, M., et al., 1989. Carbon-13 NMR and chemical characterization of particle-size separates of peats differing in degree of decomposition. Org. Geochem. 14, 393–403. Preston, C.M., Shipitalo, S.E., Dudley, R.L., et al., 1987. Comparison of 13C CPMAS NMR and chemical techniques for measuring the degree of decomposition in virgin and cultivated peat proﬁles. Can. J. Soil Sci. 67, 187–198. Raven, A.M., van Bergen, P.F., Stott, A.W., et al., 1997. Formation of long-chain ketones in archaeological pottery vessels by pyrolysis of acyl lipids. J. Anal. Appl. Pyrol. 40/41, 67–285. Stout, S.A., Boon, J.J., and Spackman, W., 1988. Molecular aspects of the peatiﬁcation and early coaliﬁcation of angiosperm and gymnosperm woods. Geochim. Cosmochim. Acta 52, 405–414. van Bergen, P.F., Bull, I.D., Poulton, P.R., and Evershed, R.P., 1997. Organic geochemical studies of soils from the Rothamsted Classical Experiments-I. Total lipids, solvent insoluble residues and humic acids from Broadbalk Wilderness. Org. Geochem. 26, 117–135. van der Heijden, E., 1994. A aombined anatomical and pyrolysis mass spectrometric study of peatiﬁed plant tissues. Ph.D. Thesis, University of Amsterdam, 1–157. van der Heijden, E. and Boon, J.J., 1994. A combined pyrolysis–mass spectrometric and light microscopic study of peatiﬁed Calluna wood isolated from raised bog peat deposits. Org. Geochem. 22, 903–919.

240

P. Buurman et al.

van Smeerdijk, D.G. and Boon, J.J., 1987. Characterisation of subfossil Sphagnum leaves, rootlets of Eracaceae and their peat by pyrolysis-high-resolution gas chromatography–mass spectrometry. J. Anal. Appl. Pyrol. 11, 377–402. Verhoeven, J.T.A. and Liefveld, W.M., 1997. The ecological signiﬁcance of organochemical compounds in Sphagnum. Acta Bot. Neerl. 46, 117–130. Xie, S., Nott, C.J., Avsejs, L.A., et al., 2003. Molecular and isotopic stratigraphy in an ombrotrophic mire for paleoclimate reconstruction. Geochim. Cosmochim. Acta 68, 2849–2862. Young, L.Y. and Frazer, A.C., 1987. The fate of lignin and lignin-derived compounds in aeaerobic environments. Geomicrobiol. J. 5, 261–293.

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

241

Chapter 11

Mineral matter, major elements, and trace elements in raised bog peat: a case study from southern Sweden, Ireland and Tierra del Fuego, south Argentina L.G. Franze´n

Introduction Peatlands are giant reservoirs in which signiﬁcant amounts of carbon have accumulated since the last glacial retreat. The greatest concentration of peatlands and the highest rate of peat formation are found in the cool, humid climates of the Northern Hemisphere. More than 95% of the total peat reserves of the world is concentrated in the temperate belt of the Northern Hemisphere; the rest is found in the humid tropics. The total peatland area in the world has been estimated at 5 106 km2 or about 3.5% of the Earth’s land surface. In Sweden, wetlands cover 93,000 km2 of which about 2/3rd is regarded as peatlands (Franze´n, 1985). Estimates of the amount of carbon stored in the peatlands of the world ranges between 120 and 400 Gt (Ajtay et al., 1979; Sjo¨rs, 1980, 1982; Adams et al., 1990; Franze´n, 1992, 1994; Franze´n et al., 1996). In a recent report Smith et al. (2004) estimated that peat makes up about 26% of all terrestrial carbon accumulated since the Last Glacial Maximum. Besides being an important factor in the global carbon cycle, peatlands are important archives of atmospheric fallout since peat formation began. The dry peat substance of raised bogs consists of combustible organic material and mineral matter deposited during peat formation. Since these peatlands are not associated with ﬂowing water, particulate matter deposited on the plant surfaces remains approximately in situ to be incorporated in the peat as successive layers of vegetation accumulate. Thus, a peat proﬁle contains a record of particulate deposition from the atmosphere. After combustion, the remaining peat ash is generally composed of about 90% soluble salts, dominantly Ca (Fredriksson, 1996). The rest is insoluble particles, mainly microscopic mineral grains (mostly quartz) deposited as condensation nuclei, volcanic ash (Persson, 1971; Pilcher and Hall, 1992, 1996) and, in most layers, phytoliths from peat-forming plants. Another source of minerals in peat is the continuous rain of extraterrestrial dust. In natural ombrotrophic peat the ash concentration is about 1% or even less, whereas in fen peats the ash content can reach 15% or even more (Fredriksson, 1996). ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09011-0

242

L.G. Franze´n

Methods Fourteen raised bogs and one blanket bog in southern Sweden and Ireland and ﬁve mires in Tierra del Fuego of southern Argentina have been investigated to determine rate of peat accumulation and geochemistry. Here some aspects are being analyzed of peat geochemistry and mineral content of the ash (Figs. 11.1a, b, c). Sampling Samples from the different bogs were taken with a 50 mm diameter Byelorussian peat corer. Sampling was always performed in the highest central parts of the peatland to get the longest record possible. Normally, bogs start to develop in a moist depression expanding laterally as they thicken (Franze´n et al., 1996). Samples were cut out in 5 cm levels, allowing for a temporal resolution of approximately 50 years (assuming a peat accumulation rate of about 1 mm/yr). Two separate cores (A and B), situated less than 25 cm from each other were taken from each bog. The A-sample was used for particle studies, and the B-sample for geochemical investigations. To avoid contamination and oxidation, the samples were immediately placed in zipper bags, which were kept cool before lab preparation. Laboratory analyses The laboratory work followed two different preparation paths. (1) The A-samples were dissolved in a NaOH solution. After this, a long sieving and decantation process, ended up with a small portion of mineral grains and fragments of charcoal. The mineral samples were placed under lid in 60 mm glass Petri dishes. The material in the Petri dishes was ﬁrst scanned under a polarizing microscope to determine the frequency of volcanic glass shards and phytoliths. The second step was to scan them under a light stereomicroscope. Particles of special interest were picked out with a thin insect needle and placed on scanning electron microscope (SEM) aluminum stubs, using double-sided carbon tape to keep them in place. These particles were studied in detail and analyzed with a scanning electron microscope – energy dispersive spectroscope (SEM–EDX). Images were taken, stored, and processed digitally. Whereas the spherules (spherical-shaped particles) reported from deep sea sediments and the Antarctic ice cap are rather large, with sizes ranging from 0.1 to 1 mm, the size of the spherules found in peat varies from less than 1 micron to a maximum of about 100 microns. Smaller sizes than 1 micron are impossible to detect by the magniﬁcation used under the light microscope, but even under the SEM, no spherule smaller than about 1 micron has been measured in the samples studied. The apparent absence of spherules smaller than 1 micron may be an artifact due to sample treatment; that is, not enough time was allowed for the smaller particles to settle out of suspension during decantation. Similarly, the

Mineral matter, major elements, and trace elements in raised bog peat

243

Figure 11.1. Sampling sites: a. Southern Sweden: Ga¨llseredsmossen (1), Lyngmossen (2), Store mosse O¨xaba¨ck (3), Do¨mle mosse (4), Hjortemossen (5), Komosse (6), Konungso¨mossen (7), Fallamossen (8), Vildmossen (9), Torstama˚la ﬂy (10) and Store mosse Kallinge (11); b. Ireland: Sluggan bog (1), Fallahogy bog (2), Owenduff bog (3) and Clara bog (4); c. Tierra del Fuego, S. Argentina: Punta Moat bog (1), Harberton bog (2), Las Cotorras mire (3), Tierra Australis bog (4), and Lago Fagnano/Tolhuin mire (5).

procedure in extracting some minerals from the peats is a difﬁcult one and operator errors can occur. The extraction of carbonate crystals, for example, is a delicate matter. Whenever a peat sample is taken out of its anaerobic environment and is exposed to the air, the crystals are rapidly dissolved in the acid peat

244

L.G. Franze´n

water. The preservation of the crystals was obtained by adding a weak NaOH solution to the peat sample in the ﬁeld, immediately after sampling. (2) The B-samples were dried and combusted at 550 1C. Ash content was calculated as the residual over the dry weight. The peat ashes were dissolved in a 50% aqua regia solution and stored in polypropylene bottles. A small portion of the dried B-samples was kept for possible 14C datings. The solutions were analyzed with an inductively coupled plasma mass spectrometer (ICP-MS), and the concentrations of 68 major and trace elements were measured. Forty-ﬁve elements measured in ombrotrophic and minerotrophic peats from Sweden and Ireland are reported in Table 11.1. With this background information, an attempt was made to evaluate the main sources of mineral matter and associated elements, and the inﬂux variations during the last 10,000 years.

Results and discussion In the following presentation, it is important to understand the fundamental background to atmospheric dust. It may be hard to imagine how clean the atmospheric environment was before the increase of the human population. After the ice age virtually every spot of bare soil was covered under the competition of plants and the only places at the local and regional environmental scales where ﬁne mineral grains could be lifted up by winds were sea and lake shores, river beds and deltas, and bare rock. Even after extensive forest ﬁres, the bare soil windows that might have been formed, were rapidly closed by pioneering plants species. Consequently, the amount of dust in the modern atmosphere is greatly inﬂuenced by human activities. In general, the natural fallout on the topmost uncompacted (about 20–30 cm) peat (equivalent of about 100 years of accumulation) may be obscured by worldwide input from activities such as agriculture, auto and airplane trafﬁc, industry, waste disposal, and nuclear bomb tests. The amount of atmospheric dust suspended and fallout on land and in the oceans is largely controlled by climate and, as a result, feedbacks between dust and climate are possible (Arimoto, 2001). The high ash content of surface peats caused by atmospheric deposition is not only found in populated areas but is a global phenomenon. I have observed the same pattern in central western Norway, Ireland, and in the remote Tunguska region of the central Siberian high plateau. Chague-Goff and Fyfe (1997) reported high ash concentrations at the base and top proﬁles of plateau bogs of the Canadian Subarctic, where the high surface concentrations were ascribed to anthropogenic sources. The high inﬂux of modern dusts has also been reported from glaciers of the Alps (Preunkert et al., 2001), Antarctic (Murozumi et al., 1969; Legrand et al., 2004) and Arctic (Zdanowicz et al., 1998; Bory et al., 2002). Whereas the natural background deposition, based on the ash content in peat in areas far from the sea shore, could be calculated at about 0.05–0.1 mm per 1000 years in pre-industrial times, the same amount can be recorded within months, or even weeks today, in built-up areas with heavy trafﬁc. Before the signiﬁcant increase in human population, most material to the Scandinavian peninsula, for example, came from the regularly re-occurring westerly

Mineral matter, major elements, and trace elements in raised bog peat

245

Table 11.1. Relative concentration of major and trace elements in peats from 11 southern Swedish and 4 Irish raised bogs. The third column shows the ratios of the relative concentrations of elements of the minerotrophic and ombrotrophic peats. The values have been shown in order of relative concentration. Si and Al were not determined due to experimental technique that does not provide a reliable measurement for these elements. Ombrotrophic peat

Minerotrophic peat

n ¼ 1264 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

Ca Mg Fe Na P K S Ti Mn Sr As Cr Ba Zn V Cu B Pb Li Ni Ga Ge Ce Se Rb Zr La Mo Nd Sc Co W I Y Sn Pr Sb Nb Sm Gd

M/O

n ¼ 453 41.4 26.9 15.1 5.4 4.9 2 9.7 7.3 4.2 4 3.8 2.9 2.5 2.3 1 872.4 768.8 690.1 415.7 317.2 207.5 173 143.7 131.4 123 114.6 73.9 69.4 66.1 51 45.6 39.6 35.5 34.6 34.1 17.3 15.3 13.6 13.5 12.8

%

%

ppm

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

Ca Fe Mg P Na K Ti S Mn Ba Sr Zn Ce Pb As Cr La Nd Cu V Y B Rb Ni Ga Pr Sm Li Zr Ge Gd Sc Dy Co Sn Se Th Mo Er U

39.1 26 12.9 6.9 3.9 3.8 1.6 1.1 8.7 6.6 4.1 3.8 3.3 2.7 2.3 2 1.8 1.7 1.3 1.3 784 715.5 530.5 500.1 488.8 460.8 312.1 304.9 272.5 268.4 254.1 229.3 161.3 159.8 136.5 119.9 118.1 100 93.8 83.3

%

%

ppm

Pr Nd La Ho Sm Er Tm Ce Dy Y Tb Yb Gd Lu Eu Th U Tl Cs Be Sc Rb Sb Sn Pb Co Cd Pd Bi Hf Ba Zr Ga Ta Ti Mn K Os Fe Zn

26.7 25.1 25 23.6 23.1 23 23 23 22.8 22.7 22.6 21.5 19.9 19.9 16.5 9.8 9 6.4 5 4.7 4.5 4.3 4 4 3.9 3.5 3.5 3.2 3.1 3 2.6 2.4 2.4 2.2 2.2 2.1 1.9 1.7 1.7 1.7

L.G. Franze´n

246 Table 11.1 (continued ) Ombrotrophic peat

Minerotrophic peat

n ¼ 1264 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67

Th Cd U Te Dy Be Ag Er Cs Yb Pd Eu Bi Tl Au Ho Tb Pt Hf Lu Tm Hg Os Rh Ta Ru Ir

M/O

n ¼ 453 12 9.4 9.2 7.8 7.1 6.6 5.5 4.1 3.8 3.7 3.7 3.3 3.1 2.5 2.5 1.4 1.4 859.9 770.5 613.3 575.2 559.7 543.6 330 277.7 191 28.8

ppb

41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67

Yb Sb Eu Ho Cd Tb Be I W Nb Cs Tl Tm Lu Pd Bi Ag Te Hf Au Os Pt Ta Hg Rh Ru Ir

79.2 61.5 54.2 33.6 32.6 31.7 30.9 29.8 23.5 22.3 19.1 16.3 13.2 12.2 11.6 9.6 7.6 5.4 2.3 1.4 939.3 719.8 614.6 493.6 368.3 141.9 35.5

ppb

Nb Ni Ge Cu Mo P Ag V Ir S Rh Sr Ca B Se Hg I Pt Ru Li Na Cr Te As W Au Mg

1.6 1.6 1.6 1.5 1.4 1.4 1.4 1.3 1.2 1.1 1.1 1 0.9 0.9 0.9 0.9 0.8 0.8 0.7 0.7 0.7 0.7 0.7 0.6 0.6 0.6 0.5

storms, which contributed to the inﬂux of marine components, such as chlorine, sodium, potassium, magnesium, calcium, strontium, boron, and iodine (Franze´n, 1990, 1991). Episodes of terrestrial material inﬂux were not as frequent, but occurred as brown, yellow, or reddish dust from deserts and semi-deserts through the strong southerly wind channels opened up when a stable high pressure is situated over Western Russia and East Europe, with strong low pressure(s) to the west over the British Isles and the North Sea. Modern desert dust periods such as these have a recurrence interval of 1–10 times during a 50-year period. The natural dust found in older parts of peat cores, consists of small sharp-edged grains of quartz and other silicate minerals, some of which is even believed to be loess transported from such remote areas of Sweden and from NW China. Most loess deposits of the world are fossil; that is, they were formed in periglacial landscapes during the last ice age. Even the desert-margin loess of Northern Africa seems to have been formed during the LGM (Last Glacial Maximum) (Coude-Gaussen, 1987). In China, however, loess

Mineral matter, major elements, and trace elements in raised bog peat

247

production has been more or less continuous over the Pleistocene in the Taklamakan/Gobi regions and brought southwards by the strong northerly winds during winter monsoon. Some of this material is likely to be lifted high in the atmosphere and dispersed over distant lands such as Western Europe. Volcanic events, strong enough to send a detectable amount of particulate matter to Scandinavia, are not very frequent. In the 5000–6000 year record of peat accumulation of southern Sweden and the 10,000 year record of Ireland, only a handful of such episodes have been recorded in the bog samples investigated, most of them from Icelandic sources, such as Hekla 1–4 (Zielinski et al., 1994), a few from Vesuvian eruptions (Zielinski et al., 1994; Cioni et al., 1999), and the largest event from the Minoan Santorini eruption of about 3650 cal yr BP (Zielinski and Germani, 1998). In such a natural background material, a fourth component could be identiﬁed in extraterrestrial material found as micrometeorites and meteorite ablation spherules. A ﬁfth category of particulate matter is the organic constituents from forest and peatland wildﬁres. Mineral matter The occurrence of mineral matter in peat is a function of several interrelated abiotic and biotic parameters (Andrejko et al., 1983). Many deposited minerals may undergo alteration after deposition, whereas others might be formed by plants or during peat development. Thus, the mineral matter in peat is of three different categories, detrital (sedimentary/transported), authigenic (sedimentary/produced in situ), and diagenetic (altered). Detrital minerals Detrital minerals are generally deposited from sources outside the peatlands. In minerotrophic peatlands, these particles are mostly transported by water, carried by sheet-ﬂow and surface-water runoff. In paralic environments, daily tidal action, tidal surges, and washovers may contribute to large amounts of such particles in low-lying peatlands. In the ombrotrophic, raised bogs, these factors can generally be ruled out. Instead eolian action becomes signiﬁcantly more important, especially in areas near coasts, riverbeds, and lakeshores. Other atmospherically derived matter includes particles from volcanic dust clouds and microscopic extra-terrestrial particles referred to as cosmic dust. Furthermore, with increasing population, anthropogenic sources have become gradually more important. Dust particles are produced by numerous activities including agriculture, trafﬁc, and industries. The amount of anthropogenic dust, even in the most remote peatlands, is well illustrated by the variation in ash content with depth of many raised bog deposits (Fig. 11.2). The graph of Figure 11.2 illustrates a basal minerotrophy (high ash content) related to input from groundwater, the more-or-less human-unaffected ombrotrophy (low ash content) in the middle, and a renewed minerotrophy (high ash content) at the top associated with anthropogenic input. In short, the sources of detrital particles are the following. (1) Marine aerosols: seawater particles volatized in open sea or coastal wave-breaker zones (Franze´n,

248

L.G. Franze´n

Figure 11.2. Content of peat ash in Ga¨llseredsmossen bog, S. Sweden.

1990, 1991; Gustafsson and Franze´n, 1996, 2000). (2) Terrestrial particles: mineral and organic dust derived from coastal sites, periglacial areas, and arid/semiarid environments (Franze´n and Hjelmroos, 1988; Franze´n et al., 1994a, b); and tephra generated during volcanic eruptions (Persson, 1971). (3) Cosmic materials: particles entering the atmosphere, or particles generated after impact on the surface of the Earth, derived from the impactor or formed from terrestrial material. (4) Ignifacts: ﬂy ash and soot; that is, charcoal fragments and tar droplets from wildﬁres. (5) Anthropogenic materials from activities, such as agriculture, trafﬁc, and industry. (1) Marine aerosols and particles. The concentration of sea salts in the atmosphere varies with latitude, altitude, and location relative to the sea. The highest concentrations are found over the sea at low latitudes under conditions of strong winds with breaking waves even far inland from the coast. The stronger the wind, the greater is the release of drops from the ocean surface. Apart from what is driven up in the open sea, an increase in drop formation can also be expected near the coast, especially in archipelagos with a high frequency of breaking waves. In the atmosphere the small drops of saltwater are transported with the air masses as super-saturated droplets or small crystals of sea salt. The salt particles formed are 0.1–20 mm or even larger (Franze´n, 1990). Measurements of the marine deposition in southern Sweden shows a very strong decreasing gradient inland from the shore, and a strong temporal variation related to storms (Franze´n, 1990; Gustafsson and Franze´n, 1996, 2000). Even if the most marine aerosols are deposited within the ﬁrst few kilometers from the shore, signiﬁcant amounts could be recorded as far as 250 km inland after storms. The marine inﬂuence on the landscape is visible in the bedrock (as an increased rate of salt weathering on bare rocks near the coast) and vegetation (the development of salt-resistant ecosystems, and damage to trees after storms). Salts also affect soil development in areas close to the coast. Regarding peatlands, a special form develops

Mineral matter, major elements, and trace elements in raised bog peat

249

in strongly marine-inﬂuenced environments. The blanket bogs of humid Western Europe form a peat cover that gently undulates over the terrain from the valley bottoms to uphill locations, even on relatively steep gradients. The bog vegetation is often similar to that of fens in more continental areas, and rather demanding fen plants from different taxa, such as Carex and Bryales, occur frequently. The pH is generally somewhat higher (pH ca. 4.2) than in proper raised bogs (pH ca. 3.5) or poor fens, but they are still mires fed entirely by atmospheric fallout. The peats formed in the sites near the shore, often have high contents of mineral matter derived from wave-swept shores. The marine inﬂuence decreases rather quickly within a kilometer or so away from the shore. In farther inland locations, the marine inﬂuence is shown mainly by the relatively high concentrations of chlorine, sulphur, sodium, potassium, magnesium, and calcium. Other important marine elements are strontium, boron, rubidium, and iodine. Most of these elements, in particular Cl, Na, and K, are mobile and easily washed out from the surface peats, but some are trapped in the lower anaerobic part of the peat. The marine inﬂuence on bogs investigated is illustrated by the decreasing mean value of the iodine/salt ratio of the ombrotrophic parts of the various proﬁles, over distance from the coast (Fig. 11.3). (2) Terrestrial mineral particles. The terrestrial mineral concentrations in peat derive mainly from material brought in by water in the lower minerotrophic peat layers, whereas variations in higher ombrotrophic layers are commonly associated with dust derived from agricultural lands. A special category of detrital particles derives from long-range transport from loess regions of the world. For example, in the Hongyuan plateau peatlands of Sichuan province, China, the ash content of peat investigated usually varies from 25 to 50%, but may in places reach values as high as 50–70% (Bjo¨rk, 1993). The amount of particles transported by water into these soligenous

Figure 11.3. Concentration of iodine in peat ash relative to other elements plotted against distance from west coast of Sweden.

250

L.G. Franze´n

peatlands are insigniﬁcant in comparison to the eolian portion. Preliminary studies show that the inﬂux of loess into this region, since peat formation began ca. 10,000 cal yr BP, lies within the range of approximately 0.1–0.5 mm per year. I consider it possible that the source of terrestrial mineral grains in the lower ombrotrophic peats of Europe, in the pre-anthropogenic landscapes, mainly originates from East Asian sources since preliminary studies show that some larger mineral grain peaks in the European peat coincide with strong inﬂux of loess to the mires of Tibet. The amount of natural dust, including the authigenic minerals, as calculated from the peat ash contents in southern Sweden and Ireland bogs, is calculated at about 0.05–0.1 mm per 1000 years in pre-industrial times. This is about 10–100 times lower than the deposition rate of the slowest accumulating red clays of the sea ﬂoor (Seibold and Berger, 1982). In bogs close to the seashore, such as Owenduff bog of Ireland (15 km from the shore) and Lyngmossen bog in Sweden (18 km, these amounts are normally somewhat higher, whereas the ash content of inland sites normally ranges between 0.3 and 0.8% of the coastal sites) have ashes of 1–1.5%. Whereas many of the minerals of the terrigenous dust, such as feldspars, mica, amphibole, and other silicates, are eventually weathered away within the aggressive peat environment, pure quartz persists longer. Hence, in the lower portions of peat successions, quartz grains dominate. Other particles, occasionally occurring unweathered throughout the stratigraphies even in very deep peat proﬁles, are the durable grains and crystals of ilmenite (Fig. 11.4a), magnetite (Fig. 11.4b), zircon, and garnet. The raised bogs are ideal locations for the study of volcanic tephra. It is known that some historical eruptions had a noticeable impact on climate and agriculture on a hemispheric to global basis. Most signiﬁcant may be the injection of sulphur gases and particles into the stratosphere during the eruption (Zielinski and Germani, 1998). The largest eruptions may possibly have caused severe volcanic winters (Rampino et al., 1985, 1988) similar to the proposed nuclear winters or cosmic winters (Clube and Napier, 1990) following bolide impacts. A good example is the major eruption of Tambora in 1815. This gigantic eruption ejected over one hundred cubic kilometers of debris into the atmosphere, some of which rose so high into the stratosphere that it

Figure 11.4. (a) Ilmenite grain. Under the light microscope these grains have a black, glossy appearance (Fallahogy bog, N. Ireland). (b) Magnetite crystal (Sluggan bog, N. Ireland).

Mineral matter, major elements, and trace elements in raised bog peat

251

encircled the world for years. The subsequent chilling effect was almost worldwide. Crop failures occurred in North America and Western Europe and caused widespread famine. The largest increase in the particulate content of the stratosphere during the 20th century was caused by the eruption of Mt Pinatubo in June 1991. The inﬂuence of this event temporarily interrupted the trend of globally rising surface temperatures (McCormick et al., 1995). The ﬁrst attempt to use tephra as a geological marker in peat was made by Auer in 1928 (Auer, 1965). His attention was drawn to a signiﬁcant ash layer in a peat proﬁle from Tierra del Fuego, South America. In all proﬁles investigated, three ash layers were eventually found which could be used as reference levels for the paleovegetation development. Today tephrochronology is one of the standard dating techniques used in Quaternary deposits, even in areas far from volcanoes (Self and Sparks, 1981; Pilcher and Hall, 1992). Tephra is formed by explosive eruptions of viscous magmas rich in gas. Tephra has a low speciﬁc mass due to its highly vesicular structure. The small glass shards are hard to detect under light microscope, but under the polarizing microscope, the amorphous structure makes them easily distinguished from silicates having a crystalline structure. Under the SEM they appear as sharp-edged, icy forms with numerous cavities (Figs. 11.5a–f). The frequency of occurrence of glass shards varies strongly with peatland location and stratigraphical level within peat proﬁles. In the vast majority of the samples scanned under the polarizing microscopes, not a single shard may be found whereas

Figure 11.5. Volcanic glass shards. (a) Do¨mle mosse bog, Sweden, ca. 3700 cal yr BP; (b) Ga¨llseredsmossen bog, SW Sweden, ca. 3950 cal yr BP – Hekla 4. (c) Fallahogy bog, N. Ireland, ca. 4450 cal yr BP. (d) Ga¨llseredsmossen bog, SW Sweden, ca. 3950 cal yr BP – Hekla 4. (e) Owenduff bog, W. Ireland, ca. 8000 cal yr BP – Vesuvius, Italy? (f) Lyngmossen bog, SW Sweden, ca. 3650 cal yr BP – Minoan Santorini eruption.

252

L.G. Franze´n

in others, single stray shards occur. These single grains could have been reworked from previously deposited sediments or may have arrived from very remote eruption sites through stratospheric transport. In true tephra layers, volcanic shards sometimes make up for 90% or more of mineral grains (Fig. 11.5a), even if tephra horizons in the Swedish or Irish bogs at least, are never detectable in the ﬁeld but only under the microscope. In the studied Swedish peat proﬁles, the following volcanic events have been distinguished from their age and chemical composition: unknown (Owenduff bog ca. 8000 cal yr BP, possibly a Vesuvian eruption), unknown (Fallahogy bog ca. 4800 cal yr BP), Hekla 4 (ca. 4000 cal yr BP), Minoan Santorini (ca. 3650 cal yr BP), Hekla 3 (ca. 3150 cal yr BP), Hekla 1 (1104 AD), Askja (1875 AD). In the Fuegian material volcanic events are much more common where the major tephra fall occurring at ca. 7300 14C yr BP forms a 10 cm thick ash layer. Preliminary results from the study of Harberton Bog point out other signiﬁcant tephra horizons of around 11,100–11,600, 8700, 5400, 4700, 3400–3500, 3100, and 2550 14C yr BP. (3) Cosmic dust. Extraterrestrial dust particles have been recovered from various environments, most notably deep-sea sediments and ice at the South Pole (Brownlee, 1981; Taylor et al., 1998). These occurrences have established the presence of an extraterrestrial component in terrestrial deposits, which originated either as micrometeorites or products of ablation of meteorites that pass through the atmosphere (Hagen et al., 1990; Zbik and Gostin, 1995). In the case of iron meteorites, surfaceheating boils droplets of molten metal into the surrounding air. Instantly cooled and solidiﬁed, these droplets form microscopic spherules that settle to the earth along with the remaining meteoritic core. In the case of stony meteorites, the ablation products may end up with particles more or less enriched in iron. Iron has the highest boiling temperature of the principal elements, and would last longest during the atmosphere entrance. The stony meteorite spherules are glassy, and are transparent with yellow or brownish nuances to opaque and black. The gassed-off elements from both iron and stony meteorites turn up sooner or later at the surface of the earth, caught by gravitation. Both iron and stony meteorites appear with a fusion crust that is normally enriched in iron. The annual inﬂux of cosmic material has been calculated as 1.6–2.0 109 g yr1 to 1.5 1011 g yr1 (D’Almeida et al., 1991; Kane and Chester, 1993; Ceplecha, 1996). Distributed over the whole world this means ca. 0.03–0.3 103 g m2 yr1 or about the magnitude 105–103 of the pre-industrial ash content in ombrotrophic peats measured in Swedish and Irish peat stratigraphies. An additional feature associated with extraterrestrial activity is the micro-tektites that are formed from terrestrial or cosmic material released during a larger meteorite impact on Earth’s surface. The likelihood of ﬁnding such particles in Holocene peats is small. Aside from the many writings about the Siberian Tunguska event of 1908 (Dolgov et al., 1971; Vasilyev et al., 1971; Nazarov et al., 1983; Kolesnikov et al., 1998), a recent study from the Kaali impact crater of Estonia reports on the ﬁndings of impact ejecta (allochthonous minerals and iridium) dating back at ca. 800–400 BC (Veski et al., 2001). Cosmic particles/substances in peat samples can be detected by the presence of irregular micrometeorites and of magnetic and nonmagnetic spherules, and because of anomalous geochemical signatures. This latter feature will be discussed in the geochemical section below.

Mineral matter, major elements, and trace elements in raised bog peat

253

True micrometeorites are extremely rare in the peat samples. Throughout the peat samples studied that are more than 2000, only a few tens of micrometeorites have been found. Under the light microscope they appear as dull, gray to black, and could be identiﬁed by their magnetic properties. Under the electron microscope, their surface appears melted as if the sharp edges of a crushed mineral grain had been melted off. The surface is often patterned with deep cavities or distinct micro-regmaglyptic structure. The best indication, however, that they are micrometeorites, is the SEM–EDX analyses that show the small, but signiﬁcant content of nickel. Figures 11.6a–d show four possible micrometeorites found in peat from Harberton bog, Tierra del Fuego, Argentina. Their elemental composition is shown in Table 11.2. Spherules (microscopic spherical particles) are common and occur in almost all peats studied, but their concentration varies in different parts of the peat proﬁle. They are easily detected because of their shape and the mainly shining glossy appearance. The origin of the spherules has been a matter of debate since they were ﬁrst described from deep-sea deposits (Murray and Renard, 1891; Pettersson and Fredriksson, 1958), but were subsequently reported from many other environments.

Figure 11.6. Micrometeorites. (a) Lyngmossen bog, SW Sweden, ca. 750 cal yr BP. (b) Harberton bog, Tierra del Fuego, S. Argentina ca. 450 cal yr BP. (c) Harberton bog, Tierra del Fuego, S. Argentina, ca. 700 cal yr BP. (d) Moat bog, Tierra del Fuego, S. Argentina, ca. 700 cal yr BP.

Table 11.2. The elementary composition (in%; SEM–EDX) of the surface of 4 micrometeorites (chondrites) found in peat (Figs. 11.6a–d). The relative enrichment of Fe, Cr, and Ni at the surface, compared to the composition of ordinary chondrites, as reported by Wood (1963), may be explained by ablation processes during atmospheric entrance. Fig. 11.6a

Fig. 11.6b

Fig. 11.6c

Fig. 11.6d

Na2O MgO Al2O3 SiO2 CaO TiO2 Cr2O3 MnO FeO NiO

0.20 24.49 1.33 34.75 0.73 0.19 0.63 0.43 35.15 0.41

0.15 15.23 0.99 25.57 0.65 0.12 3.49 0.52 49.75 2.15

0.25 21.60 2.35 35.18 4.06 0.05 0.32 0.40 34.90 0.60

1.07 6.38 6.47 25.46 4.36 0.27 2.86 0.43 50.18 1.42

Sum

98.31

98.62

99.71

98.90

254

L.G. Franze´n

It soon became obvious that it was almost impossible to distinguish between natural spherules and spherules formed by different human activities. The modern world is full of such spherules and they are produced as the result of a large variety of different human activities including iron spherules made by welding or metal cutting, insulation material such as glass and stone wool, furnace insulation bricks, atomic explosions, industrial spherules from steel works or combustion of solid fuels, and trafﬁc pavement paint, all of which might be naturally deposited or may be contaminants introduced during the sampling or later laboratory analyses. Various methods have been tried to sort out cosmic iron spherules from industrial ones, for instance using the presence of wu¨stite (Taylor et al., 1996), a form of magnetite, which is believed to form in the upper atmosphere under low oxygen pressures. The natural terrestrial spherules are believed to derive either from volcanism, forest wildﬁres, or lightning. Whereas spherules in marine deposits have been reported frequently over the years, their occurrence in peat is much less studied. Notable studies are those in Estonia (Raukas, 1997; Raukas et al., 2001) and in the Tunguska region of Siberia (Dolgov et al., 1971; Vasilyev et al., 1971; Nazarov et al., 1983; Kolesnikov et al., 1998). The micro-spherules found in peat proﬁles are of two kinds: (a) shiny metallic grey or black magnetic spherules, and (b) transparent or opaque nonmagnetic spherules. (a) Under the light microscope, the black magnetic spherules normally appear glassy and are easily distinguished from other round objects by this property. Some magnetic spherules, however, are dull but have a metallic appearance. Under the SEM, the glossy and smooth-surface spherules (Fig. 11.7a) are readily distinguishable from the rough-surface, dull particles. The degree of dullness of the latter is generally determined by the grade of which different structures (Figs. 11.7b, c) have developed on their surfaces. These surface structures are likely formed by supercooling of the ﬂuid during the quenching process. The solidiﬁcation process that begins from the outer shell of a ﬂuid metallic drop may also explain the occurrence of cavities (Fig. 11.7d), and holes (Fig. 11.7e) inside the spherules. According to Carusi et al. (1972), these cavities are formed by gas bubbles, which expand in the melted droplets. The cellular structures may also be found on the inner walls of these cavities (Fig. 11.7f). SEM–EDX analyses show that the vast majority of these magnetic spherules are made of pure magnetite. (b) The nonmagnetic spherules are less common than the magnetic ones, and they do not exceed 5–10% of the total number of spherules. These glassy spherules have wide range of colors and transparencies spanning from opaque and black to smoky and gray, brownish to even colorless and transparent. The color generally reﬂects the content of iron; the more iron the darker the spherule. The composition and appearance of a few of these spherules are listed in Table 11.2. Many of the glassy spherules have vesicles, visible under the light microscope, but not in the SEM. Whereas most of these spherules are perfectly polished, pearl-like objects (Fig. 11.8a) some of them are covered with small depressions, giving them a golf ball-like appearance. Others are perforated with deep holes and cavities (Figs. 11.8b, c). Whereas many of these surface structures are weathering phenomena (Fig. 11.8d), others are likely formed during the solidiﬁcation processes (Fig. 11.8e). In some cases, cracks

Mineral matter, major elements, and trace elements in raised bog peat

255

Figure 11.7. Black magnetic sperules. (a) Black magnetic spherule with smooth surface. It appears glossy under the light microscope (Do¨mle mosse bog, Sweden, ca. 4300 cal yr BP). (b) Black magnetic spherule with well-developed, dendritic surface structures. Small round objects are glossy magnetite spherules (Clara bog, central Ireland, ca. 400 cal yr BP). (c) Black magnetic spherule with polygon plate surface structures (Harberton bog, Tierra del Fuego, S. Argentina, ca. 450 cal yr BP). (d) Black magnetic spherule with a yarn-ball-like dendritic surface structure. The spherule appears to be hollow with some round entrance cavity (Ga¨llseredsmossen bog, SW Sweden, ca. 400 cal yr BP). (e). Black magnetic spherule. It appears to be hollow with a big hole. Weakly developed dendritic surface structures (Harberton bog, Tierra del Fuego, S. Argentina, ca. 1450 cal yr BP). (f) Crushed black magnetic spherule. The hollow interior exposes a well-developed, dendritic surface structures (Sluggan bog, N. Ireland, ca. 400 cal yr BP).

occur during cooling (Fig. 11.8f) and in rare cases, impact structures such as craters (Figs. 11.8g, h), shock-wave rings, or frozen splash-up cascades, witness collisions with smaller objects during the atmospheric ﬂight. Isolated spherules have been found disseminated in most peat samples investigated; however, the large concentrations in rather distinct layers occur as well. These layers never coincide with volcanic horizons, or with high concentrations of much charcoal and tar. This is the reason both volcanic and wildﬁre origin for these spherules is ruled out and the extraterrestrial origin is preferred. Whereas the spherules reported from deep-sea sediments and the Antarctic ice cap are rather large, with sizes ranging from 0.1 to 1 mm, the size of the spherules found in peat varies from less than 1 micron to a maximum of about 100 microns. Smaller sizes than 1 micron are impossible to detect by the magniﬁcation used under the light microscope, but even under the SEM no spherule larger than 1 micron has been measured in the samples studied. The apparent absence of spherules smaller than 1 micron may be an artifact due to sample treatment; that is, not enough time was allowed for the smaller particle to settle out of suspension during decantation. (4) Ignifacts (charcoal and tar). The term ignifact is here intended to mean any item shaped, deformed, affected, or produced by ﬁre. By deﬁnition, such items would include ﬁre cracked stones, melted rock fragments or grains of minerals, pieces of

256

L.G. Franze´n

Figure 11.8. (a) Dark brown glass spherule. Harberton bog, Tierra del Fuego, S. Argentina, ca. 1450 cal yr BP. (b) Brown transparent glass spherule (Fallahogy bog, N. Ireland, ca. 650 cal yr BP). (c) Colorless transparent spherule patterned with small cavities (Do¨mle mosse bog, Sweden, ca. 3700 cal yr BP). (d) Brownish green transparent glass spherule with weathering pits. The small cavities could also have been formed by microbiological activity (Sluggan bog, N. Ireland ca. 750 cal yr BP). (e) Colorless transparent glass spherule with a deep cavity that was probably formed during solidiﬁcation. Smaller attached spherules have the same composition as the main sperule (Sluggan bog, N. Ireland, ca. 750 cal yr BP). (f) Yellowish transparent spherule. The cracked surface structure was probably caused by rapid cooling (Harberton bog, Tierra del Fuego, S. Argentina, ca. 1450 cal yr BP). (g) Transparent brownish greenish glass spherule from Lyngmossen bog ca. 1450 cal yr BP. The crater formed is probably an impact structure. The interior of the crater as well as the bright crater rim, have a very high iron content compared to the spherule itself. (h) Detail from 8 g. The spherule was most likely hit by a much smaller (ca.1 micron), fast rotating iron spherule, which entered at the right side (a) and went out at the left (b). Note the halved outlet crater to the left. A gaseous explosion probably formed the large crater.

wood deformed by ﬁre, charcoal, and tar. Both charcoal and tar can be used as indicators of paleo-ﬁre frequency and climate (Tolonen, 1985; Pitka¨nen, 2000). Ignifacts could have several different local or distant sources, such as wildﬁres on other peatlands, forests, or grasslands. Charcoal is not only found in peat formed during the Holocene, but also in coal (Scott and Jones, 1994) and lignite (Parish and Lamberson, 1998). Autochthonous ignifacts such as macroscopic or microscopic pieces of charred wood, herbs, and mosses are formed by ﬁres in the peatland whereas allochthonous ignifacts are microscopic particles brought in by wind (Simmons and Innes, 1988; Carcaillet et al., 2001). They can all be used to detect periods of wildﬁre activity (Brussel, 1988). Ignifacts in peat have also been used to explain vegetation shifts in the peatlands at speciﬁc times where charcoal horizons occur (Alm et al., 1992). During the Holocene, charcoal occured during preferential times in peat proﬁles (Wu Hong, 1992). In lower Holocene peats in southernmost South America, charcoal particles occurred intermittently since about 13,000 14C yr BP and in great abundance since 11,000 14C yr BP (Markgraf, 1993). Preliminary results from Harberton bog, Tierra del Fuego, show that charcoal and tar are present, more or less frequently abundant, from the bottom layers dating back at ca. 12,000 14C yr BP to the present. The preferred occurrence of ignifacts, as observed by the author, in the lower parts of peatlands in southern Sweden and Ireland, indicate a warm and dry Late Atlantic/

Mineral matter, major elements, and trace elements in raised bog peat

257

sub-Boreal with frequent and widespread wild ﬁres, followed by a moist and cool sub-Atlantic with more sporadic ﬁres. The occurrence of ignifacts in the upper levels of the Swedish material is also strongly correlated with the abundance of pollen from coniferous trees, especially Norwegian spruce, indicating that coniferous forest communities are more susceptible to ﬁre than deciduous environments. The frequency of major natural forest ﬁres has been calculated in Finland from peat deposits to occur on average every 80–85 years during the last 3000 years (Tolonen, 1985). Because of the presence or crop pollen in cores, Tolonen claims that prior to 2000 14C yr BP ﬁres were primarily natural. Three progressive stages of human interference were revealed according to Tolonen, due to temporary slash-andburn clearance (300–700 AD), intensive, continual slash-and-burn (700–900 AD) until the middle ages and modern times, and predominantly arable cultivation from about 1700 AD. Hence, ﬁres could either have natural causes such as volcanism (Scott and Jones, 1994; Wilmshurst and McGlone, 1996; Samaniego et al., 1998), lightning (Scott and Jones, 1994; McCafferty and Owen, 1996), or anthropogenic activities such as slash-and-burn clearances and regular large-scale burn-beating as detected from peat deposits in East Africa from ca. 4800 14C yr BP (Hamilton et al., 1986), England from ca. 5000 14C yr BP (Simmons and Innes, 1988), Ireland from ca. 5000 14C yr BP (Dodson, 1990), Scotland from 5950–5700 14C yr BP (Hirons and Edwards, 1990) and New Zealand from ca. 3500 14C yr BP (Rogers and McGlone, 1994). Simmons and Innes (1988) pointed out that periods of forest disturbance coincide with charcoal layers in peat proﬁles, whereas disperse microscopic charcoal occurs during phases of relative woodland stability. The presence of charcoal and tar particles was also recorded in all peat cores analyzed in my study for mineral composition. Although some larger pieces of charred wood were retrieved during sampling, most ignifacts were recognized under the microscope (5–100 mm). Charred particles included coniferous pollen grains, microscopic pieces of wood, and herb tissues. Tar particles consisted of large spherules with an inverted golf ball-like surface (Fig. 11.9a), large spherules reminding one of mulberries (Fig. 11.9b), small, perfectly spherical to elongated fusiform particles with a glossy appearance (Figs. 11.9c, d), irregularly shaped tar fragments (Fig. 11.9e), and solidiﬁed tar foam (Fig. 11.9f). The temporal distribution of ignifacts in the peat cores from southern Sweden showed a rather distinctive pattern (Fig. 11.10). A high inﬂux of charcoal and tar generally occurs in the lower parts of the core in decomposed fen and carr peats older than 4000 14C yr BP. Charcoal is less abundant in the overlying strongly decomposed Sphagnum peat, and both charcoal and tar are rare and only occur in speciﬁc layers in the uppermost weakly decomposed ombrotrophic peat, less than ca. 2500 14C yr BP. There is no regional difference in the oldest, minerotrophic peat layers, and it seems that wildﬁres were common throughout southern Sweden the ﬁrst few thousand years after peat formation began. The ignifacts were mainly of local types with a high share of charred herbs of the early sedge-dominated fens. The distribution of ignifacts in peat cores of southern Sweden indicate a ca. 550-year periodic variation between higher and lower concentrations (Fig. 11.10). This suggests recurrent variation of moister and dryer conditions that regulated the occurrence of wildﬁres. Notable dry conditions occurred between around 4500 14C yr BP and 4000 14C yr BP,

258

L.G. Franze´n

Figure 11.9. Tar spherules. (a) Large tar spherule with a coarse granular surface structure. (b) Large tar particle of the ‘mulberry’ type. (c) Small black tar spherules with a smooth surface. These spherules are almost impossible to distinguish from black and glossy magnetic spherules in light microscope. (d) Elongated fusiform tar particle. (e) Irregularly shaped tar fragment with vesicles. In the SEM these particles are reminiscent of volcanic glass shards. (f) Solidiﬁed tar foam.

Figure 11.10. Temporal distribution of ignifacts (tar and charcoal fragments) in the Swedish peat samples studied. Ignifact frequency was recorded in a relative scale 0–5, where 0 indicates no ignifacts and 5 many ignifacts. Five standards were used for the higher classes. A running mean has been used to smooth the graph and interpolate between classes. (Age in cal yr BP).

Mineral matter, major elements, and trace elements in raised bog peat

259

whereas a longer period of moister conditions is bracketed between high peaks of ca. 2900 14C yr BP and ca. 1900 14C yr BP. The high peaks of surface peat far from builtup areas can probably be attributed to local domestic heating. Authigenic minerals Authigenic minerals are produced in situ by biogenic activities such as peat-forming plants, microbial activity, and geochemical formative (precipitation) or transformation (diagenetic) processes. The biogenic particles include chrysomonad cysts, diatom frustules, and sponge spicules. Some peat-producing plants such as various Graminoides accumulate opaline silica within the plant, which is released eventually in the peat deposits as phytoliths (Andrejko et al., 1983; Carnelli et al., 2004). In the European material investigated, the phytolith frequency varies greatly with depth, from absence to more than 95% of the mineral matter of some layers. There is no regional difference in the occurrence of phytoliths, as they occur more or less frequently in all peatlands in Sweden, Ireland, and Tierra del Fuego. Biogenic silica particles observed in peat are shown in Figures 11.11a–f.

Figure 11.11. (a) Sponge spicules (Ephydatia). The small round object is a pyrite spherule. (Ga¨llseredsmossen bog, SW Sweden, from minerotrophic peat ca. 8600 cal yr BP). (b) Phytolith from minerotrophic peat (species unknown). Under the polarizing microscope these particles are easily confused with volcanic glass. (c) Phytolith from ombrotrophic peat (species unknown). This type of phytolith is the most common in the peat cores studied, sometimes making up for 99% of all mineral particles. (d) Phytolith from minerotrophic peat (species unknown – probably Graminidae). This type can also be confused with volcanic glass under the polarizing microscope. (e) Bottle-shaped phytolith from minerotrophic peat: second most common type after the sponge spicules (species unknown). The whole phytolith is covered by distinct perforations, 2–4 mm in diameter, possibly as a result of microorganism digestion. (f) the most common type of phytolith in ombrotrophic peat. The surface is patterned by deep cavities, possibly a result of microbiological activity.

260

L.G. Franze´n

Many of the larger biogenic particles exhibit dissolution features of a possible bioerosional origin. These features are presumed to result from assimilative activities of microorganisms such as bacteria, endolithic fungi, or diatoms. The observed features consist of both distinct perforations, 2–4 mm in diameter (Fig. 11.11e) and pitlike structures or depressions (Fig. 11.11f) (Andrejko et al., 1983). Some authigenic particles precipitated under anaerobic conditions (geochemicalgenerated materials). Iron is a common element in most peats, and exists in reduced form in the waterlogged soils as hydrated magnetite (Fe3O4 H2O or Fe3(OH)8) along with some hydrotrolite (FeS nH2O). If anaerobic conditions persist, these precipitates may age and produce the typical minerals of reduced sediments, magnetite (Fe3O4), and pyrite (FeS2) (Sikora and Keeney, 1983). The pyrite mostly occurs in peat as small spherical nodules made up by cubic pyrite crystals; rarely as single crystals (Figs. 11.12a, b). Pyrite is most commonly found in the lower minerotrophic layers, rarely in overlying ombrotrophic peats. Calcium is the most common element in most peat ashes (Table 11.1), and can form rhombic carbonate crystals (Figs. 11.13a, b). In bogs, the main supply of calcium comes from marine aerosols (Franze´n, 1990), which can be dry deposited, or dissolved in the waters of rain and snow. The formation mechanisms of such

Figure 11.12. (a) Pyrite spherules from Clara bog, Ireland, bottom minerotrophic peat layers (ca. 9700 cal yr BP). (b) Pyrite spherules composed of irregularly shaped pyrite crystals (detail of Fig. 12a).

Figure 11.13. (a) Calcite crystals from ombrotrophic peat (Hjortemossen bog, S. Sweden). (b) Calcite crystals found in bog peat. The crystals have a typical rhombic form (Hjortemossen bog, S. Sweden).

Mineral matter, major elements, and trace elements in raised bog peat

261

Figure 11.14. (a) Feldspar grain from ombrotrophic peat. The irregular cavities were formed by differential weathering process or microbiological activity. (b) Mineral grain where relatively less resistant minerals were removed by weathering.

carbonate crystals are still not fully understood, but they are never found in the upper oxygen-rich zone but only in the lower anaerobic parts of ombrotrophic peat. Similar authigenic carbonate rhombs are found in the anoxic parts of marine sediments, formed on the substratum of deposited foraminifers (Pederstad and Aagaard, 1985), calcite spherulites (Verreccia et al., 1995) or cementing organic-rich sediments (Irwin et al., 1977). Weathering may alter or dissolve completely some mineral or mineral aggregates. In the latter case, relative afﬁnity to weathering of the various components may result in irregular forms or cavities (Fig. 11.14a) or a grain ﬁlled with pores formed like the vanished crystal (Fig. 11.14b). Many forms are the result of leaching or dissolution of detrital mineral matter by the surrounding or percolation organic or humic acids within the interstitial waters of the peat deposit (Andrejko et al., 1983).

Major and trace elements in peat The peatlands normally have a proﬁle consisting of an upper aerobic portion, called acrotelm, and a lower anaerobic portion, called catotelm. A transition layer that is essentially at the ground-water table position separates the two parts. In the catotelm, the metabolism of organisms reduces the concentration of oxygen, resulting in anaerobic conditions with reduced forms of different elements. The ability of the waterlogged peat soils to retain heavy metals is important, because these environments are receptors of wastes in human inﬂuenced settings. It is believed that many elements, such as copper and zinc, form organometallic complexes with organic matter. With a persistently high water table, maintaining anoxic conditions, these complexes are kept intact. However, if the water table is lowered by natural causes or by drainage operations, oxidation takes place through percolating surface waters or exposure to air, these organometallic complexes are disintegrated, and the previously ﬁxed elements can be transported out of the system. Since many trace and major elements are retained in the growing peat column, they can be used as archives of the atmospheric input of trace metals located upwind from anthropogenic sources. Metal concentrations in near-surface ombrotrophic peats are sometimes comparable to those found in mineral soils, showing signiﬁcant

262

L.G. Franze´n

surﬁcial enrichment over natural background levels (Perkins et al., 2000; Dollar et al., 2001; Souch et al., 2002). In the following sections, the occurrence of a few trace elements is discussed, such as lead and rare earth elements (REE). Lead gave a uniform signal in all studied bogs, and could be used to make a time correlation in cores where 14C datings were sparse. Lead is a relatively common element in Earth’s crust, but rare in extraterrestrial material (1/15 to crustal) and seawater (1/25,000 to crustal). Along with copper, silver, and gold, lead is probably one of the ﬁrst metals used by humans. Probably, the Egyptians already had the knowledge of metallic lead 5000 years ago. The Babylonians and the Romans used lead plates to write on with metallic pencils. In the ancient Rome and Greece, lead was used in coins, weights, cooking pots, and water pipes. Lead was also used to reinforce stone constructions. The ore mineral galena was initially mined by Grecian and Phoenician colonists in Spain, an industry that was continued by the Romans. Other early galena mines occur in the Harz Mountains and Bohemia of Central Europe and in Cumbria and Derbyshire in Northern England. The transport ways from these sources to the Swedish and Irish bogs could possibly be connected to the general atmospheric circulation pattern with dominating ﬂows from SW. The mining of the lead deposits and the subsequent working of the ore most probably created some dust that could have been lifted up and added to the air masses. The common dated stratigraphic distribution of Pb according to the ICP-MS analyses is shown in Figure 11.15. The peak between 500 BC and 500 AD could have two causes: lead mining and working, and cultivation of North African territories by the Roman Empire. In the latter case, the cultivation of those arid and semi-arid soils could have led to soil degradation similar to what still occurs today, as it is shown by the frequent dustfalls of African material throughout northern Europe (Valentin, 1902; Franze´n et al., 1994 a,b, 1995). The occurrence of lead in peat, and lake sediments, has been reported by Renberg et al. (2000, 2001) and in polar ice caps by Dibb and

Figure 11.15. Concentration of lead (Pb) in the Swedish raised bogs. The graph is a running mean (n ¼ 5) for all measurements made.

Mineral matter, major elements, and trace elements in raised bog peat

263

Jaffenzo (1997). Since the major part of bog deposition in southern Sweden and in Ireland has a marine (and cosmic) origin, any extra addition from crustal sources, such as loess or volcanic ash, would increase the relative proportions between lead and the other elements. More information on lead in peat is given in other parts of this book (particularly in Chapter 21; but also in Chapters 4, 9, 17 and 18). The REE include scandium (Sc), yttrium (Y), and the periodic group of elements (lanthanides) ranging from atomic numbers 57 to 71. All lanthanides form trivalent cations, and only two can form other oxidation states: Eu2+ and Ce4+ (Cornell, 1993). The lanthanides are, on a relative scale, about 100 times more common in terrestrial (continental crust) material than in cosmic dust, and a factor of 105–104 more common than in seawater, hence they could be used to evaluate the degree of terrestrial inﬂuence in peat deposits. A high REE concentration would indicate a strong inﬂuence from terrestrial sources, whereas a lower concentration could indicate a relatively higher inﬂux of marine aerosols and/or cosmic dust. The transition from minerotrophy into ombrotrophy in bog proﬁles developed from fens, carrs, and swamps is clearly deﬁned by the lanthanides signal. Figure 11.16 shows the distribution of lanthanum (La) in the proﬁle from Ga¨llseredsmossen bog, situated some 40 km inland from the west coast of Sweden (Fig. 11.1a), where the transition from minerotrophy to ombrotrophy at a depth of around 500 cm is clearly visible. Bogs, which were formed directly on the mineral soil as a result of paludiﬁcation, normally lack this and have an overall low REE concentration. The top layers of all investigated bogs also show a relatively high REE signal due to modern deposition of dust deposited as a result of different human activities. A dominating inﬂux of cosmic material can be detected by looking at the relative concentration of the REE elements (Boynton, 1984). Bulk Earth concentrations of REE, normalized by dividing them by the concentrations in a reference chondrites sample, plot in quasi-horizontal line (Cornell, 1993). In terrestrial (continental crust)

Figure 11.16. Concentrations of lanthanum in the Ga¨llseredsmossen bog, S. Sweden (note logarithmic scale on vertical axis).

264

L.G. Franze´n

Figure 11.17. Chondrite-normalized concentrations of lanthanides. The upper line shows the average crustal composition; the lower line shows the concentrations in sample 163 (805–810 cm) ca. 8700 cal yr BP, from Fallahogy bog of Northern Ireland. The almost perfectly horizontal graph indicated a dominantly cosmic origin of deposited dust in this sample.

material, instead, the lighter REE are relatively more common than the heavier ones, and europium (Eu) is depleted in respect to the other lanthanides. Plagioclase has a strong afﬁnity for the reduced form Eu2+, depleting it in magmas from which it crystallizes, leading to depletion in the residual ﬂuids and in later precipitated minerals; thus, continental rocks are generally characterized by the europium depletion (Fig. 11.17a) (McLennan and Taylor, 1981; Cornell, 1993). These relationships can be utilized to establish the concentration of cosmic versus Earth crust-derived particles in peat ash. The horizontal graph for sample 163 from Fallahogy bog of Northern Ireland, with no Eu depletion, indicates, for example, that cosmic dust inﬂux strongly dominated over the terrestrial sources (Fig. 11.17b). The sample presented in Figure 11.17 is only one of a row of six samples or 30 cm of peat, with a similar horizontal distribution of chondrite normalized lanthanides. With the peat growth rate calculated from the 14C datings this means that cosmic material inﬂux dominated over a period of ca. 400 years between 8600 and 9000 cal yr BP. The distinct decrease of the lanthanides in the transition from minerotrophy to ombrotrophy; that is, from fen to bog, was found in all peat stratigraphies with this developmental history. Hence, the lanthanide curve can be used to distinguish this transition, more precise than that traditionally used, but a not equally distinct decrease in ash content (for a comparison see Fig. 11.2).

Summary and ﬁnal remarks Bogs are important archives for atmospheric deposition since peat formation began after deglaciation. Their stratigraphic record is much shorter than what is provided

Mineral matter, major elements, and trace elements in raised bog peat

265

by Arctic and Antarctic ice cores. However, the signal is very good, at least in preindustrial times, with the total amount of non-biological material mineral matter (including solids and soluble salts) precipitated into peatlands being calculated at about 0.05–0.1 mm per 1000 years. The results presented here show that the deposition varies largely over time, for mineral particles as well as for chemical elements. The mineral fraction of peat and the different element concentrations can be used as a proxy for climate changes. For example, variation of trace elements with a marine signature, such as iodine, signals an increased inﬂux of marine aerosols and hence the intensity of cyclone activity. The inﬂux of soil particles might be associated with variations of beach processes, for example, but also, if telecorrelated with Chinese sources, might tell us something about the variations of winter/summer monsoon activities and the subsequent loess production in that area. The particulate charcoal, or ignifacts, indicative of ﬁres in peatlands or in their surroundings, can be used as a proxy for climate types. Low frequencies indicate high cyclone activity with few thunderstorms whereas a large inﬂux of ignifacts indicates more continental conditions with a higher frequency of convective cloud types and subsequent thunderstorms. The investigations show that the stratigraphic occurrence of cosmic particulate matter, such as micro-spherules, and the relation between the elements of the lanthanide group show up with a similar pattern in all the bogs investigated in Sweden, Ireland, and Tierra del Fuego. Preliminary results indicate that a correlation exists between the timing of samples with greater concentration of cosmic materials and that of climate deterioration (colder and wetter climates) as indicated by historical and geological data information. This implies that the studies of dust ﬂuxes and their connections to climate are important, not only for the study of paleoclimatic variations, but also for impacts on the future climate change scenarios.

References Adams, J., Faure, H., Faure-Denard, L., et al., 1990. Increases in terrestrial carbon storage from the Last Glacial Maximum to the present. Nature 348, 711–714. Ajtay, G.L., Ketner, P., and Duvigneaud, P., 1979. Terrestrial primary production and phytomass. In: Bolin, B., Degens, E., Kempe, S., and Ketner, P. (Eds), The Global Carbon Cycle, SCOPE 13. Wiley, Chichester, pp. 129–182. Alm, J., Tolonen, K., and Starr, M., 1992. Determination of recent apparent carbon accumulation in peat using dated ﬁre horizons. In: Vasander, H. and Starr, M. (Eds), Proceedings of the International Workshop on Carbon Cycling in Boreal Peatlands and Climate Change. Suo 43, 191–194. Andrejko, M.J., Cohen, A.D., and Raymond, R., Jr., 1983. Origin of mineral matter in peat. In: Raymond, R., Jr. (Ed.), Mineral Matter in Peat. Its Occurrence, Form and Distribution. Proceedings of Workshop on Mineral Matter in Peat, September, Los Alamos, pp. 3–25. Arimoto, R., 2001. Eolian dust and climate; relationships to sources, tropospheric chemistry, transport and deposition. Earth-Sci. Rev. 54, 29–42. Auer, V., 1965. The Pleistocene of Fuego-Patagonia, Part IV: Bog proﬁles. Annales Academiae Scientiarum Fennicae, Ser. AIII, Geologica-Geographica 80, 160pp. Bjo¨rk, S., 1993. The Hongyuan wetland research project. An ecological and technical feasibility study of peat mining in Hingyuan, Sichuan, China. Summary prepared for the Sichuan Institute of Natural Resources, Lund, Sweden, 100pp.

266

L.G. Franze´n

Bory, A.J.M., Biscaye, P.E., Svensson, A., and Grousset, F.E., 2002. Seasonal variability in the origin of recent atmospheric mineral dust at North GRIP. Greenland. Earth Planet. Sci. Lett. 196, 123–134. Boynton, W.V., 1984. Geochemistry of rare earth elements: meteorite studies. In: Henderson, P. (Ed.), Rare Earth Element Geochemistry. Elsevier, Amsterdam, pp. 63–114. Brownlee, D.E., 1981. Extraterrestrial components. In: Emiliani, C. (Ed.), The Oceanic Lithosphere. Wiley, New York, pp. 733–762. Brussel, M.R., 1988. Mid and late Holocene pollen diagrams and Polynesian deforestation, Wanganui District, New Zealand. NZ J. Bot. 26, 431–451. Carcaillet, C., Bouvier, M., Frechette, B., Larouche, A.C., and Richard, P.J.H., 2001. Comparison of pollen-slide and sieving methods in lacustrine charcoal analyses for local and regional ﬁre history. Holocene 11, 467–476. Carnelli, A.L., Theurillat, J.-P., and Madella, M., 2004. Phytolith types and type-frequencies in sub-alpine plant species of the European Alps. Rev. Palaeobot. Palynol. 129, 39–65. Carusi, A., Coradini, A., and Fulchignoni, M., 1972. Genesis of the structures of the black magnetic sperules. Periodico di Mineralogia 41, 363–371. Ceplecha, Z., 1996. Luminous efﬁciency based on photographic observations of the lost-city ﬁreball and implications for the inﬂux of interplanetary bodies onto Earth. Astron. Astrophysics, 311, 329–332. Chague-Goff, C. and Fyfe, W.S., 1997. Effect of permafrost on geochemistry in a Canadian peat plateau bog. Appl. Geochem. 12, 465–472. Cioni, R., Santacroce, R., and Sbrana, A., 1999. Pyroclastic deposits as a guide for reconstructing the multi-stage evolution of the Somma-Vesuvius caldera. Bull. Volcanol. 60, 207–222. Clube, V. and Napier, B., 1990. The Cosmic Winter. Basil Blackwell, Oxford, 307pp. Cornell, D.H., 1993. Rare earths from supernova to superconductor. Pure and Appl. Chem. 65, 2453–2464. Coude-Gaussen, G., 1987. The perisaharan loess: sedimentological characterization and paleoclimatical signiﬁcance. GeoJournal 15, 177–183. D’Almeida, G.A., Koepke, P., and Shettle, E.P., 1991. Atmospheric aerosols: global climatology and radiative characteristics. Deepak publishing, QC882.42-D148 Dibb, J.E. and Jaffenzo, J.L., 1997. Air-snow exchange investigations at Summit, Greenland; an overview. J. Geophys. Res., C, Oceans 102, 26795–26807. Dodson, J.R., 1990. Fine resolution pollen analysis of vegetation history in the Lough Adoon Valley, Co. Kerry, western Ireland. Rev. Palaeobot. Palynol. 64, 235–245. Dolgov, Y.A., Vasilyev, N.V., Shugurova, N.A., et al., 1971. Chemical composition of silicate spherules from peat near the impact point of the Tunguska Meteorite. Transactions of the USSR Academy of Sciences; Earth Science Sections 200, 212–214. Dollar, N.L., Souch, C.J., Filipelli, G.M., and Mastalerz, M., 2001. Chemical fractional of metals in wetland sediments: Indiana dunes National Lakeshore. Environ. Sci. Technol. 35, 3608–3615. Franze´n, L. 1985. Peat in Sweden – a method to calculate the resources. Avhandling, GUNI Rapport 21, 241pp. Franze´n, L. and Hjelmroos, M., 1988. A coloured snow episode on the Swedish west coast, January 1987. A quantitative study of air borne particles. Geograﬁska Annaler 70A, 235–243. Franze´n, L.G., 1990. Transport, deposition and distribution of marine aerosols over southern Sweden during dry westerly storms. AMBIO 19, 180–188. Franze´n, L.G., 1991. The changing frequency of gales of the Swedish west coast and its possible relation to the increased damage to coniferous forests of southern Sweden. Int. J. Climatol. 11, 769–793. Franze´n, L.G., 1992. Can earth afford to lose the wetlands in the battle against the increasing greenhouse effect. IPC Proceedings of the 9th International Peat Congress, Uppsala, Sweden 1992. Int. Peat J., Special Issue 1, 1–18. Franze´n, L.G., 1994. Are wetlands the key to the ice age cycle enigma? AMBIO 23, 300–308. Franze´n, L.G., Chen, D., and Klinger, L.F., 1996. Principles for a climate regulation mechanism during the late Phanerozoic era, based on carbon ﬁxation in peatforming wetlands. AMBIO 25, 435–442. Franze´n, L.G., Hjelmroos, M., Ka˚llberg, P., et al., 1994. Yellow dust – a case study of long-range transport of dust in Northern Fennoscandia, March 1991. Atmos. Environ. 28, 3629–3646.

Mineral matter, major elements, and trace elements in raised bog peat

267

Franze´n, L.G., Hjelmroos, M., Ka˚llberg, P., et al., 1995. The Saharan dust episode of south and central Europe, and northern Fennoscandia, March 1991. Weather 50, 313–318. Franze´n, L.G., Mattsson, J.O., Ma˚rtensson, U., et al., 1994. Yellow snow over the Alps and Subarctic from a dust storm in Africa, March 1991. AMBIO 23, 233–235. Fredriksson, D., 1996. Peat resources in Sweden. In: Lappalainen, E. (Ed.), Global Peat Resources. International Peat Society. Jyska¨, Finland, pp. 137–144. Gustafsson, M.E.R. and Franze´n, L.G., 1996. Dry deposition and concentration of marine aerosols in a coastal area, SW Sweden. Atmos. Environ. 30, 977–989. Gustafsson, M.E.R. and Franze´n, L.G., 2000. Inland transport of marine aerosols in southern Sweden. Atmos. Environ. 34, 313–325. Hagen, E.H., Koeberl, C., and Faure, G., 1990. Extraterrestrial spherules in glacial sediment, Beardmore Glacier area, Transatlantic Mountains. Contribution to Antarctic Research I, Antarctic Res. Ser. 50, 19–24. Hamilton, A., Taylor, D., and Vogel, J.C., 1986. Early forest clearance and environmental degradation in southwest Uganda. Nature 320, 164–167. Hirons, K.R. and Edwards, K.J., 1990. Pollen and related studies at Kinloch, Isle of Rhum, Scotland, with particular reference to possible early human impacts on vegetation. New Phytologist 116, 715–727. Irwin, H., Curtis, C., and Coleman, M., 1977. Isotopic evidence for source of diagenetic carbonates formed during burial of organic rich sediments. Nature 269, 209–213. Kane, T.J. and Chester, S.G., 1993. Lidar observations of the meteoric deposition of mesospheric metals. Science 259, 1297–1300. Kolesnikov, E.M., Stepanov, A.I., Goridko, E.A., and Kolesnikova, N.V., 1998. Element and isotopic anomalies in peat from the Tunguska explosion (1908) area are probably traces of cometary matter. Meteoritics and Planet. Sci. 33 (Suppl.), A85. Legrand, M., Preunkert, S., Jourdain, B., and Aumont, B., 2004. Year-round records of gas and particulate formic and acetic acids in the boundary layer at Dumont d’Uville, coastal Antarctica. J. Geophys. Res., D: Atmospheres 109 (6), D06313 1–11. Markgraf, V., 1993. Younger Dryas in southernmost South America: an update. Quatern. Sci. Rev. 12, 351–355. McCafferty, E.D. and Owen, D.E., 1996. Stratigraphic Study of Elk Park Wetland, Gilpin County, Colorado. U.S. Geological Survey, Reston, VA, U.S.A., 12pp. McCormick, M.P., Thomason, L.W., and Trepte, C.R., 1995. Atmospheric effects of the Mt Pinatubo eruption. Nature 367, 399–404. McLennan, S.M. and Taylor, S.R., 1981. Role of subducted sediments in island-arc magmatism: constraints from REE patterns. Earth Planet. Sci. Lett. 54, 423–430. Murozumi, M., Chow, T.J., and Patterson, C., 1969. Lead, dust, and salt in ﬁrn and ice from Camp Century and Byrd station. Antarctic J. United States 4, 218–219. Murray, S. and Renard, A.F., 1891. Rept. Sci. Results Voyage H.M.S. Challenger, Neill and Co., Vol. 7, 327pp. Nazarov, M.A., Korina, M.I., Kolesov, G.M., and Vasilyev, N.V., 1983. The Tunguska event; mineralogical and geochemical data. Proceedings of the Lunar and Planetary Science Conference 14, Part 2, pp. 548–549. Parish, J.T. and Lamberson, M.N., 1998. Cretaceous (Nanushuk Group, Albanian-Cenomanian) wetland environments of the North Slope of Alaska. Geological Society of America, 1998 annual meeting (Abstracts with programs), Vol. 30, p. 366. Pederstad, K. and Aagaard, P., 1985. Interstitial waters and diagenetic changes of marine sediments from Oslofjord and Skagerrak. Norsk Geologisk Tidskrift 4, 287–299. Perkins, S.M., Filipelli, G.M., and Siuch, C.J., 2000. Airborne trace metal contamination of wetland sediments at Indiana Dunes National Lakeshore. Water Air Soil Pollut. 122, 231–260. Persson, C., 1971. Tephrochronological investigation of peat deposits in Scandinavia and on the Faroe Islands. Swe. Geol. Surv. Ser. C. 656, 3–33. Pettersson, H. and Fredriksson, K., 1958. Magnetic spherules in deep-sea deposits. Paciﬁc Science 12, 71–81.

268

L.G. Franze´n

Pilcher, J.R. and Hall, V.A., 1992. Towards a tephrochronology for the Holocene of the north of Ireland. Holocene 2, 255–259. Pilcher, J.R. and Hall, V.A., 1996. Tephrochronological studies in northern Ireland. Holocene 1, 100–105. Pitka¨nen, A., 2000. Fire frequency and forest structure at a dry site between AD 440 and 1110 based on charcoal and pollen records from a laminated lake sediment in eastern Finland. Holocene 2, 221–228. Preunkert, S., Legrand, M., and Wagenbach, D., 2001. Sulfate trends in a Col du Dome (French Alps) ice core: a record of anthropogenic sulfate levels in the European midtroposphere over the twentieth century. J. Geophys. Res., D Atmosphere 106 (23), 31991–32004. Rampino, M.R., Self, S., and Stothers, R.B., 1988. Volcanic winters. Ann. Rev. Earth Planet. Sci. 16, 73–99. Rampino, M.R., Stothers, R.B., and Self, S., 1985. Climatic effects of volcanic eruptions. Nature 313, 272. Raukas, A., 1997. An attempt to use microimpactites in establishing the age of impact events on the example of the Kaali crater ﬁeld, Sphaerula. International J. IGCP 384, 32–41. Raukas, A., Tiirmaa, R., Kaup, E., and Kimmel, K., 2001. The age of Ilumetsa meteorite craters in Southeast Estonia. Meteoritics and Planet. Sci. 36, 1507–1514. Renberg, I., Bindler, R., and Bra¨nnvall, M.-L., 2001. Using the historical atmospheric lead-deposition record as a chronological marker in sediment deposits in Europe. Holocene 11, 511–516. Renberg, I., Bra¨nnvall, M.-L., Bindler, R., and Emteryd, O., 2000. Atmospheric lead pollution history during four millennia (2000 BC–2000 AD) in Sweden. AMBIO 29, 150–156. Rogers, G.M. and McGlone, M.S., 1994. A history of Kaiparoro clearing and the limits of Nothofagus in the northern Tararua Range, New Zealand. NZ J. Bot. 32, 463–482. Samaniego, P., Monzier, M., Robin, C., and Hall, M.-L., 1998. Late Holocene eruptive activity at Nevado Cayambe volcano, Equador. Bull. Volcanol. 59, 451–459. Scott, A.C. and Jones, T.P., 1994. The nature and inﬂuence of ﬁre in Carboniferous ecosystems. Palaeogeog. Palaeoclimatol. Palaeoecol. 106, 91–112. Seibold, E. and Berger, W.H., 1982. The Sea Floor – An Introduction to Marine Geology. Springer, New York, 288pp. Self, S. and R.S.J. Sparks (Eds), 1981, Tephra studies. Riedel, Dordrecht, . Sikora, L.J. and Keeney, D.R., 1983. Further aspects of soil chemistry under anaerobic conditions. In: Gore, A.J.P. (Ed.), Mires: Swamp, Bog, Fen and Moor. General Studies. Elsevier, Amsterdam, Vol. A, pp. 247–256. Simmons, I.G. and Innes, J.B., 1988. Late vegetational history of the North York Moors. X. Investigations on East Bilsdale Moor. J. Biogeogr. 15, 299–324. Sjo¨rs, H., 1980. Peat on earth: multiple use or conservation. AMBIO 9, 303–308. Sjo¨rs, H., 1982. The zonation of northern peatlands and their importance for the carbon balance of the atmosphere. In: Gopal, B., Turner, R.E., Wetzel, R.G., Whigham, D.F. (Eds), Wetlands: Ecology and Management. Proceedings of the First International Wetlands Conference, New Dehli, India, 10–17 September, pp. 11–14. Smith, L.C., MacDonald, G.M., Velichko, A.A., et al., 2004. Siberian peatlands a net carbon sink and global methane source since the early Holocene. Science 303, 353–356. Souch, C.J., Filipelli, G.M., Dollar, N., and Perkins, S., 2002. Accumulation rates of airborne heavy metals in wetland. Phys. Geogr. 23, 21–43. Taylor, P.L., Nusbaum, R.L., Fronabarger, A.K., et al., 1996. Magnetic spherules in coastal plain sediments, Sullivan’s Island, South Carolina, USA. Meteoritics and Planet. Sci. 31, 77–80. Taylor, S., Lever, J.H., and Harvey, R.P., 1998. Accretion rate of cosmic spherules measured at the South Pole. Nature 392, 899–903. Tolonen, M., 1985. Palaeoecological record of local ﬁre history from a peat deposit in SW Finland. Ann. Bot. Fenn. 22, 15–29. Valentin, J., 1902. Der Staubfall vom 9. bis 12 Ma¨rz 1901. Berichte der Akademie vin Wissenschaften, Wien, 50pp. Vasilyev, N.V., Bronskiy, B.I., Demin, D.V., et al., 1971. Silicate spherules in peat near the point of impact of the Tunguska Meteorite. Transactions of the USSR Academy of Sciences; Earth Science Sections 199, 212–214.

Mineral matter, major elements, and trace elements in raised bog peat

269

Verreccia, E.P., Freytet, P., Verrecchia, K.E., and Dumont, J.-L., 1995. Spherulites in calcrete laminar crusts: biogenic CaCO3 precipitation as a major contributor to crust formation. J. Sed. Res. A65, 690–700. Veski, S., Heinsalu, A., Kirimae, K., et al., 2001. Ecological catastrophe with the impact of the Kaali meteorite about 800–400 B.C. on the island of Saaremaa, Estonia. Meteoritics and Planet. Sci. 36, 1367–1375. Wilmshurst, J.M. and McGlone, M.S., 1996. Forest disturbance in the central North Island, New Zealand, following the 1850 BP Taupo eruption. Holocene 6, 399–411. Wood, J.A., 1963. Physics and chemistry of meteorites. In: Middlehurst, B.M. and Kuiper, G.P. (Eds), The Moon, Meteorites and Comets. University of Chicago Press, Chicago, pp. 337–401. Wu Hong, 1992. Black carbon in polar ice cores; the palaeo-record of forest ﬁre activity over the past 13,000 years. M.Sc Thesis, Dalhouise University, Halifax, Canada, 64pp. Zbik, M. and Gostin, V.A., 1995. Morphology and internal structure of Antarctic cosmic dust spherules: possible links to meteorite fusion crusts. Proc. NIRP Symp. Antarct. Meteorites 8, 339–351. Zdanowicz, C.M., Zielinski, G.A., and Wake, C.P., 1998. Characteristics of modern atmospheric dust deposition in snow on the Penny ice cap, Bafﬁn Island, Arctic Canada. Tellus, Ser. B, Chem. Phys. Meteorol. 50, 506–520. Zielinski, G.A. and Germani, M.S., 1998. New ice-core evidence challenges the 1620s BC age for the Santorini (Minoan) eruption. J. Archaeol. Sci. 25, 279–289. Zielinski, G.A., Mayewski, P.A., Meeker, L.D., et al., 1994. Record of volcanism since 7000 B.C. from the GISP2 Greenland Ice Core and implications for the Volcano-Climate System. Science 264, 948–952.

This page intentionally left blank

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

271

Chapter 12

Consequences of increasing levels of atmospheric nitrogen deposition on ombrotrophic peatlands: a plant-based perspective L. Bragazza

Introduction Nitrogen (N) is a fundamental nutrient for forms of life on Earth. Most of the N (about 98%) is lithospheric and, being ﬁrmly linked to inorganic mineral phases, plays little part in the biogeochemical cycle of the element. The second big N reservoir is the atmosphere, containing about 2% of the total N of the Earth. The biosphere and the hydrosphere contain the lowest portion of total N (o0.01%), although N in the biosphere is highly reactive and rapidly recycled. In spite of the fact that the total amount of N in the lithosphere, hydrosphere, and atmosphere is greater than all other macronutrients combined, that is, carbon (C), phosphorus (P), oxygen (O), and sulfur (S) (Mackenzie, 1998), N is the element least readily available for living organisms. In other words, there is a seeming paradox in that all terrestrial organisms live in an atmosphere containing 78% by volume of N, yet N is commonly the limiting nutrient in an ecosystem. The reason of course is that only a very small fraction of N in the biosphere is in a form available to plants. The largest amount of N in the biosphere is atmospheric N2, a molecular species, chemically un-reactive under ambient conditions. The reactive N includes all those N forms biologically and chemically active such as ammonia (NH3), ammonium (NHþ 4 ), nitrogen oxide (NOx), nitric acid (HNO3), nitrous oxide (N2O), nitrate (NO 3 ), and organic N forms (proteins, enzymes, urea, nucleic acids). Under natural conditions, reactive N is created by biological ﬁxation and lightning, with the former processes assumed to ﬁx about 130–330 Tg yr1, two orders of magnitude greater than the latter (Galloway et al., 1995). In addition, another characteristic of the N cycle in a pre-human world is the balance between biological ﬁxation and denitriﬁcation without signiﬁcant N accumulation within ecosystems or redistribution among ecosystems (Galloway et al., 1995). The N cycle has been gradually but dramatically altered as human population has increased, especially since the Industrial Revolution. In particular, on land the anthropogenic ﬁxation of reactive N is equal to the amount of naturally ﬁxed N ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09012-2

272

L. Bragazza

(Galloway et al., 1995), the driving force being the need to feed a population that has been growing exponentially (Frink et al., 1999). Speciﬁcally, the main causes for increasing supplies of N in natural and semi-natural ecosystems (in decreasing order of magnitude) are: the widespread use of synthetic fertilizers obtained through the Haber–Bosch process, cultivation of leguminous crops, and fossil fuel combustion. The massive supply of reactive N in the environment is causing deleterious effects from deep in the ground to high in the stratosphere. Although the synthetic ﬁxation of N has provided a means to support the development of human society, the unwise use of fertilizers causes many concerns not only in a pure ecological context but also in relation to the negative socio-economical effects for human life (Townsend et al., 2003). On the whole, the major concerns related to the altered N cycle can be brieﬂy summarized as follows: (1) Production of tropospheric ozone with connected ecological human-health problems. (2) Alteration of nutrient balance of natural ecosystems with additional negative effects on biodiversity. (3) Acidiﬁcation and eutrophication of aquatic ecosystems. (4) Contribution to global warming and stratospheric ozone depletion. Alteration of the N cycle has particularly dramatic effects on those ecosystems that are both nutrient-poor and dependent on atmospheric inputs for their nutrient and mineral supplies, a typical example being the ombrotrophic peatland (bog). Peatlands are formed where there is an imbalance between plant biomass production and plant biomass decomposition causing a net accumulation of partially decomposed plant remnants (peat). Such processes take place where there is a surplus of precipitation over evapotranspiration, associated with suitable geomorphologic conditions favoring the creation of anoxic conditions. Ombrotrophic peatlands are unaffected by local groundwater and maintain their own suspended water table. Bogs receive nutrients exclusively from atmospheric sources, so they become mineral and nutrient poor (Bridgham et al., 1996). From a biological point of view, the most peculiar feature of ombrotrophic peatlands is the presence of a dense mat of Sphagnum plants forming the bulk of living and dead biomass. Sphagnum is a genus of bryophytes adapted to thrive under low-nutrient availability (van Breemen, 1995). Like all bryophytes Sphagnum plants absorb nutrients by direct inﬂux through the single cell-layered leaves. Few vascular plants are able to live in ombrotrophic peatlands due to the low-nutrient availability, the acidity of the substrate, the anoxic conditions, and the low temperature of the rooting environment (van Breemen, 1995). Peatlands are carbon sinks (Gorham, 1991). To what extent increased nitrogen deposition can alter the biogeochemistry and biodiversity of bogs is becoming a matter of concern, particularly because no decrease of N inputs is foreseeable. The objective of this chapter is therefore to analyze the effect of increasing N availability in ombrotrophic peatlands, with particular attention to the effects on plant ecology at both species and community level.

Consequences of increasing atmospheric nitrogen deposition

273

Effects of increasing N input: the plant species level Mosses and vascular plants employ different mechanisms of nutrition. Consequently, effects on the two taxonomic groups are discussed here separately. Nevertheless, strong interactions exist between mosses and vascular plants, as is reported below. Mosses Most of the studies on N effects on peatland mosses focus on the genus Sphagnum that forms the bulk of the biomass in bogs. Sphagnum mosses have been demonstrated to absorb exogenous N at greater rates under low than under high-atmospheric N input (Bragazza et al., 2004). This behavior is related to the N-limited conditions experienced by Sphagnum plants growing in relatively unpolluted areas. Under increasing amounts of N availability, Sphagnum plants gradually become N-saturated, and K and P become the limiting nutrients (Fig. 12.1). The depositional threshold of (bulk) N deposition separating N-limited from P+K co-limited conditions has been estimated to be ca. 1 g m2 yr1 for European bogs, a depositional threshold which is suggested as the N-critical load for European bogs (Gunnarsson and Rydin, 2000; Bragazza et al., 2004). This critical load is reﬂected in N/P and N/K ratios of living Sphagnum plants 430 and 3.5, respectively (Fig. 12.1). Saturation of the Sphagnum layer under high atmospheric N input is accompanied by an increase of dissolved inorganic (DIN) and organic (DON) N forms in bog-pore þ water. Increased DIN (NO 3 and NH4 ) concentrations have been suggested to be a consequence of reduced N retention ability of the moss layer under N saturating

Figure 12.1. Variation of N: P and N: K ratios in Sphagnum plants along a natural gradient of bulk atmospheric N deposition in Europe (after Bragazza et al., 2004).

274

L. Bragazza

conditions of Sphagnum tissues (Bragazza et al., 2005). Instead, increasing concentrations of DON have been explained as an enhanced production of organic exudates by Sphagnum plants as a means of mitigating the negative effects of high-N concentrations in living tissues (Bragazza and Limpens, 2004). An important role of plant exudates has also been claimed by Freeman et al. (2004) to explain the increase of DOM carbon export from bogs under increasing CO2 concentrations. Different studies of N fertilization experiments have demonstrated a decrease of linear growth of Sphagnum under enhanced N availability (Gunnarsson and Rydin, 2000; Berendse et al., 2001; Limpens et al., 2003a; Gunnarsson et al., 2004). The decrease was much more consistent the closer the Sphagnum was to the level of N saturation in the plant tissue. Indeed, for Sphagnum plants that are N-limited, an additional N input initially favors the moss growth, but beyond the saturation point in the tissues, further input of N appears detrimental to Sphagnum growth (van der Heijden et al., 2000). Two main reasons may explain such a pattern: (1) the metabolic toxicity of excess N accumulated in Sphagnum tissues as expressed by increasing amino acids contents and changes of internal N relocation between the capitulum and the stem (Limpens and Berendse, 2003a; Bragazza et al., 2005); and (2) the increased interaction with other organisms favored by enhanced inorganic N availability in pore water (Heijmans et al., 2002a; Limpens et al., 2003a; see also below the ‘community level’ topic). Different atmospheric N sources affect the N-isotopic signature of Sphagnum plants (Bragazza et al., 2005). Indeed, when NOx forms predominate in atmospheric precipitation, the d15N signature of Sphagnum tends to have less negative values than when NHx forms predominate (Fig. 12.2). Dominance of NOx forms in atmospheric

Figure 12.2. Variation of mean isotopic d15N signature in ombrotrophic Sphagnum plants from 13 bogs located in the Czech Republic (CZ), Sweden (S), Norway (N), Finland (FIN), Denmark (DK), the Netherlands (NL), United Kingdom (UK), Italy (I), Ireland (IRE), and Slovenia (SLO). Concentrations of NH4 and NO3 in bulk precipitation were obtained by national organizations responsible for precipitation monitoring (N isotopic signature of Sphagnum plants were redrawn after Bragazza et al., 2005).

Consequences of increasing atmospheric nitrogen deposition

275

precipitation is mainly due to fossil–fuel combustion, whereas the main sources of NHx forms are related to agricultural activities. Vascular plants Fertilization experiments carried out in ombrotrophic peatlands reported an increase of vascular plant biomass and vascular plant cover under increasing N inputs. For example, Tomassen et al. (2003) showed an increase of Molinia caerulea biomass after N addition, although they did not ﬁnd an increase of Betula pubescens that suffered a P-limitation under enhanced N availability. Limpens et al. (2003a) found a positive relationship between M. caerulea and Rhynchospora alba biomass with concentration of inorganic N in pore water, but also in this case the interaction with other factors such as water table depth and P availability may signiﬁcantly affect above-ground production of vascular plants. Heijmans et al. (2001) reported an increase of vascular plant biomass after N fertilization, particularly in the case of the shallow rooting Vaccinium oxycoccus. Chapin et al. (2004) did not show any signiﬁcant effects of N fertilization on graminoids, whereas shrubs responded positively to N fertilization. The response of vascular plants was strongly species-speciﬁc in the experiment carried out by Thormann and Bayley (1997) with some species increasing and other decreasing their above-ground biomass production. Similar individual responses of vascular plants to environmental manipulations were also found when other environmental parameters were modiﬁed, such as temperature and peat moisture (Weltzin et al., 2003). In general, the response of bog vascular plants to N additions is not only speciesspeciﬁc but also even within the same life forms the responses can be different. This conclusion makes it more difﬁcult to predict the response to eutrophication of vascular plant communities as a whole. Two main reasons can be claimed to explain such variability: (1) the role played by other environmental factors in affecting above biomass production under enhanced N availability, in particular P availability and water-table position, and (2) the short term of fertilization experiments.

Effects of increasing N input: the community level The effects of increased N depositions at the community level is addressed looking at some major processes that are supposed to be particularly sensitive to eutrophication; that is, gas emission ﬂuxes, plant inter-speciﬁc relationships, as well as peat accumulation and litter decomposition. Gas fluxes The role of increased N deposition on the rates of methane (CH4) and carbon dioxide (CO2), two gases studied in bog ecology for their role in global warming, are assessed.

276

L. Bragazza

Methane (CH4) is the result of degradation of organic matter under anoxic conditions. The amount of CH4 emissions from northern peatlands is estimated to be about 7% of total global methane emission to the atmosphere (Bartlett and Harriss, 1993), the latter corresponding to about 540 Tg yr1 (Mosier, 1998). The rates of CH4 emission from a peatland is the result of the balance between CH4 production by methanogenic bacteria and CH4 oxidation by methanotrophic bacteria. Indeed, once methane has been produced in the anoxic layer, it can reach the atmosphere by ebullition, diffusion, or passing through the aerenchyma of vascular plants. When CH4 ascends through an oxic region of a bog it can be oxidized to CO2, and its initial concentration may decrease up to about 80% (Fechner and Hemond, 1992). Vascular plants play an important role in methane dynamics by supplying methanogenic bacteria with an organic substrate through litter deposition and, particularly, root exudates (Conrad, 1996). In addition, vascular plants favor the transport of methane via aerenchymous tissues (Joabsson and Christensen, 2001). Vascular plants can also play a suppressing role on methane balance through its oxidation in the rhizosphere just before the onset of aerenchymous transport (Conrad, 1996). The effects of N additions on methane emissions from bogs are contrasting. One of the main reasons can be attributed to the short duration of fertilization experiments that do not allow the detection of the gradual changes associated with long-term, chronic inputs of atmospheric N. Nitrogen addition can affect directly or indirectly CH4 emissions. Direct effects include both inhibition of methanogenesis (decreased emission) by nitrate (Watson and Nedwell, 1998), and the inhibition of oxidation (increase emission) by ammonium (Crill et al., 1994). Indirect effects primarily concern the alteration of vascular plant cover due to increased nutrient availability (Nykanen et al., 2002). The consequences of increased N input on methane emission from peatlands are, therefore, related to the predominance of positive or negative effects of improved N availability. For example, Aerts and de Caluwe (1999), on the basis of laboratory incubation experiments, suggested that long-term exposure of peatlands to high-N input leads to higher methane emissions. A moderate increase of CH4 emission following N addition was also reported by Saarnio et al. (2000) and was explained by the relatively short-lasting fertilization that did not permit the excess N to reach the deeper peat layers. In ﬁeld experiments lasting six years, Nykanen et al. (2002) showed a signiﬁcant increase of CH4 emission particularly on nutrient poor sites where N addition increased the cover of Eriophorum vaginatum. Granberg et al. (2001) reported no, or slightly positive effects of, N deposition on CH4 emissions, with substantial negative effects where sedge cover was high. The authors explained the negative role of sedges by an increased allocation of C to shallow roots, enhanced by superﬁcial nutrient additions, so decreasing the development of deep roots. In a European survey of CH4 emission, Silvola et al. (2003) reported no signiﬁcantly higher CH4 emission rates under increasing levels of N additions, explaining such a trend with the minor or null response of vegetation cover to N addition. The main sources of CO2 in bogs are related to aerobic and anaerobic decay of organic matter, oxidation of CH4, and rhizosphere respiration. The measurements of CO2 ﬂuxes can be a way to assess decomposition rates alternative to the measure of mass-loss of plant remnants kept in mesh bags (the so called ‘litter bags’ method). Anyway, to be sure that CO2 emissions exclusively reﬂect the decomposition activity

Consequences of increasing atmospheric nitrogen deposition

277

of soil organisms it is necessary to exclude any other CO2-producing sources not related to litter decay, namely rhizosphere respiration. This is the reason why CO2 measurements in peat decomposition studies are carried out through peat incubation under controlled laboratory experiments (Hogg, 1993; Aerts and de Caluwe, 1999; Scanlon and Moore, 2000). On the other hand, ﬁeld studies of CO2 ﬂuxes using chamber or tower measurements permit the assessment of the balance between the amount of CO2 ﬁxed by photosynthesis and the amount emitted by total community respiration, obtaining the net ecosystem exchange (NEE) of CO2. If NEE is positive then the peatland acts as a C sink, but if aerobic respiration dominates over photosynthesis, then NEE is negative and the peatland is a C source. Emission CO2 rates during litter decomposition are particularly affected by C quality of the peat (Bridgham and Richardson, 1992), and by aeration of the peat (Scanlon and Moore, 2000). On the other hand, inter-annual variability of NEE is strongly related to water-table position (Waddington and Roulet, 2000; Laﬂeur et al., 2003), which regulates the moisture content of the Sphagnum affecting the photosynthetic yield of the moss layer (Silvola and Aaltonen, 1984), and to the peat temperature that affects the degree of community respiration (Bridgham and Richardson, 1992). Few papers deal with the relationships between CO2 ﬂuxes and N deposition both in incubation experiments, and in ﬁeld NEE measurements. For example, in a laboratory incubation experiment using peat columns Aerts and De Caluwe (1999) reported a negative effect of increased atmospheric N input on CO2 emission. The claimed reason was two sided. On one side, high-N inputs were supposed to cause an acidiﬁcation of the soil through nitriﬁcation of NHþ 4 N (Likens et al., 1996); on the other side, highN supplies were supposed to negatively affect the activity of lignin-degrading enzymes of the fungi community (Waldrop et al., 2004). Different results were obtained by Saarnio et al. (2003) in a ﬁeld experiment where high-N inputs did not signiﬁcantly alter the soil CO2 emissions. In a fertilization experiment carried out in a bog in the Italian Alps during a very dry summer it was found that increasing N deposition mitigated the negative effects of dryness by reducing CO2 emission from hummocks (that is, relatively drier microhabitats), and increasing CO2 retention by carpets (relatively wetter microhabitats) compared to the control plots (Fig. 12.3). Additional input of phosphorus (P) greatly increased the NEE of CO2 reducing the nutrient constraints of plants (Fig. 12.3). The positive role of N addition in favoring C accumulation or, at least, in reducing C emission, can be likely related to the N-limited conditions of Sphagnum plants actually receiving, in the study alpine bog, a N input lower than the critical load suggested for European bogs (Bragazza et al., 2004). It is worth remembering that these gas ﬂuxes represent short-term effects, which do not incorporate changes in vegetation and peat composition as a result of long-term exposure to increased atmospheric N deposition.

Inter-specific competitive relationships Competition plays an important role in affecting the structure and the functioning of all ecosystems. The outcome of competitive interaction depends on the metabolic

278

L. Bragazza

Figure 12.3. Average (7standard deviation) net ecosystem exchange (NEE) of CO2 in ﬁve hummocks and ﬁve carpets in an alpine bog on southeastern Alps of Italy during summer 2003 in relation to three different fertilization treatments. N0P0 ¼ control plots receiving distilled water; N3P0 ¼ plots receiving an additional N input of 3 g m2 yr1 and no phosphorus (P0); N3P1 ¼ plots receiving an additional N input (see above) plus and additional P input of 0.25 g m2 yr1. Negative NEE indicates CO2 emission, whereas positive NEE indicates CO2 uptake. Averages are based on gas measurements performed four times at each habitat during the growing season (Bragazza et al., unpublished data).

requirements of the different species and/or the different individuals. Natural or human-induced changes in environmental conditions are expected to alter the competitive ability of interacting species with cascade consequences on ecosystem stability. When bogs are subject to increased N availability, experimental data showed an alteration of competitive equilibriums both among the different Sphagnum species, and between Sphagnum plants and vascular plants. For example, Gunnarsson et al. (2004) have reported an alteration of competitive equilibriums between Sphagnum balticum and S. lindbergii under increased N inputs. Although both these two Sphagnum species can naturally coexist in ombrotrophic peatlands (though S. lindbergii can also be found in minerotrophic peatlands in mid- and southern Sweden), a different ability to tolerate increasing exogenous N availability caused an increase of S. lindbergii cover with detrimental effects on S. balticum. The outcome can be explained by taking into account the major tolerance to N eutrophication of minerotrophic than strictly ombrotrophic Sphagnum species (Jauhiainen et al., 1998). Limpens et al. (2003c) have reported another example of alteration of competitive equilibria where S. fallax was shown to take an advantage over the co-existing S. papillosum and S. magellanicum under increased nutrient input, and the competitive ability of S. fallax was much more effective if a suitable range of environmental conditions were met, primarily in relation to P availability and water-table position (Fig. 12.4). Alteration of competitive equilibriums between Sphagnum and vascular plants is becoming a major concern in bog conservation strategies due to the competitive advantage shown by vascular plants under increased N availability. As we have

Consequences of increasing atmospheric nitrogen deposition

279

Figure 12.4. Mean percentage expansion (71 standard error) of Sphagnum fallax intermixed with S. magellanicum and S. papillosum in a Dutch bog during three years of nitrogen (N) and phosphorus (P) fertilization. The background of bulk atmospheric N deposition was about 2 g m2 yr1. Control ¼ plots receiving only distilled water; N ¼ plots receiving 4 g N m2 yr1; P ¼ plots receiving 3 g P m2 yr1. Expansion of S. fallax was based on percentage cover variation calculated using the point intercept method (after Limpens et al., 2003b).

Figure 12.5. Biomass (a) and 15N allocation (b) for three different bog compartments, that is, the apical living portion of Sphagnum magellanicum (S. mag 0–5 cm), deep peat (Peat 15–30 cm), and vascular plants (Vasc. plants) under ambient and increased N deposition in a Dutch mesocosm experiment. Biomass changes are based on a three-year long fertilization experiment, whereas 15N allocation is calculated as percentage of total 15N amount retrieved 15 months after addition. Signiﬁcant differences inside each compartment between the N treatments are indicated: * po0.05, ** po0.01 (after Heijmans et al., 2002a).

previously seen, under N saturation Sphagnum mosses fail to absorb N inputs so that an increase of N concentration in pore water is observed (Limpens et al., 2003a; Bragazza and Limpens, 2004; Bragazza et al., 2005). The reduced N retention by a Sphagnum layer favors the expansion of vascular plants through a greater N availability in the rhizosphere (Fig. 12.5).

280

L. Bragazza

Figure 12.6. Average percentage cover (7 standard deviation) of N. ossifragum in relation to Sphagnum cover in six Swedish poor-fens. Sphagnum cover was obtained bulking cover data of Sphagnum magellanicum and S. affine. The number of replicates within each Sphagnum cover class was between 4 and 14 (after Malmer et al., 2003).

Although Sphagnum plants and vascular plants are supposed to rely on different nutrient sources (Malmer et al., 2003), a combination of nutrient alteration and increased shading by vascular plants seems to explain the competitive disadvantage experienced by Sphagnum mosses under increased N availability. Direct competition for light has been reported by Malmer et al. (2003) between Narthecium ossifragum (a perennial, clonal vascular plants) and Sphagnum mosses (Fig. 12.6). In particular, mire sites characterized by a greater cover of N. ossifragum also showed a lower cover of the moss layer. The negative effect of shading has been suggested as taking place above a threshold of vascular plant cover greater than 50–60% (Heijmans et al., 2002b; Limpens et al., 2003a). In addition, a high cover of vascular plants exerts an additional negative effect on Sphagnum layer through litter deposition (Berendse et al., 2001). Another example of alteration of inter-speciﬁc interaction triggered by high-N inputs was studied by Limpens et al. (2003b), concerning a parasitic fungus, epiphytic algae, and Sphagnum. In particular, infection intensity by Lyophyllum palustre (a fungus fairly common in bogs) and expansion of epiphytic algae on Sphagnum plants were positively associated with increasing N concentration in Sphagnum tissues. The outcomes were a reduced vitality of Sphagnum plants due to defoliation caused by fungus infection, as well as reduced photosynthetic capacity caused by the expansion of epiphytic algae over Sphagnum. On the whole, alteration of inter-speciﬁc competitive ability forces vegetation to change over time and space much faster than natural succession rates under pristine conditions. For example, rapid ﬂoristic changes over 13 years in a United Kingdom bog were explained by Hogg et al. (1995) to be related to increasing atmospheric N deposition causing acidiﬁcation and eutrophication. Similar effects associated with N deposition were reported by Gunnarsson et al. (2000) who noted a stronger

Consequences of increasing atmospheric nitrogen deposition

281

decrease of water pH from 1945 to 1995 in a Swedish bog in correspondence with relatively less acidophilus plant communities. The effects of forced acidiﬁcation due to both N and sulfur deposition caused a decrease of species richness in the marginal fen areas, but not in the more acidic inner portion of the bog (Gunnarsson et al., 2000). Peat accumulation and peat decomposition Can increased atmospheric N deposition stimulate carbon (C) accumulation in ombrotrophic peatlands? Peat accumulation rates depend on the balance between biomass production and biomass decomposition. Because Sphagnum plants tend to increase their biomass and their vertical growth under N-limited conditions if exogenous N increases, we can then suppose that under N unsaturated conditions C accumulation can be enhanced under increased N availability. Such a conclusion has been corroborated by Turunen et al. (2004) who found an increase of C accumulation rates over the last 50 years in eastern Canadian peatlands receiving a maximum amount of N deposition of 0.7–0.8 g m2 yr1 (Fig. 12.7). Litter decomposition plays a fundamental role in controlling net C accumulation in bogs so that the consequences that N eutrophication could have on peat decomposition are a matter of concern, particularly if peatlands are forced to change from C sinks to C sources. Decomposition rates primarily depend on microhabitat conditions (particularly in relation to soil aeration), litter chemistry, and microbial activity in the soil (Aerts, 1997). Slow decomposition rates in peatlands are related to the poor-nutrient content of peat litter (low N and P concentration) and to the large concentrations of phenolic compounds in Sphagnum plants (van Breemen, 1995). Whether enhanced N availability can ameliorate the litter chemistry is becoming a matter of debate. Some papers support the view that increased endogenous N concentrations in plant litter

Figure 12.7. Mean total C mass (7 1 standard error) in hollows and hummocks during the last 50 years in eastern Canadian peatlands under increasing levels of atmospheric N deposition (after Turunen et al., 2004).

282

L. Bragazza

enhances the short-term decomposition (Aerts et al., 2001; Limpens and Berendse, 2003b), but others do not show any signiﬁcant effect (Rochefort et al., 1990; Bridgham and Richardson, 2003). On the whole, the initial quality of plant litter is an important factor in controlling decomposition rates in bogs, particularly through the concentration of soluble phenols that are responsible for hampering litter decomposition (Bridgham and Richardson, 2003). This is in accordance with the positive relationship found by Limpens and Berendse (2003b) between N/C quotient of litter and decomposition rates. Anyway, it is worth remembering that other nutrients such as P can constrain decomposition (Aerts et al., 2001). Recent ﬁndings indicate that increased exogenous nutrient availability does not affect the decay rates, at least over a short period (Bridgham and Richardson, 2003; Limpens and Berendse, 2003b). Instead, a major role is played by endogenous-nutrient concentrations of peat litter, which, in their turn, are related to the initial tissue chemistry of the living plants. On the other hand, plant tissue chemistry in nutrient-poor environments changes very gradually under increasing N inputs so that a signiﬁcant effect of increasing N deposition on decomposition rates is expected to occur only when the original plant species with low-tissue quality will be replaced by plant species with better litter quality (Bridgham and Richardson, 2003). This means that an expansion of vascular plants is expected to cause an increase in decomposition rates. If this is the case then long-term C-accumulation rates in peatlands may decrease as a result of increased atmospheric N deposition.

Conclusions Under a globally increasing trend of atmospheric N depositions, no ecosystem on Earth seems to escape the threat of ongoing eutrophication. The consequences of this forced, anthropogenic fertilization were demonstrated to alter the composition and functioning of bogs so strongly as to likely convert these peatlands from net C sinks into net C sources. To obtain a better understanding of the effects associated with exogenous N inputs on bogs some topics deserve further attention in future studies. In particular, a few of the questions that need answers are:

(1) What is the feedback between increasing N deposition and increasing CO2 concentration on plant litter quality and related-decomposition rates? (2) To what extent are CO2 and CH4 ﬂuxes affected under a scenario of future global warming and increasing eutrophication? (3) What is the capacity of bogs to return to their initial-ombrotrophic conditions if N deposition should decrease?

Trying to answer these and other questions should help to shed light on the responses of peatlands to the global change by contributing a better understanding of the feedbacks between world biomes and human-induced alterations.

Consequences of increasing atmospheric nitrogen deposition

283

References Aerts, R., 1997. Climate, leaf litter chemistry and leaf litter decomposition in terrestrial ecosystems: a triangular relationship. Oikos 79, 439–449. Aerts, R. and de Caluwe, H., 1999. Nitrogen deposition effects on carbon dioxide and methane emissions from temperate peatland soils. Oikos 84, 44–54. Aerts, R., Walle´n, B., Malmer, N., and De Caluwe, H., 2001. Nutritional constraints on Sphagnum-growth and potential decay in northern peatlands. J. Ecol. 89, 292–299. Bartlett, K.B. and Harriss, R.C., 1993. Review and assessment of methane emissions from wetlands. Chemosphere 26, 261–320. 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 Biol. 7, 591–598. Bragazza, L. and Limpens, J., 2004. Dissolved organic nitrogen dominates in European bogs under increasing atmospheric N deposition. Global Biogeochem. Cycles 18, GB4018, doi: 10.1029/2004GB002267. Bragazza, L., Limpens, J., Gerdol, R., et al., 2005. Nitrogen concentration and d15N signature of ombrotrophic Sphagnum mosses at different N deposition levels in Europe. Global Change Biol. 11, 106–114. Bragazza, L., Tahvanainen, T., Kutnar, L., et al., 2004. Nutritional constraints in ombrotrophic Sphagnum plants under increasing atmospheric nitrogen depositions in Europe. New Phytol. 163, 609–616. Bridgham, S.D., Pastor, J., Janssens, J.A., et al., 1996. Multiple gradients in peatlands: a call for a new paradigm. Wetlands 16, 45–65. Bridgham, S.D. and Richardson, C.J., 1992. Mechanisms controlling soil respiration (CO2 and CH4) in southern peatlands. Soil. Biol. Biochem. 24, 1089–1099. Bridgham, S.D. and Richardson, C.J., 2003. Endogenous versus exogenous nutrient control over decomposition and mineralization in North Carolina peatlands. Biogeochemistry 65, 151–178. Chapin, C.T., Bridgham, S.D., and Pastor, J., 2004. pH and nutrient effects on above-ground net primary production in a Minnesota, USA bog and fen. Wetlands 24, 186–201. Conrad, R., 1996. Soil micro-organisms as controllers of atmospheric trace gases (H2, CO, CH4, OCS, N2O, and NO). Microbial Rev 60, 609–640. Crill, P.M., Martikainen, P.J., Nykanen, H., and Silvola, J., 1994. Temperature and N fertilization effects on methane oxidation in a dried peatland soil. Soil Biol. Biochem. 26, 1331–1339. Fechner, E.J. and Hemond, H.F., 1992. Methane transport and oxidation in the unsaturated zone of a Sphagnum peatland. Global Biogeochem. Cycles 6, 33–44. Freeman, C., Fenner, N., Ostle, N.J., et al., 2004. Dissolved organic carbon export from peatlands under elevated carbon dioxide levels. Nature 430, 195–198. Frink, C.R., Waggoner, P.E., and Ausubel, J.H., 1999. Nitrogen fertilizer: retrospect and prospect. Proc. Natl. Acad. Sci. USA 96, 1175–1180. Galloway, J.N., Schlesinger, W.H., Levy, H. II, et al., 1995. Nitrogen ﬁxation: anthropogenic enhancement-environmental response. Global Biogeochem. Cycles 9, 235–252. Gorham, E., 1991. Northern peatlands: role in carbon cycle and probable responses to climatic warming. Ecol. Appl. 1, 182–195. Granberg, G., Sundh, I., Svensson, B.H., and Nilsson, M., 2001. Effects of temperature, and nitrogen and sulfur deposition, on methane emission from a boreal mire. Ecology 82, 1982–1998. Gunnarsson, U., Granberg, G., and Nilson, M., 2004. Growth, production and interspeciﬁc competition in Sphagnum: effects of temperature, nitrogen and sulphur treatments on a boreal mire. New Phytol. 163, 349–359. Gunnarsson, U. and Rydin, H., 2000. Nitrogen fertilization reduces Sphagnum production in bog communities. New Phytol. 147, 527–537. Gunnarsson, U., Rydin, H., and Sjo¨rs, H., 2000. Diversity and pH changes after 50 years on the boreal mire Skattlo¨sbergs Stormosse, Central Sweden. J. Veg. Sci. 11, 277–286. Heijmans, M.M.P.D., Berendse, F., Arp, W.J., et al., 2001. Effects of elevated carbon dioxide and increased nitrogen deposition on bog vegetation in the Netherlands. J. Ecol. 89, 268–279. Heijmans, M.M.P.D., Klees, H., and Berendse, F., 2002b. Competition between Sphagnum magellanicum and Eriophorum angustifolium as affected by raised CO2 and increased N deposition. Oikos 97, 415–425.

284

L. Bragazza

Heijmans, M.M.P.D., Klees, H., de Visser, W., and Berendse, F., 2002a. Effects of increased nitrogen deposition on the distribution of 15N-labeled nitrogen between Sphagnum and vascular plants. Ecosystems 5, 500–508. Hogg, E.H., 1993. Decay potential of hummock and hollow Sphagnum peats at different depths in a Swedish raised bog. Oikos 66, 269–278. Hogg, P., Squires, P., and Fitter, A.H., 1995. Acidiﬁcation, nitrogen deposition and rapid vegetational change in a small valley mire in Yorkshire. Biol. Conserv. 71, 143–153. Jauhiainen, J., Silvola, J., and Vasander, H., 1998. The effects of increased nitrogen deposition and CO2 on Sphagnum angustifolium and S. warnstorfii. Ann. Bot. Fenn. 35, 247–256. Joabsson, A. and Christensen, T.R., 2001. Methane emissions from wetlands and their relationship with vascular plants: an Arctic example. Global Change Biol. 7, 919–932. Laﬂeur, P.M., Roulet, N.T., Bubier, J.L., et al., 2003. Inter annual variability in the peatland-atmosphere carbon dioxide exchange at an ombrotrophic bog. Global Biogeochem. Cycles 17(2), 1036, doi: 10.1029/2002GB001983. Likens, G.E., Driscoll, C.T., and Buso, D.C., 1996. Long-term effects of acid rain: response and recovery of a forest ecosystem. Science 272, 244–245. Limpens, J. and Berendse, F., 2003a. Growth reduction of Sphagnum magellanicum subjected to heavy nitrogen deposition: the role of amino nitrogen concentration. Oecologia 135, 339–345. Limpens, J. and Berendse, F., 2003b. How litter quality affects mass loss and N loss from decomposing Sphagnum. Oikos 103, 537–547. Limpens, J., Berendse, F., and Klees, H., 2003a. N deposition affects N availability in interstitial water, growth of Sphagnum and invasion of vascular plants in bog vegetation. New Phytol. 157, 339–347. Limpens, J., Raymakers, J.T.A.G., Baar, J., et al., 2003b. The interaction between epiphytic algae, a parasitic fungus and Sphagnum as affected by N and P. Oikos 103, 59–68. Limpens, J., Tomassen, H.B.M., and Berendse, F., 2003c. Expansion of Sphagnum fallax in bogs: striking the balance between N and P availability. J. Bryol. 25, 83–90. Mackenzie, F.T., 1998. Our Changing Planet, an Introduction to Earth System Science and Global environmental Change. Upper Saddle River, Prentice-Hall, NJ, 486pp. Malmer, N., Albinsson, C., Svensson, B.M., and Walle´n, B., 2003. Inferences between Sphagnum mosses and vascular plants: effects on plant community structure and peat formation. Oikos 100, 469–482. Mosier, A.R., 1998. Soil processes and global change. Biol. Fertil. Soils 27, 221–229. Nykanen, H., Vasander, H., Huttunen, J.T., and Martikainen, P.J., 2002. Effect of experimental nitrogen load on methane and nitrous oxide ﬂuxes on ombrotrophic boreal peatland. Plant Soil 242, 147–155. Rochefort, L., Vitt, D.H., and Bayley, S.E., 1990. Growth, production, and decomposition dynamics of Sphagnum under natural and experimentally acidiﬁed conditions. Ecology 71, 1986–2000. Saarnio, S., Ja¨rvio¨, S., Saarinen, T., et al., 2003. Minor changes in vegetation and carbon balance in a boreal mire under a raised CO2 and NH4NO3 supply. Ecosystems 6, 46–60. Saarnio, S., Saarinen, T., Vasander, H., and Silvola, J., 2000. A moderate increase in the annual CH4 efﬂux by raised CO2 and NH4NO3 supply in a boreal oligotrophic mire. Global Change Biol. 6, 137–144. Scanlon, D. and Moore, T., 2000. Carbon dioxide production from peatland soil proﬁle: the inﬂuence of temperature, oxic/anoxic conditions and substrate. Soil Sci. 165, 153–160. Silvola, J. and Aaltonen, H., 1984. Water content and photosynthesis in the peat mosses Sphagnum fuscum and S. angustifolium. Ann. Bot. Fenn. 21, 1–6. Silvola, J., Saarnio, S., Foot, J., et al., 2003. Effects of elevated CO2 and N deposition on CH4 emissions from European mires. Global Biogeochem. Cycles 17(2), 1068, doi:10.1029/2002GB001886. Thormann, M.N. and Bayley, S.E., 1997. Response of aboveground net primary plant production to nitrogen and phosphorus fertilization in peatlands in southern boreal Alberta, Canada, Wetlands 17, 502–512. Tomassen, H.B.M., Smolders, A.J.P., Lamers, L.P.M., and Roelfs, J.G.M., 2003. Stimulated growth of Betula pubescens and Molinia caerulea on ombrotrophic bogs: role of high levels of atmospheric nitrogen deposition. J. Ecol. 91, 357–370. Townsend, A.R., Wowarth, R.W., Bazza, F.A., et al., 2003. Human health effects of a changing global nitrogen cycle. Front. Ecol. Environ. 1, 240–246.

Consequences of increasing atmospheric nitrogen deposition

285

Turunen, J., Roulet, N.T., Moore, T., and Richard, P.J.H., 2004. Nitrogen deposition and increased carbon accumulation in ombrotrophic peatlands in eastern Canada. Global Biogeochem. Cycles 18, GB3002, doi:10.1029/2003GB002154. van Breemen, N., 1995. How Sphagnum bogs other plants. Trends Ecol. Evol. 10, 270–275. van der Heijden, E., Verbeek, S.K., and Kuiper, P.J.C., 2000. Elevated atmospheric CO2 and increased nitrogen deposition: effects on C and N metabolism and growth of the peat moss Sphagnum recurvum P. Beauw. var. mucronatum (Russ.) Warnst. Global Change Biol. 6, 201–212. Waddington, J.M. and Roulet, N.T., 2000. Carbon balance of a boreal patterned peatland. Global Change Biol. 6, 87–97. Waldrop, M.P., Zak, D.R., and Sinsabaugh, R.L., 2004. Microbial community response to nitrogen deposition in northern forest ecosystems. Soil Biol. Biochem. 36, 1443–1451. Watson, A. and Nedwell, D.B., 1998. Methane production and emission from peat: the inﬂuence of anions (sulphate and nitrate) from acid rain. Atm. Environ. 32, 3239–3245. Weltzin, J.F., Bridgham, S.D., Pastor, J., et al., 2003. Potential effects of warming and drying on peatland plant community composition. Global Change Biol. 9, 141–151.

This page intentionally left blank

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

287

Chapter 13

Microbial diversity in Sphagnum peatlands D. Gilbert and E.A.D. Mitchell

Introduction Peat accumulates because the net production of organic matter exceeds its decomposition by microorganisms. Peatlands, and especially Sphagnum-dominated peatlands, were at one time erroneously believed to be devoid of microbial life. In reality, and despite the successful use of Sphagnum mosses as surgical dressings, diapers, or menstrual pads, Sphagnum mosses and peatlands in general are home to a high diversity of microorganisms. The numbers of published works on the subject are relatively low, probably because of the technical difﬁculties due to the abundance of organic matter particles that make direct observations challenging and the range and variability of water content of the substrate (Given and Dickinson, 1975; Gilbert et al., 2000b). The existing studies fall into two categories: the taxonomical approach and the functional approach. In the taxonomical approach, the different groups of microorganisms (bacteria, protists, fungi, and micro-metazoa) are usually studied independently and the focus is usually on the taxonomy of the species restricted to peatlands. By contrast, little or no attention is given to abundance, biomass, and auto- and heterotrophic production. Recent studies of whole microbial communities, including all groups of microorganisms and measurements of abundance, biomass, and auto- and heterotrophic production, have demonstrated how microbial communities can be useful, ﬁrst for the characterization of environmental conditions (Fisher et al., 1998) at the surface of peatlands, and second for assessing the effects of perturbations on these ecosystems (Gilbert, 1998a, b; Mitchell et al., 2003). In a long-term perspective (millennia), these studies are related to the use of microbial indicators such as testate amoebae for the reconstruction of past environmental conditions in paleoecological studies (Charman et al., 1999; Charman, 2001; Mitchell et al., 2001). In the functional approach, the focus is on the role of microorganisms in the cycling of nutrients, mainly carbon and nitrogen, in the ecosystem. These processes remain poorly understood in peatlands, partly because of the lower funding for research in these ecosystems as compared to others more directly economically ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09013-4

288

D. Gilbert and E.A.D. Mitchell

relevant. Nevertheless, the important role of peatlands in the global carbon cycle, as carbon pools, sinks, and sources, and the increasing concern about the anthropogenic inﬂuence on the greenhouse effect, have resulted in a recent increase in the number of studies in microbial ecology (methanogenic and methanotrophic bacteria) and biogeochemistry of peatlands (Edwards et al., 1998). In a similar way, research on peatland restoration is now starting to include aspects of microbial ecology such as microbial density or respiration (Croft et al., 2001), or microbial community structure (Chapman et al., 2003). Our goal in this review is to synthesize the existing knowledge on various aspects of microbial ecology in peatlands with a special focus on studies that include data on abundance and biomass. Thus, although bacterial and fungal communities are by far the most studied microbial groups in peat soils, this review focuses more on other microbial groups (auto- and heterotrophic protists and micro-metazoa). We also aim to demonstrate the interest of integrating these groups of microorganisms, and in particular testate amoeba, in studies of present or past perturbations of peatlands. Examples of some common testate amoebae are illustrated in Figure. 13.1, and a selection of microorganisms in living Sphagnum is illustrated in Figure. 13.2.

Microbial diversity in peatlands Overview of the sampling, observation, and biomass estimation methods Sampling and fixation Because peatlands represent intermediate conditions between mineral soils and aquatic environments, the sampling and extraction methods for microorganisms depend on the kind of material sampled (peat, litter, or mosses) and the water content of the samples. Peat samples are usually taken as cores, which are subsequently sliced (usually in 1–10-cm-thick slices) to analyze speciﬁc depths. Although peat coring is relatively easy–because of the absence of coarse mineral material, the local presence of woody remains can cause problems and the water saturation creates a suction effect that makes removing the cores difﬁcult and creates risks of damaging them during extraction. In this case, the main risk is a compaction of the peat that makes it difﬁcult to establish with conﬁdence the exact depth of a given sample. Another risk in case of very decomposed peat is the upward or downward movement of material during coring that can cause microbiological contamination. To overcome these problems speciﬁc methodologies and equipments have been developed over the years. One such example is a double corer that allows the extraction of intact peat cores for the top 1–1.5 m of peat (Buttler et al., 1998). Sampling in Sphagnum or other mosses can depend on the water content. In very wet conditions (bog pools or wet fens), it is possible to sample water by simply exerting a pressure on the moss surface. The advantage of this method is that it does not destroy the vegetation. However, this method is not optimal, because many microorganisms may remain attached to the mosses. A more reliable method is to

Microbial diversity in Sphagnum peatlands 289

Figure 13.1. Examples of testate amoebae. (a) Trigonopyxis arcula. (b) Hyalosphenia subflava. (c) Bullinularia indica. (d) Nebela tincta. (e) Nebela militaris. (f) Assulina muscorum. (g) Assulina seminulum. (h) Arcella arenaria. (i) Hyalosphenia elegans. (j) Physochila (Nebela) griseola. (k) Hyalosphenia papilio. (l) Centropyxis aculaeta. (m) Amphitrema flavum. (n) Placocista spinosa. (o) Difflugia bacillifera. (p) Nebela carinata. (q) Amphitrema wrightianum (Scale bars indicate approximately 50 mm except for Figure f (A. muscorum) where scale is 20 mm.

290

D. Gilbert and E.A.D. Mitchell

Microbial diversity in Sphagnum peatlands

291

sample the mosses and extract the microorganisms. One important aspect to keep in mind when preparing a sampling protocol is the spatial heterogeneity of microbial communities. Mitchell et al. (2000a), for example, demonstrated that a signiﬁcant spatial heterogeneity existed in testate amoebae communities within a macroscopically homogeneous 40 60 cm surface of Sphagnum magellanicum. Spatial autocorrelation was signiﬁcant up to a distance of about 15 cm and the structure of communities was signiﬁcantly correlated to the almost ﬂat micro-topography (maximum height difference within the surface: 6.6 cm). Therefore, although sampling a small number of mosses, or even only the capitulum (top 1 cm) of a Sphagnum moss may yield a sufﬁcient number of microorganisms for community characterization, if the aim of the study is to characterize the general community structure of a study area, then it is preferable to sample a larger surface, or to pool several small samples from a given area. The question of extracting microorganisms from Sphagnum has not yet been studied in detail with the exception of testate amoebae (Hendon and Charman, 1997). Nevertheless, the microscopic observation of Sphagnum mosses reveals that some microorganisms are able to enter the large, hollow hyaline cells though the small pores (mostly 5–15 mm) that connect the inside of the cells with the surrounding environment. Although these spaces are clearly easily accessible to bacteria and small protists such as ﬂagellates, even relatively large microorganisms such as testate amoebae, and micro-metazoans such as nematode and rotifers have been observed in these cells. Microorganisms living in these spaces beneﬁt from the double advantage of a physical barrier against predators and a wet environment that allows them to remain active longer during periods of dry weather.

Figure 13.2. Microorganisms in living Sphagnum and their role. Living Sphagnum micro-ecosystem: (a) Sphagnum mosses dominate vast expanses of nutrient-poor peatlands. They produce organic matter and provide habitat for microorganims. Some cells, the hyalocysts, lose their content and retain water. Microorganisms can live either between the leaves or inside the hyalocysts. Microbial photosynthesis: (b, c) Cyanobacteria: in the green part of Sphagnum mosses, non-ﬁlamentous cyanobacteria (b. Choococcus sp.) or ﬁlamentous cyanobacteria (c. Anabanena sp) use the sunlight to produce organic matter. Some species also ﬁx atmospheric nitrogen in specialized cells, the heterocysts (see arrow, picture c). (d, e) Autotrophic protists : green algae (d. Penium sp.) and diatoms (e. Eunotia sp.) are two of the most important groups of microalgae in peatlands. (f, g) Mixotrophic protists: in Sphagnum, many protozoa are mixotrophic. Some testate amoebae (f. Amphitrema flavum) and ciliates (g. Uronema sp.) contain symbiotic algae. This symbiosis probably gives these protozoa a trophic advantage and helps them meet their nutritional requirements in nutrient-poor habitats. Microbial food chains: (h–j) Hetrotrophic protists: microbial food webs are complex. Many ﬂagellates (i. heterotrophic ﬂagellates observed in epiﬂuorescence microscopy after a primuline coloration) and ciliates (g, h. Uronema sp. 1 and 2) prey on bacteria (see arrow in picture i1) or algae. Testate amoebae have a very broad range of prey. Some species are able to catch very large preys such as rotifera, ciliates, and even nematodes (j. two individuals of the species Nebela collaris (stars) are eating a rotifera, see arrow). (k) Small-size metazoa: Rotifera (k. Bdelloı¨ da) and Nematoda, most of which are believed to be predators of bacteria, represent a minor proportion of the relative microbial biomass. Decomposition: (l) Hetrotrophic bacteria: bacteria are the most abundant microbial group in Sphagnum and play a major role in the decomposition of labile organic matter. In picture 1, millions of bacteria (grey points, see arrow) are degrading a testate amoeba (Hyalosphenia papilio), just after its death. (m, n) Fungi: fungi degrade the more resistant organic matter (cellulosic) (m. fungi hyphae). Many fungal spores are also found in Sphagnum (n. conidia of Helicoons sp.).

292

D. Gilbert and E.A.D. Mitchell

The ﬁxation and preservation of peatland microorganisms is not different from that of other environments (glutaraldehyde, formaldehyde) (Fisher et al., 1998; Gilbert et al., 1998a, b), but, for Sphagnum moss samples, the water content (about 95%) must be taken into consideration, as it will cause a dilution of the ﬁxing solution. Observation and counting The enumeration of microorganisms in peatlands can be done using several methods. None of these methods is speciﬁc to peatlands but we present them brieﬂy here. Culture methods for bacteria and fungi on gels were mainly used in the early studies of peat microbiology. These methods underestimate the abundance and diversity of microorganisms in soils and water by 90–99% because the culture conditions have little in common with the conditions of the natural environment (Given and Dickinson, 1975). This problem is likely to be even more acute in acidic peatlands because of the unique combination of environmental conditions (low nutrient, pH, temperature, redox state). For this reason, the direct counting method is considered to be the most reliable method for estimating microorganism densities. Bacteria are enumerated under epiﬂuorescence microscopy after being stained by a ﬂuorochrome, usually DAPI (Porter and Feig, 1980). In peat, the abundance of minute peat particles (o2 mm) makes the counting more difﬁcult by masking the bacteria and absorbing the ﬂuorochrome. Thus, higher concentrations of ﬂuorochrome are needed as well as the assistance of an image analysis system capable of accumulating the light signal over several tenths of a second. Picocyanobacteria (with diameter o2 mm) can also be enumerated using epiﬂuorescence microscopy using their autoﬂuorescence property. Heterotrophic ﬂagellates and other small size protists can be counted using ﬂuorochromes such as primulin (Caron, 1983). Larger organisms, such as cyanobacteria (42 mm), protists, and micro-metazoa, are most commonly analyzed using the Utermo¨lh’s (1958) method that consists in placing a volume of water containing microorganisms in a plankton chamber slide, letting them settle, and counting them using an inverted microscope. This method allows the enumeration of organisms and the estimation of biovolumes, which can then be converted to biomass. However, this method does not allow identifying all of the microorganisms at the species level, which takes some additional preparation. In addition, the Uthermo¨l method is time-consuming despite the help of imageanalyzing systems. The observation of fungal hyphae and spores using this method is not difﬁcult but it seems that it is very difﬁcult to extract fungi from Sphagnum mosses or from the peat in which they grow. Thus, direct methods certainly underestimate the biomass of fungi by an unknown proportion. For autotrophic microorganisms, epiﬂuorescence can also be used with plankton chambers, but it is usually not possible to identify the algae to the species level; for this, 1000 magniﬁcation and special preparations are usually required. Testate amoebae are unusual among peatland microorganisms in that they build a shell (called ‘test’) that is very resistant to physical and chemical degradation. For this reason, and the relative ease of identiﬁcation based on the test morphology, testate amoebae are the best-studied group

Microbial diversity in Sphagnum peatlands

293

of protists in peatlands. Testate amoebae are usually studied using a standard upright microscope. Phase contrast microscopy is useful for the observation of genera such as Euglypha, Trinema, and Corythion, which produce siliceous scales almost invisible under bright-ﬁeld microscopy. The analysis of microbial RNA or DNA allows assessment of the diversity of microorganisms, and to some extent their abundance in the environment. This and other molecular methods have the advantage of being faster than direct counts. Furthermore, they provide information on the genotypic diversity and not only the morphological diversity. This is especially important for bacteria, for which direct observation methods do not allow distinguishing the different species, as well as for some groups of protists, for which the taxonomy in insufﬁciently documented in peatlands so that only broad morphological groups can be distinguished using direct methods of observation. The coupling of direct observations and molecular techniques, such as the FISH method of ﬂuorescence in situ hybridization should allow in future more precise enumeration of different groups of bacteria and minute, hyaline protists. Other techniques, such as those using the microbial fatty acids (phospholipid fatty acids, or PLFA) have been used in a few studies to describe the structure of microbial communities (Borga et al., 1994; Sundh et al., 1995; Sundh et al., 1997; Schmidt et al., 2000; Cole et al., 2002). Biomass evaluation Using direct methods, the abundance of microorganisms and the biomass of each species or morphological type can be determined. From the morphology and biometric measurements of cells or organisms it is possible to estimate relatively precisely their speciﬁc biovolume and then, by taking into account the relative abundance of each type and, by using conversion factors, to convert the biovolume into biomass. However, none of the conversion factors currently used were determined from peatland microorganisms; instead conversion factors from marine or freshwater environments or from soils are used (Gilbert et al., 1998a, b; Mitchell et al., 2003). The biomass of microorganisms can also be determined using the fumigation-extraction method for total microbial biomass and using the ergosterol extraction method (for fungi). Individual groups: diversity, abundance, biomass Prokaryotes (1) Methanogenic bacteria constitute a very important functional group of microorganisms in deep peat as they represent one of the main potential sources of CH4 emission to the atmosphere. Using a DNA ampliﬁcation method, Hales et al. (1996) found the diversity of methanogenic bacteria from a peatland in northern England to be very limited. However, Edwards et al. (1998), Galand et al. (2002; 2003), and Sizova et al. (2003) have identiﬁed numerous strains of methanogenic bacteria, some of which were hitherto unknown in deep peat. It also appears that different species of

294

D. Gilbert and E.A.D. Mitchell

methanogens are sensitive to the heterogeneity of the environment as attested by both horizontal and vertical patterns. For example, in an Eriophorum lawn of a Finnish fen, Galand et al. (2002) observed that the upper waterlogged layers contained methanogens belonging to a previously unknown group of Archaea related to the Methanomicrobiales, which they named ‘fen cluster’, whereas the deeper layers contained hydrogenotrophs belonging to the Rice cluster I. Interestingly, but perhaps not surprisingly, Galand et al. (2003) also observed differences in the near-surface Archaea communities between micro-sites (hummock versus lawn), whereas those occurring deeper in the peat were less variable. (2) Heterotrophic bacteria constitute the most abundant microbial group in peat. The most frequently encountered species in ombrotrophic peatlands belong to the genera Bacillus, Pseudomonas, Achromobacter, and Arthrobacter (Given and Dickinson, 1975). According to Greaves et al. (1973), the size of cells is generally between 0.3 and 1 mm. Direct counting methods using ﬂuorochromes coloration and epiﬂuorescence microscopy (Hobbie et al., 1977) yield abundance and biomass estimates of heterotrophic bacteria in peat ranging from about 109 to 1010 bacteria per gram peat dry weight and from about 106 to 107 bacteria per milliliter of water extracted from Sphagnum. These numbers are 103–104 times higher than those resulting from culture methods (Table 13.1). At the surface, although the abundance of bacteria is globally positively correlated to temperature, a decrease in the abundance of bacteria and other microorganisms is common during the summer. This phenomenon is especially clear in drained peatlands where the summer drought is most marked (Gilbert, 1998; Potila and Sarjala, 2004). In the peat itself, the abundance of bacteria decreases with depth (Francez ,1991; Sundh et al., 1997) and seasonal variations are less pronounced (Sundh et al., 1997). Methanotrophic bacteria that live at the interface between the aerobic and anaerobic zone beneﬁt directly from the production of methane by Archea in the deeper peat. Their abundance has been estimated to be between 1.0 and 4.3 104 cell mL 1 using the most probable number method in ﬁltered pore water (Williams and Crawford, 1983), and between 0.1 and 51.0 106 cells g 1 of peat using ﬂuorescent antibodies (Vecherskaya et al., 1993) and the phospholipids fatty acids method (Sundh et al., 1995). (3) Cyanobacteria play an important role in peatlands because some species are able to ﬁx atmospheric nitrogen and can thus enrich the ecosystem in this oftenlimiting nutrient (Basilier, 1980). The input of nitrogen due to these organisms is signiﬁcant. It is estimated to be about equivalent to atmospheric wet and dry deposition (Hemond, 1983). Cyanobacteria can be divided into four orders. Three of them comprise ﬁlamentous forms. Among them, two orders (the Nostocales and the Stigonematales) produce heterocysts and are able to ﬁx atomospheric nitrogen in oxic conditions (Anabaena). The fourth order, the Chroococcales, comprise unicellular (and in some case colonial forms) and non-heterocystic species (Chlorococcus). The dominant genera of Cyanobacteria in peatlands are Anabaena, Aphanocapsa, Aphanothece, Chroococcus, Eucapsis, and Synechococcus (Dell’Uomo, 1981; Lederer, 1995a,b). For practical reasons, cyanobacteria are usually enumerated simultaneously with microalgae. Abundance and biomass results are therefore presented further together with microalgae.

Densities and biomasses of heterotrophic bacteria in Sphagnum peatlands.

Environment sampled Type

Density

Sphagnum water Sphagnum water Sphagnum water Sphagnum water Sphagnum water Sphagnum Sphagnum Mixed peat Peat (Review) Peat: 0–50 cm Peat: 5–70 cm Peat: 5 cm Peat: 5 cm Peat: 5 cm Peat: 15 cm Peat: 25 cm Peat: 30 cm Peat: 32 cm Peat: 32 cm Peat: 50 cm Peat: 80 cm

5–140 103 cells mL–1 2.2 104 cells mL–1 7.3 106 cells mL–1 4.4 107 cells mL–1 2.6–13.2 107 cells mL–1 0.8–22 -106 cells g–1(DW)

Aerobic Aerobic Total Total Total Total Total Aerobic Aerobic Total Methanotrophic Aerobic Aerobic Cellulolytic Aerobic Aerobic Aerobic Aerobic Anaerobic Aerobic Aerobic

Biomass

Method

Culture Culture 0.5 mgC mL–1 Fluorescence 3.9 mgC mL–1 Fluorescence Fluorescence Fluorescence 0.09–1.0 mgC g–1 (DW) Fluorescence 6 –1 0.2 10 cells g (DW) Culture 0.03–5.5 106 cells g–1(DW) Culture 2–11.9 109 cells g–1(DW) Fluorescence 0.3-51 106 cells g–1(DW) Indirect 1.61 1010 cells g–1(DW) 4.5 mgC g–1 (DW) Culture 6 –1 0.08–90 10 cells g (DW) Culture 2.7–0.87 105 cells g–1(DW) Culture 4-6.3 106 cells g–1(DW) Culture 2.1 104 cells mL–1 Culture 3.5 106 cells g–1(DW) Culture 1.5 1010 cells m–2 0.3 g m–2 (DW) Culture 4 108 cells m–2 Culture 7 103 cells mL–1 Culture 2 103 cells mL–1 Culture

Reference Grolie`re (1977) Francez (1991) Gilbert et al. (1998a) Gilbert et al. (1998a) Fisher et al. (1998) Mitchell (1999) Mitchell et al. (2003) Gardiner (1975) Given and Dickinson (1975) Greaves et al. (1973) Sundh et al. (1995) Clarholm and Rossswall (1980) Hiroki and Watanabe (1996) Hiroki and Watanabe (1996) Croft et al. (2001) Francez (1991) Waskman & Stevens (1929) Martin et al. (1982) Martin et al. (1982) Francez (1991) Francez (1991)

Microbial diversity in Sphagnum peatlands

Table 13.1.

Note:DW, dry weight.

295

296

D. Gilbert and E.A.D. Mitchell

Fungi Microscopic fungi, together with heterotrophic bacteria, are the main decomposers of organic matter in soils. This group includes a high diversity of forms, many of which grow as ﬁlaments. In northern peatlands the diversity estimates range from 22 to 55 species (Nilsson et al., 1992; Czeczuga, 1993; Thormann et al., 2001). The main genera encountered belong to the three higher groups of fungi: Zygomycota, Ascomycota, Basidiomycota, as well as to the imperfect fungi (Deuteromycota). The abundance of viable fungi ranges between 105 and 106 units per gram of dry peat (Table 13.2). However, as for bacteria, this kind of enumeration provides little useful information because it does not correlate to the real fungal biomass. From the direct observation of ﬁlaments, which represent up to 12.4 m g–1 dry peat, Collins et al. (1978) estimated this biomass at 1.4 mg g–1 dry peat in the litter of a raised bog. Using direct counts of ﬁlaments and spores, Mitchell et al. (2003) estimated the fungal biomass to range between 0.1 and 0.2 mgC g–1 Sphagnum dry weight at the surface of ﬁve different European Sphagnum-dominated peatlands, whereas Gilbert et al. (1998b) estimated it at 0.8 mgC mL–1 in Sphagnum extraction water. As noted earlier, it seems likely that values obtained from Sphagnum water extraction underestimate the real fungal biomass substantially because most of the ﬁlaments are tightly bound to organic matter. The ratios among heterotrophic bacteria, yeasts, and fungi in the top 3 cms of peat and during spring was estimated to be 1:4:5 by Wynn-Williams (1982), whereas Golovchenko et al. (1994) estimated fungi to represent 80–99% of the total microbial biomass in peat. The relative higher importance of fungi as compared to heterotrophic bacteria in peatlands is likely due to their higher tolerance to acidity. In support of this, cellulose-degrading fungi were shown to be more abundant in peatlands than their bacterial counterparts (Hiroki and Watanabe, 1996). Fungi respond to ecological gradients and experimental changes in environmental conditions. However, the addition of nitrogen or agricultural fertilizers, exposure to elevated atmospheric CO2, and manipulations of solar ultraviolet light B (UVB) did not cause any signiﬁcant change in fungal biomass (Gilbert et al., 1998a, b; Mitchell et al., 2003; Robson et al., 2004). But this lack of overall response may, however hide a species-speciﬁc response as attested by the response of fungi to the manipulation of solar UVB in a peatland in Tierra del Fuego (Robson et al., 2004) and the response of some micro-fungi to the nutrient content of plant litter (Thormann et al., 2003). The respective role of fungi in litter decomposition in peatlands is still relatively poorly understood. Thormann et al. (2002) have found 55 species of fungi to be associated with Sphagnum fuscum. In a detailed study, they found nine of these to be able to degrade tannic acid and cellulose and generally concluded that fungi were able to degrade a wide variety of carbon substrates. These same authors (Thormann et al., 2003) further demonstrated the existence of a temporal succession of fungal communities during the decomposition of Sphagnum fuscum probably due to the changes in the biochemical characteristics of the substrate during the decomposition process.

Densities and biomasses of fungi in Sphagnum peatlands.

Environment sampled

Type

Sphagnum water Sphagnum Sphagnum Peat: surface Peat: surface Peat: surface Peat: surface Peat: surface Peat: surface Peat: 5 cm Peat: 5 cm Peat: 5 cm Peat: 15 cm Peat: 25 cm Peat: 50 cm Peat: 80 cm

Hyphae Hyphae Total Total Total Total Yeast Hyphae Total Total Hyphae Cellulolytic Total Total Total Total

Density 5

0.5–18.4 10 ind. g

Biomass

Method

Reference

0.81 mgC mL–1

Direct Direct Direct Culture Culture Culture Culture Direct Culture Culture Direct Culture Culture Culture Culture Culture

Gilbert et al. (1998b) Mitchell (1999) Mitchell et al. (2003) Waskman and Stevens (1929) Gardiner (1975) Francez (1991) Given and Dickinson (1975) Given and Dickinson (1975) Francez (1991) Hiroki and Watanabe (1996) Collins et al. (1978) Hiroki and Watanabe (1996) Croft et al. (2001) Francez (1991) Francez (1991) Francez (1991)

1

0.03–0.4 mgC. g–1 (DW) 5

–1

10 ind. g (FW) 2.5 105 ind. g–1 (DW) 4.8 105 ind g–1 (DW) 2.9 106 ind. g–1 (DW) 160–320 m cm–3 2.7 103 ind. mL–1 0.02–10 106 ind. g–1 (DW) 1–12.4 m g–1 (DW) 0.5–30 104 ind. g–1 (DW) 0.2–3.2 105 cells g–1 (DW) 1.3 103 ind. mL–1 0.2 103 ind. mL–1 0.1 103 ind. mL–1

0.4–5.2 mgC g–1 (DW)

Microbial diversity in Sphagnum peatlands

Table 13.2.

Note: DW, dry weight; FW, Fresh weight.

297

298

D. Gilbert and E.A.D. Mitchell

Microalgae From a taxonomic standpoint, microalgae (autotrophic protists) and cyanobacteria are the best-studied groups of microorganisms in peatlands, together with testate amoebae. Autotrophic protists are highly polyphyletic and the organisms that are grouped under this name are spread across at least four of the eight major groups of eukaryotes (Baldauf, 2003). For example, green algae, such as Desmids, are related to higher plants, whereas diatoms are related to brown algae and euglenoids belong to yet another major group of eukaryotes and are related to the human parasites Leishmania and Trypanosoma (Baldauf, 2003). The only common characteristics of these different groups of organisms are that they are eukaryotes, contain chloroplasts, and are either unicellular or aggregated into colonies with no vascular tissue. When all the different taxonomic groups are pooled together, microalgae are the most diverse group of microorganisms in peatlands although estimates vary considerably. Villeret (1955) identiﬁed 198 species of microalgae and cyanobacteria in peatlands of Brittany, whereas Cosandey (1964) found over 400 species in a Swiss peatland, Compe`re (1980) listed 248 species in peatlands of France and Belgium, and Mataloni (1999) observed 299 species in seven peatlands of Tierra del Fuego. However, Wutrich and Matthey (1978, 1980) identiﬁed 362 species of diatoms alone in a single peatland of the Jura Mountains of Switzerland, and estimate this group to represent 20 to 60% of the total diversity in peatland ponds. Mataloni cites several other earlier studies of microalgae in peatlands in which more information can be found (Mataloni and Tell, 1996; Mataloni, 1999). According to Dell’Uomo (1981), the 240 species identiﬁed in a small peaty lake in Italy were dominated by the green algae (46% of the species, 17% being desmids) and diatoms (27%), whereas euglenoids and cyanobacteria each represented 8%. Among the numerous green algae found in Sphagnum, desmids constitute the dominant group. The genera Closterium, Staurastrum, Cosmarium, Micrasterias, and Mesotaenium are thus frequently cited (Dell’Uomo, 1981; Dell’Uomo and Agostinelli, 1990; Dell’Uomo and Pellegrini, 1993a, b; Tomaszewicz, 1994; Gilbert, 1998a, b). Pennate diatoms are more abundant than centric forms in peatlands, the genera Eunotia, Achnanthes, Cymbella, Frustulia, Gomphonema, Navicula, Nitzschia, and Pinnularia being the most common (Dell’Uomo, 1981; Gilbert, 1998). Other groups of microalgae are less frequent, but genera, such as Cryptomonas (Cryptophyceae), Euglena, and Trachelomonas (Euglenophyceae) commonly occur. Mataloni (1999) studied the spatial succession of microalgae along a hydrological gradient from open water in bog pools to emerged mosses 320 cm from the pool margin in seven peatlands of Tierra del Fuego. Along these transects, the diversity (genera and species richness) decreased. The relative frequency of all desmids decreased and they were replaced by species from other taxonomic groups that could tolerate the drier conditions, lower pH and higher conductivity (Mataloni, 1999). In the same study, Mataloni also observed that microalgal communities of bog pools differed in relation to the morphological and chemical characteristics of the pools. These observations conﬁrm those of Kingston who noted a decrease in diversity from rich fen to poor fen and to bog hummocks (Kingston, 1982). In agreement with these studies, Poulı´ ckova´ et al. (2004) showed that the distribution of the diatom species

Microbial diversity in Sphagnum peatlands

299

responded primarily to the water depth and pH in acidic mineral poor spring fens of the Carpathian Mountains. Few data concerning the abundance of microalgae are available (Table 13.3). The abundance data range between 103 and 107 cell L–1, depending on authors and sites, which corresponds to between 3 and 1041 mg L–1 of chlorophyll a and 3.8 mgC L–1 (Table 13.3). According to Gilbert (1998a, b), diatoms represent 34% of the total algal biomass, euglenoids 22%, and other microalgae (mostly green algae) 44% in Sphagnum fallax. The seasonal succession of microalgae is strongly inﬂuenced by climatic conditions. Duthie (1965) and Gilbert (1998a, b) have both clearly shown the existence of two peaks of development for microalgae, the ﬁrst in spring and the second in fall. Like other microorganisms, microalgae are susceptible to be passively transported by animals. For example, Wutrich and Matthey (1978) washed 12 snipes (Gallinago gallinago L.) caught near a Swiss lowland lake. They estimated the overall number of diatoms to exceed 3 million individuals, or about 250,000 per bird. A total of 119 species belonging to 29 genera were observed in these extracts. In a follow-up study, these same authors also estimated the potential for various aquatic insects to transport diatoms. From the analysis of 44 insects belonging to 11 species, they estimated the total number of diatoms to be 13,215 belonging to 24 species. Wind alone was estimated to bring 350,000 diatoms per square meter over a one-year period (Wuthrich and Matthey, 1980). Thus, signiﬁcant numbers of microorganisms can be transported by migratory birds, insects, and wind (Wuthrich and Matthey, 1978, 1980). Heterotrophic protists Heterotrophic protists include here the three groups (heterotrophic ﬂagellates, ciliates, and amoebae) that traditionally were commonly referred to as ‘protozoa’. This name is no longer valid because the protozoa are not monophyletic (they do not share a common ancestor that would not be shared by other groups of protists). (1) Heterotrophic flagellates. Detailed data on heterotrophic ﬂagellates living in peatlands are very scarce. Different species of heterotrophic ﬂagellates belonging to the genera Chilomonas, Monosiga, and Monas were observed by Henebry and Cairns (1984) on polyurethane supports immersed in peatlands water. Their abundance reached about 107 cell L 1 (Table 13.4). (2) Ciliates. Ciliates were studied quite extensively by Grolie`re (1974–1975, 1975, 1976, 1977). Among the identiﬁed genera, some are ubiquitous (Paramecium, Cyclidium, Urotricha, Prorodon, Spirostomum, Nassula, Frontonia, Vorticella, and Spathidium), whereas others, such as Bryometopus and Climacostomum, are speciﬁc to peatlands. Grolie`re (1974–1975, 1975) described many species found almost exclusively in Sphagnum. Measurements done on Sphagnum water extract yielded an estimated abundance ranging between 0 and 4.2 106 cells L 1 (Table 13.4). Overall abundance peaks were observed at the beginning and end of summer, but this general pattern masks the strong variability among species (Gilbert, 1998a, b; Grolie`re, 1977). (3) Naked amoebae and testate amoebae. Both naked amoebae and testate amoebae occur in peatlands, but testate amoebae (also referred to as testaceans) have been

300

Table 13.3.

Densities and biomasses of algae and cyanobacteria in Sphagnum peatlands. Type

Pool water Pool water Pool water Periphyton Sphagnum water Sphagnum water Sphagnum water Sphagnum water Sphagnum water Sphagnum water Sphagnum water Sphagnum water Sphagnum water Sphagnum water Sphagnum water Sphagnum water Sphagnum Sphagnum Sphagnum water Sphagnum water Sphagnum water Sphagnum Sphagnum

Total Total Total Total Total Total Total Chlorophyceae Chrysophyceae Desmids Desmids Diatoms Diatoms Diatoms Dinophyceae Nanoalgae Microalgae Microalgae Cyanobacteria Cyanobacteria Cyanobacteria Cyanobacteria Cyanobacteria

Note: DW, dry weight.

Density

Biomass

Method

Reference

3–13 mgChl a L 1 10–03 mgChl a L–1

Direct Direct Direct Direct Direct Direct Direct Direct Direct Direct Direct Direct Direct Direct Direct Direct Direct Direct Direct Direct Direct Direct Direct

Henebry and Cairns (1984) Schoenberg and Oliver (1988) Mataloni and Tell (1996) Schoenberg and Oliver (1988) Duthie (1965) Gilbert et al. (1998a) Gilbert et al. (1998a) Duthie (1965) Malatoni and Tell (1996) Tomascewicz (1994) Duthie (1965) Duthie (1965) Malatoni and Tell (1996) Gilbert et al. (1998a) Malatoni and Tell (1996) Schoenberg and Oliver (1988) Mitchell et al. (2003) Mitchell et al. (2003) Malatoni and Tell (1996) Gilbert et al. (1998a) Gilbert et al. (1998a) Mitchell et al. (2003) Mitchell et al. (2003)

0.5–6.4 106 indiv. L–1 0.4–4 10 3

0.1–10 10 cells L

1

mgChl a. L–1

–1

79.6 mgChl a L 1 179 103 mgChl a L–1 0.1–7 103 cells L–1 0–5 106 cells L–1 0.6–125 103 cells L 1 0.1–5 103 cells L–1 0.1 8 103 cells L–1 0–9 106 cells L–1 0.2–4.4 107 cells L–1 0–1.3 106 cells L–1

0.44 mgC L–1 10–53 mgChl a L 1 0.1–1.6 10 1 mgC g–1 (DW)

0.3–12 104 cells g–1 0–2.8 106 cells L–1 0.2–22.4 107 cells L–1 4.5 107 cells L–1 0.6–64 103 cells g

1

0.48 mgC L–1 0.54 mgC L–1 0.1–1.8 10 2 mgC g–1 (DW)

D. Gilbert and E.A.D. Mitchell

Environment sampled

Densities and biomasses of protozoa and micrometazoa in Sphagnum peatlands.

Environment sampled

Type

Density

Sphagnum Sphagnum Sphagnum Sphagnum Sphagnum Sphagnum Sphagnum Sphagnum Sphagnum Sphagnum Sphagnum Sphagnum Sphagnum Sphagnum Sphagnum

Heterotrophic ﬂagellates Heterotrophic ﬂagellates Ciliates Ciliates Ciliates Testate amoeba Testate amoeba Testate amoeba Testate amoeba Rotifera Rotifera Rotifera Nematoda Nematoda Nematoda

0–2.8 104 cells g 1 (DW) 1.6 107 cells L–1 0–0.7 103 cells g–1 (DW) 0–2.6 106 cells L–1 4.2 106 cells L–1 0.8–4.6 104 shells g–1 (DW) 0.6–1.1 103 cells g–1 (DW) 1.2 105 cells L–1 0.13–23 102 shells cm–3 2.9–8.2 104 ind. m–2 1–191 ind. g–1 (DW) 5.3 104 ind. L–1 3.2–11.4 104 ind. m–2 3–142 ind. g–1 (DW) 3.2 104 ind. L–1

water water water

water

water

water

Biomass 0.14 mgC L–1

1.01 mgC L–1

0.94 mgC L–1 29–58 mgC. m–2 (FW) 0.27 mgC. L–1 6–21 mgC. m–2 (FW) 0.02 mgC. L–1

Method

Reference

Direct Direct Direct Direct Direct Direct Direct Direct Direct Direct Direct Direct Direct Direct Direct

Mitchell et al. (2003) Gilbert et al. (1998a) Mitchell et al. (2003) Grolie`re (1977) Gilbert et al. (1998a) Warner (1987) Mitchell et al. (2003) Gilbert et al. (1998a) Tolonen et al. (1992) Francez (1986) Mitchell et al. (2003) Gilbert et al. (1998a) Francez (1986) Mitchell et al. (2003) Gilbert et al. (1998)

Microbial diversity in Sphagnum peatlands

Table 13.4.

Note: DW, dry weight.

301

302

D. Gilbert and E.A.D. Mitchell

much more intensively studied in the second part of the 20th century and many new studies have been published in recent years (Bonnet, 1958; Heal, 1964; Scho¨nborn, 1965; Couˆteaux, 1969; Meisterfeld, 1979; Beyens and Chardez, 1984; Warner, 1987; Tolonen et al., 1992; Charman, 1997; Woodland et al., 1998; Bobrov et al., 1999; Mitchell et al., 1999; Mitchell et al., 2000a,b; Booth, 2002; Gilbert et al., 2003). The most common species in Sphagnum-dominated peatlands of Europe and North America are listed in Table 13.5. Like microalgae, testate amoebae are polyphyletic. They belong to at least two taxonomically distinct groups, the testate amoebae with ﬁlose pseudopodia (mostly the Euglyphida) and the Arcellinida, or testate amoebae with lobose pseudopodia (Meisterfeld, 2002a,b; Nikolaev et al., 2005). However, because of the key characteristics these two groups of organisms share (i.e., presence of a shell, size, generation time, and general feeding habits) they are usually studied together in ecological and paleoecological studies of peatlands. Testate amoebae seem to be less affected by climatic variations than other microorganisms and are regularly present throughout the year, thanks in part to their ability to encyst during unfavorable periods. Their abundance is high and can reach 4.6 104 individuals (ind.) per gram dry weight of Sphagnum and 1.2 105 ind. L–1 (Table 13.5, Meisterfeld, 1977). Several authors have studied the vertical micro-distribution of testate amoebae in Sphagnum. One of the main observations being that mixotrophic species such as Hyalosphenia papilio, Heleopera sphagni, and Amphitrema spp., (that contain symbiotic microalgae inside their cytoplasm) preferentially colonize the uppermost, photosynthetic part of the mosses, where their endo-symbionts can photosynthesize, whereas heterotrophic species are found at all depths but dominate the community in the lower part of the mosses (Heal, 1962; Scho¨nborn, 1963; Meisterfeld, 1977; Mitchell and Gilbert, 2004). The composition of testate amoebae communities is primarily controlled by the moisture regime and to a lesser extent by pH (Warner and Chmielewski, 1992; Tolonen et al., 1994; Charman and Warner, 1997; Mitchell et al., 1999; Bobrov et al., 2002; Booth, 2002; Lamentowicz and Mitchell, 2005). In addition, Tolonen et al. (1992) found testate amoebae to be correlated to the trophic status and the concentration of mineral nutrients such as calcium. Finally, the relative importance, in terms of biomass, of testate amoebae in the microbial community as well as the broad range of organisms on which they feed (bacteria, fungi, protists, and micro-metazoa) suggest that they play a key role in the microbial food webs in peatlands (Gilbert et al., 1998a; Mitchell et al., 2003). Micro-metazoa (1) Rotifers. According to the classiﬁcation of Sieburth et al. (1978), small metazoans can be considered as microorganisms when their size does not exceed 200 mm. This is the case for many rotifers, which are also considered as one of the component of microbial food webs (1993). Batut (1965), Francez (1981, 1987, 1988), and Bledzki and Ellison (2003) have established species lists for peatland rotifers. In terms of biomass, the Bdelloidea are dominant and mostly include species of the genera Philodina and Habrotrocha. In addition, many Monogononta species of the genera

Most common testate amoebae taxa occurring in Sphagnum moss samples from peatlands of Europe and North America.

Taxon

Arcellinida Assulina muscorum GREEFF Nebela tincta (LEIDY) Corythion dubium TARANEK Assulina seminulum (EHRENBERG) Hyalosphenia papilio LEIDY Phryganella acropodia (HERTWIG & LESSER) Nebela militaris PENARD Hyalosphenia elegans LEIDY Euglypha compressa (CARTER) Amphitrema flavum ARCHER Euglypha strigosa (EHRENBERG) Euglypha laevis (EHRENBERG) Bullinularia indica (PENARD) Nebela tincta v. major DEFLANDRE Euglypha ciliata EHRENBERG Physochila griseola JUNGb Heleopera sphagni (LEIDY) Arcella arenaria GREFFd Euglypha rotunda WAILES Heleopera rosea PENARD Heleopera petricola LEIDY Trigonopyxis arcula (LEIDY) Difflugia leidyi WAILES

Percentage occurrencec

Taxonomic group a

Filosea

Uncertain

x x x x x x x x x x x x x x x x x x x x x x x

94.0 72.6 71.0 61.8 59.6 55.2 53.9 51.1 44.5 42.9 41.6 40.4 39.7 37.9 37.2 36.6 35.6 33.8 31.2 26.5 25.6 25.6 25.2

Relative frequency Mean

Max

Median

SE

8.5 10.8 6.0 2.3 8.6 4.1 3.5 7.6 2.9 4.9 2.5 1.0 1.0 3.3 1.7 3.5 3.3 1.2 0.9 0.9 1.0 1.2 1.8

73.5 86.1 67.4 53.7 91.4 67.6 53.1 77.6 76.6 66.9 35.3 26.5 46.5 66.7 46.9 35.2 65.4 26.1 21.0 19.3 25.7 31.8 36.9

5.7 3.5 1.6 0.9 1.7 0.7 0.5 0.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.5 0.9 0.6 0.3 0.9 0.4 0.4 0.7 0.4 0.6 0.3 0.1 0.2 0.5 0.2 0.4 0.5 0.2 0.1 0.1 0.2 0.2 0.3

303

Source: Compiled from the following: Kishaba and Mitchell, 2005; Mitchell and Gilbert, 2004; Mitchell et al., 1999, 2000a, b; Tolonen et al., 1992; Warner, 1987; and unpublished data. a Testate amoebae with ﬁlose pseudopodia. b Synonym: Nebela griseola. c Calculated as 100 number of samples in which the species was recorded/total number of samples. d Includes Arcella catinus.

Microbial diversity in Sphagnum peatlands

Table 13.5.

304

D. Gilbert and E.A.D. Mitchell

Lecane, Colurella, Trichocerca, and Euchlanis occur in peatlands. Francez (1981) identiﬁed 142 species in various peatlands of Auvergne (France). In addition, he observed that the abundance and average size of these organisms was higher in fens (8.2 104 ind. m–2) than bogs (2.9 104 ind. m–2) (Table 13.4). This trend is probably related to differences in moisture content and pH, both of which are lower in bogs than fens (Francez, 1987; Pejler and Berzins, 1993). (2) Nematodes. Contrary to rotifers, nematodes are usually not included as an element of microbial food webs because their size is usually comprised between 0.2 and 1 mm and their feeding habits can be very varied. However, these organisms can be very abundant in Sphagnum and can have a signiﬁcant impact in bacterial and fungal populations (Ingham et al., 1985). According to Wasilewska (1991) the relative abundance between bacterivorous and fungivorous nematodes on one side and phytophagous nematodes on the other is dependant on the moisture content of the environment. The abundance of nematodes is about 105 ind. m–2, 1–200 mm g–1, or 40 ind mL–1 (Table 13.4). (3) Other micro-metazoa. Other microscopic metazoans occur in peatlands, but their abundance and biomass is limited and therefore they are unlikely to play a signiﬁcant role in organic matter decomposition or microbial food webs. This is the case for gastrotrichs, a little known group of organisms of which 21 species have been described in a peatland complex of the north of Italy by Balsamo and Todaro (1993). Three other interesting groups of minute animals are the tardigrads (Tardigrada), also called water bears, the oribatid mites (Arachnida), and the ﬂatworms (Platyhelminthes). In addition, Sphagnum mosses, when they are sufﬁciently humid, can harbor Harpacticoid copepods (Crustacea; Copepoda) and, in the wettest parts of peatlands, Cladocera (water ﬂeas) (Francez, 1986). All these organisms are in most cases larger than 200 mm and are therefore not considered as microorganisms.

Total microbial biomass and relative importance of the different groups Using the fumigation-extraction method, the total microbial biomass was estimated to represent 1.7–16 mgC g–1 dry weight of peat at the surface of peatlands. It decreases with depth to reach between 1.2 and 2.5 mgC g–1 peat dry weight (Table 13.6; Williams and Silcock, 1997; Brake et al., 1999; Baum et al., 2003). The microbial C:N ratio also decreases with depth, from 12.1 at the surface to 8.9 at 0.75 m (Francez et al., 2000). Total microbial biomass values estimated from direct counts are substantially lower than those obtained using the fumigation-extraction method, probably because of the underestimation of the fungal biomass. The relative importance of different microbial groups, in terms of biomass has been estimated in July in a Sphagnum fallax-Carex rostrata-poor fen (Gilbert et al., 1998b). Heterotrophic bacteria represented 17% of the total microbial biomass, autotrophic microorganisms (algae and cyanobacteria) about 50%, heterotrophic protists 26%, fungi 2%, and rotifers 5%. In a study of ﬁve European peatlands, Mitchell et al. (2003) obtained the following proportions: bacteria 40–49%, fungi 15–36%, heterotrophic protists 6–32%, microalgae 1–22%, micro-metazoa (rotifers and nematodes) 5–9%, and cyanobacteria 0–4%. Overall it appears that the ratio of

Microbial diversity in Sphagnum peatlands Table 13.6.

305

Total C, N, and P microbial biomasses in Sphagnum peatlands.

Environment sampled

Biomass

Method

Reference

Sphagnum Sphagnum water Peat: 0–10 cm Peat: 0–10 cm Peat: 0–30 cm Peat: 0–50 cm Peat: 2–15 cm Peat: 75–100 cm Peat: 5–25 cm Peat: 0–10 cm Peat: 0–10 cm Peat: 0–30 cm Peat: 20–170 cm Peat: 5–25 cm Peat: 0–30 cm Peat: 0–50 cm

0.1–0.2 mgC g–1 (DW) 12.9 mgC mL–1 2–16 mgC g–1 (DW) 1.7–4.2 mgC g–1 (DW) 1.2–3.2 mgC g–1 (DW) 0.4–1.1 mgC g–1 (DW) 5.2–9.1 mgC g–1 (DW) 1.2–2.5 mgC g–1 (DW) 20–250 mgC g–1 0.35 mgN g–1 (DW) 0.3–0.6 mgN g–1 (DW) 0.1–0.5 mgN g–1 (DW) 0.19–0.22 mgN g–1 (DW) 25–60 mgN L–1 0.01–0.08 mgP g–1 (DW) 0.004–0.043 mgP g–1 (DW)

Direct Direct Fumigation-extraction Fumigation-extraction Fumigation-extraction Fumigation-extraction Fumigation-extraction Fumigation-extraction Fumigation-extraction Fumigation-extraction Fumigation-extraction Fumigation-extraction Fumigation-extraction Fumigation-extraction Fumigation-extraction Fumigation-extraction

Mitchell et al. (2003) Gilbert (1998) Potila and Sarjala (2004) Francez et al. (2000) Baum et al. (2003) Brake et al. (1999) Croft et al. (2001) Francez et al. (2000) Williams and Silcock (1997) Francez et al. (2000) Potila and Sarjala (2004) Baum et al. (2003) Francez et al. (2000) Williams and Silcock (1997) Baum et al. (2003) Brake et al. (1999)

Note: DW, dry weight.

autotrophic photosynthetic microorganisms/heterotrophic microorganisms drops when the moisture content decreases (Gilbert, 1998; Mitchell et al., 2003).

Functional importance of microbial communities in peatlands Heterotrophic activities: organic matter decomposition and nutrient cycling Fungi and bacteria are the primary decomposers of complex organic molecules produced by plants. The fungi produce cellulolytic exoenzymes that are more efﬁcient than those of bacteria and are generally responsible for most of the plant organic matter degradation (Davet, 1996), whereas bacteria occur preferentially in the vicinity of hyphae to beneﬁt from the molecules of low molecular weight resulting from the activity of fungal exoenzymes (Clarholm, 1994). At the surface of peatlands, the two main sources of organic matter are the particulate organic matter resulting from the degradation of plant litter and the organic matter excreted by microorganisms. The two communities of decomposer microorganisms, bacteria and fungi, coexist and feed on this source of food. Peatlands accumulate between 5 and 15% of the organic matter produced in a year (Clymo, 1983). This means that about 90% of the net primary production is lost, mainly through the respiration of organic matter by bacteria and fungi in the acrotelm. The total microbial respiration has been estimated at between 0.5 and 7.7 mgC g–1 h–1 (dry weight) at the surface and decreases with depth (Table 13.6). Inversely, the production of methane only takes place in the waterlogged peat where anoxic conditions prevail. Methane production can reach 0.2 mgC g–1 h–1 (dry weight) (Table 13.7). Other microbial activities, such as atmospheric nitrogen ﬁxation and methane oxidation have been measured in a few cases (Table 13.7). The estimation of microbial respiration is essential for the establishment of the overall carbon budget of

306

Table 13.7.

Microbial activities in Sphagnum peatlands. Activity

Variable measured

Method

Reference

Sphagnum water

0–2478 pmol L–1 h–1

Leucine incorporation

Fisher et al. (1998)

Sphagnum water Sphagnum water Sphagnum water Sphagnum water Peat: 5–10 cm Peat: 10–25 cm Peat: 0–15 cm Peat: 0–50 cm Peat: 0–75 cm Peat: 15–30 cm Peat: 75–170 cm Peat: 0–40 cm Peat: 10–75 cm Peat: 75—170 cm Peat: 4–9 cm Peat: 0–20 cm

0.7–6.1 mgC L–1 h–1 1.8 mgC L–1 h—1 5–261 mgC L–1 h–1 0.2–45.8 mgC L–1 h–1 9.9–23.5 mgC g–1 h–1 (DW) 2.6–10.9 mgC g–1 h—1 (DW) 3.8 mgC g–1 h—1 (DW) 0.5–3.8 mgC g–1 h–1 (DW) 5.8–7.7 mgC g–1‘h–1 (DW) 1.4 mgC g–1 h–1 (DW) 1.1–2.1 mgC g–1 h–1 (DW) 0.3 mgC g–1 h–1 (DW) 0.2–0.4 10 3 mgC g–1 h–1 (DW) 0.1–0.2 mgC g–1 h–1 (DW) 7.5 mgC g–1 h–1 (DW) 1.1–2.9 mgN g–1 h–1 (DW)

Microbial carbon incorporation Heterotrophic activity Heterotrophic activity Photosynthetic activity Photosynthetic activity Total CO2 respiration Total CO2 respiration Total CO2 respiration Total CO2 respiration Total CO2 respiration Total CO2 respiration Total CO2 respiration Microbial CH4 emission Microbial CH4 emission Microbial CH4 emission Microbial CH4 consumption Microbial N2 ﬁxation

Amino acid incorporation Amino acid incorporation 14 C incorporation 14 C incorporation Peat incubation Peat incubation Peat incubation Peat incubation In situ fluxes Peat incubation In situ fluxes Peat incubation In situ fluxes In situ fluxes Peat incubation Peat incubation

Gilbert et al. (1998a) Gilbert (1998) Gilbert et al. (1998a) Gilbert (1998) Williams and Silcock (1997) Williams and Silcock (1997) Fisk and Schmidt (1996) Brake et al. (1999) Francez et al. (2000) Fisk and Schmidt (1996) Francez et al. (2000) Krumholz et al. (1995) Francez et al. (2000) Francez et al. (2000) Krumholz et al. (1995) Krumholz et al. (1995)

Note: DW, dry weight.

D. Gilbert and E.A.D. Mitchell

Environment sampled

Microbial diversity in Sphagnum peatlands

307

peatlands and the role they play as C sinks. In the perspective of peatland exploitation for horticulture and successive regeneration, Francez et al. (2000) have demonstrated that, during the restoration process, cutover peatlands can remain carbon sources for many years following the re-establishment of Sphagnum mosses. Microorganisms play an especially important functional role at the surface of peatlands by recycling nutrients before they become incorporated in the peat. These processes are crucial for the functioning of peatlands which otherwise depend essentially from nutrient inputs through wet and dry deposition. Nitrogen, which is usually the limiting nutrient in peatlands except when they are subject to high N pollution levels (Aerts et al., 1992), is one of the main causes for inter-speciﬁc competition (Francez and Loiseau, 1999). Sphagnum mosses are especially efﬁcient at retaining nutrients from precipitation in spring during the growth period (Clymo, 1963; Van Breemen, 1995), whereas microorganisms are especially efﬁcient in keeping nutrients in summer (Li and Vitt, 1997). Moreover, microorganisms appear as a key group for nitrogen recycling in the acrotelm (Humphrey and Pluth, 1996; Croft et al., 2001; Potila and Sarjala, 2004), most of these microbial processes occurring in the ﬁrst 10 cm of the peat (Francez and Loiseau, 1999).

Microbial primary production The primary production from vascular plants and Sphagnum mosses has been much studied in peatlands. By contrast, microbial primary production has, to our knowledge hardly been studied at all. In a Sphagnum fallax-dominated peatland from the center of France, Gilbert (1998) estimated this production at about 3 gC m–2 yr–1, or less than 1% of plant net primary production. Similarly, according to Thuillier (1998), the production of Sphagnum and vascular plants is 200 times greater than microbial primary production. Furthermore, microorganisms are quickly lysed after the death of the organisms, as attested by the fumigation-extraction experiments realized in peat. In these conditions, it seems justiﬁed to consider the necromass originating from photosynthetic microorganisms to be quantitatively negligible for peat accumulation processes.

The microbial loop The microbial loop concept initially derived from studies of marine and lake ecosystems has been extended to soils and then to peatlands (Azam et al., 1983; Porter et al., 1985; Clarholm, 1994; Coleman, 1994; Gilbert et al., 1998a). In peatlands, the trophic chain based on photosynthetic assimilation by microalgae and cyanobacteria is relatively marginal. Instead, the main source of organic matter comes from the decomposition of Sphagnum mosses and vascular plants that grow on the peatland surface. Furthermore, in the microbial loop of aquatic ecosystems, bacteria constitute the essential link between organic matter and micro-zooplankton. In peatlands, it would appear logical to include the fungi in the microbial loop. Indeed, although

308

D. Gilbert and E.A.D. Mitchell

fungi and bacteria are two very different kinds of organisms, they both use dissolved organic matter and both fall prey to heterotrophic protists and micro-metazoa (Yeates and Foissner, 1995; Gilbert et al., 2000a, 2003). The second level of the microbial loop includes the heterotrophic protists, which, in Sphagnum peatlands, are essentially the testate amoebae and the ciliates. Ciliates, and especially the smallest among them, are the main bacterivorous organisms at the surface of peatlands (Gilbert, 1998). By contrast the precise role of testate amoebae is harder to deﬁne. In soils, naked amoebae are responsible for the regulation of bacterial populations, owing to their ability to move in the interstitial volumes of the soil and because they are better adapted than ﬁltering microorganisms to ingest bacteria forming bioﬁlms on the surface of soil particles (Clarholm, 1981). At the surface of Sphagnum peatlands, the available pore space is much greater than in mineral soils and the bacteria are mostly associated with particulate organic matter in suspension in the water. The physical structure of the environment thus seems more favorable to the development of testate amoebae, which can easily move between the leaves of Sphagnum. Furthermore, these heterotrophic protists have an additional advantage over naked amoebae in that they beneﬁt from the protection of a shell against predators. However, peatland testate amoebae do not appear to have exactly the same trophic role as naked amoebae in mineral soils. Indeed, the range of the types of food they ingest seems wider than that of naked amoebae, and reﬂects the full diversity of available prey in Sphagnum: heterotrophic bacteria, fungal hyphae and spores, microalgae, heterotrophic protists, small micro-metazoa (Gilbert et al., 2000a, 2003). This feeding behavior has been interpreted as a compensatory solution that does not allow them an optimal development (Couˆteaux and Pussard, 1983). However, in case of food shortage, this behavior does provide testate amoebae an advantage over other predatory microorganisms. Finally, because they represent several trophic levels, testate amoebae constitute a key trophic link in the microbial communities of Sphagnum peatlands (Gilbert, 1998).

Transfer to higher trophic levels The role of the meso- and macro-fauna in the trophic webs of peatlands has been little studied (Arndt, 1993). Depending on the water content of Sphagnum mosses, it is possible to observe many groups of organisms (oligochaeta, insect larvae and adults, crustaceans, mollusks, arachnids, and others) (Francez, 1984, 1986; Nowak and Pilipiuk, 1997). Recent studies in soils show that many detritivorous species live in the litter (Nieminen and Seta¨la¨, 1998). Furthermore, according to Bonkowski and Schaefer (1997), oligochaetes ingest naked amoebae in soils. Similarly, in peatlands it is likely that an important part of the energy and matter needed for the development of the meso- and macro-fauna come from the consumption of microorganisms, be it selective or fortuitous through the ingestion of bulk soil particles. Empty tests (shells) of testate amoebae found in the digestive tract of earthworms seems to attest of this (Gilbert, 1998). In the absence of speciﬁc studies on this subject, it is however difﬁcult to reach deﬁnite conclusions on the signiﬁcance of trophic interactions among mesoand macro-fauna and microorganism at the surface of peatlands.

Microbial diversity in Sphagnum peatlands

309

Practical applications Biomonitoring Beyond their functional importance in the recycling of nutrients through the decomposition of organic matter, microorganisms are also a subject of interest for ecologists because of their value as bioindicators of the quality of natural, perturbed, or regenerating ecosystems. Heterotrophic protists, and especially the testate amoebae are well suited for such application (Buttler et al., 1996; Foissner, 1997, 1999; Jauhiainen, 2002; Laggoun-De´farge et al., 2004). Sphagnum peatlands are fragile ecosystems whose functioning strongly depends on the maintenance of an appropriate level of moisture and low concentrations and inputs of nutrient. Any modiﬁcation of hydrology or nutrient status will therefore have a strong impact on the structure and functioning of the ecosystem, and may, for example, cause the peatland to stop being a carbon sink and instead act as a carbon source. Perturbations may be caused by direct human impact through peat harvesting, drainage for forestry purposes, or fertilization for agriculture. But peatlands may also be affected indirectly by human activities, for example, through increases in nitrogen and sulfur deposition, atmospheric carbon dioxide concentrations, or global warming (Aerts et al., 1992; Jauhiainen et al., 1998; Lee, 1998; Berendse et al., 2001; Freeman et al., 2002, 2004; Mitchell et al., 2003). What can microorganisms tell us about these changes? Drainage and nutrient inputs modify the structure and functioning of microbial communities in Sphagnum. The main effect on the functioning of the peatland ecosystem is an increase in rates of organic matter mineralization. For example, three years after the conversion of a peatland to agricultural land, the abundance of fungi decreased whereas that of bacteria and actinomycetes increased tenfold (Kuster, 1993). Furthermore, the addition of agricultural fertilizers (NPKCa and PKCa) to a Sphagnum-dominated peatland caused a strong increase in the relative abundance of heterotrophic bacteria, diatoms, and ciliates whereas testate amoebae and other microalgae decreased (Gilbert et al., 1998a). Similarly, controlled burning of the surface vegetation and litter with the aim of increasing plant productivity for fodder production caused an increase in the abundance of euglenes and diatoms (Kuster, 1993). Finally according to Francez (1991), mowing induced a decrease in the diversity of rotifers over the entire surface of a peatland. A study on the impact of nitrogen addition on the surface of peatlands showed that the structure and functioning of microbial communities respond fast to perturbations (Gilbert et al., 1998b). Four months after the onset of the experiment, the addition of 5 gN m2 caused the biomass of autotrophic microorganisms, chlorophyll a, and microbial primary production to increase strongly, whereas the biomass and activity of heterotrophic microorganisms increased less clearly. These observations suggest that, with increasing N deposition rates, Sphagnum mosses become less efﬁcient in taking up N from precipitation, leaving a higher proportion of the input to be used by microorganisms and, deeper in the peat by vascular plants. This mechanism is in agreement with experimental evidence of a competitive shift in favor of vascular plants with higher N deposition rates (Berendse et al., 2001).

310

D. Gilbert and E.A.D. Mitchell

The ongoing increase in atmospheric CO2 concentrations may also affect microbial communities indirectly through its effect on plants. In an in situ experimental study, Mitchell et al. (2003) found that the abundance and biomass of heterotrophic bacteria increased and the biomass of testate amoebae decreased under raised CO2 levels. Whereas the effect on bacteria is likely due to increased production and release of labile organic compounds by plants, the reason for the decrease in testate amoebae is less straightforward. Interestingly, testate amoebae were also negatively affected by fertilization (Gilbert et al., 1998a, b) and experimentally reduced UVB (Searles et al., 2001) in peatlands, and the number of large heterotrophic protists decreased under elevated CO2 in mineral soils (Treonis and Lussenhop, 1997). Thus, it appears that testate amoebae in general are systematically affected by experimental changes in their environment, but to date no satisfactory explanation has been proposed for these responses. Could it simply be that their central position in the microbial food webs makes them more likely to suffer from any changes in the structure of the microbial community? Conversely, the central position and higher number of links in the food web should ensure testate amoebae a higher resilience. So, the observed high responsiveness of testate amoebae to changes could be due merely to the higher number of studies on this group.

Paleoecology The extent to which microorganisms can be useful for paleoecological studies depends on the preservation of recognizable body parts. Several groups of microorganisms can provide valuable information on past ecological conditions and especially diatoms, testate amoebae and fungal spores (Van Geel, 1978; Beyens, 1985; Tolonen, 1986; Campbell et al., 1997; Kuhry, 1997). Other groups such as chrysophyceans (Smol, 1990), copepods, cladocerans (Hann, 1990), and the rotifer genus Habrotrocha also occur in peat (Warner and Chengalath, 1988). However, in Sphagnum peatlands, of these three microbial groups the testate amoebae are most commonly used in paleoecological studies. Whereas the potential for using this group of organisms in paleoecology was recognized long ago (Harnisch, 1927), the development of quantitative numerical approaches (transfer functions) has made this tool much more valuable to paleoecologists ( Tolonen, 1986; Beyens and Chardez, 1987; Warner, 1987; Charman, 2001). Transfer functions describe the relationship between species and an environmental parameter of interest (such as depth to water table) statistically and then apply this relationship to fossil assemblages to provide estimates of changes in the environmental parameter through time. Testate amoebae have several advantages that make them useful: they are diverse (usually about 10–30 species in a given sample, often over 100 species across an ecological gradient), numerous (1 cm3 of peat is usually enough to extract a sufﬁcient number of shells), and they allow quantitative inference of the moisture conditions and pH of the precise coring location. Thus they are now often used in parallel to pollen and spores, which provide information on the surrounding vegetation as well as the peatland in general. Through a combined, multiproxy approach, a sounder image of the history of a site can be achieved. Judging by the impressive number of recent publications

Microbial diversity in Sphagnum peatlands

311

using testate amoebae in paleoecology, we expect this tool to become a standard component of the peatland paleoecologist’s toolbox (Chiverrell, 2001; Booth and Jackson, 2003; Langdon et al., 2003; Mauquoy and Barber, 2002; Wilmshurst et al., 2002, 2003; Charman et al., 2004).

Open research questions The complexity of micro-environmental conditions present in peatlands and the challenges associated with the proper sampling for microorganisms in this environment may, in part, explain the fact that microorganisms are still being much less studied than other components of the ecosystem. Very few studies have provided data on the abundance and diversity of various groups of microorganisms, despite the fact that most scientists agree that microorganisms play key roles in the functioning of ecosystems (nutrient cycling, methanogenesis, and others). When microorganisms are included in ecological studies of peatlands most often the methods used do not provide much detail (microbial biomass determined using the fumigation-extraction method), or only the result of their activity is measured (methane emissions, decomposition rates, carbon balance). Analyzing the structure of microbial communities by separating the major groups (heterotrophic bacteria, fungi, cyanobacteria, microalgae, different groups of heterotrophic protists, micro-metazoa) provides more information than total microbial biomass, but this approach to microbial community structure may still hide species–speciﬁc responses. The microbial world is clearly as complex, if not more so, than the macroscopic world. The size of these organisms and the spatial and temporal scale of environmental inﬂuences they respond to represent important challenges for ecologists. However, until we learn more about the microbial world we will not be able to achieve a full understanding of the functioning of peatland ecosystems. There is no a-priori reason to continue to study microbial communities with a black-box approach or not consider them at all. It is therefore fundamental to study the spatial and temporal changes in biomass for each of the microbial functional groups. This information should then be integrated in order to generate hypotheses about the functioning of the microbial loop that can be experimentally tested using ﬁeld manipulative experiments or mesocosm studies. Microorganisms respond fast to changes in their environment and therefore constitute early indicators of perturbations in peatlands and other ecosystems. This is especially interesting in the case of indirect perturbations such as long-distance nutrient or pollutant transport and deposition, which may be difﬁcult to quantify owing to spatial and temporal variability in production, transport, and deposition. Furthermore, using microorganisms as bioindicators in seminatural or natural ecosystems could also be an alternative to measurements that are technologically more intensive, practically more difﬁcult to set up in remote locations, and much more expensive. In such cases, a monitoring program can easily be established with point-time samplings of water, mosses, or soil samples, ideally combined to a paleoecological approach in order to compare the range of present changes with past ones. In order for microorganisms to be more useful as a tool for ecologists and paleoecologists, and to better understand the functioning of peatland ecosystems, we

312

D. Gilbert and E.A.D. Mitchell

clearly need much more baseline data on the structure and functioning of microbial communities, their response to ecological gradients in different types of peatlands, in natural conditions and under a broad range of perturbations and subsequent recovery. References Aerts, R., Wallen, B., and Malmer, N., 1992. Growth-limiting nutrients in Sphagnum-dominated bogs subject to low and high atmospheric nitrogen supply. J. Ecol. 80, 131–140. Arndt, H., 1993. Rotifers as predators on components of the microbial web (Bacteria, Heterotrophic Flagellates, Ciliates) – a review. Hydrobiologia 255, 231–246. Azam, F., Fenchel, T., Field, J.G., et al., 1983. The ecological role of water column microbes in the sea. Mar. Ecol. Pro. Ser. 10, 257–263. Baldauf, S.L., 2003. The deep roots of eukaryotes. Science 300, 1703–1706. Balsamo, M. and Todaro, M.A., 1993. Gastrotrichi del Trentino: le viotte de Monte Bodone. Studi Trentini di Science Naturali. Acta Biol. 70, 9–22. Basilier, K., 1980. Fixation and uptake of nitrogen in Sphagnum blue-green-algal associations. Oikos 34, 239–242. Batut, J., 1965. Etude de la faune submicroscopique de quelques tourbie`res a` Sphagnum. Hydrobiologia 25, 239–276. Baum, C., Leinweber, P., and Schlichting, A., 2003. Effect of chemical conditions in re-wetted peats on temporal variation in microbial biomass and acid phosphatase activity within the growing season. Appl. Soil Ecol. 22, 167–174. 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. Glob. Change Biol. 7, 591–598. Beyens, L., 1985. On the Subboreal climate of the Belgian Campine as deduced from diatom and testate amoeba analyses. Rev. Palaeobot. Palynol. 46, 9–31. Beyens, L. and Chardez, D., 1984. Testate amoebae (Rhizopoda, Testaceae) from Southwest Ireland. Arch. Protistenkd. 128, 109–126. Beyens, L. and Chardez, D., 1987. Evidence from testate amoebae for changes in some local hydrological conditions between c. 5000 BP and c. 3800 BP on Edgeøya (Svalbard). Polar Res. 5, 165–169. Bledzki, L.A. and Ellison, M.A., 2003. Diversity of rotifers from northeastern USA bogs with new species records from North America and New England. Hydrobiologia 497 (1), 53–62. Bobrov, A.A., Charman, D.J., and Warner, B.G., 1999. Ecology of testate amoebae (Protozoa: Rhizopoda) on peatlands in western Russia with special attention to niche separation in closely related taxa. Protist 150, 125–136. Bobrov, A.A., Charman, D.J., and Warner, B.G., 2002. Ecology of testate amoebae from oligotrophic peatlands: speciﬁc features of polytypic and polymorphic species. Biol. Bull. Rus. Acad. Sci. 29, 605–617. Bonkowski, M. and Schaefer, M., 1997. Interactions between earthworms and soil protozoa: A trophic component in the soil food web. Soil Biol. 29, 499–502. Bonnet, L., 1958. Les the´camoebiens des Bouillouses. Bull. Soc. Hist. Nat. Toulouse 93, 529–543. Booth, R.K., 2002. Testate amoebae as paleoindicators of surface-moisture changes on Michigan peatlands: modern ecology and hydrological calibration. J. Paleolimnol. 28, 329–348. Booth, R.K. and Jackson, S.T., 2003. A high resolution record of late-Holocene moisture variability from a Michigan raised bog, USA. Holocene 13, 863–876. Borga, P., Nilsson, M., and Tunlid, A., 1994. Bacterial communities in peat in relation to botanical composition as revealed by phospholipid fatty-acid analysis. Soil Biol. Biochem. 26, 841–848. Brake, M., Ho¨per, H., and Joergensen, R.G., 1999. Land use-induced changes in activity and biomass of microorganisms in raised bogs at different depths. Soil Biol. Biochem. 31, 1489–1497. Buttler, A., Grosvernier, P., and Matthey, Y., 1998. A new sampler for extracting undisturbed surface peat cores for growth pot experiments. New Phytol. 140, 355–360.

Microbial diversity in Sphagnum peatlands

313

Buttler, A., Warner, B.G., Grosvernier, P., and Matthey, Y., 1996. Vertical patterns of testate amoebae (Protozoa: Rhizopoda) and peat forming vegetation on cutover bogs in the Jura, Switzerland. New Phytol. 134, 371–382. Campbell, D.R., Duthie, H.C., and Warner, B.G., 1997. Post-glacial development of a kettle-hole peatland in southern Ontario. Ecoscience 4, 404–418. Caron, D.A., 1983. Technique for enumeration of heterotrophic and phototrophic nanoplankton, using epiﬂuorescence microscopy and comparison with other procedures. Appl. Environ. Microbiol. 46, 491–498. Chapman, S., Buttler, A., Francez, A.-J., et al., 2003. Commercial exploitation of peatlands and maintenance of biodiversity – a conﬂict between economy and ecology. Front. Ecol. Environ. 1, 525–532. Charman, D.J., 1997. Modelling hydrological relationships of testate amoebae (Protozoa: Rhizopoda) on New Zealand peatlands. J. Roy. Soc. NZ 27, 465–483. Charman, D.J., 2001. Biostratigraphic and palaeoenvironmental applications of testate amoebae. Quatern. Sci. Rev. 20, 1753–1764. Charman, D.J., Brown, A.D., Hendon, D., and Karofeld, E., 2004. Testing the relationship between Holocene peatland palaeoclimate reconstructions and instrumental data at two European sites. Quatern. Sci. Rev. 23, 137–143. Charman, D.J., Hendon, D., and Packman, S., 1999. Multiproxy surface wetness records from replicate cores on an ombrotrophic mire: implications for Holocene palaeoclimate records. Quatern. Sci. Rev. 14, 451–463. Charman, D.J. and Warner, B.G., 1997. The ecology of testate amoebae (Protozoa: Rhizopoda) in oceanic peatlands in Newfoundland, Canada: Modelling hydrological relationships for palaeoenvironmental reconstruction. Ecoscience 4, 555–562. Chiverrell, R.C., 2001. A proxy record of late Holocene climate change from May Moss, northeast England. J. Quat. Sci. 16, 9–29. Clarholm, M., 1981. Protozoan grazing of bacteria in soil – impact and importance. Microb. Ecol. 7, 343–350. Clarholm, M., 1994. The microbial loop in soil. In: Ritz, K., Dighton, J., and Giller, K.E. (Eds), Beyond the Biomass. Wiley, Chichester, pp. 221–230. Clarholm, M. and Rosswall, T., 1980. Biomass and turnover of bacteria in a forest soil and a peat. Soil Biol. Biochem. 12, 49–57. Clymo, R.S., 1963. Ion exchange in Sphagnum and its relation to bog ecology. Ann. Bot. 27, 309–324. Clymo, R.S., 1983. Peat. In: Gore, A.J.P. (Ed.), Mires: Swamp, Bog, Fen, and Moor. General Studies. Elsevier, Amsterdam, Vol. A, pp. 159–224. Cole, L., Bardgett, R.D., Ineson, P., and Hobbs, P.J., 2002. Enchytraeid worm (Oligochaeta) inﬂuences on microbial community structure, nutrient dynamics and plant growth in blanket peat subjected to warming. Soil Biol. Biochem. 34, 83–92. Coleman, D.C., 1994. The microbial loop concept as used in terrestrial soil ecology studies. Microb. Ecol. 28, 245–250. Collins, V.G., D’Sylva, D.T., and Latter, P.M., 1978. Microbial populations in peat. In: Heal, O.W. and Perkins, D.F. (Eds), Production Ecology of British Moors and Montane Grasslands. Springer, Berlin, pp. 94–112. Compe`re, P., 1980. Quelques algues de la re´gion de Bruˆly (France, Ardennes; Belgique, Namur). Bull. Soc. Roy. Bot. Belg. 112, 151–165. Cosandey, F., 1964. La tourbie`re des Tenasses sur Vevey. Mate´riaux pour le leve´ ge´obotanique de la Suisse 45, 1–320. Couˆteaux, M.-M., 1969. The´camoebiens muscicoles de Gaume et de Moyenne-Belgique. Rev. Ecol. Biol. Sol. 6, 413–428. Couˆteaux, M.-M. and Pussard, M., 1983. Nature du re´gime alimentaire des protozoaires du sol. In: LeBrun, P., Andre´, H.M., De Medts, A., et al. (Eds), New Trends in Soil Biology, Proceedings of the VIII. International Colloquium of Soil Biology, Louvain-la-Neuve (Belgium), pp: 179–195. Croft, M., Rochefort, L., and Beauchamp, C.J., 2001. Vacuum-extraction of peatlands disturbs bacterial population and microbial biomass carbon. Appl. Soil Ecol. 18, 1–12. Czeczuga, B., 1993. Aquatic fungi of the Gorbacz and Ostrowki Peatbogs. Acta Mycol. 28, 69–75.

314

D. Gilbert and E.A.D. Mitchell

Davet, P., 1996. Vie microbienne du sol et production ve´ge´tale. Institut national de la Recherche Agronomique, INRA, Paris Cedex, pp. 383. Dell’Uomo, A., 1981. Studio algologico del bacino torboso-palustre del Laghestel (Trento). Studi Trentini di Science Naturali. Acta Biol. 58, 169–230. Dell’Uomo, A. and Agostinelli, A., 1990. Florula desmidiologica del trentino-alto Adige: le torbiere di nova Ponente e del Doss le Grave. Studi Trentini di Science Naturali 66, 83–111. Dell’Uomo, A. and Pellegrini, E., 1993a. Desmids from a peat-bog in the Northern Apennines (Italy). Arch. Hydrobiol. /Suppl., Algological Studies 68, 27–38. Dell’Uomo, A. and Pellegrini, E., 1993b. Desmids of a small peat-bog in the Lomasona Fen (Trento, Northern Italy). Cryptogam. Algol. 14, 191–198. Duthie, H.C., 1965. A study of the distribution and periodicity of some algae in a bog pool. J. Ecol. 53, 343–359. Edwards, C., Hales, B.A., Hall, G.H., et al., 1998. Microbiological processes in the terrestrial carbon cycle: methane cycling in peat. Atmos. Environ. 32, 3247–3255. Fisher, M.M., Graham, J.M., and Graham, L.E., 1998. Bacterial abundance and activity across sites within two northern Wisconsin Sphagnum bogs. Microb. Ecol. 36, 259–269. Fisk, M.C. and Schmidt, S.K., 1996. Microbial responses to nitrogen additions in alpine tundra soil. Soil Biol. Biochem. 28, 751–755. Foissner, W., 1997. Protozoa as bioindicators in agroecosystems, with emphasis on farming practices, biocides, and biodiversity. Agric. Ecosyst. Environ. 62, 93–103. Foissner, W., 1999. Soil protozoa as bioindicators: pros and cons, methods, diversity, representative examples. Agric. Ecosyst. Environ. 74, 95–112. Francez, A.-J., 1981. Rotife`res de quelques tourbie`res d’Auvergne. Ann. Stat. Biol. Besse-en-Chandesse 15, 276–287. Francez, A.-J., 1984. Ecologie des peuplements de rotife`res sessiles des lacs-tourbie`res d’Auvergne (France). Bull. Ecol. 15, 231–237. Francez, A.-J., 1986. Sphagnum microfauna in two peatbogs of the French Massif Central. Suo 37, 1–6. Francez, A.-J., 1987. Successions e´cologiques dans les tourbie`res: Le peuplement de rotife`res du lac-tourbie`re de Chambedaze (Puy-de-Doˆme, France). Bull. Ecol. 18, 31–38. Francez, A.-J., 1988. Le peuplement de rotife`res libres de deux lacs-tourbie`res du Puy-de-Doˆme (France). Vie Milieu 38, 281–292. Francez, A.-J., 1991. Production primaire et accumulation de matie`re organique dans les tourbie`res a` sphaignes des monts du Forez. Inﬂuences des activite´s humaines sur leur fonctionnement et leur e´volution. Ph.D Thesis, Univ. Pierre et Marie Curie, Paris 6., Paris, France, 397 pp. Francez, A.-J., Gogo, S., and Josselin, N., 2000. Distribution of potential CO2 and CH4 productions, denitriﬁcation and microbial biomass C and N in the proﬁle of a restored peatland in Brittany (France). Eur. J. Soil Biol. 36, 161–168. Francez, A.-J. and Loiseau, P., 1999. The fate of mineral nitrogen in a fen with Sphagnum fallax klinggr and Carex rostrata stokes (Massif-central, France). Can. J. Bot.-Rev. Can. Bot. 77, 1136–1143. Freeman, C., Fenner, N., Ostle, N.J., et al., 2004. Export of dissolved organic carbon from peatlands under elevated carbon dioxide levels. Nature 430, 195–198. Freeman, C., Ostle, N., and Kang, H.J., 2002. An enzymatic ‘latch’ on a global carbon store. Nature 409, 149. Galand, P.E., Fritze, H., and Yrja¨la¨, K., 2003. Microsite-dependent changes in methanogenic populations in a boreal oligotrophic fen. Environ. Microbiol. 5, 1133–1143. Galand, P.E., Saarnio, S., Fritze, H., and Yrja¨la¨, K., 2002. Depth related diversity of methanogen Archea in Finnish oligotrophic fen. FEMS Microb. Ecol. 42, 441–449. Gardiner, J.J., 1975. The inﬂuence of fertilisers upon microbial activity in peat. II-Calcium and nitrogen. Irish Forestry 32, 101–114. Gilbert, D., 1998. Les communaute´s microbiennes a` la surface des tourbie`res a` sphaignes: structure, fonctionnement et impact des apports de fertilisants. Ph.D Thesis, Laboratoire de Biologie Compare´e des Protistes, University Blaise Bascal (Clermont II), Clermont-Ferrand, France, 133pp. Gilbert, D., Amblard, C., Bourdier, G., and Francez, A.-J., 1998a. The microbial loop at the surface of a peatland: structure, function, and impact of nutrient input. Microb. Ecol. 35, 83–93.

Microbial diversity in Sphagnum peatlands

315

Gilbert, D., Amblard, C., Bourdier, G., and Francez, A.J., 1998b. Short-term effect of nitrogen enrichment on the microbial communities of a peatland. Hydrobiologia 374, 111–119. Gilbert, D., Amblard, C., Bourdier, G., et al., 2000a. Le re´gime alimentaire des the´camoebiens. Ann. Biol.-Paris. 39, 57–68. Gilbert, D., Francez, A.-J., Amblard, C., and Bourdier, G., 2000b. Microbial communities at the surface of Sphagnum peatlands: good indicators of Human disturbances?. Bull. Ecol. 30, 45–52. Gilbert, D., Mitchell, E.A.D., Amblard, C., et al., 2003. Population dynamics and food preferences of the testate amoeba Nebela tincta major-bohemica-collaris complex (Protozoa) in a Sphagnum Peatland. Acta Protozool. 42, 99–104. Given, P.H. and Dickinson, C.H., 1975. Biochemistry and microbiology of peats. In: Paul, P.A. and Mc Laren, A.D. (Eds), Soil Biochemistry. Marcel Dekkets, New York, Vol. 3, pp. 123–212. Golovchenko, A.V., Polyanskaya, L.M., Dobrovol’skaya, T.G., et al., 1994. The spatial distribution and structure of microbial complexes in bog forest Ecosystems. Eurasian Soil Sci. 26, 78–89. Greaves, M.P., Weatley, R.E., Shepherd, H., and Knight, A.H., 1973. Relationship between microbial populations and adenosine triphosphate in a basin peat. Soil Biol. Biochem. 5, 685–687. Grolie`re, C.A., 1974–1975. Etude de quelques cilie´s hyme´nostomes des eaux acides de la re´gion de Besse en Chandesse. Ann. Stat. Biol. Besse-en-Chandesse 9, 79–109. Grolie`re, C.A., 1975. Descriptions de quleques cilie´s hypotriches de tourbie`res a sphaignes et des e´tendues d’eau acides. Protistologica 11, 481–498. Grolie`re, C.A., 1976. Ecology of Sphagnum Infusoria. J. Protozool. 23, A11. Grolie`re, C.A., 1977. Contribution a` l’e´tude de quelques cilie´s des sphaignes: II-Dynamique des populations. Protistologica 13, 335–352. Hales, B.A., Edwards, C., Ritchie, D.A., et al., 1996. Isolation and identiﬁcation of methanogen-speciﬁc DNA from blanket bog peat by PCR ampliﬁcation and sequence analysis. Appl. Environ. Microbiol. 62, 668–675. Hann, B.J., 1990. Cladocera. In: Warner, B.G. (Ed.), Methods in Quaternary Ecology. Reprint series, Geoscience Canada, St. John’s, Newfoundland, Vol. 5, pp. 81–91. Harnisch, O., 1927. Einige Daten zur recenten und fossilen testaceen Rhizopodenfauna der Sphagnen. Arch. Hydrobiol. 18, 345–360. Heal, O.W., 1962. The abundance and microdistribution of testate amoebae (Protozoa: Rhizopoda) in Sphagnum. Oikos 13, 35–47. Heal, O.W., 1964. Observations on the seasonal and spatial distribution of testaceans (Protozoa: Rhizopoda) in Sphagnum. J. Anim. Ecol. 33, 395–412. Hemond, H.F., 1983. The nitrogen budget of Thoreau gog. Ecology 64, 99–109. Hendon, D. and Charman, D.J., 1997. The preparation of testate amoebae (Protozoa: Rhizopoda) samples from peat. Holocene 7, 199–205. Henebry, M.S. and Cairns, J. Jr., 1984. Protozoan colonization rates and trophic status of some freshwater wetland lakes. J. Protozool. 31, 456–467. Hiroki, M. and Watanabe, M.M., 1996. Microbial community and rate of cellulose decomposition in peat soils in a mire. Soil Sci. Plant Nutr. 42, 893–903. Hobbie, J.E., Delay, R.J., and Jasper, S., 1977. Use of Nuclepore ﬁlters for counting bacteria by ﬂuorescence microscopy. Appl. Environ. Microbiol. 33, 1225–1228. Humphrey, W.D. and Pluth, D.J., 1996. Net nitrogen mineralization in natural and drained fen peatlands in Alberta, Canada. Soil Sci. Soc. Am. J. 60, 932–940. Ingham, R.E., Trofymow, J.A., Ingham, E.R., and Coleman, D.C., 1985. Interactions of bacteria, fungi, and their nematode grazers: effects on nutrient cycling and plant growth. Ecol. Monogr. 55, 119–140. Jauhiainen, J., Silvola, J., and Vasander, H., 1998. The effects of increased nitrogen deposition and CO2 on Sphagnum angustifolium and S. warnstorfii. Annales Botanici Fennici 35, 247–256. Jauhiainen, S., 2002. Testacean amoebae in different types of mire following drainage and subsequent restoration. Europ. J. Protistol. 38, 59–72. Kingston, J.C., 1982. Association and distribution of common diatoms in surface samples from northern Minnesota peatlands. Nova Hedwigia. 73, 333–346. Kishaba, K. and Mitchell, E.A.D., 2005. Changes in testate amoebae (Protists) communities in a small raised bog. A 40-year study. Acta Protozool. 44, 1–12.

316

D. Gilbert and E.A.D. Mitchell

Krumholz, L.R., Hollenback, J.L., Roskes, S.J., and Ringelberg, D.B., 1995. Methanogenesis and Methanotrophy within a Sphagnum Peatland. FEMS Microbiol. Ecol. 18, 215–224. Kuhry, P., 1997. The palaeoecology of a treed bog in western boreal Canada: a study based on microfossils, macrofossils and physico-chemical properties. Rev. Palaeobot. Palynol. 96, 183–224. Kuster, E., 1993. The microbiology of peat. In: Heatwaite, A.L. and Gottlich, K.H. (Eds), Mires: Processes, Exploitation and Conservation. Wiley, Chichester, New York, pp. 311–324. Laggoun-De´farge, F., Mitchell, E.A.D., Gilbert, D., et al., 2004. Biochemical characteristics of peat organic matter and distribution of testate amoebae in two naturally regenerating cutover Sphagnum peatlands of the Jura Mountains. In: Paı¨ vanen, J. (Ed.), Proceedings of the 12th International Peat Congress, Tampere, Finland, Vol. 1, pp. 383–384. Lamentowicz, M. and Mitchell, E.A.D., 2005. The ecology of testate amoebae (Protists) in Sphagnum in north-west Poland in relation to peatland ecology. Microb. Ecol. 50, 48–63. Langdon, P.G., Barber, K.E., and Hughes, P.D.M., 2003. A 7500-year peat-based palaeoclimatic reconstruction and evidence for an 1100-year cyclicity in bog surface wetness from Temple Hill Moss, Pentland Hills, southeast Scotland. Quatern. Sci. Rev. 22, 259–274. Lederer, F., 1995a. A new species of Cyanodictyon (cyanoprokaryota, Chroococcales) from peat-bogs in the Sumava Mountains, Czech Republic. Preslia, Praha 67, 117–121. Lederer, F., 1995b. Several little known cyanobacteria/cyanoprokaryota from peat-bogs in the Sumava Mountains. Czech Republic Algological Studies 79, 57–65. Lee, J.A., 1998. Unintentional experiments with terrestrial ecosystems – ecological effects of sulfur and nitrogen pollutants. J. Ecol. 86, 1–12. Li, Y.H. and Vitt, D.H., 1997. Patterns of retention and utilization of aerially deposited nitrogen in boreal peatlands. Ecoscience 4, 106–116. Martin, N.J., Siwasin, J., and Holding, A.J., 1982. The bacterial population of a blanket peat. J. Appl. Bacteriol. 53, 35–48. Mataloni, G., 1999. Ecological studies on algal communities from Tierra del Fuego peat bogs. Hydrobiologia 391, 157–171. Mataloni, G. and Tell, G., 1996. Compartive analysis of the phytoplankton communities of a peat bog from Tierra del Fuego (Argentina). Hydrobiologia 325, 101–112. Mauquoy, D. and Barber, K., 2002. Testing the sensitivity of the palaeoclimatic signal from ombrotrophic peat bogs in northern England and the Scottish Borders. Rev. Palaeobot. Palynol. 119, 219–240. Meisterfeld, R., 1977. Die horizontale und vertikale Verteilung der Testaceen (Rhizopoda: Testacea) in Sphagnum. Arch. Hydrobiol. 79, 319–356. Meisterfeld, R., 1979. Zur Systematik der Testaceen (Rhizopoda, Testacea) in Sphagnum. Eine REMUntersuchung. Arch. Protistenkd. 121, 246–269. Meisterfeld, R., 2002a. Order Arcellinida Kent, 1880. In: Lee, J.J., Leedale, G.F., and Bradbury, P. (Eds), The Illustrated Guide to the Protozoa. Society of Protozoologists, Lawrence, Kansas, USA, Vol. 2, pp. 827–860. Meisterfeld, R., 2002b. Testate amoebae with ﬁlopodia. In: Lee, J.J., Leedale, G.F., and Bradbury, P. (Eds), The Illustrated Guide to the Protozoa. Society of Protozoologists, Lawrence, Kansas, USA, Vol. 2, pp. 1054–1084. Mitchell, E.A.D., Borcard, D., Buttler, A.J., et al., 2000a. Horizontal distribution patterns of testate amoebae (Protozoa) in a Sphagnum magellanicum carpet. Microb. Ecol. 39, 290–300. Mitchell, E.A.D., Buttler, A., Grosvernier, P., et al., 2000b. Relationships among testate amoebae (Protozoa), vegetation and water chemistry in ﬁve Sphagnum-dominated peatlands in Europe. New Phytol. 145, 95–106. Mitchell, E.A.D., Buttler, A.J., Warner, B.G., and Gobat, J.M., 1999. Ecology of testate amoebae (Protozoa: Rhizopoda) in Sphagnum peatlands in the Jura Mountains, Switzerland and France. Ecoscience 6, 565–576. Mitchell, E.A.D. and Gilbert, D., 2004. Vertical micro-distribution and response to nitrogen deposition of testate amoebae in Sphagnum. J. Eukaryot. Microbiol. 51, 485–495. Mitchell, E.A.D., Gilbert, D., Buttler, A., et al., 2003. Structure of microbial communities in Sphagnum peatlands and effect of atmospheric carbon dioxide enrichment. Microb. Ecol. 46, 187–199.

Microbial diversity in Sphagnum peatlands

317

Mitchell, E.A.D., van der Knaap, W.O., van Leeuwen, J.F.N., et al., 2001. The palaeoecological history of the Praz-Rodet bog (Swiss Jura) based on pollen, plant macrofossils and testate amoebae (Protozoa). Holocene 11, 65–80. Nieminen, J.K. and Seta¨la¨, H., 1998. Enclosing decomposer food web: implications for community structure and function. Biol. Fertil. Soils 26, 50–57. Nikolaev, S.I., Mitchell, E.A.D., Petrov, N.B., et al., 2005. The testate lobose amoebae (order Arcellinida Kent, 1880) ﬁnally ﬁnd their home within Amoebozoa. Protist 156, 191–202. Nilsson, M., Baath, E., and So¨derstro¨m, B., 1992. The microfungal communities of a mixed mire in northern Sweden. Can. J. Bot. 70, 272–276. Nowak, E. and Pilipiuk, I., 1997. The inﬂuence of drainage on enchytraeids (Enchytraeidea, Oligochaeta) of fens in the Biebrza ice-marginal valley. Ekol. Pol.-Pol. J. Ecol. 45, 423–440. Pejler, B. and Berzins, B., 1993. On the ecology of mire rotifers. Limnologica 23, 295–300. Porter, K.G. and Feig, Y.S., 1980. The used of DAPI for identifying and counting aquatic microﬂora. Limnol. Oceanogr. 25, 943–948. Porter, K.G., Sherr, E.B., Sherr, B.F., et al., 1985. Protozoa in planktonic food webs. J. Protozool. 32, 409–415. Potila, H. and Sarjala, T., 2004. Seasonal ﬂuctuation in microbial biomass and activity along a natural nitrogen gradient in a drained peatland. Soil Biol. Biochem. 36, 1047–1055. Poulı´ ckova´, A., Ha´jkova´, P., Krenkova´, P., and Ha´jek, M., 2004. Distribution of diatoms and bryophytes on linear transects through spring fens. Nova Hedwigia. 78, 411–424. Robson, T.M., Pancotto, V.A., Ballare´, C.L., et al., 2004. Reduction of solar UV-B mediates changes in the Sphagnum capitulum microenvironment and the peatland microfungal community. Oecologia 140, 480–490. Schmidt, I.K., Ruess, L., Baath, E., et al., 2000. Long-term manipulation of the microbes and microfauna of two subarctic heaths by addition of fungicide, bactericide, carbon and fertilizer. Soil Biol. Biochem. 32, 707–720. Schoenberg, S.A. and Oliver, J.D., 1988. Temporal dynamics and spatial variation of algae in relation to hydrology and sediment characteristics in the Okefenokee Swamp, Georgia. Hydrobiol. 162, 123–134. Scho¨nborn, W., 1963. Die Stratigraphie lebender Testaceen im Sphagnetum der Hochmoore. Limnologica 1, 315–321. Scho¨nborn, W., 1965. Untersuchungen u¨ber die Zoochlorellen-Symbiose der Hochmoor-Testaceen. Limnologica 3, 173–176. Searles, P.S., Kropp, B.R., Flint, S.D., and Caldwell, M.M., 2001. Inﬂuence of solar UV-B radiation on peatland microbial communities of southern Argentina. New Phytol. 152, 213–221. Sieburth, J.M., Smetacek, V., and Lenz, J., 1978. Pelagic ecosystem structure: heterotrophic compartments of the plankton and their relationship to plankton size-fractions. Limnol. Oceanogr. 23, 1256–1263. Sizova, M.V., Panikov, N.F., Tourova, T.P., and Flanagan, P.W., 2003. Isolation and characterization of oligotrophic acido-tolerant methanogenic consortia from a Sphagnum peat bog. FEMS Microb. Ecol. 45, 301–315. Smol, J.P., 1990. Freshwater algae. In: Warner, B.G. (Ed.), Methods in Quaternary Ecology. Reprint series, Geoscience Canada, St. John’s, Newfoundland, Vol. 5, pp. 3–14. Sundh, I., Borga, P., Nilsson, M., and Svensson, B.H., 1995. Estimation of cell numbers of methanotrophic bacteria in boreal peatlands based on analysis of speciﬁc phospholipid fatty acids. FEMS Microb. Ecol. 18, 103–112. Sundh, I., Nilsson, M., and Borga, P., 1997. Variation in microbial community structure in two boreal peatlands as determined by analysis of phospholipid fatty acid proﬁles. Appl. Environ. Microb. 63, 1476–1482. Thormann, M.N., Currah, R.S., and Bayley, S., 2003. Succession of microfungal assemblages in decomposing peatland plant. Plant Soil 250, 323–333. Thormann, M.N., Currah, R.S., and Bayley, S.E., 2001. Microfungi isolated from Sphagnum fuscum from a southern boreal bog in Alberta, Canada. Bryologist 104, 548–559. Thormann, M.N., Currah, R.S., and Bayley, S.E., 2002. The relative ability of fungi from Sphagnum fuscum to decompose selected carbon substrates. Can. J. Microbiol. 48, 204–211.

318

D. Gilbert and E.A.D. Mitchell

Thuillier, L., 1998. Dynamique du carbone et de l’azote dans les tourbie`res solige`nes en conditions naturelles et modiﬁe´es. Ph.D Thesis, University Rennes I, France, 151 pp. Tolonen, K., 1986. Rhizopod analysis. In: Berglund, B.E. (Ed.), Handbook of Holocene Palaeoecology and Palaeohydrology. Wiley, Chichester, pp. 645–666. Tolonen, K., Warner, B.G., and Vasander, H., 1992. Ecology of Testaceans (Protozoa, Rhizopoda) in mires in Southern Finland. 1. Autecology. Arch. Protistenkd. 142, 119–138. Tolonen, K., Warner, B.G., and Vasander, H., 1994. Ecology of Testaceans (Protozoa, Rhizopoda) in Mires in Southern Finland. 2. Multivariate-Analysis. Arch. Protistenkd. 144, 97–112. Tomaszewicz, G.H., 1994. Abundance and composition of the desmid ﬂora in a series of peat pits, in relation to pH and some other habitat parameters. Biol. Bratislava 49, 519–524. Treonis, A.M. and Lussenhop, J.F., 1997. Rapid response of soil protozoa to elevated CO2. Biol. Fertil. Soils 25, 60–62. Utermo¨lh, H., 1958. Zur vervollkommnung der quantative phytoplankton-methodik. Mitteilungen aus Institut Verhein Limnologie 9, 1–38. Van Breemen, N., 1995. How Sphagnum bogs down other plants. Trends Ecol. Evol. 10, 270–275. Van Geel, B., 1978. Paleoecological study of Holocene peat bog sections in Germany and Netherlands, based on analysis of pollen, spores and macroscopic and microscopic remains of fungi, algae, Cormophytes and animals. Rev. Palaeobot. Palynol. 25, 1–120. Vecherskaya, M.S., Galchenko, V.F., Sokonova, E.N., and Samarkin, V.A., 1993. Activity and species composition of aerobic methanotrophic communities in tundra soils. Curr. Microbiol. 27, 181–184. Villeret, S., 1955. Contribution a` la biologie des algues des tourbie`res a` sphaignes. Ph.D Thesis, Universite´ de Paris, 246pp. Warner, B.G., 1987. Abundance and diversity of testate amoebae (Rhizopoda, Testacea) in Sphagnum peatlands in Southwestern Ontario, Canada. Arch. Protistenkd. 133, 173–189. Warner, B.G. and Chengalath, R., 1988. Holocene fossil Habrotrocha angusticollis (Bdelloidea: Rotifera) in North America. J. Paleolimnol. 1, 141–147. Warner, B.G. and Chmielewski, J.G., 1992. Testate amoebae (Protozoa) as indicators of drainage in a forested mire, northern Ontario, Canada. Arch. Protistenkd. 141, 179–183. Wasilewska, L., 1991. The effect of moisture content of drained peat soils on soil nematode communities. Polish Ecol. Stud. 17, 191–201. Waskman, S.A. and Stevens, K.R., 1929. Contribution to the chemical composition of peat: V. The role of microorganisms in peat formation and decomposition. Soil Sci. 27, 315–340. Williams, B.L. and Silcock, D.J., 1997. Nutrient and microbial changes in the peat proﬁle beneath Sphagnum magellanicum in response to addition of ammonium nitrate. J. Appl. Ecol. 34, 961–970. Williams, R.T. and Crawford, R.L., 1983. Microbial diversity of Minnesota peatlands. Microb. Ecol. 9, 201–214. Wilmshurst, J.M., McGlone, M.S., and Charman, D.J., 2002. Holocene vegetation and climate change in southern New Zealand: linkages between forest composition and quantitative surface moisture reconstructions from an ombrogenous bog. J. Quat. Sci. 17, 653–666. Wilmshurst, J.M., Wiser, S.K., and Charman, D.J., 2003. Reconstructing Holocene water tables in New Zealand using testate amoebae: differential preservation of tests and implications for the use of transfer functions. Holocene 13, 61–72. Woodland, W.A., Charman, D.J., and Sims, P.C., 1998. Quantitative estimates of water tables and soil moisture in Holocene peatlands from testate amoebae. Holocene 8, 261–273. Wuthrich, M. and Matthey, W., 1978. Les diatome´es de la tourbie`re du Cahot (Jura Suisse) 2: associations et distribution des espe`ces caracte´ristiques. Schwiz. Z. Hydrol. 40, 87–103. Wuthrich, M. and Matthey, W., 1980. Les diatome´es de la Tourbie`re du Cachot (Jura Neuchaˆtelois) III. Etude de l’apport e´olien et du transport par les oiseaux et insectes aquatiques 42, 269–284. Wynn-Williams, D.D., 1982. Simulation of seasonal changes in microbial activity of maritime Antarctic peat. Soil Biol. Biochem. 14, 1–12. Yeates, G.W. and Foissner, W., 1995. Testate amebas as predators of nematodes. Biol. Fert. Soils 20, 1–7.

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

319

Chapter 14

Peatland hydrology J. Holden

Introduction Other than protruding hummocks and some peat surface lawns, most of the mass of any peatland will spend its time in a waterlogged state. The hydrology of peatlands is fundamental to their development and degradation. Peatland classiﬁcation is often based on the dominant sources of water (groundwater or precipitation) or hydroecological patterns occurring within that environment. There were important Russian studies in the mid 20th century, which examined ﬂow through upper peat layers (Ivanov, 1948, 1953), but until recently most peatland hydrology had focused on the water budget (inputs and outputs) with relatively little attention given to hydrological processes such as overland ﬂow, inﬁltration, or pipeﬂow. Few studies gave much attention to the hydrological processes generating or attenuating storm runoff. The movement of water is a controlling ecological factor in peatlands (Hammond et al., 1990; Ingram, 1991), yet little work has been done on the spatial heterogeneity of surface and subsurface ﬂow in peatland environments, or on the spatial structuring of hydraulic peat properties (Chappell and Ternan, 1992; Baird, 1995). Neither of the reviews of peat hydrology by Ingram (1983) or Gilman (1994) discussed the spatial and temporal nature of runoff production within these environments in any detail, and very little mention was made of the inﬁltration process (Holden and Burt, 2002a). Peatland hydrology inﬂuences oxygen and gas diffusion rates, redox status, nutrient availability and cycling, and species composition and diversity; it is important for water resource management, ﬂooding and stream water quality. Peatlands accumulate carbon over time because immersion by water reduces the rate of decay so that it is less than the rate of production. Being such large stores of carbon, a decrease in water immersion (water table lowering) can convert peatlands into major sources of atmospheric carbon. In addition to gaseous carbon release, degraded peats also suffer from increased carbon losses in soluble and particulate forms. Changes to water ﬂow paths and water ﬂuxes across and through peatlands caused by environmental change may inﬂuence how soluble and particulate carbon is removed from peatlands. Therefore in order to predict the consequences of environmental change ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09014-6

320

J. Holden

on peatlands, whether the change is direct, such as drainage or restoration strategies, or inadvertent, such as climate change or chemical deposition, an understanding of the temporal and spatial variability of hydrological processes is required. This chapter provides an overview of recent developments in our understanding of peat hydrology. It is impossible to cover the hydrology of all types of peatland in depth within such a short space. Instead this chapter will focus on a small number of peatland types, such as blanket peat, with which the author is most familiar. Nevertheless the chapter will highlight important hydrological themes and distinguish between hydrological processes operating in different peat types where possible. A later chapter (Holden et al., 2006 – this book, Ch. 22) deals with an analysis of the hydrology of drained peatlands. This chapter begins by outlining components of the peatland water budget, followed by an overview of the traditional conceptual hydroecological model of peatlands known as the acrotelm–catotelm model. The overuse of the acrotelm–catotelm model has meant that the spatial and temporal nature of ecohydrological functioning in peatlands has been largely ignored. Although the acrotelm–catotelm model has broad utility, it ignores the important roles of macropores and soil pipes in connecting deep and shallow parts of a peat proﬁle, and that surface vegetation, topography and preferential ﬂow paths are important threedimensional components of peat hydrology.

Water sources and water budget for peatlands Peat is decaying organic matter that has accumulated under water-saturated conditions. Formation of peat therefore occurs in areas where precipitation or groundwater inﬂux exceeds evaporation and transpiration losses and where there is a slow rate of decomposition. Peatlands are likely to form in regions with high precipitation, such as upland areas of the temperate and boreal zones or in lowland areas where shallow gradients, impermeable substrates or topographic convergence maintain saturation. Classiﬁcation of peatland types is generally related to two fundamental factors: source of nutrients and source of water. Bogs are ombrotrophic peatlands dependent on precipitation for water and nutrient supply, whereas minerotrophic peatlands (fens) are reliant on groundwater for water and nutrient supply. Bogs are therefore highly acidic (pHo4) and contain low amounts of calcium and magnesium, whereas minerotrophic peats are less acidic and tend to be base rich. Blanket bogs form as a deposit that drapes over the local topography. In some areas there are raised bogs where the peat has grown into a dome with a halo of ‘lagg’ fen, overlying level mineral terrain or an inﬁlled basin (Bragg and Tallis, 2001). However, Charman (2002) suggest that raised bogs and blanket bogs are simply end-points of an ecological continuum. Traditionally water budget approaches have been used to examine the hydrological functioning of peats: P+G ¼ D(M, R)+Q+E where P is the precipitation, G the groundwater inﬂow, M the soil water storage, R the surface storage, Q the runoff, and E the evapotranspiration. To develop a water budget for a peatland requires good hydrometry. The hydrology of a wetland site can usually only be assessed on the basis of local data (Gilman, 1994).

Peatland hydrology

321

Inputs (1) Precipitation. Generally, ombrotrophic peatlands occur only in areas where there is over 600 mm of precipitation per year. However, fens may require very little rainfall as they are groundwater dominated. Many blanket bogs in the Northern Hemisphere require precipitation in excess of 1500 mm depending on evapotranspirative losses and topography. In some areas fog deposition may be an important contributor. Price (1992a) found that fog deposition was 50% of rainfall in a Newfoundland blanket bog. Evaporation was also restricted during fog and rain episodes so that the addition of water by fog combined with low-evaporation rates was critical to bog development. (2) Groundwater inflow. Subsurface springs may provide water to fen basins, acid valley and basin mires. Additionally ﬂooding may be the source of water for valley ﬂoodplain peats. In areas of tropical and temperate climate it is usual for peatlands to receive water from more than one source. In contrast, in arid areas away from coasts, rivers and lake systems, groundwater is the only signiﬁcant water component such as in the South Park fens of Colorado.

Stores (1) Water levels. The simplest data to collect, and frequently the only data available as time series, are water levels (Gilman, 1994). Peatlands have excess water that exerts an inﬂuence on the soil properties and the range and distribution of plant species. Peatland water tables usually lie close to the surface. Peatlands act as a large store of water. However, this water store may not act as a reservoir to maintain streamﬂow during dry weather (see below) as much of it is held tightly within the peat. Figure 14.1 presents water table data for the ﬁrst six months of 2004 from the Upper Wharfedale blanket peatland in Yorkshire, UK (541 130 5100 N latitude, 21 130 3800 W longitude). The dominant vegetation is Eriophorum sp. with some Sphagnum cover. Peat occurs on slopes up to 151. The peats are from 0.5 to 2.5-m deep, and formed during the Holocene underlain by glacial boulder clay. The climate is oceanic with 1800 mm of precipitation per year and a mean annual temperature of 61C. For 75% of the time the water table is within 10 cm of the surface. Only for short periods during the summer does the water table decline to levels below this. Even then, the maximum water table depth recorded from the site between December 2002 and December 2004 was 25 cm, which is very shallow when compared to other soil types. Saturated peat tends to be 90–98% water by volume. However, even above the water table, peat can still hold large volumes of water (approximately 90–95% water by volume). Even small amounts of rainfall are enough to raise the water table. At the site shown in Figure 14.1, data are available at 5-min intervals and the water table rises at a mean recharge of 16 mm h1. Rainfall intensity is the dominant control upon the rate of rise. When the water table is at 150-mm depth, 1 mm rainfall induces a rise of 29 mm; when the water table is within 50 mm of the surface, 1 mm of rainfall only induces a rise of 8 mm. Any change in water table elevation in upper horizons of

322

J. Holden

Figure 14.1. Automatically collected water table data from a dipwell in Oughtershaw, a small peat catchment in Upper Wharfedale, UK, for the ﬁrst six-month period of 2004.

less decomposed peat therefore represents considerably more water than a corresponding change in deeper, denser peats (Boelter, 1964). These relationships give a measure of the air-space volume (in effect, the speciﬁc yield fraction of porosity) in the peat that can be ﬁlled with water. Typically, the peat within the top 50 mm of the proﬁle will have four times as much air-space volume as peat at 150-mm depth. However, this is strongly controlled by the dominant peat-forming vegetation. In fen peatlands derived from reed and sedge remains, the peat has a much higher speciﬁc yield and some contain very large pores giving them a large hydraulic conductivity (Baird et al., 2004). Variations in water table elevation can result in signiﬁcant changes to ground surface level in peatlands. This is a particular problem in drained peats (Holden et al., 2006 – this book, Ch. 22) but the phenomenon can also be observed seasonally in intact peats. Hutchinson (1980) discovered a seasonal rise of 50 mm or more in Holme Fen Post, East Anglia, UK, and movements of the surface of Crymlyn Bog between 1985 and 1989 varied between 5 and 12% of the water table movement (Gilman, 1994). (2) Surface storage. In many peatlands, water is stored on the peat surface in pools. Peat pools have been found on every continent except Antarctica but are particularly characteristic of northern peatlands. The pools often form intricate networks ranging in size from localized systems to vast complexes covering the regional landscape (Glaser, 1998). The layout of pool systems typically reﬂects the morphology of peat landforms because pools are consistently orientated perpendicular to the slope. The dynamics of pool formation is still poorly understood in most environments, although are better known in permafrost peatlands. Porewater migrates to the freezing

Peatland hydrology

323

planes within the peat, creating massive bodies of ground ice. When these bodies rupture the insulating peat cover then ground ice is exposed to summer warmth and melts. The peat then collapses forming a cavity that can ﬁll with water. However, no single mechanism has been proposed for pool formation in other peatlands. In many areas pool development is cyclical and in a period of 30 years a pool can form and then inﬁll with vegetation. In other areas pools can remain for hundreds of years. There is a fundamental distinction between primary pools, present since the onset of peat formation and secondary pools, formed at a relatively late stage of peatland development. There is evidence from unpublished measurements (A.J. Baird, pers. comm.) that pools are local hotspots of carbon dioxide and methane loss. The reasons for this are not clear, especially given that pools are often underlain by welldecomposed, recalcitrant peat. One possibility is that pools receive waters via pipe ﬂow that are saturated in carbon dioxide and methane which degasses when experiencing lower pressures in the pool. Another is that dissolved carbon transported via pipes and matrix ﬂow to pools is mineralized to produce carbon dioxide and methane.

Losses (1) Evapotranspiration. There are two main components of water loss in peatlands; evapotranspiration and runoff (both surface and subsurface ﬂow). Atmospheric loss is often the largest component of water loss from peatlands (Baird et al., 2004). At the ecosystem scale, the energy, water and gas exchange processes are strongly coupled (Petrone et al., 2004). In addition to climatological parameters, the terrain, groundwater level and plant distributions control evapotranspirative release. Evapotranspiration was assessed by Kellner (2001) for Stormossen mire, a boreal Sphagnum bog in northern Sweden. Different Sphagnum moss species were found to have different capabilities to hold and transport water and so the surface wetness and the cover of vascular plants tended to vary among the micro-relief. This was reﬂected in the spatial variation in soil temperatures and the ground heat ﬂux was almost twice as much in hollows as in ridges. The Bowen ratio (sensible heat ﬂux divided by latent heat ﬂux) did not change with temporal ﬂuctuations in peat wetness but depended mainly on net radiation, air temperature and relative humidity. Therefore there was a trend of a falling Bowen ratio during the day and during the summer season. More than half the evapotranspiration originated from the moss surface, with a larger part in the pre-leaf period. Where there was a relatively large quantity of non-transpiring biomass (brown leaves, Ericaceae stems, and dead grass) sheltering the ground, there was a high Bowen ratio and large surface resistance to evapotranspiration. In Canadian peatlands, evaporation from non-vascular Sphagnum mosses was shown to be well below potential evaporation (Campbell and Williamson, 1997) compared with relatively efﬁcient latent heat transfer by vascular sedges. Price (1996) noted that daily net radiation and evaporation ﬂux from a Sphagnum-dominated surface were similar to a bare peat surface the latter having an effective capillary water supply. Figure 14.2 shows that for the site presented in Figure 14.1 the water table level is sometimes controlled almost entirely by evapotranspiration. During dry periods

324

J. Holden

Figure 14.2. Diurnal steps in the water table from the dipwell presented in Figure 1 for July 3rd – 8th, 2004.

when the water table drops below about 8 cm there are clear diurnal patterns with very little change during night time hours. Evans et al. (1999) found comparable diurnal steps and these steps have been reported elsewhere (Charman, 2002). Thus, once the water table drops more than a few centimeters below the surface, the peat appears to be able to retain its water without allowing any further free drainage. Any falls in water table below 8 cm are controlled by evapotranspiration alone. Groundwater–evapotranspiration models and general atmospheric-loss models continue to be developed for peatlands, particularly with reference to restoration and modeling of global climate change. In particular Bowen ratio, Priestly–Taylor and Penman–Montieth models are being reﬁned for peatlands environments across the world (Baird et al., 2004). (2) Streamflow. Ombrotrophic peatland catchments tend to have very ﬂashy hydrological regimes. In the continental bogs of Minnesota (Bay, 1969), the blanket bogs of Newfoundland (Price, 1992b), and in British blanket peats (Evans et al., 1999; Holden and Burt, 2003a) streamﬂows were found to be dominated by highpeak ﬂows and discontinuous summer ﬂows. Response to rainfall was rapid, and peat streams tended to have hydrographs with steep recessional curves and minimal baseﬂow. Peat stream discharge was investigated in the 11.4 km2 Upper Tees catchment in the UK and is shown in Figure 14.3. The catchment is covered by blanket peat of up to 3-m depth dominated by Eriophorum vaginatum, Calluna vulgaris, and Sphagnum species (mainly S. rubellum, S. papillosum and S. magellanicum). A typical intact peat proﬁle consists of an upper 5 cm of poorly humiﬁed (H2–H3 on the Von Post (1922) scale) black brown-colored peat with living roots and a crumb structure. Below this to 10 or 15 cm depth the peat tends to be brown and slightly humiﬁed (H3–H4) with occasional bands of light brown Sphagnum peat overlying a darker brown Eriophorum – Calluna– Sphagnum peat (H4). The peat then very gradually becomes more humiﬁed

Peatland hydrology

325

Figure 14.3. Discharge for the 1999 water year from the 11.4 km2 Trout Beck blanket peat catchment, northern England, based on 15 min gauging (after Evans et al., 1999).

with depth. By 1.5 m into the proﬁle the peat is highly humiﬁed with decomposition almost complete (H9). Frequently there are well-preserved remains of birch found at the base of the peat that overlies light colored gray clay with sandstone boulders. The clay is often gleyed and waterlogged. Maritime air masses from the North Atlantic Ocean dominate the climate that has been classiﬁed as ‘subarctic oceanic’ with a mean precipitation of 1982 mm and mean annual temperature of 5.31C. Within the Upper Tees catchment ﬂows below 0.5 m3 s1 occurred 75% of the time but only 21% of the total discharge occurred during this time. The pattern is indicative of minimal groundwater ﬂow from the peat. The poor maintenance of baseﬂow is a problem for water companies in the UK so that during drought periods, reservoirs supplied by upland peats tend to be at risk of severe depletion. This is despite the maintenance of high water tables in peatlands even during dry periods. Mean storm runoff/rainfall ratios from the Tees catchment are around 40% (Evans et al., 1999), with annual catchment efﬁciencies in excess of 72%. This is high and considering the rapidity of the runoff suggests a limited rainfall storage capacity. There tends to be a close correspondence of rainfall and runoff. Figure 14.4 illustrates that even if the winter periods were wet, this would not be sufﬁcient to maintain summer ﬂows because bog peatlands have little storage capacity for rainwater due to the maintenance of high water tables. Thus, in contradiction to an oftenexpressed view (ﬁrst expounded by Turner, 1757), peatlands do not always behave like a sponge (Ingram, 1983); rather water is released rapidly following rainfall or snowmelt. Baseﬂows are often poorly maintained and many small tributaries dry up completely after only a week without rain. At the individual storm level, Figure 14.5 illustrates the effect of water table recharge, on runoff production, after a dry summer spell. Here rainfall in excess of 5 mm h1 produces minimal hydrograph response while the water table is deep below

326

J. Holden

Figure 14.4. Monthly rainfall and runoff totals for the 1999 water year from Trout Beck (after Evans et al., 1999).

the surface. Later in the storm, just under 4 mm h1 rainfall is sufﬁcient to trigger a rapid hydrograph rise. The difference between the two responses is controlled by the level of the water table, which was at 24 cm depth before the storm but had risen to within 5 cm of the surface when the hydrograph response was triggered. Response to initial rainfall is much more rapid when the water table is close to the surface as indicated by Figure 14.5b and the result is a greater peak ﬂow. Thus it appears that streamﬂow is dominated by runoff associated with peatland saturation (saturationexcess overland ﬂow – see below) rather than by overland ﬂow associated with higher rainfall intensities than the inﬁltration capacity of the peat (inﬁltration-excess overland ﬂow) (Holden and Burt, 2003a, b, c). Similar results were found at the plot-scale by Holden and Burt (2002a) who performed a series of rainfall simulation experiments on blanket peat. Some peatlands do contribute more to sustaining baseﬂow but these tend to be ones that are connected to a much wider hydrological system. For example Roulet (1990) found the mean annual daily runoff from a southern Ontario peatland was 46 mm day1. However, it is important to note that the peatland itself had little effect on the magnitude of the ﬂux that was actually controlled by groundwater discharge through the peatland (Roulet, 1988; Burt, 1995). In certain topographic locations some peatlands will inﬂuence regional ﬂow regimes by intercepting catchment runoff and storing some of the storm waters. The impact of this would be to reduce peak ﬂows. However, this will depend on the size and location of the peatland relative to the drainage network (Heathwaite, 1995) and the time of year (Ogawa and Male, 1986). In spring there may be a lower water storage capacity to store water than in summer and, in fact, the attenuation may be more strongly related to the size and duration of a particular storm event. During the winter months, even those peatlands that normally reduce peak ﬂows may actually contribute to a higher ﬂoodpeak as they will be fully saturated and may be draining water from the upper layers. Thus, Burt (1995) concluded that, for British peatlands at least, they rarely act to attenuate

Peatland hydrology

327

Figure 14.5. Hydrograph and water table data from two storms in the Trout Beck catchment illustrating the importance of near surface water tables in generating runoff. (a) July 6th, 1995. (b) May 22nd, 1996.

328

J. Holden

ﬂow and are much more likely to contribute to storm runoff due to their propensity for rapid saturation. He suggested that only where a peatland lies between groundwater sources and the river can exert some ameliorating inﬂuence on downstream hydrology. Devito et al. (1996) found that during seasons with large water inputs, swamps in the Canadian Shield peatland were hydrologically connected to uplands, and that overland ﬂow dominated in the peatland. Quinton and Roulet (1998) and Glenn and Woo (1997) found that some Canadian peatlands operate as a single source area with rapid runoff response when the water table exceeded the depression storage capacity of the peatland pools. Relatively slower responses occurred when pools became disconnected into separate microcatchments during drier periods. Peatlands with permafrost and those disturbed by beaver pools may behave in an even more lagged way. Soil moisture deﬁcits during dry years in boreal, forested peatlands can have a signiﬁcant impact on the magnitude of the subsequent spring runoff peak (Hillman, 1998; Woo and Young, 1998). This is in contrast to the less lagged response shown in Figure 14.5 for a blanket peatland in the UK, which does not have a permafrost or permanent winter snow cover.

Introduction to hillslope hydrology The streamﬂow is the end product of a range of runoff production processes that have occurred within a peatland. Runoff pathways (surface and subsurface) are important in their own right because they will control the rate at which water moves to the stream channel, as well as having an inﬂuence on its solute and sediment content. Water that moves through subsurface soil layers tends to have a very different chemistry from that which has solely moved over the soil surface, for example. Here I will give a brief introduction to runoff production processes before moving on to discuss how those processes operate within peatlands. Further detail on hillslope hydrology is provided by texts such as Kirkby (1976) and Ward and Robinson (2000). It should also be noted that the nature of ﬂow processes in peat catchments has until recently been poorly understood (Holden and Burt, 2003a, c). It is now known that the hydrological processes operating on hillslopes range from inﬁltration-excess overland ﬂow to saturation-excess overland ﬂow, through subsurface ﬂow within the matrix, within macropores and through natural pipes, to ﬂow through the underlying bedrock or sediments (Fig. 14.6). The relative importance of the ﬂow processes in any catchment varies with climate, topography, soil character, vegetation cover and land use, and may vary seasonally at one location with antecedent moisture and with precipitation intensity and duration (Burt, 1996). The runoff processes are by no means independent of one another and water traveling over the surface at one point may later take the form of subsurface ﬂow through the matrix and then ﬂow through macropores, for example. It is important to distinguish between the different forms of overland ﬂow and subsurface ﬂow because the speed of water movement and the nature of nutrient and sediment ﬂuxes are often controlled by the ﬂow process. For example, there are important differences between inﬁltration-excess overland ﬂow and saturation-excess overland ﬂow. Inﬁltration-excess overland ﬂow is produced when the rainfall intensity

Peatland hydrology

329

Figure 14.6. The main hillslope runoff pathways (after Holden, 2005b).

is greater than the inﬁltration rate and the overland ﬂow therefore consists of water that has not been within the soil. This type of surface runoff is most likely on soils with low inﬁltration capacity (the ‘partial contributing area’ concept, Betson, 1964). Saturation-excess overland ﬂow can occur at much lower rainfall intensities and is produced when the soil proﬁle is completely saturated; the water at the surface is a mixture of water that has been within the soil mass that is returning to the surface from upslope and fresh rainwater. Saturation-excess overland ﬂow can occur for long periods after rainfall has ceased, particularly at the foot of a hillslope where the soil continues to be supplied by water draining from upslope (Burt, 1996; Holden, 2005 b). The source areas (parts of the hillslope which contribute runoff) for saturationexcess overland ﬂow may be very different from those for inﬁltration-excess overland ﬂow and will vary over time (the ‘variable source area concept’, Hewlett, 1961). The differences between runoff produced by these processes has often been determined hydrochemically (Ogunkoya and Jenkins, 1991), with saturation-excess overland ﬂow generally having a greater solute concentration than inﬁltration-excess overland ﬂow (since it is a mixture of old soil water and precipitation unable to inﬁltrate into the soil). Clearly both forms of overland ﬂow are also capable of transporting sediment over the surface of a peatland. Subsurface ﬂow may be generated through the soil or peat matrix or by ﬂow through macropores or pipes. Research has indicated that macropores can be important in solute transport through soils (Beven and Germann, 1982). Macropore ﬂow has been shown to develop in peats that have been cut and air-dried to supply

330

J. Holden

Irish power stations (Holden, 1998), but until recently little work had been done on macropore ﬂow in intact peats (Baird, 1997a; Holden et al., 2001). Water ﬂow through pipes is a much-neglected process, yet these subsurface features are present in many peatlands (Elgee, 1912; Ingram, 1983; Price, 1992a; Jones et al., 1997; Holden and Burt, 2002b; Holden, 2005a). The occurrence of pipes in many environments is strongly associated with faunal activity, but this appears unlikely in most acidic peats. There have been few detailed surveys of pipe density or contribution to runoff production in peat catchments. Jones (1981), Jones and Crane (1984), and Jones et al. (1997) worked on the shallow peaty podzols of mid-Wales where pipes were typically found at soil horizon interfaces and could be responsible for up to 50% of the runoff generation. Price (1992a), Holden and Burt (2002b), Holden et al. (2002), and Holden (2005a) reported pipes in deeper blanket peat. The acrotelm–catotelm model Since the mid-20th century Russian scientists have adopted a diplotelmic system for understanding the functioning of peatlands. This comprises an upper active peat layer with a high hydraulic conductivity and ﬂuctuating water table, and a more inert lower layer which corresponds to the permanently saturated main body of peat (Ivanov, 1948, 1953, 1981; Lopatin, 1949; Romanov, 1968). This layering system for analyzing peatlands became widely accepted from the late 1970s and is now used regularly in ecohydrological and peat development modeling and budgeting (Ingram, 1982, 1983, 1991; Kirkby et al., 1995; McKillop et al., 1999; Hilbert et al., 2000; Holden and Burt, 2003c). Ingram (1978, 1983) noted that the distinction between the upper, periodically aerated, partly living soil layer (acrotelm) and the lower anaerobic layer (catotelm), which is dead except for the aerenchymatous roots of helophytic angiosperms, is an important concept and fundamental to any understanding of the hydrology, ecology, and pedology of mires. According to Ingram’s deﬁnition, the acrotelm is affected by a ﬂuctuating water table (the lowest water table depth is therefore the base of the acrotelm), has a high hydraulic conductivity and a variable water content, is rich in peat-forming aerobic bacteria and other microorganisms and has a live matrix of growing plant material. The catotelm has water content invariable with time, a small hydraulic conductivity, is not subject to air entry and is devoid of peat-forming aerobic microorganisms. The acrotelm–catotelm model implies that most runoff production and nutrient transfer will occur within the upper peat layer, close to or at the peat surface. This would match the ﬁndings of the catchment streamﬂow and water table studies discussed above.

Runoff processes within peatlands Matrix and overland flow Some of the detailed process-based measurements of runoff production at the hillslope and plot scale by Holden and Burt (2003a), and data shown for the ﬁrst time in

Peatland hydrology

331

Table 14.1. Percent of runoff collected in automated through ﬂow troughs from peat layers in Upper Wharfedale, December 2002–December 2004. Peat layer (depth, cm)

Percent runoff from hillslope

0–1 1–8 8–20 420

74 21 5 o0.01

Table 14.1 ﬁt the acrotelm–catotelm model. Table 14.1 is based on data from an undisturbed blanket peatland hillslope in Upper Wharfedale, UK. Most runoff (74%) is produced from the surface of the peat and most of the rest from the upper 20 cm of the peat proﬁle. Lateral ﬂow at greater depths is restricted such that runoff contribution from these layers is low. Hence the low hydraulic conductivities found at relatively shallow depths within the blanket peat (Rycroft et al., 1975; Holden et al., 2001) result in minimal ﬂow contributions from most of the peat mass. The main applications of the diplotelmic model have been to ombrotrophic mires and particularly raised bogs. However, whereas some evidence from raised bogs suggests that the catotelm is, effectively, impermeable (van der Schaaf, 2002a, b), in other raised bogs the hydraulic conductivity has been measured to be several orders of magnitude higher and the catotelm does produce a signiﬁcant water ﬂow (Ingram, 1982; Chanson and Siegel, 1986; Glaser et al., 2004b). It should therefore be noted that large differences can occur between bogs of apparently very similar form. These differences have yet to be fully examined. It is clear that different peats have different properties of their acrotelm and catotelm. Ingram and Bragg (1984), for example, view the acrotelm of araised mires as comprising poorly decomposed, high hydraulic conductivity Sphagnum peat often 50 cm thick. Such an acrotelm is uncommon in many blanket peats except in localized ﬂushes or hollows. Hence the nature of the peat surface and upper peat layers are likely to be very important factors in determining runoff production processes within different types of peatland. Boelter (1964) made an important but often ignored point that the hydrologic role of any peatland depends on the type of peat found in the peatland. Many model calculations rely heavily on measurements of hydraulic conductivity of the peats. However, measurements of hydraulic conductivity in peat soils are rarely within one order of magnitude error bands (Holden et al., 2001). Furthermore, there is evidence that the hydraulic conductivity of peat soils can vary over several orders of magnitude over just a few meters vertically or horizontally (Rycroft et al., 1975; Holden and Burt, 2003b). This makes groundwater ﬂow modeling difﬁcult since the size of computational cells is usually greater than the scale at which signiﬁcant variability in hydraulic parameters occur (Bromley and Robinson, 1995; Baird, 1997b). Traditionally, rigid soil theory has been misapplied to calculate the hydraulic conductivity (K) of peat soils that are, in fact, compressible. Baird and Gaffney (1994) used compressible soil theory in a fen peat and found that there were likely to be large errors in the earlier rigid-soil calculations applied to peats. Holden and Burt (2003b)

332

J. Holden

used compressible soil theory to calculate K in a blanket peat in the UK. They found values of K in the top 80 cm of the peat (mean of 2.4 106 cm s1) were an order of magnitude lower than those found by Baird and Gaffney (1994) in the Somerset Levels. Hydraulic conductivity is often assumed to decrease with depth, but Holden and Burt (2003b) found that, other than a higher K in the top 10 cm of the peat, there was no signiﬁcant systematic variation in K with depth. Beckwith et al. (2003), however, used a modiﬁed cube method in the laboratory to determine peat K and did ﬁnd signiﬁcant K-depth relationships from a cutover raised bog, though the relationship was not simple. What they suggested was that particular layers (presumably related to the vegetation properties of that layer) had characteristic K values. This did not necessarily mean that K declined with depth but that K was related to the properties of a particular peat layer. Chason and Seigel (1986) tested the vertical patterns of K in a raised bog, and two fen peatlands. In all three there was considerable vertical variability in K measured. On a larger scale, Holden and Burt (2003b) found that K varied signiﬁcantly between sampling sites across a catchment and a hillslope (p ¼ 0.002) and suggested in some circumstances that this may be more important than vertical differences in K. These results highlight the difﬁculties for generalizing catchment-scale hydraulic conductivity based on only a few measurements. Hillslopescale and catchment-scale variability were found to be much greater than plot-scale variability. The woody peats of continental North America may be even more prone to high K variability than the oceanic peats of Western Europe (Charman, 2002). There have been recent advances in the study of subsurface water ﬂow in peatlands including the North American and UK work on K determination and the effect of biogenic gas bubbles (Beckwith and Baird, 2001; Baird and Waldron, 2003; Beckwith et al., 2003). Glaser et al. (2004a), Kellner et al. (2004) and Baird et al. (2004) have shown that gas ebullition can occur from both deep and shallow peat. Gas pockets can build up within peatlands and may be released episodically. The build up of biogenic gas bubbles has been shown to have a signiﬁcant impact on water ﬂow through peatlands. Blocking of pores by gas bubbles can reduce K by 5–8 times (Beckwith and Baird, 2001). Rainfall simulation experiments on peats have conﬁrmed the dominance of overland ﬂow on both vegetated and bare peat surfaces (Holden and Burt, 2002a). Overland ﬂow occurs across peatlands with and without Sphagnum. Ingram and Bragg (1984) and Ingram (1991) suggested that the acrotelm itself possesses the essential characteristics of a layer that suppresses overland ﬂow, and is thus selfsustaining. This is particularly important because Sphagnum, for example, has no roots and could in theory be washed away by overland ﬂow (Bragg and Tallis, 2001). However, overland ﬂow can frequently be seen over both long-established and regenerating Sphagnum without resulting in erosion (Fig. 14.7) and ﬁeld observations suggest Sphagnum carpets can survive rapid and deep overland ﬂow. However, the experiments of Holden and Burt (2002a) did indicate that there are relationships between ﬂow production and vegetation cover. This is likely to be related to peat properties determined by vegetation cover at a particular site. It is well known that particular vegetation types prefer different water table conditions (Ingram, 1983); furthermore the vegetation may directly affect peat structure through the root structures, litter deposition and the accumulation of the peat deposit.

Peatland hydrology

333

Figure 14.7. Overland ﬂow across a Sphagnum carpet in the Upper Tees catchment, UK.

Figure 14.8 presents precipitation (Fig. 14.8a), streamﬂow in a tributary of the Tees River (Fig. 14.8b), and hydrographs from a storm event on a blanket peat hillslope (Fig. 14.8c–g) during 1999. Data were collected using automatic runoff troughs connected to tipping bucket ﬂow recorders. Overland ﬂow on the footslope site is more prolonged than on upslope sites. As the hillslope drains, return ﬂow is produced on gentler slopes producing saturation-excess overland ﬂow on the footslopes for a longer period than seen on steeper hillslope sections or at the crest of the hill. When overland ﬂow has ceased on the footslope, the ﬂow record from the 5-cm deep trough indicates that the near-surface layers of the peat continue to drain. Runoff from 5 cm depth tends to be more prolonged than at the surface with much more rounded and less peaky hydrograph forms. This is indicative of a limited ﬂow capacity of the near-surface layer of blanket peat and of the dominance of saturation-excess runoff generation. Flow from the surface and 5-cm troughs is

334

J. Holden

Figure 14.8. Runoff production from monitoring sites across a small peat hillslope, Julian days 236–241, 1999 (August 26–29, 1999). (a) precipitation; (b) trout beck; (c) topslope overland ﬂow; (d) midslope overland ﬂow; (e) footslope overland ﬂow; (f) midslope throughﬂow at 1–5 cm depth; (g) footslope throughﬂow at 1–5 cm depth (after Holden and Burt, 2003a).

Peatland hydrology

335

ephemeral. Given the minimal contribution of ﬂow from deeper layers in the peat (no ﬂow recorded for peat below 10 cm depth at this site) this indicates that peatlands can release their gravitationally free water rapidly following rainfall. Figure 14.9 presents runoff dynamics on the slope at different stages of ﬂow recession. Here overland ﬂow (or at least surface ponding) was recorded by a network of 250 crest stage tubes over almost the entire hillslope at the peak of the storm at 0300, Julian day 239, 1999; that is, August 27, 1999 (in this case it is the day of the year starting counting from January 1; not to be confused with the day of the ‘Julian calendar’) (Fig. 14.9a). Small-scale microtopographic differences could be found on the hillslope but the measurement network allows the general pattern of hillslope runoff production to be displayed. As the hillslope drains after the rainfall has stopped, the more gently sloping top and footslope regions continue to produce overland ﬂow with the steeper slopes producing ﬂow just below the surface at 3 cm

Figure 14.9. Minimum depth of ﬂow from the peat surface on a small peatland hillslope, Julian day 239–240, 1999 (August 27–28, 1999 as monitored by crest-stage tubes. (a) 0300 h of day 239; (b) 0900 h of day 239; (c) 2100 h of day 239; (d) 0900 h of day 240 (after Holden and Burt, 2003a). Contours are in meter above local datum, OLF refers to overland ﬂow.

336

J. Holden

(Fig. 14.9b). By 2100 hr (Fig. 14.9c) the surface-saturated zone of peat only exists on the hillslope toe regions whereas steeper areas drain to produce ﬂow down to depths of 6 cm and occasionally 9 cm. After 0900 h of Julian day 240 (August 28, 1999) there is only very slow change (Fig. 14.9d). Drainage of free water available in the upper soil layers of the slope is sufﬁciently rapid such that within 30 h the hillslope has reached a quasi-equilibrium state with water tables stabilized. Runoff from almost the whole hillslope becomes minimal. The only fully saturated area is on the right ﬂank of the hillslope where monitoring has indicated that the peat is almost permanently waterlogged due to poor drainage. Thus, topography is important for determining dominant runoff process contributions even on low-gradient peat. This is an obvious statement but is often neglected due to the oversimpliﬁcation when using the acrotelm–catotelm model. The steeper midslope sections of the slope produce overland ﬂow less frequently than shallower slopes. This suggests that the midslope sections produce more subsurface runoff that collects at the bottom of the slope, and, due to impeded drainage, manifests itself as return ﬂow. Mean water tables were higher on shallower slopes, notably at the crest of the hillslope and at the foot of the slope. The spatial pattern of overland ﬂow development on peatland hillslopes clearly shows that saturation-excess overland ﬂow occurs and that inﬁltration-excess overland ﬂow is rare. The spatial pattern of peat saturation has a dominant inﬂuence on runoff production in most peatlands. This does not invalidate the diplotelmic model, but reminds us that spatial variation in hillslope runoff generation is not unimportant. On many peatland slopes it is not only topography that controls the spatial and temporal patterning of runoff production. Hillslope-scale preferential pathways for water have been identiﬁed, usually related to peat with higher hydraulic conductivity. These ﬂow lines have sometimes been attributed to headstreams that were originally developed in mineral ground, but have become overgrown by peat rather than collapsing later. Ingram (1967) also identiﬁed water tracks in peats where preferential ﬂow seemed to occur.

Preferential flow Tracer studies (Baird and Gaffney, 2000) and tension inﬁltrometer tests have shown that macropores (here deﬁned as pores greater than 1 mm in diameter) are important pathways for runoff within peatlands. Baird (1997a) found between 51 and 78% of the ﬂux moved through macropores in a fen peat, whereas Holden et al. (2001) found macropore contributions to inﬁltration were around 35% in a blanket peat. Sphagnumcovered peat had a signiﬁcantly greater macroporosity and hydraulic conductivity down to 20 cm depth than other surface cover types. Macroporosity was found to be greatest at 5 cm depth, being slightly lower at the surface, but declining rapidly below 5 cm depth. This is to be expected given that the most dense root networks and least densely spaced Sphagnum branches would be in the upper part of the peat proﬁle (Ingram, 1983). Macroporosity has been found to increase following drought (Holden and Burt, 2002 c) and peat drying has been associated with permanent structural changes (Holden et al., 2006 – this book, Ch. 22). Holden and Burt (2003 a) reported a site where rapid runoff was recorded emerging from the base of a blanket peat layer at

Peatland hydrology

337

the interface with an underlying clay. It was clear that a macropore was allowing surface water to percolate rapidly to the peat base and then run laterally across the clay interface. There was no slow drainage of the peat at this site and ﬂow was only activated when there was rainfall and overland ﬂow occurring. These ﬁndings suggest that we need to move beyond the acrotelm–catotelm model in peatlands if we are to more accurately represent hydrological processes that are occurring within these environments. Natural soil pipes (Fig. 14.10) are larger continuous forms of macropores and can transport signiﬁcant amounts of water, sediment and solutes through peatlands. Holden and Burt (2002b) measured pipe discharge in a blanket peat catchment in northern England. Response times from all the pipes were short, even from pipes deep within the peat, and peak ﬂow from a single pipe was as great as 4.6 L min1. Overall, pipeﬂow contributed around 10% of the streamﬂow volume at the site. On the rising and falling limbs of the stream hydrograph, however, pipes could contribute up to 30% of streamﬂow (Fig. 14.11). Most pipes in the study catchment responded to low rainfall intensities and totals even after a dry period. In contrast to the mainly ephemeral systems examined by Gilman and Newson (1980) there is no evidence to suggest a minimum rainfall threshold for pipeﬂow. It seems that pipes in deep blanket peat are well connected to the surface, receiving drainage far more quickly and in greater volumes than would be expected simply from diffuse seepage through the overburden. Holden and Burt (2002b) suggested that most of the pipes receive their water from saturation-excess overland ﬂow and near-surface ﬂow; the water enters pipe networks near the surface or where the pipes are open to the surface. Macropores may provide bypass routes for water into pipe networks, and

Figure 14.10. Two natural soil pipes emerging at a peatland streambank in the Upper Tees catchment, UK. The camera case is shown for scale. The lower (left) pipe does not ﬂow as frequently as the upper pipe from which ﬂow can be seen emerging in the photograph.

338

J. Holden

Figure 14.11. Pipeﬂow contributions to streamﬂow in a small blanket peat catchment during Julian days 241–272, 1999 (August 29–September 29, 1999) (after Holden and Burt, 2002a).

pipe formation itself has been linked to crack formation in the peat following dry weather (Gilman and Newson, 1980; Jones, 1981; Jones et al., 1997). Collapse features are common, allowing surface water to readily enter the pipe network. However, little is known about the initiation of pipes or their enlargement by erosion. Often sediment is deposited on the peat and vegetation surface, where a pipe has overﬂowed during a storm event. This sediment can contain a large proportion of mineral material from the underlying substrate (Fig. 14.12). The existence of pipes and macropores therefore opens the way for the ﬂuxes of water, sediment and nutrient contributions from deep within and below the peat rather than simply by rapid transfer through the acrotelm. Therefore the conceptual model of water and nutrient supply within even the most ombrotrophic bogs needs to be re-assessed. The study by Holden and Burt (2002b) is the only detailed study of pipeﬂow in a peatland anywhere in the world. There were some, more limited, measurements of pipe ﬂow in some Arctic peatlands by Quinton and Marsh (1998) and Carey and Woo (1999) who found pipe drainage to be important (Price and Waddington, 2000). There have been no studies, that the author is aware of, that examine in any detail the role of peatland pipes in sediment, dissolved gas or solute production. Weekly sampling on some shallow peaty podzols of the Maesnant catchment, in mid-Wales suggested 15% of annual stream sediment yield came from the pipes (Jones and Crane, 1984), and Jones (1990) speculated that this value was more likely to be 25% when the unmonitored pipes were taken into account. Jones (2004) showed that for Maesnant the areas of piping yielded more sediment to the stream than the areas without piping. Pipes have been found in peatlands across the world but relatively little is known about the number and extent of soil pipes in most peatland (or any other) environments. Commonly pipes are only reported where they issue onto streambanks or where their roofs have collapsed. However, as Jones et al. (1997) noted, many pipes are not directly connected to streambanks. Furthermore, Holden et al. (2002),

Peatland hydrology

339

Figure 14.12. A surface mineral sediment deposit that originated from approximately 3 m below the surface. The deposit occurred during a storm event when a soil pipe overﬂowed onto the peat surface. This provides the surface with sediment and nutrients that would otherwise be limited by rainfall in a blanket peatland.

Holden and Burt (2002b) and Terajima et al. (2000) showed that soil pipe dimensions and depths could be very different just a few meters upslope. Soil pipes are commonly not just formed as a single conduit but can form complex drainage networks with branching tributaries (Holden et al., 2002). The main difﬁculty with collecting data in order to examine pipe distribution in peatlands has been the lack of appropriate techniques for detecting and mapping subsurface soil pipes. Recently, Holden et al. (2002) reported on the successful application of ground-penetrating radar (GPR) for identifying soil pipes and their hydrological connectivity (Holden, 2004). This technique allows soil pipes to be identiﬁed from the ground surface without disturbance. Several factors that might control pipe distribution in peatlands have been suggested. These include topographic position, slope gradient, aspect, and land management. Many authors, such as Jones (1981) and Price (1992 b) have suggested that piping is more common on steeper slopes and that ﬂatter slopes (such as hilltops) with limited drainage area per unit contour length are less susceptible to soil piping (Jones et al., 1997). This is because steeper hydraulic gradients and greater volumes of water ﬂow are more likely to result in pipe formation (with greater shear stresses on macropore and pipe walls). Based on an initial ﬁeld survey, however, Holden et al. (2002) suggested that, in blanket peat catchments, soil pipe densities may be much greater on more gentle slopes (such as valley ﬂoors and hill tops) but they had very limited data with which to back up this hypothesis. Therefore, Holden (2005a) used a systematic GPR survey of 160 peat catchments to test this theory. He found piping in all 160 catchments with a mean of 69 pipes per kilometer of GPR transect. Topographic position was found to be a signiﬁcant control on soil pipe frequency (po0.001). Topslopes and toeslopes were found to have signiﬁcantly higher densities

340

J. Holden

of soil pipes than midslopes. However, slope gradient and topographic index (a measure of likely saturation related to the area draining to given point divided by the slope) were not signiﬁcant. One of the reasons that slope may not be an important factor here (unlike on other soil types) is that peats do not readily form on slopes greater than 151, due to the need for waterlogged conditions. The topographic position control outlined above for piping is strikingly similar to many surface erosion features seen in peatlands and suggests that both processes are linked. Bower (1960) described two types of gully erosion in blanket peats. Type 1, consisting of a close network of freely and intricately branching gullies, occurred on hilltops of deep peat (41.5 m). In contrast, more open, linear gully systems with less branching (Type 2) were more common on midslopes where gullies form sub-parallel trenches running downslope. Data presented in Figure 14.13 illustrates the variability of K and dry bulk density (DBD) on one of the survey hillslopes examined by Holden (2005a). There is a much greater variability in K and DBD on the topslope and footslope parts of this hillslope than in the midslope section. This was consistent across the slopes sampled by Holden (2005a). The standard deviation of K for midslope plots was only 6.5 106 cm s1, compared to 20.4 and 16.2 106 cm s1 for top and footslope plots. In addition, the average of K was slightly higher on the midslopes. Peatland piping may therefore be a function of inherent soil properties on different parts of the hillslope. The dominance of the topographic control suggests that piping and landform development are intimately related. Some models simply propose that pipes result in landform change (topographic depressions or gullying) or that existing topography promotes enhanced ﬂow in concentrated areas of hillslopes which promotes piping (Jones, 1990). However, I propose that the nature of the underlying topography (and

Figure 14.13. Saturated hydraulic conductivity and dry bulk density proﬁles for a peatland hillslope in Caithness, northern Scotland by slope position. (a) Topslope; (b) midslope; (c) footslope.

Peatland hydrology

341

its associated drainage conditions) promotes differential build up of the peat deposits. This occurs because of the development of micropools and larger bog pool systems on hilltops and footslopes that are colonized by a mosaic of plants with specialist positions within the microtopography. The remains of these plants are then incorporated into the peat as it thickens, resulting in a peat of variable properties throughout its proﬁle. In addition, bog pool development tends to be a cyclic process; pools disappear from one spot in a peatland while new ones form elsewhere as differential plant growth in pools and hummocks interacts with an ever-changing local topography while the peat deposit thickens (Weber, 1902; Clymo, 1991; Glaser, 1998). This means that peat where bog pools have been present is inherently more susceptible to preferential ﬂow and piping. It was shown above that shallow throughﬂow dominated as the runoff process on blanket peat midslopes. This compares with footslopes and topslopes where saturation-excess overland ﬂow dominated and there was more switching between overland ﬂow and throughﬂow dependent on antecedent conditions. Better-drained midslopes therefore have a more uniform structure, less variable K, and more uniform (spatially and temporally) runoff production. The associated midslope plant formations tend to be more homogeneous. Midslopes are therefore less susceptible to wandering and branching pipe networks. This homogeneity combined with gradient will allow pipe branching to be at a minimum on midslopes. Jones (1981) noted that pipes are common where hydraulic conductivity changes abruptly. Results from peatlands suggest that there can be up to two orders of magnitude variation in K, vertically and laterally, over just 10 cm of peat. This is particularly prominent on top and footslopes. Notably, these are also sites where the change in hydraulic gradient is greatest. On convex upslope and concave downslope hillslopes the hydraulic gradient increases toward the midslope from the topslope, and then decreases from the midslope to the footslope. Thus, greater changes in hydraulic gradients and peat heterogeneity combine on footslopes and hilltops to increase the propensity for wandering preferential ﬂow and pipe formation. In addition as peat depths are often greater on hilltops and toes, more (and larger) pipes can form. Holden (2004), for example, showed that pipe networks could overlap each other vertically in the soil proﬁle and yet not be hydrologically connected. This is less likely in a shallower peat because the pipes are more likely to connect to form one pipe. There are also links between other peatland landscape features and pipes which require further research. In some catchments there are areas where bog pools are common and are associated with piping (Holden et al., 2002). Many of the pipes seem to originate in pool–hummock complexes and some pipes drain directly into gully heads. Further work is required to test whether the pipe networks are of a similar form to the surface features and to search for further process links between piping and landform development in peatlands.

Conclusions Traditional research and modeling in peatlands has focused on black-box water budget approaches that mask the true complexity of peatland hydrology. Any

342

J. Holden

research into peat water movements has also traditionally been conﬁned to the acrotelm–catotelm school of thought. This chapter has presented recent research on hydrological processes operating within peatlands. I have shown that the application of the acrotelm–catotelm model needs to be rethought. In particular there is a spatial and temporal functioning of peatland hydrology that has previously been ignored and this can be partly related to a reliance on the simplicity of the model. While saturation-excess overland ﬂow dominates in most peatlands there are clear links between runoff production and topography, vegetation, and preferential ﬂowpaths at a range of scales. Preferential ﬂow processes have, until recently, been largely ignored in peatlands as scientists have focused on taking measurements of matrix hydraulic conductivity. Many, though not all peatlands, are characterized by rapid saturation, ﬂashy runoff and poor maintenance of base ﬂow. As such, peatlands are often sources of storm ﬂow rather than attenuators of ﬂoods. However, this will depend on the topographic and drainage network context of the peatland. Those peatlands that do supply baseﬂow tend to be those that are receiving water from groundwater inﬂux and it is not the peat itself that results in the baseﬂow maintenance, rather, it is the connected groundwater aquifer that is responsible. Many of the examples used in this chapter were from blanket peat catchments, with which the author is most familiar. There remains much to be done to understand the operation of hydrological processes not only in these peatlands but also to understand the global hydrology of all types of peatlands, including those which we currently know very little about such as tropical peat swamps and Andean glacial valley peats. Peatland decomposition not only increases atmospheric carbon but it destroys the peat mass and with it the archeological and paleoclimatologic archive. With climate change there will be changes to precipitation, vegetation, evapotranspiration, and groundwater levels. There is a need for more work on the global hydrological importance of peatlands and on predictions of how they will react to climate change. We are only just starting to understand the role of soil pipes in peat catchments but we know little about their formation processes or their role in landform development. We also know little about their role in altering the in situ stratigraphic record held by peatlands. Further research on soil pipes in peat is required. The science of ecohydrology in peatlands is also just emerging and there is a need for hydrologists, biogeochemists and ecologists to work together to understand whole-system functioning. With the importance of global carbon cycling in today’s scientiﬁc remit, and the high sensitivity of peatlands to change, it would be prudent to ensure that we continue to push the peatland hydrology research frontier. This is because it is the hydrology that drives carbon storage and ﬂux within peatland systems. References Baird, A.J., 1995. Hydrological investigations of soil water and groundwater processes. In: Hughes, J.M.R. and Heathwaite, A.L. (Eds), Hydrology and Hydrochemistry of British Wetlands. Wiley, Chichester, pp. 111–129.

Peatland hydrology

343

Baird, A.J., 1997a. Field estimation of macropore functioning and surface hydraulic conductivity in a fen peat. Hydrol. Process. 11, 287–295. Baird, A.J., 1997b. Continuity in hydrological systems. In: Wilby, R. (Ed.), Contemporary Hydrology: Towards Holistic Environmental Science. Wiley, Chichester, pp. 25–58. Baird, A.J. and Gaffney, S.W., 1994. Cylindrical piezometer responses in a humiﬁed fen peat. Nord. Hydrol. 25, 167–182. Baird, A.J. and Gaffney, S.W., 2000. Solute movement in a drained fen peat: a tracer study in a Somerset (UK) wetland. Hydrol. Process. 14, 2489–2503. Baird, A.J., Surridge, B.W.J., and Money, R.P., 2004. An assessment of the piezometer method for measuring the hydraulic conductivity of a Cladium mariscus – Phragmites australis root mat in a Norfolk (UK) fen. Hydrol. Process. 18, 275–291. Baird, A.J. and Waldron, S., 2003. Shallow horizontal groundwater ﬂow in peatlands is reduced by bacteriogenic gas production. Geophys. Res. Lett. 30, 2043. Bay, R.R., 1969. Runoff from small peatland watersheds. J. Hydrol. 9, 90–102. Beckwith, C.W. and Baird, A.J., 2001. Effect of biogenic gas bubbles on water ﬂow through poorly decomposed blanket peat. Water Resour. Res. 37, 551–558. Beckwith, C.W., Baird, A.J., and Heathwaite, A.L., 2003. Anisotropy and depth-related heterogeneity of hydraulic conductivity in a bog peat. I: Laboratory measurements. Hydrol. Process. 17, 89–101. Betson, R.P., 1964. What is watershed runoff?. J. Geophys. Res. 69, 1541–1552. Beven, K.J. and Germann, P., 1982. Macropores and water ﬂow in soils. Water Resour. Res. 18, 1311–1325. Boelter, D.H., 1964. Water storage characteristics of several peats in situ. Soil Sci. Soc. Am. Proc. 28, 433–435. Bower, M.M., 1960. Peat erosion in the Pennines. Adv. Sci. 64, 323–331. Bragg, O.M. and Tallis, J.H., 2001. The sensitivity of peat-covered upland landscapes. Catena 42, 345–360. Bromley, J. and Robinson, M., 1995. Groundwater in raised mire systems: models, mounds and myths. In: Hughes, J.M.R. and Heathwaite, A.L. (Eds), Hydrology and Hydrochemistry of British Wetlands. Wiley, Chichester, pp. 95–109. Burt, T.P., 1995. The role of wetlands in runoff generation from headwater catchments. In: Hughes, J.M.R. and Heathwaite, A.L. (Eds), Hydrology and Hydrochemistry of British Wetlands. Wiley, Chichester, pp. 21–38. Burt, T.P., 1996. The hydrology of headwater catchments. In: Petts, G.E. and Calow, P. (Eds), River Flows and Channel Forms. Blackwell, Oxford, pp. 6–31. Campbell, D.I. and Williamson, J.L., 1997. Evaporation from a raised peat bog. J. Hydrol. 193, 142–160. Carey, S.K. and Woo, M-K., 1999. Snowmelt hydrology of two subarctic slopes, southern Yukon, Canada. Nord. Hydrol. 29, 331–346. Chappell, N. and Ternan, L., 1992. Flow path dimensionality and hydrological modeling. Hydrol. Process. 6, 327–345. Charman, D., 2002. Peatlands and Environmental Change. Wiley, Chichester. Chason, D.B. and Seigel, D.I., 1986. Hydraulic conductivity and related physical properties of peat, Lost River Peatland, northern Minnesota. Soil Sci. 142, 91–99. Clymo, R.S., 1991. Peat growth. In: Cushing, E.J. and Shane, L.C. (Eds), Quaternary Landscapes. University of Minnesota Press, Minneapolis, Minnesota, pp. 76–112. Devito, K.J., Hill, A.R., and Roulet, N., 1996. Groundwater-surface water interactions in headwater forested wetlands of the Canadian Shield. J. Hydrol. 181, 127–147. Elgee, F., 1912. The Moorlands of North-Eastern Yorkshire. A Brown, London. Evans, M.G., Burt, T.P., Holden, J., and Adamson, J., 1999. Runoff generation and water table ﬂuctuations in blanket peat: evidence from UK data spanning the dry summer of 1995. J. Hydrol. 221, 141–160. Gilman, K., 1994. Hydrology and Wetland Conservation. Wiley, Chichester. Gilman, K. and Newson, M.D., 1980. Soil Pipes and Pipeﬂow; a Hydrological Study in Upland Wales. British Geomorphological Research Group Research Monograph No. 1, Geo Books, Norwich. Glaser, P.H., 1998. The distribution and origin of mire pools. In: Standen, V. Tallis, J.H. and Meade, R. (Eds.), Patterned Mires and Mire Pools: Origin and Development; Flora and Fauna. Proceedings. University of Durham, British Ecological Society, London, 6–7 April, pp. 4–25.

344

J. Holden

Glaser, P.H., Chanton, J.P., Morin, P., et al., 2004a. Surface deformations as indicators of deep ebullition ﬂuxes in a large northern peatland. Global Biogeochem. Cycles 18, 1003. Glaser, P.H., Hansen, B.C.S., Seigel, D.I., et al., 2004b. Rates, pathways and drivers for peatland development in the Hudson Bay Lowlands, northern Ontario, Canada. J. Ecol. 92, 1036–1053. Glenn, M.S. and Woo, M.-K., 1997. Spring and summer hydrology of a valley-bottom wetland, Ellesmere Island, Northwest Territories, Canada. Wetlands 17, 321–329. Hammond, R.F., Van der Krogt, G., and Osinga, T., 1990. Vegetation and water tables on two raised bog remnants in County Kildare. In: Doyle, G.J. (Ed.), Ecology and Conservation of Irish Peatlands. Roy. Irish Academy, Dublin, pp. 121–134. Heathwaite, A.L., 1995. Overview of the hydrology of British wetlands. In: Hughes, J.M.R. and Heathwaite, A.L. (Eds), Hydrology and Hydrochemistry of British Wetlands. Wiley, Chichester, pp. 11–20. Hewlett, J.D., 1961. Watershed management. Report for 1961 Southeastern Forest Experiment Station, US Forest Service, Ashville, NC, pp. 62–66. Hilbert, D.W., Roulet, N., and Moore, T., 2000. Modelling and analysis of peatlands as dynamical systems. J. Ecol. 88, 230–242. Hillman, G.R., 1998. Flood wave attenuation by a wetland following a beaver dam failure on a second order boreal stream. Wetlands 18, 21–34. Holden, J., 2004. Hydrological connectivity of soil pipes determined by ground penetrating radar tracer detection. Earth Surf. Proc. Land. 29, 437–442. Holden, J., 2005a. Controls on soil pipe density in blanket peat uplands. J. Geophys. Res. 110, F010002 doi: 10.1029/2004JF000143. Holden, J., 2005b. Catchment hydrology. In: Holden, J. (Ed.), An Introduction to Physical Geography and the Environment. Pearson, Harlow, pp. 300–326. Holden, J. and Burt, T.P., 2002a. Inﬁltration, runoff and sediment production in blanket peat catchments: implications of ﬁeld rainfall simulation experiments. Hydrol. Process. 16, 2537–2557. Holden, J. and Burt, T.P., 2002b. Piping and pipeﬂow in a deep peat catchment. Catena 48, 163–199. Holden, J. and Burt, T.P., 2002c. Laboratory experiments on drought and runoff in blanket peat. Eur. J. Soil Sci. 53, 1–15. Holden, J. and Burt, T.P., 2003a. Runoff production in blan.ket peat covered catchments. Water Resour. Res. 39, 1191 doi: 10.1029/2003WR002067. Holden, J. and Burt, T.P., 2003b. Hydraulic conductivity in upland blanket peat; measurement and variability. Hydrol. Process. 17, 1227–1237. Holden, J. and Burt, T.P., 2003c. Hydrological studies on blanket peat: the signiﬁcance of the acrotelmcatotelm model. J. Ecol. 91, 86–102. Holden, J., Burt, T.P., and Cox, N.J., 2001. Macroporosity and inﬁltration in blanket peat: the implications of tension disc inﬁltrometer measurements. Hydrol. Process. 15, 289–303. Holden, J., Burt, T.P., and Vilas, M., 2002. Application of ground penetrating radar to the identiﬁcation of subsurface piping in blanket peat. Earth Surf. Proc. Land. 27, 235–249. Holden, J., Chapman, P.J., Lane, S.N., and Brookes, C. 2006 (this book, Ch. 22). Impacts of artiﬁcial drainage of peatlands on runoff production and water quality. In: Martini, I.P., Matı´ nez Cortizas, A., and Chesworth, W. (Eds), Peatlands: Evolution and Records of Environmental and Climatic Changes. Elsevier, Amsterdam. Holden, N.M., 1998. By-pass of water through laboratory columns of milled peat. Int. Peat J. 8, 13–22. Hutchinson, J.N., 1980. The record of peat wastage in the East Anglian Fenlands at Holme Post, 1848–1978. J. Ecol. 68, 229–249. Ingram, H.A.P., 1967. Problems of hydrology and plant distribution in mires. J. Ecol. 55, 711–724. Ingram, H.A.P., 1978. Soil layers in mires: function and terminology. J. Soil Sci. 29, 224–227. Ingram, H.A.P., 1982. Size and shape in raised mire ecosystems: a geophysical model. Nature 297, 300–303. Ingram, H.A.P., 1983. Hydrology. In: Gore, A.J.P. (Ed.), Mires: Swamp, Bog, Fen and Moor. General Studies. Elsevier, Amsterdam, Vol. A, pp. 67–158. Ingram, H.A.P., 1991. Introduction to the ecohydrology of mires in the context of cultural perturbation. In: Bragg, O.M., Hulme, P.D., Ingram, H.A.P., and Robertson, R.A. (Eds), Peatland Ecosystems and Man: An Impact Assessment. Department of Biological Sciences, Dundee University/International Peat Society, Helsinki, pp. 67–93.

Peatland hydrology

345

Ingram, H.A.P. and Bragg, O.M., 1984. The diptotelmic mire: some hydrological consequences reviewed. Proceedings of the 7th International Peat Congress, Dublin, pp. 220–234. Ivanov, K.E., 1948. Filtration in the top layer of convex mire massifs. Meteorol. Gidrol. 2, 46–59 (In Russian). Ivanov, K.E., 1953. Gidrologiya Bolot. [Hydrology of Mires]. Gidrometeoizdat, Leningrad. Ivanov, K.E., 1981. Water Movement in Mirelands. Academic Press, London. Jones, J.A.A., 1981. The Nature of Soil Piping: A Review of Research. British Geomorphological Research Group Monograph Series 3. Geo Books, Norwich. Jones, J.A.A., 1990. Piping effects in humid lands. In: Higgins, C.G. and Coates, D.R. (Eds.), Groundwater Geomorphology: The Role of Subsurface Water in Earth-Surface Processes and Landforms. Geology Society of America, Special Paper 252, pp. 111–137. Jones, J.A.A., 2004. Implications of natural soil piping for basin management in upland Britain. Land Degrad. Dev. 15, 325–349. Jones, J.A.A. and Crane, F.G., 1984. Pipeﬂow and pipe erosion in the Maesnant experimental catchment. In: Burt, T.P. and Walling, D.E. (Eds), Catchment Experiments in Fluvial Geomorphology. Geobooks, Norwich, pp. 55–72. Jones, J.A.A., Richardson, J.M., and Jacob, H.J., 1997. Factors controlling the distribution of piping in Britain: a reconnaissance. Geomorphology 20, 289–306. Kellner, E., 2001. Surface energy ﬂuxes and control of evapotranspiration from a Swedish Sphagnum mire. Agr. Forest Meteorol. 110, 101–123. Kellner, E., Price, J.S., and Waddington, J.M., 2004. Pressure variations in peat as a result of gas bubble dynamics. Hydrol. Process. 18, 2599–2605. Kirkby, M.J. (Ed.), 1976, Hillslope Hydrology. Wiley, Chichester, . Kirkby, M.J., Kneale, P.E., Lewis, S.L., and Smith, R.T., 1995. Modelling the form and distribution of peat mires. In: Hughes, J.M.R. and Heathwaite, A.L. (Eds), Hydrology and Hydrochemistry of British Wetlands. Wiley, Chichester, pp. 83–93. Lopatin, A.V., 1949. The hydrological signiﬁcance of raised mires. Vest. Leningrad. Gos. Univ. 2, 37–49 (In Russian). McKillop, R., Kouwen, N., and Soulis, E.D., 1999. Modeling the rainfall-runoff response of a headwater wetland. Water Resour. Res. 35, 1165–1177. Ogawa, H. and Male, J.W., 1986. Simulating the ﬂood mitigation role of wetlands. J. Water Res. Plann. Manage. 112, 114–128. Ogunkoya, O.O. and Jenkins, A., 1991. Analysis of runoff pathways and ﬂow contributions using deuterium and stream chemistry. Hydrol. Process. 5, 271–282. Petrone, R.M., Price, J.S., von Waldow, H., and Waddington, J.M., 2004. Effects of a changing surface cover on the moisture and energy exchange of a restored peatland. J. Hydrol. 295, 198–210. Price, J.S., 1992a. Blanket bog in Newfoundland: Part 2. Hydrological processes. J. Hydrol. 135, 103–119. Price, J.S., 1992b. Blanket bog in Newfoundland: Part 1. The occurrence and accumulation of fog water deposits. J. Hydrol. 135, 87–101. Price, J.S., 1996. Hydrology and microclimate of a partly restored cutover bog, Quebec. Hydrol. Process. 10, 1263–1274. Price, J.S. and Waddington, J.M., 2000. Advances in Canadian wetland hydrology and biogeochemistry. Hydrol. Process. 14, 1579–1589. Quinton, W.L. and Marsh, P., 1998. Melt water ﬂuxes, hillslope runoff and streamﬂow in an arctic permafrost basin. Proceedings of the 7th International Conference on Permafrost, 23–27 June, Yellowknife, pp 921–926. Quinton, W.L. and Roulet, N.T., 1998. Spring and summer hydrology of a subarctic patterned wetland. Arctic Alpine Res. 30, 285–294. Romanov, V.V., 1968. Hydrophysics of Bogs. (Translated by N. Kaner). Israel Program of Scientiﬁc Translations, Jerusalem. Roulet, N., 1988. Groundwater ﬂux in a headwater wetland in southern Ontario. In: Bardecki, M.J. and Patterson, N. (Eds), Ontario Wetlands: Inertia or Momentum. Proceedings of a conference. Federation of Ontario Naturalists, Toronto, pp. 301–308. Roulet, N., 1990. The hydrological role of peat-covered wetlands. Can. Geogr. 34, 82–83.

346

J. Holden

Rycroft, D.W., Williams, D.J.A., and Ingram, H.A.P., 1975. The transmission of water through peat: II ﬁeld experiments. J. Ecol. 63, 557–568. Terajima, T., Sakamoto, T., and Shirai, T., 2000. Morphology, structure and ﬂow phases in soil pipes developing in forested hillslopes underlain by a Quaternary sand-gravel formation, Hokkaido, northern main island in Japan. Hydrol. Process. 14, 713–726. Turner, N., 1757. An Essay on Draining and Improving Peat Bogs; in which their Nature and Properties are Fully Considered. Baldwin and Pew, London. van der Schaaf, S., 2002a. Bog hydrology. In: Schouten, M.G.C. (Ed.), Conservation and Restoration of Raised Bogs: Geological, Hydrological and Ecological Studies. Duchas. Department of the Environment and Local Government, Dublin, pp. 54–109. van der Schaaf, S., 2002b. Analysis of the Hydrology of Raised Bogs in the Irish Midlands: A Case Study of Raheenmore Bog and Clara Bog. Wageningen Agricultural University, The Netherlands, 375pp. Von Post, L., 1922. Sveriges Geologiska Undersoknings torvinventering och nogra av dess hittils vunna resultat (SGU peat inventory and some preliminary results). Svenska Mosskulturforeningens Tidskrift, Jonkoping, Sweden 36, 1–37. Ward, R.C. and Robinson, M., 2000. Principles of Hydrology. McGraw-Hill, London. Weber, C.A., 1902. Uber die vegetation und Entstehung des Hochmoors von Augstumal im Memeldelta mit vergleichenden Aus blacken aut andere Hochmoore der erde. Paul Parey, Berlin. Woo M-K. and Young, K., 1998. Characteristics of patchy wetlands in a polar desert environment, Arctic Canada. Proceedings of the 7th International Conference on Permafrost, 23–27 June, Yellowknife, pp. 1141–1146.

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

347

Chapter 15

Hydrogeology of major peat basins in North America P.H. Glaser, D.I. Siegel, A.S. Reeve and J.P. Chanton

Introduction Most of the world’s peatlands are concentrated in boreal and subarctic regions where the cool moist climate and glaciated landscape maintain waterlogged soils across extensive areas (Gore, 1983a, b). The formation of unusual surface patterns in these peatlands has inspired intense interest for over a century (Weber, 1902; Auer, 1920; Sjo¨rs, 1963; Washburn, 1979; Gore, 1983a, b). However, current scientiﬁc interest is focused on the role of northern peatlands as a globally important source or sink for greenhouse gases (Kivinen and Pakarinen, 1981; Gorham, 1991). Because peatlands are concentrated in regions that have signiﬁcantly warmed over the past few decades, further warming is likely to produce an important positive or negative feedback to the global climate system (Moore et al., 1998; Khalil, 2000; Wigley and Schimel, 2000). Predicting the response of these peatlands to future climatic change, however, is problematic because of the complex interplay among climate, groundwater, and ecosystem processes across heterogeneous geologic landscapes. Amongst of the most favorable places to elucidate these complex interactions are the large peat basins of boreal America (Fig. 15.1a). These large peat basins developed across regions of low relief where a cool, moist climate, poor drainage, and the absence of topographic obstructions permitted peatlands to form an interlocking web across 60–90% of the regional landscape. This extensive web of waterlogged soils restricts the spread of wildﬁres (Glaser, 1987; Johnson, 1992), while greatly limiting human access and impacts. The low frequency of disturbance in these large peat basins permitted highly developed surface patterns to form that resemble geomorphic landforms on scales from less than 100 m2 to over 200 km2. These peat landforms are visible on remote sensing imagery and provide important indicators for the vegetation assemblages, topography, ranges in water chemistry, and surface hydrology (Sjo¨rs, 1948, 1963, 1983; Heinselman, 1963, 1970; Glaser et al., 1981, 2004c, b; Glaser and Janssens, 1986; Glaser, 1992 a, b, c; Vitt et al., 1994). Recent investigations in the glacial Lake Agassiz region and Hudson Bay Lowland have shown that these peat landforms are also sensitive indicators for groundwater ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09015-8

348

P.H. Glaser et al.

Figure 15.1. Topographic setting of large peat basins in boreal North America. (a) Location of major peat basins in North America (Shuttle radar image composite; visualization by Paul Morin). (b) Topographic setting of the Albany River (1) and Glacial Lake Agassiz peatland (2) study areas with respect to major physiographic features including the valleys formed by the Des Moines (3) and James (4) lobes of the Laurentide Ice Sheet (Visualization based on SRTM digital elevation data; after Morin and Thorleifson, 2004).

Hydrogeology of major peat basins in North America

349

ﬂow systems and geologic features in the underlying glacial deposits and bedrock (Glaser, 1987, 1989; Glaser et al., 1990, 1997, 2004c, b; Siegel, 1992). In this chapter, we review over 25 years of investigation that link groundwater and peatland interactions in two of the largest peat basins of North America: the glacial Lake Agassiz peatlands (GLAP) in northern Minnesota, USA and the Albany River (ALBR) peatlands in the Hudson Bay Lowland of northern Ontario, Canada (Fig. 15.1b).

Central problems The large peat basins of North America and Eurasia deﬁed comprehensive study until recently because of their great size, remote locations, and lack of access. Nevertheless, the pioneering work of Kulczynski (1949) in the Polesie of eastern Europe, Sjo¨rs (1963) in the Hudson Bay Lowland of northern Canada, and Heinselman (1963) in the glacial Lake Agassiz region of northern Minnesota provided a different perspective on the genesis, evolution, and carbon balance of peatlands. The spread of peatlands across a broad lowland implies a regional rise in the water table driven by either physical, chemical, or biotic processes. All three of these pioneering studies indicated that paludiﬁcation (spread of peatlands across poorly drained lands) was the dominant process of peatland genesis in large peat basins and that terrestrialization (lake inﬁlling) was only locally important. However, the principal drivers for peatland genesis were not well established. Kulczynski (1949) hypothesized that the growth of peatlands in valleys at the lower reaches of watersheds impeded drainage upslope leading to a regional rise in the water table. Sjo¨rs (1963) suggested that beaver dams have a similar effect in initiating peat formation in the Hudson Bay Lowland. Heinselman (1963), however, suggested that the spread of peatlands across the bed of glacial Lake Agassiz in northern Minnesota was initiated by a change to a cooler and moister regional climate after 5000 14C yr BP. As peat accumulates the absorptive organs of green plants become separated from direct contact with mineral soil. As a result the transport of solutes in ﬂowing water largely determines the supply of (1) inorganic bases that neutralize the organic acids released from decaying vegetation and (2) mineral nutrients such as calcium that exert an important control on plant distributions. Rates of ﬂow in porous media are dependent on the size, geometry, and interconnectivity of the pores, but a more general measure of permeability is provided by hydraulic conductivity (K), which can be measured in the ﬁeld or laboratory (Freeze and Cherry, 1979). A wide range of methods were ﬁrst used by peatland scientists to show that hydraulic conductivity decreases exponentially with increasing bulk density and the degree of decomposition of different peat types (Baden and Eggelsmann, 1963; Boelter, 1969; Rycroft, Williams, and Ingram, 1975; Boelter and Verry, 1977). On the basis of these studies Ivanov (1975) and Ingram (1983) developed a two-layered model of peat deposits that is widely used in biogeochemical and hydrological studies. This imprecise but useful model proposed that ﬂuxes of water, gas, and solutes are largely conﬁned to an upper-aerobic layer, called the acrotelm in which the relatively undecomposed peat has a high-hydraulic conductivity (Ingram, 1978, 1982, 1983). In contrast ﬂuxes are negligible in the deeper anaerobic peat, called the catotelm because of its greater

350

P.H. Glaser et al.

degree of decomposition and much lower values for hydraulic conductivity (Holden, 2006 – this book, Ch. 14). According to this model the surface waters in large peatlands should become progressively diluted by precipitation as runoff drains from nearby mineral uplands into the interior of a peat basin. As surface waters are acidiﬁed by the release of organic acids from decaying vegetation (Hemond, 1980, 1990; Gorham et al., 1985; Reeve et al., 1996; Glaser et al., 2004b; Siegel et al., in press) the peat basin should become covered by large raised bogs, which receive all their water and salts solely from precipitation. However, the regional studies of Kulczynski (1949), Sjo¨rs (1963), and Heinselman (1963, 1970) showed that large peat basins are usually dominated by fens, whereas raised bogs are restricted to speciﬁc physiographic settings, such as watershed divides between rivers. Sjo¨rs (1963) and Heinselman (1970) further noted that fen water tracks containing minerotrophic vegetation arise within the interior of large raised bogs where there are no outcrops of mineral soil. The source for the alkalinity and inorganic solutes needed to maintain fens in the interior of bogs larger than 20 km2 could only come from the underlying mineral soils leading Sjo¨rs (1963) and Boelter and Verry (1977) to speculate that groundwater upwelled through small ‘windows’ in the otherwise impermeable peat cover. Sjo¨rs (1963) also proposed that base cations released from decomposing vegetation may supply the alkalinity necessary to raise the pH of bogs waters from 4.2 to 5, which is typical for fens. Computer simulations of groundwater ﬂow in the GLAP of northern Minnesota suggest an alternative explanation (Siegel, 1983). He proposed that groundwater mounds under raised bogs drive local-recharge cells that ﬂow downward through the underlying glacial deposits to the underlying bedrock, but then curve upward to discharge into a water track at the margin of the bog. The Siegel model assumed that the vertical hydraulic conductivity of the glacial deposits were similar to that of the peat, and that, despite the slow rates of ﬂow, even a small input of groundwater to the peat surface was sufﬁcient to transform bog waters with a pH of 4 to fen waters with a pH of 6. These assumptions were soon tested by some of the ﬁrst groundwater investigations in peatlands in North America (Chason and Siegel, 1986; Wilcox et al., 1986; Siegel and Glaser, 1987). Groundwater investigations in large peat basins were initially focused on the GLAP in northwestern Minnesota and the ALBR peatlands in the Hudson Bay Lowland. Study areas The ALBR study area is located along the lower ALBR drainage in northern Ontario from Albany Forks to Blackbear Island (Figs. 15.2a, b, c; Glaser et al., 2004c, b). Peatlands cover 90% of the regional landscape and include a wide range of raised Figure 15.2. Albany River (ALBR) study area in the Hudson Bay Lowlands. (a) Peatland vegetation types and physiographic units. The solid lines delineate abandoned river channels, whereas the dashed lines indicate the boundary between the low-lying plain and mounded plateau; boxes indicate location of detailed studies (Glaser et al., 2004a). The vegetation types are deﬁned in Table 1. (b) Drainage system. (c) Visualization of the SRTM elevational data for the study area (courtesy of Paul Morin). The white arrows mark the path of the Albany River, whereas the black arrows show abandoned drainage channels. (Note that the geographic orientation (north) of the image is reversed from that in Fig. 15.2a.)

Hydrogeology of major peat basins in North America

351

352

P.H. Glaser et al.

bogs, fen water tracks, and spring-fen channels. The ﬁrst ﬁeld investigations in this study area utilized aerial photographs and Landsat MSS imagery, and relied on lowresolution topographic maps for elevation control (Coombs, 1954; Dean, 1959; Pala and Weischet, 1982). More recently, higher-resolution Landsat TM imagery and Shuttle Topography Radar Mission (SRTM) digital elevation data have revealed the occurrence of two distinct landform units in this study area: a low-lying plain underlain by clayey silt (cf. till plain of Glaser et al., 2004c) and a mounded plateau that is underlain by till, sand, and gravel (cf. moraine of Glaser et al., 2004c). The plain is dissected by the ALBR, which ﬂows through a gorge 30 m deep and a system of parallel rivers some of which ﬂow within a series of mega-scale lineations (sensu Clark, 1993). The plateau has a high number of kettle lakes, a series of linear mineral outcrops on its northern ﬂank (Glaser et al., 2004c), and till depths of 45 m near its summit (Hogg et al., 1953). The study area is located near the center for postglacial rebound in North America and land surfaces have risen by 60–120 m over the past 6000 years. Glaser et al. (2004c,b) suggest that isostatic uplift over the past 6000 years has altered drainage patterns and perturbed groundwater ﬂow systems in this area providing a natural experiment for analyzing the response of peatlands to groundwater ﬂow systems. The GLAP study area in northeastern Minnesota occupies the Beltrami Arm of glacial Lake Agassiz, which is bordered to the south by moraines and till plains and to the north and northeast by bedrock of the Precambrian Canadian Shield (Wright, 1972). The lacustrine plain has gentle relief that is locally interrupted by sand and gravel beach ridges of glacial Lake Agassiz. The plain is underlain by 30–50 m of calcareous glacial deposits composed of silt, clay, sand, and gravel, which cover metamorphic bedrock. Peatlands cover 56% of the regional landscape and include large raised bogs over 160 km2 in area and patterned water tracks up to 150 km2 in area and a few smaller spring-fen systems (Figs. 15.3a, b; Glaser, 1992a, b, c). The regional climate seems too dry to support large areas of peatland since there is only a slight annual moisture surplus and frequent droughts (Baker et al., 1967, 1979; Glaser et al., 1997). The groundwater ﬂow systems in this area are therefore sensitively adjusted to the climate and are easily perturbed by the frequent droughts that can last from weeks to several years (Table 15.1).

Properties of large peat basins in North America Physiographic setting On a continental scale, topographic factors largely determine the aerial extent of peatlands within the moist boreal belt of North America. In the Labrador/Ungava region of eastern Canada, for example, the rugged topography of the Precambrian Canadian Shield restricts peatlands to small topographic depressions, and prevents peat-forming centers from spreading outward to cover large areas of the regional landscape. In central and western Canada, however, glacial processes altered the topography of the Precambrian Canadian Shield creating three extensive lowlands that currently have a high regional peat cover (Zoltai and Pollet, 1983). The Hudson

Hydrogeology of major peat basins in North America

353

Figure 15.3. Glacial Lake Agassiz peatlands (GLAP) study area in northwestern Minnesota. (a) Peatland types (after Glaser, 1992a; Glaser et al., 1997). (b) Sparse drainage system (contrast it with the one of the Albany River area, Fig. 15.2b). The sparse drainage pattern reduces water losses from runoff and groundwater discharge and is probably a key factor permitting extensive peatlands to grow in this dry region.

354

Table 15.1.

Properties of the vegetation units used in Fig. 15.2a.

Peatland type

pH

Ca

Dominants

Landform

Hydrology

Raised bogs Sphagnum lawn

3.7–3.9

0.4–1.6

Carex oligosperma Chamaedaphne calyculata Sphagnum magellanicum S. capillifoilum

Nonforested bog Semi-forested bog

Recharge

Bog/hummock/forest

3.7–4

0.2–2

Picea mariana Kalmia angustifolia Sphagnum capillifoilum Sphagnum fuscum

Forested bog Semi-forested bog Nonforested bog

Recharge

Sphagnum lawn/poor fen

2.0–4.0

2.0–3.4

Carex oligosperma Chamaedaphne calyculata Sphagnum magellanicum S. capillifoilum

Nonforested bog Semi-forested bog Internal water track

Lateral

Patterned fens

5.7–6.9

2.9–10.8

Carex lasiocarpa Rhychospora alba Scorpidium scorpiodes Sphagnum subsecundum

Water track

Lateral

Spring-fen channels

5.9–6.4

12.9–19.5

Scirpus cespitosus Carex lasiocarpa Scorpidium scorpiodes Sphagnum warnstorfii

Spring-fen channel

Discharge

Swamp forests

5.4–5.9

9–15.7

Larix laricina Carex chrodorrhiza Tomenhypnum nitens Pleurozium schreberi

Water tracks

Lateral

Fens

P.H. Glaser et al.

Hydrogeology of major peat basins in North America

355

Bay Lowland, Lake Agassiz region, and Lake McConnell lowlands are clearly deﬁned by digital elevation models (DEM) based on the new SRTM elevation data (Fig. 15.1a). A subset of these data was expanded to show the Lake Agassiz and Hudson Bay Lowland peat basins in more detail (Fig. 15.1b; Morin and Thorliefson, 2005). The evolution of these peat basins was largely shaped by the dynamics of the Laurentide Ice Sheet from 18,000 to 7800 14C yr BP. The ice sheet had several ice domes that shifted in location during the periods of ice advance and retreat. By 10,000–8000 14C yr BP, a prominent ice dome developed in southeastern Hudson Bay that was connected by ice divides to other ice domes to the east and northwest (Dyke and Prest, 1987; Dredge and Cowan, 1989; Dyke et al., 1989). Glacial ice moving outward from the divides and ice domes planed the landscape of the lowland by eroding the soft carbonate bedrock and transporting till over the Precambrian Canadian Shield to the south, east, and west. The Shield was also modiﬁed by ‘ice streams’ feeding the James and Des Moines glacier lobes, which created prominent tongue-shaped valleys farther to the south in Minnesota (Fig. 15.1b). The landscape was also altered by the formation of extensive proglacial lakes and a major marine transgression during and immediately after deglaciation. Large proglacial lakes formed along the southern and western margin of the Laurentide Ice Sheet after 13,000 14C yr BP as glacial meltwater was impounded by the retreating ice mass, which blocked the drainage to the north (Teller, 1985; Dyke et al., 1989). The largest of these proglacial lakes was Lake Agassiz to the south, which submerged the valleys formed by the James and Des Moines glacial lobes (Fig. 15.1b), and spilled over to submerged large areas from northwestern Minnesota to western Ontario and southern Saskatchewan (Teller and Clayton, 1983; Teller, 1985; Tarnocai and Stolbovoy, 2006 – this book, Ch. 2). To the west, Lake McConnell expanded over the lowland now occupied by Great Slave and Great Bear lakes. The sea re-entered the Hudson/James Bay area after deglaciation of the Hudson Strait about 7800 14C yr BP, and the Hudson Bay Lowland was ﬂooded by the large glacial Tyrrell Sea (Dyke et al., 1989). The Hudson Bay Lowland area was at the time depressed by the loading of the Laurentide Ice Sheet (Andrews and Peltier, 1989).

Genesis of the peat basins Peatlands spread across these three basins at different times in response to different environmental drivers. In the Hudson Bay Lowland stratigraphic data from the ALBR region indicates that the extensive peat cover developed upon emergence of new land from the early postglacial Tyrrell Sea (Glaser et al., 2004c, b). Peat cores from the ALBR region (Glaser et al., 2004c) have a sharp mineral contact with the basal glacial and/or glaciomarine sediments, which lack evidence for signiﬁcant exposure to an oxidizing environment. The basal mineral sediments in the cores lack weathering horizons typical of upland soils, contain minerals that require anoxia for their preservation, and contain silt grains devoid of chemical etch pits (Reeve et al., 1996; Glaser et al., 2004b). In addition, the deeper pore waters under these inland peatland sites still retain a marine chemical signature with high-chloride concentrations and at some sites a 1:1 sodium to chloride ratio (Reeve et al., 1996). The

356

P.H. Glaser et al.

basal-organic stratum, moreover, contains pollen and plant macrofossil assemblages indicative of the modern salt or brackish marshes that characterize the modern coast of Hudson and James bays. On the basis of these peat cores Glaser et al. (2004b) proposed that the differential pattern of isostatic rebound in the Hudson Bay Lowland, relative to that of adjacent areas of the Precambrian Canadian Shield, was one of the primary environmental drivers for peat formation in this large peat basin. The landscape of the lowland has been rising isostatically since early deglaciation at an average rate comparable to that of Fennoscandia (Andrews and Peltier, 1989). Maximum isobars of uplift are in southeastern Hudson Bay and decrease outward in quasi-concentric rings across the Hudson Bay Lowland and adjacent parts of the Precambrian Canadian Shield. Rates of uplift from 4000 to 1000 14C yr BP show a similar pattern (Andrews and Peltier, 1989) and are still quite high in the lowland equaling about 1 m/century at the mouth of the ALBR (Hunter, 1970; Webber et al., 1970). This pattern of uplift creates an unusual geomorphic setting in which the major rivers arising on the Precambrian Canadian Shield ﬂow toward regions that are closer to the center for isostatic rebound (Starkel, 1979; Bloom, 1998). Glaser et al. (2004c, b) therefore proposed that the greater differential uplift of the lower reaches of these rivers during postglacial times decreased the regional gradient, impeded drainage, and drove a rise in the regional water table that allowed peatlands to spread over the landscape. This view contrasts with some previous inferences that climate (Halsey et al., 2000) or the activity of beaver (Sjo¨rs, 1963) was major determinant of peat formation in the Hudson Bay Lowland and other large peat basins of North America. Farther south, the growth of peatlands across the Lake Agassiz region was delayed by unfavorable climatic conditions during the early and mid-Holocene (Heinselman, 1963, 1970). Glacial Lake Agassiz retreated from northeastern Minnesota around 11,000 14C yr BP (Teller, 1985), and the exposed lakebed was then colonized by vegetation indicative of modern prairie (Janssen, 1968). Pollen diagrams from lakes distributed across eastern North America show that prairie vegetation spread eastward from western Minnesota to Illinois from 10,000 to 5000 14C yr BP in response to the warm dry climate of the early and mid Holocene (Jacobson et al., 1987). After 5000 14C yr BP the prairie–forest border shifted back toward the west and large peatlands began spreading across the Lake Agassiz region as the climate became cooler and moister. The general warming trend of the early to mid Holocene was driven by increased solar radiation and a strengthening of westerly ﬂow across the Midwest (COHMAP, 1988). In contrast the cooling trend after 6000 14C yr BP was produced by a reduction in summertime solar radiation, a weakening of westerly ﬂow, and a strengthening of southerly monsoonal ﬂow from the Gulf of Mexico. The climatic control on peat formation is particularly apparent in the Lake Agassiz region of northwestern Minnesota and southeastern Manitoba where large peatlands extend to the prairie/forest border as mapped prior to the European settlement. Although climatic change was primarily responsible for the regional rise in the water table, the effect of higher recharge rates was ampliﬁed by the low-regional relief and a sparse drainage system that evolved under the previous dry climate of the early to mid Holocene (Fig. 15.3b; Wright, 1972; Glaser et al., 1997). This factor may account for the unexpectedly high rate of peat accumulation in these peatlands,

Hydrogeology of major peat basins in North America

357

reported by Glaser and Janssens (1986) and Gorham et al. (2003), which exceed that found in moister climatic regions to the north and east. Peat landform patterns The large peat basins of boreal America have four principal types of surface patterns that may be distinguished on the basis of their landform morphology, surface-water chemistry (principally pH and Ca), species assemblages, and hydrology. Because these surface patterns are of sufﬁcient size to be readily visible on remote sensing imagery they have been termed peat landforms or vegetation-landform patterns (Glaser et al., 1981). The following treatment is largely based on peat landforms that have not been modiﬁed by permafrost that complicates the dynamics of peatland hydrology and development. This treatment follows the terminology developed by Sjo¨rs (1948, 1963), Heinselman (1963, 1970), and Glaser and others (Glaser et al., 1981, 1997, 2004c, b; Glaser and Janssens, 1986; Glaser, 1992a, b, c) rather than the slightly different terminology developed by Canadian scientists (National Wetlands Working Group, 1986). A fuller account of peat landforms associated with permafrost can be found in Zoltai (1972), Washburn (1979), National Wetlands Working Group (1986), and Vitt et al. (1994). The four main types of peat landforms in the GLAP and ALBR study areas are the following: (1) Raised bog landforms. The interior of a raised bog is always higher than its margins causing runoff to drain outward from a central crest, mound, or plateau to the bog margin. Bogs are generally dominated by species of Sphagnum, and a small ﬂora of acidophilus species. Bogs may be forested or non-forested, but never have fen-indicator species such as Carex lasiocarpa, C. livida, C. chordorrhiza, and Betula pumila var. glandulifera (sensu Sjo¨rs, 1963; Heinselman, 1963, Glaser et al., 1981; 2004c; Jeglum and Cowell, 1982.). The surface waters consistently have a pH lower than 4.2 and calcium concentrations less than 2 mg L 1. Raised bogs are said to be ombrotrophic because they are fed solely from precipitation. (2) Fen water track landforms. These landforms have a ﬂat or slightly concave topography that channel runoff across an expanse of peatland. Water tracks may have streamlined vegetation stands parallel to the slope, networks of sinuous pools and peat ridges oriented perpendicular to the slope or ﬁelds of streamlined peat islands (which may have the properties of either bog or fen landforms). Water tracks may be forested (cf. featureless water tracks or swamp forests; Glaser, (1992a, b) semi-frosted, or non-forested. Fen landforms are marked by various fen-indicator species such as C. lasiocarpa, C. livida, C. chordorrhiza, and B. pumila var. glandulifera, which increase in number and relative abundance with an increase in the pH (4.2–47) and calcium (42 mg L 1) of the surface waters or variation in micro-topography. Fens are said to be minerotrophic because they derive at least part of their water and inorganic solutes from groundwater or surface waters that have drained from mineral uplands.

358

P.H. Glaser et al.

(3) Spring-fen channels. These landforms are slightly concave, narrow, and nonforested channels that often form an anastomosing web or a series of parallel rows. They are generally smaller in area and narrower than most fen water tracks and are further distinguished by having an assemblage of extremely rich fen-indicator species such as Tofieldia glutinosa, Scirpus hudsonianus, Parnassia paulstris, and Thuja occidentalis (sensu Sjo¨rs, 1963; Glaser, 1992b) and surface waters with a pH higher than 6.8 and calcium concentrations higher than 20 mg L 1. (4) Bog complexes. In large peat basins, raised bogs may spread outward to cover areas larger than 100 km2. These large bogs generally have sharp streamlined margins trimmed by fen-water tracks and also tend to be fragmented into streamlined lobes or ovoid-shaped islands by smaller fen-water tracks that arise within the center of the bog massif. Regional surveys indicate that any bog larger than 20 km2 has these internal strips of fen vegetation, which indicate some common developmental mechanism that is intimately related to the hydrology (Glaser et al., 1981, 2004c; Glaser, 1987).

Hydrological models for large peat basins The close relationship between peatland ecosystems and hydrological processes was ﬁrst established by studies of small peatlands and blanket bogs in northern Europe. On the basis of general observations, early workers such as King (1685), Rennie (1810), and Dau (1823) inferred that waterlogged soils retard the decomposition of plant litter, which then accumulates to form peat. They also observed that changing water levels control the characteristic plant assemblages on hummocks, hollows, and pools found on many peatlands. However, their most important insight was that the source of water that ﬂows through a peatland determines its character. The isolation of a peatland from waters that have percolated through mineral soil, for example, was considered the primary determinant for the dominance of Sphagnum and the impoverished acidophilus ﬂora on raised bogs. A series of studies in northern Europe used chemical tracers, such as dissolved calcium, to infer the source of the surface waters on peatlands. Weber (1902), for example, concluded that the very low lime (CaO) content of surface waters from the central plateau of a raised bog in eastern Prussia could only be derived from precipitation. However, he found much higher lime concentrations in some streams and ‘rullen’ (water tracks) arising from the bog margins that he inferred were derived from groundwater upwelling from a buried sand lens under the bog plateau (Weber, 1902). Du Rietz (1949) and Sjo¨rs (1948, 1950, 1963) later used the pH and calcium concentrations of peat waters to distinguish ombrotrophic (rain nourished) from the various classes of minerotrophic (mineral–soil nourished) vegetation assemblages. Although the system developed by Sjo¨rs and Du Rietz had broad applications across the boreal belt of North America and Eurasia, direct measurements of groundwater ﬂow in peat deposits based on hydraulic head and pore-water chemistry were not reported until the 1980s (Wilcox et al., 1986; Siegel and Glaser, 1987). The impetus for some of the ﬁrst groundwater investigations in large peat basins was the occurrence of vegetation patterns that could not be readily explained by the

Hydrogeology of major peat basins in North America

359

local surface hydrology. First, an apparent inconsistency was discovered from the surveys of large peat basins in Eastern Europe (Kulczynski, 1949) and North America (Heinselman, 1963; Sjo¨rs, 1963), which determined that these large areas of peatland were dominated by fen vegetation despite the declining inﬂuence of runoff draining from mineral uplands. A second problem was the appearance of fen waters tracks in the middle of large raised bogs where there were no surface exposures of mineral soil. Sjo¨rs (1963) speculated that alkalinity could be produced by enhanced decomposition of bog peat, but subsequent experimental and modeling work showed that decomposing Sphagnum releases strong organic acids that increase the acidity of the surrounding water (Hemond, 1980; Gorham et al., 1985). Finally, a third problem was the development of bogs, fens, and spring fens in close proximity to each other on peat-covered landscapes nearly devoid of mineral exposures or other surface sources for alkalinity. The ﬁrst detailed groundwater investigation in a large peat basin focused on a raised bog and adjacent spring-fen mound in the Lost River portion of the GLAP (Chason and Siegel, 1986; Siegel and Glaser, 1987; Glaser et al., 1990). The objective was to determine if the development of these contrasting peatland types was linked to groundwater ﬂow systems. Since 1980s there has also been an expanding number of groundwater investigations of small peatlands, which have established the general occurrence of vertical or horizontal ﬂow systems in the deeper peat despite low values for vertical (KV) and horizontal (Kh) hydraulic conductivity (Hemond and Goldman, 1985; Wilcox et al., 1986; Roulet, 1990; McNamara et al., 1992; Price and Maloney, 1994; Devito et al., 1997; Waddington and Roulet, 1997; Fraser et al., 2001). These studies have also shown the signiﬁcance of ﬂow systems for transporting dissolved solutes through a peat deposit. In this chapter, however, we will focus on our work in two large peat basins where an unusual environmental setting has helped to elucidate the major drivers for groundwater ﬂow systems and their impact on peatland ecosystems.

Hydrogeologic investigations in large North American peat basins Peatland hydrology in an isostatically rising landscape Three different groundwater models have been proposed to explain groundwater/ peatland interactions in the southern Hudson Bay Lowland. Roulet and McKenzie (1998) showed that a two-dimensional ﬁnite-difference model based on MODFLOW (McDonald and Harbaugh, 1984; Harbaugh and McDonald, 1996) could explain the major vegetation zonation of the lowland. Their model predicts that precipitation recharges the regional groundwater system under a zone of raised bog complexes in the interior of the lowland. Groundwater then ﬂows laterally toward the coast within calcareous glacial deposits and discharges to the land surface within a coastal zone of fens. The higher pH and inorganic solute concentrations in the surface waters of these fens is apparently supplied by groundwater transport according to this model. Reeve et al. (2000, 2001a, b) assumed that lateral groundwater ﬂow systems tend to develop in large gently sloping peat basins where the peat is underlain by mineral sediments with a low-hydraulic conductivity. According to this model, lateral ﬂow

360

P.H. Glaser et al.

systems should develop across low-lying plains in the Hudson Bay Lowland where deposits of glaciomarine silts and clays are thickest and most extensive. Their coupled groundwater ﬂow-transport model showed that inorganic solutes from the underlying calcareous glaciomarine sediments would be transported upward to the surface by transverse dispersion and dispersive mixing along long lateral ﬂow paths (Reeve et al., 2001a). This model would suggest that the Hudson Bay Lowland should be covered with fen vegetation unless some counteracting surface process (release of strong organic acids from decaying Sphagnum peat or the development of local ﬂow systems) permitted the spread of bogs. Vertical ﬂow systems (such as recharge or discharge cells) would be expected to develop where the peatlands grow over more permeable deposits (Reeve et al., 2000) or in wide-interﬂuvial areas (Freeze and Cherry, 1979; Fetter, 2000). A vegetation survey based on peat landform patterns, however, showed that a 24,000 km2 study area along the ALBR consisted of 55% fen, 35% bog, and 10% water or mineral soil (Fig. 15.2a; Glaser et al., 2004c). The general working hypothesis of this study assumed that the local hydrogeologic setting determines the distribution of raised bogs, patterned fens, and spring fens across the regional landscape. Raised bogs, for example, should be restricted to hydrogeologic settings that (1) promote the development of local recharge mounds that isolate the surface waters from groundwater in the mineral substratum by driving ﬂow downward, and (2) determine the capacity for these water-table mounds to migrate upward and saturate new layers of peat. Both of these processes favor the growth of bog-forming species of Sphagnum that can acidify a site through the release of organic acids with strongly acidic functional groups (Hemond, 1980, 1990; Reeve et al., 1996; Glaser et al., 2004c; Siegel et al., in press). In contrast, spring fens should be restricted to sites where groundwater discharges and provides a continuous supply of circum-neutral waters at the land surface, whereas fen water tracks will develop farther downslope from these discharge sites where there is still sufﬁcient alkalinity to prevent the spread of bog-forming Sphagnum. Alternatively, water tracks could be fed by the upward transport of inorganic solutes from the mineral substratum by transverse dispersion or dispersive mixing along longer lateral-ﬂow paths (Reeve et al., 2001a). According to these hypotheses (1) raised bogs should develop over recharge zones at the top of mounded plateau or within the interﬂuvial divides formed by the parallel rivers, abandoned river channels, and mega-scale lineations (sensu Clark, 1993), (2) spring fens will form where local ﬂow systems discharge along the margins of this mounded plateau and the ﬂanks of the deeper furrows, and (3) fen water tracks will form on the broad surfaces of the low-lying plain where inorganic solutes are supplied to the peat surface through transverse dispersion driven by variations in lateral ﬂow. Field studies largely support these hypotheses (Glaser et al., 2004c). The mounded plateau is covered by small raised bogs that have surface waters with low pH (44.0), low calcium (42 mg L 1), and bog vegetation with no fen indicator species. Hydraulic head decreased with depth on the two bogs sampled in 1992 indicating that they were local recharge zones for groundwater (Reeve, 1996; Glaser et al., 2004c). On the northern ﬂanks of the mounded plateau, however, small raised bogs are dissected by rows of spring-fen channels, which have the chemical and biotic properties of an

Hydrogeology of major peat basins in North America

361

extremely rich fen (pH46.8; Ca concentrations420 mg L 1). These spring fens developed over sand and gravel deposits, and at these sites hydraulic head increased with depth indicating that the northern edge of this plateau is a regional discharge zone for ﬂow systems that originate under the summit of the mounded plateau. Rows of springfen channels also line the ﬂanks of the mega-scale lineations on the plateau itself indicating that these sites are also discharge zones for local ﬂow systems producing distinct spectral signatures on Landsat TM imagery (Glaser, 1989; Glaser et al., 2004a). By contrast, the low-lying plain is covered by fen water tracks or large bog complexes depending on the local hydrogeologic setting. The largest fen water tracks, for example, occur downslope from seepage faces that arise from the ﬂanks of the mounded plateau or at the extreme downgradient portions of local watersheds. These water tracks have surface waters typical of a poor to rich fen (pH: 4.5–6.4; Ca: 3–10 mg L 1) and were usually located in zones of lateral ﬂow indicated by a negligible change in hydraulic head with depth. Longer lateral ﬂow paths, however, are locally interrupted on the low-lying plain by the dense system of parallel ﬂowing rivers. These rivers act as discharge zones for local recharge mounds that develop under the interﬂuvial divides (Freeze and Cherry, 1979). Raised bogs consistently develop in this hydrogeologic setting and the initial stages of this process are apparent where tributary streams or rills are eroding headwards into broad areas of fen with tamarack trees (Fig. 15.4). Glaser et al. (2004b) used an

Figure 15.4. Air photograph showing the effect of ﬂuvial erosion on raised bog formation. Raised bogs are located in the interﬂuves between rivers and smaller tributary streams that are eroding headwards into a featureless fen with tamarack trees. The image covers an area 7 km across.

362

P.H. Glaser et al.

Figure 15.5. Dupuit simulations indicating the effect of river incision on the maximum height of the water-table mound and peat accumulation in an interﬂuvial divide. Notice the shift in the location of the water divide upgradient due to deeper incision of the stream on the left (after Glaser et al., 2004b). The values used for recharge and hydraulic conductivity in the model are provided.

analytical model based on the widely used Dupuit assumption (Dupuit, 1863; Freeze and Cherry, 1979; Fetter, 2000) to show that the width of an interﬂuvial divide imposes a geologic control on how high a raised bog can grow within this hydrogeologic setting. These simple simulations also show how river incision could arrest the growth of a raised bog and alter the geometry of its water-table mound (Fig. 15.5). This analysis indicates that the development of a raised bog is determined by the growth and dynamics of a recharge mound, which is also constrained by the local climate and geologic setting. As rivers incise into their beds, the location of the groundwater divide between two rivers will shift closer to the river with the higher water level (Fig. 15.4). However, in the ALBR region some raised bogs have their crest located closest to the downgradient river in apparent disequilibrium with the prevailing hydraulic gradient (Glaser et al., 2004b). An explanation for this anomaly is that the geometry of the river network has changed during postgalcial times as the major rivers shifted their course and cut new channels. This also provides an explanation for the timing of bog formation across this study area that was probably linked to the formation of new interﬂuvial divides (Glaser et al., 2004b). Geomorphic evidence of river avulsion is fairly common along the emerging coastal zone of Hudson and James bays (Martini et al., 1980a, b; Martini, 1981, 2006 – this book, Ch. 3). In addition, both Landsat TM imagery (Fig. 15.2a) and SRTM visualizations (Fig. 15.2c; Glaser et al., 2004b) show many abandoned river channels with underﬁt streams across the lowland including several that now cross major drainage divides. Local ice dams can produce river avulsion during the ﬂush of spring

Hydrogeology of major peat basins in North America

363

runoff (freshets), but the longer-term tilting of the landscape driven by isostatic uplift may also be important. Bloom (1998), for example, noted that the tilting of postglacial shorelines, which is on the order of 1 m km 1 is comparable to that of many river segments. River gradients may be even lower within the ALBR study area where the ALBR has an apparent gradient of only 25 cm km 1 making it susceptible to perturbations caused by isostatic uplift or ice dams. One of the most enigmatic hydrological features of the ALBR region is the appearance of fen water tracks within the interior of large raised bogs. The Dupuit simulations of Glaser et al. (2004b) suggest that all bogs have a tendency to ﬂood if they occupy an interﬂuvial divide with a width greater than 5 km; so these water tracks probably function as drains for excess runoff. However, neither groundwater discharge nor transverse-dispersive mixing seem likely mechanisms to explain the water chemistry in these fens. Groundwater would most likely discharge to the rivers rather than interﬂuvial areas occupied by many of these bogs, and the internal water tracks arise too close to the bog crest for the transverse mixing mechanism to be effective. As a result, Glaser et al. (2004c, b) suggested that some internal mechanism related to the production of biogenic gases might be important. Some support for this hypothesis is provided by the measurement of zones of overpressure in several bogs and fens within the study area (Reeve, 1996; Glaser et al., 2004c).

Response of peatland flow systems to climatic thresholds Groundwater investigations in the glacial Lake Agassiz region of northern Minnesota were ﬁrst focused on explaining the development of two peat mounds in the Lost River peatland (Fig. 15.6). One of these mounds had the chemical and biological properties of a raised bog, whereas the other was typical of an extremely rich fen (Glaser et al., 1990). The development of these mounds could only be explained by groundwater ﬂow since they were located less than 1 km apart and the entire surrounding area was covered with peat. This study showed that (1) a raised bog and an adjacent spring-fen mound were both located over a regional discharge zone for groundwater, (2) circumneutral groundwater upwelled into the deeper peat of both mounds but only discharged at the surface of the spring-fen mound, (3) a small recharge cell under the raised bog ﬂushed the uppermost peat strata (0–100 cm depth) and deﬂected upwelling groundwater to the bog margins, (4) head gradients reversed seasonally under the bog, and the peat surface rose by as much as 10 cm during the study period, and (5) an unexpected zone of overpressure persisted at about 1 m depth under the bog (Almendinger et al., 1986; Siegel and Glaser, 1987; Glaser et al., 1990). A regional survey was then conducted across the GLAP to determine the relationship of groundwater ﬂow systems to the peat landform patterns. The preliminary hypothesis for groundwater–peatland interactions in this peat basin proposed that (1) raised bogs are located in recharge zones for groundwater, (2) fen water tracks are located in zones of lateral ﬂow or groundwater discharge, and (3) spring-fens are located in areas of groundwater discharge (Glaser, 1992b). Each of these different ﬂow regimes is related to the hydraulic properties of the deposits that underlie the

364

P.H. Glaser et al.

Figure 15.6. Aerial photograph of the Lost River peatland, northern Minnesota. The image covers an area approximately 2 km across.

peatlands suggesting a close coupling among hydrology, geology, and peatland development. The regional applicability of the hypothesis was established by a synoptic study of 32 bog, fen, and spring fen landforms across the 7600 km2 GLAP study area (Romanowicz et al., 1993, 1995; Siegel et al., 1995; Glaser et al., 1997). The ﬁrst survey was conducted at the height of a 3-year drought in 1990, whereas the same sites were re-sampled the following year after several months of normal precipitation. Hydraulic head gradients did not vary under the spring fens and water tracks between the two sampling periods indicating that these sites remained discharge zones or zones of lateral ﬂow, respectively. However, the head gradients under the raised bogs reversed during this same period, similar to the reversals previously observed in the Lost River study (Fig. 15.7). At the height of the drought, all of the raised bogs were located over discharge zones for groundwater. The upwelling groundwater, however, did not affect the surface vegetation because the water table in 1990 had fallen 1–2 m below the peat surface. The following year the water table rose closer to the peat surface after 2 months of normal precipitation and a local recharge ﬂow-cell developed in the upper portion of the peat proﬁle. Although the vertical ﬂow systems within raised bogs may reverse rapidly on a seasonal or annual basis, the proﬁles of solute concentrations in pore waters change much more slowly on decadal time-scales. This ﬁnding is based on a time-series set of ﬁeld measurements from the Lost River peatland and a one-dimensional solute transport model based on the advection-dispersion equation (Siegel et al., 1995; Glaser et al., 1996).

Hydrogeology of major peat basins in North America

365

Figure 15.7. Hydraulic-head measurements from a raised bog in the Red Lake peatland in 1990 (left) and 1991 (right). All measurements of head are relative to the water table.

Both ﬁeld data and modeling simulations show that the geographic distribution of recharge, discharge, and lateral ﬂow systems in the GLAP are closely coupled to the hydraulic properties of the glacial deposits that underlie the peat. Vertical ﬂow systems generally occur in peat landforms that overlie permeable sand and gravel deposits, whereas lateral ﬂow systems preferentially occur where the peat landforms accumulated over less permeable silty or clayey lakebeds (Glaser et al., 1997; Reeve et al., 2000). In addition, heterogeneities in the thick glacial deposits that underlie the peatlands are an important determinant for the geometry of regional ﬂow systems. Boldt (1986) and later Reeve et al. (2001b) showed how the location for discharge zones in large peatlands may be determined by heterogeneities in the geologic substratum, such as buried sand and gravel lenses and bedrock ridges. Seismic reﬂection proﬁling (Miller et al., 1992) in this region later conﬁrmed the existence of these buried bedrock ridges under peat landforms where groundwater discharges at the land surface. Siegel (1981, 1983), in contrast, showed how the growth of even slight water-table mounds under raised bogs in the GLAP could drive local recharge cells that extend deep into the underlying glacial deposits. Field investigations showed that interactions between local and regional ﬂow systems in the GLAP are more complex than those originally simulated by the Siegel 1981 model (Siegel, 1981). In this relatively dry region, raised bogs preferentially develop over regional discharge zones for groundwater that upwells into the deeper portion of the peat proﬁle. During wet years, water-table mounds under the bogs drive ﬂow downward into the deeper peat and deﬂect the upwelling groundwater laterally to the bog margins (Fig. 15.8). After periods of drought, however, the watertable mounds dissipate and upwelling groundwater can then rise higher in the peat proﬁle (Siegel et al., 1995). The onset of severe droughts across the GLAP about 1200 14C yr BP, apparently was responsible for the discharge of groundwater at the

P.H. Glaser et al.

366

Figure 15.8.

Diagram of climate-driven ﬂow systems under raised bogs in the GLAP (after Siegel et al., 1995).

spring-fen mound at Lost River that abruptly converted a raised bog into an extremely rich fen (Glaser et al., 1990, 1996). Groundwater also upwelled into the proﬁle of the other raised bog at Lost River, but apparently from another recharge zone at lower elevation since the upwelling groundwater never reached the peat surface. This explanation is supported by d18O and dD of pore waters, which indicate that pore waters from the two peat mounds at Lost River have different recharge zones (Hogan et al., 2000; Siegel et al., 2001). These climatically driven ﬂow systems also directly affect biogeochemical processes within the entire peat proﬁle by determining the supply and vertical transport of nutrients, labile organic substrates, terminal electron acceptors, and other dissolved solutes. Biogeochemical drivers for groundwater flow in large peat basins In large peat basins, microbial metabolism within the deeper peat strata is closely coupled to groundwater ﬂow systems. Methanogenesis and fermentation reactions may be stimulated by downward transport of labile carbon substrates exuded from plant roots in local recharge cells, or by the upward transport of inorganic nutrients and bases from the underlying mineral soil by advection in groundwater discharge zones. Methane dissolved in the deeper pore waters of bogs and fens of the GLAP, for example consistently has radiocarbon ages that are 1000–2000 years younger than that of the adjacent peat (Fig. 15.9; Chanton et al., 1995; Chasar et al., 2000, 2001; Chanton et al. 2004). Isotopic mixing equations indicate that 20–80% of this methane was derived from modern carbon substrates apparently transported downward

Hydrogeology of major peat basins in North America

367

Figure 15.9. Depth proﬁles for radiocarbon ages of dissolved methane and solid-phase peat in the bogs and fens from the Glacial Lake Agassiz peatlands, northern Minnesota.

from living plant roots. Similar ﬁndings were reported for smaller peatlands in Canada and England (Aravena et al., 1993; Charman et al., 1994, 1999). These results indicate that the downward transport and rapid recycling of modern primary production by microbes may be responsible for the unexpectedly high concentrations of dissolved methane found by Romanowicz et al. (1993, 1995). Transport processes may also be responsible for the occurrence of metabolic hot spots where high rates of methanogenesis enrich the surrounding pore waters in deuterium (Siegel et al., 2001). Climatic and tectonic drivers that alter groundwater ﬂow systems will also have an important impact on rates of methanogenesis and fermentation in large peat basins. The regional drought cycle in the GLAP, for example controls the elevation of the mixing interface between upwelling groundwater and downward ﬂowing recharge in the peat proﬁles of raised bogs (Siegel et al., 1995; Glaser et al., 1996, 1997). During

368

P.H. Glaser et al.

moist periods the growth of the water-table mounds under raised bogs drives recharge water deeper into the peat proﬁle increasing the supply of labile root exudates (Chanton et al., 1995; Chasar et al., 2000, 2001). In contrast during droughts groundwater upwells higher into the peat proﬁle ﬂushing these root exudates from the pore ﬂuids, but this effect may be counter-balanced by the increasing supply of mineral nutrients. Climatic factors also have a strong inﬂuence on the storage and release of biogenic gases in the deeper peat. Large volumes of free-phase gas in the form of discrete bubbles form conﬁning layers in the peat producing local zones of overpressure ﬁrst observed at Lost River (Siegel and Glaser, 1987), but subsequently found across raised bogs in the GLAP and ALBR area (Romanowicz et al., 1993, 1995; Reeve, 1996; Glaser et al., 1997, 2004a). During droughts these semi-elastic conﬁning layers, episodically rupture in response to external drivers such as changes in the water-table elevations or atmospheric pressure or an internal excess gas accumulation (Rosenberry et al., 2003; Glaser et al., 2004a). These ruptures trigger a massive release of methane to the atmosphere that may exceed 35 g CH4 m 2 per event and associated topographic oscillations in excess of 16 cm in a 4-hour period (Glaser et al., 2004a). Overpressured gas reservoirs in the deeper peat will alter the hydrodynamical interactions between local and regional ﬂow systems that are normally controlled in peatlands by the regional climate and hydrogeologic setting. The superposition of geologically deﬁned ﬂow-systems with overpressures generated by microbial metabolism creates an additional level of complexity in considering the hydrogeology of large peatlands.

Problems, prospects, and future directions Synoptic and site-intensive investigations in the GLAP and Hudson Bay Lowland indicate that peatland ecosystems in large peat basins are closely adjusted to groundwater ﬂow systems. Thus, the dynamics of these systems may be elucidated by groundwater models if adequate ﬁeld data is available to set realistic model parameters and boundary conditions. However, the unique properties of peatlands raise important challenges for most modeling investigations. The growth of peat, for example, continually alters the elevation of the land surface and the topography of the water table modifying ﬂow systems on centennial or millennial time scales. The original groundwater model formulated by Siegel showed how the formation of even small peat mounds on the relatively ﬂat plain of Lake Agassiz can create local ﬂow systems that transport inorganic solutes from the underlying glacial deposits to the peat surface (Siegel, 1981, 1983). Although this model needed to be modiﬁed after detailed ﬁeld data became available, it still clariﬁed important feedbacks between the development of peat landforms and groundwater ﬂow. Attempts to build coupled groundwater/peat accumulation models have been limited to analytical approaches based on the Dupuit assumption ﬁrst applied to raised bogs by Ingram (1982) and later developed by Clymo (1991, 1992), Winston (1994), Hilbert et al. (2000), Belyea and Clymo (2001), and Belyea and Malmer (2004). A similar analytical approach based on Dupuit ﬂow has been used to estimate

Hydrogeology of major peat basins in North America

369

the maximum elevation (‘limiting height’) that the water table and peat surface can rise in any given hydrogeologic setting (Glaser et al., 2004b). These analytical models are particularly valuable for sensitivity analysis to determine the effect of a singlemodel parameter by varying its value, while holding the other model parameters constant. However, a serious limitation of simple analytical models is that they assume isotropic conditions, simple geometry, and idealized boundary conditions (Konikow and Bredehoeff, 1992). Analytical models are therefore difﬁcult to apply to actual ﬁeld conditions where the model parameters and other factors vary in time and space. This limitation is particularly important in boreal peatlands where the material properties of the porous media are transient. Peat is an elastic material that deforms readily in response to even small tensile or shear stresses. Surface oscillations in excess of 30 cm both vertically and horizontally, for example, were recorded in the Red Lake peatland by a local global positioning system (GPS) network (Glaser et al., 2004a). These surface oscillations, which occurred in 6–17 h, were attributed to the release of gas bubbles from the deeper peat. Vertical oscillations of similar magnitude were previously recorded in the Lost River peatland using laser and electronic levels (Almendinger et al., 1986). Gas bubbles produced by methanogenesis and fermentation reactions may also alter the hydraulic properties of the deeper peat by obstructing peat pores (Brown et al., 1989; Reynolds et al., 1992; Beckwith and Baird, 2001; Fraser et al., 2001), forming conﬁning layers (Romanowicz et al., 1995), and generating zones of overpressuring (Rosenberry et al., 2003; Glaser et al., 2004a; Strack et al., 2005; Kellner et al., in press). Peat pores can also expand and contract in response to ﬂushing with salt solutions (Ours et al., 1997) a phenomenon known as dilation, which was previously observed in salt marsh sediments (Nuttle et al., 1990). These elastic properties of peat are partially the result of a very high water content (490% wet weight), liquid limit, and high ﬁber content (Hobbs, 1986). It is not surprising that vertical oscillations of the peat surface of 2–3 cm were originally ascribed to changes in water storage (Weber, 1902; Ingram, 1983; Roulet, 1991). Seasonal changes in water storage may also be sufﬁcient to generate transient zones of overpressure because of the slow dissipation of excess pore pressure following a drop in the water table, which unloads the peat column (Waddington and Roulet, 1997; Devito et al., 1997; Fraser et al., 2001; Van Seters and Price, 2001). Recent modeling studies of this effect in large peat basins shows that this mechanism is capable of driving local hydraulic reversals producing transient zones of overpressure (Reeve et al., in press). However, a large body of independent data (described above) shows that ﬂow reversals in the GLAP were instead driven by climatic changes that control interactions between local and regional ﬂow systems. In addition, changes in water storage can only account for small (2–3 cm) topographic oscillations in peatlands, whereas the larger topographic deformations in excess of 10 cm can only be produced by the storage and release of large volumes of gas from the deeper peat. This disagreement between a model simulation and ﬁeld data illustrates both the value and limitations of mathematical models. Different mathematical models and even multiple calibrations of the same model can produce the same result. Since no model calibration is unique, independent sets of ﬁeld data are essential for discriminating between model simulations that realistically ﬁt ﬁeld conditions from those

370

P.H. Glaser et al.

that do not (Konikow and Bredehoff, 1992; Oreskes et al., 1994). Models therefore serve as important tools for formulating multiple working hypotheses that can be realistically tested by ﬁeld data or for interpreting large complex datasets. Models are therefore best used as a means not an end (Clymo, 1992). They often can suggest the most promising hypotheses to pursue with lab and ﬁeld studies and also serve as a means to make sense of complex datasets. When used with this objective, hydrogeologic models offer a simplifying approach for untangling the complex interactions among climate, hydrology, and peatland carbon balance in a changing environment. Conclusions Recent global warming should alter the net carbon balance of northern peatlands and their current net exchange of greenhouse gases to the atmosphere. Current attempts to model the response of these large terrestrial ecosystems to future climatic change are beset with uncertainties regarding the nature of these climatic changes (such as temperature, precipitation, seasonality), the response of groundwater ﬂow systems, and the feedbacks with biological processes particularly those regulating primary production and decomposition. At present the dynamics of local and regional ﬂow systems in large peat basins has only been imperfectly characterized. Over 20 years of study in the GLAP for example has continued to uncover new layers of complexity not only with regard to the geometry of these ﬂow systems but also their interactions with climate, geology, and biological processes (such as production of greenhouse gases). Nevertheless, studies in the GLAP and ALBR showed that bogs, fens, and spring fens are sensitively adjusted to groundwater ﬂow systems and the distribution of these ecosystems can be predicted from the local hydrogeologic setting. These large peat basins also provide an important natural laboratory to elucidate the drivers for peatland development. Climatic drivers for groundwater ﬂow systems and peatland development are most apparent in the unusual climatic setting of the GLAP, whereas geomorphic drivers are evident in the isostatically rising landscape of the ALBR. The peat landforms visible on aerial photographs and satellite imagery provide a faithful indicator for the distribution of different groundwater ﬂow systems over the landscape since raised bog landforms are consistently associated with groundwater mounds and recharge systems, whereas fen landforms are found in zones of discharge or lateral ﬂow. These landforms therefore provide valuable tools for calibrating and testing mathematical models that seek to simulate climate/groundwater/peatland interactions at the local and regional scale. Despite the complexity of the interactions among climate, groundwater, and peatland carbon balance these models should be capable of identifying the most important gaps in ﬁeld data and guiding ﬁeld research. References Almendinger, J.C., Almendinger, J.E., and Glaser, P.H., 1986. Topographic ﬂuctuations across a springfen and raised bog in the Lost River peatland, northern Minnesota. J. Ecol. 74, 393–401.

Hydrogeology of major peat basins in North America

371

Andrews, J.T. and Peltier, W.R., 1989. Quaternary geodynamics of Canada. In: Fulton R.J. (Ed), Quaternary Geology of Canada and Greenland. Geological Suvey of Canada no. 1 (also Geological Society of America, The Geology of North America, vol. K-1), Ottawa, pp. 543–572. Aravena, R., Warner, B.G., and Charman, D.J., 1993. Carbon isotopic composition of deep carbon gases in an ombrogenous peatland, northwestern Ontario. Radiocarbon 35, 271–276. Auer, V., 1920. Uber die Entstehung der Stra¨nge auf den Torfmooren. Acta Forestr. Fenn. 12, 1–145. Baden, W. and Eggelsmann, R., 1963. Zur Durchla¨ssigkeit der Moorbo¨den. Z. Kult. Tech. 4, 226–254. Baker, D.G., Haines, D.A., and Stubb, J.H., 1967. Climate of Minnesota. Part V. Precipitation facts, normals, and extremes. University of Minnesota Agriculture Experiment Station Technical Bulletin 254, 1–43. Baker, D.G., Nelson, W.W., and Kuehnast, E.L., 1979. Climate of Minnesota. Part XII – the hydrologic cycle and soil water. University of Minnesota Agriculture Experiment Station Technical Bulletin 322, 1–23. Beckwith, C.W. and Baird, A.J., 2001. The effect of biogenic gas bubbles on water ﬂow through poorly decomposed blanket peat. Water Resour. Res. 37, 551–558. Belyea, L.R. and Clymo, R.S., 2001. Feedback control of the rate of peat formation. Proc. Roy. Soc. London B 268, 1315–1321. Belyea, L.R. and Malmer, N., 2004. Carbon sequestration in peatland patterns and mechanisms of response to climate change. Global Change Biol. 10, 1043–1052. Bloom, A.L., 1998. Geomorphology. A Systematic Analysis of Quaternary Landforms. Prentice-Hall, Engelwood Cliffs. Boelter, D.H., 1969. Physical properties of peats related to degree of decomposition. Soil Sci. Soc. Am. Proc. 33, 606–609. Boelter, D.H. and Verry, E.S., 1977. Peatland water in the northern Lake States. United States Department of Agriculture Technical Report NC-31, pp. 1–22. Boldt, D.R., 1986. Computer simulations of groundwater ﬂow in a raised bog system, Glacial Lake Agassiz peatlands, northern Minnesota. M.S. Thesis, Syracuse University, Syracuse, N.Y. USA. Brown, A., Mathur, S.P., and Kushner, D.J., 1989. An ombrotrophic bog as a methane reservoir. Global Biogeochem. Cycles 3, 205–213. Chanton, J.P., Bauer, J., Glaser, P.H., et al., 1995. Radiocarbon evidence for the substrates supporting methane formation within northern Minnesota peatlands. Geochim. Cosmochim. Acta, 59, 3663–3688. Chanton, J.P., Chasar, L.C., Glaser, P., and Siegel, D., 2004. Carbon and hydrogen isotopic effects in microbial methane from terrestrial environments. In: Flanagan, L.B., Ehleringer, J.R., and Pataki, D.E. (Eds), Stable Isotopes and Biosphere-Atmosphere Interactions. Elsevier–Academic Press, Amsterdam. Charman, D.J., Aravena, R.A., Bryant, C.L., and Harkness, D.D., 1999. Carbon isotopes in peat, DOC, CO2, and CH4 in a Holocene peatland on Dartmoor, southwest England, U.K. Geology. 27, 539–542. Charman, D.J., Aravena, R.A., and Warner, B.G., 1994. Carbon dynamics in a forested peatland in northeastern Canada. J. Ecol. 82, 55–62. Chasar, L.S., Chanton, J.P., Glaser, P.H., and Siegel, D.I., 2000. Methane concentration and stable isotope distribution as evidence of rhizospheric processes: comparison of a fen and bog in the Glacial Lake Agassiz peatland complex. Ann. Bot. 86, 655–663. Chasar, L.S., Chanton, J.P., Glaser, P.H., et al., 2001. Radiocarbon and stable carbon isotopic evidence for transport and transformation of dissolved organic carbon, dissolved inorganic carbon, and CH4 in a northern Minnesota peatland. Global Biogeochem. Cycles 14, 1095–1108. Chason, D.B. and Siegel, D.I., 1986. Hydraulic conductivity and related physical properties of peat, Lost River peatland, northern Minnesota. Soil Sci. 142, 91–99. Clark, C.D., 1993. Mega-scale glacial lineations and cross-cutting ice-ﬂow landforms. Earth Surf. Syst. Proc. Land. 18, 1–29. Clymo, R.S., 1991. Peat growth. In: Cushing, E.J. and Shane, L.C. (Eds), Quaternary Landscapes. University of Minnesota Press, Minneapolis. Clymo, R.S., 1992. Models of peat growth. Suo 43, 127–136. COHMAP, 1988. Climatic changes of the last 18,000 years: observations and model simulations. Science 241, 1043–1052.

372

P.H. Glaser et al.

Coombs, D.B., 1954. The physiographic subdivisions of the Hudson Bay Lowland south of 60 degrees north. Geogr. Bull. 6, 1–16. Dau, J.H.C., 1823. Neues Handbuch u¨ber den Torf. J.C. Hinrichsche Buchhandlung, Leipzig, Germany, 240pp. Dean, W.G., 1959. Physiography and vegetation of the Albany River map area, northern Ontario an aerial photograph reconnaissance. Ph.D. Thesis, McGill University, Montreal, 391pp. Devito, K.J., Waddington, J.M., and Branﬁreun, B.A., 1997. Flow reversals in peatlands inﬂuenced by local groundwater systems. Hydrol. Process. 11, 103–110. Dredge, L.A. and Cowan, W.R., 1989. Quaternary geology of the southwestern Canadian Shield. In: Fulton, R.J. (Ed.), Quaternary geology of Canada and Greenland (Chapter 3). Geological Survey of Canada no 1 (also: Geological Society of America, The geology of North America, Vol. K-1), pp. 214–249. Dupuit, J., 1863. Etudes theoriques et pratiques sur le mouvement des eaux dans les canaux decouverts et a travers les terrains permeables. Dunot, Paris. Du Rietz, G.E., 1949. Huvdenheter och huvudgra¨nser i svensk myrvegetatio. Summary: main units and main limits of Swedish mire vegetation. Svensk Botnisk tidskrift, 48, 174–187. Dyke, A.S. and Prest, V.K., 1987. Paleogeography of northern North America 18,000 to 5,000 years ago. Geological Survey of Canada, Map 1703A, Scale 1:12,500,000. Dyke, A.S., Vincent, J-S., Andrews, J.T., et al., 1989. The Laurentide ice sheet and an introduction to the Quaternary geology of the Canadian Shield. In: Fulton, R.J. (Ed.), Quaternary geology of Canada and Greenland (Chapter 3). Geological Survey of Canada no 1 (also: Geological Society of America, The geology of North America, vol. K-1), Ottawa, pp. 178–189. Fetter, C.W., 2000. Applied Hydrogeology. Prentice-Hall, Englewood Cliffs, NJ. Fraser, C.J.D., Roulet, N.T., and Laﬂeur, P.M., 2001. Groundwater ﬂow patterns in a large peatland. J. Hydrol. 246, 142–154. Freeze, R.A. and Cherry, J.A., 1979. Groundwater. Prentice-Hall, Englewood Cliffs, NJ, 604pp. Glaser, P.H., 1987. The ecology of patterned boreal peatlands of northern Minnesota: A community proﬁle. U.S. Fish and Wildlife Service Biological Report 85 (7.14), 98pp. Glaser, P.H., 1989. Detecting ecologic and hydrogeochemical processes in large peat basins with Landsat TM imagery. Remote Sensing of Environ. 28, 109–119. Glaser, P.H., 1992. Raised bogs in eastern North America: regional controls on species richness and ﬂoristic assemblages. J. Ecol. 80, 535–554. Glaser, P.H., 1992. Peat landforms. In: Wright, H.E. Jr. and Cofﬁn, B.A. (Eds), Patterned Peatlands of Northern Minnesota. University of Minnesota Press, Minneapolis, pp. 3–14. Glaser, P.H., 1992. Vegetation and water chemistry. In: Wright, H.E. Jr. and Cofﬁn, B.A. (Eds), Patterned Peatlands of Northern Minnesota. University of Minnesota Press, Minneapolis, pp. 15–26. Glaser, P.H., Bennett, P.C., Siegel, D.I., and Romanowicz, E.A., 1996. Paleo-reversals in groundwater ﬂow and peatland development in the Lost River peatland, northern Minnesota, USA. Holocene 6, 413–421. Glaser, P.H., Chanton, J.P., Morin, P., et al., 2004. Surface deformations as indicators of deep ebullition ﬂuxes in a large northern peatland. Global Biogeochem. Cycles, 18, GB1003. Glaser, P.H., Hansen, B.C.S., Siegel, D.I., and Reeve, A.S., 2004. Rates, pathways, and drivers for peatland development in the Hudson Bay Lowlands, northern Ontario. J. Ecol. 92, 1036–1053. Glaser, P.H. and Janssens, J.A., 1986. Raised bogs in eastern North America: transitions in landforms and gross stratigraphy. Can. J. Bot. 64, 395–415. Glaser, P.H., Janssens, J.A., and Siegel, D.I., 1990. The response of vegetation to hydrological and chemical gradients in the Lost River Peatland, northern Minnesota. J. Ecol. 78, 1021–1048. Glaser, P.H., Siegel, D.I., Reeve, A.S., et al., 2004. Tectonic drivers for vegetation patterning and landscape evolution in the Albany River region of the Hudson Bay Lowlands. J. Ecol. 92, 1054–1070. Glaser, P.H., Siegel, D.I., Romanowicz, E.A., and Shen, Y.P., 1997. Regional linkages between raised bogs and the climate, groundwater, and landscape features of northwestern Minnesota. J. Ecol. 85, 3–16. Glaser, P.H., Wheeler, G.A., Gorham, E., and Wright, H.E. Jr., 1981. The patterned peatlands of the Red Lake peatland, northern Minnesota: vegetation, water chemistry, and landforms. J. Ecol. 69, 575–599.

Hydrogeology of major peat basins in North America

373

Gore, A.J.P. (Ed.), 1983a, Mires: swamp, bog, fen, and moor. General Studies. Elsevier, Amsterdam, Vol. A. Gore, A.J.P. (Ed.), 1983b, Mires: swamp, bog, fen, and moor. Regional Studies. Elsevier, Amsterdam, Vol. B. Gorham, E., 1991. Northern peatlands: role in the carbon cycle and probable responses to climatic warming. Ecol. Appl. 1, 182–195. Gorham, E., Eisenreich, S.J., Ford, J., and Sandtelmann, M.V., 1985. The chemistry of bog waters. In: Stumm, W. (Ed.), The Chemical Processes in Lakes. Wiley, New York, pp. 330–363. Gorham, E., Janssens, J.A., and Glaser, P.H., 2003. Rates of peat accumulation during the postglacial period in 32 sites from Alaska to Newfoundland, with special emphasis on northern Minnesota. Can. J. Bot. 81, 429–438. Halsey, L.A., Vitt, D.H., and Gignac, L.D., 2000. Sphagnum-dominated peatlands in North America since the Last Glacial Maximum: their occurrence and extent. Bryologist 103, 334–352. Harbaugh, A.W. and McDonald, M.G., 1996. User’s documentation for Modﬂow-96. An update to the U.S. Geological Survey Modular Finite-Difference Groundwater Groundwater Flow Model. U.S. Geological Survey Open File Report 96-486, pp. 1–222. Heinselman, M.L., 1963. Forest sites, bog processes, and peatland types in the Glacial Lake Agassiz region, Minnesota. Ecol. Monogr. 33, 327–372. Heinselman, M.L., 1970. Landscape evolution, peatland types, and the environment in the Lake Agassiz Peatland Natural Area, Minnesota. Ecol. Monogr. 40, 235–261. Hemond, H.F., 1980. Biogeochemistry of Thoreau’s bog, Concord, Massachusetts. Ecology 50, 507–526. Hemond, H.F., 1990. Acid neutralizing capacity, alkalinity, and acid-base status of natural waters containing organic acids. Environ. Sci. Technol. 24, 1468–1489. Hemond, H.F. and Goldman, J.C., 1985. On non-Darcian water-ﬂow in peat. J. Ecol. 73, 579–584. Hilbert, D.W., Roulet, N., and Moore, T., 2000. Modelling and analysis of peatlands as dynamical systems. J. Ecol. 88, 230–242. Hobbs, N.B., 1986. Mire morphology and the properties and behaviour of some British and foreign peats. Q. J. Eng. Geol. London, 19, 7–80. Hogan, J.F., Blum, J.D., Siegel, D.I., and Glaser, P.H., 2000. 87Sr/86Sr as a tracer of groundwater discharge and precipitation recharge in the Glacial Lake Agassiz Peatlands, northern Minnesota. Water Resour. Res. 36, 3701–3710. Hogg, N., Slatterly, J., and Wilson, A., 1953. Drilling in the James Bay Lowland: Part I – Drilling by the Ontario Department of Mines. Ontario Department of Mines, Annual Report 61, Part 6, pp. 115–140. Holden, J., 2006. (this book, Chapter 14). Peatland hydrology. In: Martini, I.P., Matı´ nez, C.A., and Chesworth, W. (Eds), Peatlands: Evolution and Records of Environmental and Climatic Changes. Elsevier, Amsterdam. Hunter, G.T., 1970. Postglacial uplift at Fort Albany, James Bay. Can. J. Earth Sci. 7, 547–548. Ingram, H.A.P., 1978. Soil layers in mires: function and terminology. J. Soil Sci. 29, 224–227. Ingram, H.A.P., 1982. Size and shape in raised mire ecosystems: a geophysical model. Nature 297, 300–303. Ingram, H.A.P. 1983, Hydrology. In: Gore, A.J.P. (Ed.), Mires: Swamp, Bog, Fen, and Moor. General Studies. Elsevier, Amsterdam, Vol. A, pp. 67–158. Ivanov, K.E., 1975. Vodoobmen v bolotnykh landshaftakh. Gidrometeoizdat, Lenningrad [Water Movement in mirelands, Thomson, A., and Ingram, H.A.P. (Trans.), Academic Press, London, 1981]. Jacobson, G.L., Webb T. III, and Grimm, E.C., 1987. Patterns and rates of vegetation change during the deglaciation of eastern North America. In: Ruddiman, W.F. and Wright, H.E. Jr., (Eds), Geology of North America, Vol. K-3, North America and Adjacent Oceans during the Last Deglaciation. Geological Society of America, pp. 277–288. Janssen, C.R., 1968. Myrtle Lake: a late- and post-glacial pollen diagram from northern Minnesota. Can. J. Bot. 46, 1397–1408. Jeglum, J.K. and Cowell, D.W., 1982. Wetland ecosystems near Kinoje Lakes, southern interior Hudson Bay Lowland. Natural. Canad. 109, 621–635. Johnson, E.A., 1992. Fire and Vegetation Dynamics. Cambridge University Press, Cambridge.

374

P.H. Glaser et al.

Kellner, E., Waddington, J.M., and Price, J.S., (in press). Dynamics of biogenic gas bubbles in peat: potential effects on water storage and peat deformation. Water Resour. Res. Khalil, A.K., 2000. Atmospheric Methane: Its Role in the Global Environment. Springer, Berlin. King, W., 1685. On the bogs and loughs of Ireland. Phil. Trans. Roy. Soc. London, 15 (170), 948–960. Kivinen, E. and Pakarinen, P., 1981. Geographical distribution of peat resources and major peatland complex types in the world. Ann. Acad. Scient. Fenn. A, III, Geolog.-Geograph. 132, 5–28. Konikow, L.F. and Bredehoeff, J.D., 1992. Ground-water models cannot be validated. Adv.Water Resour. 15, 75–83. Kulczynski, S., 1949. Peat bogs of Polesie. Me´m. de l’Acad. Pol. Sci. Lett., Classe Sci. Math. et Naturell., Se´r. B: Sci. Nat. 15, 1–356. Martini, I.P., 1981. Morphology and sediments of the emergent Ontario coast of James Bay, Canada. Geogr. Ann. 63, 81–94. Martini, I.P., 2006. (this book, Chapter 3). The cold-climate peatlands of the Hudson Bay Lowland, Canada: brief review of recent work. In: Martini, I.P., Matı´ nez, C.A., and Chesworth, W. (Eds), Peatlands: Evolution and Records of Environmental and Climatic Changes. Elsevier, Amsterdam. Martini, I.P., Cowell, D.W., and Wickware, G.M., 1980a. Geomorphology of southwestern James Bay: a low energy, emergent coast. In: McCann, S.B. (Ed.), The Coastline of Canada. Geological Society of Canada, Ottawa, Paper 80–10, pp. 293–301. Martini, I.P., Morrison, R.I.G., Glooschenko, W.A., and Protz, R., 1980. Coastal studies in James Bay, Ontario. Geosci. Canada 7, 11–21. McDonald, J.M. and Harbaugh, A.W., 1984. A modular three-dimensional ﬁnite difference groundwater ﬂow model. U.S. Geological Survey Open File Report 83-875, 528pp. McNamara, J.P., Siegel, D.I., Glaser, P.H., and Beck, R.M., 1992. Hydrologic controls on peatland development in the Malloryville wetland, New York (USA). J. Hydrol. 140, 279–296. Miller, P., Shaw, G.H., Glaser, P.H., and Siegel, D.I., 1992. Bedrock topography beneath the Red Lake Peatlands. Abstract, Geological Society of America, National Meeting, Cincinnati, OH, October 1992. Moore, T.R., Roulet, N.T., and Waddington, J.M., 1998. Uncertainty in predicting the effect of climatic change on the carbon cycling of Canadian peatlands. Climate Change 40, 229–245. Morin, P.J. and Thorliefson H., 2005. Digital elevation model for the upper Midwest and Canada. GSA Today 14. National Wetlands Working Group, 1986. Wetlands of Canada, (C.D.A. Rubec Co-ordinator). Polyscience, Montreal. Nuttle, W.K., Hemond, H.F., and Stolzenbach, K.D., 1990. Mechanisms of water storage in salt marsh sediments: the importance of dilation. Hydrol. Process. 4, 1–13. Oreskes, N., Shrader-Frechette, K., and Belitz, K., 1994. Veriﬁcation, validation, and conﬁrmation of numerical models in the earth science. Science 263, 641–646. Ours, D.P., Siegel, D.I., and Glaser, P.H., 1997. Chemical dilation and the dual porosity of humiﬁed bog peat. J. Hydrol. 196, 348–360. Pala, S. and Weischet, W., 1982. Toward a physiographic analysis of the Hudson bay-James bay Lowland. Natural. Canad. 109, 637–651. Price, J.S. and Maloney, D.A., 1994. Hydrology of a patterned bog-fen complex in southeastern Labrador, Canada. Nordic Hydrol. 25, 313–330. Reeve, A.S., 1996. Numerical and multivariate statistical analysis of hydrogeology and geochemistry in large peatlands. Ph.D. Dissertation, Syracuse University, Syracuse, N.Y. Reeve, A.S., Evensen, R., Glaser, P.H., et al., (in press). Flow path oscillations in transient ground-water simulations of large peatland systems. J. Hydrol. Reeve, A.S., Siegel, D.I., and Glaser, P.H., 1996. Geochemical controls on peatland pore-water from the Hudson Bay Lowlands: a multivariate and statistical approach. J. Hydrol. 181, 285–304. Reeve, A.S., Siegel, D.I., and Glaser, P.H., 2000. Simulating vertical ﬂow in large peatlands. J. Hydrol. 227, 207–217. Reeve, A.S., Siegel, D.I., and Glaser, P.H., 2001. Simulating dispersive mixing in large peatlands. J. Hydrol. 242, 103–114. Reeve, A.S., Warzocha, J., Glaser, P.H., and Siegel, D.I., 2001. Regional ground-water ﬂow modeling of the Glacial Lake Agassiz peatlands, Minnesota. J. Hydrol. 243, 91–100.

Hydrogeology of major peat basins in North America

375

Rennie, H., 1810. Essays on the Natural History and Origin of Peat Moss. Constable, Edinburgh. Reynolds, W.D., Brown, D.A., Mathur, S.P., and Overend, R.P., 1992. Effect of in-situ gas accumulation on the hydraulic conductivity of peat. Soil Sci. 153, 397–408. Romanowicz, E.A., Siegel, D.I., Chanton, J.P., and Glaser, P.H., 1995. Temporal variations in dissolved methane deep in the Lake Agassiz peatlands, Minnesota (USA). Global Biogeochem. Cycles 9, 197–212. Romanowicz, E.A., Siegel, D.I., and Glaser, P.H., 1993. Hydraulic reversals and episodic methane emissions during drought cycles in mires. Geology 21, 231–234. Rosenberry, D.O., Glaser, P.H., Siegel, D.I., and Weeks, E.D., 2003. Use of hydraulic head to estimate volumetric gas content and ebullition ﬂux in northern peatlands. Water Resour. Res. 39, 1066. Roulet, N.T., 1990. Hydrology of a headwater basin wetland: groundwater discharge and wetland maintenance. Hydrol. Process. 4, 387–400. Roulet, N.T., 1991. Surface level and water table ﬂuctuations in a subarctic fen. Arctic Alpine Res. 23, 303–310. Roulet, N.T. and McKenzie, J.W., 1998. Role of groundwater in determining the pattern of peatlands in the Hudson Bay lowlands. Geol. Soc.Am. Annu. Meet. Abstr.Programs, 30, A–119. Rycroft, D.W., Williams, D.J.A., and Ingram, H.A.P., 1975. The transmission of water through peat. I. Rev. J. Ecol. 63, 535–556. Siegel, D.I., 1981. Hydrogeologic setting of the Glacial Lake Agassiz peatlands, northern Minnesota. U.S. Geological Survey Water Resources Investigations, 81-24a, 1–30. Siegel, D.I., 1983. Groundwater and the evolution of patterned mires, Glacial Lake Agassiz peatlands, northern Minnesota. J. Ecol. 71, 913–923. Siegel, D.I., 1992. Groundwater hydrology. In: Wright, H.E. Jr. and Cofﬁn, B.A. (Eds), Patterned Peatlands of Northern Minnesota. University of Minnesota Press, Minneapolis, pp. 59–69. Siegel, D.I., Chanton, J.P., Glaser, P.H., et al., 2001. Estimating methane production rates in bogs and landﬁlls by deuterium enrichment of pore-water. Global Biogeochem. Cycles 15, 967–975. Siegel, D.I. and Glaser, P.H., 1987. Groundwater ﬂow in a bog-fen complex, Lost River peatland, northern Minnesota. J. Ecol. 75, 743–754. Siegel, D.I., Glaser, P.H., So, J., and Janecky, D.R., (in press). Dynamic balance between organic acids and circumneutral groundwater in a large boreal peat basin. J. Hydrol. Siegel, D.I., Reeve, A.S., Glaser, P.H., and Romanowicz, E., 1995. Climate-driven ﬂushing of pore water in humiﬁed peat. Nature 374, 531–533. Sjo¨rs, H., 1948. Myrvegetation i Bergslagen. Acta Phytogeographica Suecica 21, 1–299. Sjo¨rs, H., 1950. On the relation between vegetation and electrolytes in North Swedish mire waters. Oikos 2, 241–258. Sjo¨rs, H., 1963. Bogs and fens on Attawapiskat River, northern Ontario. Natl. Mus. Can. Bull. Contrib.Bot. 171, 1–31. Sjo¨rs, H. 1983. Mires of Sweden. In: Gore, A.J.P. (Ed.), Mires: Swamp, Bog, Fen, and Moor. Regional Studies. Elsevier, Amsterdam, Vol. B, pp. 69–94. Starkel, L., 1979. Typology of river valleys in the temperate zone during the last 15,000 years. Acta universitatis ouluensis. Series A., 82, Geologica 3, 9–18. Strack, M., Kellner, E., and Waddington, J.M., 2005. Dynamics of biogenic gas bubbles in peat and their effects on peatland biogeochemistry. Global Biogeochem. Cycles 19, GB1003. Tarnocai, C. and Stolbovoy, V., 2006. (this book, Chapter 2). Northern peatlands: their characteristics, development and sensitivity to climate change. In: Martini, I.P., Matı´ nez, C.A., and Chesworth, W. (Eds), Peatlands: Evolution and Records of Environmental and Climatic Changes. Elsevier Amsterdam. Teller, J.T., 1985. Glacial Lake Agassiz and its inﬂuence on the Great Lakes. In: Karrow, P.F. and Calkin, P.E. (Eds), Quaternary Evolution of the Great Lakes. Geological Association of Canada, Special Paper 30, pp. 1–16. Teller, J.T. and Clayton, L. (Eds), 1983. Glacial Lake Agassiz. Geological Association of Canada, Special Paper No. 26. Van Seters, T. and Price, J.S., 2001. The impact of peat harvesting and natural regeneration on the water balance of an abandoned bog, Quebec. Hydrol. Process. 15, 233–248.

376

P.H. Glaser et al.

Vitt, D.H., Halsey, L.A., and Zoltai, S.C., 1994. The bog landforms of continental western Canada in relation to climate and permafrost patterns. Arctic Alpine Res. 26, 1–13. Waddington, J.M. and Roulet, N.T., 1997. Groundwater ﬂow and dissolved carbon movement in a boreal peatland. J. Hydrol. 191, 122–138. Washburn, A.L., 1979. Geocryology. Edward Arnold, London. Weber, C.A., 1902. U¨ber die Vegetation und Entstehung des Hochmoors von Augstumal im Memeldelta mit vergleichenden Ausblicken auf andere Hochmoore der Erde. Paul Parey, Berlin. Webber, P.J., Richardson, J.W., and Andrews, J.T., 1970. Postglacial uplift and substrate age at Cape Henrietta, southeastern Hudson Bay, Canada. Can. J. Earth Sci. 7, 317–325. Wigley, T.M.L. and Schimel, D.S., 2000. The Carbon Cycle. Cambridge University Press, Cambridge. Wilcox, D.A., Shedlock, R.J., and Henderson, W.H., 1986. Hydrology, water chemistry and ecological relations in the raised mound of Cowles Bog. J. Ecol. 74, 1103–1117. Winston, R.B., 1994. Models of the geomorphology, hydrology, and development of domed peat bodies. Geol. Soc. Am. Bull. 106, 1594–1604. Wright Jr., H.E., 1972. Physiography of Minnesota. In: Sims, P.K. and Morey, G.B. (Eds), Geology of Minnesota: A Centennial Volume. Minnesota Geological Survey, Minneapolis, pp. 515–560. Zoltai, S.C., 1972. Palsas and peat plateaus in central Manitoba and Saskatchewan. Canadian J. Forest Res. 2, 291–302. Zoltai, S.C. and Pollet, F.C., 1983. Wetlands in Canada: their classiﬁcation, distribution, and use. In: Gore, A.J.P. (Ed.), Mires: Swamp, Bog, Fen, and Moor. Regional Studies. Elsevier, Amsterdam, Vol. B, pp. 69–94.

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

377

Chapter 16

Slope instability and mass movements in peat deposits A.P. Dykes and K.J. Kirk

Introduction Catastrophic failures of peat deposits have been recorded from many parts of the world, but particularly from Ireland (comprising Northern Ireland (NI) and the Republic of Ireland (RoI)) where peat-cutting activities in the 18th and 19th centuries caused a number of large and damaging failures of lowland raised bogs (Grifﬁth, 1821; Kinahan, 1897a, b; Sollas et al., 1897). Elsewhere in the world, one of the largest recorded peatland failures was the collapse of an upland peat swamp in Australia in 1998: around 5 million m3 of peat was displaced from a valley bog into a reservoir, the continuing operation of which may be at risk (Beder, 2001). This failure has also been attributed to peat-extraction activities (O’Loughlin, 1998). Widespread interest in the topic of peat instability was generated in Ireland by the failure of Knocknageeha Bog, County Kerry, in December 1896, which also involved 5–6 million m3 of peat but killed a family of eight people as well as damaging roads and agricultural land (Cole, 1897; Latimer, 1897; Sollas et al., 1897). Most attention has been paid to slope instability in British and Irish peatlands because approximately 60% of all recorded peat failures are in Ireland, with a further 20% in Great Britain (the United Kingdom (U.K.) excluding NI), although the true incidence of peat failures in the rest of the world is uncertain and may become signiﬁcant in the future in response to global climatic and environmental changes. Events in late 2003 generated a wide regional awareness of the topic throughout the U.K. and the RoI. On 19 September 2003, a severe storm in the early morning triggered 20 large peat slides south of Lerwick in southern Shetland, Scotland, the landslides and ﬂoodwater causing an estimated 5 million USD of damage. In the evening of the same day, a highly localized thunderstorm in the northwest corner of County Mayo, RoI, triggered 35 landslides on one small mountain, several of them peat slides, which destroyed one house, damaged local infrastructure and killed many sheep. The cost of this event was initially thought to be around 11 million USD. Just two weeks later at Derrybrien in County Galway, RoI, a small peat failure on 2 October 2003 preceded an enormous landslide on the 16th, the latter involving around 500,000 m3 of peat; both occurred in forest-covered blanket peat on gently sloping upland terrain. ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09016-X

378

A.P. Dykes and K.J. Kirk

The disparity between the global distribution of peatlands and the global distribution of recorded peat failures (Fig. 16.1) has been commented on previously (Tallis, 2001); it arises either because failures in remote regions are rarely observed and almost never reported, or because peat failures are genuinely rare events outside the British Isles. Table 16.1 lists all the known occurrences of peat failures outside the British Isles, of which two were attributed to peat extraction activities (FeldmeyerChriste and Mulhauser, 1994; Beder, 2001) and two resulted from tipping of excavation spoil onto peat (Hungr and Evans, 1985; Dykes, unpublished observations). The remainder appears to have occurred naturally, but in association with different contributory factors. Instability, often leading to catastrophic failure, arises when forces that act within a peat body to resist the downslope pull due to gravity are exceeded by the disturbing effect of the gravitational force. At the most fundamental level, the total disturbing force is a function of the mass and thickness of the peat on the slope and the geometry of the slope. The resisting force is provided by the rheological properties of the peat and of the mineral substrate; that is, the strength of the materials and how they behave in response to particular stress conditions. The balance between these two sets of forces depends on the combined inﬂuences of other site-dependent (internal) and external factors (Table 16.2). There is little consensus between previous authors of which factors, in which combination(s), might be the most common and most inﬂuential in giving rise to instability. This is of concern if the future hazard from peat instability, particularly with respect to climate change scenarios, peatland management activities, and/or engineering projects on peatlands, is to be adequately assessed and managed. Systematic research into mass movements in peat has only recently begun (Kirk, 2001; Mills, 2002; Dykes and Kirk, 2000, 2001; Warburton et al., 2003, 2004), with the 2003 events outlined earlier demonstrating the urgent need

Figure 16.1. Global distribution of peatlands (after Charman, 2002) and recorded peat failures.

Slope instability and mass movements in peat deposits Table 16.1.

379

Reported examples of peat stability outside the British Isles.

Date

Location

Peat typea

Failure typeb

Source

1763, autumn

Treuenfeld, northern Germany Falkland Islands, South Atlantic Falkland Islands, South Atlantic Campbell Island, New Zealand Auckland Island, New Zealand Tierra del Fuego, Argentina Scho¨nberg, Oberbayern, southern Germany northern Sarawak (Borneo), Malaysia Northeast Spain

RB

BB

Sollas et al. (1897)c

BL

BF

Bailey (1879)

BL

BF

Barkly (1887)

BL

PS?

Campbell (1981)d

BL

PS?

BL

PS?

Aston (1909) and Campbell (1981)d Gallart et al. (1994)e

BL

BS/PS

Vidal (1966)

FP

PF

Wilford (1966)

BL

PS?

B BL

PS BF?

A. Martı´ nez Cortizas (pers. comm., 2004)f Shroder (1976) Wilson et al. (1993)

BL

PF

RB

BB

BL

PS

Selkirk (1996)g

FP(f)

PF

Beder (2001)

FP(b)

PF

A.P. Dykes (unpublished data)

1878, November 1886, June 19th20th century 19th20th century 195770 1960, June

1961, May ‘recent’ 20th century ‘recent’ 20th century early 1980s 1982, spring 1987, September

19901993 1998, August

2003, earlymid a

Nyika Plateau, Malawi Falkland Islands, South Atlantic Prince Rupert, British Columbia, Canada Mouille de la Vraconne National Park, SwitzerlanD Macquarie Island, South Paciﬁc Wingecarribee Swamp, New South Wales, Australia Cordillera Blanca, Peru´

Hungr and Evans (1985) Feldmeyer-Christe and Mulhauser (1994)

BL ¼ blanket bog, RB ¼ raised bog, FP ¼ fen peat (f ¼ fen, b ¼ basin bog: after Hobbs, 1986). (see text) BB ¼ bog burst, BF ¼ bogﬂow, BS ¼ bog slide, PS ¼ peat slide, PF ¼ peat ﬂow. c According to Sollas et al., several German scholars wrote possible explanations for bog failures throughout the 19th century, but no other examples presented. d According to Campbell (i) landsliding erosion was ‘common’ in ‘organic soils’ (including peat) on Campbell Island, and (ii) Aston reported ‘frequent bog-slides’ on Auckland Island e Eight peat failures dated to within this period, with a few other much older ones. f Features observed in this area are not yet conﬁrmed as peat failure scars. g According to Selkirk, ‘many’ peat slides have occurred on Macquarie Island before and since 1980 in addition to the seven dated examples discussed in the paper. b

A.P. Dykes and K.J. Kirk

380 Table 16.2. 1993).

Factors that may contribute to mass movement in peat deposits (after Selby,

Types of factors Factors contributing to high shear stress (1) Removal of lateral support (2) Overloading (3) Transitory stresses (4) Removal of underlying support

(5) Lateral pressure Factors contributing to low shear strength (1) Composition and properties (2) Physico-chemical reactions (3) Effects of pore water

(4) Changes in structure (5) (Relict) structures

Major relevant mechanisms Stream erosion; previous mass movement; excavation by humans Weight of rain or snow; engineering ﬁlls or waste piles Vibrations from human activities; earthquakes Undercutting by running water; subsurface erosion; escape of underlying plastic or semi-liquid material; creation of lakes or reservoirs Water within micro- or macropore spaces; mobilization of residual stress Botanical composition; ﬁber content; degree of humiﬁcation Cation exchange; gas production Buoyancy effects; reduction of capillary tension; viscous drag of moving water in subsurface pipes Spontaneous ﬂuidization; progressive creep; reactivation of earlier shears Natural and artiﬁcial planes of weakness; layered peat over impermeable mineral substrate

for such work. Table 16.3 lists the most recent peat failures, many of which are the focus for this research. This chapter will examine the nature of peat instability and what is known about the possible causes and mechanisms of mass movements in peat deposits. It will identify and discuss wider issues that are raised by previously published accounts of peat failures, in particular concerning the nature and (engineering) properties of peat and how these relate to the observed behavior of peat. Warburton et al. (2004) recently examined some of the key research issues concerning the hydrology of blanket peat-covered hillslopes and how this may determine or control instability in those systems. This chapter highlights some of the other fundamental questions and suggests where future research efforts might be most effectively directed.

Failure types Different types of mass movements in peat can be recognized, and several terms have been used to refer to peat failures in published accounts, most commonly bogﬂow,

Recorded peat failures of the last ten years (19952004 inclusive).

Date

Location

Peat typea

Failure typeb

Details

1995, 2 February

Hart Hope Burn, North Pennines, northern England Cuilcagh Mountain (East), County Cavan, Ireland Coldcleugh Head, North Pennines, northern England Stony River, Slieve Anierin, County Leitrim, Ireland Cuilcagh Mountain (East), County Cavan, Ireland Wingecarribee Swamp, New South Wales, Australia

BL

PS

Warburton et al. (2003)

BL BL

BS PS

Kirk (2001) Mills (2002)

BL

BF

BL FP(f)

BS PF

BL

PS

Kirk (2001) and Yang and Dykes (unpublished data) A.P. Dykes (unpublished data Beder (2001) and National Trust of Australia (undated) Dykes and Kirk (2001)

BL

PS

Kirk (2001)

BL

PS

A.P. Dykes (unpublished data)

BL

PS

A.P. Dykes (unpublished data)

FP(b)

PF

A.P. Dykes (unpublished data)

ca 1997 1997/1998 ca 1998 ca 1998 1998, 9 August

1998, 25 October

PF

A.P. Dykes (unpublished data)

BL

PS(20+)

2003, 19 September

Dooncarton Mountain, County Mayo, Ireland

BL

PS(30+)

J. Warburton and A.P. Dykes (unpublished data) A.P. Dykes and J. Warburton (unpublished data)

2000, 2 March 2002, 31 July 2002, 31 July 2003, earlymid 2003

381

BL

2003, 19 September

Cuilcagh Mountain (North), Co. Fermanagh, Northern Ireland Cuilcagh Mountain (North), Co. Fermanagh, Northern Ireland Tooleyshaw Moor, South Pennines, northern England Wessenden Head Moor, South Pennines, northern England Tambille valley, near Chavin, Cordillera Blanca, Peru´ Annacarriga River, Slieve Bearnach, County Clare, Ireland Southern Mainland, Shetland Islands, Scotland

Slope instability and mass movements in peat deposits

Table 16.3.

382

Table 16.3 (continued ) Date

Location

Peat typea

Failure typeb

2003, 2 October

BL

PF

BL

PF

2004, midlate

Derrybrien, Slieve Aughty, County Galway, Ireland Derrybrien, Slieve Aughty, County Galway, Ireland Birk Sike, North Pennines, northern England

BL

PS

2004, midlate

Clough Head, Lake District, northern England

BL

PS

2003, 16 October

a b

Details

J. Warburton (pers. comm., 2004) J. Warburton (pers. comm., 2004)

BL ¼ blanket bog, RB ¼ raised bog, FP ¼ fen peat (f ¼ fen, b ¼ basin bog: after Hobbs, 1986). (see text) BB ¼ bog burst, BF ¼ bogﬂow, BS ¼ bog slide, PS ¼ peat slide, PF ¼ peat ﬂow.

A.P. Dykes and K.J. Kirk

Slope instability and mass movements in peat deposits

383

bog burst, bog slide, and peat slide. These were used interchangeably and thus could not be relied on to identify the nature of the failure(s) being reported. Indeed, the failures in the Yellow River catchment, County Leitrim, RoI, in June 1986, were described as two peat slides by Coxon et al. (1989) and as ‘‘a bog burst’’ that ‘‘ y took the form of three separate slides’’ (Large, 1991, p. 356). To confuse matters further, Hungr and Evans (1985) reported a peat ﬂow. The earlier, more prosaic term ‘debacle’, used in Ireland, had a speciﬁc meaning: bursting forth (Kinahan, 1897a). If different types of peat failures are to be reliably identiﬁed, then the basis for any classiﬁcation should ideally be to consider the type of peat deposit involved, the morphology of the failure, the material that has failed, and the mechanism of failure. Dykes and Kirk (2001) proposed an interim use of terminology to describe different types of peat failures, explained below, in the absence of a more formal classiﬁcation scheme. (1) Bog bursts: failures of raised bogs (bog peat) involving the break-out of semiliquid highly humiﬁed basal peat. Several examples of bog bursts in Ireland are described by Sollas et al. (1897). In some cases, the bog surface was observed to suffer increasingly large and violent ripple or wave-like movements prior to bursting (Kapanihane Bog, County Limerick, RoI, 1697, Molyneux, 1697; bog near Dundrum, County Tipperary, RoI, 1788, Sollas et al., 1897) or swelled from within during the hours or days prior to bursting (Fairloch Moss, County Antrim, NI, 1835, Sollas et al., 1897), but most involved the break-out of the semiliquid peat from peat-cutting excavations, often vertical faces several meters high. The failed bogs were typically several meters thick initially, but following the failure they left a valley up to a few hundred meters wide, sometimes as much as 12 km long and several meters deep below the level of the remaining bog. The subsided margins of the excavated valleys tended to be characterized by large concentric crevasses, partially ﬁlled from below by semi-liquid peat displaced by the subsided blocks (Grifﬁth, 1821; Sollas et al., 1897). (2) Bogflows: failures of blanket bogs (that is, bog peat) dominated by semi-liquid highly humiﬁed basal peat ﬂowing downslope from within the source area. Although similar to bog bursts in nature, they are usually much smaller in terms of area and volume of failed bog and extent of deposits and downstream impacts. The reported bogﬂows occur on slopes of no more than about 51 covered with at least 1.5 m thickness of blanket peat, and usually include disturbance of less that 100,000 m3 of peat (Tomlinson, 1981; Hendrick, 1990; Alexander et al., 1986; Feehan and O’Donovan, 1996). The most consistent morphological characteristics are the subsided blocks or strips of intact peat, comprising the acrotelm and varying thicknesses of the catotelm, along the margins of the source areas, torn free from the in situ peat but left behind as the lower catotelm peat escaped from beneath. Usually these intact rafts form concentric features within arcuate heads of bogﬂow failures (Fig. 16.2a). Occasionally, rafts or blocks tens of meters in diameter may be displaced downslope intact, carried by the moving lower catotelm peat. (3) Bog slides: failures of blanket bogs (bog peat) dominated by sliding of intact peat from the source area over a failure surface within the lower part of the peat proﬁle. They tend to occur on moderate slopes (around 4151) covered with peat around 12 m thick, and involve a few hundred to a few tens of thousands of cubic meters

384

A.P. Dykes and K.J. Kirk

Figure 16.2. Examples of peat failure types. (a) Bogﬂow: Slieve Anierin, County Leitrim, RoI, a few months after the event in 1998. (b) Bog slide: Cuilcagh Mountain (E2 in Fig. 16.5), County Cavan, RoI, showing in situ basal peat originally up to 10 cm thick on the failure plane with stones exposed in 1993 by 14 months of post-failure erosion (photo courtesy J. Gunn). (c) Peat slide: Dooncarton Mountain, County Mayo, RoI, eight days after the event on 19 September 2003. (d) Peat ﬂow: Tambille Valley, northern Peru´, a few months after the event in 2003.

of peat (Colhoun, 1966; Wilson and Hegarty, 1993). These failures may produce morphologies similar to bogﬂows or to peat slides (see below). Any thickness of catotelm peat remaining in the source area is intact and in situ (Fig. 16.2b). (4) Peat slides: failures of blanket peat-covered hillslopes dominated by sliding of the entire peat mass from the source area over a failure surface controlled by the higher shear strength of the mineral substrate; that is, at the peat–mineral interface or within the substrate below the base of the peat (as distinct from bog slides). Selkirk (1996) reported peat slides on 321 slopes involving as much as 0.8 m of peat (Macquarie Island); at the other extreme the upper part of the Hart Hope peat slide (North Pennines, England) had a gradient of 51 (Warburton et al., 2003) and the average slope of the 2001 Wessenden Head Moor peat slide (South Pennines, England) was just 4.51. They usually involve a peat thickness of less than 1.5 m and a thickness of mineral substrate material of less that 0.5 m, and failure volumes only occasionally exceed a few thousand cubic meters. These failures tend to form clearly deﬁned scars with irregular margins and little failed material (perhaps only a few intact blocks of peat) left behind, revealing a mineral surface (Fig. 16.2c). Displaced rafts of peat may have a layer of mineral substrate material still attached.

Slope instability and mass movements in peat deposits

385

(5) Peat flow: this term can be used to refer to two speciﬁc failure scenarios that do not ﬁt the previous criteria. The ﬁrst is the failure of (valley) fen or fen–bog transitional peat deposits. Only three such examples are known (Peru´, Australia, and Malaysia; Table 16.1). The second is the development of a ﬂow failure in peat (any type) as a result of head-loading. The cause, mechanism, and morphology of these latter failures are different from bogﬂows as described above, and again only ﬁve examples are known (Peru´ and Canada: Table 16.1; Slieve Bearnach, County Clare, RoI, in 2003, and two at Derrybrien, County Galway, RoI, in 2003).

Causes of instability in peatlands Instability of hillslopes can be considered to result from the combined effects of preparatory factors and trigger factors (Crozier, 1986). Preparatory factors are site conditions that have developed so as to render a slope susceptible to failure under the inﬂuence of a speciﬁc trigger event, the latter normally being an external inﬂuence acting on the slope. Instability of peatlands can be examined using the same approach, but the issue is complicated by the need to consider both natural processes and the particular human activities associated with peatland management and exploitation. The development of reliable methods for assessing the potential hazard from peat mass movements will rely on accurate assessments of the inﬂuence of these human activities, not least because they can include both preparatory and trigger effects. The difﬁculty lies in the necessary reliance on published accounts of peat failures to address this issue. For each case in which one or more anthropogenic preparatory inﬂuences have been identiﬁed, a critical question arises: would this failure have occurred in the absence of the anthropogenic factor(s). Dykes and Kirk (2001) answered this question by means of a quantitative analysis, showing that the 1998 peat slide on Cuilcagh Mountain, NI, would probably not have occurred if a particular drainage ditch had not been cut a few years before. The dumping of peat spoil or other material onto in situ peat constitutes the most obvious anthropogenic trigger factor, the cause of the mass movement being shear failure of the in situ peat. The Canadian example (Table 16.1) involved up to 6 m thickness of peat being tipped onto 2–3 m thick blanket peat on 51 slopes over 4 days. The edge of the spoil tip failed, and the rapid undrained loading (sensu Hutchinson and Bhandari, 1971) of adjacent previously unaffected blanket peat by displaced spoil triggered a signiﬁcant peat ﬂow (Hungr and Evans, 1985). Figure 16.2d shows the recent example from 4200 m altitude in the Peruvian Andes, where a 5–6 m high gravel tip built on valley ﬂoor fen (or transitional) peat failed. The rotational shear failure of the downslope side of the tip extended up to 2 m into the underlying peat, resulting in a peat ﬂow on a 3o–41 down-valley gradient. The Derrybrien and Slieve Bernach failures (RoI) in 2003 involved similar scenarios. The other type of anthropogenic trigger, not reported since the turn of the 20th century in Ireland, was the manual extraction of peat from cut faces in raised bogs. In some cases, these cut faces were up to 9 m high (bog of Kilmaleady, King’s County, RoI, 1821, and bog near Dunmore, County Galway, RoI, 1873) although

386

A.P. Dykes and K.J. Kirk

the 1.5–3 m deep cutting at Knocknageecha Bog, County Kerry, RoI, 1896, was perhaps more typical (Sollas et al., 1897). The cause of these bog bursts was the removal of cut peat that also removed the lateral and conﬁning support for the dominantly semi-liquid lower layers of peat, allowing this material to break out due to high water pressure within the bog. Such failures are now highly unlikely in the British Isles as there are no raised bogs of any size still subject to peat cutting in this manner. In most of the reported peat mass movements, failure was triggered by a naturally occurring heavy rainfall event either involving very high intensities (Carling, 1986; Acreman, 1991) or large aggregated depths of rainfall over a few days (Wilson and Hegarty, 1993; Hendrick, 1990). One exception is the 1984 Straduff Townland bogﬂow, County Sligo, Ireland, that occurred after just 25 mm of rain, spread over three days, fell onto blanket peat that may have still contained cracks formed during the preceding very dry summer (Alexander et al., 1986). Very dry antecedent moisture conditions have been implicated in several other previous reports (Carling, 1986; Acreman, 1991; Wilson and Hegarty, 1993). The primary role of rainwater is to increase the pore-water pressure acting on a potential failure plane within or below the lower part of the peat proﬁle. As porewater pressure increases, the effective shear strength of the peat or the mineral substrate is reduced as the water pressure supports some of the weight of the material above, reducing the frictional resistance to shearing. The difﬁculty of explaining many peat failures is to account for the rapid transfer of excess water to the base of a largely impermeable, saturated, peat proﬁle (Warburton et al., 2004). In some cases desiccation cracks or tine cuts (vertical cuts made by modern peat extraction equipment) probably provide the required route for the rainwater, and pre-forestry tunnel plowing (similar in effect to tine cutting) appears to have had the same effect in one example (Hendrick, 1990). Pre-failure tension cracks may have fulﬁlled the same role in others. Subsurface pipes (Fig. 16.3a) provide another mechanism for transferring rainwater into the base of a potential failure site (Carling, 1986; Dykes and Kirk, 2001; Warburton et al., 2003), but their precise roles are not yet certain. A potentially useful technique for locating and tracing the subsurface courses of peat pipes using ground-penetrating radar (GPR) has only recently been developed (Holden, 2004), although this cannot reveal the shapes or sizes of the pipes detected. Detailed research into the continuity of peat pipes is needed to assess whether they are more likely to contribute to peat failures by allowing high and possibly artesian (Carling, 1986) water pressures to develop where they give way to diffuse matrix ﬂow, or whether they help to maintain stability by providing effective under-drainage for the peat mass on a hillslope. The susceptibility of a slope to failure will arise from the geometric conﬁguration of the slope and its peat cover, and the nature and properties of the materials on the slope. Undercutting of a peat mass at the bottom of a slope by a stream (Delap and Mitchell, 1939) removes downslope support for the peat. A similar condition occurs at the edges of escarpments. Escarpment bogﬂows (Fig. 16.2a) have been reported by Mitchell (1935) and Alexander et al. (1985, 1986), both of whom identiﬁed a bank or wall of ﬁrm (drier) peat on the escarpment edge holding back the peat above. In all

Slope instability and mass movements in peat deposits

387

Figure 16.3. Features of blanket bogs associated with peat failures. (a) Natural pipe (indicated by black arrow) within clay substrate 40 cm below the base of the peat (pen for scale): Cuilcagh Mountain (N4 in Fig. 16.5), NI, a month after the event in 1998. (b) Vertical tine cuts (indicated by white arrows) exposed in a peat slide on Dooncarton Mountain, County Mayo, RoI, eight days after the event on 19 September 2003. (c) Degraded drainage ditch (indicated by white arrows) found to have been partially responsible for this peat slide: Cuilcagh Mountain, NI (N4 in Fig. 16.5) (P indicates the location of the pipe shown in a).

such cases, the morphological evidence indicates that failure began by loss of basal support at the escarpment edge as the ﬁrm lower wall of peat gave way, and propagated upslope by retrogressive unloading of the adjacent bog. In other cases, a less pronounced convex break of slope appears to have led to a similar scenario. Bishopp and Mitchell (1946) described a well-drained, ﬁrm margin of peat at a sharp break of slope between an upper slope of less than 61 and a lower slope of 121: the ﬁrm edge gave way to allow the failure to propagate upslope. The bogﬂow described by Tomlinson (1981) also involved a convex break of slope, with the steeper lower segment having a drier peat type and, being ﬁrmer throughout, a narrower failure scar. What is not clear from Tomlinson’s account is that the break of slope, although obvious in the ﬁeld, is very small, increasing from 41 to 5.51 downslope. Morphological evidence obtained in 2004 indicates that this failure also began at, or immediately above, the break of slope. Few examples of peat failure at concave breaks of slope have been reported (failure immediately above concave break, Wilson and Hegarty, 1993; failure across concave break, Mitchell, 1938; Hendrick, 1990; Walker and Gunn, 1993; failure immediately below concave break, Tomlinson and Gardiner, 1982), none of which suggest a clear causal link. It therefore appears that planar or convex slopes, or planar slopes with a convex break, are

388

A.P. Dykes and K.J. Kirk

the usual topographic locations of peat failures, some of which may also coincide with natural surface drainage lines or ﬂushes (Selkirk, 1996; Warburton et al., 2003, 2004). The presence of a particular substrate feature in some hillslopes has been identiﬁed as a possible contributory factor in the occurrence of peat slides. Acreman (1991, p. 175) reported that the failure scar surfaces of three peat slides in southern Scotland were ‘‘associated with an indurate ferruginous soil layer at 20 cm below peat/soil interface’’. A similar thin layer of brittle dark brown material a few millimeters thick was observed in the failure scar surfaces of some of the steeper peat slides on Macquarie Island, and analyzed to reveal it to be a humic gel complex known as dopplerite (Selkirk, 1996). Both these reports emphasize the probable role of the material as an impermeable layer impeding drainage and, thus, promoting failure at or below the base of the peat. Features corresponding with both Acreman’s indurated layer and Selkirk’s dopplerite have been found in several of the 2003 landslides on Dooncarton Mountain and older failures on other nearby hills in County Mayo, RoI. The precise nature of these features, and their roles in the occurrence of failures of thin blanket peat on steeper slopes, has yet to be quantitatively determined.

Anthropogenic causes Peat mass movements with anthropogenic triggers are rare, and have been discussed previously in this section to highlight the potential consequences of inappropriate human actions. Other anthropogenic causes of failures comprise activities that change the overall condition of the peat deposit so that it becomes more susceptible to failure if subjected to a natural trigger event, usually a high magnitude rainfall event. The bog burst at la Vraconne, Switzerland in 1988 during a 180 mm overnight rainfall event had been subjected to peat cutting from the 17th to mid-20th centuries (Feldmeyer-Christe and Mulhauser, 1994), and the collapse of the Wingecarribee Swamp in Australia in 1998 was initiated by the failure of an unsupported excavation face at the upstream side of a peat-mining dredge pool during a violent storm producing 300 mm of rainfall (O’Loughlin, 1998). Modern peat extraction in Ireland results in near-vertical tine cuts, parallel slits down through the bog roughly a meter apart, often well over a meter deep but barely a few centimeters wide. These have been implicated in a bogﬂow at Tullynascreen Townland, County Sligo, RoI (Alexander et al., 1985) and in two of the large peat slides on Dooncarton Mountain (Fig. 16.3b). Although the precise contribution the tine cuts made to the instability in both these cases is not yet certain, rapid accumulation of rainwater at the base of the peat and the loss of structural integrity of the peat mass normally provided by the tensile strength of the peat, particularly the acrotelm layer, require detailed investigation. Artiﬁcial drainage channels and boundary ditches have been identiﬁed as contributory factors in a number of peat mass movements. Ditches may have one of the two possible effects that could lead to a catastrophic failure. The ﬁrst possibility is that a ditch cut across a sloping bog eliminates downslope support for the bog above

Slope instability and mass movements in peat deposits

389

the ditch that would otherwise be provided by continuity of the peat mass. This was the case demonstrated by Dykes and Kirk (2001) using a modeling approach in the failure analysis (Fig. 16.3c). The 16 ha peat slide on the north side of Glencastle Hill, County Mayo, RoI, in 1867, was also attributed to the lack of support for the bog at the base of the slope due to a mearing (dividing) ditch having been cut (Kinahan, 1897a). A large bogﬂow on Slieve Bloom, County Laois, RoI in 1988 (Feehan and O’Donovan, 1996) may also have been inﬂuenced in this way. More commonly, ditches have been implicated as causal factors because they could have transferred critical additional volumes of storm runoff water into the failure zones either directly (Tomlinson and Gardiner, 1982; Wilson et al, 1996; Wilson and Hegarty, 1993) or indirectly through connecting natural pipes (Carling, 1986). Drains associated with plowing for forestry planning were thought to have directly contributed to one Irish failure (Hendrick, 1990).

Peat properties Any attempt to understand peat instability must take account of the bulk physical and geotechnical properties of a peat body as well as smaller-scale structures within the peat that result from the botanical and ecological development of the peat. The relationships between botanical composition of peat and its geotechnical properties were highlighted as a fundamental issue by Hobbs (1986), although there have been few studies undertaken to investigate them. The degree of humiﬁcation of peat is probably a key property of peat to which many of the other geotechnical properties are related. The von Post method for determining humiﬁcation in the ﬁeld remains acceptable to engineers. Carlsten (1993) advocates Landva et al.’s (1986) earlier view that the complete von Post classiﬁcation system for describing peat, as presented in English by Landva and Pheeney (1980) and Hobbs (1986), should be used by geotechnical engineers. This system allows properties of the peat having potential geotechnical implications such as wetness, and the content of ﬁbers, roots and/or woody fragments, to be recorded. The Radforth classiﬁcation of peat structure (Radforth, 1969) can be useful for describing peat with low ash contents (Landva et al., 1983), and a method for describing peat devised by Troels-Smith (1955) can be used with the von Post system to provide a more comprehensive description of the peat (Warburton et al., 2003). Fibers within peat can strongly inﬂuence the hydraulic conductivity and, more critically, the shear strength. Landva and La Rochelle (1983) suggest that the apparent cohesive strength of peat experiencing zero normal stress in their tests must be due to the entanglement of ﬁbers, but published values of peat shear strength are few, highly variable, and mostly do not relate to humiﬁed basal peat (Table 16.4). One of the reasons for the lack of published data is the difﬁculty of obtaining reliable test results from peat samples. Some relationships between properties will change during tests to determine, for example, unconﬁned compressive strength (Hanrahan, 1964) or indeed shear strength. The latter is particularly difﬁcult to resolve, given that the unit weight of saturated peat may be equivalent to, or even less than, that of water, so that there is effectively no normal stress on a potential shear surface within a peat

A.P. Dykes and K.J. Kirk

390 Table 16.4.

The range of published values for shear strength of peat (after Kirk, 2001).

Original source

Peat type/characteristics

Cohesion, c (kPa)

Hanrahan (1954)

Sphagnum peat with high moisture contents Peat with low moisture contents Remoulded H4 Sphagnum peat

high

Adams (1965) Hanrahan et al. (1967) Hollinshead and Raymond (1972) Landva and La Rochelle (1983) Marachi et al. (1983) Kirk (2001)

Sphagnum peat (H3, mainly ﬁbrous) Fibrous delta peat with low moisture contents Ombrotrophic blanket peat 35125 cm depths

Internal friction angle, f(1) 5

0 5.56.1 4.0

36.643.5 34

2.44.7

27.135.4

18.0 2.78.2

28 26.130.4

mass. This fact makes peat particularly susceptible to rotational shear failure under applied loads (Hanrahan, 1954). However, Hanrahan (1954) reported the shear strength of peat to be entirely cohesive. MacFarlane (1969) contradicted this, citing work by Adams (1965) that showed conclusively that peat is essentially a frictional material, suitable for effective stress analysis. It is difﬁcult to draw any conclusions regarding the shear strength of peat because there is no consistency of peat types or testing procedures between the published examples. The second reason for the scarcity of relevant published shear strength data for peat is that engineers have been concerned with peat under applied loads, such as road foundations or embankments. Geotechnical properties of in-situ peat under natural conditions have only recently started to be investigated with respect to environmental (rather than engineering) instability (Kirk, 2001; Yang and Dykes, unpublished data). The use of a shear vane to measure in-situ undrained shear strengths of peat is common engineering practice but suffers from the uncertainty of the inﬂuence of peat ﬁbers. A detailed study by Landva (1980) concluded that vane shear tests of peat are likely to give misleading indications of the shear strength and are of little engineering use. Helenelund (1967) had earlier found that the tensile strength of ﬁbrous peat may provide a better estimate of the shear strength than any vane test results, although the nature of any relationship between tensile strength and shear strength remains uncertain. Notwithstanding the demonstrated unreliability of the absolute shear strength values obtained from vane tests, these tests can possibly be used to indicate patterns of relative strength variations with depth through a peat deposit (MacFarlane, 1969; Kirk, 2001). In blanket peat up to 3 m deep on the eastern slopes of Cuilcagh Mountain, County Cavan, Ireland, Kirk (2001) found that the undrained shear strength appeared to increase with depth below the surface, as did the degree of humiﬁcation. The general pattern of depth variations appears counter-intuitive given the nature of bog slide and bogﬂow failures in which the lower peat layers clearly failed, and the fact that more humiﬁed peat might be expected to contain fewer residual ﬁbers. Increasing humiﬁcation had previously been shown to correspond

Slope instability and mass movements in peat deposits

391

with lower water contents and lower liquid limits (Hobbs, 1986), higher dry bulk densities and lower hydraulic conductivities (Boelter, 1969), implying that the higher strengths are associated with the decrease in particle size and thus closer packing of the organic particles and the consequently smaller pores and higher bulk density (Boelter, 1969). The lower porosity and liquid limit may cause the highly humiﬁed basal catotelm to be susceptible to failure because it would take a smaller increase in water content to bring about signiﬁcantly higher water pressures or even a change of state. On the other hand, any contrast in shear strengths between layers could control the occurrence and position of shear failure above the base of the peat proﬁle (Delap and Mitchell, 1939). Hydrological properties and characteristics of peat that have actual or potential inﬂuences on the overall stability of peat have been examined by Warburton et al. (2004), and the general hydrological behavior of peat bogs is discussed by Holden et al. (2006 – this book, Ch. 22). The hydrological, geotechnical, and other physical properties of peat such as water content, ash content, bulk density, and speciﬁc gravity, can vary considerably in three dimensions at the scale of a few centimeters as a consequence of spatial and temporal changes in the constituent botanical assemblages (Tallis, 2001) and the environmental conditions inﬂuencing the decomposition process. This often makes it very difﬁcult to determine the actual physical characteristics of peat that has failed. However, characterizing failed peat is a necessary line of investigation in order to determine whether particular combinations of peat properties might indicate a susceptibility to failure. It is also necessary to understand the nature and characteristics of the overall peat mass, rather than simply the peat material, in a manner analogous to the study of slope instability in hard rock masses (Selby, 1993). In other words, the types, distributions, and properties of structural features within peat bodies may be more important for failure than the peat itself. Structures such as desiccation cracks, tension cracks, tine cuts, or subsurface pipes were highlighted previously as potential (preparatory) hydrological causes of failure. They also provide structural weaknesses through the peat mass, thus reducing its overall in-situ strength. Other structures derive from the formation of the peat. Landva and Pheeney (1980) described aelotropic (banded) peat having discrete horizontal layers that are easily pulled apart vertically but possessing high horizontal tensile strengths. Another structural feature observed in exposed blanket peat proﬁles at several Irish failures, during ﬁeld investigations in 2004, was the presence of thin (about 1 cm) and irregular cracks, lenses, or (slightly thicker) pockets of gray slurry contained or conﬁned within the lower 30–50 cm of the catotelm. The slurry was typically around 96–98% organic (loss on ignition at 550 1C for 3 h: Skempton and Petley, 1970; Hobbs, 1986) with a ‘water content’ 41000% – the water content of peat is normally determined as the ‘mass of water per unit mass of oven-dry solids’, then multiplied by 100 and reported as a percentage (see glossary in this book for example of calculation). The three-dimensional continuity of these structures is not known but they are associated with unusual gray colored peat when freshly exposed and the peat mass breaks apart with no resistance along them. The exact nature and implications of these and other structures for peat instability have yet to be quantitatively investigated.

392

A.P. Dykes and K.J. Kirk

Failure mechanisms Peat slides and bog slides occur when the shear strength of a particular layer of material within a peat-covered hillslope, or the shear strength provided by the contact plane between two layers, is exceeded by the disturbing forces acting on the slope in the same way as for any other shallow translational landslide (Nash, 1987). The position of the shear surface will usually be the shallowest layer at which this condition arises, and can often be associated with distinct characteristics of the peat (or underlying substrate) proﬁles. The peat slides at Slieve-an-Orra, NI, 1980, failed at the interface between a sandy, organic rich till and the clayey till beneath, up to 30 cm below the base of the peat (Tomlinson and Gardiner, 1982) and the peat slides on the north side of Cuilcagh Mountain, NI, in 1998 (Dykes and Kirk, 2000, 2001) and 2000 failed up to 0.5 m below the peat at the interface between a pale clayey till and a dark clay derived from in-situ weathered shale. Carling (1986) noted peat slides in the North Pennines, England, 1983, to have failed within a single clay layer around 20 cm below the peat where the clay was not reinforced by roots from the original peat vegetation. In these cases, the shear failure is partially controlled by effective stresses imposed by the mineral substrate. In bog slides, by contrast, shear failure occurs within the peat which would be assumed to be saturated and, thus, almost buoyant in many cases. Failure must therefore be attributed to a structural weakness within the peat deposit. At Slieve Rushen, County Cavan, RoI, 1965, Colhoun (1966, p. 206) observed that ‘‘the upper surface of the [black amorphous] bottom peat seems to have acted as a slip plane over which the blocks of ﬁbrous peat moved’’. A similar ‘‘discontinuity between the basal peat and overlying peat’’ was found by Wilson and Hegarty (1993, p. 599) in the Skerry Hill slides, NI, 1991. The Conaghra bog slide, County Mayo, RoI, 1986, involved only the upper 50 cm of peat (effectively the acrotelm) sliding over in-situ catotelm across a 5 ha area of 2.51–4.01 slope. This suggests that a distinct plane of weakness should be found if the peat proﬁle at the latter failure is examined. The head-loaded peat ﬂows also appear to have begun as shear failures under the loads generated by the spoil tips, but with the peat structure breaking down to form a ﬂow as the undrained loading effects propagated downslope ahead of and because of the progressive loss of strength, similar perhaps to the development of mudﬂows due to remolding of clays with high water contents along a shearing surface (Selby, 1993). The failure mechanisms in bogﬂows and bog bursts are less clear, not least because there has never been any quantitative demonstration of the semi-ﬂuid state of catotelm peat reported in many accounts of these types of failure. Sollas et al. (1897) were certain that the large bog bursts involved liquid catotelm breaking through the conﬁning ﬁbrous acrotelm crust (or breaking out from peat-cutting excavations that penetrated the lower layers of the bogs) due to accumulation of excess water within the bogs from heavy rainfall or, as in the case of Fairloch Moss, from an inﬂux of subterranean water. Hemingway and Sledge (1946, p. 282) concluded that 46 mm of rain in 3 h had been sufﬁcient to raise the pressure of the basal watery peat within Danby Low Moor bog (North Yorkshire Moors, England) to the point at which it generated ‘‘an upburst of violence sufﬁcient to throw back and overturn the adjacent crustal peat’’.

Slope instability and mass movements in peat deposits

393

Some bogﬂows were also reported to have involved liquid peat escaping from beneath the acrotelm, the latter tearing and subsiding as a result (Delap et al., 1932; Mitchell, 1935; Hendrick, 1990). Kirk (2001) described a slope on the east side of Cuilcagh Mountain, County Cavan, RoI, that has been deformed by downslope movement of the catotelm peat beneath an acrotelm layer that was clearly strong enough to withstand the disturbance (Fig. 16.4). The sizes, patterns, and extent of the tension cracks between subsided strips of acrotelm at the head of this failure seem to indicate movement of the catotelm, which must therefore be interpreted to have been at least semi-ﬂuid. The primary failure mechanism may turn out to be collapse of the peat matrix structure due to disturbance, described by Bishopp and Mitchell (1946, p. 153) as ‘‘some process analogous to a reversal of phase’’, the very high water content and known sensitivity of amorphous peat (MacFarlane, 1969) resulting in behavior similar to that of quick clays (Selby, 1993). The key research question is to identify the cause(s) and nature of the disturbance. This could simply be a critical increase of the pore water pressure causing water to be released from plant cell structures into adjacent voids (Bell, 2000) at the micro-scale and/or mechanical disaggregation of the peat mass at the larger scale as the liquid limit is exceeded and pore water pressures exceeding the hydrostatic pressure start to support the weight of the material above (Selby, 1993). The possible hydrological processes that might be associated with this failure mechanism have been reviewed and summarized by Warburton et al. (2004). The other disturbance that might cause this loss of strength and ﬂow development is the initiation of shear failure. This mechanism of initiation of ﬂow failures is well known from mass movements in clay or other mineral sediments (Hungr et al., 2001). It begins for the same reason that shearing begins in bog slides and peat slides, but

Figure 16.4. Proﬁle of the bulging slope on the east side of Cuilcagh Mountain, RoI (BS in Fig. 16.5). The labeled positions identify proﬁle data presented and discussed by Kirk (2001).

394

A.P. Dykes and K.J. Kirk

develops differently. If the shearing is rapid relative to the permeability of the peat matrix (a condition known as undrained shearing) the associated micro-scale deformations are likely to generate locally enhanced pore water pressures and loss of peat matrix structure (remolding) resulting in catastrophic local loss of strength. The extent of this liqueﬁed zone tends to increase with a positive feedback effect. In many of the recorded bog bursts and bogﬂows, the location at which the bog ﬁrst gave way was the lower margin (raised bog) or downslope end of the failure, the characteristic morphologies resulting from semi-ﬂuid catotelm peat ﬂowing toward that point and dragging intact blocks of acrotelm (sometimes with upper catotelm) peat downslope with it (Fig. 16.2a). The distinction from peat ﬂows (as identiﬁed earlier) is that in the latter, the ﬂow results from remolding under shear in response to undrained loading from upslope. A ﬁnal possible explanation for the outﬂow of peat slurry from some Irish bog bursts is that the central part of a raised bog may have accumulated over a spring, and that an extensive body of peat fragments in suspension with no development of cohesion or ﬁssile structure may have been overgrown and effectively entombed by the encroachment of normally developing bog (Kinahan, 1897a). Kinahan suggested that if a natural sub-peat drainage path from these (semi) liquid interiors became blocked, increased water pressure from the continuing inﬂow would eventually cause the bog to burst. Thus, it seems that the catotelm peat within some raised bogs may have never been solid or ﬁrm as is normally the case. There are no data to indicate how common this scenario might be in peatlands globally.

Case example of blanket bog instability: Cuilcagh Mountain, Ireland Cuilcagh Mountain straddles the border between NI (County Fermanagh) and the RoI (County Cavan). The middle slopes (ca. 300–500 m elevation) of the northern and eastern sides of the mountain comprise typical upland blanket bog environments in varying conditions from pristine to badly degraded, a reﬂection of the history and land use of the area. An extensive initial phase of research (Kirk, 2001) established that these 24 km2 of blanket bog have experienced at least 47 mass movements (Fig. 16.5). Table 16.5 summarizes the characteristics and timings of the ﬁve groups of failures indicated in Figure 16.5. The failure dates are based on available 1:10,000 scale aerial photographs taken in 1981, 1989, 1991, and 1995, and ﬁeld surveys during 1997–2005. Average failure depths are based on ﬁeld observations with an estimated allowance for post-failure shrinkage of the peat around the scar margins. Average volumes are based on scar dimensions obtained from ground surveys, air photographs, and the estimated failure depths. Those volumes include peat and mineral substrate in most cases (see below). Scar gradients are the averages of the upper parts of the scars of four failures from each group. The measured North and Northwest scars were all convex, but three of the measured East failures occurred around sharp concave breaks of slope. It is clear that failures of deeper peat on the less steep eastern slopes involved much larger volumes of material. These East failures were originally identiﬁed as bogﬂows largely due to the presence of extensive

Slope instability and mass movements in peat deposits

395

Figure 16.5. Peat failures on Cuilcagh Mountain, Ireland, up to April 2005 (after Kirk, 2001).

Table 16.5. Frequency and general characteristics of blanket bog failure on Cuilcagh Mountain (after Kirk, 2001). See text for explanations of the data. Failure group N NW E S NE a b

Failure frequency according to known age ranges Pre-1981

1981–1989

12 7 3

7(+2a) 4(+1a)

Pre-1989

3 5 1

1992-2000 1(2)b 1 3

Mean depth (m)

Mean volume (m3)

Mean scar gradient (1)

1.25 1.25 2.0 1.5 1.75

4,550 3,325 13,775 3,450 35,650

7.5 8.5 5.25

These failures did not create new scars, they extended previous scars. Although a single failure, the 1998 peat slide created two separate distinct scars.

crescentic tension cracks around parts of the scars, some deposits of peat slurry within parts of the scars, and the extensive downstream deposits of peat transported by ﬂoodwater from the 1992 failure and surveyed within a few days of the event (Walker and Gunn, 1993). However, more recent ﬁeld surveys of peat failures throughout Ireland and access to previously unknown photographs of the 1992 failure (Fig. 16.2b) suggest that most of the East Cuilcagh failures may in fact have been bog slides (according to the deﬁnition presented earlier): the blocks and strips of

A.P. Dykes and K.J. Kirk

396

peat between the crescentic tension cracks have generally not subsided greatly relative to adjacent intact bog (subsidence may be more due to shrinkage associated with post-failure drying and oxidation than to loss of a thickness of catotelm), extensive parts of the source areas show complete removal of the peat with few remaining blocks or rafts, and many of those blocks that remain comprise the full thickness of the peat and have been tipped onto their sides rather than ﬂoated upright on catotelm slurry. Correct identiﬁcation of the type of failure is important if patterns of failures are to be correctly interpreted, although this remains difﬁcult in a few cases (see below). Nevertheless, it appears probable that most of the failures on Cuilcagh Mountain simply involved shearing of the weakest plane within the peat/ substrate proﬁle relative to the forces acting at each site. In the 1981 air photographs, 27 peat failures were identiﬁed. Eighteen of these were well vegetated, some having new peat accumulations occupying the scar surfaces. The other nine failures appeared relatively fresh and were therefore assumed to be no more than a few years old by that time. Between 1981 and 1991, 18 further peat failures occurred, one of which may have involved a ﬂow component although the recent re-interpretations suggest probably not. Most of the others involved shallow failures of the clay beneath the peat on steeper (101–181) slopes with the entire depth of peat (ca. 0.7 m) being removed. Between 1991 and 2001 ﬁve failures occurred, each signiﬁcant for a different reason. No failures have occurred in the E, N or NW areas between March 2000 and April 2005. Table 16.5 shows this pattern of peat loss and sediment production dominated by frequent small failures shortly before and during the 1980s. Almost 20 ha of mostly high-quality blanket peat have been lost in total, which amounts to about 0.8% of the peat on the northern and eastern sides of Cuilcagh Mountain (Table 16.6). The higher frequency of failure events during the 1980s compared with the 1990s are thought to be associated with increased sheep stocks and uncontrolled burning on the steeper northern slopes that degraded the bog by disrupting the acrotelm and living vegetation (Gunn, 2000), although the exact nature of this indirect link remains unclear. Table 16.6. Sediment production and loss of blanket bog (including some of the underlying clay) from peat failures on Cuilcagh Mountain. Time period

Number of failures

Total area of failed bog (ha)

Mean failure area (ha)

Total volume of sediment produced (m3)

Mean failure volume (m3)

Pre-1981 (vegetated by 1981) Pre-1981 (bare scar in 1981) 1981–1991 1991–2000 Total

18

10.5

0.58

172,325

9575

9

2.3

0.25

35,350

3925

18 5 50

3.9 2.8 19.5

0.22 0.57

54,675 45,225 307,575

3050 9050

Source: Kirk (2001).

Slope instability and mass movements in peat deposits

397

The ﬁve failures of the 1990s highlight different issues relating to the occurrence of peat mass movements. The bog slide of August 1992 (E2 in Fig. 16.5; Fig. 16.6a) was triggered by a summer rainfall event producing 57 mm of rainfall within 18 h. This failure is somewhat unusual in that it has a steep (81) planar head scar from which the entire peat mass moved by sliding over a shear surface within the lower catotelm (Fig. 16.2b), then a main failure zone of just 2.51 between the head scar and a stream, the upper part of which extends into a gentle depression to the side of the head scar and displays an extensive area of crescentic tears separated by strips or rafts of intact peat. The peat slurry observed at the site was probably produced by remolding of the

Figure 16.6. Recent peat failures on Cuilcagh Mountain. (a) Bog slide of August 1992, a month after the event in 1992 (photo courtesy J. Gunn). (b) Bog slide of c. 1997 seen in May 2000. (c) Bog slide of c.1998 (E1 in Fig. 16.5) showing the ‘landslide dam’ caused by sliding of peat from the direction indicated by the arrow, photographed September 2004. (d) Peat slide of October 1998 a month after the event. (e) Peat slide of 2 March 2000 showing level head of scar comprising a dark grey clay surface and the thickness of peat and pale clay lost from the slope (the broken white line indicates the base of the peat), seen 10 days after the event.

398

A.P. Dykes and K.J. Kirk

lower catotelm as it slid (as described by Bishopp and Mitchell, 1946), rather than resulting from a ﬂow-type failure. The cause of the failure was probably the ﬂooded state of the stream, although it is not certain whether the stream eroded its bank to unload the toe of the peat slope above, or whether a small initial movement created a temporary dam that forced the impounded stream water into basal discontinuities in the peat and effectively lifted the main part of the peat blanket free of its base. The pattern of tension cracks around the scar shows that the peat slid from the steep head scar following the initial movement. This failure supplied several thousand cubic meters of peat into a ﬂooded stream that fed into the tourist show caves at Marble Arch Cave a few kilometers downstream. Even in early 2005, fragments of peat remain on stalactites near the roof of the cave passage in some places. This highlights a speciﬁc hazard at this location (fortunately the event occurred at 0400 h on a Sunday morning) and raises another: large volumes of failed peat entering watercourses can have serious consequences for water managers and for the natural ecology of affected streams, rivers and lakes (McCahon et al., 1987; Wilson et al., 1996). The failure of the mid-late 1990s, estimated to date from around 1997 based on its condition when ﬁrst discovered in early 1998, also involved the movement of peat from a steeper slope onto an almost level bog-ﬁlled valley course. It remains uncertain whether this was a bog slide or a bogﬂow as it displayed morphological evidence consistent with both types of failure; indeed, it may have involved both mechanisms. The affected peat further downslope was sufﬁciently competent that the valley bog caused the failure to remain conﬁned within a relatively small area (approx. 120 70 m) with the displaced peat thrusting under the in-situ peat around the toe margin (E6 in Fig. 16.5; Fig. 16.6b). The bog slide of ca. 1998 (E1 in Fig. 16.5; Fig. 16.6c) was similar but the tributary valley below the failure contained a stream channel. The opposite bank of the stream appears to have provided sufﬁcient resistance to cause the displaced peat to override it and come to rest, thus completely inﬁlling the original course of the stream channel over a 75 m length. This constitutes a landslide dam that could survive indeﬁnitely, as the stream has begun to develop an overﬂow course around the toe of the 8000-tonne slide mass. The volume of impounded water is only a few tens of cubic meters, so the potential hazard from the present situation is extremely low. The peat slide of 24/25 October 1998 (N4 in Fig. 16.5; Fig. 16.6d), triggered by a 60–90 mm rainfall event with a peak 60-minute intensity of up to 20 mm h1, comprised a small initial failure attributed to natural pipes and a drainage ditch (Dykes and Kirk, 2001) the displaced peat from which, sliding downslope over thin in-situ blanket bog, immediately caused a larger failure 100 m further downslope where the gradient reduced and the peat became thicker (Dykes and Kirk, 2000). The peat slide of 2 March 2000 (NW10 in Fig. 16.5; Fig. 16.6e) appears to have begun where a 12 cm diameter PVC water pipe crossed the slope in a shallow trench 30–40 m downslope from the upper margin. In 1998, the local people had to repair the pipe at this location having been broken by small-scale movement of the peat, suggesting a progressive weakening of the slope leading to failure during a large storm event (Kirk, 2001). Conventional methods for analyzing the stability of slopes were applied to several of the Cuilcagh peat failures in order to assess the suitability of this approach. Carling (1986) and Warburton et al. (2003) had previously used the inﬁnite slope

Slope instability and mass movements in peat deposits

399

model (Skempton and DeLory, 1957) to analyze the stability of peat slides, whereas Dykes and Kirk (2001) used the SLOPE/W software (Geo-Slope International Ltd.) to investigate the inﬂuences of ditches and natural pipes on the stability of the 1998 Cuilcagh peat slide (N4 in Fig. 16.5). This software can analyze slopes using most of the standard methods of limit equilibrium stability analysis (see Nash, 1987, for a review of these), but it does not include the inﬁnite slope model as an option. However, SLOPE/W can be used to analyze inﬁnite slope problems using the Ordinary Method of Slices and the more rigorous Morgenstern-Price Method (Geo-Slope, 1995). The inﬁnite slope model analyses a single slice of a two-dimensional slope proﬁle and assumes that every other part of the slope has the same characteristics. It requires six input parameters: slope gradient and depth of failure surface below ground level (geometry), cohesion and internal friction angle of the failed material (shear strength), height of water table above failure surface which determines the pore water pressure in the shear zone, and the average unit weight (bulk density 9.81) of the material above the failure surface. SLOPE/W requires the same input parameters but can represent the details of the entire two-dimensional slope proﬁle, representing each different layer of material (acrotelm peat, catotelm peat, sub-peat clay) explicitly, and as an alternative to specifying the water table position the distribution of pore water pressures throughout the slope can be imported from a prior ﬁniteelement hydrological analysis of the slope using the companion SEEP/W software. Table 16.7 shows the parameter values used to analyze the stability (with respect to shear failure) of ten of the Cuilcagh peat failures and the bulging slope referred to previously in this chapter. The pore water pressures for the SLOPE/W analyses were obtained from hydrological simulations using SEEP/W (Kirk, 2001); for the inﬁnite slope model, the water table was speciﬁed as being at the ground surface. The inﬁnite slope model also used the shear strength of catotelm peat for the E and BS failures, and the shear strength of the underlying clay for the N and NW failures. The results (Table 16.8) are presented in terms of the Factor of Safety, F, which represents the ratio of disturbing forces to resisting forces. If Fo1.0, the analysis indicates that a slope should have (or should be expected to) become unstable under the conditions analyzed. A value of F41.0 suggests that failure should not be expected, although values only slightly above 1.0 are interpreted as showing a slope that might require only a small change in its conditions (relative to those analyzed) to become unstable. For most of the failures analyzed, the results from SLOPE/W indicated that the failure position with the lowest F value corresponded with the actual failure position assumed on the basis of ﬁeld evidence, as would be expected. The large and unknown variability of actual conditions within and throughout the failed slopes can account for the fact that none of the calculated values of F wereo1.0, but most of the F values are sufﬁciently close to 1.0 to indicate marginally stable conditions (Crozier, 1986). Results from the inﬁnite slope model appear to be rather less consistent than the SLOPE/W results; it appears that the assumption of ‘inﬁnite slope’ conditions may be too much of a simpliﬁcation for these peat-covered slopes. This may be exacerbated in slopes where the average unit weight of the slope materials above the failure surface (the peat) is lower than that of water. The results in Table 16.8 do suggest that, in general, standard methods for stability analysis

400

Table 16.7. Parameter values used to analyse the pre-failure stability of peat-covered slopes on Cuilcagh Mountain (after Kirk, 2001). Failure pressure (Fig. 16.5) E2 E3 E6 E7 BS N1 N15 N19 NW5 NW6 NW10

Slope geometry

Acrotelm properties

Catotelm properties

Pore water along failure surface

b (o )

z (m)

L (m)

c0 (kPa)

f0 (o)

gs (kN/m3)

c0 (kPa)

f0 (o)

gs (kN/m3)

Min.uw (kPa)

Max.uw (kPa)

7–8 3–4 5–6 4–5 4–10 6–7 7–8 9 6–12 10 10

2.0 2.0 2.0 2.0 1.5–2.7 1.5 1.0 1.0 1.5 1.5 0.7

105 425 110 105 90 155 90 65 230 40 150

4.2

30.8

9.8

2.6

30.4

9.9

3.2

30.4

9.8

4.0

28.8

9.9

18 18 20 20 2 14 8 6 12 10 4

20 20 20 20 26 20 10 10 14 14 6

A.P. Dykes and K.J. Kirk

Notes:1. b ¼ slope gradient, z ¼ depth of failure surface, L ¼ length of failure, c0 ¼ cohesion, f0 ¼ angle of internal friction, gs ¼ unit weight (saturated in all these cases), uw ¼ pore water pressure. 2. All E/BS failures have the same acrotelm/catotelm properties, as do all N/NW failures. 3. All failures are assumed to have underlying clay properties c0 ¼ 2.8 kPa, f0 ¼ 26.51, gs ¼ 16.5 kN/m3 (Dykes and Kirk, 2001).

Slope instability and mass movements in peat deposits

401

Table 16.8. Factor of safety obtained from stability analyses of peat failures on Cuilcagh Mountain (Kirk, 2001). Failure (Fig. 16.5)

E2 E3 E6 E7 BS N1 N15 N19 NW5 NW6 NW10 a

Value of F obtained from SLOPE/W for a failure surface located

Within basal catotelm

At peat/clay interface

Within the clay

1.6a 2.0a 1.9a 1.3a 1.3a 1.9 2.5 2.1 2.1 2.5 4.5

1.8 2.1 1.9 1.2 1.3 1.5a 1.8 2.3 1.5a 2.1 3.9

1.8 2.3 2.2 1.4 1.4 1.2 1.5a 1.4a 1.3 1.4a 1.5a

Value of F obtained using the Inﬁnite Slope Model for failure at the assumed actual failure surface at its average depth

1.1 2.2 1.1 1.7 1.3 2.4 2.3 2.0 1.3 1.3 2.4

Assumed position of actual failure surface.

might be successfully applied to peat instability problems provided realistic values for the input parameters are available. There are several possible reasons why the values of F are all too high, which reﬂect many of the issues highlighted previously in this chapter. Firstly, the shear strength values obtained from ﬁeld samples for use in these analyses (Kirk, 2001) may not be fully representative of the actual strength conditions. Secondly, the stability models assume that each material is uniform and homogeneous throughout, which must be considered highly unlikely in reality. Thirdly, the possibility that at least some of the failures were associated with artesian water pressures within subsurface pipes has not been examined in this study (but was demonstrated as being critical for the 1998 peat slide (N4 in Fig. 16.5) by Dykes and Kirk, 2001). Finally, failure mechanisms other than rapid ﬁrst-time shearing cannot always be represented in these analyses, even if appropriate data are available. The two possibilities at Cuilcagh are that plastic deformation and creep of basal catotelm peat preceded failure at reduced strength conditions, or that weathering and progressive failure of the underlying clay, possibly involving micro-shearing, led to failure at strength conditions reduced to residual in places (Carling, 1986; Selby, 1993).

Future research directions The potential impacts of peat failures have been highlighted in recent years as including damage to properties and infrastructure (houses and roads at Dooncarton

402

A.P. Dykes and K.J. Kirk

and Shetland; reservoir at Wingecarribee Swamp), water resources (contaminated streams and lakes at Derrybrien and Slieve Bernach), forestry (Derrybrien) and farming (livestock losses at Dooncarton). It is fortunate that no tourists were killed by the 1992 bog slide at East Cuilcagh (Walker and Gunn, 1993) or the 1988 bogﬂow at Slieve Bloom, County Laois, RoI (Feehan and O’Donovan, 1996). From a management perspective, however, the need is to be able to assess the possible hazard from peat instability that may arise as a result of (1) the exceedance of natural intrinsic thresholds of peat accumulation and development, (2) the more frequent incidence of high intensity rainfall events predicted by climate change models for the UK and Ireland, and/or (3) inappropriate utilization or management of peatlands by individuals, local communities, or commercial enterprises. Conventional slope stability analyses have been used in a few previous studies of peat failures (Carling, 1986; Dykes and Kirk, 2001; Warburton et al., 2003), though perhaps with some degree of uncertainty. On the basis of the initial research at Cuilcagh, there appears to be no reason in principle why this approach should not work for peat instability assessment and analysis, provided accurate parameter values and ground conditions can be determined, to correctly describe peat strength and pore water pressure in the analysis. In some cases, back-analysis of failures that have occurred might provide realistic estimates of ﬁeld conditions at the time of failure, if enough is known about the intrinsic variability of peat properties. However, for the most part recent research has served to identify the questions that need to be answered, there being as yet insufﬁcient data from which to construct quantitative models of peat failure conditions. Several areas of future research can be identiﬁed, some of which have been discussed earlier this chapter, and are summarized below. (1) Peat hydrology is examined in detail by Holden et al. (2006 – this book, Ch. 22), but the speciﬁc research issues relevant to peat instability identiﬁed by Warburton et al. (2004) include: (a) the precise hydrological roles of pipes in and below the peat, (b) the coupling between water in the peat and in the underlying substrate, and (c) the rapid transmission of rainwater directly to the base of peat proﬁle, particularly in relation to desiccation shrinkage and cracking during dry periods. (2) Geotechnical properties of peat relevant to road-building or other construction activities are reasonably well known (Hobbs, 1986; Carlsten, 1993; Bell, 2000). Those relating to natural slope instability, including shear strength and rheology and their relationships with other physical, chemical, hydrological, and biological/botanical properties of peat, are highly uncertain. The nature and inﬂuence of the tensile strength of peat, and of structural variations within the peat mass, must also be investigated. (3) Long-term changes to blanket peat deposits may signiﬁcantly control the incidence of failure in some cases. Notwithstanding the possible wider environmental changes that may affect peatlands as a consequence of global climate change, the peat mass will experience continual change as humiﬁcation proceeds, altering the properties of the peat and, thus, its behavior, but may also change by way of increasing in thickness as accumulation continues. Bower (1961) and Tallis (1964) suggested that loss of peat might be associated with a critical depth being attained

Slope instability and mass movements in peat deposits

403

such that its properties give rise to instability; such thresholds often similarly control instability of soil-covered hillslopes (Carson and Petley, 1970; Dykes, 2002). The combined effects of accumulation and humiﬁcation may directly determine the susceptibility of deep blanket bog to failure (bogﬂow or bog slide type) because of the progressive changes to the peat, which may also include the formation of pockets of gas by differential decomposition. Compression of the peat due to continued accumulation may trap this gas within pore spaces, having the same effect as high pore water pressures. Trapped gas, which may comprise up to 10% of the catotelm peat (Clymo, 1984), and changes in acidity have been suggested to affect the strength of peat at depth (Bishopp and Mitchell, 1946) but these are difﬁcult to determine in the ﬁeld. Long-term changes to properties of mineral substrates beneath thin blanket bog may strongly inﬂuence the occurrence of peat slides, but these factors lie outside the scope of this chapter and are addressed extensively by the wider landslides literature. (4) The role of peat failures as components of changing peatland environments has not been examined in this chapter as this topic has not thus far been systematically addressed. The frequency of failures on Cuilcagh Mountain perhaps highlighted the possible ecological importance of such events, although ecological and geomorphologic impacts have been investigated in some cases (McCahon et al., 1987; Wilson et al., 1996; Warburton et al., 2003). Recovery (revegetation) of peat failure scars has been reported occasionally (Large, 1991; Wilson and Hegarty, 1993) and systematically monitored at the site of the Swiss failure (Table 16.1) (Feldmeyer-Christe, 1995, 2002), but this topic also needs detailed research in the British Isles where most of the world’s peat failures occur within very small pockets of surviving peatland.

References Acreman, M., 1991. The ﬂood of July 25th 1983 on the Hermitage Water, Roxburghshire. Scot. Geogr. Mag. 107, 170–178. Adams, J.I., 1995. The engineering behaviour of a Canadian muskeg. Proceedings of 6th International Conference on Soil Mechanics and Foundation Engineering, Vol. 1, pp. 3–7. Alexander, R.W., Coxon, P., and Thorn, R., 1985. Bog ﬂows in south-east Sligo and south-west Leitrim. In: Thorn, R. (Ed.), Field Guide No. 8: Sligo and West Leitrim. Irish Assoc. Quat. Studies, pp. 58–80. Alexander, R.W., Coxon, P., and Thorn, R.H., 1986. A bog ﬂow at Straduff Townland, County Sligo. Proc. Roy. Irish. Acad. 86B, 107–119. Aston, B.C., 1909. The soils and soil-formers of the subantarctic islands. In: Chilton, C. (Ed.), The subantarctic islands of New Zealand. Government Printer, Wellington, Vol. 2, pp. 745–777. Bailey, A., 1879. A letter from Acting Governor Bailey to Governor Callaghan. Q. J. Geol. Soc. London, XXXV, 96–97. Barkly, A., 1887. A letter from the Lieutenant-Governor of the Falkland Islands, communicated by H.M. Secretary of State for the Colonies. Q. J. Geol. Soc. London XLII, 2 . Beder, S., 2001. The collapse of Wingecarribee wetlands. Web Site prepared on behalf of Robertson Environmental Protection Society (REPS): www.uow.edu.au/arts/sts/sbeder/wingecarribee/ (accessed 12 April 2005). Bell, F.G., 2000. Engineering Properties of Soil and Rocks. Blackwell, Oxford.

404

A.P. Dykes and K.J. Kirk

Bishopp, D.W. and Mitchell, G.F., 1946. On a recent bog-ﬂow in Meenacharvy Townland Co. Donegal. Sci. Proc. Roy. Dublin Soc. 24, 151–156. Boelter, D.H., 1969. Soil and water management and conservation. Physical properties of peats as related to degree of decomposition. Soil Sci. Soc. Am. Proc. 33, 606–609. Bower, M.M., 1961. The distribution of erosion in blanket peat bogs in the Pennines. Trans. Inst. Brit. Geogr. 29, 17–30. Campbell, I.B., 1981. Soil pattern of Campbell Island. NZ J. Sci. 24, 111–135. Carling, P.A., 1986. Peat slides in Teesdale and Weardale, northern Pennines, July 1983: Description and failure mechanisms. Earth Surf. Proc. Land 11, 193–206. Carlsten, P., 1993. Peat – geotechnical properties and up-to-date methods of design and construction. State-of-the-Art-Report. Swedish Geotec. Inst. Report No. 215, Linko¨ping. Carson, M.A. and Petley, D.J., 1970. The existence of threshold hillslopes in the denudation of the landscape. Trans. Inst. Brit. Geogr. 49, 71–93. Charman, D., 2002. Peatlands and Environmental Change. Wiley, Chichester. Clymo, R.S., 1984. Sphagnum-dominated peat bogs: A naturally acid ecosystem. Phil. Trans. Roy. Soc. London B 305, 487–499. Cole, G.A., 1897. The bog-slide of Knocknageeha, in the County of Kerry. Nature 55 (1420), 254–256. Colhoun, E.A., 1966. The debris-ﬂow at Glendalough, Co. Wicklow and the bog-ﬂow at Slieve Rushen, Co. Cavan January, 1965. The Irish Naturalist 15, 199–206. Coxon, P., Coxon, C.E., and Thorn, R.H., 1989. The Yellow River (County Leitrim, Ireland) ﬂash ﬂood of June 1986. In: Beven, K. and Carling, P. (Eds), Floods: Hydrological, Sedimentological and Geomorphological Implications. Wiley, Chichester, pp. 199–217. Crozier, M.J., 1986. Landslides: Causes, Consequences and Environment. Croom Helm, London. Delap, A.D., Farrington, A., Praeger, R.L., and Smyth, L.B., 1932. Report on the recent bog-ﬂow at Glencullin, Co Mayo. Sci. Proc Roy. Dublin Soc. 20, 181–192. Delap, A.D. and Mitchell, G.F., 1939. On the recent bog-ﬂow in Powerscourt Mountain Townland Co. Wicklow. Sci. Proc. Roy. Dublin Soc. 22, 195–198. Dykes, A.P., 2002. Weathering-limited rainfall-triggered shallow mass movements in undisturbed steepland tropical rainforest. Geomorphology 46, 73–93. Dykes, A.P. and Kirk, K.J., 2000. Morphology and interpretation for a recent multiple peat slide event on Cuilcagh Mountain, Northern Ireland. In: Bromhead, E., Dixon, N., and Ibsen, M-L. (Eds), Landslides in Research, Theory and Practice. Thomas Telford, London, Vol. 1, pp. 495–500. Dykes, A.P. and Kirk, K.J., 2001. Initiation of a multiple peat slide on Cuilcagh Mountain, Northern Ireland. Earth Surf. Proc. Land. 26, 395–408. Feehan, J. and O’Donovan, G., 1996. The Bogs of Ireland: An Introduction to the Natural, Cultural and Industrial Heritage of Irish Peatlands. University College Dublin, Dublin, 399–419. Feldmeyer-Christe, E., 1995. La Vraconnaz, une tourbie`re en movement. Dynamique de la ve´ge´tation dans une tourbie`re soumise a` un glissement de terrain. Bot. Helv. 105, 55–73. Feldmeyer-Christe, E., 2002. Onze ans de dynamique de la ve´ge´tation dans une tourbie`re soumise a` un glissement de terrain. Bot. Helv. 112, 103–120. Feldmeyer-Christe, E. and Mulhauser, G., 1994. A moving mire – the bog burst of la Vraconnaz. In: Gru¨nig, A. (Ed.), Mires and Man. Mire conservation in a Densely Populated Country – The Swiss Experience. Excursion Guide and Symposium Proceedings of the 5th Field Symposium of the International Mire Conservation Group (IMCG) to Switzerland 1992. Swiss Federal Institute for Forest, Snow and Landscape Research, Birmensdorf, pp. 181–186. Gallart, F., Clotet-Perarnau, N., Bianciotto, O., and Puigdefa`bregas, J., 1994. Peat soil ﬂows in Bahı´ a del Buen Suceso, Tierra del Fuego (Argentina). Geomorphology 9, 235–241. Geo-Slope, 1995. SLOPE/W User Manual. Geo-Slope International Ltd., Calgary. Grifﬁth, R., 1821. Report relative to the moving bog of Kilmaleady, in the King’s County, made by Order of the Royal Dublin Society. J. Roy. Dublin Soc. 1, 141–144. Gunn, J., 2000. Karst catchment protection: the Cuilcagh Mountain Park initiative, County Fermanagh, Northern Ireland. In: Barany-Kevei, I. and Gunn, J. (Eds), Essays in the Ecology and the Conservation of Karst. Acta Universitatis Szegediensis/International Geographical Union, Szeged (Hungary), pp. 121–133.

Slope instability and mass movements in peat deposits

405

Hanrahan, E.T., 1954. An investigation of some physical properties of peat. Ge´otechnique 4, 108–123. Hanrahan, E.T., 1964. A road failure on peat. Ge´otechnique 14, 185–202. Hanrahan, E.T., Dunne, J.M. and Sodha, V.G., 1967. Shear strength of peat. Proceedings of the Geotechincal Conference, Oslo, Vol. 1, pp. 193–198. Helenelund, K.V., 1967. Vane tests and tension tests on ﬁbrous peat. Proceedings of the Geotechnical Conference, Oslo, Vol. 1, pp. 199–203. Hemingway, J.E. and Sledge, W.A., 1946. A bog-burst near Danby-in-Cleveland. Proceedings of the Leeds Philosophical and Literary Society IV, 276–284. Hendrick, E., 1990. A bog ﬂow at Bellacorrick Forest, Co Mayo. Irish Forestry 47, 32–44. Hobbs, N.B., 1986. Mire morphology and the properties and behaviour of some British and foreign peats. Q. J. Eng. Geol. 19, 7–80. Holden, J., 2004. Hydrological connectivity of soil pipes determined by ground-penetrating radar tracer detection. Earth Surf. Proc. Land 29, 437–442. Holden, J., Chapman, P.J., Lane, S.N., and Brookes, C., 2006 (this book, Chapter 22). Peatland hydrology. In: Martini, I.P., Martı´ nez Cortizas, A., and Chesworth, W. (Eds), Peatlands: Evolution and Records of Environmental and Climatic Changes. Elsevier, Amsterdam. Hollingshead, G.W. and Raymond, G.P., 1972. Field loading tests on muskeg. Can. Geotech. J. 9, 278–289. Hungr, O. and Evans, S.G., 1985. An example of a peat ﬂow near Prince Rupert, British Columbia. Can. Geotech. J. 22, 246–249. Hungr, O., Evans, S.G., Bovis, M.J., and Hutchinson, J.N., 2001. A review of the classiﬁcation of landslides of the ﬂow type. Env. Eng. Geosci. VII, 221–238. Hutchinson, J.N. and Bhandari, R.K., 1971. Undrained loading, a fundamental mechanism of mudﬂows and other mass movements. Ge´otechnique 21, 353–358. Kinahan, G.H., 1897a. Peat bogs and debacles. Trans. Inst. Civil Eng. Ireland 26, 98–123. Kinahan, G.H., 1897b. Bog slides and debacles. Nature 55 (1421), 268–269. Kirk, K.J., 2001. Instability of blanket bog slopes on Cuilcagh Mountain, N.W. Ireland. Unpublished Ph.D Thesis, University of Huddersﬁeld, U.K. Landva, A.O., 1980. Vane testing in peat. Can. Geotech. J. 17, 1–19. Landva, A.O., Korpijaakko, E.O., and Pheeney, P.E., 1983. Geotechnical classiﬁcation of peats and organic soils. In: Jarrett, P.M. (Ed.), Testing of Peats and Organic Soils. ASTM Special Technical Publication 820, Philadelphia, pp. 37–51. Landva, A.O., Korpijaakko, E.O.,and Pheeney, P.E., 1986. Notes on the original von Post peat and peatland classiﬁcation system. Advances in Peatlands Engineering, Proceedings, Ottawa, pp. 17–29. Landva, A.O. and La Rochelle, P., 1983. Compressibility and shear characteristics of Radforth peats. In: Jarrett, P.M. (Ed.), Testing of peats and organic soils. ASTM Special Technical Publication 820, Philadelphia, pp. 157–191. Landva, A.O. and Pheeney, P.E., 1980. Peat fabric and structure. Can. Geotech. J. 17, 416–433. Large, A.R.G., 1991. The Slievenakilla bog burst: investigations into peat loss and recovery on an upland blanket bog. Irish Naturalist J. 23, 354–359. Latimer, J., 1897. Some notes on the recent bog-slip in the Co. Kerry. Trans. Inst. Civil Eng. Ireland 26, 94–97. MacFarlane, I.C., 1969. Engineering characteristics of peat. In: MacFarlane, I.C. (Ed.), Muskeg Engineering Handbook. University of Toronto Press, Toronto, pp. 78–126. Marachi, N.D., Dayton, D.J., and Dare, C.T., 1983. Geotechnical properties of peat in San Joaquin delta. In: Jarrett, P.M. (Ed.), Testing of Peats and Organic Soils. ASTM Special Technical Publication 820, Philadelphia, pp. 207–217. McCahon, C.P., Carling, P.A., and Pascoe, D., 1987. Chemical and ecological effects of a Pennine peatslide. Environ. Pollut. 45, 275–289. Mills, A.J., 2002. Peat slides: morphology, mechanisms and recovery. Unpublished Ph.D Thesis, University of Durham, UK. Mitchell, G.F., 1935. On a recent bog-ﬂow in the County Clare. Sci. Proc. Roy. Dublin Soc. 21, 247–251. Mitchell, G.F., 1938. On a recent bog-ﬂow in the Co. Wicklow. Sci. Proc. Roy. Dublin Soc. 22, 49–54. Molyneux, W., 1697. Kapanihane bog ﬂow, near Charleville, County Limerick. Phil. Trans. Roy. Dublin Soc. XIX (Oct), 714–716.

406

A.P. Dykes and K.J. Kirk

Nash, D., 1987. A comparative review of limit equilibrium methods of stability analysis. In: Anderson, MG. and Richards, K.S. (Eds), Slope Stability. Wiley, Chichester, pp. 11–75. National Trust of Australia, undated. Wingecarribee Swamp – scene of one of Australia’s worst ecological disasters. www.nsw.nationaltrust.org.au/pmalert.html (accessed 7 July 2004). O’Loughlin, E., 1998. Contribution to ABC (Australia) Radio National radio programme: Earthbeat – 29/ 08/98: Wingecarribee Swamp. Transcript at: www.abc.net.au/rn/science/earth/stories/s12078.htm (accessed 7 July 2004). Radforth, N.W., 1969. Classiﬁcation of Muskeg. In: MacFarlane, I.C. (Ed.), Muskeg Engineering Handbook. University of Toronto Press, Toronto, pp. 31–52. Selby, M.J., 1993. Hillslope Materials and Processes. Oxford University Press, Oxford. Selkirk, J.M., 1996. Peat slides on subantarctic Macquarie Island. Z. Geomorph. N.F. 105, 61–72. Shroder, J.T., 1976. Mass movement on Nyika Plateau, Malawi. Z. Geomorph. 2, 56–77. Skempton, A.W. and DeLory, F.A., 1957. Stability of natural slopes in London Clay. Proceedings, 4th Int. Conference Soil Mechanics and Foundation Engineering (London), Vol. 2 , pp. 378–381. Skempton, A.W. and Petley, D.J., 1970. Ignition loss and other properties of peats and clays from Avonmouth, King’s Lynn and Cranberry Moss. Ge´otechnique 20, 343–356. Sollas, W.J., Praeger, R.L., Dixon, A.F., and Delap, A., 1897. Report of the committee appointed by the Royal Dublin Society to investigate the recent bog-ﬂow in Kerry Sci. Proc. Roy. Dublin Soc. VIII, 475–510. Tallis, J., 2001. Bog bursts. Biologist 48, 218–223. Tallis, J.H., 1964. Studies on southern Pennine peats III. J. Ecol. 52, 323–353. Tomlinson, R.W., 1981. A preliminary note on the bog burst at Carrowmaculla, Co. Fermanagh, November 1979. Irish Naturalists J. 20, 313–316. Tomlinson, R.W. and Gardiner, T., 1982. Seven bog slides in the Slieve-An-Orra Hills, County Antrim. J. Earth Sci. Roy. Dublin Soc. 5, 1–9. Troels-Smith, J., 1955. Karakterisering af lose jordater (Characterization of unconsolidated sediments). Danmarks Geologiske Undersogelse IV (3), 1–73. Vidal, H., 1966. Die Moorbruchkatastrophe bei Scho¨nberg/Oberbayern am 13/14.6.1960. Zeitschrift deutsch. geol. Ges. Jahrgang 1963 115, 770–782. Walker, C. and Gunn, J., 1993. A peat ﬂow in the catchment of Marble Arch Caves, Ireland. (Abstract). 3rd International Geomorphology Conference. McMaster University, Programme with Abstracts, p. 267. Warburton, J., Higgett, D., and Mills, A., 2003. Anatomy of a Pennine peat slide. Earth Surf. Proc. Land 28, 457–473. Warburton, J., Holden, J., and Mills, A.J., 2004. Hydrological controls of surﬁcial mass movements in peat. Earth-Sci. Rev. 67, 139–156. Wilford, G.E., 1966. A peat landslide in Sarawak, Malaysia, and its signiﬁcance in relation to washouts in coal seams. J. Sed. Petrol. 36, 244–247. Wilson, P., Clark, R., McAdam, J., and Cooper, E.A., 1993. Soil erosion in the Falkland Islands: an assessment. Appl. Geogr. 13, 329–352. Wilson, P., Grifﬁths, D., and Carter, C., 1996. Characteristics, impacts and causes of the Carntogher bogﬂow, Sperrin Mountains, Northern Ireland. Scot. Geogr. Mag. 112, 39–46. Wilson, P. and Hegarty, C., 1993. Morphology and causes of recent peat slides on Skerry Hill, Co. Antrim, Northern Ireland. Earth Surf. Proc. Land 18, 593–601.

407

C. Peatlands as multi-signal archives of environmental changes Section C considers peatlands as archives of important data that are potentially useful in recording the details of past-environmental (including climatic) changes, especially over the last 10,000 years or so (Chapter 18). As such, they complement similar data from archives such as lake sediments and soils. The record imprinted in mires of pre-historical and historical human activities are examined (Chapters 17, 20, 21) mainly through the analysis of mercury and lead. Chapter 20 looks at the possibility that mercury may be mobilized in a peatland, so that caution must be exercised in making the assumption that analysis of a given proxy element gives values that have been unmodiﬁed since deposition.

This page intentionally left blank

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

409

Chapter 17

Using bog archives to reconstruct paleopollution and vegetation change during the late Holocene T.M. Mighall, S. Timberlake, D.A. Jenkins and J.P. Grattan

Introduction Peatlands cover approximately 5 106 km2 of the Earth’s surface. These archives store records of past Holocene climate and environmental change and represent ideal databases to understand both natural and human-induced changes on natural ecosystems such as mining and metallurgy (Martı´ nez Cortizas and Weiss, 2002). Records of atmospheric metal pollution contained within archives including ice cores, peat bogs (peatlands) and lake sediments, have provided evidence that air, soil and water pollution – impacts that are often regarded as the product of modern technological development (Ngriau, 1996) – date back to ancient times (Hong et al., 1996; Bra¨nnvall et al., 1999; Mighall et al., 2002a) although both Jenkins (1988) and Ngriau (1996) suggest that the use of geochemical records preserved in such archives have been relatively unexplored. The British Isles is an ideal location to redress this research gap. Not only is it well endowed with peatlands and lakes, it also has a long history of metal mining and working (West et al., 1997). Chambers and Charman (2004) estimate that between 1.6 and 1.5 million hectares of original peatland exist in the United Kingdom, with a further 1.18 million hectares in Ireland. Indeed a sizeable proportion of the ombrotrophic peat in Britain is now largely conﬁned to the uplands such as Dartmoor, the North Pennines and the Cambrian mountains of central Wales: the same areas that contain large quantities of metalliferous ore (Figs. 17.1a, b). The British Isles contain a considerable wealth of metalliferous ores, with 16 major oreﬁelds, containing eight major non-ferrous metals and non-metallic minerals; lead (Pb), zinc (Zn), gold (Au), silver (Ag), copper (Cu), tin (Sn), ﬂuorspar (CaF2) and barite (Ba4SO4), have been exploited from late prehistoric times (Department of the Environment and Minerals Division, 1994). The largest lead oreﬁelds are located in the North Pennines, the Peak District, Shropshire and northeast Wales, whereas smaller deposits occur in the Mendips, Shropshire, southern Scotland and central Wales (Cranstone, 2001) (Fig. 17.1a). ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09017-1

410

T.M. Mighall et al.

Using bog archives to reconstruct paleopollution

411

Figure 17.1. (a) Location of principal oreﬁelds in the British Isles (from the Department of Environment and Minerals Division, 1994). (b) Distribution of blanket peat in the British Isles. (Source: JNCC international Designations Database, SNH Uplands Database and Queen’s University Belfast Peatland Survey. Source of map outlines: Maps in minutes).

412

T.M. Mighall et al.

Given the wealth of metalliferous deposits in the British Isles it is not surprising that there is evidence of mining and metalworking extending back four millennia. The excavation of these early workings has revolutionized our knowledge with regard to the origins and development of mining and metallurgy. For example, evidence for Bronze Age copper mining has now been ﬁrmly established in southwest Ireland (O’Brien, 1994, 2004), the Central Pennines (Barnatt and Thomas, 1998; Timberlake, 2002b) and in north and mid-Wales (Dutton and Fasham, 1994; Jenkins, 1995; Timberlake, 2003). Of 16 possible prehistoric mining sites identiﬁed in the central Welsh oreﬁeld, six have now been radiocarbon-dated to the Early Bronze Age (Timberlake, 1998, 2002b). Iron Age/Romano-British and medieval ironworking sites are also common with examples in northwest Wales (Crew, 1986, 1989, 1991a), Exmoor, the Forest of Dean and the Weald (Cleere and Crossley, 1985). This research has largely concentrated on surveying and dating early mines, mining technology, smelting technology and metal provenance studies (Craddock, 1994, 1995). This chapter reviews the use of biological and geochemical archives stored within bogs to reconstruct and shape our understanding of the impact of mining and metalworking on the landscape. Given the abundance of suitable paleoenvironmental archives and archeological and historical evidence of mining and metalworking, it will mainly use examples from the British Isles (Fig. 17.1a) and, in a few cases, from Europe. The use of paleoenvironmental archives, including the application of trace element analysis, pollen and microscopic charcoal analysis and mineral magnetic records contained within bogs, is discussed with reference to published examples and future developments are advocated. We argue that multi-proxy paleoenvironmental investigations can challenge archeological-based assumptions about the origins of metallurgy, the longevity of mining and metalworking on a local and/or regional scale, as well as provide evidence of the scale and nature of vegetation change and pollution caused by metallurgical activities.

The metallurgical industry as an agent of vegetation change Until recently, the role of mining and metalworking as an agent of vegetation change was largely neglected because the major focus of paleoecological research has been to examine the extent to which agriculture has created our cultural landscape (Chambers, 1993). Edwards (1999) suggested that in many instances extensive removal of woodland for smelting purposes, perhaps as part of a managed system, would have occurred during the Bronze and Iron Ages ‘‘but the detection of such activity in the pollen record, preferably associated with parallel studies of sediment chemistry, is relatively under-investigated’’. Vegetation change during prehistory Effectively, the interpretation of vegetation changes associated with mining and metalworking from pollen records is only plausible when there is archeological or

Using bog archives to reconstruct paleopollution

413

other evidence, such as metal deposition histories, that speciﬁcally identify phases of metallurgical activity. Archeological evidence suggests that miners and metalworkers exploited woodland for timber and charcoal. Wood and charcoal of various trees, especially oak and hazel, have been consistently found in spoil heaps and working areas of prehistoric mines such as Copa Hill, Cwmystwyth, mid-Wales (Timberlake, 2003) and Mount Gabriel, County Cork, Ireland (O’Brien, 1994) but also within other prehistoric and historic industrial contexts including lead, iron and tin works, potteries, lime burning and glassmaking (Gale, 2003). The use of wood for ﬁre setting, lighting and drainage have led to the suggestion that large areas of woodland must have been exploited to sustain mining and/or smelting operations, the implication being the impact on woodlands must have been severe (Voss, 1988). In contrast, Mighall and Chambers (1993) and Mighall et al. (2000a,b) have suggested that the impact of Bronze Age mining was negligible both at Copa Hill (Fig. 17.2) and Mount Gabriel. Here a series of short-lived, small scale declines in arboreal pollen, mainly Quercus (oak) and Corylus avellana-type (hazel), have been recorded during the known operation of the mine, thus total arboreal percentages are not permanently affected. Selective deforestation by prehistoric miners also took place in Europe. In particular, Fagus, but also Quercus and Corylus, were exploited to supply wood for the extraction and working of metals. Selective woodland clearances during the Late Bronze Age are recorded in pollen diagrams at a time when Cu, Ag and Au were extracted in the Mt Beuvray region of the Morvan, northern Massif Central in France. Total arboreal pollen percentages eventually recover to their pre-metalworking values, suggesting that such activities did not have a permanent impact although the composition of woodland changed (Monna et al., 2004). At Mount Gabriel, a permanent decline in arboreal (trees and shrubs) pollen percentages almost immediately follows after the cessation of prehistoric copper mining (Mighall and Lageard, 1999). In this instance, woodland appears to have been primarily removed to create land for agriculture, as cereal-type pollen and other herbs often associated with human activity are recorded. Mighall et al. (2000a) suggest that the miners may have managed the woodland to ensure there were sufﬁcient supplies or, alternatively, the demand for timber and charcoal was relatively small on an annual basis. An examination of tree-rings of wood recovered from mine adits suggests that some form of woodland management took place (McKeown, 1994). A similar pattern of vegetation change occurs close to prehistoric ironworking sites. Vegetation changes associated with a Late Iron Age and Romano-British ironworking hillfort of Bryn y Castell in upland southern Snowdonia in northwest Wales suggest that permanent woodland clearance occurred before the site was used for the manufacture of iron and only small-scale declines in certain arboreal taxa can be correlated with archeological evidence for the occupancy of the site. Betula and Alnus were most adversely affected, with the minor loss of Quercus and Corylus and the pollen data suggest that woodland recovered to its pre-ironworking level except within the immediate vicinity of the hillfort (Mighall and Chambers, 1997). Given the often temporary and small-scale nature of vegetation changes that occurred as a result of mining and/or metallurgical activities, it would be easy without archeological evidence for palynologists to dismiss such subtle changes as natural ﬂuctuations or the result of low-intensity agricultural activity. Clearly multi-proxy

414 T.M. Mighall et al.

Figure 17.2. Percentage pollen diagram and Pb and Cu concentrations for sampling site CH2, Copa Hill, Cwmystwyth, mid Wales. Pollen spores are expressed as percentages of total land pollen. Five hundred land pollen grains were counted for each level (Details of the site and methods used can be found in Mighall and Chambers (1993) and Mighall et al. (2002a,b). Horizontal shading across the diagram represents the known period of Bronze Age copper mining and the Roman period).

Using bog archives to reconstruct paleopollution

415

investigations which use archeological evidence to provide a chronology and information about the nature and scale of activity, combined with pollen and/or chemical data preserved in peatlands or lake sediments, could be a better way to help elucidate the impact of mining and metalworking on the landscape. Vegetation change during Roman and historical times The impact of mining and/or metalworking became more intense from Roman times onwards. Pollen data from sites in Britain and Ireland suggest that a more sustained period of woodland clearance occurred after the cessation of prehistoric mining at Copa Hill and Mount Gabriel. It commenced during the Iron Age, ca 2395735 14 C yr BP at Copa Hill (Fig. 17.2) with the demise of Ulmus, Alnus, Quercus and Fraxinus. The lack of metal enrichment in the peat at this time suggests that the mines were abandoned or only subject to very minor exploitation. It appears, therefore, the loss of woodland was primarily the result of agricultural expansion rather than mining as cereal-type pollen and other non-arboreal pollen with cultural afﬁnities, such as Plantago lanceolata, Rumex sp., Potentilla-type, Chenopodiaceae and Asteroidaceae, are all recorded. However, subsequent larger-scale changes in vegetation appear to have been strongly inﬂuenced by mining. After a slight recovery, tree and shrub percentages fell from 90 to 72 cm at which point total arboreal percentages reached their lowest recorded level with Quercus, Alnus, Betula and Corylus avellana-type all affected. The transformation of a well-wooded landscape into one dominated by heather-rich blanket peat and acidic grasslands coincides with a rise in Pb concentrations attributed to Roman mining and smelting in the Ystwyth valley (Mighall et al., 2002b). However, mining did not occur in isolation and there is also evidence that farming continued throughout the Roman period as cereal-type pollen is recorded at 72 cm, whereas taxa such as Plantago lanceolata and Rumex spp. are recorded more regularly in slightly higher percentages. Thus, the changes in vegetation were the result of a dual agricultural and mining economy. When mining stops the concentrations of Pb decrease and the total arboreal pollen percentages then slowly recover. This reaches pre-Roman values by about 60 cm suggesting that woodland has regenerated within approximately 275 years. The size of any impact on vegetation appears to be related to the scale and duration of mining. Although the combination of mining and agriculture led to permanent changes at Copa Hill, the impact of iron production may not have been so detrimental if mining and/or metal production was a local activity. Estimates from a study of a settlement near Joldelund, Nordfriesland (Do¨rﬂer, 1995) suggest that the amount of ore required to meet local demands and therefore that the amount of timber and charcoal requirements for iron smelting and for smithing was equivalent to that consumed for house building and domestic purposes. At this site, nearly 500 slag pits used between ca 350 and 450 AD were discovered over an area of 8 ha. Do¨rﬂer (1995) suggested that iron production for local consumption did not have a long-lasting impact on woodland in the Schleswig-Holstein area of northern Germany during the Roman Iron Age. Selective deforestation and the creation of large openings within the forest canopy appear to typify regional vegetation changes associated with mining and

416

T.M. Mighall et al.

metalworking elsewhere in Europe from Roman times onwards. In the Mt Beuvray region, Fagus pollen values collapsed as the concentrations of isotopic Pb increased during the Late Iron Age and peak at the apogee of the Aeduan civilization, a major metalworking culture (Monna et al., 2004). A decline in woodland, especially Fagus, which disappears completely from the pollen record, and a corresponding increase in Pb concentrations that also occurs in Mt Beuvray, is suggestive of another phase of local medieval mining. The decline of Fagus has also been associated with iron and copper production in the Johnsbach and Radmer valleys in Steiermark, eastern Austria, from the 12th to 6th century, respectively. Little change is recorded in the arboreal pollen sum between the 12th and 14th centuries suggesting that sporadic mining activity had no adverse effect on woodland, but from the 14th century (ca cal AD 1400–1650) slightly lower arboreal pollen percentages coincide with the expansion of the Radmer copperworks. Fagus pollen percentages decrease permanently as beech charcoal was used in the copperworks, but the eventual rise of other tree taxa such as Picea, Alnus and Pinus suggest that overall forest cover may not have been adversely affected (Marshall, 2003). In contrast to other studies, Cu concentrations and inﬂux values did not increase during the operation of the Radmer copperworks (Marshall, 2003). Fagus also declined in other large-scale medieval mining areas because it was preferentially used for charcoal (Sercelji, 1988; Galop and Jalut, 1994). The absence of Fagus as a major British woodland component precluded its selection by miners or metalworkers. Gale (2003) only recorded Fagus charcoal at one Roman and one medieval ironworking site and a Roman lime-burning kiln. At Copa Hill, Pb concentrations again begin to increase in the peat core from 40 cm onwards and the start of this rise is considered to date to the early medieval period (Fig. 17.2). At this point, the total arboreal pollen taxa also begin to decline. First Quercus percentages fall, followed by Betula, Alnus, and ﬁnally Corylus avellana type. The loss of woodland coincided with the earliest documented period of historical Pb–Zn mining in the Ystwyth valley, thus it is feasible that local woodland was exploited for use in the mines at this time. However, the increase in cereal-type pollen, Rumex sp., Potentilla type, Artemisia type Plantago lanceolata and Pteridium spores suggest that agricultural activities also intensiﬁed during this time. Although total arboreal pollen percentages do not fall below 10% within the top 40 cm of the Copa Hill pollen diagram, Leland in 1536 and 1539 (in Chambers, 2003) blamed lead smelting for the disappearance of woodland in the Ystwyth valley. A similar pattern within pollen diagrams from the North Pennines oreﬁeld suggests that woodland was permanently lost from the 11th century AD onwards (Mighall et al., 2004). The destruction of woodland in parts of the British uplands contrasts with evidence from the lowlands. Documentary records during historical times suggest that woodland management helped maintain timber supplies in southern Britain (Cleere and Crossley, 1995) as well as in the western Scottish highlands (Smout, 1999). Gale (2003) expressed little doubt that fuel-based industries had a profound effect on woodland composition of lowland Britain from the medieval period until the 17th and 18th centuries, yet questioned seriously how much of a detrimental impact such industries had on the survival of woodland. Rackham (2003) and Gale (2003) both suggest that the resulting management practiced could be regarded as beneﬁcial to the long-term maintenance of woodland. The natural life of many tree species can be

Using bog archives to reconstruct paleopollution

417

extended by coppicing or pollarding by several hundred years, practices that extend back into the Neolithic in Britain. The detection of coppicing and/or pollarding in pollen diagrams has proved elusive and therefore the use of such methods to examine such claims is difﬁcult. Often woodlands were destroyed only after metal mining and/ or working ceased. Agrarian transgression therefore was probably a more serious threat to woodland survival (Gale, 2003).

Paleopollution, mining and metalworking: records from bogs The use of trace element analysis as a geoarcheologic tool still remains at an early stage of investigation with regard to mining and metalworking sites. Jenkins (1988) suggested ‘‘that its development awaits the accumulation of more background data and also experimental studies on smelting, soil redistribution and other relevant systems’’ whereas Ngriau (1996) commented that it is ‘‘somewhat surprising that paleopollution study has remained an unexplored tool in the ﬁeld of archeology’’. Both these views still remain valid today. The lack of progress in this area is due in part to paucity in early (prehistoric and medieval) industrial sites but also a limited number of researchers are currently interested in early mining and metalworking. Moreover, there was a commonly held belief that mining and metalworking processes did not cause widespread pollution before the industrial revolution, a myth that has now been refuted with evidence of metal enrichment in ice cores, lake sediments and bogs (Hong et al., 1994; Shotyk et al., 1998; Renberg et al., 2000; Weiss et al., 2002), and the growing archeological evidence for prehistoric mines throughout the British Isles and Europe. One of the fundamental assumptions associated with using the bog archive to reconstruct atmospheric pollution histories is that the metals are not prone to remobilization once they have been incorporated into the acrotelm and catotelm. A plethora of studies have been conducted to test such a hypothesis. Recent research has challenged the long-held view that Pb is relatively mobile in peat (Urban et al., 1990). Shotyk et al. (1997) argued that post-depositional Pb migration in bogs is not supported by comparisons of Pb concentrations (total, extractable or in isotopic form) in bogs and lake sediments. Mackenzie et al. (1997, 1998) also used Pb isotope ratios in peat cores dated by 210Pb to argue in favor of the quantitative retention of atmospheric lead in Scottish bogs. By comparing herbarium Sphagnum plant samples and Swiss bog lead isotope ratios with time, Weiss et al. (1999a,b) effectively demonstrated that any post-depositional lead transformation involving redox, pH and dissolved organic matter in the pore waters did not signiﬁcantly alter the record of atmospheric lead deposition. Thus, the record of lead concentrations in the peat should provide a reliable record of past metal mining and working. Another approach to test the mobility of metals in peat is to compare archives from bogs in close proximity to those mining or smelting sites for which we have reliable archeological and documentary evidence for early working. Notwithstanding the numerous variables that inﬂuence the dispersion of gaseous and particulate pollution from its source (Macklin, 1992; Davies, 1983), bogs located close to industrial sites should provide an accurate chronological record of their activity as pollutants

418

T.M. Mighall et al.

are deposited onto the mire surface. For example, the detailed archeological excavation of a prehistoric Cu mine at Copa Hill (Timberlake, 2003) afforded the opportunity to test the mobility of Cu in an ombrotrophic blanket peat close by (Mighall et al., 2002a). The chronology of the mine gallery and entrance at Copa Hill was established by radiocarbon dating of artifacts used by the miners such as antler picks, wooden drainage launders, pit-props as well as the ubiquitous charcoal produced by ﬁre setting. Mining commenced sometime before 2000 BC, and the prehistoric opencast was abandoned around 1600 BC (Timberlake, 2003). The Cu concentrations determined within the radiocarbon-dated peat proﬁle rose signiﬁcantly during the known period of Bronze Age mining (Mighall et al., 2002a). Such evidence strongly suggests that Cu remained in situ and that the bog recorded localized pollution from the mine. By such means the chronological integrity of the metals can be tested against independent evidence. Once problems of post-depositional movement have been addressed and we can conﬁdently assume the chemical record contained in the bog archive is chronologically intact, the data from bogs can then be used to assess the pollution caused by metal mining and/or smelting. Mighall et al. (2002b) were able to reconstruct the atmospheric pollution history of Pb and Zn from the blanket peat on the upper plateau of the Ystwyth valley. Pb–Zn ores had been mined from the valley sides immediately below the blanket peat virtually continuously since the late 1500s, Pb production peaking in the late 1700s, by which time the mines were also being worked for zinc. Final closure of the metal mines in the Ystwyth valley occurred during the 1930s (Hughes, 1981). The lead pollution record within these peat proﬁles shows good correspondence with the known mining history (Fig. 17.2). Similar results have also been obtained from blanket peats close to lead mines in Weardale and Upper Teesdale in the North Pennines Pb–Zn oreﬁeld (Mighall et al., 2004). Bog archives not only test archeological interpretations but atmospheric pollution studies can be used as a predictive tool in archeology, in a rather similar way to that described by Whittington and Edwards (1994) for palynology. Records of atmospheric pollution from bog archives can produce plausible and challenging data to pose and isolate problems for the archeologist. Whittington and Edwards (1994) argued that pollen analytical studies can provide not only a full picture of cultural and environmental change (including evidence of the extent of cultivation, length of occupation and possibly the degree of nomadism) but they also highlight archeological lacunae. In much the same way, chemical records in peat proﬁles offer potential information with regard to aspects of a metal mining and working site that have eluded archeologists. For example, it is commonly very difﬁcult to ascertain when metallurgical activity actually commenced, its duration and whether it took place continuously or in phases based solely on the evidence from archeological excavation. A furnace, for example, can be radiocarbon- or archeomagnetically dated, but this date only provides an indication of its age, its construction and/or when it was last used. The great majority of old metal mines are multi-period, some having been exploited repeatedly over hundreds, and sometimes thousands of years, yet later activity and a lack of dateable artifacts can make it difﬁcult to build an accurate chronological sequence of events using archeological evidence alone. The work of mining archeologists has established that some of these complexities, such as

Using bog archives to reconstruct paleopollution

419

the presence of inverted stratigraphies associated with the back ﬁll of workings, and the movement and re-working of spoil (Timberlake, 2003). These are the sorts of problems that archeology can highlight but not necessarily answer. The bog archive at least provides circumstantial evidence to resolve some of these problems. Given a sufﬁcient sampling resolution, and assuming negligible movement of metals once incorporated into the peat, it is possible that separate phases of mining and/or smelting can be determined and dated, thus providing important information about the timing and longevity of metal mining and working operations and any associated metallurgy, especially when archeological evidence is scant or confused. An example is shown in the Copa Hill lead proﬁles of Figure 17.2. Before radiocarbon dating conﬁrmed the antiquity of the copper mine at Copa Hill, Davies (1947) suggested that the mines were probably Roman. Initially, the revised age for the mine dispelled the local tradition that the Romans mined lead in the valley. However, the Pb proﬁles produced by Mighall et al. (2002b) show at discernible peak between 70 and 90 cm in four separate monoliths. By extrapolation these elevated Pb concentrations correspond to the Roman period and the authors suggested that they represented pollution from the exploitation of Pb ores in the valley. Although this pollution phase has not yet been conﬁrmed by radiocarbon dating, recent evidence supports this other hypothesis. In 1999, a presumed medieval lead-smelting site located in the bottom of the Ystwyth valley was rediscovered. Subsequent excavation of several hearths and the associated radiocarbon dates from charcoal taken from the site has since yielded dates of Roman and medieval ages (Timberlake, 2002a). The best preserved of these bole hearths, which was little more than a small-scale or trial furnace, would seem to conﬁrm lead smelting and thus by inference some sort of mining during the ﬁrst to second century AD. Atmospheric pollution histories can also be used to test archeologically constructed hypotheses. Bick (1999) questioned the orthodox view that copper was the main objective for Bronze Age miners. Instead he proposed that early mines in Wales were exploited for Pb and Ag, because the location of most of the proposed copper mines were in places with a long tradition of Pb–Zn mining where copper ore was present as only a minor constituent. Furthermore, the low grade of the copper–pyrite ores, plus the absence of smelting heaths or slags were evidence ‘‘that the ores were never smelted’’ (Bick, 1999, p. 8). Bick concluded that ‘‘it is time to think the unthinkable – mining for Pb or Pb–Ag in mid-Wales in the early to middle Bronze Age’’. Mighall et al. (2000b), using the peat chemistry records from the blanket peat on Copa Hill, challenged this assumption. The pollution records suggested only slightly elevated concentrations of Pb and no Ag within the peat that accumulated during the Bronze Age, evidence that seems to support the idea that copper was the primary ore targeted. An archeological re-examination of some of the ﬁnely crushed mineral residues associated with working areas on the mine spoil tips appear to conﬁrm this. Chalcopyrite was being crushed on stone anvils whereas galena remained as larger lumps, presumably discarded onto the spoil tips (Mighall et al., 2000b). However, the case for this is not clear-cut. There is some evidence for limited working of the lead veins and crushing of ores elsewhere within the opencast. The Copa Hill mine has thus some tentative evidence for the extraction of lead ore approximately 300–500 years before its use in metalwork, a phenomenon that might

420

T.M. Mighall et al.

represent a period of metallurgical experimentation rather than actual production (Timberlake, 2003). There is a limited amount of evidence for the use of lead within the British Isles prior to 1500 BC. Hunter and Davis (1994) described a discovery of a lead bead necklace in an early Bronze Age – Beaker grave in southeast Scotland, whereas a tin–lead alloy head was discovered at St. Columb in Cornwall (Shell, 1979) and an urn encrusted with lead foil was found at Sheuy, Co. Tipperary (Rafferty, 1961). Lead was also intentionally being alloyed with copper and tin in bronze by the end of the middle Bronze Age, a phase of metalwork known as the Acton Park II metalwork production period (Tylecote, 1986), and again at the beginning of the Late Bronze Age (Rohl and Needham, 1998), where small pieces of lead metal have also been found in increasing numbers on Late Bronze Age occupation sites (Needham and Hook, 1988). Pre-Roman exploitation of Pb is also suggested in southwest England, where undated trench mines at Charterhouse on Mendip may be much earlier in origin than previously thought. Todd (1993) discovered a pre-Roman enclosure associated with lead smelting activity and there is evidence to suggest that lead isotope ratios within some of the Wilburton Phase metalwork are associated with Mendip ore sources (Rohl and Needham, 1998). Given that there is evidence for lead objects in prehistory, as well as Pb–Ag-rich deposits across Britain with archeological evidence for early mining, these ores could be identiﬁed in order to test whether Pb and Ag were being exploited back in the Bronze Age. Of course one major advantage of using the bog archive compared to an archeological excavation is that the former site is likely to be relatively undisturbed and the analysis straightforward. Stable lead isotope analysis of the ores, as well as samples from the bog closest to the prehistoric mines, might also provide us with more conclusive evidence for an anthropogenic origin for the lead (Le Roux et al., 2004; Monna et al., 2004). There has been a great deal of speculation, research and debate concerning the history of Pb mining in the North Pennines oreﬁeld. One issue that largely remains unresolved is the nature, scale and extent of Roman lead mining. Indeed, no precise location for mining or smelting has yet been determined, a situation that lends itself to those researchers who appear to readily accept that the evidence has been destroyed or covered by debris from subsequent mining (Jones, 1986). Most of the evidence for Roman lead mining in the North Pennines oreﬁeld is circumstantial, consisting of scattered artifacts such as the discovery of pigs of lead. However, undated remains of lead smelting have been found at a site on Wolsingham South Moor in Weardale. Wooler (1924) has suggested that the origins of these lead workings date back to the Roman period. There is some circumstantial evidence to support this claim, as a Roman settlement lies less than 2 miles away at Hamsterley (Raistrick and Jennings, 1965). An examination of atmospheric pollution histories recorded in bogs across Europe provides evidence for pollution that is correlated with the Roman metal mining industry (Shotyk et al., 1996; Go¨rres and Frenzel, 1997; Martı´ nez Cortizas et al., 2002; Monna et al., 2004). Several researchers have also interpreted minor Pb peaks within bog records as examples of pollution caused by Roman Pb mining in the British Isles (West et al., 1997; Mighall et al., 2002b). Now that the chronological integrity of the Pb record contained within bogs has been resolved, the approach can

Using bog archives to reconstruct paleopollution

421

also be applied to provide circumstantial evidence for metal mining within those regions where archeological evidence is scant or inconclusive. It is thus fair to assume that if the Romans were mining and smelting lead ore within the North Pennines on a reasonable scale, a pollution signal should be recorded within those bogs that now overlay large parts of the region. However, the lack of enriched Pb concentrations during later prehistoric times and again during the time of the Roman occupation within the Rookhope and Valley Bog peat records suggests that local lead ores were either not exploited at all or else on a scale insufﬁcient to generate a pollution signal (Mighall et al., 2004). Thus, the discovery of traces of lead found in Bronze tools, plus the increasing relative abundance of lead objects at Iron Age sites, which led Raistrick and Jennings (1965) to suggest that these cultures exploited lead, deposits cannot as yet be substantiated. Neither can it be conﬁrmed that Romans mined lead in the North Pennines (Raistrick and Jennings, 1965; Jones, 1986). Indeed, it is difﬁcult to provide a reason to account for the apparent absence of a Pb pollution signal in the North Pennines bogs if Pb ores were being exploited. Crew (1990) suggested that native iron production in northwest Wales virtually ceased during the Roman occupation and the Romans imported iron from other regions of Britain such as the Weald. One possibility is that local production was small scale and imported iron satisﬁed demand. A similar situation possibly occurred with lead. There is also evidence of signiﬁcant Pb pollution in certain areas of Europe, as for example Sweden (Bra¨nvall et al., 1999), Scotland (Eades et al., 2002) and NW Spain (Martı´ nez Cortizas et al., 1997) from the Roman period.

Experimental archeology and peatland research Experimental archeology also has a role in understanding the chemical records contained within bog archives. To date only a limited number of studies have been undertaken to reconstruct the smelting of ores and the process by which a metal bloom separated from its ore has been worked into an object. Crew (1991b) and Sauder and Williams (2002) have attempted to smelt iron ore into a bloom and then worked (bloom smithing) the bloom into a metal bar. Timberlake (1994) presented the results of a tin smelt using an experimental bowl and shaft furnace. If the reconstruction process, based on evidence from archeological excavation and using the experiences gained from ethnographic studies, replicates the smelting and metalworking process realistically, it provides an opportunity to understand how metals behave in such systems and, perhaps more importantly, we can collect and determine the atmospheric pollution straight from its source, especially if ores taken from former mining sites are used. It is then possible to test whether such data ﬁngerprints match the records of atmospheric metal pollution determined from a bog archive. A major focus of archeometallurgical research has been to provenance metal artifacts to their ore source in order to determine the main mining centers and possible trade routes. Stable lead isotopes (204Pb,206Pb,207Pb and 208Pb) ratios can distinguish between natural atmospheric Pb and Pb derived from anthropogenic sources (metalliferous ores, leaded petrol). Indeed, Pb determined in Greenland ice cores and bogs in Switzerland is thought to have originated from Pb pollution caused

422

T.M. Mighall et al.

by Roman mining and metallurgy (Rosman et al., 1997; Shotyk et al., 1998). Le Roux et al. (2004) and Monna et al. (2004) used stable Pb isotopes to reconstruct successfully the atmospheric Pb pollution history from Lindow Moss near Manchester and in the Mt Beuvray region of eastern France, respectively. Whereas Le Roux et al. (2004) revealed evidence of Pb pollution extending back to the late Iron Age, the isotopic ratios recorded could not discriminate which English ores had been exploited (all have similar isotopic ratios). However, when the results of isotopic ratios prove inconclusive, a comparison between a suite of trace elements characteristic of the exploited ores and that of the pollution signal recorded in a bog itself may be fruitful. Results from an examination of 14 chalcopyrite samples from potential sources for Welsh Bronze Age Cu and that of eight samples from the Comet Lode in the Ystwyth valley, from which prehistoric miners exploited Cu (Mighall et al., 2002a), are shown in Table 17.1 and Figure 17.3. Based on this analysis and because of differences in mineralogy from other potential sources for Welsh Bronze Age Cu, Jenkins and Timberlake (1997) suggested that the variation in trace element chemistry of the Welsh ores could be used as a tool to provenance ores. From the data it would appear that there are regional trends of high Bi, Ni, Pb and Sn in midWales, with more localized concentrations of Ag, Sb and Tl. Other signiﬁcant variations occur in Co, Cd and Ge. It was concluded that the calculation of the determination of relative concentrations of three systems (Ag:Cd:Sb, Ag:Co:Ni and Ag:Ge:Ni, as mapped on triangular diagrams) between both ore sources and artifacts, and the use of these as indices for comparison, showed most promise as a best means of a discrimination tool. Such differences may also be apparent in bog archives and so reconstructing atmospheric pollution histories close to known mines with ore geochemistry data can test this approach. Currently there seems to be no clear way of distinguishing between a preserved atmospheric pollution record produced by mining and that produced by smelting. Mining will inevitably lead to greater oxidation of exposed sulphide minerals and the slow leaching of metals into the surrounding environment, although atmospheric deposition into a true ombrotrophic bog is more likely to be in the form of a ﬁne wind-blown dust, possibly also in smoke produced by ﬁresetting. Smelting may preferentially concentrate the more volatile metals such as zinc, antimony, bismuth and arsenic over lead and copper within the atmospheric fallout (Jenkins and Timberlake, 1997). Whereas it is relatively straightforward to reconstruct heavy metal enrichment in bogs as a result of mining and smelting, the detection of ironworking is much more problematic given that it is a ubiquitous element and one that is prone to post-depositional remobilization (Shotyk, 1988). One approach that has not been explored fully is the use of mineral magnetic methods to discriminate material produced by ironworking from other sources. Some of the results of a preliminary investigation are presented in Figure 17.4. Particulate pollution was collected in traps from an iron smelting experiment conducted by Peter Crew at the Snowdonia National Park Centre. Archeological evidence of medieval furnace design and the type of ore used provided the framework and initial parameters for the ironworking experiment (Crew, 2004). A local source of bog iron ore was used to produce a bloom of iron with slag. Detail of earlier experiments was provided by Crew (1991b).

Using bog archives to reconstruct paleopollution

423

Table 17.1. Trace element content of 14 chalcopyrite samples from Wales and eight samples from the Comet lode, Cwmystwyth as estimated by ICP-MS. (Average crustal abundances from Krauskopf and Bird, 1995; elements indicated in bold are chalcophilic. No values were obtainable for Ge, Se, Te or Hg; values obtained but not reported for the RE elements.). Mean/crustal

Welsh mines

(mg/g)

Min.

Max. 22.4 0.49 77 7.3 11.7 1175

6.5 0.2 18 1.9 1.3 320

0.10 0.01 12 0.67 7.4 905

3.6 1.4 36 6.6 10.4 2060

1.4 0.5 9.4 1.9 1.0 418

Average

Standard Comet lode deviation Min. Max.

Standard deviation

Li Be Ti V Cr Mn

3.00 0.11 25.6 1.48 8.19 1040

0.01 0.5 0.005 0.01 0.1 1

0.05 0.01 10.9 0.4 6.1 590

Co Ni Zn Ga As Rb Sr Y Zr Mo

21.3 27.1 490 0.67 55.5 1.27 1.39 2.57 4.58 3.06

1 0.5 5 0.05 25 0.01 0.005 0.1 0.03 2

3.2 3.7 9.0 0.16 2.7 0.37 0.06 0.08 0.17 0.24

132 84 2800 3.0 406 4.7 8.2 16 19 18

34 24 870 0.7 108 1.5 2.2 4.2 7.0 5.3

4.0 11.6 54 0.18 3.4 0.36 0.56 0.45 0.37 0.09

48 170 310 0.9 260 5.7 4.8 5.1 27 1.2

14 55 96 0.2 86 1.8 1.6 1.6 11 0.4

Ag Cd In Sn Sb Cs Ba La Ce W Tl Pb Bi Th U

21.3 23.8 5.30 55.3 21.9 0.25 1.39 1.35 4.08 0.34 1.41 1160 208 0.14 0.74

200 100 50 25 100 0.1 0.003 0.03 0.05 0.3 3 100 1000 0.01 0.3

1.85 0.47 0.08 1.51 0.75 0.16 0.13 0.06 1.37 0.03 0.19 28 4.04 0.00 0.01

65 109 14 235 172 0.33 16.4 10.4 26 1.1 4.6 3070 575 1.16 7.2

21 44 4.7 77 45 0.1 5.2 2.7 6.3 0.3 1.4 1160 201 0.3 2.0

2.6 1.6 2.3 27 2.9 0.22 0.24 0.20 1.6 0.09 0.33 109 19 0.01 0.01

6.9 5.9 20 120 13.9 0.5 12.3 0.8 2.7 1.8 4.2 2500 380 0.2 1.4

1.5 1.6 5.9 32 4.2 0.1 3.7 0.2 0.4 0.6 1.3 776 123 0.1 0.5

Mineral magnetic measurements (susceptibility and remanence) of the particulate pollution released from the furnace conﬁrmed that it was highly magnetized and therefore should be detectable in a bog. Some of the atmospheric dust was sprinkled onto a peat core made with moss peat purchased from a local garden center and analyzed using a Bartington MS2E magnetic susceptibility probe. The results suggest that very small quantities of the atmospheric dust can be detected (Fig. 17.4a).

424

T.M. Mighall et al.

Figure 17.3. Location of chalcopyrite ores samples used to provide the data presented in Table 17.1.

S-ratios are presented (Fig. 17.4b) for a range of materials used or produced during the production of iron including the atmospheric dust, roasted bog ore obtained from an excavated ironworking site and charcoal from a medieval iron bloomery at Llwyn Du, northwest Wales, and samples of peat spiked with these products. Also shown are samples of peat taken from a core close to the medieval iron bloomery and a sample of purchased moss peat to act as a control. S-ratios help discriminate the mineralogy of a sample. A ratio close to +1.0 is indicative of ferromagnetic minerals whereas a ratio of below 0.6 indicates that anti-ferrimagnetic minerals dominate the sample. The atmospheric dust, roasted bog and spiked samples have higher S-ratios when compared to the charcoal and peat samples. However, the part of the Llwyn Du core that is thought to have accumulated during the tenancy of the bloomery (by radiocarbon dating) has an elevated S-ratio compared to the part of the core that accumulated before the bloomeries were operational. Thus, the S-ratios and other

Using bog archives to reconstruct paleopollution

425

Figure 17.4. (a) Magnetic susceptibility measurements from an ironworking experiment. A core of moss peat, purchased from a local supplier, was sprinkled with charcoal, atmospheric dust and roasted bog ore to determine whether such material could be recorded in a bog. The charcoal and roasted ore were collected from archeological excavations of ironworking sites. (b) S-ratios from an ironworking experiment and a bog close to the medieval iron bloomery of Llwyn Du, northwest Wales. The peat sample is a control. The Llwyn Du and charcoal samples are taken from a bog close to a medieval iron bloomery. The atmospheric dust was collected during an experimental iron smelt by Peter Crew, Snowdonia National Park Centre.

remanence magnetic measurements (ratios of Log ARM, Log Xlf and Log SIRM – data not shown here) have potential to detect ironworking pollution in bogs. Conclusions Metal mining landscapes are a palimpsest of human activities, both agricultural and industrial which have interacted in complex ways (Chambers, 2003). Both farmers and metallurgists have helped to create the cultural landscape of today. Paleoenvironmental investigations have chronicled those changes but separating their respective roles and impacts has yet to be fully achieved. Woodland disturbance does, however, coincide with phases of elevated metal concentrations. Early mining and metalworking activities had an impact on vegetation. During prehistoric times, the impact was often small scale and non-permanent. It intensiﬁed

426

T.M. Mighall et al.

during the Roman period although woodland did recover at some sites. Permanent woodland clearance coincided with metallurgical activities from medieval times onwards. The reconstruction of atmospheric metal deposition histories close to former metallurgical sites can be useful to test element mobility in bogs. Once immobility of an element has been established, bog archives can be used as an archeological tool to understand the history of a mine or metalworking site. Lead isotope analysis has been proven in identifying the sourcing and timing of early atmospheric lead pollution. A more systematic approach is needed to use isotopic data to identify early exploitation of lead ores in the British Isles. A multi-elemental analysis of bogs, in combination with chemical data from archeological sites and ores may be useful in reconstructing paleopollution records and sourcing pollution to speciﬁc sites, especially if the lead isotopic signal cannot differentiate between ore bodies. As yet there is no evidence to suggest it is possible to distinguish smelting from other mining activities. Based on the available evidence, mineral magnetism may identify periods of ferrous mining and metalworking in bogs.

References Barnatt, J. and Thomas, G., 1998. Prehistoric mining at Ecton, Staffordshire: a dated antler tool and its context. Mining History: Bull. Peak Dist. Mines Hist. Soc. 13, 72–78. Bick, D.E., 1999. Bronze Age copper mining in Wales – fact or fantasy?. Hist. Metall. 33, 7–12. Bra¨nnvall, M-L., Bindler, R., Renberg, I., et al., 1999. The medieval metal industry was the cradle of modern large-scale atmospheric lead pollution in Northern Europe. Environ. Sci. Technol. 33, 4391–4395. Chambers, F.M. (Ed.), 1993. Climate Change and Human Impact on the Landscape. Chapman and Hall, London, 303pp. Chambers, F.M., 2003. Setting the scene. In: Murphy, P. and Wiltshire, P.E.J. (Eds), The Environmental Archaeology of Industry. Symposia of the Association for Environmental Archaeology No 20. Oxbow, Oxford, pp. 4–8. Chambers, F.M. and Charman, D., 2004. Holocene environmental change: contributions from the peatland archive. Holocene 14, 1–6. Cleere, H. and Crossley, D., 1995. The Iron Industry of the Weald. Merton Priory Press, Cardiff, 425pp. Craddock, P.T., 1994. Recent progress in the study of early mining and metallurgy in the British Isles. Historical Metallurgy 28, 69–84. Craddock, P.T., 1995. Early Mining and Metal Production. Edinburgh University Press, Edinburgh, 362pp. Cranstone, D., 2001. Industrial archaeology – manufacturing a new society. In: Newman, R. (Ed.), The Historical Archaeology of Britain c.1540–1900. Sutton Publishing, Stroud, pp. 183–210. Crew, P., 1986. Bryn y Castell hillfort – a late prehistoric iron working settlement in north-west Wales. In: Scott, B.G. and Cleere, H. (Eds), The Crafts of the Blacksmith. 1984 Symposium of the UISPP Comite pour la Siderurgie Ancienne, pp. 91–100. Crew, P., 1989. Excavations at Crawcwellt West, 1986–1989: a late prehistoric upland ironworking settlement. Archaeol. Wales 29, 11–16. Crew, P., 1990. In decline or prohibition? The end of prehistoric ironworking in north-west Wales. In: Benoit, P. and Fluzin, P. (Eds), Iron Paleometallurgy and Cultures. Proceedings of the Symposium of the Comite de la Siderurgie Anciennce, Belfort. Edition Vulcain, Paris, pp. 217–228.

Using bog archives to reconstruct paleopollution

427

Crew, P., 1991a. Late Iron Age and Roman iron production in north-west Wales. In: Burnham, B.C. and Davies, J.L. (Eds), Conquest, Co-existance and Change: Recent Aarchaeological Work in Roman Wales. Trivium 25, Lampeter, pp. 150–160. Crew, P., 1991b. The experimental production of bar iron. Hist. Metall. 25, 21–36. Crew, P., 2004. Cast iron from a bloomery furnace. HMS NEWS 57, 1. Davies, B.E., 1983. Heavy metal contamination from base metal mining and smelting: implications for man and his environment. In: Thornton, I. (Ed.), Applied Environmental Geochemistry. Academic Press, London, pp. 425–462. Davies, O., 1947. Cwmystwyth mines. Archaeol. Cambrensis 99, 57–63. Department of the Environment and Minerals Division, 1994. The reclamation and management of metalliferous mining sites. HMSO, London, 168pp. Do¨rﬂer, W., 1995. Versuch einer Modellierung des Energieﬂusses und des Rohstoffverbrauchs wa¨hrend der ro¨misch-kaiserzeitlichen Eisenverhu¨ttung in Joldelund, Ldkr. Nordfriesland. Probleme der Ku¨stenforschung im su¨dlichen Nordseegebiet 23, 175–185. Dutton, A. and Fasham, P., 1994. Prehistoric copper mining on the Great Orme, Llandudno, Gwynedd. Proc. Prehist. Soc. 60, 245–286. Eades, L.J., Farmer, J.G., MacKenzie, A.B., et al., 2002. Stable lead isotopic characterisation of the historical record of environmental lead contamination in dated freshwater lake sediment cores from northern and central Scotland. Sci. Tot. Environ. 292, 55–68. Edwards, K.J., 1999. Palynology and people: observations on the British record, In: Edwards, K.J. and Sadler, J., (Eds.), Holocene Environments of Prehistoric Britain. Quaternary Proceedings No 7, Wiley, Chichester, pp. 531–544 Gale, R., 2003. Wood-based industrial fuels and their environmental impact. In: Murphy, P. and Wiltshire, P.E.J. (Eds), The Environmental Archaeology of Industry. Symposia of the Association for Environmental Archaeology No 20. Oxbow, Oxford, pp. 30–47. Galop, D. and Jalut, G., 1994. Differential human impact and vegetation history in two adjacent Pyrenean valleys in the Arie`ge Basin, Southern France, from 3000 BP to the present. Veg. Hist. Archaeobot. 3, 225–244. Go¨rres, M. and Frenzel, B., 1997. Ash and metal concentrations in peat bogs as indicators of anthropogenic activity. Water Air Soil Pollut. 100, 355–365. Hong, S., Candelone, J-P., Patterson, C.C., and Boutron, C.F., 1994. Greenland ice evidence of hemispheric lead pollution two millennia ago by Greek and Roman Civilisations. Science 265, 1841–1843. Hong, S., Candelone, J-P., Patterson, C.C., and Boutron, C.F., 1996. History of ancient copper smelting pollution during medieval times recorded in Greenland Ice. Science 272, 246–248. Hughes, S., 1981. The Cwmystwyth Mines. British Mining No 17. Northern Mines Research Society 77pp. Hunter, F. and Davis, M., 1994. Early Bronze Age lead – a unique necklace from southeast Scotland. Antiquity 68, 824–830. Jenkins, D.A., 1988. Trace element analysis in the study of ancient metallurgy. In: Ellis-Jones, J. (Ed.), Aspects of Ancient Mining and Metallurgy, Bangor, 1986. University College, Bangor, pp. 95–105. Jenkins, D.A., 1995. Mynydd Parys copper mines. Archaeol. Wales 35, 35–37. Jenkins, D.A. and Timberlake, S., 1997. Geoarchaeological research into prehistoric mining for copper in Wales. Unpublished report to the Leverhulme Trust, pp. 141. Jones, R.F.J., 1986. Rome in the Pennines. In: Manby, T.G. and Turnbull, P. (Eds), Archaeology in the Pennines. British Archaeological Report British Series 158, Oxford, pp. 229–236. Krauskopf, K.B. and Bird, D.K., 1995. Introduction to Geochemistry. McGraw-Hill, London. Le Roux, G., Weiss, D., Grattan, J., et al., 2004. Identifying the sources and timing of ancient and medieval atmospheric lead pollution in England using a peat proﬁle from Lindow Bog, Manchester. J. Env. Monitor. 6, 502–510. MacKenzie, A.B., Farmer, J.G., and Sugden, C.L., 1997. Isotopic evidence of the relative retention and mobility of lead and radiocaesium in Scottish ombrotrophic peats. Sci. Tot. Environ. 203, 115–127. MacKenzie, A.B., Logan, E.M., Cook, G.T., and Puldford, I.D., 1998. Distributions, inventories and isotopic composition of lead in 210Pb-dated peat cores from contrasting biogeochemical environments: implications for lead mobility. Sci. Tot. Environ. 223, 25–35.

428

T.M. Mighall et al.

Macklin, M., 1992. Metal pollution of soils and sediments: a geographical perspective. In: Newson, M.D. (Ed.), Managing the Human Impact on the Natural Environment: Patterns and Processes. Belhaven Press, London, pp. 172–195. Marshall, P., 2003. Reconstructing the environmental impact of past metallurgical activities. In: Murphy, P. and Wiltshire, P.E.J. (Eds), The Environmental Archaeology of Industry. Symposa of the Association for Environmental Archaeology No 20. Oxbow, Oxford, pp. 10–18. Martı´ nez Cortizas, A., Garcı´ a-Rodeja, E., Pontevedra-Pombal, X., et al., 2002. Atmospheric Pb deposition in Spain during the last 4600 years recorded by two ombrotrophic pea bogs and implications for the use of peat as archive. Sci. Tot. Environ. 292, 33–44. Martı´ nez Cortizas, A., Pontvedra-Pombal, X., No´voa Mun˜os, J.C., and Garcı´ a-Rodeja, E., 1997. Four thousand years of atmospheric Pb, Cd and Zn deposition recorded by ombrotrophic peat bog of Penido Vello (Northwestern Spain). Water, Air and Soil Pollut. 100, 387–403. Martı´ nez Cortizas, A. and Weiss, D., 2002. Peat bog archives of atmospheric metal deposition. Sci. Tot. Environ. 292, 1–5. McKeown, S., 1994. The analysis of wood remains from mine 3, Mount Gabriel. In: O’Brien, W.F. (Ed.), Mount Gabriel: Bronze Age mining in Ireland. Bronze Age Studies 3. National University of Ireland, Galway, pp. 265–280. Mighall, T.M., Abrahams, P.W., Grattan, J.P., et al., 2002a. Geochemical evidence for atmospheric pollution derived from prehistoric copper mining at Copa Hill, Cwmystwyth, mid-Wales, U.K. Sci. Tot. Environ. 282, 69–80. Mighall, T.M. and Chambers, F.M., 1993. The environmental impact of prehistoric mining at Copa Hill, Cwmystwyth, Wales. Holocene 3, 260–264. Mighall, T.M. and Chambers, F.M., 1997. Early ironworking and its impact on the environment: palaeoecological evidence from Bryn y Castell hillfort, Snowdonia, North Wales. Proc. Prehist. Soc. 63, 199–219. Mighall, T.M., Chambers, F.M., Lanting, J.M., and O’Brien, W.F. 2000a. Prehistoric copper mining and its impact on vegetation: palaeoecological evidence from Mount Gabriel, Co. Cork, South-West Ireland. In: Nicholson, R.A. and O’Connor, T.P. (Eds), People as an Agent of Environmental Change. Symposia of the Association of Environmental Archaeology No. 16. Oxbow Books, Oxford, pp. 19–29. Mighall, T.M., Dumayne-Peaty, L., and Cranstone, D., 2004. A record of atmospheric pollution and vegetation change as recorded in three peat bogs from the northern Pennines Pb-Zn oreﬁeld. Environ. Archaeol. 9, 15–40. Mighall, T.M., Grattan, J.P., Lees, J.A., et al., 2002b. An atmospheric pollution history for lead–zinc mining from the Ystwyth Valley, Dyfed, mid-Wales, UK as recorded by an upland blanket peat. Geochem.: Explor., Environ., Anal. 2, 175–184. Mighall, T.M. and Lageard, J.G.A., 1999. The prehistoric environment. In: O’Brien, W.F. (Ed.), Sacred Ground: Megalithic Tombs in Coastal South-west Ireland. Bronze Age Studies 4. National University of Ireland, Galway, pp. 41–60. Mighall, T.M., Timberlake, S., Grattan, J.P., and Forsyth, S., 2000b. Bronze Age lead mining at Copa Hill – fact or fantasy?. Hist. Metall. 34, 1–12. Monna, F., Petit, C., Guillaumet, J-P., et al., 2004. History and environmental impact of mining activity in Celtic Aeduan territory recorded in a peat bog (Morvan, France). Environ. Sci. Technol. 38 (3), 665–673. Needham, S. and Hook, D.R., 1988. Lead and lead alloys in the Bronze Age – recent ﬁnds from Runnymeade Bridge. In: Slater E.A. and Tate, J.O. (Eds). Science and Archaeology, Glasgow, 1987. Oxford, British Archaeological Report 196, pp. 259–274. Ngriau, J.O., 1996. A history of global metal pollution. Science 272, 223–224. O’Brien, W.F., 1994. Mount Gabriel: Bronze Age mining in Ireland. Bronze Age Studies 3. National University of Ireland, Galway, 371pp. O’Brien, W.F., 2004. Ross Island: Mining, Metal and Society in Early Ireland. Bronze Age Studies 6. National University of Ireland, Galway, 768pp. Rackham, O., 2003. Ancient woodland. Castlepoint Press, Colvend, 584pp. Rafferty, J., 1961. National museum of Ireland archaeological acquisitions in 1959. J. Roy. Soc. Antiquaries Ireland 61, 43–107. Raistrick, A. and Jennings, B., 1965. A History of Lead Mining in the Pennines. Longman, London, 347pp.

Using bog archives to reconstruct paleopollution

429

Renberg, I., Bra¨nnvall, M-J., Bindler, R., and Emteryd, O., 2000. Atmospheric lead pollution history during four millennia (2000BC to 2000 AD). AMBIO 29, 150–156. Rohl, B. and Needham, S., 1998. The circulation of metal in the British Bronze Age: the application of lead isotope analysis. British Museum Occasional Paper no.102, 234pp. Rosman, K.J.R., Chisholm, W., Hong, S., et al., 1997. Lead from Carthaginian and Roman Spanish mines isotopically identiﬁed in Greenland ice dated from 600BC to 300AD. Environ. Sci Technol. 31, 3413–3416. Sauder, L. and Williams, S., 2002. A practical treatise on the smelting and smithing of bloomery iron. Hist. Metall. 36, 122–131. Sercelji, A., 1988. Palynological evidence of human impact on the forests of Slovenia. In: Salbitano, F. (Ed.), Human Inﬂuence on Forest Ecosystems Development in Europe. ESF FERN-CNR. Pitagora Editrice, Bologna, pp. 47–57. Shell, C.M., 1979. The early exploitation of tin deposits in SW England. In: Ryan, M. (Ed.), The Origins of Metallurgy in Atlantic Europe. The Stationery Ofﬁce, Dublin, pp. 251–263. Shotyk, W., 1988. Review of the inorganic geochemistry of peats and peatland waters. Earth-Sci. Rev. 25, 95–176. Shotyk, W., Cherburkin, A.K., Appleby, P.G., et al., 1996. Two thousand years of atmospheric arsenic, antimony and lead deposition recorded in an ombrotrophic peat bog proﬁle, Jura Mountains, Switzerland. Earth Planet. Sci. Lett. 145, E1–E7. Shotyk, W., Norton, S.A., and Farmer, J.G., 1997. Summary of the workshop on peat bog archives of atmospheric metal deposition. Water Air Soil Pollut. 100, 213–219. Shotyk, W., Weiss, D., Appleby, P.G., et al., 1998. History of atmospheric lead deposition since 12,370 14C yr BP recorded in a peat bog proﬁle, Jura Mountains, Switzerland. Science 281, 1635–1640. Smout, C., 1999. The myth of Caledon. Tree News, Autumn 14–17 . Timberlake, S., 1994. An experimental tin smelt at Flag fen. Hist. Metall. 28, 121–128. Timberlake, S., 1998. Survey of early metal mines within the Welsh Uplands. Archaeol. in Wales 38, 79–81. Timberlake, S., 2002a. Medieval lead-smelting boles near Penguelan, Cwmystwyth. Archaeol. Wales, 42, 45–59. Timberlake, S., 2002b. Ore prospection in the Early Bronze Age. In: Bartelheim, M., Pernicka, E., and Krause, R. (Eds), The Beginnings of Metallurgy in the Old World. Archaometrie – Freiberger Forschungen zur Altertumswissenschaft, Vol. 1, 327–358. Timberlake, S., 2003. Excavations on Copa Hill, Cwmystwyth (1986–1999): An Early Bronze Age Copper Mine within the Uplands of Central Wales. BAR British Series, Oxford, 348pp. Todd, M., 1993. Charterhouse on Mendip: interim report 1993. Somerset Archaeol. Nat. Hist. Soc. 137, 59–67. Tylecote, R.F., 1986. The Prehistory of Metallurgy in the British Isles. Institute of Metals, London, 257pp. Urban, N.R., Eisenreich, S.J., Grigal, D.F., and Schurr, K.T., 1990. Mobility and diagenesis of Pb and 210 Pb in peat. Geochim. Cosmochim. Acta. 54, 3329–3346. Voss, O., 1988. Iron production in Populonia. In: G. Sperl (Ed.), The ﬁrst iron in the Mediterranean. PACT 21. Strasbourg, Council of Europe, 91–100. Weiss, D., Shotyk, W., Appleby, P.G., et al., 1999a. Atmospheric Pb deposition since the industrial revolution recorded by ﬁve Swiss peat proﬁles: enrichment factors, ﬂuxes, isotopic composition, and sources. Environ. Sci. Technol. 33, 1340–1352. Weiss, D., Shotyk, W., Boyle, E.A., Kramers, J.D., Appleby, P.G., and Cheburkin, A., 2002. Comparative study of the temporal evolution of atmospheric lead deposition in Scotland and eastern Canada using blanket peat bogs. Sci.Total Environ. 292, 7–18. Weiss, D., Shotyk, W., Kramers, J.D., and Gloor, M., 1999b. Sphagnum mosses as archives of recent and past atmospheric lead deposition in Switerland. Atmos. Environ. 33, 3751–3763. West, S., Charman, D.J., Grattan, J.P., and Cherburkin, A.K., 1997. Heavy metals in Holocene peats from south west England: detecting mining impacts and atmospheric pollution. Water Air Soil Pollut. 100, 343–353. Whittington, G. and Edwards, K.J., 1994. Palynology as a predictive tool in archaeology. Proc. Soc. Antiq. Scot. 124, 55–65. Wooler, E., 1924. Roman lead mining in Weardale. Yorkshire Archaeol. J. 28, 93–100.

This page intentionally left blank

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

431

Chapter 18

Beyond the peat: synthesizing peat, lake sediments and soils in studies of the Swedish environment R. Bindler and J. Klaminder

Introduction Peatlands and lakes are two integral components of the Swedish landscape – a characteristic typical of many other, formerly glaciated, northern regions. Lakes cover an estimated 42,000 ha of the Swedish surface area (ca. 9%), whereas peatlands, deﬁned on the basis of at least 30 cm of accumulated organic matter, cover an estimated 64,000 ha of the surface area (ca. 15%). Consequently, research on contemporary biogeochemistry and environmental and climate changes must ideally consider important processes in both of these environments and not least interactive processes. Similarly, studies that seek to reconstruct past environmental changes, such as land-use changes, pollution and climate changes, should also consider the natural archives preserved in both of these physical environments. Whereas some environmental questions lend themselves more to exploration of one archive over the other, for example, lake sediments are more appropriate for studies of past changes in lake-water pH, synthesizing the results from both archives can generally strengthen any research on past changes and the inﬂuence such environmental changes may have on contemporary processes. An example of how these two archives can be used in a complimentary fashion leading to a stronger synthesis is given by the study of past metal deposition in Sweden and the use of this retrospective data to model the distribution and fate of these metals in boreal forest ecosystems. Forests, which cover about two-thirds of the Swedish landscape, are an important ecological, cultural and economic resource and preservation of this resource is an important research area. Using primarily lead as the focal point (Kylander et al., 2006 – this book, Ch. 21), in this chapter we show how synthesizing data from multiple sites and from different environmental media can strengthen studies of past environmental changes, as well as how these data can be used to study the inﬂuence of past changes such as atmospheric metal deposition on present-day processes (Fig. 18.1). We include cross correlation of lake sediment and peat records to improve age–depth models, synthesizing the data provided by these archives to ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09018-3

432

R. Bindler and J. Klaminder

Figure 18.1. Schematic diagram for the synthesis of peat, lake sediment and forest soil data in the study of the Swedish environment: the example of lead. Peat and lake sediments (A) are used to ﬁrst quantify past changes in lead concentrations; the chronological information from these records are used to (B) reconstruct lead deposition rates based largely on accumulation rates of lead in peat; and ﬁnally these data are combined with studies of lead in soils in order to model soil biogeochemistry (C) and long-term changes in soil lead concentrations. The map shows sites mentioned in the chapter (bogs: ﬁlled circles; lakes: open squares; and soils: open triangles).

provide long-term time series that are not available from present-day monitoring programs and ﬁnally the application of these data in the study of contemporary processes, where we focus especially on the biogeochemical cycling of lead in boreal forest soils and vegetation.

Lead in Swedish peat, lake sediments, and soils

433

This chapter presents data focused on lead in the Swedish environment. We cite studies on soils (from Bottnaryd, Ha¨sthult and Norra Kvill in southern Sweden and Arvidsjaur in the north), lake sediments (from Kalven and Koltja¨rn) and four bogs (Store Mosse, Trollsmosse, O¨nneby Mosse and Dumme Mosse).

Combining peat and lake sediment records Properties of peat and lake sediments Lake sediments and peat constitute two substantially different archives in terms of, for example, composition, biogeochemistry and hydrology. Peat represents a more open system since it is composed of a build-up of organic matter that is supported by water. From the perspective of making environmental reconstructions, bogs beneﬁt from the fact that they are supplied with nutrients and metals only via the atmosphere, although in the deepest peat layers upward diffusion from the underlying sediments can affect the distribution of some elements (Shotyk, 1996). Nutrients are continually redistributed in the uppermost peat and because of changes in water level, which can vary by as much as several decimeters, the depth of the redox boundary in the acrotelm and upper catotelm can vary substantially over the course of a year. This ﬂuctuation in redox conditions has led some researchers to question the geochemical stability of many elements in the uppermost peat; this includes metals that are considered to be generally conservative or bound tightly to organic matter, such as mercury and lead (Damman, 1978; Urban et al., 1990; Stewart and Fergusson, 1994). At the same time organic matter is continually lost from the peat – rapidly in the uppermost peat, where as much as 80% of the mass can be lost before burial in the catotelm, and thereafter much more slowly in the catotelm, where a further 10% of the original mass can be lost (Kuhry and Vitt, 1996). Lake sediments are derived from material that settles out of the lake-water column; this material may come from the surrounding catchments (mineral soil particles and terrestrial organic matter), from atmospheric deposition (pollen, soil dust, some metals and sulfate) or from within-lake productivity (diatoms, fecal pellets). Lake sediments represent relatively closed systems once the sediment has been buried below the active surface layer, which in many lakes may constitute only the unconsolidated, uppermost few centimeters. This active layer, where bioturbation and redox processes can redistribute some sediment or elements, typically comprises sediments accumulated over the past one or a few decades. Thereafter, biogeochemical processes alter some sediment properties relatively gradually. Of particular interest in Sweden is the existence of a smaller number of lakes that contain annually laminated sediments – also called ‘varves’ (from the Swedish varv: a complete cycle or revolution) – in the deepest areas of the lake basin. Varves in Swedish lakes are the product of marked seasonal changes with rapid transitions from summer to fall and winter to spring, which are typical for the climate of this region (Petterson et al., 1993). In some lakes, these seasonal transitions are coupled with rapid stratiﬁcation and subsequent oxygen deﬁciency in bottom waters for long periods during the winter and summer, which prevent bioturbation and disturbance of the

434

R. Bindler and J. Klaminder

sediment and thus contribute to preservation of the annual cycle of sediment deposition. This sediment cycle in Swedish lakes consists of a lighter-colored mineral layer laid down during snowmelt and spring discharge, a darker layer during the growing season consisting of remains from in-lake productivity (diatoms) and terrestrial organic matter, sometimes a second mineral layer during more intense fall runoff and ﬁnally a dark, ﬁne-grained layer that sediments out from the water column during winter ice cover. The advantage of varved sediment records for studying past changes is that they can provide an absolute chronology based rather simply on varve counting.

Improving chronologies of the peat record (using lake sediments) Whereas varves provide an absolute chronology, the development of chronologies in peat (and more typical non-laminated lake sediments) must rely on radiocarbon dating, calibration of radiocarbon years to calendar years and subsequent age–depth modeling. Consequently, environmental reconstructions based on varve counting can provide greater chronological control; alternatively, cross correlation of peat records with varved lake sediment records can serve as a valuable aid to anchor age models in peat, when such possibilities exist. To exemplify this, we compare the lead record – both concentrations and 206Pb/207Pb isotope ratios – from a 660-cm-deep peat sequence collected from Store Mosse in (Fig. 18.2) with the lead records from varved lake sediments from Kalven and Koltja¨rn, two lakes located in central and northern Sweden, respectively (the natural minerogenic lead fraction has been factored out of the lead concentration, leaving only the pollution contribution; see Farmer et al. (1996) or Bra¨nnvall et al. (1999) for further details on how this separation is calculated). Store Mosse is a large bog complex in the boreal–nemoral zone of southern Sweden that encompasses about 8000 ha. The peat record from Store Mosse is plotted along a calendar year scale based on calibrated radiocarbon ages (Stuiver and Reimer, 1993; Stuiver et al., 2004) and a conventional third-order polynomial regression, which passes through the 2-sigma age ranges for each of the dated sections (Fig. 18.2b). In Figure 18.2, a general agreement between the trends in pollution lead in the peat and the varved sediment records can be clearly seen; this includes features such as the well-established Roman lead peak around 1 BC, an increase in lead concentrations about 1000 AD and also the recent peak in lead in the 1970s, which is associated largely with the use of alkyl lead in gasoline. As Bra¨nnvall et al. (1997) concluded in a comparison of the long-term lead records from three lakes with those from three peat bogs (including Store Mosse, shown here), the general agreement of these two types of archives, as well as the agreement with historical records of metal production as reconstructed by Settle and Patterson (1980), is fairly convincing evidence for the stability of lead in the two environmental archives and the potential of peat as an archive for past metal deposition (see Renberg et al., 2001 for additional comparison of long-term peat, lake sediment and ice records). Two obvious differences between the peat and lake sediment records are apparent in Figure 18.2. First, the lead concentrations in pre-industrial layers are much lower in the peat and a logarithmic scale is necessary to see the small increase in lead concentrations from about 3500 years ago. Second, the 206Pb/207Pb ratios in the two archives differ. A

Lead in Swedish peat, lake sediments, and soils

435

Figure 18.2. (a) Comparison of the long-term record of lead (concentrations and 206Pb/207Pb) in Store Mosse, a bog in south-central Sweden, and two lakes, Koltja¨rn in northern Sweden and Kalven in central Sweden, which have annually laminated (varved) sediments (Bindler et al., 1999; Bra¨nnvall et al., 1999). The chronology in the lake sediments is based on varve counting, which has an estimated error in these lakes of 740 years for the ﬁrst 2000 years. (b) Diagram showing the radiocarbon-based age–depth model used to plot the lead record in Store Mosse, which is based on a conventional third-order polynomial regression that passes through the 2-sigma age ranges for each of the radiocarbon-dated levels; below the deepest-dated level a simple linear extrapolation is applied based on the accumulation rate of the peat between the two oldest radiocarbon dates, which gives a basal age of about 8000 years in accordance with existing data for this bog (Svensson, 1988; Klarqvist, 2001).

436

R. Bindler and J. Klaminder

large fraction of the lead entering the lake sediment, particularly in the two lakes shown, is derived from soil mineral particles. Old, crystalline bedrock with high initial U/Pb and Th/Pb ratios, such as the Precambrian shield bedrock of Fennoscandia and Canada, usually have evolved toward higher 206Pb/207Pb ratios (the ca. 1.5 in Kalven and Koltja¨rn) than that found in younger bedrock. The lake sediment record reﬂects a mixing of catchment-derived lead and lead derived from pollution sources. Peat receives inputs only via the atmosphere and thus the isotopic composition of the peat only reﬂects atmospheric inputs. Pre-anthropogenic aerosols have ratios of ca. 1.2 (Rosman et al., 1997; Shotyk et al., 1998), which is the value measured in Store Mosse after ca. 5500 14C yr BP when the peat transitions from a fen to an ombrotrophic phase. The comparisons of long-term records and others like this one, which are focused more on recent changes within the past two centuries, indicate that these two archives do respond to the same external changes that in this case can be directly linked to atmospheric deposition. As a further example, Farmer and co-workers in Scotland showed good consistency between peat and lake sediments as well as herbaria samples of forest mosses for lead deposited over the past two centuries (Farmer et al., 1997, 2002). Close agreement exists in the temporal pattern of lead deposition between the different environmental media – peat, lake sediments (here, varved sediments) and herbaria samples. This argues strongly against the likelihood that peaks in lead are the result of post-depositional changes that might contribute to a downward migration and accumulation of lead in older peat layers. However, as compared to the varved sediment record the timing of these features in the lead record differs slightly between the radiocarbon-dated peat records and the varved lake sediment records. In the varved sediment records from both Kalven and Koltja¨rn, the well-known Roman lead peak is centered during the period 50–150 AD, whereas the peat record indicates a somewhat earlier peak. Also, the well-deﬁned peaks in lead pollution at 1200 and 1530 in the varved sediments are also not as clearly deﬁned in the peat. The small chronological discrepancies between the peat and the varved sediment records highlight the inherent problems of age–depth modeling in peat based on radiocarbon dating, and the fact that an individual peat slice may not represent as discrete a point in time as is the case for a section of varved sediments. Using the absolute time markers given by the lead record in varves, as well as other absolute age markers such as tephra, it would be possible to improve age–depth models in peat (Renberg et al., 2001); this could be advantageous in studies of climate change, for example, where determination of rates of change may be very important. This approach was applied partly by Klaminder et al. (2003), who incorporated the Roman lead peak (using a calendar date of 50750 AD, based on the lead record from Greenland ice) in their age–depth model for three peat records used for lead studies.

Reconstructing regional long-term changes in deposition Since the lead record in lake sediments represents an integrated record reﬂecting both changes in direct atmospheric deposition to the lake surface and changes in transport of lead from surrounding catchment soils, consisting of natural soil lead and an accumulating pool of atmospherically derived lead, it is difﬁcult to model the direct

Lead in Swedish peat, lake sediments, and soils

437

atmospheric deposition of lead over time. It is possible to use stable lead isotopes to separate the total lead concentration into a pollution fraction and a naturally derived fraction originating from catchment soils (Fig. 18.2; Farmer et al., 1996; Bra¨nnvall et al., 1999), but due to processes such as sediment resuspension and focusing of ﬁnegrained material into the deeper basins of lakes (Blais and Kalff, 1995), the sediment ﬂux is considered to represent relative changes in atmospheric inputs and not necessarily absolute changes in deposition. However, because pollution lead is discernible in sediment records from at least the Roman period, it is reasonable to assume that atmospheric deposition of pollution lead must have had a discernible effect on lead concentrations in surface soils and more broadly on lead biogeochemistry. In lakes in southern Sweden, the input of pollution lead 2000 years ago exceeded the natural input of lead to sediments by a factor of two to three (Bra¨nnvall et al., 2001a). Atmospheric deposition can be modeled using lake sediments, but this requires time-consuming and costly whole-basin studies of multiple lakes, which thus far have only been applied to the timescale covered by radiometric lead (210Pb) dating (Swain et al., 1992; Engstrom et al., 1994; Fitzgerald et al., 2005). No studies have attempted to apply this approach for metal records over much longer timescales (103 years), except for estimating the whole-lake basin inventory in one lake that is the result of lead accumulation from long-term lead deposition in Europe (Bindler et al., 2001b). Peat, supported by lake sediment records, offers a less-complicated medium to model the long-term changes in atmospheric inputs of lead to the Swedish environment. In two studies in southern and central Sweden, we have examined the long-term accumulation of lead in peat records from a total of six bogs. These bogs have accumulations of ombrotrophic peat spanning as much as the past nearly 6000 years, with an additional 1000–3000 years of minerotrophic fen peat underlying the bog phase (Bindler et al., 1999; Klaminder et al., 2003). In all of these bogs the ombrotrophic peat older than about 3500 years has lead concentrations in the range of only 0.1–0.5 mg g 1. By comparison, the lead concentrations in the uppermost peat and in the organic layer of forest soils are typically three orders of magnitude higher; that is, in the range of 100–150 mg Pb g 1. Given the net peat mass accumulation rates in the older ombrotrophic peat, which are typically in the range of 30–70 g m 2 yr 1 (Svensson, 1988; Bindler et al., 1999; Klarqvist, 2001; Klaminder et al., 2003), the lead accumulation rate in the mid-Holocene prior to human inﬂuence was in the range of 1–10 mg Pb m 2 yr 1. A similar value for this time period was determined from the peat record from Etang de la Grue`re in Switzerland (Shotyk et al., 1998); this close similarity between long-term changes in lead accumulation at the Swedish and Swiss sites extends also to mercury (Bindler, 2003; Roos-Barraclough and Shotyk, 2003). Between 2500 and 3500 years ago the accumulation rate of lead began to increase. A small fraction of this initial increase can be attributed to lead transported in the atmosphere in association with soil mineral particles, as evidenced by threefold increases in lithogenic metals such as scandium, titanium and zirconium, which can be attributed to increased land use at this time period (Lagera˚s, 1996). However, a decline in the 206Pb/207Pb isotope ratio in the peat layers, beginning about 3500 years ago to values below ca. 1.2, signiﬁes an addition of anthropogenic sources. Whereas the average 206Pb/207Pb ratio of the upper continental crust and the ratio occurring in mid-Holocene peat in these bogs as well as the Greenland ice (Rosman et al., 1997)

438

R. Bindler and J. Klaminder

are about 1.2, the isotope ratio of the lead deposited on the bogs declined to values of ca. 1.17. This lower value can be explained by anthropogenic lead sources such as mining and later also coal burning, which have 206Pb/207Pb isotope ratios typically in the range of 1.16–1.18 (Fig. 18.3). By ca. 1 BC, the accumulation of lead in the bogs had increased by an order of magnitude (Fig. 18.2a). This combination of increased deposition and a change in lead source, as indicated by lead isotope ratios, was sufﬁcient to be observed even in lake sediment records. After a reduced period of accumulation following the decline of the Roman Empire, the lead accumulation rate again increased in the peat and by about 1500 AD it reached a level of around 1 mg m 2 yr 1, which was nearly two orders of magnitude greater than the accumulation rate in the mid-Holocene. Furthermore, this accumulation rate in the 1500 s was at a level comparable to the lead deposition rate in southern Sweden in the late 1990s. Lead deposition in the 1970s was an order of magnitude greater than that in the late 1990s (Ru¨hling and Tyler, 2001); the lead deposition in the 1970s was about three orders of magnitude greater the rate in the mid-Holocene (Shotyk et al., 1998; Bindler et al., 1999). Although it is generally accepted that metal accumulation rates in peat more closely reﬂect absolute deposition rates as compared to lake sediments, a study of multiple peat cores from Store Mosse indicates that there can be heterogeneity in metal accumulation rates within a single bog, even on small (o50 m) spatial scales (Bindler et al., 2004a). For lead accumulated during the period 1890–2000 AD, the inventory of lead per unit area in three cores from the bog varied by a factor of two, 0.75–1.4 g Pb m 2. Net

Figure 18.3. The range of 206Pb/207Pb isotope ratios in: (1) deeper lake sediments older than 3000 years and the C-horizon in soils (left); (2) the temporal changes in 11 radiocarbon-dated lake sediment records and in Store Mosse (open circles) during the past 4000 years (center); and (3) surface sediments and peat, the organic horizon (mor) in forest soil and major pollution sources (right). For the boxplots n refers to the number of sites, except for the pollution sources, where n represents the number of data points. Whereas the peat from Store Mosse, for example, tracks only atmospheric inputs of lead, the lake sediments reﬂect an integrated input from the atmosphere and soils. References for the data used in the ﬁgure are cited in the text.

Lead in Swedish peat, lake sediments, and soils

439

mercury accumulation in nine cores from the bog for the same 110-year time period varied by a factor of four, 0.85–3.4 mg Hg m 2. The implications of this variation from the perspective of modeling past atmospheric deposition can be exempliﬁed in Figure 18.4, which shows the accumulation of mercury during the period 1900–2000 from three cores from Store Mosse as well as one from another bog, Dumme Mosse, located about 60 km away. To overcome potential spatial variability, models of past changes should incorporate data from more than one core.

Using the peat record to determine the fate of lead in the Swedish environment So far, the potential to use the record of past metal accumulation to understand contemporary environmental processes has not been fully exploited. Most studies of the metal record in peat only quantify past changes in metal concentrations and to some extent also accumulation rates. Peat proﬁles with well-established chronologies can provide long-term environmental data series that can give us insights into timescales of environmental change not available in contemporary monitoring programs or studies of contemporary biogeochemical processes. In this section, we give an example of how the data extracted from the peat record can be used to model the background concentration of lead in boreal forest soils, speciﬁcally the organic horizon, which is most important for soil biota, and to model the cycling of lead through the soil and vegetation. The example: background level of lead in the organic horizon of boreal forest soils Podzols are acidic soils characteristic of boreal forests in Fennoscandia, Russia and North America. In Sweden these soils are typically stratiﬁed into an organic horizon

Figure 18.4. Modeled mercury accumulation in three peat cores from Store Mosse (lines) taken within a 50 m distance of each other and one peat core from Dumme Mosse, a bog located about 60 km away (diamond symbol; horizontal bars indicate time frame covered by each sample) (after Bindler, 2003; Bindler et al., 2004a).

440

R. Bindler and J. Klaminder

(a humus form speciﬁcally called ‘mor’) that covers the mineral soils like a blanket, a gray eluvial horizon (E-horizon) depleted in iron and aluminum, a dark illuvial horizon (spodic Bs-horizon) where organic complexes, iron and aluminum accumulate, and a gradation downwards into unaltered parent material (C-horizon), which is most commonly till. Bringmark and co-workers (Bringmark et al., 1998; Bringmark and Bringmark, 2001; Palmborg et al., 2001) in Sweden, as well as others elsewhere (Laskowski et al., 2003), have shown that elevated pollutant metal concentrations, such as lead and mercury originating from diffuse atmospheric pollution sources, can have a negative inﬂuence on microbial respiration in boreal forest soils. Although deposition of pollutant metals has decreased sharply in the past few decades, for example the lead deposition in the 1990s was only about 10% of that ca. 1970 (Ru¨hling and Tyler, 2001; Steinnes et al., 2005), it is not known how long it will take for soils to respond to these improvements. Time-series data on both site-speciﬁc metal concentrations in soil and site-speciﬁc metal deposition rates span only a few decades at best. This limits the understanding of the potential recovery rate of soils to different scenarios of atmospheric pollution emissions (including real-time changes) as well as derivation of a reasonable estimate for the pre-pollution concentration of pollutant metals. We know that monitoring data do not cover the full period of atmospheric pollution as evidenced by changes in lead and other metal accumulation rates in peat and lake sediments archives. This introduces several important questions particularly regarding effects on the present-day environment. For example, what was the natural content of lead or mercury in the organic horizon, which constitutes the main habitat of soil organisms and plant roots, before the advent of pollution? Answering such a question helps us to go beyond quantifying past records of metal deposition and to more fully grasp the extent of past and modern atmospheric pollution and its inﬂuence on the present-day environment. Lead and other metals derived from pollution sources have been atmospherically transported and deposited even in regions as remote as northern boreal forests and the Arctic for thousands of years. Analyses of lead isotopes across the northern hemisphere also clearly show that pollution is ubiquitous in the present-day environment even in locations far removed from emission regions; for example, a clear anthropogenic lead isotope signal is found in, besides Scandinavia (Rosman et al., 1998; Bra¨nnvall et al., 2001c; Steinnes et al., 2005), the Greenland ice record (Rosman et al., 1997) and in Arctic lake sediments, plants and soils (France and Blais, 1998; Bindler et al., 2001a) and even Arctic marine mammals such as beluga (Outridge et al., 1997). Consequently, reference conditions can no longer be measured directly and other means must be employed to estimate pre-pollution conditions. Some researchers suggest that the higher metal concentrations commonly observed in the organic horizon of soils are the result of natural processes (Rasmussen, 1998; Garrett, 2000), generally related to the strong binding capacity of organic matter or the uptake and recycling of metals by plants. The limited importance of these processes (metal retention by organic matter and plant recycling of metals) relative to atmospheric deposition can be shown by analyses of stable lead isotopes and mass balance budgets.

Lead in Swedish peat, lake sediments, and soils

441

Lead isotopes provide clear evidence that the original source of the lead in the organic horizon in soil proﬁles is not natural lead from the local soil (Bacon et al., 1995; Erel et al., 1997; Semlali et al., 2001; Watmough and Hutchinson, 2004). In Sweden, for example, the deeper, unaltered mineral soil in the C-horizon has high 206 Pb/207Pb ratios, 1.3–2.0, as is the case for deeper, older lake sediments (Bra¨nnvall et al., 2001c); such values are typical for soils and sediments derived from Precambrian crystalline bedrock (Chow, 1965). In contrast to these high ratios, the organic horizon has quite low 206Pb/207Pb isotope ratios that are comparable to the lead isotope ratios in modern aerosols (Hopper et al., 1991) and those found in the surface of bogs, 1.14–1.16 (Bindler et al., 1999, 2004a). From the organic horizon downwards into the upper C-horizon there is a gradient of increasing 206Pb/207Pb ratios indicating that the long-term input of pollution lead shown in the peat and lake sediment records has resulted in a pronounced downward migration of pollution lead in the forest soils (Figs. 18.3, 18.5). The time it takes for atmospherically deposited lead to be removed from the soil organic horizon by biogeochemical and physical migration processes is of critical importance for our understanding of past, present and also future environmental conditions. A rapid turnover of lead (at a timescale of a few years) in the organic horizon would suggest low background inventories and that the lead currently found there reﬂects inputs of recent pollution; thus, soils will respond rapidly to the decreasing atmospheric inputs over the past three decades. In contrast, a slow turnover (centuries to millennia) would suggest higher background inventories of lead in the organic horizon. In other words, the atmospheric pollution during the early Industrial Revolution and even medieval times as shown by the peat record, could still be an important component of the present-day pool of lead in the soil organic horizon. Therefore, the short-term response of soil-lead concentrations to decreased atmospheric pollution will be insigniﬁcant and any negative environmental effects

Figure 18.5. 206Pb/207Pb isotope ratios and lead concentrations and Pb concetration in soil proﬁles from four boreal forest sites in southern (Ha¨sthult, Bottnaryd and Norra Kvill) and northern (Vaksliden) Sweden (data from Bindler et al., 1999 and Bra¨nnvall et al., 2001b).

442

R. Bindler and J. Klaminder

attributed to elevated metal concentrations can be expected to continue well into the future. Because of this interest in the response of soils to improved emission controls, a number of studies have examined the biogeochemical cycling of metals in soils. Most of these ﬁrst estimates from the boreal zone originated from laboratory experiments or in situ measurements of soil water ﬂuxes of lead using lysimeters for soil solution sampling and calculations involving simple steady-state assumptions, which suggested residence times for lead in the organic horizon of several hundred years (Tyler, 1981; Bergkvist, 2001). However, as shown by past emission estimates and temporal variations in lead accumulation in environmental archives such as peat bogs, lake sediment and polar ice, steady state is not a reasonable assumption. Other studies have tried to include changing lead deposition rates into their estimates of the residence time for lead in the organic horizon. For example, in northeastern USA, Miller and Friedland (1994) and Johnson et al. (1995) included a model of lead atmospheric ﬂuxes during the 20th century, reconstructed from gasoline consumption trends, to search for a residence time of lead in the organic horizon that could explain the observed changes in lead concentration in forest ﬂoor samples collected since 1966 and 1977, respectively. Both studies estimated that the mean residence time (MRT) (the time it takes for the pool of deposited lead to decrease to the 1/e (ca. 37%) of its original amount) in the soil organic horizon in northern temperate hardwood forests was on the order of 17–35 years, whereas in higher elevation alpine spruce-ﬁr forests the MRT could be upwards to 77 years (Miller and Friedland, 1994). However, the use of a similar approach for solving turnover rates of lead in the Swedish mor, or organic horizons elsewhere in Europe, is not possible due to the long history of atmospheric pollution; hence, reconstruction of atmospheric ﬂux must consider more than 3500 years of pollution. It is here that bogs with a well-determined chronology can provide valuable long-term estimates of atmospheric ﬂuxes of lead and other metals. Recent modeling of lead biogeochemistry in Swedish soils incorporated time-series data from peat and lake sediment records; these studies suggest that the estimate for the lead residence time in the soil organic horizon is intermediate to the ﬁrst estimates by Tyler and the North American data from temperate forests; that is, in the range of 75–250 years (Bindler et al., 1999; Klaminder et al., 2005, in press). Essential to determining the fate of lead in the soil is an assessment of the importance of the upward transport of lead from the mineral soil to the organic horizon. Upward transport is a complex variable that depends on plant uptake rates from the mineral soil, in mixing of soil minerals, passive and indirect transport by soil biota and upward diffusion of lead solutes in the soil water. The essentially identical 206 Pb/207Pb isotope ratios in the soil organic horizon (1.15070.007 for samples collected from 11 sites in southern and northern Sweden in the late 1990s) and in the top 20 cm of bogs (1.15770.007, an arithmetic mean from four bogs with this depth interval corresponding to the past ca. 150 years of peat) in Sweden are a strong indicator that the lead in the uppermost soil layers is largely derived from atmospheric deposition. If upward mixing of natural lead derived from the underlying mineral soil, which has high 206Pb/207Pb ratios, and retention of this lead by organic matter were important, we would expect to ﬁnd a difference between the isotopic composition of peat and the organic horizon (Figs. 18.3 and 18.5). Aboveground

Lead in Swedish peat, lake sediments, and soils

443

vegetation has 206Pb/207Pb values that are slightly higher than those in the organic horizon, which indicates that there is a limited degree of plant uptake. For example, in southern Sweden, stemwood from Norway spruce (Picea abies) and Scots pine (Pinus sylvestris) have 206Pb/207Pb values typically in the range of 1.18–1.20 as compared to that in the organic horizon of the soils from the same sites (Bindler et al., 1999, 2004b). In a study of lead biogeochemistry in Arvidsjaur, a site in northern Sweden which receives much lower inputs of atmospheric pollution, we analyzed soil samples and several key plant species in an undisturbed boreal forest; here, we found that stemwood of spruce, pine and birch (Betula pubescens) as well as juniper (Juniperus communis) and ﬁeld-layer vegetation such as bilberry (Vaccinium myrtillus) and crowberry (Empetrum nigrum), had 206Pb/207Pb values from 1.2 to 41.3 (Klaminder et al., 2006). In contrast, the organic horizon from this site has values similar to those in atmospheric deposition recorded by peat bogs over the past 1000 years and herbaria samples of forest mosses from the study area (Fig. 18.6). This tendency toward a lead isotopic composition intermediate between the mineral soil and atmospheric pollution has also been observed in studies of tree rings from other areas, such as the UK (Watmough et al., 1999). From a mass balance perspective, Klaminder et al. (2005) combined data on lead concentrations in different boreal forest vegetation, including both tree species (spruce, pine and birch) and ﬁeld-layer plants (bilberry, heather and crowberry), with models of plant biomass to assess the importance of plant mobilization and uptake of lead in boreal forest soils and to determine actual uptake rates. Data from the lead record in peat cores from south-central Sweden and from herbaria samples of forest mosses collected in northern Sweden were included as a means of quantifying the isotopic composition of the lead deposited on the study region in recent centuries. All

Figure 18.6. (a) The 206Pb/207Pb isotope ratio recorded in peat from south-central Sweden together with herbaria samples of forest mosses collected in northern Sweden during the 20th century (left panel) and the ratios at a boreal forest site in Arvidsjaur, northern Sweden (data compiled from Bindler et al., 1999, 2004b; Klaminder et al., 2005b). (b) At the forest site the analyses included: the mineral soil and organic horizon, ﬁeld-layer plants (bilberry, crowberry, heather, mountain cranberry) and pine (shown here; spruce and birch were also analyzed), which was analyzed separately as needles, stem and also roots.

444

R. Bindler and J. Klaminder

Figure 18.7. Inventories of lead (mg Pb m 2) and the ﬂux of lead (mg Pb m 2 yr 1) from atmospheric deposition and uptake by forest vegetation in a boreal forest site, Arvidsjaur, northern Sweden (after Klaminder et al., 2005b). The ﬂux and 206Pb/207Pb ratio of atmospheric pollution are determined from ombrotrophic peat and forest mosses (see also Fig. 18.7).

of the analyzed forest vegetation had lead concentrations o1 mg Pb g 1, and nearly all samples had concentrations o0.1 mg Pb g 1. Multiplying these concentrations with the respective plant biomass per unit area, the estimated lead inventory in the vegetation of a northern Swedish forest site is 170.8 mg Pb m 2. This lead inventory in boreal forest vegetation is o1% of the amount of lead found in the organic horizon (Fig. 18.7) and only about 0.1% of the total inventory of pollution lead in the soil (Bra¨nnvall et al., 2001b). For further comparison, the accumulation of lead in peat for the period from AD 1890 to 2000 in south-central Sweden is about 1 g Pb m 2. To estimate the annual cycling rate of lead within the forest stand we multiplied the mean concentrations for each plant with their annual growth rate. This yields an average uptake of 0.170.1 mg Pb m 2 yr 1 by forest plants, where ca. 60% of the total vegetation uptake is due to spruce and where ﬁeld-layer shrubs contribute as much as 20% to the total ﬂux. The uptake of lead from the soil (mostly recycled pollution lead) and from direct atmospheric interception are estimated to be about 0.05 mg Pb m 2 yr 1 and 0.05 mg Pb m 2 yr 1, respectively. To place these estimated ﬂux rates of plant uptake of lead into perspective, the atmospheric lead ﬂux in northern Sweden during the last three centuries has been in the range of 0.5–3 mg Pb m 2 yr 1, whereas the average rate for the 20th century in south-central

Lead in Swedish peat, lake sediments, and soils

445

Figure 18.8. Modeled lead concentration in the organic horizon (mor) of boreal forest soils in Sweden during the past 5500 years. The model is based on contemporary models of lead cycling in forest soil (Johnson et al., 1995) and historical atmospheric deposition of lead based on the peat record (after Bindler et al., 1999). The model also projects future concentrations in organic horizon based on the current (2000 AD) lead deposition rate.

Sweden is in the range of 7–13 mg Pb m 2 yr 1 (Bindler et al., 2004a). So although forest vegetation does contribute to the biogeochemical cycling of lead, the relative importance of plants for the biogeochemical cycling of lead is negligible. Taking this data on the concentration of lead in vegetation and the rates of plant uptake of lead together with the data on past lead accumulation in peat records, where the pre-anthropogenic lead deposition was two to three orders of magnitude lower than modern values, we have come to the conclusion that an undisturbed soil organic horizon in the boreal forest would have contained very little lead. With turnover rates of lead and organic matter in the range of 75–250 years, the lead concentration in the organic horizon would have been o1 mg g 1, and possibly as low as 0.1 mg Pb g 1 (Fig. 18.8) (Bindler et al., 1999). As with estimates of atmospheric deposition, this estimate for the natural, pre-anthropogenic content of lead in undisturbed boreal forest soils is two to three orders of magnitude lower than those found in the environment today.

Conclusions The emphasis in this chapter has been on comparing peat and lake sediment records and linking the quantitative record of metals in peat to contemporary environmental problems, which in this case has focused on lead. Quantifying metal records in peat has been an important step, but new research needs to move beyond this and consider how to apply these data. Lead analyses, including stable isotopes, are now fairly routine and based on these analyses the historical trends of lead deposition are now well established in peat, lake sediments and even glacial ice. The biogeochemical cycling of lead has also been well researched, which allows us to make this link between the historical lead record and soil biogeochemistry. Although the focus in this chapter has been on data from Swedish studies, where there has been a more

446

R. Bindler and J. Klaminder

overt coupling between the peat record and present-day biogeochemistry, there are increasing numbers of studies that try to take advantage of the potential for longterm data series from peat and other archives. The next step is to apply the lessons learned from lead toward other ecotoxicologically important metals, such as cadmium or mercury. As an example of progress in this direction, Fitzgerald et al. (2005) coupled the historical record of mercury in sediments from several whole-lake basins with mercury biogeochemistry in an Arctic setting. Since peat and lake sediments seem to record the same changes in mercury deposition (Norton et al., 1997), there is similar promise in linking the long-term peat record of mercury and other metals with biogeochemical cycling of mercury and other important metals in forests and soils, for example, as shown here for lead.

References Bacon, J.R., Berrow, M.L., and Shand, C.A., 1995. Isotopic characterization of lead in the Scottish upland environment. Int. J. Environ. Anal. Chem. 59, 253–264. Bergkvist, B., 2001. Changing of lead and cadmium pools of Swedish forest soils. Water Air Soil Pollut. Focus 1, 371–383. Bindler, R., 2003. Estimating the natural background atmospheric deposition rate of mercury utilizing ombrotrophic bogs in south Sweden. Environ. Sci. Technol. 37, 40–46. Bindler, R., Bra¨nnvall, M.-L., Renberg, I., et al., 1999. Natural lead concentrations in pristine boreal forest soils and past pollution trends: a reference for critical load models. Environ. Sci. Technol. 33, 3362–3367. Bindler, R., Klarqvist, M., Klaminder, J. and Fo¨rster, J., 2004a. Does within-bog spatial variability of mercury and lead constrain reconstructions of absolute atmospheric deposition rates from single peat records? The example of Store Mosse. Global Biogeochem. Cycles 18, GB3020 doi 10.1029/ 2004GB002270. Bindler, R., Renberg, I., Anderson, N.J., et al., 2001a. Pb isotope ratios of lake sediments in West Greenland: inferences on pollution sources. Atmos. Environ. 35, 4675–4685. Bindler, R., Renberg, I., Bra¨nnvall, M.-L., et al., 2001b. A wholebasin study of sediment accumulation using stable lead isotopes and ﬂyash particles in an acidiﬁed lake, Sweden. Limnol. Oceanogr. 46, 178–188. Bindler, R., Renberg, I., Klaminder, J., and Emteryd, O., 2004b. Tree rings as Pb pollution archives? A comparison of 206Pb/207Pb isotope ratios in pine and other environmental media. Sci. Total Environ. 319, 173–183. Blais, J.M. and Kalff, J., 1995. The inﬂuence of lake morphometry on sediment focusing. Limnol. Oceanogr. 40, 582–588. Bra¨nnvall, M.-L., Bindler, R., Emteryd, O., and Renberg, I., 2001a. Four thousand years of atmospheric lead pollution in northern Europe: a summary from Swedish lake sediments. J. Paleolimnol. 25, 421–435. Bra¨nnvall, M.-L., Bindler, R., Emteryd, O., and Renberg, I., 2001b. Vertical distribution of atmospheric pollution lead in Swedish boreal forest soils. Water Air Soil Pollut. Focus 1, 357–376. Bra¨nnvall, M.-L., Bindler, R., Emteryd, O., et al., 1997. Stable isotope and concentration records of atmospheric lead pollution in peat and lake sediments in Sweden. Water Air Soil Pollut. 100, 243–252. Bra¨nnvall, M.-L., Bindler, R., Renberg, I., et al., 1999. The medieval metal industry was the cradle of modern large scale atmospheric lead pollution in northern Europe. Environ. Sci. Technol. 33, 4391–4395. Bra¨nnvall, M.-L., Kurkkio, H., Bindler, R., et al., 2001c. The role of pollution versus natural geological sources for lead enrichment in recent lake sediments and surface forest soils. Environ. Geol. 40, 1057–1065.

Lead in Swedish peat, lake sediments, and soils

447

Bringmark, L. and Bringmark, E., 2001. Soil respiration in relation to small-scale patterns of lead and mercury in mor layers of southern Swedish forest sites. Water Air Soil Pollut. Focus 1, 395–408. Bringmark, L., Bringmark, E., and Samuelsson, B., 1998. Effects on mor layer respiration by small experimental additions of mercury and lead. Sci. Total Environ. 213, 115–119. Chow, T.J., 1965. Radiogenic leads of the Canadian and Baltic Shield regions. Symp. Marine Geochem. 3, 169–184. Damman, A.W.H., 1978. Distribution and movement of elements in ombrotrophic peat bogs. Oikos 30, 480–495. Engstrom, D.R., Swain, E.B., Henning, T.A., et al., 1994. Atmospheric mercury deposition to lakes and watersheds. In: Baker, L. (Ed.), Environmental Chemistry of Lakes and Reservoirs. American Chemical Society, Washington, DC, pp. 33–66. Erel, Y., Veron, A., and Halicz, L., 1997. Tracing the transport of anthropogenic lead in the atmosphere and in soils using isotopic ratios. Geochim. Cosmochim. Acta 61, 4495–4505. Farmer, J.G., Eades, L.J., Atkins, H., and Chamberlin, D.F., 2002. Historical trends in the lead isotopic composition of archival Sphagnum mosses from Scotland (1838–2000). Environ. Sci. Technol. 36, 152–157. Farmer, J.G., Eades, L.J., MacKenzie, A.B., et al., 1996. Stable lead isotope record of lead pollution in Loch Lomond sediments since 1630 A.D. Environ. Sci. Technol. 30, 3080–3083. Farmer, J.G., MacKenzie, A.B., Sugden, C.L., et al., 1997. A comparison of the historical lead pollution records in peat and freshwater lake sediments from central Scotland. Water Air Soil Pollut. 100, 253–270. Fitzgerald, W.F., Engstrom, D.R., Lamborg, C.H., et al., 2005. Modern and historic atmospheric mercury ﬂuxes in northern Alaska: global sources and Arctic depletion. Environ. Sci. Technol. 39, 557–568. France, R.L. and Blais, J.M., 1998. Lead concentrations and stable isotopic evidence for transpolar contamination of plants in the Canadian High Arctic. AMBIO 27, 506–508. Garrett, R.G., 2000. Natural sources of metals to the environment. Human Ecol. Risk Assess. 6, 945–963. Hopper, J.F., Ross, H.B., Sturges, W.T., and Barrie, L.A., 1991. Regional source discrimination of atmospheric aerosols in Europe using the isotopic composition of lead. Tellus 43B, 45–60. Johnson, C.E., Siccama, T.G., Driscoll, C.T., et al., 1995. Changes in lead biogeochemistry in response to decreasing atmospheric inputs. Ecol. Appl. 5, 813–822. Klaminder, J., Renberg, I., Bindler, R., and Emteryd, O., 2003. Isotopic trends and background ﬂuxes of atmospheric lead deposition in N Europe: analyses of three ombrotrophic bogs from south Sweden. Global Biogeochem. Cycles 17, 1019–1028 doi:10.1029/2002GB001921. Klaminder, J., Renberg, I., Bindler, R., and Emteryd, O., 2005. Uptake and recycling of lead by boreal forest plants: quantitative estimates from a site in northern Sweden. Geochim. Cosmochim. Acta 69, 2485–2496. Klaminder, J., Renberg, I., Bindler, R., et al., 2006. Estimating the mean residence time of lead in the mor layer of boreal forest soils using 210-lead, stable lead and a soil chronosequence. Biogeochemistry 78, 31–49. Klarqvist, M., 2001. Peat growth and carbon accumulation rates during the Holocene in boreal mires. Swedish University of Agricultural Sciences (Silvestria 203), Umea˚. Kuhry, P. and Vitt, D.H., 1996. Fossil carbon/nitrogen ratios as a measure of peat decomposition. Ecology 77, 271–275. Kylander, M.E., Weiss, D.J., Peiteado Varela, E., et al., 2006 (this book, Chapter 21). Archiving natural and anthropogenic lead deposition in peatlands, In: Martini, I.P., Matı´ nez Cortizas, A., and Chesworth, W. (Eds), Peatlands: Evolution and Records of Environmental and Climatic Changes. Elsevier, Amsterdam. Lagera˚s, P., 1996. Farming and forest dynamics in an agriculturally marginal area of southern Sweden, 5000 BC to present: a palynological study of Lake Avego¨l in the Sma˚land Uplands. Holocene 6, 301–314. Laskowski, R., Niklinska, M., Nycz-Wasilec, P., et al., 2003. Variance components of the respiration rate and chemical characteristics of soil organic layers in Niepolomice forest, Poland. Biogeochemistry 64, 149–163. Miller, E.K. and Friedland, A.J., 1994. Lead migration in forest soils: response to changing atmospheric inputs. Environ. Sci. Technol. 28, 662–669. Norton, S.A., Evans, G.C., and Kahl, J.S., 1997. Comparison of Hg and Pb ﬂuxes to hummocks and hollows of ombrotrophic Big Heath Bog and to nearby Sargent Mt. Pond, Maine, USA. Water Air Soil Pollut. 100, 271–286.

448

R. Bindler and J. Klaminder

Outridge, P.M., Evans, R.D., Wagemann, R., and Stewart, R.E.A., 1997. Historical trends of heavy metals and stable lead isotopes in beluga (Delphinapterus leucas) and walrus (Odobenus rosmarus rosmarus) in the Canadian Arctic. Sci. Total Environ. 203, 209–219. Palmborg, C., Bringmark, L., and Bringmark, E., 2001. Microbial activity in relation to small scale patterns of heavy metals and substrate quality in spruce mor layers (Of) in southern Sweden. Water Air Soil Pollut. Focus 1, 409–424. Petterson, G., Renberg, I., Geladi, P., et al., 1993. Spatial uniformity of sediment accumulation in varved lake sediments in northern Sweden. J. Paleolimnol. 9, 195–208. Rasmussen, P.E., 1998. Long-range atmospheric transport of trace metals: the need for geoscience perspectives. Environ. Geol. 33, 96–108. Renberg, I., Bindler, R., and Bra¨nnvall, M.-L., 2001. Using the historical atmospheric lead deposition record as a chronological marker in sediment deposits in Europe. Holocene 11, 511–516. Roos-Barraclough, F. and Shotyk, W., 2003. Millennial-scale records of atmospheric mercury deposition obtained from ombrotrophic and minerotrophic peatlands in the Swiss Jura Mountains. Environ. Sci. Technol. 37, 235–244. Rosman, K.J.R., Chisholm, W., Hong, S.M., et al., 1997. Lead from Carthaginian and Roman Spanish mines isotopically identiﬁed in Greenland ice dated from 600 BC to 300 AD. Environ. Sci. Technol. 31, 3413–3416. Rosman, K.J.R., Ly, C., and Steinnes, E., 1998. Spatial and temporal variation in isotopic composition of atmospheric lead in Norwegian moss. Environ. Sci. Technol. 32, 2542–2546. Ru¨hling, A˚. and Tyler, G., 2001. Changes in atmospheric deposition rates of heavy metals in Sweden. A summary of Nationwide Swedish Surveys in 1968/70 – 1995. Water Air Soil Pollut. Focus 1, 311–323. Semlali, R.M., van Oort, F., Denaix, L., and Loubet, M., 2001. Estimating distributions of endogenous and exogenous Pb in soils using Pb isotopic ratios. Environ. Sci. Technol. 35, 4180–4188. Settle, D. and Patterson, C.C., 1980. Lead in Albacore: guide to lead pollution in Americans. Science 207, 1167–1176. Shotyk, W., 1996. Peat bog archives of atmospheric metal deposition: geochemical evaluation of peat proﬁles, natural variations in metal concentrations, and metal enrichment factors. Environ. Rev. 4, 149–183. Shotyk, W., Weiss, D., Appleby, P.G., et al., 1998. History of atmospheric lead deposition since 12,370 14C yr BP recorded in a peat bog proﬁle, Jura Mountains, Switzerland. Science 281, 1635–1640. Steinnes, E., A˚berg, G., and Hjelmseth, H., 2005. Atmospheric deposition of lead in Norway: spatial and temporal variation in isotopic composition. Sci. Total Environ. 336, 105–117. Stewart, C. and Fergusson, J.E., 1994. The use of peat in the historical monitoring of trace metals in the atmosphere. Environ. Pollut. 86, 243–249. Stuiver, M. and Reimer, P.J., 1993. Extended C14 data-base and revised Calib 3.0 C14 age calibration program. Radiocarbon 335, 215–230. Stuiver, M., Reimer, P.J. and Reimer, R., 2004. CALIB radiocarbon calibration, version 4.4. http:// calib.qub.ac.uk/calib/ Svensson, G., 1988. Bog development and environmental conditions as shown by the stratigraphy of Store Mosse mire in southern Sweden. Boreas 17, 89–111. Swain, E.B., Engstrom, D.R., Brigham, M.E., et al., 1992. Increasing rates of atmospheric mercury deposition in midcontinental North America. Science 257, 784–787. Tyler, G., 1981. Leaching of metals from the A-horizon of a spruce forest soil. Water Air Soil Pollut. 15, 353–369. Urban, N.R., Eisenreich, S.J., Grigal, D.F., and Schurr, K.T., 1990. Mobility and diagenesis of Pb and 210 Pb in peat. Geochim. Cosmochim. Acta 54, 3329–3346. Watmough, S.A., Hughes, R.J., and Hutchinson, T.C., 1999. 206Pb/207Pb ratios in tree rings as monitors of environmental change. Environ. Sci. Technol. 33, 670–673. Watmough, S.A. and Hutchinson, T.C., 2004. The quantiﬁcation and distribution of pollution Pb at a woodland in rural south central Ontario, Canada. Environ. Pollut. 128, 419–428.

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

449

Chapter 19

Occurrence and fate of halogens in mires H. Biester, A. Martı´ nez Cortizas and F. Keppler

Introduction Historic records of halogens in peat have received much less attention than those of other trace elements, such as heavy metals. This is likely because until recently halogens were considered as conservative elements, with the variations of their concentrations predominately reﬂecting atmospheric ﬂuxes and rainwater composition (Shotyk, 1997). In the last century, large amounts of persistent organic pollutants were released to the environment by human activities, and halogenated organic compounds (HOC) make up a great fraction of these naturally occurring xenobiotics. As a consequence, organohalogens were assumed to be mainly of anthropogenic origin, but today the occurrence and formation of HOC have been conﬁrmed to be widespread. Gribble (2003) recently reported the occurrence of more than 3800 organohalogen compounds, most containing chlorine, but also some containing bromine and a few, iodine. HOC are produced by living organisms or natural abiotic processes in almost all natural environments. It is a key ﬁnding that organohalogens are naturally formed in bogs (Silk et al., 1997), and that large amounts have been stored in the Earth’s peatlands (Keppler and Biester, 2003) since the end of the last glaciation, especially in high latitudes. Studies of the fate of organohalogens in the environment show that soil organic matter and anoxic organic-rich environments, such as bogs, play a key role in binding and degradation of HOC. Understanding their natural formation and degradation is therefore essential in understanding the fate of HOC of anthropogenic origin. For example, the existence of high amounts of HOC in peat could explain the recently reported formation of dioxins and furans during peat burning (Meharg and Killham, 2003). The formation of HOC in bogs is also a source of halomethanes (Dimmer et al., 2001), compounds relevant to stratospheric ozone depletion. The understanding of the mechanisms involved in the natural degradation of HOC may help to design and support natural attenuation processes in soils or groundwater contaminated by organohalogens. Today about 2 billion people, mainly children, suffer from iodine deﬁciency disorders, and our understanding of the biogeochemical cycle of iodine is still poor (Keppler et al., 2004). Many aspects of the interaction of iodine with organic matter are related to this problem. The formation of ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09019-5

450

H. Biester, A. Martı´nez Cortizas, F. Keppler

organoiodine compounds, for example, is the main cause of iodine depletion in drinking water, since these compounds (mainly as dissolved organic matter) are usually removed by ﬁltration with activated carbon. In the late 1970s, the group parameter AOX (adsorbable organic halogens) was developed by Ku¨hn et al. (1977) to determine organic halogen pollution from anthropogenic sources. This method was later used to investigate the content of organohalogen compounds in soils, peats and plants (Mu¨ller and Schmitz, 1985; Asplund and Grimvall, 1991). Studies of samples of pre-industrial origin have demonstrated that the widespread occurrence of organohalogens in soils is mainly a result of the biological halogenation of organic matter (Keppler and Biester, 2003). Enzymes such as haloperoxidases and halogenases are thought to be involved in the natural halogenation of organic compounds in the environment (Shaw and Hager, 1959; Reddy et al., 2002). Chlorine, for example, can be bound to aromatic structures in humic substances by reactions involving enzymes such as chloroperoxidases (Flodin et al., 1997), and binding of iodine to phenolic groups in humic substances has also been reported (Warner et al., 2000). However, abiotic processes related to organohalogen formation in soil have been described (Keppler et al., 2000). Most studies on the formation of organohalogens focus on organochlorine compounds in soils, since chlorine is the most abundant halogen in natural environments. An overview of the role of chlorine in natural systems can be found in Winterton (2000). Recent research has challenged the long applied principle in biogeochemical modeling that soils act neither as a sink nor as a source of chlorine. In this context, chlorine is assumed to be a conservative element with respect to water. However, a recent review of the natural chlorine cycle (O¨berg, 2002) demonstrates that it does not behave conservatively in soils; and that it is involved in a multitude of processes such as the formation of volatile, semi-volatile and high-molecular organic compounds. Even less is known about the fate of bromine and iodine in soils. Studies on the dynamics of iodine and bromine indicate that both elements also form stable organohalogen compounds in soils, through interaction with humic substances, which limits their availability for plant uptake (Ra¨dlinger and Heumann, 2000). Although chlorine in plants mainly exists in the inorganic form, it is both incorporated into organic macromolecules (Myneni, 2002) and volatilized by abiotic methylation processes (Hamilton et al., 2003) during the degradation of plant material. Moreover, the amount of organically bound chlorine increases during the decomposition and humiﬁcation, indicating that degradation and chlorination of organic matter are concurrent processes in soils (Flodin et al., 1997). In contrast to chlorine, no information is available on the behavior of bromine and iodine in response to the degradation of plant material. Limited research has been conducted on the historical distribution of halogens in peat (Chague´-Goff and Fyfe, 1996; Shotyk, 1997; Silk et al., 1997; Keppler and Biester, 2003; Biester et al., 2004), and the data given in this chapter are based on a comparatively small number of publications. The largest data set available is derived from a study of three peat bogs located in the Magellanic Moorlands (Southern Chile), and many aspects of the fate of halogens in peatlands discussed here are based on this data set.

Occurrence and fate of halogens in mires

451

Sources of halogens in peatlands Atmospheric wet and dry depositions are the major pathways of halogens, and in most studies historical halogen deposition was reconstructed using ombrotrophic mires (those receiving elements exclusively by atmospheric deposition). At minerogenic sites, additional ﬂuxes through catchment run-off have to be considered. Analyses of halogen concentrations in rainwater at bog sites are scarce (Shotyk, 1997; Biester et al., 2004), and ﬂuxes through dry deposition are mostly unknown. According to Hultberg and Grennfeld (1992), dry deposition of chlorine from marine aerosols can exceed precipitation ﬂuxes in coastal areas. Table 19.1 gives an overview of chloride concentrations in rain and in surface waters of bogs at different locations. It shows that chloride concentrations are much higher at maritime sites than in continental sites. However, it is not exactly known to what extent chloride concentrations in surface waters reﬂect rainwater composition, and what is the role of peat-forming plants in the release or uptake of chlorine.

Halogens in peat Total organic halide (TOX) concentrations in peat show a wide range of values reaching up to more than 2000 mg kg1 in maritime bogs (Keppler and Biester, 2003). In some studies halogen concentrations in peat are given only as TOX, providing no speciﬁc data on concentrations of chlorine, bromine or iodine. Owing to the much higher deposition of chlorine, this element accounts for the largest part of the halogens stored in bogs. Data on iodine in peat are limited to a small number of Table 19.1. Chlorine concentrations in rain and surface water of peat bogs at different locations (after Shotyk, 1997, and original dataa). Region Continental sites Southern Poland Minnesota, USA NW Ontario, Canada Maritime sites NW Spain (inland) SW Finland N England NW Spain (coastal areas) NW Scotland W British Columbia Canada Falkland Islands a

Average chloride (mg L1)

References

0.2–0.3 0.770.8 0.570.4

Tolpa and Gorham (1961) Verry (1975) Vitt and Bayley (1984)

0.9–1.8 1.1–1.2 2.2–6.5 7.0–11.8 18.3–29.6 23.4

Martı´ nez Cortizas (unpub.) Tolonen et al. (1979) Gorham (1956) Martı´ nez Cortizas (unpub.) Pearsall (1956) Malmer et al. (1992)

72.2

Gorham and Cragg (1960)

The data for bogs from NW Spain were obtained with the ﬁnancial support of projects REN200309228-CO2-01 (Spanish Ministry of Science and Technology) and PGIDIT03PXIB20002PR (Xunta de Galicia).

H. Biester, A. Martı´nez Cortizas, F. Keppler

452

publications (Maw and Kempton, 1982; Biester et al., 2004; Keppler and Biester, 2003), whereas a greater data set is available for bromine (Maw and Kempton, 1982; Shotyk, 1997; Roos-Barraclough et al., 2002; Biester et al., 2004). Data on ﬂuorine concentrations in peat could not be found in the literature. An overview on halogen concentrations in peat is given in Table 19.2. The data show that concentrations are highest in oceanic bogs, and decrease with distance from the ocean following the decrease in halogen concentrations in rain. Concentrations of halogens in peat are usually signiﬁcantly higher than in other kinds of soil, which suggests that the presence of organic matter plays a key role in the retention of halogens. Studies on bromine and iodine in soils have shown that both are bound to organic matter (Yamoda, 1968; Yuita, 1994). Maw and Kempton (1982) report that more than 99% of bromine in peat from eight different locations was organic bromine, and other studies have conﬁrmed that halogens in peat predominately exist in organic form (Chague´-Goff and Fyfe, 1996; Silk et al., 1997; Keppler and Biester, 2003; Biester et al., 2004). Although the chemical nature of organohalogens in peat is not exactly known, Silk et al. (1997) could show that dioxin- and furan-like organohalogens are formed naturally in pre-industrial peat. Recent studies on halogens in two oceanic bogs in southern Chile (Magallenic Moorlands) show that 80–95% of total halogens exist as organic forms (Keppler and Biester, 2003; Biester et al., 2004) (Fig. 19.1). The enzymes involved in the halogenation Table 19.2.

Overview of halogen concentrations in peat from different locations.

Location NW Spain

Total chlorine (mg kg1) 660–3900

Total bromine (mg kg1)

Total iodine (mg kg1)

Reference

24–341

27–64 10–21

Martı´ nez Cortizas (unpub.) Biester et al. (2004) Biester (unpub.) Maw and Kempton (1982) Maw and Kempton (1982) Maw and Kempton (1982) Maw and Kempton (1982) Maw and Kempton (1982) Maw and Kempton (1982) Maw and Kempton (1982) Maw and Kempton (1982) Bindler (unpub.) Harty et al. (1991)

Patagonia (bog) Patagonia (fen) Canada

350–1200 n.a. n.a.

50–200/186 30–105/53 42

Finland

n.a.

12

4.1

Ireland

n.a.

110

17.7

Latvia

n.a.

19.9

6.3

Lithuania

n.a.

10.6

3

Liverpool (UK)

n.a.

115.7

8.9

Poland

n.a.

20.5

3.9

Somerset

n.a.

61

Sweden Jamaica

230–870 5700–27,000

11–70 160–2200

5.5

11.5 n.a. 3–30

Occurrence and fate of halogens in mires

453

Figure 19.1. Historical records of total halogen (TCl, TBr, TI) and organohalogen (OCl, OBr, OI) concentrations in a peat bog (SKY) located in the Magellanic Moorlands, Chile (after Biester et al., 2004).

of organic matter, such as haloperoxidases, demand oxic conditions that are not met in the deeper sections of the bog (the catotelm). Thus, the formation of organohalogens is restricted to the upper aerated section (the acrotelm) (Biester et al., 2004). Once, the humiﬁed plant material enters the catotelm the formation of organohalogens appears to cease. Although several dehalogenation processes have been described in anoxic environments, they have not yet been investigated in bogs. The high amounts (41000 mg kg1) of organohalogens found in bogs indicate that halogens have been historically enriched during the last 10,000 years of the Holocene. This is in agreement with a recent model of terrestrial iodine ﬂuxes (Fig. 19.2), which suggests that bogs are one of the largest terrestrial sinks of atmospheric-derived iodine (Keppler et al., 2004). The processes leading to the formation of organohalogens in bogs are probably different for chlorine than for bromine and iodine. Plant uptake and formation of organohalogens during decomposition of plant material has been suggested as the major pathway for the formation of organochlorines in peat, whereas this process seems to be of minor importance for the formation of organobromine and organoiodine. The uptake of chlorine (as chloride) by plants is signiﬁcantly higher than that of bromine and iodine. Chlorine concentrations in plants can range between 700 and

454

H. Biester, A. Martı´nez Cortizas, F. Keppler

Figure 19.2. Model of terrestrial iodine ﬂuxes (modiﬁed after Keppler et al., 2004). (Gg ¼ giga gram ¼ 109 g.)

21,000 mg kg1 dw (Bowen, 1979), whereas average bromine and iodine contents are in the range of 1–5 mg kg1 dw and 0.1–0.4 mg kg1 dw, respectively, and there is a signiﬁcant enrichment of bromine and iodine in soil as compared to plant material (Yamoda, 1968). Atmospherically derived bromine and iodine can also be directly retained by humiﬁed plant material and the formation of organobromine and organoiodine compounds is a result of direct sorption of these halogens to humic acids, or of direct bromination or iodination of the organic matter by microbial activity (Ra¨dlinger and Heumann, 2000).

Net accumulation and retention rates of halogens Net accumulation rates of halogens in peat bogs from different locations may vary considerably, mainly because of the different climatic conditions (Table 19.3). Since sea-spray aerosols are the primary source at least for chlorine and bromine (a large portion of iodine is emitted as CH2I2 by seaweed (Kolb, 2002)), atmospheric deposition of halogens decreases with distance to the coast, and halogen accumulation rates in oceanic bogs are therefore generally higher than at continental sites. Accumulation rates vary considerably, not only among sites, but also within a single bog. In most cases, the highest accumulation rates are found in the uppermost peat sections. Within-core variability is highest for chlorine and lowest for iodine. Accumulation rates of chlorine vary up to 18-fold, whereas those of bromine and iodine

Occurrence and fate of halogens in mires

455

Table 19.3. Mean net halogen accumulation rates in bogs at different locations (numbers in brackets show rates of atmospheric halogen deposition) (age in 14C yr BP). Location (age)

Chloride (mg m2 yr1)

Bromine (mg m2 yr1)

Iodine (mg m2 yr1)

Reference

Switzerland (0–3000)

10–94

o1–7

n.a.

NW Spain (0–5000)

20–150 (15000)

Chile (0–6000) Sweden (0–4100)

12–216 (600–3000) 1.7–12 (5.7–23.4) 10–173 0.03–0.9

Roos-Barraclough et al. (2002) Martı´ nez Cortizas (unpub.) Biester et al. (2004) Bindler (unpub.)

9–14 (340)

1–11 (n.a.) 0.4–1.2 (1–2.6) n.a.

n.a. ¼ Data not available.

by three- to sevenfold (Table 19.3). Some authors have suggested that the variations reﬂect changes in atmospheric deposition, leading to the conclusion that atmospheric ﬂuxes to bogs must have increased by a similar factor (up to 18 times) in the past ca. 100 years (Table 19.3). For example, Roos-Barraclough et al. (2002) attributed a 7–14 fold increase of bromine accumulation rates in modern times to anthropogenic emissions. Changes in climatic conditions, such as increasing precipitation rates are unlikely to account for this large increase in the accumulation of halogens in such a short time. The fraction of atmospheric halogen ﬂuxes derived from anthropogenic sources is not exactly known, but anthropogenic emissions are an unlikely explanation for the increase in halogen accumulation rates in bogs located at such remote sites as the Magellanic Moorlands. The available data on halogen retention in peat indicates that proportional retention of atmospheric-derived bromine and iodine is in general much higher than that of chlorine. Records of historic halogen accumulation rates do not show a clear trend of decreasing accumulation in peat with depth/age, so that the amount of halogens lost during peat diagenesis is assumed to be low. Nevertheless, some studies suggest that signiﬁcant amounts of halogens can be released from peat through formation of volatile organohalogens such as halomethanes (Dimmer et al., 2001), or as soluble organohalogens via the aquatic pathway. The estimated retention of bromine and iodine ranges between 8–50% and 36–46%, respectively (Roos-Barraclough et al., 2002; Biester et al., 2004), whereas retention rates of chlorine are usually much lower, not exceeding 2% of the deposited chlorine (Shotyk, 1997; Biester et al., 2004). The main reason for the much lower retention of chlorine is its higher atmospheric ﬂux, which exceeds that of bromine and iodine by two to three orders of magnitude. In addition, there are also probable differences between the three halogens in the extent of organohalogen formation.

Halogen and peat accumulation Recent research demonstrates that humiﬁcation and loss of carbon during peat decomposition have a strong inﬂuence, not only on the concentrations of organically bound elements such as halogens (or mercury), but also on their accumulation rates (Biester et al., 2003, 2004). The accumulation rates increase with increasing peat

456

H. Biester, A. Martı´nez Cortizas, F. Keppler

accumulation rates and are always highest in the uppermost, slightly decomposed peat, as mentioned above. This suggests that they are overestimated for lightly decomposed sections and that historical accumulation rates derived from differently decomposed peat are not directly comparable. This latter assumption is based on the fact that peat bogs are open systems, which loose high amounts of mass during the diagenesis of the organic matter. Mass loss of peat is not compensated by proportional increases in the concentration of organically bound elements or changes in density, so the accumulation rates do not remain constant during peat decomposition processes (Biester et al., 2003). As a consequence, the accumulation rates of organically bound elements have to be normalized to a reference decompositional state (i.e., normalizing to carbon accumulation in a highly decomposed reference section, which could be seen as the ﬁnal stage of the peat decomposition process). This approach is supported by the observation that corrected accumulation rates of mercury (after normalizing for carbon accumulation) derived from bogs are in better agreement with those derived from lake sediments (Biester et al., 2003). These results seem to apply to the accumulation of halogens in peat. Our studies in Chilean and Spanish bogs have shown that pre-industrial accumulation rates of bromine and selenium mirror those of Hg, suggesting that these elements are also affected by peat decomposition. In some cases bromine accumulation was used to correct the pre-industrial variations in atmospheric Hg ﬂuxes (Roos-Barraclough et al., 2002). However, normalizing to the net accumulation of carbon leads to the same results as does normalizing to bromine or other organically bound elements such as selenium. Thus, normalization based on accumulation rates of another organically bound element does not correct for changes in atmospheric ﬂuxes (an external process) but for differences in the degree of peat decomposition and net carbon accumulation (an internal process). Table 19.4 shows chlorine and peat (mass) accumulation rates in young and old sections of bogs at different locations in Europe and in southern Chile. As described for the Chilean bogs, halogen accumulation at European locations increase in the young, less humiﬁed peat layers simultaneously with increasing peat accumulation rates. The extent of this enrichment is not likely to have been caused by variations in atmospheric ﬂuxes induced by climatic changes but by peat decomposition. Thus, the link between halogen accumulation and peat decomposition must be a general process in bogs, and the interpretation of halogen accumulation rates should consider the effect of diagenetic changes on peat accumulation. However, this new approach is in an early stage and more data from different sites are necessary to better understand the inﬂuence of diagenetic processes on halogen records in bogs.

Halogens in peat porewaters The most common way to investigate the release of halogens from peat is to determine halogen concentrations in porewaters. Bog porewaters are usually collected using in situ diffusion equilibrium samplers or peepers (Bendell-Young and Pick, 1997; Steinmann and Shotyk, 1997) or sipping techniques (Romanowicz et al., 1993; Biester et al., 2006). Thus, it is worth mentioning that the available data on halogen

Occurrence and fate of halogens in mires

457

Table 19.4. Within-core variation of net chlorine and peat accumulation rates in peat bogs at different locations (BP ¼ 14C yrs BP). Location and age Switzerland 1892–1991 AD 2110 BP–1892 AD Switzerland 1893–1991 1730 BP–1893 AD Sweden 1850–2001 4100 BP–1850 AD NW Spain ca. 1980–1999 5000 BP Chile (GC1) 1880–1999 2020 BP–1880 AD Chile (SKY) 1090 BP–1950 AD 4110–1850 BP

Peat mass (g m2 yr1)

Reference

65

240

10

32

Steinmann and Shotyk (1997) Steinmann and Shotyk (1997)

94

271

12

33

108 14

118 60

Bindler (unpub.) Bindler (unpub.)

112

120

61

46

Martı´ nez Cortizas (unpub.) Martı´ nez Cortizas (unpub.)

216 52

65 28

Biester et al. (2004) Biester et al. (2004)

18 10

23 15

Biester et al. (2004) Biester et al. (2004)

Chlorine (mg m2 yr1)

Steinmann and Shotyk (1997) Steinmann and Shotyk (1997)

concentrations in porewaters were obtained by different analytical techniques and that most investigations are focused on inorganic halogen species only, neglecting soluble organohalogen compounds. Concentrations of chemical elements in peat porewaters may reﬂect the release of solutes from the surrounding peat, on the assumption that the low hydraulic conductivity limits the vertical ﬂow of water (Ingram, 1982). Nevertheless, some studies have shown that climatic ﬂuctuations of short duration may temporarily reverse the vertical direction of ﬂuid ﬂow through the peat, although this has little effect on porewater chemistry (Romanowicz et al., 1993). Long drought periods (3–5 years), however, can produce signiﬁcant changes in porewater chemistry (Siegel et al., 1995). The release of halogens from bogs is rarely investigated and most studies consider that halogen concentrations in porewater mainly reﬂect halogen concentrations in the rain. Nevertheless, the data obtained for the Chilean bogs and in other studies (Table 19.5) indicate that halogen concentrations in porewater are signiﬁcantly higher than concentrations in rain (Shotyk, 1997; Steinmann and Shotyk, 1997; Biester et al., 2006). Additional inputs by fog or dry deposition of sea-salt aerosols

H. Biester, A. Martı´nez Cortizas, F. Keppler

458

Table 19.5. Halogen concentrations in rain and porewater at different locations in the Magellanic Moorlands (GC 1, SKY; Biester et al., 2006) and NW Spain (PDC). Location (proﬁle depth/age)

Chloride (mg L1) mean

Bromine (mg L1) mean

Iodine (mg L1) mean

GC1 (120 cm/2000 BP) SKY (244 cm/6000 BP) PDC (190 cm/ca. 5300 BP)

0.5/15 0.4/4 9/14

3.9/122 3.8/47 200/90

0.48/6.8 0.68/11 n.a.

have been proposed to explain the high halogen concentrations in oceanic bogs (Shotyk, 1997). The difference between porewater and rain is highest for chlorine and bromine (15–30 ) at locations with high inputs of halogens by sea spray. Iodine shows a similar enrichment (ca. 20 ) at all investigated sites. Evapotranspiration and the relatively low retention rates by peat are probably related to this strong enrichment in porewaters. For example, in the Chilean bogs sodium and chlorine concentrations in porewater were similarly high, indicating that most chlorine is directly derived from atmospheric deposition and not released by the peat (Biester et al., 2006). It appears that chlorine concentration in porewater is mainly controlled by precipitation. However, in a Spanish bog mean bromine concentrations in rain were 2 times higher than in porewater (Table 19.5). Despite the differences in bromine concentrations in rain between the Chilean and the Spanish bog, Br accumulation rates and also Br concentrations in porewater are very similar in both bogs, indicating that Br concentrations in porewater are controlled by release processes from the peat and only to a minor extent by atmospheric deposition. Release of halogens from peat Halogen release from organic substrates has been mainly investigated in studies dealing with the fate of organohalogens (mostly organochlorines) released from anthropogenic sources, or with the development of remediation technologies to clean up groundwaters contaminated with chlorinated organic solvents. Up to now, little is known about dehalogenation of naturally formed organohalogens, in particular of organobromine and organoiodine compounds. Laboratory studies have shown that dehalogenation can occur under aerobic and anaerobic conditions and that halogens can be cleaved enzymatically or through reductive dehalogenation from the organic substrate. An overview on dechlorination processes can be found in Winterton (2000), and in Van Pe´e and Unversucht (2003). In Chilean bogs, halogen concentrations in porewater show strong variations throughout the peat column, and these variations are mostly independent of halogen concentrations in peat. Halogen concentrations in porewater correlate with changes in the C/N ratios in some sections of the peat proﬁles suggesting a relationship between the release of halogens and the degree of peat decomposition. In most cases, this relationship is more pronounced for bromine and iodine than for chlorine (Fig. 19.3b). In contrast to halogen concentrations in peat, which increase with

Occurrence and fate of halogens in mires

459

Figure 19.3. Percentage of released bromine, iodine and chlorine compared to released, dissolved organic carbon (DOC) in a Patagonian bog (SKY).

increasing humiﬁcation (lower C/N ratios) (Fig. 19.3a), halogen concentrations in porewater show the opposite trend, with higher values in less decomposed peat sections (higher C/N ratios) (Fig. 19.3b). This relationship is weak in bogs that are generally highly decomposed (GC1 and PDC in Table 19.6; C/N ratios in this bog are only half those found in other Chilean bogs). The dependency of halogen release on the degree of peat decomposition (mass loss) is even clearer if expressed as the proportion of released halogens (calculated on the basis of total halogen concentrations in peat) instead of concentrations (Fig. 19.4). In addition, the proportion of dissolved organic carbon (DOC) in porewater (related to total C in peat) shows the same trend to parallel the C/N ratios (Fig. 19.4). The ﬁnding that DOC release from peat decreases with increased peat humiﬁcation as a result of increasing condensation of the organic molecules has been demonstrated in several studies (Kalbitz and Geyer, 2002; Blodau et al., 2004). For the Chilean bogs the best correlation was found between proportions of bromine and iodine, and the proportion of DOC released (Fig. 19.4). Since a large amount of bromine (54–69%) and most iodine (88–93%) in peat porewater exist as soluble organohalogens, the relationship to DOC suggests that both halogens are predominately released as organohalogen compounds (Biester et al., 2006). In contrast, the relationship between the chlorine and DOC is weak in all investigated bogs, and the amount of organochlorines in porewater is estimated to be generally low (Fig. 19.4).

H. Biester, A. Martı´nez Cortizas, F. Keppler

460

Table 19.6. Mean concentrations of halogens in porewater and in peat, and mean proportions (in italics) of released halogens as well as ratios of dissolved organic carbon (DOC) and halogens in porewater compared to peat decomposition patterns expressed as C/N ratios in bogs of Patagonia (GC 1, SKY, PBR; Biester et al., 2006) and NW Spain (PDC). Halogen concentration Porewater Cl (mg L1)/% Br (mg L1)/% I (mg L1)/% DOC (mg L1)/% Peat Cl (mg kg1) Br (mg kg1) I (mg kg1) DOC/Cl DOC/Br DOC/I C/N

GC1 (Chile)

SKY (Chile)

PBR (Chile)

PDC (Spain)

15/8 122/0.5 6.8/0.54 22/0.03

4/11 47/1 11/1.1 48/0.1

10/15 97/2.2 21/2.1 42/0.1

15/13.7 90/0.6 12/0.03 36/0.07

1084 186 19 8,6 3492 44,360 27

366 48 11 41 9612 59,540 48

596 60 13 15 5972 26,770 58

1263 165 39 8 2664 33,150 30

Data on organochlorines in peat porewater have not yet been published. Organochlorine concentrations in lakes and rivers can range between 11 and 185 mg L1, and highest values were found in streams draining peatlands (Asplund and Grimvall, 1991). Assuming that similar amounts of organochlorine occur in peat porewater, o3% of total chlorine in porewater would exist as organochlorines (Asplund and Grimvall, 1991; O¨berg, 2002). Thus, most chlorine in porewaters is chloride originating from rain or released through reductive dechlorination. However, as discussed above, most chlorine in porewater is likely to be derived directly from rain and dry deposition, and enriched by evapotranspiration and low retention. Changes in peat decomposition and DOC release have been shown to explain most of the variations of bromine and iodine release from peat, and also partly of chlorine. Comparison of the proportions of released halogens in different peat bogs also highlights the key role of peat decomposition and DOC release. The data in Table 19.6 show proportions of halogens and DOC released from three bogs in the Magellanic Moorlands (Chile) and one from Spain, compared to halogen concentrations in peat and the mean degree of peat decomposition (expressed as C/N ratios; Biester et al., 2006). The proportion of chlorine in porewater is 7–10 times higher than those of bromine and iodine. Moreover, proportions of released halogens and DOC were generally lower in those bogs (GC1, PDC) where the degree of peat decomposition was highest (lowest C/N ratios), although halogen concentrations in peat and porewater were higher than in the less decomposed bogs. Thus, the degree of peat decomposition and the proportion of DOC seem to be the most important factors controlling the release of bromine and particularly of iodine, and concentrations of halogen in peat seem to be of minor importance. However, halogen concentrations in peat have inﬂuence on those of organohalogens in porewater,

Occurrence and fate of halogens in mires

461

Figure 19.4. Comparison of halogen concentrations in peat (a) and peat porewater (b) to the degree of peat decomposition (mass loss) expressed by carbon/nitrogen ratios (C/N) in a peat bog (SKY) located in the Magellanic Moorlands, Chile (after Biester et al., 2004, 2006).

so that molar ratios of carbon and halogens are smaller in the porewater of bogs where halogen concentrations in peat are relatively high (Table 19.6). It is not yet known how much of the accumulated halogens in peat are lost during the decomposition of peat and further investigations are needed. Comparisons

462

H. Biester, A. Martı´nez Cortizas, F. Keppler

between minerogenic and ombrogenic environments might help to understand the fate of halogens in peat. Minerogenic peats are generally more humiﬁed and loose higher amounts of their initial mass through mineralization than ombrogenic peat. Data from a minerogenic and an ombrogenic site located in the same area and receiving similar amounts of bromine by atmospheric deposition showed large differences in bromine concentrations in peat (Biester et al., 2006). Despite the fact that minerogenic mires receive additional bromine input from catchment runoff, average bromine concentrations in peat were 3.5 times lower than in the ombrogenic peat. This suggests that a large amount of the bromine was not retained by the highly decomposed peat, or is more likely lost during peat mineralization at the minerogenic site.

Conclusions Our understanding of the cycles of halogens is changing dramatically. They were long held to be conservative, non-reactive elements, whose concentrations in soils reﬂect exclusively atmospheric inputs. Now it is clear that complex cycling takes place, in which organic matter and microbial activity play central roles. It has also changed in the sense that HOC have been demonstrated to form in almost all natural environments and are not only released by anthropogenic activities. The few but increasing number of studies conducted in peatlands has revealed that these wetland ecosystems constitute signiﬁcant reservoirs of halogens at a global scale. Large amounts of halogens have been accumulating since the beginning of their formation (up to some 10,000 years or more), mostly as HOC. Concentrations of halogens are usually greater than in other type of soils, although the sequence of abundance is the same (chlorine4bromine4iodine). Within the characteristic vertical distribution of the contrasting geochemical environments of mires (the acrotelm and the catotelm) the halogenation of peat organic matter seems to occur in the upper aerated sections coupled to mass loss and decomposition. The latter processes are climate-dependent, speciﬁcally of variations in superﬁcial wetness, and thus halogen records reﬂect to a large extent the effect of climate changes on the mire mass balance. The release of halogens from the peat has also a strong climatic component reﬂected by the degree of peat decomposition and the potential DOC release. If, as it has been suggested (Freeman et al., 2001), wet and warm periods produce an increased DOC release, the release rates of halogens, especially of bromine and iodine, will also increase during such climatic conditions. It is therefore reasonable to assume that the release of halogens will increase in those bogs where wetness increases under present-day warming (in high-altitude mires). On the other hand, climate warming may cause increasing frequency of drought events, and dry falling of peat lands, which will diminish export of DOC, and therefore also of halogens. Despite the general agreement regarding a comprehensive framework on halogen research there are many areas that demand more work. There is a need for extending the spatial and temporal representativeness of the records, for the identiﬁcation of the chemical nature of the organohalogens, the enzymes and processes involved in

Occurrence and fate of halogens in mires

463

halogenation/dehalogenation reactions and, ultimately, for the external and internal conditions controlling them. It seems obvious that other organically bound elements may respond to peat transformations in a similar way to organohalogens, but it may be possible that other elements (if not all) are also affected directly or indirectly by the same transformations. Investigating this issue is perhaps the most important challenge faced by peat researchers regarding the usefulness of peat records in the reconstruction of atmospheric ﬂuxes of elements.

References Asplund, G. and Grimvall, A., 1991. Organohalogens in nature. Environ. Sci. Technol. 258, 1346–1350. Bendell-Young, L. and Pick, F.R., 1997. Base cation composition of pore water, peat and pool water of ﬁfteen Ontario peatlands: implications for peatland acidiﬁcation. Water Air Soil Pollut. 96, 155–173. Biester, H., Keppler, F., Putschew, A., et al., 2004. Halogen retention, organohalogens, and the role of organic matter decomposition on halogen enrichment in two Chilean peat bogs. Environ. Sci. Technol. 387, 1984–1991. Biester, H., Martı´ nez Cortizas, A., Birkenstock, S., and Kilian, R., 2003. Historic mercury records in peat bogs. The role of peat decomposition and mass losses. Environ. Sci. Technol. 3, 32–39. Biester, H., Selimovic, D., Hemmerich, S., and Petri, M., 2006. Release of halogens from peat bogs – the role of peat decomposition and dissolved organic matter (DOM). Biogeosciences 3, 53–64. Blodau, C., Basiliko, N., and Moore, T.R., 2004. Carbon turnover in peatland mesocosms exposed to different water tables. Biogeochemistry 67, 331–351. Bowen, H.J.M. (Ed.), 1979, Environmental Chemistry of the Elements. Academic Press, London, . Chague´-Goff, C. and Fyfe, W., 1996. Geochemical and petrographical characteristics of a domed bog, Nova Scotia: a modern analogue for temperate coal deposits. Org. Geochem. 242, 141–158. Dimmer, C.H., Simmonds, P.G., Nickless, G., and Bassford, M.R., 2001. Biogenic ﬂuxes of halomethanes from Irish peatland ecosystems. Atmos. Environ. 35, 321–330. Flodin, C., Johansson, E., Boren, H., et al., 1997. Chlorinated structures in high molecular weight organic matter solated from fresh and decaying plant material and soil. Environ. Sci. Technol. 31, 2464–2468. Freeman, C., Evans, C.D., Monteith, D.T., et al., 2001. Export of organic carbon from peat soils. Nature 412, 6849. Gorham, E., 1956. On the chemical composition of some waters from the Moor House Nature Reserve. J. Ecol. 44, 375–382. Gorham, E. and Cragg, J.B., 1960. The chemical composition of some bog waters from the Falkland Islands. J. Ecol. 48, 175–181. Gribble, G.W., 2003. The diversity of naturally occurring organohalogen compounds. Chemosphere 52, 289–297. Hamilton, J.T.G., McRoberts, W.C., Keppler, F., et al., 2003. Chloride methylation by plant pectin: an efﬁcient environmentally signiﬁcant process. Science 301, 206–209. Harty, E.M., Lalor, G.C., and Robotham, H., 1991. Element concentrations in Jamaican peat. Environ. Geochem. Health 13/4, 197–202. Hultberg, H. and Grennfeld, G., 1992. Sulphur and seasalt deposition as reﬂected by throughfall and runoff chemistry in forested catchments. Environ. Pollut. 75, 215–222. Ingram, H.A.P., 1982. Size and shape in raised mire ecosystems: a geophysical model. Nature 297, 300–303. Kalbitz, K. and Geyer, S., 2002. Different effects of peat degradation on dissolved organic carbon and nitrogen. Org. Geochem. 33, 319–326. Keppler, F. and Biester, H., 2003. Peatlands: a major sink for naturally formed organic chlorine. Chemosphere 52, 451–453. Keppler, F., Biester, H., Putschew, A., et al., 2004. Organoiodine formation during humiﬁcation in peatlands: a key process in terrestrial iodine cycling. Environ. Chem. Lett. 1, 219–223.

464

H. Biester, A. Martı´nez Cortizas, F. Keppler

Keppler, F., Eiden, R., Niedan, V., et al., 2000. Halocarbons produced by natural oxidation processes during degradation of organic matter. Nature 403, 298–301. Kolb, C.E., 2002. Iodine’s air of importance. Nature 417, 597–598. Ku¨hn, W., Fuchs, F., and Sontheimer, H., 1977. Untersuchung zur Bestimmung des organisch gebundenen Chlors mit Hilfe eines neuartigen Anreicherungsverfahrens. Z. Wasser-Abwasser-Forsch. 6, 192–194. Malmer, N., Horton, D.G., and Vitt, D.H., 1992. Element concentrations in mosses and in surface waters of Western Canadian mires relative to precipitation chemistry and hydrology. Ecography 15, 114–128. Maw, G.A. and Kempton, R.J., 1982. Bromine in soils and peat. Plant Soil 65, 103–109. Meharg, A. and Killham, K., 2003. The burning of coastal peats – a major pre-industrial revolution source of dioxins and furans. Nature 421, 909–910. Mu¨ller, G. and Schmitz, W., 1985. Halogenorganische Verbindungen in aquatischen Sedimenten: anthropogen und biogen. Chemiker Zeitung. 109, 415–417. Myneni, S., 2002. Formation of stable chlorinated hydrocarbons in weathering plant material. Science 295, 1039–1041. O¨berg, G., 2002. The natural chlorine cycle – ﬁtting the scattered pieces. Appl. Microbiol. Biotechnol. 58, 565–581. Pearsall, W.H., 1956. Two blanket-bogs in Sutherland. J. Ecol. 44, 493–516. Ra¨dlinger, G. and Heumann, K.G., 2000. Transformation of iodide in natural and wastewater systems by ﬁxation on humic substances. Environ. Sci. Technol. 34, 3932–3936. Reddy, C.M., Xu, L., Drenzek, N.D., et al., 2002. A chlorine isotope effect for enzyme-catalyzed chlorination. J. Am. Chem. Soc. 124, 14526–14527. Romanowicz, E.A., Siegel, D.I., and Glaser, P.H., 1993. Hydraulic reversals and episodic methane emissions during drought cycles in mires. Geology 21, 231–234. Roos-Barraclough, F., Martı´ nez Cortizas, A., Garcia-Rodeja, E., and Shotyk, W., 2002. A 14,500 year record of the accumulation of atmospheric mercury in peat: volcanic signals, anthropogenic inﬂuences and a correlation to bromine accumulation. Earth Planet. Sci. Lett. 6334, 1–18. Shaw, P.D. and Hager, L.P., 1959. Biological chlorination IV. Peroxidative nature of enzymatic chlorination. J. Am. Chem. Soc. 81, 6527–6528. Shotyk, W., 1997. Atmospheric deposition and mass balance of major and trace elements in two oceanic peat bog proﬁles, northern Scotland and Shetland Islands. Chem. Geol. 138, 55–72. Siegel, D.I., Reeve, A.S., Glaser, P.H., and Romanowicz, E.A., 1995. Climate-driven ﬂushing of pore water in peatlands. Nature 374, 531–533. Silk, P.J., Lonergan, G.C., Arsenault, T.L., and Boyle, C.D., 1997. Evidence of natural organochlorine formation in peat bogs. Chemosphere 3512, 2865–2880. Steinmann, P. and Shotyk, W., 1997. Chemical composition, pH, and redox state of sulphur and iron in complete vertical porewater proﬁles from two Sphagnum peat bogs, Jura Mountains, Switzerland. Geochim. Cosmochim. Acta 61, 1143–1163. Tolonen, K., Raikamo, E., Lahti, T., and Viista, J., 1979. Siikaneva mire complex. Excursion guide, International Symposium on Classiﬁcation of Peat and Peatlands, Hyytia¨la¨, Finnland, September. International Peat Society, Helsinki, 26pp. Tolpa, S. and Gorham, E., 1961. The ionic composition of waters from three Polish bogs. J. Ecol. 49, 127–133. Van Pe´e, K.-H. and Unversucht, S., 2003. Biological dehalogenation and halogenation reactions. Chemosphere 522, 299–312. Verry, E.S., 1975. Streamﬂow chemistry and nutrient yields from upland-peatland-watersheds in Minnesota. Ecology 56, 1149–1157. Vitt, D.H. and Bayley, S., 1984. The vegetation and water chemistry of four oligotrophic basin mires in northwest Ontario. Can. J. Bot. 62, 1485–1500. Warner, J.A., Casey, W.H., and Dahlgren, R.A., 2000. Interaction kinetics of I2aq with substituted phenols and humic substances. Environ. Sci. Technol. 34, 3180–3185. Winterton, N., 2000. Chlorine: the only green element – towards a wider acceptance of its role in natural cycles. Green Chem. 2, 173–225. Yamoda, Y., 1968. Occurrence of bromine in plants and soil. Talanta 15/11, 1135–1141. Yuita, K., 1994. Overview and dynamics of iodine and bromine in the environment. Jpn. Agric. Res. Q. 28, 90–99.

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

465

Chapter 20

Mercury in mires H. Biester, R. Bindler and A. Martı´ nez Cortizas

Introduction Long-range transport of mercury (Hg) emitted from anthropogenic sources is well documented. Coal burning, waste incineration, and chlor-alkali plants located at mid-latitudes are common sources, and the atmospheric transport of mercury to remote areas and subsequent deposition are well studied (Nriagu and Pacyna, 1988; Steinnes and Andersson, 1991; Slemr and Langer, 1992; Mason et al., 1994; Pacyna and Keeler, 1995; AMAP, 1998; Hermanson, 1998). Although there are uncertainties over the importance of natural geologic sources for the mercury found in peat and lake sediments (Rasmussen, 1994; Fitzgerald et al., 1998), it is now well accepted that the evidence for an increase in anthropogenic mercury emissions relative to natural sources since the Industrial Revolution, is unambiguous. In the early 1990s, Mason et al. (1994) estimated that 70–80% of the modern atmospheric mercury ﬂux can be attributed to anthropogenic sources. A better understanding of the behavior of mercury in the environment is needed for a number of reasons. For example, increased biomagniﬁcation of mercury in aquatic food chains, especially in ﬁsh, and enhanced accumulation in remote areas such as the Arctic have been observed in the last few decades. Mercury toxicity in aquatic ecosystems is of particular concern, with the role of methylmercury (MeHg) being critical. This compound can be concentrated by more than a million times in the aquatic food chain (Grigal, 2002). Biogeochemical studies and monitoring programs, which include direct measurements of wet deposition or indirect measurements based on biomonitoring of forest mosses, have established that anthropogenic activities have affected the global cycling of mercury. Although a precise link has yet to be made between the increased content of mercury in biota and the increased accumulation rates observed in natural environmental archives, such as peat, lake sediments, and glacial ice, there is broad consensus that these archives provide a means to reconstruct atmospheric deposition trends at local, regional, and global scales (Jackson, 1997; Fitzgerald et al., 1998).

ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09020-1

466

H. Biester, R. Bindler, A. Martı´nez Cortizas

Most importantly, the natural archives allow us to look at aspects of the mercury cycle on timescales not available in monitoring programs. This chapter gives an overview of the present knowledge of mercury enrichment in mires and bogs, which may be inﬂuenced both by external and internal processes, and progress in determining past deposition rates of mercury. Bogs in particular have been widely used to evaluate long-term records of atmospheric mercury deposition with the goals of estimating the natural, pre-anthropogenic ﬂux of mercury and quantifying the increase in atmospheric mercury ﬂuxes during the industrial age. The use of data from bogs has deﬁnite advantages over other archives. For example, lake sediment cores, unlike ombrotrophic peat cores, are inﬂuenced by additional inputs to the lake from the surrounding catchment, and separating atmospheric from catchment contributions of mercury in lake sediment studies is a difﬁcult and complicated task (Swain et al., 1992; Fitzgerald et al., 2005). Ice cores, on the other hand, have the disadvantage of a limited geographical distribution. Bogs are found at nearly all latitudes where precipitation is sufﬁcient, and consequently, they have received increasing attention as archives of historical deposition of metals.

Atmospheric deposition and retention of mercury Three major mercury species occur in the atmosphere: gaseous elemental Hg(0), gaseous inorganic Hg(II) compounds (reactive gaseous mercury), and particulatephase mercury (Schroeder and Munthe, 1998). Elemental mercury is dispersed globally, and its residence time in the atmosphere is estimated to be 1–2 years (Fitzgerald and Mason, 1996), which is sufﬁciently long to allow some mixing between the northern and southern hemispheres. Transport of Hg(II) and particulate mercury ranges between tens to hundreds of kilometers depending on particulate size and mass (Schroeder and Munthe, 1998). More than 95% of atmospheric mercury is Hg(0), but the relative amounts of mercury species are source dependent and can be altered by oxidation–reduction reactions (Iverfeldt and Lindqvist, 1986; Munthe, 1992). Although atmospheric mercury is dominated by Hg(0), gaseous Hg(II) is much more soluble and is the dominant form in precipitation (Porcella, 1994; Fitzgerald and Mason, 1996). Wet deposition of mercury may vary to a large extent depending on local emission sources and weather conditions, but most values for recent deposition in the literature range between 5 and 15 mg m2 yr1 in the northern hemisphere, with a mean value of 10 mg m2 yr1 (Grigal, 2002). Data on dry mercury deposition are comparatively rare, but Lamborg et al. (1995) estimated that about 25% of the total mercury deposition directly to lakes in north central Wisconsin (USA) was dry deposition of particulate mercury, and Rea et al. (2001) estimated about 20% of the mercury deposited as throughfall in a mixed forest setting was dry deposition. The retention of mercury by peat in general and more speciﬁcally under changing surface wetness has not been investigated. Because of the high afﬁnity of mercury to bind to humic substances it is most likely that mercury in peat is mainly retained through binding to humiﬁed organic matter (Benoit et al., 1994), preferentially to

Mercury in mires

467

reduced sulfur groups (Skyllberg et al., 2003), and that plant uptake is of minor importance. Moreover, it is generally assumed that organically bound mercury is stable and immobile in peat as soon as it enters the catotelm, the permanently anoxic part of a bog. Little is known about the occurrence of deﬁned mercury species in peat. Formation of Hg(0) as a result of Hg(II) reduction by humic acids (Alberts et al., 1974; Allard and Arsenie, 1991) and subsequent degassing most likely occurs as is the case for soils. Martı´ nez Cortizas et al. (1999) assumed that the existence of Hg(0) and labile Hg(II) species in peat is dependent on climatic conditions. They concluded from thermal release experiments that Hg(0) is the dominant form of Hg in peat formed during cold and dry climates, whereas Hg(II) species are dominant in warmer and wetter periods. Although production of methylmercury (CH3Hg+) in mires and its importance for surface waters is an important area of research (Branﬁreun et al. 1999), there are few data on mono- or di-methylmercury in ombrotrophic peat. In one Swedish bog (Bindler, unpublished data) concentrations of methylmercury are in the range from 0.1 to 12 ng g1, which represents from 1% to as much as 35% of the total Hg in the peat samples. This range is similar to values reported for upland peatlands and mires (r7 ng meHg g1; Branﬁreun and Roulet, 2002).

Mercury concentrations in peat Concentrations of mercury in peat do not necessarily reﬂect atmospheric deposition because concentrations do not depend solely on atmospheric ﬂuxes, but also on net peat accumulation rates (Biester et al., 2003). Moreover, atmospheric mercury mainly exists in a gaseous form (Hg(0)) and only a small portion is bound to particles. Because mercury and other crustal metals are completely decoupled, mercury concentrations cannot be normalized to conservative elements such as titanium or aluminum to compensate for changes in peat accumulation and to distinguish between natural and anthropogenic fractions, as has been done for other metals such as lead (Shotyk et al., 2002). Mercury concentrations in ombrotrophic peat dated to pre-industrial (prior to ca. 150 cal yr BP) or pre-historical times (ca. 2000 cal yr BP or earlier) show generally low values in the range of ca. 5–50 ng Hg g1. Important natural sources of atmospheric mercury are volcanic eruptions, degassing of the Earth crust, emission from soils, and biologically induced evasion of mercury from the oceans (Vandal et al., 1993). In some studies, increased mercury concentrations in pre-industrial peat sections have been assigned to volcanic eruptions (Roos-Barraclough et al., 2002), a relationship that can be seen in the recent glacial record from the Upper Fremont Glacier in Wyoming, USA (Schuster et al., 2002), but other peat studies see no direct link between volcanic emissions and mercury enrichment in peat (Pheiffer-Madsen, 1981; Biester et al., 2003). Higher mercury concentrations in the uppermost (youngest) peat are attributed to increased atmospheric deposition caused by anthropogenic emission. In contrast, higher values in the deepest and often more minerogenic parts of peat cores are considered to be inﬂuenced by external factors such as inﬂuxes from groundwater or weathered bedrock. Mercury concentrations in peat formed in the past 200 years

H. Biester, R. Bindler, A. Martı´nez Cortizas

468 Table 20.1.

Geographical variability of mercury concentrations in bogs and fens.

Location

Hg ng g1 Peat type

Age (cal yr BP)

Reference

Denmark Denmark Greenland Scandinavia Switzerland Switzerland Canada Sweden S Chile S Chile Spain Minnesota Faroe Islands Slovenia

67–741 5–280 15–180 12–194 10–280 40–280 20–100 10–260 10–157 60–570 20–405 40–280 70–500 74–380

200 3000 3000 120 12,000 2000 5900–800 5500 15,000 11,000 5000 1750 AD (210Pb date) 5240 n.d.

Pheiffer-Madsen (1981) Shotyk et al. (2003) Shotyk et al. (2003) Jensen and Jensen (1991) Roos-Barraclough et al. (2002) Roos-Barraclough et al. (2002) Givelet et al. (2004) Bindler (2003) Biester et al. (2003) Franze´n et al. (2004) Martı´ nez Cortizas et al. (1999) Benoit et al. (1994) Shotyk et al. (2005) Biester (unpub.)

Ombrogenic Ombrogenic Minerogenic Ombrogenic Ombrogenic Minerogenic Minerogenic Ombrogenic Ombrogenic Minerogenic Ombrogenic Ombrogenic Minerogenic Ombrogenic

vary from o100 to 4700 ng g1 (Table 20.1). Mercury concentrations are generally higher in the mid-latitudes of the northern hemisphere, where most of the emission sources are, and the enrichment of mercury in peat in northern Scandinavia (Jensen and Jensen, 1991; Bindler et al., 2004) and the Arctic indicates mercury transport from mid- to high-latitudes. Single peat cores are commonly used to reconstruct regional mercury deposition rates, because peat cores are assumed to reﬂect absolute deposition rates. However, in a comparison of mercury records from nine hummock cores collected from a 2000 m2 area in a Swedish bog, Bindler et al. (2004) showed that the maximum mercury concentration and cumulative inventories (the total amount of mercury accumulated per square meter during the past 110 years) both vary by a factor of 4 among the cores (maximum concentrations from 130 to 460 ng Hg g1 and 110-year inventories of 0.85–3.4 mg Hg m2). Lead concentrations and inventories varied also, but the difference among cores was within a factor of 2. This variation for mercury within the investigated bog was greater than the difference between any one of these cores and single cores measured from other bogs located within a distance of 60 km. Malmer and Walle´n (1999) observed similar within-bog spatial variations for atmospherically supplied 210Pb, which were related to micro-topography. Bindler et al. (2004) attributed the large variations in mercury concentrations and inventories to small spatial-scale differences in vegetation and micro-topography on the surface of the bog suggesting an inﬂuence on the interception and retention of mercury. They concluded that reconstructions of past mercury accumulation that are based only on single peat cores do not necessarily provide a representative ﬂux for the entire bog. To overcome the potential constraints of single records, they recommended that data from multiple sites or at least multiple cores be used to scale up to regionally valid models of past mercury deposition, which is an approach commonly used in paleolimnology (Lamborg et al., 2002; Rippey and Douglas, 2004). In addition to these differences in solid-phase concentrations, Branﬁreun (2004) found that porewater chemistry (methylmercury, sulfate, and DOC) varied spatially in relation to micro-topography. Such a variation in chemistry suggests that internal

Mercury in mires

469

processes, such as the bacterially mediated processes that produce methylmercury, may be important for peat chemistry.

Mercury concentrations and peat decomposition Mercury concentrations in peat can be strongly inﬂuenced by peat decomposition (Biester et al., 2003). Mercury is enriched in highly decomposed peat sections, whereas concentrations are lower in relatively less decomposed peat sections because the loss of peat mass is lower. Based on this, most changes in mercury concentrations in bogs have been explained by differences in peat decomposition (mass loss). Diagenetic effects within the peat are superimposed on any changes in mercury concentration that may be related to past changes in atmospheric ﬂuxes (wet and dry deposition). This relationship between decomposition and mercury concentrations can be shown for bogs from different locations. Figure 20.1 shows depth proﬁles of mercury concentrations and carbon/nitrogen (C/N) ratios in peat cores from Chile (Magellanic Moorlands). The C/N ratio is commonly used as an indicator for the degree of decomposition of peat and mass loss (Kuhry and Vitt, 1996; Malmer and Walle´n, 2004), where lower C/N ratios indicate a higher degree of peat decomposition (greater carbon loss). In the three peat cores, lower C/N ratios are clearly associated with higher mercury concentrations. Peat decomposition and mass loss in bogs is mainly controlled by changes in the water table (surface wetness). Peat decomposition is highest during dry periods, when formerly anaerobic peat sections become aerated and metabolized (Malmer and Walle´n, 2004). Thus, mercury is enriched in highly decomposed peat sections as a result of mass loss (carbon) during mineralization of organic matter. Mercury enrichment results from aerobic decay related to a lowered water table, which affects

Figure 20.1. Historical mercury concentrations in peat compared to peat decomposition (C/N ratios) in historical peat proﬁles taken from bogs in southern Chile (Skyring, Patagonia), NW Spain (PDC), and Sweden (Dumme Mosse).

470

H. Biester, R. Bindler, A. Martı´nez Cortizas

previously deposited mercury and peat. This ﬁnding is opposite to what is expected from changes in atmospheric deposition, where higher ﬂuxes and higher concentrations of metals in peat are expected during wet periods and vice versa. Whereas the water table in a bog may normally ﬂuctuate by about 10–15 cm (Damman, 1978), which might affect peat that is tens to a hundred years in age; lower water tables in dry periods and the aeration of formerly anaerobic peat can affect peat that is up to a thousand years old; and the subsequent enrichment of mercury is always younger than the deposition of mercury. The role of peat decomposition in the enrichment of mercury in peat also becomes clear when mercury concentrations in minerogenic mires (fens) are compared to those in bogs located in the same area. An important assumption in such a comparison is that the fens and bogs in question are sufﬁciently close that relatively similar levels of atmospheric deposition occur. A second assumption is that groundwater is not an important additional source of mercury, which is reasonable given the low hydraulic conductivity and low porewater mercury concentrations typically measured in peat (Branﬁreun et al., 1999). As compared to peat in bogs, peat in fens usually shows a higher degree of decomposition because the higher availability of nutrients and the higher pH in fens support more intense decomposition by microorganisms. The higher degree of peat decomposition and also the higher turnover of biomass in fens cause a stronger enrichment of mercury than in bogs. Figure 20.2 shows historical mercury records derived from a fen (GC2) and a bog (GC1) in Patagonia located within ca. 1 km of each other. Mercury concentrations in the fen are on average four times higher than in the bog. Although some of the mercury in the fen may be from groundwater, the most important explanation for a fourfold higher concentration is that the lower net mass accumulation in the fen peat as compared to the bog indicates a higher turnover of organic mass, and that mercury is enriched in the fen peat during the mineralization of the organic matter. Support for an enrichment of mercury in fen peat is provided by other studies where peat from mires that are geographically close have been examined. Table 20.1 shows the generally higher mercury concentrations in minerogenic peat relative to ombrotrophic peat from comparable locations. Even for raised bogs that were initially fens, higher mercury concentrations typically are found in the underlying minerotrophic peat. For example, in Dumme Mosse (Fig. 20.1; Bindler, 2003) mercury concentrations in the underlying fen peat (4350 cm depth) are in the range of 25–230 ng g1 as compared to values of about 10 ng g1 in the overlying ombrotrophic peat (from 60 to 350 cm depth). Furthermore, Biester et al. (2003) observed a sixfold increase in mercury concentrations in the peat ca. 25 cm above a tephra layer in a peat record from Patagonia. The tephra layer provided a ready source of nutrients that contributed to increased microbial peat decomposition, which correlates with a higher concentration of mercury. The peat section with enriched mercury concentrations, which dates to ca. 2700 cal yr BP, does not coincide with any regional volcanic activity, thereby excluding volcanic emissions as an explanation for this enrichment (Biester et al., 2003). The relationship between peat decomposition and mercury concentrations clearly indicates that the concentration of mercury in peat is signiﬁcantly inﬂuenced by diagenetic processes (in Dumme Mosse, for example, C/N ratios and mercury

Mercury in mires

471

Figure 20.2. Historical records of mercury in a bog (GC1) and a fen (GC2) in the same area of the Magellanic Moorlands, Chile (after Franze´n et al., 2004).

concentrations are highly correlated (R2 of 0.67)). The effects of these processes have to be considered if mercury concentrations in different bogs are to be compared and used to study differences in atmospheric ﬂuxes. Although the mass balance of mercury in lakes and lake sediments is well researched (Fitzgerald et al., 2005), in bogs it is not understood. For example, based on background concentrations in peat in the range of 10–30 ng g1 and the fact that deeper peat only constitutes 10–20% of the original peat mass, the original mercury concentrations must have been in the range of only 0.5–3 ng g1. This range is quite low in comparison to the concentrations measured in the uppermost peat and living plant material, typically ca. 40–200 ng Hg g1.

Mercury accumulation rates Mercury accumulation rates in peat cores are considered to reﬂect atmospheric ﬂuxes more accurately than concentrations, because it is assumed that accumulation rates

H. Biester, R. Bindler, A. Martı´nez Cortizas

472 Table 20.2.

Mercury accumulation rates derived from bogs and fens at different locations.

Location

Factor of Background Maximum (mg m2 yr1) (mg m2 yr1) increase

Reference

SW Greenland Patagonia, Chile Ontario, Canada Ontario, Canada Ontario, Canada Minnesota, USA NW Spain NW Spain NW Spain S. Sweden (average) DM SM1 SM2 SM3 TM LM Denmark EGr, Switzerland TGe, Switzerland Mean Median

0.4 1 1.4 1.4 1.4 4.3 0.8 1.8 0.8 0.73 0.6 0.8 0.8 0.8 0.9 0.6 – 1 1.6 1.2 0.5

Shotyk et al. (2003) Biester et al. (2003) Givelet et al. (2003) Givelet et al. (2003) Givelet et al. (2003) Benoit et al. (1994) Martı´ nez Cortizas et al. (1999) Martı´ nez Cortizas et al. (1999) Martı´ nez Cortizas et al. (1999) Bindler et al. (2004) Bindler (2003) Bindler et al. (2004) Bindler et al. (2004) Bindler et al. (2004) Bindler et al. (2004) Bindler (2003) Shotyk et al. (2003) Roos-Barraclough and Shotyk (2003) Roos-Barraclough and Shotyk (2003)

164 63 54 89 141 38 87 144 44 37 24 16 44 63 29

410 63 39 63.4 101 9 109 80 55 50 41 20 55 79 32

184 29 43 64 43

29 27 68 37

are independent of net mass accumulation. Most of the data on mercury accumulation in bogs are from sites located in the northern hemisphere (Grigal, 2003; references in Table 20.2). Data from the southern hemisphere are still very limited. Median pre-industrial mercury accumulation rates in peat are about 1 mg m2 yr1 and vary within a comparatively small range of 0.6–1.7 mg m2 yr1 at most sites (Table 20.2). This accumulation in peat is about one-third of the background atmospheric mercury deposition rate of 3–3.5 mg m2 yr1 estimated from lake sediments, after the accumulation rate of mercury in lake sediments has been corrected for the inﬂuence of catchment size and sediment focusing (Engstrom et al., 1994; Lorey and Driscoll, 1999; Lamborg et al., 2002; Fitzgerald et al., 2005). A clear relationship between background mercury accumulation rates and climate conditions or geographical location remains elusive. Modern mercury accumulation rates in bogs at different locations vary across a wide range from 16 to 184 mg m2 yr1, which corresponds to an average increase in the industrial age by a factor of 69 (9–410, median 38) compared to the background values in peat (Table 20.2). The highest mercury accumulation rates in the industrial period occur at higher latitudes, where they exceed those found in bogs of midlatitudes by factors of ca. 3–4. The highest reported modern mercury accumulation rates are from a bog in Denmark (184 mg m2 yr1) and a fen in southern Greenland (164 mg m2 yr1) (Shotyk et al., 2003). The high modern mercury accumulation rates reported for some Arctic sites, such as southern Greenland, suggest a long-range transport of mercury from middle to higher latitudes, as is also reported for other contaminants such as persistent organic pollutants (Mackay et al., 1995). However, the average mercury accumulation rate

Mercury in mires

473

derived from bogs is a factor of 6 (up to 18) higher than the average mercury deposition rates in the 1990s of ca. 10 mg m2 yr1, determined from direct measurements of wet deposition (Grigal, 2002). There is also poor consistency in the timing of the maximum Hg accumulation rate in the late 20th century derived from bogs at different locations. Whereas peat, lake sediment, and ice data show a consistent record for the past deposition of lead and its maximum inﬂuence about 1970 (Weiss et al., 1999; Renberg et al., 2001), the peak accumulation of mercury varies from 1950 to the present both between and within regions. This temporal variation could be the result of the proximity of speciﬁc sites to point sources and differences in emission histories, but could also result from differences in dating methodology and modeling. For example, 14C bomb-pulse-dated cores generally suggest an earlier peak in mercury deposition in the 1950s (Givelet et al., 2003; Shotyk et al., 2003), whereas 210Pb-dated cores suggest a peak two decades later (Norton et al., 1997; Roos-Barraclough and Shotyk, 2003). Dating of peat cores and modeling chronologies are clearly important for detailed comparison of trends and accumulation rates.

Mercury accumulation rates and inﬂuence of peat diagenesis The term ‘mercury accumulation rate’ (or more accurately, the net accumulation rate) describes the amount of mercury that has been retained in a peat section per unit area within a given time period. The calculation of mercury accumulation rates is based on mercury concentration, sample thickness and density, and the time represented by a peat layer. This deﬁnition neglects all processes that cause losses of mercury through different pathways, such as the release of mercury by soluble organic complexes or re-emission as volatile Hg(0). It does not consider any changes in mercury retention caused by different plants, changes in the bog topography (particularly in regard to the speciﬁc coring site), or changing climatic conditions. The use of mercury accumulation rates to reconstruct past atmospheric mercury deposition assumes that all variables except deposition are not altered through time; that is, the net accumulation of mercury is the same as the original deposition rate. As stated earlier, the common assumption is also that atmospheric mercury deposition and the subsequent accumulation of mercury in peat are uniform across the surface area of a bog. This requires that the factors used to calculate mercury accumulation rates (peat mass accumulation and mercury concentrations) show a linear relationship through hundreds or thousands of years of peat formation and decomposition. Organic matter in bogs undergoes dramatic changes during its diagenesis. One of the major problems affecting metal records in peat is the intense loss of mass during humiﬁcation and mineralization of organic matter and the related enrichment of metals. The mass loss is most intense during the initial phase of organic matter decay in the acrotelm, where as much as 90% of the original plant mass is lost, whereas another 50% of the remaining organic matter is mineralized within a period of ca. 2000 years (Kuhry and Vitt, 1996; Malmer and Walle´n, 1999, 2004). In most bogs a strong increase in mercury accumulation rates occurs in the upper peat sections. In Spain, where the mining of mercury began at least 2000 years ago, an increase in mercury accumulation rates in a bog in the northwest starts as early as

474

H. Biester, R. Bindler, A. Martı´nez Cortizas

2000 years ago, when it increased by 1.5–1.8 times the pre-historical background values (Martı´ nez Cortizas et al., 1999). This early increase may be anthropogenic or it may be natural, but the lack of supporting evidence in other archives such as lake sediment, ice cores, or peat cores from other areas, makes this interpretation uncertain. A possible explanation for major temporal changes in peat mercury records is suggested by Biester et al. (2003, 2004). They believe that accumulation rates (as well as concentrations) of mercury and other organically bound elements are inﬂuenced by peat decomposition processes. They observed a pre-industrial twofold increase in accumulation rates of mercury and other elements in bogs located in the Magellanic Moorlands (531S latitude), which could not be explained by pollution or climatic changes. Since carbon (mass) accumulation rates show a parallel increase to mercury and other organically bound elements, they concluded that the increase in mercury accumulation is probably inﬂuenced by the increase in mass accumulation. The relationship between mass and mercury accumulation especially in the uppermost peat sections is not fully understood. The parameters used to calculate mercury accumulation rates in bogs (density, thickness, and mercury concentration) can change through time in a non-linear way. Mass loss in bogs not only increases mercury concentrations, but it also prevents density from increasing in deeper peat layers, where compaction is assumed to increase. Another important question in this context is, whether the peat mass/mercury relation stays constant through time and space or if there is, for example, lateral mass movement especially at the acrotelm/catotelm boundary, which might cause thinning of peat sections and consequently reduction of calculated accumulation rates. As a ﬁrst attempt Biester et al. (2003) normalized mercury accumulation to carbon accumulation rates (correcting for mass loss) and found that the corrected mercury net accumulation rate in recent peat is about 3–5 times greater than in the pre-industrial period. This increase agrees with values derived from lake sediments, which also indicate a 3–5 times increase in the late 20th century compared to pre-industrial values. Similarly, increases in mercury accumulation rates of about 30 times in bogs in Spain (Martı´ nez Cortizas et al., 1999) are reduced by 80% when corrected for carbon accumulation. The method of correcting mercury accumulation rates based on carbon accumulation could be also applied to other organically bound elements such as bromine or selenium (Fig. 20.3).

Hg accumulation and climate variation Accumulation rates of mercury and bromine were found to show similar patterns in the peat cores from Spain (Fig. 20.3), Patagonia (Biester et al., 2003, 2004), and Dumme Mosse in Sweden (R2 ¼ .55; unpublished data) as well as in peat cores from Switzerland and Canada (Roos-Barraclough and Shotyk, 2003; Givelet et al., 2004). For the Swiss and Canadian studies the authors interpreted the variations in bromine accumulation as reﬂecting changes in past atmospheric ﬂuxes. Assuming that there is a relationship between natural mercury deposition and bromine deposition, they then corrected mercury accumulation rates for changes in natural atmospheric ﬂuxes

Mercury in mires

475

Figure 20.3. Net accumulation rates of mercury, bromine, selenium, and carbon in a Spanish bog.

using bromine in order to estimate the increase in mercury caused by anthropogenic emissions. If wet deposition is seen as the most important pathway of atmospheric mercury deposition to bogs, this approach would imply that mercury accumulation rates in peat should increase during wet periods. However, data from Spanish bogs suggest a contrary relationship with climate. The accumulation of mercury, bromine, and selenium correlate with climate proxies (pollen and non-pollen palynomorphs; Mighall et al., 2006), indicating that elevated mercury accumulation rates occurred during relatively dry periods as a result of increased peat decomposition. The higher frequency of low water table events in bogs during dry periods causes increased peat mineralization and mass loss and a related increase in mercury concentrations as a result of aeration of the upper peat layers. The observation that this process affects all organically bound elements to a similar extent explains the good correlation between mercury and bromine in many bogs. The assumption that bromine can be used to estimate the natural fraction of mercury deposition is also problematic because the long-term trends in bromine accumulation in peat do not follow patterns observed in glacial records. This suggests that mechanisms other than atmospheric deposition inﬂuence the net accumulation of bromine in peat (Biester et al., 2004). In other words, mercury accumulation rates in bogs are not predominately controlled by atmospheric deposition. Climatic conditions have an important inﬂuence

476

H. Biester, R. Bindler, A. Martı´nez Cortizas

on the concentration of mercury in bogs, mainly through changes in peat decomposition and to a lesser extent through variations in derived atmospheric ﬂuxes. Resolving the effects of peat diagenesis on mercury accumulation rates in the peat would likely improve the comparability of the mercury records in peat and lake sediments.

Outlook In this chapter, we have focused on the dependency of mercury enrichment in bogs on climate. Changes in climate may signiﬁcantly alter the biogeochemical cycling of mercury in wetlands and soils and inﬂuence its release to aquatic environments. It is clear that we need to understand better the relationship between mercury accumulation rates, peat diagenesis, and climate. There are obvious differences in the estimates of atmospheric mercury deposition between lake sediments and bogs and also among bogs. We suggest that careful incorporation of results from studies of carbon dynamics and peat mass accumulation can be an important step for improving our understanding of past mercury cycling in bogs.

References Alberts, J.J., Schindler, J.E., Miller, R.W., and Nutter, D.E., 1974. Elemental mercury evolution mediated by humic acid. Science 184, 895–896. Allard, B. and Arsenie, I., 1991. Abiotic reduction of mercury by humic substances in aquatic system – an important process for the mercury cycle. Water Air Soil Pollut. 56, 457–464. AMAP, 1998. Arctic Pollution Issues: A State of the Arctic Environment Report. Arctic Monitoring and Assessment Programme, Oslo. Benoit, J.M., Fitzgerald, W.F., and Damman, A.W.H., 1994. Historical atmospheric mercury deposition in the mid-continental U.S. as recorded in an ombrogenic peat bog. In: Huckabee, J. and Watras, C. (Eds), Mercury Pollution: Integration and Synthesis. Lewis Publishers, Chelsea, MI, pp. 187–202. Biester, H., Keppler, F., Putschew, A., et al., 2004. Halogen retention, organohalogens, and the role of organic matter decomposition on halogen enrichment in two Chilean peat bogs. Environ. Sci. Technol. 387, 1984–1991. Biester, H., Martı´ nez Cortizas, A., Birkenstock, S., and Kilian, R., 2003. Effect of peat decomposition and mass loss on historic mercury records in peat bogs from Patagonia. Environ. Sci. Technol. 37, 32–39. Bindler, R., 2003. Estimating the natural background atmospheric deposition rate of mercury utilizing ombrotrophic bogs in south Sweden. Environ. Sci. Technol. 37, 40–46. Bindler, R., Klarqvist, M., Klaminder, J., and Fo¨rster, J., 2004. Does within-bog spatial variability of mercury and lead constrain reconstructions of absolute deposition rates from single peat records? The example of Store Mosse, Sweden. Global Biogeochem. Cycles 18, GB3020. Branﬁreun, B.A., 2004. Does microtopography inﬂuence subsurface pore water chemistry? Implications for the study of methylmercury in peatlands. Wetlands 24, 2007–2011. Branﬁreun, B.A. and Roulet, N.T., 2002. Controls on the fate and transport of methylmercury in a boreal headwater catchment, northwestern Ontario. Hydrol. Earth Syst. Sci. 6, 785–794. Branﬁreun, B.A., Roulet, N.T., Kelly, C.A., and Rudd, J.W.M., 1999. In situ sulphate stimulation of mercury methylation in a boreal peatland: toward a link between acid rain and methylmercury contamination in remote environments. Global Biogeochem. Cycles 13, 743–750. Damman, A.W.H., 1978. Distribution and movement of elements in ombrotrophic peat bogs. Oikos 30, 480–495.

Mercury in mires

477

Engstrom, D.R., Swain, E.B., Henning, T.A., et al., 1994. Atmospheric mercury deposition to lakes and watersheds. In: Baker, L. (Ed.), Environmental Chemistry of Lakes and Reservoirs. American Chemical Society, Washington, DC, pp. 33–66. Fitzgerald, W.F., Engstrom, D.R., Lamborg, C., et al., 2005. Modern and historic atmospheric mercury ﬂuxes in northern Alaska: global sources and Arctic depletion. Environ. Sci. Technol. 39, 557–568. Fitzgerald, W.F., Engstrom, D.R., Mason, R.P., and Nater, E.A., 1998. The case of atmospheric mercury contamination in remote areas. Environ. Sci. Technol. 32/1, 1–7. Fitzgerald, W.F. and Mason, R.P., 1996. The global mercury cycle: oceanic and anthropogenic aspects. In: Baeyens, W., Vasilievm, O., and Ebinghaus, R. (Eds), Regional and Global Mercury Cycles: Sources, Fluxes and Mass Balances. Kluwer, The Netherlands, pp. 85–108. Franze´n, C., Kilian, R., and Biester, H., 2004. Natural sources of mercury enrichment in a minerogenic peat bog in Southern Patagonia (531S). J. Environ. Monitor. 6, 466–472. Givelet, N., Roos-Barraclough, F., Goodsite, M.E., et al., 2004. Atmospheric mercury accumulation rates between 5900 and 800 calibrated years BP in the high Arctic of Canada recorded by peat hummocks. Environ. Sci. Technol. 38, 4964–4972. Givelet, N., Shotyk, W., and Roos-Barraclough, F., 2003. Rates and predominant anthropogenic sources of atmospheric mercury accumulation in southern Ontario recorded by peat cores from three bogs: comparison with natural ‘‘background’’ values (past 8000 years). J. Environ. Monitor. 5, 935–949. Grigal, D.F., 2002. Inputs and outputs of mercury from terrestrial watersheds: a review. Environ. Rev. 10, 1–39. Grigal, D.F., 2003. Mercury sequestration in forests and peatlands: a review. J. Environ. Qual. 32, 393–405. Hermanson, M.H., 1998. Anthropogenic mercury deposition to arctic lake sediments. Water Air Soil Pollut. 101, 309–321. Iverfeldt, A˚. and Lindqvist, O., 1986. Atmospheric oxidation of elemental mercury by ozone in the aqueous phase. Atmos. Environ. 20, 1567–1573. Jackson, T.A., 1997. Long-range atmospheric transport of mercury to ecosystems, and the importance of anthropogenic emissions – a critical review and evaluation of published evidence. Environ. Rev. 5, 99–120. Jensen, A. and Jensen, A., 1991. Historical deposition rates of mercury in Scandinavia estimated by dating and measurement of mercury in cores of peat bogs. Water Air Soil Pollut. 56, 769–778. Kuhry, P. and Vitt, D.H., 1996. Fossil carbon/nitrogen ratios as a measure of peat decomposition. Ecology 77, 271–275. Lamborg, C., Fitzgerald, W., Vandal, G., and Rolfhus, K., 1995. Atmospheric mercury in northern Wisconsin: sources and species. Water Air Soil Pollut. 80, 189–198. Lamborg, C.H., Fitzgerald, W.F., Damman, A.W.H., et al., 2002. Modern and historic atmospheric mercury ﬂuxes in both hemispheres: global and regional mercury cycling implications. Global Biogeochem. Cycles 16 (4), Art. No. 1104. Lorey, P. and Driscoll, C.T., 1999. Historical trends of mercury deposition in Adirondack lakes. Environ. Sci. Technol. 33, 718–722. Mackay, D., Wania, F., and Schroeder, W.H., 1995. Prospects for modelling the behavior and fate of mercury globally and in aquatic systems. Water Air Soil Pollut. 80, 941–950. Malmer, N. and Walle´n, B., 1999. The dynamics of peat accumulation on bogs: mass balance of hummocks and hollows and its variation throughout a millennium. Ecography 22, 736–750. Malmer, N. and Walle´n, B., 2004. Input rates, decay losses and accumulation rates of carbon in bogs during the last millennium: internal processes and environmental changes. Holocene 14, 111–117. Martı´ nez Cortizas, A., Pontevedra-Pombal, X., Garcı´ a-Rodeja, E., et al., 1999. Mercury in a Spanish peat bog: archive of climate change and atmospheric metal pollution. Science 284, 939–942. Mason, R.P., Fitzgerald, W.F., and Morel, F.M.M., 1994. The biogeochemical cycling of elemental mercury: anthropogenic inﬂuences. Geochim. Cosmochim. Acta. 58, 3191–3198. Mighall, T.M., Martı´ nez Cortizas, A., and Biester, H., 2006. Proxy climate and vegetation change during the last ﬁve millennia in NW Iberia: pollen and non-pollen palynomorph data from ombrotrophic peat bogs in the Xistral Mountains. Rev. Palaeobot. Palynol. 141, 203–223.

478

H. Biester, R. Bindler, A. Martı´nez Cortizas

Munthe, J., 1992. The aqueous oxidation of elemental mercury by ozone. Atmos. Environ. 26A, 1461–1468. Norton, S.A., Evans, G.C., and Kahl, J.S., 1997. Comparison of Hg and Pb ﬂuxes to hummocks and hollows of ombrotrophic Big Heath Bog and to nearby Sargent Mt. Pond, Maine, USA. Water Air Soil Pollut. 100, 271–286. Nriagu, J.O. and Pacyna, J.M., 1988. Quantitative assessment of worldwide contamination of air, water and soils by trace metals. Nature 333, 134–139. Pacyna, J.M. and Keeler, G.J., 1995. Sources of mercury in the Arctic. Water Air Soil Pollut. 80, 621–632. Pheiffer-Madsen, P., 1981. Peat bog records of atmospheric mercury deposition. Nature 293, 127–130. Porcella, D.B., 1994. Mercury in the environment: biogeochemistry. In: Watras, C.J. and Huckabee, J.W. (Eds), Mercury Pollution: Integration and Synthesis. Lewis Publishers, Boca Raton, pp. 3–19. Rasmussen, P.E., 1994. Current methods of estimating atmospheric ﬂuxes in remote areas. Environ. Sci. Technol. 28, 2233–2241. Rea, A.W., Lindberg, S.E., and Keeler, G.J., 2001. Dry deposition and foliar leaching of mercury and selected trace elements in deciduous forest throughfall. Atmos. Environ. 35, 3453–3462. Renberg, I., Bindler, R., and Bra¨nnvall, M.-L., 2001. Using the historical atmospheric lead deposition record as a chronological marker in sediment deposits in Europe. Holocene 11, 511–516. Rippey, B. and Douglas, R.W., 2004. Reconstructing regional-scale lead contamination of the atmosphere (1850–1980) in the United Kingdom and Ireland using lake sediments. Global Biogeochem. Cycles 18, Art. No. GB4026, 10.1029/2004GB002305. Roos-Barraclough, F., Martı´ nez Cortizas, A., Garcia-Rodeja, E., and Shotyk, W., 2002. A 14,500 year record of the accumulation of atmospheric mercury in peat: volcanic signals, anthropogenic inﬂuences and a correlation to bromine accumulation. Earth Planet. Sci. Lett. 6334, 1–18. Roos-Barraclough, F. and Shotyk, W., 2003. Millennial-scale records of atmospheric mercury deposition obtained from ombrotrophic and minerotrophic peatlands in the Swiss Jura Mountains. Environ. Sci. Technol. 37, 235–244. Schroeder, W.H. and Munthe, J., 1998. Atmospheric mercury – an overview. Atmos. Environ. 32, 809–822. Schuster, P.F., Krabbenhoft, D.P., Naftz, D.L., et al., 2002. Atmospheric mercury deposition during the last 270 years: a glacial ice core record of natural and anthropogenic sources. Environ. Sci. Technol. 36, 2303–2310. Shotyk, W., Goodsite, M.E., Roos-Barraclough, F., et al., 2003. Anthropogenic contributions to atmospheric Hg, Pb and As accumulation recorded by peat cores from southern Greenland and Denmark dated using the 14C ‘‘bomb pulse curve’’. Geochim. Cosmochim. Acta 67, 3991–4011. Shotyk, W., Goodsite, M.E., Roos-Barraclough, F., et al., 2005. Accumulation rates and predominant atmospheric sources of natural and anthropogenic Hg and Pb on the Faroe Islands since 5420 14C yr BP recorded by a peat core from a blanket bog. Geochim. Cosmochim. Acta 69, 1–17. Shotyk, W., Weiss, D., Heisterkamp, M., et al., 2002. A new peat bog record of atmospheric lead pollution in Switzerland: Pb concentrations, enrichment factors, isotopic composition, and organolead species. Environ. Sci. Technol. 36, 3893–3900. Skyllberg, U., Qian, J., Frech, W., et al., 2003. Distribution of mercury, methylmercury and organic sulphur species in soil, soil solution and stream of a boreal forest catchment. Biogeochemistry 64, 53–76. Slemr, F. and Langer, E., 1992. Increase in global atmospheric concentrations of mercury inferred from measurements over the Atlantic Ocean. Nature 355, 434–437. Steinnes, E. and Andersson, E.M., 1991. Atmospheric deposition of mercury in Norway: temporal and spatial trends. Water Air Soil Pollut. 56, 391–404. Swain, E.B., Engstrom, D.R., Brigham, M.E., et al., 1992. Increasing rates of atmospheric mercury deposition in midcontinental North America. Science 257, 784–787. Vandal, G.M., Fitzgerald, W.F., Boutron, C.F., and Candelone, J.P., 1993. Variations in mercury deposition to Antarctica over the past 34,000 years. Nature 362, 621–623. Weiss, D., Shotyk, W., Appleby, P.G., et al., 1999. Atmospheric Pb deposition since the industrial revolution recorded by ﬁve Swiss peat proﬁles: enrichment factors, ﬂuxes, isotopic composition, and sources. Environ. Sci. Technol. 33, 1340–1352.

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

479

Chapter 21

Archiving natural and anthropogenic lead deposition in peatlands M.E. Kylander, D.J. Weiss, E. Peiteado Varela, T. Taboada Rodriguez and A. Martı´ nez Cortizas

Introduction Peat deposits have proven to be excellent terrestrial geochemical archives able to record accurately the timing of atmospheric dust deposition and that of several heavy metals as far back as the early Holocene (Shotyk et al., 2001; Weiss et al., 2002a). This has allowed not only reliable determination of natural background concentrations in the air during the last and before the commencement of major human activities ca. 3000 14C yr BP, but also to establish the onset and duration of major pollution episodes and their sources. Much of what is known about heavy metal deposition in peatlands comes from the study of lead. This element is often chosen as it has been, and still is, one of the major pollutants affecting environmental and human health alike; it is largely immobile in bogs covering the time span of major interest. It has an isotopic composition that allows in a nearly unique way accurate source assessment and use of isotope ratios as proxies for pollutant dispersal and climate change. This chapter gives an overview of atmospheric lead deposition and some of the geochemical tools, in particular lead isotopes and enrichment factors (EF), used to decipher the information stored in bogs. We take a look at historical reconstructions of lead deposition made using bogs at several locations, paying particular attention to the discussed interpretive tools.

Lead from a geological perspective Understanding the geochemistry of lead in its natural geological context is pivotal to comprehending human impacts and how to use the various geochemical techniques (such as elemental or isotope ratios) in peat archive analysis. Using the classiﬁcation of Goldschmidt, Pb is a chalcophile element – along with other major metals used in ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09021-3

480

M.E. Kylander et al.

human civilization including Zn, Cu, and Ag – and therefore prefers the sulphide phase to that of the silicate or iron phase (Krauskopf and Bird, 1995). Consequently, terrestrial lead is generally associated with massive sulphide deposits, but also occurs in the crystal structures of rock forming silicates (K-feldspar) and oxides of common crustal rocks. Comparing the Pb content in major global reservoirs, most Pb is locked into the lithosphere (soils, sediments, ca. 5 1019 g) followed by the hydrosphere (ca. 1016 g), and the biosphere (ca. 1012 g). Only a small amount of Pb is stored in the atmosphere (ca. 1010 g), but this reservoir serves as the major transport pathway (Reuer and Weiss, 2002). The main natural source of Pb in sediments and soils (including bogs) is derived from rock weathering. With respect to mineralogical forms, the primary form of Pb in nature is galena (PbS) but its oxidation products like plattnerite (PbO2), cerussite (PbCO3), and anglesite (PbSO4) are also important in the environment. These minerals generally have low solubility (Stumm and Morgan, 1996).

Geochemical tools and interpretation Lead isotopes – powerful tracers and proxies The variations in isotopic composition that exist for many chemical elements have often been exploited in the Earth Sciences. Isotopic systems used include stable, light, non-radiogenic isotopes such as C, O, and S and heavy radiogenic isotopes such as Pb, Sr, Nd, and Os. Whereas variations in stable non-radiogenic isotope compositions are due to mass-dependent fractionation following kinetic or equilibrium processes, heavy elements show very small mass differences relative to the absolute mass, for mass-dependent variations to be exploited. Nevertheless, different stable isotope ratios of heavier elements such as Pb, Nd, and Sr occur in nature because their isotopes are formed by the radioactive decay of different parent nuclides (Rollinson, 1993). Lead isotope geochemistry Lead has four different stable isotopes ranging in atomic mass units from 204 to 208 (204Pb, 206Pb, 207Pb, 208Pb; see Dickin, 1995; Rollinson, 1993; and Doe, 1970, for details). Their natural abundances in lead ores and minerals varies due to the different U and Th decay constants and concentrations, producing 206Pb, 207Pb, and 208Pb from 238U, 235U, and 232Th, respectively. In contrast, 204Pb is not generated by radioactive decay. The distribution of U, Th, and Pb within the lithosphere is highly variable. Uranium and Th are incompatible in silicate minerals, whereas Pb is incompatible in maﬁc minerals but compatible in K-feldspars and apatite. These fractionation processes enrich the upper continental crust (UCC) in U, Th, and Pb and result in high U/Pb and Th/Pb continental crust ratios (Fig. 21.1). In the lower crust, U is depleted relative to Pb, but in the upper crust U is enriched. The distribution of Th within the crust is not well understood, but all three elements are depleted in the

Archiving natural and anthropogenic lead deposition in peatlands

481

Figure 21.1. Development of U/Pb (open symbols) and Th/Pb (closed symbols) ratios in the three geochemical reservoirs of the upper crust (triangles), lower crust (circles), and upper mantle (diamonds). The enrichment of U and Th in the upper crust and the decrease of U and Th due to radioactive decay are clearly seen (Ga ¼ billion of years ago).

mantle relative to the crust (Wedepohl, 1995). Depending on the age of formation, the initial U, Th, and Pb concentrations, and the geological history, the Pb isotope ratios of rocks and minerals can differ signiﬁcantly from one location to another (Dickin, 1995). This simple observation allows, ﬁrstly, the identiﬁcation of different Pb sources in the environment using the unique isotopic character of the geochemical reservoir from which the Pb derives, and secondly, recognition of mixing between isotopically distinct sources. Importantly, Pb isotopes detect changes that may otherwise go unnoticed in concentration data. Table 21.1 consigns isotope ratios to different major geochemical reservoirs including the continental crust (upper, middle, lower) and mantle (depleted and enriched) (Rollinson, 1993). Whereas the different crustal reservoirs are important in forming soil dust, the mantle reservoirs are likely important in tracing possible volcanic sources as these can include crustal and mantle material. The UCC is characterized by U and Th enrichment, and thus higher 206Pb, 207Pb, and 208Pb concentrations. The lower continental crust has lower U/Pb and Th/Pb ratios than modern mantle values, so, 206Pb, 207Pb, and 208Pb isotope concentrations are very low and can be used to distinguish between lower crust and mantle reservoirs. The contrasting variation of 206Pb and 207Pb, both produced from U, is a consequence of their differing radioactive decay rates (Fig. 21.2). During the early history of the Earth, 235U decayed rapidly relative to 238U. Abundances of 207Pb are therefore an extremely sensitive indicator of an old source. Today, however, 235U is largely extinct thus in the recent history of the Earth 238U decay is more prominent, and consequently, 206Pb abundances show a greater spread than that of 207Pb (Rollinson, 1993).

M.E. Kylander et al.

482 Table 21.1.

The isotopic character of crust and mantle reservoirs (after Rollinson, 1993). 238

Reservoir Continental crust

Pb

235

U–

207

Pb

232

Th –

208

Pb

High U/Pb

High U/Pb

High Th/Pb

Middle crust

High 206Pb/204Pb U-depleted Low 206Pb/204Pb

High 207Pb/204Pb U-depleted Low 207Pb/204Pb

Severe U depletion Very low 206 Pb/204Pb ca. 14.0 Low U/Pb

Severe U depletion Low 207Pb/204Pb

High 208Pb/204Pb Moderate high Th Moderate high 208 Pb/204Pb Severe Th depletion Low 208Pb/204Pb

ca. 14.7 Low U/Pb

Th/U ¼ 2.470.4

Low 206Pb/204Pb ca. 17.2–17.7 Low U/Pb

Low 207Pb/204Pb ca. 15.4 Low U/Pb

Low 208Pb/204Pb ca. 37.2–37.4 Th/U

Depleted mantle

EM I

EM II Bulk Earth

206

Upper crust

Lower crust

Enriched mantle

U–

Low 207Pb/204Pb Low 208Pb/204Pb Low 206Pb/204Pb ca. 17.6–17.7 ca. 15.46–15.49 ca. 38.0–38.2 High 207Pb/204 Pb and 208Pb/204Pb at a given 206Pb/204Pb 206 207 208 Pb/204Pb Pb/204Pb Pb/204Pb 18.470.3 15.5870.08 38.970.3

Figure 21.2. Temporal development of 206Pb (open symbols) and 207Pb (closed symbols) concentrations with time. The isotopes are normalized to 204Pb and shown are eroded crust (squares) and the three geochemical reservoirs of the upper crust (triangles), lower crust (circles), and upper mantle (diamonds) (Ga ¼ billion of years ago).

Archiving natural and anthropogenic lead deposition in peatlands

483

Application of lead isotopes The application of lead isotopes to trace anthropogenic sources was ﬁrst suggested four decades ago. Measuring stable lead isotope variability within leaded gasoline, coals, and aerosols, Chow et al. (1975) made the critical observation that a large isotopic range is found in natural and synthetic materials, such as a 14% range in 206 Pb/207Pb in leaded gasoline between Bangkok, Thailand (1.072), and Santiago de Chile, Chile (1.238). As there is no fractionation of isotopes during low- or hightemperature processes, the isotopic dissimilarities arising from the differing origins of Pb used (low 206Pb, 207Pb, and 208Pb concentrations, and high 206Pb, 207Pb, and 208Pb concentrations in Pb derived from old- and young-ore deposits, respectively) are preserved in gasoline exhausts and other industrial sources to aerosols; this allows the tracing of source inputs to the atmosphere. Since then tracing using Pb isotopes has proven a powerful tool in other ﬁelds such as sedimentary source tracing (Landmeyer et al., 2003) and pollution assessment (Gulson et al., 1994). A thorough description of lead isotope techniques (graphical and mathematical) used for source assessment and contribution calculations is given elsewhere (Albare´de, 1995). Here we aim to highlight a few important constraints. In general, source assessment is done using three-isotope plots. Because the three radiogenic isotopes of Pb have different parent nuclides and two parent elements (U and Th), it is crucial to include all three isotopes (206Pb, 207Pb, 208Pb) in such plots. Using only one isotope ratio and assuming two-component mixing without actually proving it using three-isotope plots is dangerous and likely overestimates certain source contributions. The least abundant isotope, 204Pb (approximately 1.4%), is often excluded as it is measured with the least precision. It is important to know that excluding 204Pb from three-isotope plots can miss additional sources, especially when assessing pollution on a local scale where isotopic variations are small (Carignan and Garie´py, 1995; Spiro et al., 2004). This is partly because differences in any of the three radiogenic isotopes relative to 204Pb are larger than to the radiogenic isotope used (due to the addition of radiogenic Pb with time). It is also in part due to the fact that as 204Pb is non-radiogenic, only normalizing to this isotope allows normalization to a variable independent of time and initial U and Th concentrations. An example of the essential information provided by normalizing to 204Pb in three-isotope plots is given in Figures 21.3a,b. Here the known isotope ﬁelds of possible lead sources to two Swiss bogs are plotted: pre-anthropogenic aerosols (PAA) and UCC, ores from Broken Hill Mine (used extensively in Europe from the late 19th century), European coal, incinerator ﬂy ash, and Swiss car exhaust. The linear array on the 206Pb/207Pb vs. 206Pb/204Pb plots suggests two components mixing between PAA/UCC and Broken Hill Mine ores. Conversely, the 207Pb/204Pb vs. 206Pb/204Pb plot suggests that there are multiple sources contributing to the archived aerosol in varying amounts at different times (Weiss et al., 1999). Enrichment factors (EF) – gauging the magnitude Using EF for gauging anthropogenic impact on elemental cycling was ﬁrst introduced by Rahn (1971, 1976) and has been extensively employed since to evaluate

484

M.E. Kylander et al.

Figure 21.3. 206Pb/207Pb vs.206Pb/204Pb and 207Pb/204Pb vs.206Pb/204Pb three-isotope plots for two different bog sites in Western (a) and Central Switzerland (b) showing ﬁelds of possible lead sources and mixing end members (from Weiss et al., 1999; used with the permission of the American Chemical Society, Environ. Sci. Technology).

anthropogenic enrichment of Pb and other heavy metals in surﬁcial systems of the Earth. The intuitive notion behind EF is that natural processes may have varying ﬂuxes of elements, but the ratio between a given element (M) and a lithogenic one (X) (conservative, of natural origin), to acquire a M/X ratio, will remain constant without human perturbation of the cycle of M. EF are calculated by normalizing M/X ratios of record samples to the same ratio in a reference material (Rahn, 1971, 1976; Duce et al., 1975), thus assuming that the latter characterizes the source of the material. To calculate EF for lead, for example, the following equation is used: PbEFReference;X ¼ ðPb=X ÞSample =ðPb=X ÞReference where PbEFReference,X is used to communicate which reference material and which element are used for normalization in the equation, (Pb/X)Sample is the ratio of lead to the selected lithogenic element in a given sample, and (Pb/X)Reference is the ratio of lead to the same selected lithogenic element in a reference material. The reference can be external (the M/X ratio of the continental crust, a given geological material, soil, and so on), or internal (the average of a set of samples from the pre-pollution age of the same record). The main assumption in the use of EF is the relative constancy in the ratios of the elements and, thus, of the mineralogical composition of the deposited dust. If the mineral composition of the airborne dust deposited in a bog remains constant, the ratio between any given pair of conservative elements must also keep constant through time. Constancy is critical for the use of a reference lithogenic

Archiving natural and anthropogenic lead deposition in peatlands

485

element to estimate anthropogenic enrichments due to atmospheric pollution. Where the pre-pollution M/X ratios of a bog record show little or no variation they can reasonably be used to estimate the natural component of the accumulation of element M in anthropogenic times, and also to obtain an estimation of the anthropogenic ﬂuxes. Constancy, however, seems to be the exception rather than the rule. In bog research it has become common practice to calculate EF to evaluate the enrichment/depletion of deposited elements and their sources. More recently however, some authors have challenged the validity of EF in environmental studies and recommended abandoning their indiscriminate use (Reimann and De Caritat, 2000), based precisely on those limitations given previously by Rahn (1976). These limitations include elemental fractionation across the crust–air and sea–air interfaces leading to enrichments or depletions of some elements and the choice of a representative normalizing element and reference material. Other researchers have also suggested a need to review the applicability of EF in bog research oriented towards reconstruction of human activities in present and past atmospheric metal pollution (Weiss et al., 1997; Martı´ nez Cortizas et al., 2002a, b). Here, we brieﬂy discuss the applicability of EF, by reviewing some previous works as well as introducing some examples to evaluate their limitations and the factors from which these arise.

The Swiss experience Table 21.2 summarizes the PbEF values given in the literature for eight ombrotrophic and minerotrophic mires in Switzerland. We have chosen this data set because PbEF were calculated using more than one lithogenic reference element (Sc, Ti, and/ Table 21.2. Maximum enrichment of Pb in eight bogs from Switzerland (EGr: Etang de la Gruere; PRd: Praz Rodet; SwM: Scho¨pfenwaldmoor; HAG: Hagenmoos; TGe: Tourbiere de Genevez; GdL: Gola di Lago; SUO: Suossa; MAU: Mauntschas). PbEFs were calculated to Sc, Ti, and/or Zr. Ratios were normalized to the UCC and the average value of a set of samples of the same peat record. Egr Sc-UCC Sc-Peat Ti-UCC Zr-UCC

2,3,7,9

420 301–954,5,6 1208 5207

Note: 1 Shotyk (1996a); 2 Shotyk (1996b); 3 Shotyk et al. (1996); 4 Shotyk et al. (1997); 5 Shotyk et al. (1998); 6 Weiss et al. (1999); 7 Shotyk et al. (2000); 8 Shotyk (2002); 9 Shotyk et al. (2002b).

PRd 7

350 786 – –

SwM – 106 – 5407

HAG – – – 7007

TGe

GdL

7

7

380 874,6 1158 –

900 1096 – –

SUO 7

400 – – –

MAU – – 708 1307

486

M.E. Kylander et al.

or Zr) and, in some cases, was normalized to two different reference materials: the UCC, or a set of samples of the same peat record, denoted PbEFUCC and PbEFP, respectively. The effect of applying different lithogenic elements can only be evaluated when the UCC is used as the reference material. In this case Zr produced higher enrichment estimates than Sc or Ti. For Etang de la Gruyere, Zr gave a maximum Pb enrichment four times greater than that of Ti, whereas for Tourbiere de Genevez, Sc produced a PbEFUCC three times greater than that obtained with Ti. If a comparison is made between reference values, it appears that normalizing to UCC ratios overestimates the enrichments by a factor of four (Etang de la Gruere, Praz Rodet, Tourbiere de Genevez) to eight times (Gola di Lago) compared to normalizations made using ratios from the same peat record. In light of these results it is questionable whether the use of different lithogenic elements is a reasonable procedure for evaluating metal enrichments in peat records when we intend to compare different areas. For example, the maximum PbEFUCC in Gola di Lago is by far the greatest of those determined for Swiss bogs (2.6 times greater than any other) suggesting a highly polluted site, but the PbEFP is comparable to those of other bogs suggesting a similar degree of pollution. The use of ratios from the same peat record is logically a more realistic reference than any external one as it better reﬂects the mineralogy of the deposited dust. But the question arises as to whether or not dust composition remains constant when a record covers several thousands of years. For Etang de la Gruere, two maximum PbEFP were published, both normalized to the peat record but using different reference periods, resulting in values that differ by more than three times (30 vs. 95)(Shotyk, 1996a, Shotyk et al., 1997; Weiss et al., 1999). This is in agreement with the ﬁnding of Shotyk et al. (2002a) that in Etang de la Guere the Pb/X ratios for samples belonging to the Holocene optimum (8000–6000 cal yr BP) are higher than those of any other pre-pollution period. This effect was more pronounced when EF were calculated using Zr as the reference element. It also may happen that the lithogenic element used for the estimation of the enrichments is representative of a certain mineralogy or even has a non-mineral origin (volcanoes, sea salt, biogenic releases).

Some examples from northwestern Spain The Swiss mires are separated by relatively large distances, have somewhat different local lithology/mineralogy and represent a different history of human activity and inﬂuence on the environment. This complex situation introduces some uncertainties with respect to the above commented aspects, since there are many possible factors responsible for the observed variations. To obtain a deeper insight into the importance of the reference element and background material used we calculated PbEF in three bogs dated to 5000 cal yr BP from NW Spain: Borralleiras de Cal Grande, Penido Vello, and Pena da Cadela. These are located at increasing altitudes (600, 800, and 1000 m, approximately) and at relatively short distances to one another in a north to south transect covering ca. 10 km. Table 21.3 shows each lithogenic reference element used for normalization based on both UCC values (Wedepohl, 1995) and the average ratio of samples of the base of the

K

Ti

Ga

Rb

UCC

Peat

UCC

Peat

UCC

Peat

UCC

Y

Zr

Peat

UCC

Peat

UCC

Peat

BLL

Maximum (Industrial) Present Roman period

173 59 73

34 12 14

43 24 7

58 32 10

15 10 5

16 10 6

89 24 22

45 12 11

– – –

– – –

94 32 13

68 23 9

PVO

Maximum (Industrial) Present Roman period

70 20 42

18 5 11

86 46 28

27 15 9

11 5 7

13 6 9

26 6 31

10 2 12

20 5 10

18 5 9

89 50 37

21 12 9

PDC

Maximum (Industrial) Present Roman period

158 72 42

47 21 13

78 56 15

27 19 8

16 14 12

23 20 17

125 49 40

31 12 10

45 32 14

46 32 15

146 92 27

81 51 15

Source: Data obtained with the ﬁnancial support of projects REN2003-09228-CO2-01 (Spanish Ministry of Science and Technology) and PGIDIT03PXIB20002PR (Xunta de Galicia).

Archiving natural and anthropogenic lead deposition in peatlands

Table 21.3. PbEFs calculated using six different reference elements and two different reference materials, UCC (data from Wedepohl, 1995) and peat (average ratio of peats from pre-anthropogenic times, from 4000 to 5000 years before present), at sites Borralleiras de Cal Grande (BLL), Penido Vello (PVO) and Pena da Cadela (PDC) in Spain.

487

M.E. Kylander et al.

488

cores (PbEFP, calculated from samples between 4000 and 5000 cal yr BP). Given is the maximum lead enrichment from industrial times as well as Roman and recent times. The PbEFUCC are similar to the PbEFP only when Ga and Y are used as reference elements. The use of the other elements resulted in higher enrichments: PbEFUCC are mostly between two and ﬁve times greater than PbEFP, indicating that the values of the UCC are not representative in a general way. If we now consider the PbEFP, a number of facts become evident: (1) Ga always estimated the lowest enrichment; (2) the sequence of conservative elements is almost the same for the three bogs, with Zr and Ti giving the highest PbEFP; and (3) the variability of the estimates is high for the maximum PbEFP of industrial times and recent PbEFP with a coefﬁcient of variation (CV) of 35–64, but smaller for the Roman maximum with a coefﬁcient of variation of 14–30, and Penido Vello shows signiﬁcantly lower PbEFp for the industrial maximum and present enrichment. These results suggest that the mineralogy of the source area, the number of mineral phases hosting the elements (Rahn, 1976), the amount of deposited dust and processes occurring during dust transport (physical-chemical fractionation linked to climate changes) and in the peat after deposition (post-depositional redistribution and peat mass transformations) may play important roles on the concentration of lithogenic elements in the peat. The lower the number of host mineral phases of an element and their abundance in the dust source areas the more variable its record can be as its ﬂux to the bog is likely to be undetectable at some time, as with Zr. On the other hand an element that is hosted by many abundant mineral phases, such as Ga, which substitutes for Al in aluminosilicates, is less dependent on the source area and the physical properties of the minerals, so its ﬂuxes will not tend to suffer such dramatic changes. Particle density and size are also suggested as affecting physical fractionation during transport with denser, larger particles not being transported as long distances as smaller, lighter particles (Schu¨tz and Rahn, 1982; Schu¨tz, 1989; Martı´ nez Cortizas et al., 2002a; Shotyk et al., 2002b).

The lead story told in peatlands It was the pioneering work of Lee and Tallis (1973) that introduced the use of bogs as archives for reconstructing lead depositional history. Since then a large body of work has accumulated in this ﬁeld and the peat cores analyzed, although mostly conﬁned to Europe, vary widely in terms of temporal coverage. For purposes of discussion we divide the history of lead into ﬁve main periods: pre-anthropogenic, ancient, preindustrial, industrial, and recent. Varying amounts of paleoenvironmental, written historical, geochemical and archeological evidence, represent each period. Geochemistry is the tool most relied upon by environmental scientists but the strength of an interpretation is contingent on agreement between all information sources. The pre-anthropogenic period (o3000

14

C yr BP)

The establishment of pre-anthropogenic aerosol lead concentrations and isotope ratios is important in terms of quantifying the magnitude of present day heavy metal

Archiving natural and anthropogenic lead deposition in peatlands

489

pollution (as in EF calculations) and setting realistic environmental policies. To date however, there is a paucity of information available from pre-anthropogenic times in peat cores. An exception to this is the landmark study by Shotyk et al. (1998), which analysed an ombrotrophic core from the Jura Mountains of Switzerland. It was the ﬁrst long-term core, dating to 12,370 14C yr BP, that comprehensively showed the relationship between natural and anthropogenic Pb contributions, Pb isotopes and changes in dust ﬂuxes as indicated by Sc as the lithogenic element (Fig. 21.4). In this record pre-anthropogenic times are taken to be those pre-dating 3000 14C yr BP. The mid-Holocene pre-anthropogenic background dusts were geochemically deﬁned by a period with constant and low Pb and Sc ﬂuxes, even Pb/Sc ratios that agree with UCC values, as well as stable isotope-ratio signals. Speciﬁcally, dusts from this period, derived from the natural weathering of terrestrial surfaces, were characterized by an average 206Pb/207Pb ratio of ca. 1.20, a Pb ﬂux of 0.010 mg m2 yr1, and a PbEFP,Sc (using background Sc concentrations from the same core) below 2. It is speculated that during this time the dusts deposited came from the Scandinavian Shield, and later, the Sahara. Closer examination of the pre-anthropogenic period in the Swiss core shows two major climatic events. Large increases in dust deposition at 8230 and 10,590 14C yr BP and a smaller increase in dust deposition at about 5320 14C yr BP are demonstrated using lithogenic lead ﬂuxes ratios. The oldest event at 10,590 14C yr BP,

Figure 21.4. Shown are changes in PbEFSc normalized using the peat background value (a), and 206 Pb/207Pb lead isotope changes (b) that have occurred in lead deposited at a bog in western Switzerland over the last 12,370 14C yr BP. The major periods of lead history discussed are identiﬁed (after Shotyk et al., 1998). (BP years are 14C yr BP).

M.E. Kylander et al.

490

during which ﬂuxes exceeded background levels by 35 times, corresponds to the Younger Dryas climate event that saw the expansion of dry-dusty areas and reduced vegetation cover. Al/Sc, La/Sc, Zr/Sc, Hf/Sc, and REE/Sc (REE ¼ Rare Earth Elements) peaks suggest increased dust deposition caused by the drastic climate and a change in source, and therefore mineralogy, of the dusts. The elevated dust ﬂux at about 8230 14C yr BP exceeds background ﬂuxes by 3.5 times and shows analogous changes in dust source and/or transport to those seen in the Younger Dryas. Evidence from other archives suggests that this event could have been a cold event similar to the Little Ice Age that occurred ca. 1300–1800 AD. The shift that occurred at 5320 14C yr BP, although in part resulting from the increase in sedentary lifestyles and thus erosion from agriculture, is also an established period of Saharan drying (Shotyk et al., 1998, 2001). There were thus several signiﬁcant periods of natural change well before the on-set of anthropogenic inﬂuence.

The ancient period (3000– 1600

14

C yr BP)

The process of cupellation wherein precious metals were reﬁned from Ag–Pb ores was discovered in 3500 BC; in this process lead was often simply a by-product of silver mining (Nriagu, 1998). Isotope ratio determinations by Klaminder et al. (2003) shows that at their study site in Sweden two pollution events at 3400–3000 14C yr BC and 2100–1800 14C yr BC required the introduction of a lead source with a 206 Pb/207Pb as low as 1.16, since the natural background value derived from local soils is 41.3. Given the nearly identical isotope ratios at several sites, long distance transport of deposited lead is evoked. As identiﬁed natural sources all had 206 Pb/207Pb 41.17, the authors suggest that these events are evidence of pollution from early metal-using civilizations. Ores from ancient mining areas around the Mediterranean do indeed have a 206Pb/207Pb o1.17, but this work has yet to be conﬁrmed by further geochemical analysis. Why then were these pollution events not recorded in the Swiss core whose earliest pollution event occurred at 1221–1257 14C yr BC? A possible explanation is that the natural Swiss background 206Pb/207Pb ratio of ca. 1.20 is closer to the signature of the European ores being mined, and is therefore less sensitive to such a change. Much of the prosperity of Phoenician, Greek, and Roman civilizations relied on their ability to extract mineral wealth, especially silver. Lead was perhaps most successfully mined by the Romans who exploited every major Ag–Pb deposit in the Mediterranean and Western Europe and had an annual production of ca. 50,000 tons at the peak of the empire (Nriagu, 1998). Peat cores in Switzerland (Shotyk et al., 1998, 2001, 2002a), Spain (Martı´ nez Cortizas et al., 2002a), and Sweden (Klaminder et al., 2003) have recorded this activity through changes in PbEF and lead isotope ratios. In the Swiss core, lead isotope ratios for the period between the Roman invasion of Spain in 2nd century BC, to the height of mining production during 1st and 2nd centuries AD, start to decrease from the pre-anthropogenic value and stabilizes at a 206Pb/207Pb ratio of ca. 1.18. This reduction is a consequence of the addition of lead from Iberian Peninsula ores, which produced 40% of the worldwide Pb production during these times and have 206Pb/207Pb values ranging between

Archiving natural and anthropogenic lead deposition in peatlands

491

ca. 1.16 and 1.17 (Gro¨gler et al., 1966; Pomie´s et al., 1998). Here, calculated PbEFP,Sc increases to over two for the ﬁrst time and remain so until present day (Shotyk et al., 1998). A similar trend was seen in the 206Pb/207Pb ratios from the Spanish core where the samples selected for isotope analysis from pre-Roman times had a higher value of 41.2 compared to a value of 1.182 in the sample representing Roman times (Martı´ nez Cortizas et al., 2002a). The same is not true, however, for an English core spanning 2000 BC to 1800 AD Considering Pb ﬂuxes alone the increased deposition from 200 BC to 200 AD would easily be attributable to Roman mining. The lead isotopes, on the other hand tell a different story; three-isotope plots of 208Pb/206Pb vs. 206Pb/207Pb show that lead deposited at this time has isotopic values close to English Pb ores rather than Spanish ores. This suggests that England received little Pb pollution from Spanish mines and that there was an established mining industry two-and-a-half centuries prior to Roman arrival (Le Roux et al., 2004).

The pre-industrial period (1600– 200

14

C yr BP)

The transition to the pre-industrial period is made at the collapse of the Roman Empire and the exhaustion of the mines by 4th century AD. The decrease in Pb production during the Dark Ages in Europe sees a slight return to background PbEF and increases in Pb isotope ratios towards pre-anthropogenic values in all four cores discussed thus far (Swiss, Swedish, English, Spanish). An exception would be that found by Kempter et al. (1997) where exceptionally high PbEFCrustal,Ti were found in the Harz Mountains of Germany compared to seven sites around Central and Northern Europe. It is at this same location that a large portion of medieval mining later occurs. Beginning in the 8th century AD this is reﬂected in the Swiss core by again a decrease in 206Pb/207Pb ratios and an increase in PbEFP,Sc to ca. 15 (Shotyk et al., 1998). In the English core the high and stable Pb ﬂuxes were maintained until 1500 AD, with the exception of a small decrease at 1200 AD linked to the Black Death and England’s Hundred Years War with France. English ores remained the primary Pb source to this particular bog (Le Roux et al., 2004). Medieval mining is demonstrated in the Spanish core by increased PbEFUCC,Ti over this time period (Martı´ nez Cortizas et al., 2002a). The estimated global Pb production between 1000 and 1500 AD is between 4000–7000 tons per year (Nriagu, 1998). When comparing PbEF for the peat cores discussed it is apparent that in terms of Pb release, the Roman times were far more striking than medieval times. During the Metallurgical Revolution (1450–1750 AD) the discovery of a new method for recovering silver from copper ores severed the historical link between Ag and Pb (Nriagu, 1998).

The industrial period (ca. 1800– 1970 AD) With the Industrial Revolution mining operations were no longer powered by wood burning, but used fossil fuels (especially coal) and tall chimneystacks at industrial sites increased the dispersal of pollutants. The import of more economically produced lead from the USA, Canada, and Australia, also began. From the Industrial

492

M.E. Kylander et al.

Revolution to present there is an enormous increase in PbEF in all studied cores. In the Swiss core PbEFP,Sc reach over 20. The isotope ratios fall sharply in a steep twostep decrease in 206Pb/207Pb starting 1843 AD. The ﬁrst-step decline in isotope ratios is extreme reaching a 206Pb/207Pb of ca. 1.16. This less radiogenic value comes from the import of Australian material, the signature of which is exempliﬁed by ore from Broken Hill, which has a 206Pb/207Pb o1.04, well below the Swiss natural background. The second drive towards unradiogenic values comes from the introduction of leaded gasoline in 1936. The engineer Thomas Midgley Jr. stumbled on tetraethyl lead as a gasoline additive in 1921 and, despite its well documented toxicity, it was introduced in the early 1920s by General Motors and used until the late 1970s and early 1980s (Kitman, 2000). The use of tetraethyl lead in gasoline has contributed more lead to the environment than any other activity. This is mirrored in the Swiss core by a PbEFP,Sc of 99.1 and a ﬂux 1570 times the natural background rate at peak tetraethyl lead use in 1979. The 206Pb/207Pb ratio falls to its lowest value of ca. 1.12 at this point. This isotope ratio ﬁts well with the blend of lead ores used in gasoline used in Switzerland in the 1970s (Shotyk et al., 1998, 2001). From the pre- and post-industrial threshold to present there are many more peat cores analysed than from earlier times and the isotope-ratio data are more variable among these cores. This is because where mining was previously the main lead source, a variety of modern sources exists including: high-temperature industrial processes (steel and non-ferrous metal production), fossil fuel combustion (leaded gasoline, oil, and coal), and waste incineration. Due to increased import and export activity in modern times it is difﬁcult to constrain contributions from particular sources. Such is the case with the work of Weiss et al., (1999) discussed previously. All peat samples and identiﬁed Pb sources, on the three-isotope plots of Figures 21.3a,b, fall between the two extreme signatures of the natural end-member (as represented by the pre-anthropogenic aerosol/UCC) and the very unradiogenic Broken Hill Mine ore end-member. Based on these plots alone it is impossible to clearly deﬁne which source dominated when. What is certain is that simple two-component mixing is unlikely. Using historical data and three-isotope plots it appears that coal dominated until the late 19th and early 20th century, followed by ﬂy-ash and then leaded gasoline in the 1970s. Notably, isotope ratio variations were found among sites. For example, in contrast to sites in western and central Switzerland, isotope ratios in a core from southern Switzerland indicate that industrial lead sources are more important than automotive sources (Weiss et al., 1999). Nova´k et al. (2003) found a comparable trend in eight cores taken around the Czech Republic where coal, leaded gasoline and ore smelting all had variable site-to-site contributions. At many locations in Europe the decline in 206Pb/207Pb ratios, as seen by Shotyk et al. (1998), in the industrial period is observed. This pattern occurs again in several bogs from Switzerland (Fig. 21.5; Weiss et al., 1999) and in some variation in Scotland (Farmer et al., 1997; Weiss et al., 2002a), Sweden (Bra¨nnvall et al., 1997), and the Czech Republic (Nova´k et al., 2003). This trend was not observable in one of the few non-European cores analysed: a core from Canada (Weiss et al., 2002a). In this core no clear isotopic trend could be established for post-industrial times. In comparison to a Scottish core of a similar 200-year time span it showed much lower isotopic variation (1.4% vs. 5.2%) as a consequence of its less radiogenic and more

Archiving natural and anthropogenic lead deposition in peatlands

493

Figure 21.5. 206Pb/207Pb isotope ratios plotted versus age determined by 210Pb in the Swiss peat cores GdL, EGr, PRd, TGe and SwM (see Table 21.2 for abbreviations). The two-step decline that marks industrialisation (slope mA, calculated using the latter four cores) and the introduction of leaded gasoline (slope mB, calculated using the same cores) are apparent. Note as well the return towards preanthropogenic values in recent times (from Weiss et al., 1999; used with the permission of the American Chemical Society, Environ. Sci. Technology).

variable pre-industrial 206Pb/207Pb ratios. The pre-industrial 206Pb/207Pb signature of ca. 1.16 is most likely derived from the weathering of the Canadian Shield and is much closer to the signature for leaded gasoline in Europe. Additionally, this core saw a much later onset of industrial pollution in Canada (as seen in lead concentration and Pb/Ti ratios) than in Scotland. Both cores however, showed the impact of industry (coal burning, mining) during the ﬁrst half of the 20th century followed by leaded gasoline in the second half of the 20th century (Weiss et al., 2002a).

Recent times (1920 AD to present) A subsurface PbEF maximum and 206Pb/207Pb ratio minimum during the late 1970s and 1980s is seen throughout Europe (Dunlap et al., 1999; Weiss et al., 1999, 2002b; Shotyk et al., 1998; 2001; Nova´k et al., 2003). From this time to present there is a decrease in PbEF, Pb ﬂuxes and an increase in 206Pb/207Pb ratios back towards more radiogenic values. For example, in the Swiss core the PbEFP,Sc are below 50 and a 206 Pb/207Pb ratio of ca. 1.13 is found in the topmost sample dated 1991, consistent

494

M.E. Kylander et al.

with the phasing out of leaded gasoline through European Union legislation (Shotyk et al., 1998). Modern lead concentrations and Pb isotopes in aerosols have revealed that unleaded gasoline has reduced much of the environmental burden, but new sources are emerging that had been formerly overshadowed by leaded gasoline use (Monna et al., 1997). These new sources are yet to be fully assessed in peat archives.

Final thoughts and future directions Except perhaps for local aspects that may provide clues into middle and small-scale pollution events in connection with socio-economical development (Monna et al., 2004; Martı´ nez Cortizas et al., 2005; Mighall et al., 2006 – this book, Ch. 17) the chronology of Pb pollution is now fairly well established. Applications of Pb isotopes in bogs not only conﬁrmed the global impact of human activities shown in ice core records but also identiﬁed the major sources. This led to as a major environmental policy achievement: the phasing out of leaded gasoline. It is a prime example of where scientiﬁc observation and public awareness provided the necessary catalyst for governmental regulations. There are, however, many areas that require development. (1) Comparison between different areas using a homogenous treatment is a priority objective, since it will help to understand human impacts on the cycles of many elements. (2) We need to develop a consistent method for Pb atmospheric pollution estimation in peat records, so differences in the values truly reﬂect human impacts and are not artifacts introduced by the calculation methods or uncertainties arising from the parameters used. (3) The general validity of EF to evaluate temporal changes in elemental ﬂuxes for a given location has been questioned. We suggest shifting from ﬁnding a single, proper reference to a process-oriented multivariate approach (Simeonov et al., 1999, 2000). Interpretation of records has to concentrate on the identiﬁcation and modelling of the processes responsible for pre-pollution background variations of the elements of interest. Processes are general but may be expressed as a different set of variables in each record, depending on the connection between pre- and post-depositional conditions (local lithology, soil weathering, erosion patterns, physical fractionation during transport, wind intensity, climate changes). (4) There is a need to identify new Pb sources as Pb concentrations remain signiﬁcantly above natural background values. (5) There is a need to expand the geographic coverage of peat cores, as most work thus far has been concentrated in Europe. Recent advances in modelling conﬁrm inter-continental transport and the need for its assessment (Holloway et al., 2003). Establishing a network that reconstructs past and present Pb ﬂuxes around the globe will help to assess pollutant ﬂuxes on a hemispheric scale. (6) Tetraethyl lead is still used in gasoline in the developing world and puts the lives of 1.7 billion people at risk of lead poisoning. This problem is only likely to escalate given that the car population is increasing rapidly.

Archiving natural and anthropogenic lead deposition in peatlands

495

References Albare´de, F., 1995. Introduction to Geochemical Modelling. Cambridge University Press, Cambridge. Bra¨nnvall, M.-L., Bindler, R., Emteryd, O., et al., 1997. Stable isotope and concentration records of atmospheric lead pollution in peat and lake sediments in Sweden. Water Air Soil Pollut. 100, 243–252. Carignan, J. and Garie´py, C., 1995. Isotopic composition of epiphytic lichens as a tracer of the sources of atmospheric lead emissions in southern Que´bec, Canada. Geochim. Cosmochim. Acta 59, 4427–4433. Chow, T.J., Snyder, C.B., and Earl, J.L., 1975. Isotope ratios of lead as pollutant source indicators. UN, FAO and IAEA Symposium, IAEA-SM 191/4, 95–105. Dickin, A.P., 1995. Radiogenic Isotope Geology. Cambridge University Press, Cambridge. Doe, R.B., 1970. Lead Isotopes. Springer, New York. Duce, R.A., Hoffman, G.L., and Zoller, W.H., 1975. Atmospheric trace metals at remote northern and southern hemisphere sites: pollution or natural?. Science 187, 59–61. Dunlap, C.E., Steinnes, E., and Flegal, A.R., 1999. A synthesis of lead isotopes in two millennia of European air. Earth Planet. Sci. Lett. 167, 81–88. Farmer, J.G., MacKenzie, A.B., Sugden, C.L., et al., 1997. A comparison of historical lead pollution records in peat and freshwater lake sediments from central Scotland. Water Air Soil Pollut. 100, 253–270. Gro¨gler, N., Geiss, J., Gru¨nenfelder, M., and Hontermans, F.G., 1966. Isotopenuntersuchungen zur bestimmung der herkunft ro¨mischer bleirohre und bleibarren. Zeitschr. Naturforsch. 212 (7), 1167–1172. Gulson, B., Mizon, K., Law, A., et al., 1994. Sources and pathways of lead in humans from the Broken Hill mining community-an alternative use of exploration methods. Econ. Geol. 89, 889–908. Holloway, T., Fiore, A., and Hastings, M.G., 2003. Intercontinental transport of air pollution: will emerging science lead to a new hemispheric treaty? Environ. Sci. Technol. 37, 4535–4542. Kempter, H., Go¨rres, M., and Freznel, B., 1997. Ti and Pb concentrations in rainwater-fed bogs in Europe as indicators of past anthropogenic activities. Water Air Soil Pollut. 100, 367–377. Kitman, J.L., 2000. The secret history of lead. The Nation 270 (11), 11–43. Klaminder, J., Renberg, I., and Bindler, R., 2003. Isotopic trends and background ﬂuxes of atmospheric lead in northern Europe: analysis of three ombrotrophic bogs from Sweden. Glob. Biogeochem. Cycles 17, 1019–1028. Krauskopf, K.B. and Bird, J.D., 1995. Introduction to Geochemistry. McGraw-Hill, London. Landmeyer, J.E., Bradley, P.M., and Bullen, T.D., 2003. Stable lead isotopes reveal a natural source of high lead concentrations to gasoline-contaminated groundwater. Environ. Geol. 45, 12–22. Lee, J. and Tallis, J., 1973. Regional and historical aspects of lead pollution in Britain. Nature 245, 216–220. Le Roux, G., Weiss, D., Grattan, J., et al., 2004. Identifying the sources and timing of ancient and medieval pollution in England using a peat proﬁle from Lindow Bog, Manchester. J. Environ. Monitor. 6, 502–510. Martı´ nez Cortizas, A., Garcı´ a-Rodeja, E., Pontevedra-Pombal, X., et al., 2002a. Atmospheric Pb deposition in Spain during the last 4600 years recorded by two ombrotrophic peat bogs and implications for the use of peat as archive. Sci. Tot. Environ. 292, 33–44. Martı´ nez Cortizas, A., Garcı´ a-Rodeja Gayoso, E., and Weiss, D., 2002b. Peat bog archives of atmospheric metal deposition. Sci. Total Environ. 292, 1–5. Martı´ nez Cortizas, A., Mighall, T., Pontevedra-Pombal, X., et al., 2005. Linking changes in atmospheric dust deposition, vegetation evolution and human activities in NW Spain during the last 5,300 years. Holocene 15, 1–9. Monna, F., Galop, D., Carozza, L., Tual, M., et al., 2004. Environmental impact of early Basque mining and smelting recorded in a high ash minerogenic peat deposit. Sci. Total Environ. 327, 197–214. Monna, F., Lancelot, J., Croudace, I.W., et al., 1997. Pb isotopic composition of airborne particulate material from France and the southern United Kingdom; implications for Pb pollution sources in urban areas. Environ. Sci. Technol. 31, 2277–2286. Nova´k, M., Emmanuel, S., Vile, M.A., et al., 2003. Origin of lead in eight central European peat bogs determined from isotope ratios, strengths, and operation times of regional pollution sources. Environ. Sci. Technol. 37, 437–445.

496

M.E. Kylander et al.

Nriagu, J.O., 1998. Enhanced: tales told in lead. Science 281, 1622–1623. Pomie´s, C., Cocheriea, A., Guerrota, C., et al., 1998. Assessment of the precision and accuracy of leadisotope ratios measured by TIMS for geochemical applications: example of massive sulphide deposits (Rio Tinto, Spain). Chem. Geol. 144, 137–149. Rahn, K.A., 1971. Sources of trace elements in aerosols – an approach to clean air. PhD Thesis, University of Michigan, Ann Arbour, Xerox University Microﬁlms (Order No. 72-4956), 325pp. Rahn, K.A., 1976. The chemical composition of the atmospheric aerosol. Graduate School of Oceanography, University of Rhode Island, Technical Report, 265pp. Reimann, C. and De Caritat, P., 2000. Intrinsic ﬂaws of element enrichment factors (EFs) in environmental geochemistry. Environ. Sci. Technol. 34, 5084–5091. Reuer, M.K. and Weiss, D.J., 2002. Anthropogenic lead dynamics in the terrestrial and marine environment. Philos. Trans. R. Soc. Lond. A 360, 2889–2904. Rollinson, H.R., 1993. Using Geochemical Data. Longman Scientiﬁc, Harlow. Schu¨tz, L., 1989. Atmospheric mineral dust: properties and source markers. In: Leinen, M. and Sartheim, M. (Eds), Paleoclimatology and Paleometeorology: Modern and Past Patterns of Global Atmospheric Transport. NATO Series C: Mathematical and Physical Sciences. Kluwer Academic Publishers, London, Vol. 282, pp. 359–384. Schu¨tz, L. and Rahn, K.A., 1982. Trace-element concentrations in erodible soils. Atmos. Environ. 16, 171–176. Shotyk, W., 1996a. Peat bog archives of atmospheric metal deposition: geochemical evaluation of peat proﬁles, natural variations in metal concentrations, and metal enrichment factors. Environ. Rev. 4, 149–183. Shotyk, W., 1996b. Natural and anthropogenic enrichments of As, Cu, Pb, Sb, and Zn in ombrotrophic versus minerotrophic peat bog proﬁles, Jura Mountains, Switzerland. Water Air Soil Pollut. 90, 375–405. Shotyk, W., 2002. The chronology of anthropogenic, atmospheric Pb deposition recorded by peat cores in three minerogenic peat deposits from Switzerland. Sci. Total Environ. 292, 19–31. Shotyk, W., Blaser, P., Gru¨nig, A., and Cheburkin, A.K., 2000. A new approach for quantifying cumulative, anthropogenic, atmospheric lead deposition using peat cores from bogs: Pb in eight Swiss peat bog proﬁles. Sci. Total Environ. 249, 257–280. Shotyk, W., Cheburkin, A.K., Appleby, P.G., et al., 1996. Two thousand years of atmospheric arsenic, antimony, and lead deposition recorded in an ombrotrophic peat bog proﬁle, Jura Mountains, Switzerland. Earth Planet. Sci. Lett. 145, E1–E7. Shotyk, W., Cheburkin, A.K., Appleby, P.G., et al., 1997. Lead in three peat bog proﬁles, Jura Mountains, Switzerland: enrichment factors, isotopic composition, and chronology of atmospheric deposition. Water Air Soil Pollut. 100, 297–310. Shotyk, W., Krachler, M., Martı´ nez Cortizas, A., et al., 2002a. A peat bog record of natural, preanthropogenic enrichments of trace elements in atmospheric aerosols since 12,370 14C yr BP, and their variation with Holocene climate change. Earth Planet. Sci. Lett. 199, 21–37. Shotyk, W., Weiss, D., Appleby, P.G., et al., 1998. History of atmospheric lead deposition since 12,370 14C yr BP recorded in a peat bog proﬁle, Jura Mountains, Switzerland. Science 281, 1635–1640. Shotyk, W., Weiss, D., Heisterkamp, M., et al., 2002b. New peat bog record of atmospheric lead pollution in Switzerland: Pb concentrations, enrichment factors, isotopic composition and organolead species. Environ. Sci. Technol. 36, 3893–3900. Shotyk, W., Weiss, D., Kramers, J.D., et al., 2001. Geochemistry of the peat bog at Etang de la Gruere, Jura Mountains, Switzerland, and its record of atmospheric Pb and lithogenic trace metals (Sc, Ti, Y, Zr, and REE) since 12,370 14C yr BP. Geochim. Cosmochim. Acta 65, 2337–2360. Simeonov, V., Massart, D.L., Andreev, G., and Tsakovski, S., 2000. Assessment of metal pollution based on multivariate statistical modelling of ‘hot spot’ sediments from the Black Sea. Chemosphere 41, 1411–1417. Simeonov, V., Tsakovski, S., and Massart, D.L., 1999. Multivariate statistical modelling of coastal sediments data. Toxicological Environ. Chem. 72, 81–92. Spiro, B., Weiss, D.J., and Purvis O.W., etal., 2004. Pb isotopes in lichen transplants – transient records of diverse sources around the Karabash smelter, Urals, Russia. Environ. Sci. Technol. 38, 6522–6528.

Archiving natural and anthropogenic lead deposition in peatlands

497

Stumm, W. and Morgan, J.J., 1996. Aquatic Chemistry. Wiley, New York. Wedepohl, K.H., 1995. The composition of the continental crust. Geochim. Cosmochim. Acta 59, 1217–1232. Weiss, D., Shotyk, W., Appleby, P., et al., 1999. Atmospheric Pb deposition since the Industrial Revolution recorded by ﬁve Swiss peat proﬁles: enrichment factors, ﬂuxes, isotopic composition, and sources. Environ. Sci. Technol. 33, 1340–1352. Weiss, D., Shotyk, W., Boyle, E.A., et al., 2002a. Comparative study of the temporal evolution of atmospheric lead deposition in Scotland and eastern Canada using blanket peat bogs. Sci. Total Environ. 292, 7–18. Weiss, D., Shotyk, W., Cheburkin, A.K., et al., 1997. Atmospheric lead deposition from 12,400 to ca. 2,000 yr BP in a peat bog proﬁle, Jura Mountains, Switzerland. Water Air Soil Pollut. 100, 311–324. Weiss, D., Shotyk, W., Page, S.E., et al., 2002b. The geochemistry of major and selected trace elements in a forested peat bog, Kalimantan, SE-Asia, and its implications on past atmospheric dust deposition. Geochim. Cosmochim. Acta 66, 2307–2323.

This page intentionally left blank

499

D. Direct human impact on peatlands Section D deals with problems associated with the drainage and utilization of peat by human beings. Chapter 22 is specifically concerned with structural and biological changes, whereas Chapter 23 considers the response of drained peatlands to their use for agricultural purposes.

This page intentionally left blank

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

501

Chapter 22

Impacts of artiﬁcial drainage of peatlands on runoff production and water quality J. Holden, P.J. Chapman, S.N. Lane and C. Brookes

Introduction Excess moisture is critical to the maintenance of peatlands, which are therefore very sensitive to changes in supply of water (Heathwaite, 1995). Changes that occur at the local scale can, in accumulation, have a global impact. Drainage of peat for example, can alter carbon–sink–source relationships that can increase atmospheric CO2 (Roulet, 1990; McNeil and Waddington, 2003). The relative position of the water table within the peat ultimately controls the balance between accumulation and decomposition and therefore peat stability. Peat is very sensitive to changes in hydrology that may be brought about by climate or land management change. Greater aeration above the water table increases decomposition in unsaturated conditions relative to saturated conditions below, so having fundamental implications for properties and attributes above and below the water table. Artiﬁcial drainage of peatlands has been practiced for centuries in the interests of agricultural demand, forestry, horticultural and energy properties of peat and alleviation of ﬂood risk. Pumping water from fens, and subsequent afforestation can also contribute to lowering of peat water tables. However, several problems have been associated with these drainage activities. Peatland drainage is often hypothesized as increasing ﬂooding (Robinson, 1986; Guertin et al., 1987; Lane, 2001). There are also problems related to water quality, erosion, loss of archeological preservation within peat and ecosystem destruction. This chapter attempts to shed light on the nature and extent of these problems and review the progress made in understanding hydrological and hydrochemical processes associated with drainage of peats. The chapter overviews traditional peat drainage practice and associated literature before highlighting more recent processbased research that illustrates that artiﬁcial drainage of peatlands is unsustainable. It will become clear that there are a range of feedback mechanisms that occur following drainage, some of which are non-reversible, and result in major challenges for wetland restoration. The paper then discusses the future needs for wetland research and peatland restoration. A case study using high-resolution topographic data and hydrological modeling tools to aid peatland drainage restoration strategies is presented. ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09022-5

502

J. Holden et al.

However, we are still a long way from being able to couple the modeling with processes incorporating irreversible physicochemical change to peat.

History and extent of drainage Many countries have witnessed vast amounts of artiﬁcial peatland drainage including The Netherlands, Finland, Russia, Ireland, The Philippines, Indonesia, New Zealand, Canada and the UK. In Ireland drainage of peats and gleys has been reported since 1809 (Common, 1970; Wilcock, 1979). Most of the Irish peat drainage was associated with the aim of reducing ﬂooding and for peat extraction purposes, but drainage schemes altered and accelerated after the Second World War due to the need to increase livestock production on upland farms (Stephens and Symons, 1956; Common, 1970). In Northern Ireland there are only 169 km2 of intact peat left compared with 1190 km2 of total peatland (Cooper et al., 1991). In New Zealand, where peat soils cover more than 180,000 ha, peatlands were extensively drained for farmland in the 1970s with little regard to their ecological or environmental value (Bowler, 1980). In Britain, land drainage commenced before Roman times and there are records of it in Domesday (Darby, 1956). Britain is one of the most extensively drained lands in Europe (Baldock, 1984) and drainage expanded during the 17th century accompanying land tenure, enclosure and reclamation of the Anglian Fens. In the following hundred years, peat shrinkage and subsidence associated with the pumped removal of water from the fens meant that more and more water had to be removed to render the drainage works useful (Cole, 1976). Until the 20th century most drainage activity had focused on improving fens for agriculture by lowering the water table. After 1900 drainage was also directed toward ﬂood alleviation. The feed-Britain era after the Second World War saw government grants for expansion in drainage works paid at 70%, particularly in agriculturally marginal upland peatlands. In the 1960s and 1970s, 100,000 ha of blanket peat per year were drained in Britain (Robinson and Armstrong, 1988). Surface drains in raised or blanket peat are often contoured, or constructed in a herringbone shape with short lateral feeder ditches collecting into a central ditch (Fig. 22.1). Single isolated ditches are sometimes used for tapping springs or other natural seepages (Stewart and Lance, 1983). Peatland drainage in the uplands is often carried out with the purpose of lowering the water table to improve the vegetation and remove the surface water hazard for grazing and game birds. However, there is little evidence that increase in game populations following drainage and peatlands cannot sustain large increases in stocking density. Thus, Newson (1992) suggested that upland drainage was backed by very limited rationale with low economic beneﬁts and potentially major environmental impacts. In addition to agricultural drainage, about 15 million hectares of northern peatlands have been drained for forestry, mainly in Northern and Eastern Europe (Paavilainen and Paivanen, 1995). In Britain, about 200,000 ha of deep peatland and 350,000 ha of shallow peats have been afforested with coniferous plantations since 1945. To ensure successful establishment of trees on peatlands, the water table must ﬁrst be lowered. In Scandinavia, Finland, Russia, Canada, Ireland and Britain, drainage by closely spaced plough furrows has taken place (Fig. 22.2). The result is

Impacts of artificial drainage of peatlands

503

Figure 22.1. An aerial photograph showing the nature of peatland drainage patterns in northern England for an area of 4 km2. The photograph was taken by the UK Natural Environment Research Council (NERC) during a ﬂight on August 6th, 1995 from NERC site 94.9, ﬂight run 6, and is NERC photograph number 8886.

short-term change in runoff production while the drains are active (David and Ledger, 1988; Prevost et al., 1999; Anderson et al., 2000) and in the long-term when the forest establishes. Increased interception results in greater evaporation on trees and enhanced evapotranspiration encourages drying of the peat and the development of shrinkage cracks (Holden et al., 2004). In Finland, 5.7 million hectares of peatlands have been drained, so that now one-quarter of the country’s forested land consists of drained peatland (Laiho et al., 1998). In Scotland, 25% of Caithness and Sutherland peatlands have been affected by differing intensities of drainage associated with afforestation (Ratcliffe and Oswald, 1988). This area recently became the focus of major conservation protest and international condemnation (Charman, 2002).

Impact of peat drainage on catchment hydrology The impact of artiﬁcial drainage on the hydrological response of peatland catchments was ﬁrst fully investigated experimentally by Conway and Millar (1960). They

504

J. Holden et al.

Figure 22.2. The close spacing of peatland forest drains between planted trees that are now maturing.

worked on four small (2 ha) peatland catchments at Moor House in the English north Pennines; two had natural drainage channels and two had artiﬁcial networks of surface drains. They concluded that runoff production in blanket peat was more rapid where artiﬁcial drainage had taken place. There was an increased sensitivity of runoff response to storm rainfall with peak ﬂows both higher and earlier. In contrast, intact basins exhibited a smoother storm hydrograph with greater lag times. Their water balance calculations have been interpreted by some to suggest that intact basins could retain signiﬁcantly more water than drained basins. However, a number of other small investigations followed which showed both conﬂicting and corroborating results. Burke (1967), for example, investigated water balances in a drained peatland at Glenamoy, Ireland. In contrast to results from Moor House, runoff tended to be quicker from the undrained part of the bog (where the water table was very close to the surface). In the drained bog the water table was often 45–60 cm deep and runoff from the catchment was much slower. The reason given for this was that in the drained catchment most of the runoff ﬂowed as throughﬂow to the drains, whereas in the undrained catchment runoff was generated as saturation-excess overland ﬂow and could be transmitted much quicker from the catchment. Runoff/ rainfall ratios from the undrained Glenamoy catchment were only 23.4% compared to 79.2% from the drained catchment (Burke, 1975a,b). Similar results were reported for German peatlands by Baden and Eggelsmann (1970) and demonstrate the importance of enhanced understanding of the effect of land management practices on the hydrology of peatlands. Several reasons have been cited for the differences between Glenamoy and Moor House including different peat properties (McDonald, 1973) and drain spacing (Robinson, 1980, 1985). The drains at Moor House were 0.5 m deep and 14 m apart compared to Glenamoy where they are about twice as deep and four times closer

Impacts of artificial drainage of peatlands

505

together. Of course, Burke (1967) had already established that drain density was an important factor at Glenamoy showing that water table was only affected within 2 m of the drains. Since the aim of any drainage work was to lower water table, a drain spacing of 4 m was therefore required (and used) at that site. Thus the low hydraulic conductivity of peatlands (Holden, 2006 – this book, Ch. 14) frequently renders drainage operations unsuccessful or uneconomic because extremely close ditch spacing is required in order to signiﬁcantly lower the water table through the drawdown process, although this will depend on the properties of the peat (Huikari, 1968; Boelter, 1972; Hudson and Roberts, 1982). Conway and Millar (1960) had never established whether their drains signiﬁcantly affected water table and thus a 14 m spacing was established without soil properties being considered. Stewart and Lance (1991) later showed that water table was only affected within 0.5 m of the Moor House ditches. Ahti (1980) found that ﬂood peaks increased as ditch spacing decreased, presumably related to the increased channel density. For Burke (1967, 1975a,b), however, closer ditch spacing resulted in a greater effect on water table, increased temporary storage and a subdued runoff response to rainfall with lower ﬂood peaks. Clearly, the effects are more complex depending on local site conditions and probably relate to the site-speciﬁc interactions of precipitation, peat type, drain spacing and topographic ﬂow routing. The latter is often neglected but is probably the most important factor and is discussed in the case study later in this chapter. Table 22.1 provides a list of papers that have examined hydrological response to artiﬁcial drainage in peatlands and shows the range of conﬂicting and corroborating reports on peatland response to drainage. Moklyak et al. (1975) presented quantitative evidence from a peatland area in the Ukraine showing that drainage can both reduce and increase total runoff from peatlands within the same area. Out of the ﬁve catchments investigated, three had reduced annual runoff and ﬂood peaks following drainage, one had an increase in annual runoff and ﬂood peak and one catchment had no signiﬁcant change in ﬂow regime. Several hypotheses for each type of change were discussed and these are listed in Table 22.2 along with hypotheses presented by Robinson (1980) for the response of the Coalburn catchment in northern England. However, in neither case were measurements of processes other than streamﬂow performed. Several studies have been based on examination of river ﬂow and water balance at the large catchment-scale rather than at the hillslope or plot scale. These have tended to be ‘before and after’ studies, some of which were qualitative (Lewis, 1957; Oliver, 1958) rather than quantitative. Detailed discussion of the papers listed in Table 22.1 is provided in Holden et al. (2004) and so they are not discussed in depth here. The catchment-scale studies suffer from poor data availability, and conclusions tend to be rather piecemeal or anecdotal. Often river ﬂow records were not available for periods before or during drainage operations and fail to cope with high-ﬂow measurement. In addition, there have been no reported nested studies where hillslopes and catchments have been monitored in tandem to investigate response to artiﬁcial drainage. No hillslope studies have involved ‘before and after’ monitoring of response to peat drainage. The papers listed in Table 22.1 all provide water balance approaches and they either simply present the results with limited explanation, or provide some explanations but have no corroborating ﬁeld evidence for these explanations. There

J. Holden et al.

506 Table 22.1.

Reported hydrological effects of peatland drainage (after Holden et al., 2004).

Lewis (1957) Oliver (1958) Howe and Rodda (1960) Conway and Millar (1960) Mustonen (1964) Burke (1967) Howe et al. (1967) Baden and Eggelsmann (1970) Institute of Hydrology (1972) Moklyak et al. (1975) Heikurainen (1968) Ahti (1980) Robinson (1980, 1986) Newson and Robinson (1983) Guertin et al. (1987) Gunn and Walker (2000)

Affect on temporary storage

Affect on ﬂood peak

Affect on annual runoff

Quantitative assessmenta

Processes measured (other than stream ﬂow)

Decrease

Increase Increase Increase

Increase Increase

C C X

X X X

Increase

Increase

H

X

Increase Decrease Increase Decrease

Increase Increase

H H C H

X Water table X X

Increase

Increase

C

X

Increase or decrease Increase

Increase or decrease Decrease

Increase or decrease

C

X

H

X

Decrease Decrease

Increase Increase

Increase

H H

X X

Decrease

Increase

C

X

X

X

H

X

Decrease Increase Increase

Increase Decrease

Increase

Increase

a

C ¼ Large catchment data within which some parts of the catchment have been artiﬁcially drained, H ¼ small subcatchment or artiﬁcially drained hillslope monitored.

have been few instances of hydrological process-based measurement within the catchments themselves. The effects of ditching may depend on where in the catchment the disturbance takes place. For example, drainage of part of a catchment may result in delayed runoff from hillslopes where peak ﬂows had previously occurred before the main river channel peak. The result could be that drainage increases the peak discharge in the catchment because the timings of the catchment and drained subcatchment peak ﬂows correspond (Fig. 22.3). Hence, even though drainage may result in a reduction in the ﬂood peak at the hillslope scale, the net result may be an increase at the catchment-scale depending on where in the catchment the drainage operations took place and how that part of the catchment responds (Holden et al., 2004). Little work has been done on this aspect of peatland hydrology and clearly a catchment modeling approach is required (Lane et al., 2004). It does suggest, however, that the same land management practice will have a different impact on river ﬂow depending on its topographic and spatial context and the nature of slope-stream coupling at that site (Lane et al., 2003). Additionally there is a scaling issue. As the distance from the

Impacts of artificial drainage of peatlands

507

Table 22.2. Reasons for increased or decreased ﬂood peak and storage following drainage as hypothesized (but not tested) by Moklyak et al. (1975) and Robinson (1980). Change

Explanations

Decreased ﬂood peak and annual water storage

Increased ﬂood peak and annual storage

Direct precipitation in channels Straightening, deepening and vegetation clearance of ditches and

Decreased hydraulic conductivity Loss of overland ﬂow Increased storage in upper peat layers Increased storage on subsidence depressions Increased evapotranspiration related to vegetation change

channels

Decreased evapotranspiration on drained, uncultivated land, with

more bare peat Increased surface and groundwater slope Greater exposure of previously conﬁned aquifers Increase in drainage density Decrease in peat water retention ability due to hydrophobic reaction to drying

drains to the study point increases, so the possibility of attenuation effects will increase, and there is a greater chance that the signature will be masked. Thus as the spatial scale of approach increases, it becomes more difﬁcult to identify changes in river ﬂow records resulting from drainage on headwater hillslopes. For catchments where drainage of peat decreases the ﬂood response from disturbed hillslopes, this is because the soil, catchment and ditch characteristics have enabled water tables to fall and thus the desired response of the slope to drainage is achieved. However, a fall in water table is often accompanied by increased peat decomposition at the surface and in subsidence of the peat mass (see below). Thus the drainage operation becomes unsustainable. In other areas where drainage seems to increase ﬂood response from a catchment, this tends to be where ditches have a very limited effect on water table. Thus the ditches simply act to increase the speed at which surface storm water can escape from the catchment as storage properties are not signiﬁcantly altered. In these cases the drainage activity has not succeeded in achieving its underlying objectives, even in the short term, and may cause problems downstream. Impact of peat drainage on soil properties Hydrological implications Many drained peatland catchments exhibit increases in low ﬂows (Baden and Eggelsmann, 1970; Mustomen and Seuna, 1971; Heikurainen et al., 1978; Ahti, 1980; Robinson, 1980, 1985). This has sometimes been attributed to catchment dewatering. The drained Glenamoy catchment was estimated to lose 1000 mm of water per year

508

J. Holden et al.

Figure 22.3. One of the possible effects of hillslope drainage on the river ﬂood wave. The graphs are theoretical hydrographs (time on x-axis and relative discharge on y-axis). Despite the drainage activity causing the small tributary catchment to have a lower ﬂood peak, this has still resulted in a higher overall ﬂood peak in the main channel due to ﬂood wave synchronicity.

(Burke, 1975a) through slow drainage of the peat. Whereas lowering of the water table increases short-term (storm-event) water storage and makes the runoff response to rainfall less sensitive, in the medium-term water is lost from the catchment. This, of course, was partially the intention, but in peatlands this has often been found to be unsustainable because of associated feedback mechanisms. In the long term as peatlands dewater they are also liable to subside (Anderson et al., 1995) so that, in fact, the temporary increase in water storage capacity may be lost. The catchment may then start to behave in a more ﬂashy way with concomitant increases in ﬂood risk (Fig. 22.4). Burke’s study was not maintained over a sufﬁcient length of time to establish whether these effects occurred at Glenamoy, but certainly relaxation times are important elements that have been ignored in most peat drainage studies. Drainage of fens has been associated with severe shrinkage and decomposition of the peat. Large pumping operations in southeast England have had to be implemented to keep pace with the subsidence of the soil surface. The shrinkage occurs because as the water table is lowered, the upper peat collapses causing bulk density to

Impacts of artificial drainage of peatlands

509

Figure 22.4. A theoretical ﬁgure showing potential change in water storage and ﬂood peak (y-axis) over time following peatland drainage (x-axis) for a case where dewatering was followed by subsidence. No values are shown on the axes because the relationships are theoretical. The immediate impact of drainage is to increase water storage through lowering of the water table. However, over longer timescales as the peat subsides the water storage capacity is reduced.

increase by up to 63% in the upper 40 cm within a few years of drainage (Silins and Rothwell, 1998). The subsidence is associated with physical breakdown and consolidation of dry peat in surface layers and accelerated mineralization of organic matter (Prus-Chacinski, 1962; Egglesmann, 1972; Ivanov, 1981). The dry surface increases capillary action resulting in more water being removed from the subsurface layers. Hence the whole peat mass dries and shrinks (undisturbed peat tends to be 90% water by mass and 300% by volume; Hobbs, 1986). With shrinkage and consolidation, drain life is severely reduced (Prus-Chacinski, 1962; Bowler, 1980) and many peatlands change topographical shape around drains. Macropores are ordinarily important pathways for runoff generation in peat (Baird, 1997; Holden et al., 2001); Silins and Rothwell (1998) suggested that peatland subsidence is associated with the collapse of readily drainable macropores. However, Holden and Burt (2002b) found permanent structural changes to blanket peats in the English north Pennines subject to drought simulation in the laboratory. This leads to changes in the hydrological routing of water through the peat tested with increased macropore ﬂow and ﬂow through deeper peat, with consequential reductions in overland ﬂow. Turbulent ﬂow in macropores may result in macropore enlargement. Larger macropores are known as soil pipes (Holden, 2006 – this book, Ch. 14) and are common features of peatlands (Glaser and Janssens, 1986; Holden et al., 2002; Holden and Burt, 2002a, 2003a, c; Jones, 2004). They transport large quantities of water, solute and sediment, and may be important in peatland carbon cycling. Holden (2005) surveyed 160 blanket peat catchments for soil pipes using groundpenetrating radar (GPR). Soil pipes were found in all catchments. GPR does not provide information on pipe size but it was found that the mean frequency of piping was 69.2 km1 of surveyed transect (standard error ¼ 2.1) and a maximum of

510

J. Holden et al.

466.7 km1. Land drainage exerted the most important control on pipe frequency. Out of the 960 plots investigated, 171 had surface drains. Pipe frequency was signiﬁcantly greater in the drained plots compared to intact plots at po0.001, with a mean of 127.4 pipes km1 (standard error ¼ 6.2) for drained sites and 56.6 pipes km1 (standard error ¼ 2.0) for non-drained sites. Holden (2005) tested whether this result was simply a function of drainage taking place where piping was already very intense and found that this was unlikely to be the case. It was shown that peatland drainage therefore induces soil piping. Surface drains expose a bare peat surface to summer sunshine, resulting in increased desiccation cracking (Fig. 22.5; Holden et al., 2004), and to enhanced winter needle ice formation, which also disturbs the peat structure. In addition, the drains reduce the downslope saturation of the hillslope (by redirecting ﬂow along the contour to the

Figure 22.5. Desiccation cracking on the side of a peatland drain. The peat is a blanket peat about 1 m deep but it is dry and well oxidized on the drain edges.

Impacts of artificial drainage of peatlands

511

stream and preventing water from following its natural course downslope), thereby inducing cracking further away from the ditch walls and ﬂoors. Once cracks open up and a peat surface dries, it often fails to achieve full restoration (it can become partly hydrophobic). In other words, the peat can no longer hold as much water as its structure is permanently altered (Hobbs, 1986; Egglesmann et al., 1993). This promotes conditions conducive to pipe network development across drained hillslopes. Thus enhanced subsurface ﬂow may take place in an environment that was previously dominated by surface ﬂow. This may have implications for water quality, carbon loss, sediment production and landform development (linked to pipe-induced gully development).

Chemical implications The lowering of the water table following drainage leads to a number of processes taking place within the peat that affects both its physical and chemical properties. An increase in air-ﬁlled porosity affects microbial processes and thus decomposition rates. The oxygen promotes aerobic decomposition, which occurs over 50 times faster than anaerobic decomposition (Clymo, 1983). The oxygen also enhances the mineralization of nutrients, particularly the carbon-bound nitrogen and sulfur and the organically bound phosphorus. The top cubic meter of deep organic soils can contain as much as 20,000 kg nitrogen (N), 10,000 kg sulfur (S), 500 kg phosphorus (P) and 500,000 kg of carbon (C) (Miller et al., 1996), so even an increase in mineralization of just 1% per year has the potential to generate large losses of these elements. The loss of nutrients may in turn affect the fertility of peat. For example, De Mars et al. (1996) found that drainage of a Polish fen resulted in P and potassium (K) limitation as a result of aeration of topsoil, accelerated decomposition and increased nutrient release. Peatland drainage has been hypothesized to change peatlands from C sinks to C sources to the atmosphere as a result of increased oxidation of organic matter (Laine and Minkkinen, 1996), although studies from forested peatlands suggest that C density and retention in peat may increase after drainage (Domish et al., 1998; Minkkinen and Laine, 1998) due to increased organic C ﬂow from tree stands into soil and consequent retention in the peat. A number of studies have observed that the exchangeable cation content in drained peats is lower than in undisturbed peats and total concentrations of N and P often increase, whereas K always decreases in the topsoil (0–20 cm) of peat after drainage (Laiho et al., 1998; Sundstrom et al., 2000). In Canada, Wells and Williams (1996) investigated the impact of ditch spacing on soil nutrients in both bog and fen peats. They observed that in bog peats bulk density, total N concentrations (mg g1) and total contents (kg ha1) of N, P, K, Ca and iron (Fe) were signiﬁcantly higher in the 3 m ditch spacing compared to the 15 m ditch spacing. They concluded that increases in total nutrient contents in drained bog peats could be attributed mainly to increased bulk density. In contrast, they observed that bulk density and most nutrient contents of fen peats were not signiﬁcantly affected by drainage. The increase in total N concentrations (mg g1) observed in the topsoil of peat after drainage is due to an increase in the retention of N by microbial immobilization as the plant residues in the peat decompose and total N is increased per unit volume

512

J. Holden et al.

of peat (Wells and Williams, 1996), which also results in a lowering of the C:N ratio. However, many studies have also observed that drainage and lowering of the water table results in an increase in N mineralization (Williams, 1974; Williams and Wheatly, 1988), in response to an increase in oxygen and the number of ammonifying and nitrifying bacteria. The response of N mineralization to water table lowering is not always predictable. For example, Williams (1974) observed that lowering the water table to 18 cm signiﬁcantly decreased the amount of N mineralized in the top 10 cm of peat but that further lowering of the water table to 34 cm increased mineralization in the top 10 cm. Mineralization–immobilization responses of soil N to peatland drainage depend largely on the change in peat decomposition rate, which is regulated by environmental and substrate factors. Environmental factors include temperature, redox potential and pH. Substrate factors include stage of decomposition, organic matter quality, nutrient content, chemistry of the soil solution and the presence of chemical and biological inhibitors to microbial activity. Although lowering the water table should eliminate poor aeration as the foremost limitation to mineralization, the improved aeration may have little impact on mineralization rates if temperature, pH or nutritional constraints still inhibit microbial activity. For example, Humphrey and Pluth (1996) observed that N mineralization rates did not respond to drainage in peat at pH 4.0 but were signiﬁcantly stimulated in peat at pH 7.2. Updegraff et al. (1995) observed that aerobic N mineralization was at least twice as high as anaerobic mineralization in bog peats but not in sedge soils, and thus suggested that the sensitivity of N mineralization to aeration status depended on substrate characteristics related to the quality and quantity of organic matter. These studies therefore suggest large heterogeneity of N dynamics to drainage across the landscape depending on the interacting inﬂuence of environmental and substrate factors.

Impact of peat drainage on water chemistry As well as changes in runoff generation and soil properties, installation of drainage ditches has an impact on water chemistry. Sometimes where peatland drainage appears to have little effect on the hydrological regime of the catchment, it can have signiﬁcant effects on soil and drainage water quality (MAFF, 1980). Many studies have observed that installation of drainage ditches usually increases the leaching of nutrients. For example, in blanket peat, large increases in ammonium (NH4) concentrations have been observed following drainage (Lundin, 1991; Sallantaus, 1995; Miller et al., 1996) and water table lowering (Adamson et al., 2000), but only small changes in nitrate (NO3) concentrations. This suggests that whereas the organisms responsible for ammoniﬁcation beneﬁted from drainage, those responsible for nitriﬁcation did not do so to the same extent. However, increased NO3 and base cation losses have been reported from less acidic peats (Burt et al., 1990; Lundin, 1991; Freeman et al., 1993). Sallantaus (1995) observed a net loss of Ca, Mg and K from drained catchments compared to undrained catchments, where inputs and outputs of these nutrients were more or less balanced. Astrom et al. (2001) observed that forest ditching resulted in an increase in concentrations of suspended sediment, Ca, Mg, manganese (Mn) and

Impacts of artificial drainage of peatlands

513

aluminum (Al), a decrease in total organic C (TOC) and an increase in pH from 4.4 to 5.4 in stream water. In Scotland, Miller et al. (1996) observed initial increases in NH4–N and silicon (Si) due to losses from the exposed peat in the drains. Studies that have investigated the impact of drainage on dissolved organic C (DOC) concentration (which is also associated with discolored water, a major problem for the potable water supply industry) have observed contradictory results. Clausen (1980), Edwards et al. (1987) and Mitchell and McDonald (1992) found that drained catchments produced much more discolored (DOC rich) water than undrained catchments. In contrast, Moore (1987) observed only minor changes in stream DOC concentrations in drained and harvested bogs, compared to undisturbed peatlands in southern Quebec. Results from studies investigating the role of lower water tables on DOC production and export are also contradictory, with some studies observing an increase (Tipping et al., 1999), whereas others had a decrease in DOC (Freeman et al., 2004), or no signiﬁcant changes (Blodau et al., 2004). Adamson et al. (2000) investigated the impact of water table drawdown in blanket peat on soil solution composition during a drought period at a catchment dominated by a blanket peat in northern Britain. They observed a decline in DOC and dissolved organic N (DON) as well as a large increase in sulfate (SO4)2 and H+ ion concentrations at 10 cm depth when the water table dropped to 40 cm below the surface + of the peat. Many other studies have also reported an increase in SO2 ions 4 and H in soil (Devito and Hill, 1999) and stream waters draining peatlands following periods of drought (Bayley et al., 1986; Bottrell et al., 2004). These observations have been attributed to the oxidation of reduced sulfur stored in the peat as well as mineralization of organic sulfur to dissociated sulfuric acid. It is likely that some of the H+ ions released replace base cations held on exchange sites resulting in the marked increase in Na, Mg and Ca concentrations and ionic strength observed by Adamson et al. (2000). Freeman et al. (1993) manipulated water tables on laboratory peat columns collected from a valley bottom wetland in mid-Wales and also observed + a large increase in concentrations of SO2 ions and a decline in DOC. 4 and H In Canada, Prevost et al. (1999) investigated the impact of drainage on soil solutions collected from 20 and 40 cm depth and at 1.5, 5 and 15 m from the center of each ditch. They observed that solute content was enhanced by drainage, the effect being generally proportional to ditch closeness for S and Mg, whereas increases in N, Na, K and Ca were mainly observed within 5 m of the ditch and at 20 cm depth. This increase in solutes was associated with slight decreases in pH, and coincided with an increase in soil temperature, a decrease in moisture content and accelerated decomposition rates observed within the top 30 cm and close to the ditches where water table drawdown was greatest (Prevost et al., 1999). In fen peats, water is often pumped from the land, which results in the rapid lowering of the water table and transfer of solutes from peat to ditch. In Somerset, UK, Heathwaite (1987) observed that SO4 concentrations were at least three times higher in pumped-drained ditches compared to watercourses and that Ca and Mg concentrations were at least twice as high in pumped ditches. Green (1974) noted that decreases in downstream water quality following drainage installation could commonly not be associated directly with ditching, but rather with the activities surrounding it such as increased use of fertilizers to assist plantation

514

J. Holden et al.

establishment (Williams and Wheatly, 1988). Many studies, especially in Scandinavia and Finland, have investigated losses of P and K from forest fertilization (Karsisto, 1970; Kaunisto and Mailanen, 1992). In Scotland, Miller et al. (1996) reported losses of 1–2 kg P ha1 (of the 58 kg P ha1 applied) and 25–35 kg K ha1 (of the 108 kg K ha1 applied) in drainage water in the year after application and noted the growth of moss and algae in the main drainage channels. Additionally, where drainage ditches penetrate the mineral soil beneath the peat there may be other effects such as Al and Mn release (Reynolds and Hughes, 1989; Astrom et al., 2001) or an increase in pH of drainage waters. Most studies that have investigated the impact of artiﬁcial ditching on water chemistry have observed changes in concentrations and ﬂuxes of solutes in the short term. However, the duration of the drainage effects on water chemistry is not known as few studies have continued monitoring for more than ﬁve years. In addition, most studies have monitored the chemistry of drainage water rather than the soil solution, and few studies have linked these measurements to soil processes. Therefore it is not known in detail to what extent and by which mechanisms various solutes are released and leached in artiﬁcially drained catchments. Compared to forested peatlands, there is little information on the impacts of drainage on water chemistry in unforested blanket and shield peatlands.

Impacts of peat drainage on erosion In addition to the effects of increased peak ﬂows resulting in downstream changes in sediment ﬂux and bank erosion, ditching can lead to severe degradation of peatland soils themselves in some areas (Mayﬁeld and Pearson, 1972). Ditches can quickly become deep wide channels and supply large amounts of peat material to the river network. Drains cut to 50 cm depth may erode to several meters (Fig. 22.6). Studies of the geomorphologic impacts of peat drainage are rare. Figure 22.7 presents some data from northern England, which shows the importance of slope in controlling drain erosion or inﬁlling. The data plotted are for values of enlargement of the crosssectional area of the drain with respect to the size that the drain was cut. Therefore negative values are where natural drain inﬁlling has occurred. Natural inﬁlling of drains can often occur on gentle slopes under 41. The ﬁgure suggests that drains on slopes below 21 rarely erode, whereas drains on slopes over 41 rarely inﬁll. Erosion tends to become more severe as slope increases (po0.001). Usually vertical incision is associated with the undercutting of ditch sides and block failure. The rate of erosion usually slows once the underlying substrate is reached but many drains with a large catchment area feeding into them can continue to erode into the substrate. Nickpoint retreat at the conﬂuence of drains is often observed. Enhanced piping in drained peatlands (see above) may also exacerbate sediment losses. Such erosion not only causes on-site degradation and carbon loss, but can cause downstream problems including reservoir inﬁlling and smothering of gravel-bed spawning grounds. In the Ribble and Hadden catchments, northern England, the salmon catches fell during the eight years following drainage from 1400 yr1 to 380 yr1, whereas in the nearby Lune, where there had been no drainage, catches

Impacts of artificial drainage of peatlands

515

Figure 22.6. An eroded peatland drain that has incised into the underlying substrate by forming a large gully approximately 2.5 m deep. Note the rucksack on the left that provides scale.

remained stable (Stewart, 1963). In the River Nuorittajoki in northern Finland, Laine (2001) observed that the recapture rates of stocked yearling salmon were lower in rifﬂes receiving high inputs of particulate matter from drained peatlands than in rifﬂes receiving a considerably smaller loading. In addition, the size of the salmon was inversely related to the estimated load of particulate matter to the rifﬂe. However, little is known about the full impact of peatland drainage on sediment movement or ecology in upland areas and more work needs to be done in this area. Peatland drainage has also been linked to slope instability. Mass movements of peat, usually reported as bog bursts or peat slides have been well documented over the last 150 years (Warburton et al., 2004; Dykes and Kirk, 2006 – this book, Ch. 16). These mass movements transport vast quantities of material from slopes and some peat slides have been known to be larger than 1 km2. Many peat mass movements in both the UK and

516

J. Holden et al.

Figure 22.7. The role of slope in peatland drain erosion. Data are taken from three sites in northern England from March 2004 and are based on cross-sectional areas of drains relative to the original size of the drains when they were ﬁrst cut in 1976. Negative values of cross-sectional area of erosion indicate that the drain has partially inﬁlled.

Ireland have occurred in conjunction with artiﬁcial drainage where failure occurs along the artiﬁcial drainage line. Ditches are often found at the margin of failure scars and have been cited as possible contributors to failure and subsequent mass movement (Tomlinson, 1981; Wilson and Hegarty, 1993; Dykes and Kirk, 2001; Warburton et al., 2004). Peatland restoration and the utility of ﬁeld and modeling approaches Despite continued peatland drainage for afforestation, extraction and agriculture, there is now a realization among the public and policy-makers that degradation of an important terrestrial store and associated ecosystem destruction are not desirable. There is therefore a drive to protect undisturbed sites from disturbance and restore damaged sites. However, full reversal to a previous pristine state is rarely possible and so the goals for peat restoration schemes should be viewed in light of this. Peatlands are complex systems where multiple processes operate in combination. A signiﬁcant amount of work toward ecological restoration has taken place in peatland areas but a great deal of this work has been carried out on a pragmatic or even an ad hoc basis. This reﬂects both the urgency of the requirement to protect important sites, and the frequent shortfalls in available funding. Whereas there is a body of knowledge relating to the hydrological processes of peatlands, too often managers, through time and resource constraints, have been required to act with only a limited understanding of the functioning of their particular site. Commonly, when ecological restoration is attempted, several interventions are employed at the same time. Restoration work has often been completed with limited prior monitoring, and it has therefore been difﬁcult to sustain scientiﬁc assessments for a sufﬁcient time period in order to evaluate success (Carpenter and Lathrop, 1999) or to disentangle the precise effects of particular interventions. Commonly, wetland landscapes have such

Impacts of artificial drainage of peatlands

517

disparate relaxation times that process responses are difﬁcult to identify. Many laudable results have been achieved by the hard work and detailed on-the-ground knowledge of managers but there remain many sites where restoration has been a hitand-miss affair, where time and money have been wasted because the hydrological functioning of the system has been poorly understood. Peatland restoration (or rehabilitation, as full reversal is not possible) often involves two steps: ﬁrst, the re-establishment of high water tables and, second, the recolonization of important peat forming species such as Sphagnum. The primary aim of the hydrological management of damaged and fragmentary peats is normally to minimize water loss through a strategy of ditch blockage or through some attempt at sealing the boundary of the mire to prevent the loss of water. Techniques have been applied at a wide variety of scales and costs, often without detailed monitoring to assess the effectiveness of the works. In areas where surface drains have been cut, many organizations are seeking resources to block them. However, this is an expensive strategy and there is a range of unresolved issues associated with drain blockage. The main issues are (1) the very high cost of ditch blocking; (2) determining the most effective methods of blockage; (3) the uncertain impacts of blockage on river ﬂow and water quality; and (4) the uncertain response of the peat and vegetation in the context of permanent structural and chemical changes that may have taken place following water table lowering. There are therefore a series of research requirements. These range from practical experiments on blockage design and conditions conducive to optimum vegetation recovery to the development of tools for helping practitioners determine which drains are more important to block so that resources could be efﬁciently targeted and drain blocking prioritized. Additionally, a modeling approach that would assist in examining the impacts of drain blocking on stream ﬂow and water quality is also required. Examples of recent research in these areas are given below.

Water table and vegetation recolonization Water table recovery in peatlands following ditch blocking can be relatively rapid (Mawby, 1995; Price et al., 2003). However, that is not to say that vegetation or hydrochemical recovery will follow. Maltby (1997) and Bragg and Tallis (2001) emphasized that the peatland biodiversity assemblage is highly vulnerable to perturbation. Changes to peat pH and nutrient status as a result of drainage can also make ecological restoration difﬁcult. Price (1997) tested a range of water management approaches that attempted to ameliorate conditions limiting Sphagnum regeneration in North America. Water table depth was found not to be a good indicator of water availability at the peat surface due to decomposition of the surface layers. Simply blocking ditches caused good water table recovery during the wet spring period, but the water table recession was much faster and greater than in an undisturbed area. Price (1997) suggested that in addition to blocking ditches to recreate a water table regime comparable to that in a natural area, more aggressive management techniques such as creating open reservoirs and using straw mulch (which increased soil moisture by 10–15%) may be required. It may often be necessary to seed vegetation on the surface of a damaged bog in addition to hydrological restoration and protection of existing vegetation. Sphagnum diaspores, for

518

J. Holden et al.

example, can be spread across the surface of the bog. These may need additional protection by mulching to enable establishment (Price et al., 1998; Rochefort et al., 2003).

Modeling approaches Modeling results from the Upper Wharfedale case study catchment in the UK have been encouraging. High-resolution topographic data were collected using Light Detection and Ranging (LiDAR) to create a digital elevation model that had a precision of 12 cm in the elevations (Lane et al., 2004). It is important to use such high-resolution data in peatland environments because very small differences in topography can be important for ﬂow routing, saturation and ecology. From these data the topographic index of 2 m 2 m grid cells was calculated. The topographic index ln(A/tan b) is a measure of upslope area (A) draining to a given point per unit contour length divided by the slope angle (tan b). The drains were added to the digital elevation model using GIS. The index was then recalculated after the drains were added into the topography. Figure 22.8 provides a map, for a small proportion of the catchment, of the change in topographic index induced by the presence of the drainage channels. Of course, an important effect of the drains on the peatland is to reduce the topographic index downslope. This is primarily why peatland drains on hillslopes reduce saturation. They prevent water that would have otherwise followed the topographic drainage route from continuing downslope and instead intercept ﬂow and transport it directly to the stream. However, what is more important about this map is that it allows us to determine which drains have the biggest effect on the topographic index. It can be seen, for example, that the dense ditch network labeled A is not as important as some of the ditches on the steeper northerly slopes of the catchment labeled B. This is an illustration of how the effect of a particular drain on peatland saturation will be dependent on the topographic context of that drain. A practitioner can examine the type of map shown in Figure 22.8 and make decisions about resource allocation on a drain-by-drain or hillslope-by-hillslope basis. This is important, given that there are over 200 km of drains in the 13 km2 of upper Wharfedale alone. These maps can also be tied into ecological patterns and models of DOC production related to water table drawdown. However, this work has not yet been done. By utilizing a hydrological model such as TOPMODEL (Beven and Kirkby, 1979) it is also possible to investigate the impact of blocking drains on river ﬂow and ﬂood wave synchronization. In the 13 km2 Upper Wharfedale catchment the drains appear to reduce water travel times by 20 min on average (Lane et al., 2003). At very high topographic resolution there are a number of problems that have to be overcome when using a topographically driven ﬂow model and these were dealt with by Lane et al. (2004). However, by investigating some of these problems it was also possible to illustrate another research requirement. The model differentiates between overland ﬂow and subsurface ﬂow. Figure 22.9 presents the predicted streamﬂow hydrograph for the catchment when different sized grid cells are used for the investigation. There is actually very little difference in predicted output between scales. A good prediction of total stream discharge from the peatland might be easy to achieve using either coarse or ﬁne-scale topographic data. However, Figure 22.9 illustrates that

Impacts of artificial drainage of peatlands

519

Figure 22.8. A map of the change in topographic index for a small part of the Upper Wharfedale study catchment at 541 130 5100 N latitude, 21 130 3800 W longitude (after Lane et al., 2003). The main river channel in the center of the ﬁgure ﬂows from west to east. A location with gray shading means that it had its topographic index reduced by the presence of a peatland drain. The darker the shading, the greater the reduction in the topographic index. Thus darker shading means that the propensity for resaturation of the hillslope will be greater if the land drain is removed by blocking. The legend scale is measured in standard deviations. Thus if the mean difference is 1 and the standard deviation of the difference is 2, then a difference of 3 is one standard deviation above the mean and would be mapped as 1 (3 minus 1, which equals 2, and which equals 1 standard deviation). The mean and standard deviation are a summary of all pixel values on the difference map. The map thus indicates how unusual the difference in a pixel is compared to all other measured differences. Labels X and Y refer to discussion in the text.

as the scale becomes ﬁner the components of the stream hydrograph made up from an overland ﬂow component and a subsurface ﬂow component alter quite markedly. Therefore the predictions of overland ﬂow and subsurface ﬂow may not be correct and are scale dependent. This scale dependency of ﬂow partitioning is important from a water quality perspective as different solutes and particulates will be transported by surface and subsurface ﬂowpaths. Further work is therefore required to ensure that the processes are correctly represented even though model optimization at the catchment scale might be easy to achieve. Intensive ﬁeld research at the site is also ongoing in order to assist in model calibration and validation.

Ditch blocking techniques Natural revegetation of ditches and disturbed peatlands has been observed. If ditches are not maintained they can ﬁll in with vegetation and sediment, losing their

520

J. Holden et al.

Figure 22.9. Impact of model resolution (2 m, 16 m, 64 m) on hydrograph prediction (total ﬂow, overland ﬂow and subsurface ﬂow) using TOPMODEL in the Wharfedale study catchment.

effectiveness in water removal (Fisher et al., 1996). Indeed, this benign neglect of ditches may be one of the simplest management strategies proposed to return peats toward a favorable condition. Inﬁlling often starts where peat has slumped onto the drain ﬂoor and is colonized by mosses and later by rushes and sedges. If unshaded,

Impacts of artificial drainage of peatlands

521

the ﬂoor could regrow with Sphagnum. The tendency of drains to inﬁll depends on the type of material forming the ﬂoor, the slope angle (Fig. 22.7) and the drainage area feeding the drain. However, in most peatlands artiﬁcial dams are required to block and revegetate ditches (Van Seters and Price, 2001). Often peat and plastic ditch plugs are unsuitable for ditch blocking where slopes are steep. Here ditch water can scour around and under the plugs. On steep slopes ditch blocking can be very demanding on resources and Figure 22.10 shows where wooden stakes and dams have been installed on an Irish peatland with only 3 m spacing. Calluna bails have been used in some upland peats (Geltsdale, Cumbria, UK) where the seed bank and nutrients are local (unlike the use of straw bails which have also been used in some areas). These allow water to ﬂow through the bails, but slow the velocity and allow sediment to slowly accumulate. The aim is to avoid further scour erosion around the ditch plugs and allow the ditch to slowly inﬁll with sediment and vegetation. However, the cheapest and most successful ditch blocking practice in blanket peats, where the ditches have not scoured too deeply, is to scoop out some peat adjacent to the ditch and then very ﬁrmly pack the peat into the ditch. The scour hole thereby created from the removed scoop of peat allows water to leave the dam and rewet the adjacent peat via overland ﬂow (Fig. 22.11). This may only work, however, where a good thick contact can be maintained between the plug of peat and the ditch ﬂoor and walls. Thresholds of recovery and non-reversible trajectories There appear to be boundary conditions beyond which peatlands cannot be restored. A suitable depth of peat left in situ is often required, particularly if that peat is only

Figure 22.10. Intensive ditch blocking on a peatland near Enniskillen, Ireland.

522

J. Holden et al.

Figure 22.11. Ditch blocking to allow water to pond and escape back out onto the peatland surface.

supplied with water and nutrients by rainwater. If the upper, slightly humiﬁed peat layer is still intact then recovery may be more likely. Once the peat starts to regenerate it will eventually become self-sustaining and artiﬁcial water tables will no longer be needed, but a sufﬁcient hydrological integrity of the peatland complex is a necessary prerequisite. Many peatlands which are at their climatic margins will be more sensitive to drainage and less likely to recover. However, exceptions to this include the large peatlands of the Glacial Lake Agassiz region of northwestern Minnesota (Glaser et al., 1981, 1997). Despite the dry climate, peatlands cover 56% of the landscape and include many raised bogs and patterned water tracks (mainly related to groundwater inputs). Early 20th century drainage ditches appear to have only a minor impact on peat hydrology and vegetation patterns and so any peat rehabilitation will be more likely. Most conceptual ideas about peatland restoration are based on the idea of returning a peatland as close to the functioning of an undisturbed site as possible. The relationship between hydrological conditions in an undisturbed peatland and those within an artiﬁcially drained peatland, however, may exhibit signiﬁcant differences. This includes increased matrix hydraulic conductivity due to drying, and preferential ﬂow through desiccation cracks may also result in a far greater overall hydraulic conductivity than would normally be the case in undisturbed mires. The increased heterogeneity in the hydraulic conductivity across the mire is of great signiﬁcance where ﬂow predictions are made, particularly if a distributed model is to be used (Holden and Burt, 2003b). In addition, whereas ditch blocking practices are already underway, further research is required to examine whether these wetland restoration strategies are able to cope with enhanced piping and to ensure that piping is adequately taken into account when developing peat management plans. It is also unknown to what extent the chemical changes to peats and peat pore waters impact

Impacts of artificial drainage of peatlands

523

on vegetation and water quality in the short or long term following peatland rewetting. An important example includes the enhanced release of DOC that may be the result of peatland rewetting. Impacts may be spatially distributed in relation to topographic context. When considering peatland restoration we must ask the question: restoration to what? The climate today is different from that of the early Holocene when many peatlands formed. Therefore a peatland restored in today’s climate may well develop along an entirely different trajectory from peatlands a few thousand years ago. When restoring peatlands do we simply want to maintain current ecological functions (Charman, 2002) or do we want to allow peatlands ecosystems and their hydrochemistries to develop in new directions? The latter may not be avoidable. Judging the success of peatland restoration must then depend on our perception of peatland functions and process understanding.

Conclusions Most studies associated with artiﬁcial drainage of peatlands have been black-box water balance studies with limited measurement of the hydrological processes. These have shown that there can be contradictory hydrological responses of peatland catchments to drainage. However, more recent process-based studies are revealing the mechanisms behind these different results. These mainly relate to changes in ﬂow routing and connectivity between slopes and channels in relation to catchment and topographic context (which can both increase and decrease ﬂood peaks, and lag times), and to the lag times of feedback mechanisms as peatlands respond to drainage. For example, if a drained peatland produces reduced peak ﬂows at ﬁrst because of enhanced storage capability, this may not be maintained in the long term. Dewatered peats tend to subside. As this happens the capacity to store water will be reduced. Simultaneously, physicochemical changes to dried peats reduce the water holding capacity of the pores. Thus peak ﬂows may increase in the long term. Drainage of peats also results in exacerbated erosion, changes to peat water chemistry and runoff quality and release of terrestrial carbon. Subsurface erosion seems to be enhanced in blanket peats following drainage and further work is required to test this for other peatland types. In addition, artiﬁcial drainage rarely occurs in isolation; burning, grazing, afforestation and fertilization can all accompany drainage. Thus the effectiveness of any peatland restoration strategy will depend on how well integrated catchment management strategies are and how well we understand the interacting mechanisms. Much more work is required to examine the hydrological and hydrochemical processes surrounding artiﬁcial drainage and peatland restoration. In particular, restoration strategies will need to account for permanent structural and chemical changes that take place to peats when they are drained. It should not be expected that restoration will necessarily result in a reversal back to a peatland which functions in the way that it did when it was in pristine state. Therefore, whereas it is possible to use high-resolution spatial modeling techniques to guide practitioners and to enable them to spatially distribute management strategies and resources in peatlands, such models have not yet been coupled to models of

524

J. Holden et al.

irreversible (or reversible) physicochemical response. Whereas this chapter has shown how we have made great progress in both areas, we are still a long way from coupling the two components. References Adamson, J.K., Scott, W.A., Rowland, A.P., and Beard, G.R., 2000. Ionic concentrations in a blanket peat bog in northern England and correlations with deposition and climatic variables. Eur. J. Soil Sci. 51, 1–15. Ahti, E., 1980. Ditch spacing experiments in estimating the effects of peatland drainage on summer runoff. Int. Assoc. Hydrol. Sci. Publ. 130, 49–53. Anderson, A.R., Pyatt, D.G., and White, I.M.S., 1995. Impacts of conifer plantations on blanket bogs and prospects of restoration. In: Wheeler, B.D., Shaw, S.C., Fojt, W., and Robertson, R.A. (Eds), The Restoration of Temperate Wetlands. Wiley, Chichester, pp. 533–548. Anderson, A.R., Ray, D., and Pyatt, D.G., 2000. Physical and hydrological impacts of blanket peat afforestation at Bad a Cheo, Caithness: the ﬁrst 5 years. Forestry 73, 467–478. Astrom, M., Aaltonen, E.K., and Koivusaari, J., 2001. Impact of ditching in a small forested catchment on concentrations of suspended material, organic carbon, hydrogen ions and metals in stream water. Aquat. Geochem. 7, 57–73. Baden, W. and Eggelsmann, R., 1970. Hydrological budget of high bogs in the Atlantic region, Proceedings of the Third International Peat Congress 1968, Quebec, Department of Energy, Mines and Resources, Ottawa, pp. 260–311. Baird, A.J., 1997. Field estimation of macropore functioning and surface hydraulic conductivity in a fen peat. Hydrol. Process. 11, 287–295. Baldock, D., 1984. Wetland Drainage in Europe. International Institute for Environment and Development, Nottingham. Bayley, S.E., Behr, R.S., and Kelly, C.A., 1986. Retention and release of sulphur from a freshwater wetland. Water Air Soil Pollut. 31, 101–114. Beven, K.J. and Kirkby, M.J., 1979. A physically-based, variable contributing area model of basin hydrology. Hydrol. Sci. Bull. 24, 43–69. Blodau, C., Basiliko, N., and Moore, T.R., 2004. Carbon turnover in peatland mesocosms exposed to different water table levels. Biogeochemistry 67, 331–351. Boelter, D.H., 1972. Water table drawdown around an open ditch in organic soils. J. Hydrol. 15, 329–340. Bottrell, S., Coulson, J., Spence, M., et al., 2004. Impacts of pollutant loading, climate variability and site management on the surface water quality of a lowland raised bog, Thorne Moors, E. England, UK. Appl. Geochem. 19, 413–422. Bowler, D.G., 1980. The Drainage of Wet Soils. Hodder and Stoughton, London. Bragg, O.M. and Tallis, J.H., 2001. The sensitivity of peat-covered upland landscapes. Catena 42, 345–360. Burke, W., 1967. Principles of drainage with special reference to peat. Irish Forestry 24, 1–7. Burke, W., 1975a. Aspects of the hydrology of blanket peat in Ireland. Hydrology of marsh-ridden areas, Minsk, USSR, 1972. IAHS Studies and Reports in Hydrology, 19, Unesco Press, pp. 171–182. Burke, W., 1975b. Effect of drainage on the hydrology of blanket-bog. Irish J. Agric. Res. 14, 145–162. Burt, T.P., Heathwaite, A.L., and Labadz, J.C., 1990. Runoff production in peat covered-catchments. In: Anderson, M.G. and Burt, T.P. (Eds), Process Studies in Hillslope Hydrology. Wiley, Chichester, pp. 463–500. Carpenter, S.R. and Lathrop, R.C., 1999. Lake restoration: capabilities and needs. Hydrobiologia 395/396, 19–28. Charman, D., 2002. Peatlands and Environmental Change. Wiley, Chichester. Clausen, J.C., 1980. The quality of runoff from natural and disturbed Minnesota peatlands. Proceedings of Sixth International Peat Congress, Duluth, Minnesota, pp. 523–532. Clymo, R.S., 1983. Peat. In: Gore, A.J.P. (Ed.), Mires: Swamps, Bog, Fen and Moor. General Studies. Elsevier, Amsterdam, Vol. A, pp. 159–224.

Impacts of artificial drainage of peatlands

525

Cole, G., 1976. Land drainage in England and Wales. J. Inst. Water Eng. 30, 345–367. Common, R., 1970. Land drainage and water use in Ireland. In: Stephens, N. and Glasscock, R.E. (Eds), Irish Geographical Studies, in Honour of E. Estyn Evans. Queens University Belfast, Belfast, pp. 342–359. Conway, V.M. and Millar, A., 1960. The hydrology of some small peat-covered catchments in the northern Pennines. J. Inst. Water Eng. 14, 415–424. Cooper, A., Murray, R., and McCann, T., 1991. The Environmental Effects of Blanket Peat Exploitation. University of Ulster, Coleraine. Darby, H.C., 1956. The Draining of the Fens. Cambridge University Press, Cambridge. David, J.S. and Ledger, D.C., 1988. Runoff generation in a plough-drained peat bog in southern Scotland. J. Hydrol. 99, 187–199. De Mars, H., Wassen, M.J., and Peeters, W., 1996. The effect of drainage and management on peat chemistry and nutrient deﬁciency in the former Jegrznia-ﬂoodplain (NE-Poland). In: chemical and physical dynamics of fen hydrology. Nederlandse Geograﬁsche Studies 203, 51–68. Devito, K.J. and Hill, A.R., 1999. Sulphate mobilization and pore water chemistry in relation to groundwater hydrology and summer drought in two conifer swamps on the Canadian Shield. Water Air Soil Pollut. 113, 97–114. Domish, T., Finer, L., Karsisto, M., et al., 1998. Relocation of carbon from decaying litter in drained peat soils. Soil Biol. Biochem. 30, 1529–1536. Dykes, A.P. and Kirk, K.J., 2001. Initiation of a multiple peat slide on Cuilcagh Mountain, Northern Ireland. Earth Surf. Proc. Land. 26, 395–408. Dykes, A.P. and Kirk, K.J., 2006. Slope stability and mass movements in peat. In: Martini, I.P., Matı´ nez Cortizas, A., and Chesworth, W. (Eds), Peatlands: Evolution and Records of Environmental and Climatic Changes. Elsevier, Amsterdam. Edwards, A., Martin, D. and Mitchell, G. (Eds), 1987. Colour in upland waters. Proceedings of Yorkshire Water/Water Research Centre Workshop, Leeds, 29 September. Egglesmann, R., 1972. Physical effects of drainage in peat soils of the temperate zone and their forecasting. Int. Assoc. Hydrol. Sci. Publ. 105, 69–77. Egglesmann, R., Heathwaite, A.L., Gross-Braukmann, G., et al., 1993. Physical processes and properties of mires. In: Heathwaite, A.L. and Gottlich, K. (Eds), Mires, Process, Exploration and Conservation. Wiley, Chichester, pp. 171–262. Fisher, A.S., Podniesinski, G.S., and Leopold, D.J., 1996. Effects of drainage ditches on vegetation patterns in abandoned agricultural peatlands in central New York. Wetlands 16, 397–409. Freeman, C., Fenner, N., Ostle, N.J., et al., 2004. Export of dissolved organic carbon from peatlands under elevated carbon dioxide levels. Nature 430, 195–198. Freeman, C., Lock, M.A., and Reynolds, B., 1993. Impacts of climate change on peatland hydrochemistry: a laboratory-based experiment. Chem. Ecol. 8, 49–59. Glaser, P.H. and Janssens, J.A., 1986. Raised bogs in eastern North America: transitions in landforms and gross stratigraphy. Can. J. Bot. 64, 395–415. Glaser, P.H., Siegel, D.I., Romanowicz, E.A., and Shen, Y.P., 1997. Regional linkages between raised bogs and the climate, groundwater, and landscape features of northwestern Minnesota. J. Ecol. 85, 3–16. Glaser, P.H., Wheeler, G.A., Gorham, E., and Wright, H.E. Jr., 1981. The patterned peatlands of the Red Lake peatland, northern Minnesota: vegetation, water chemistry, and landforms. J. Ecol. 69, 575–599. Green, F.H.W., 1974. Changes in artiﬁcial drainage, fertilisers and climate in Scotland. J. Environ. Manage. 2, 107–121. Guertin, D.P., Barten, P.K., and Brooks, K.N., 1987. The peatland hydrologic impact model: development and testing. Nordic Hydrol. 18, 79–100. Gunn, J. and Walker, C., 2000. Peat cutting and bog restoration. In: Gunn, J. (Ed.), The Geomorphology of Culcagh Mountain, Ireland: A Field Guide for the British Geomorphological Research Group Spring Field Meeting, May 2000. Limestone Research Group, University of Huddersﬁeld, Huddersﬁeld, pp. 15–25. Heathwaite, A.L., 1987. Chemical transformations in a drained fen peat. Unpublished Ph.D. Thesis, University of Bristol.

526

J. Holden et al.

Heathwaite, A.L., 1995. Overview of the hydrology of British wetlands. In: Hughes, J.M.R. and Heathwaite, A.L. (Eds), Hydrology and Hydrochemistry of British Wetlands. Wiley, Chichester, pp. 11–20. Heikurainen, L., 1968. Results of draining peatland for forestry in Finland. Transactions of the Second International Peat Congress, Leningrad, USSR, 1963, pp. 773–780. Heikurainen, L., Kenttamies, K., and Laine, J., 1978. The environmental effects of forest drainage. Suo 29, 49–58. Hobbs, N.B., 1986. Mire morphology and the properties and behaviour of some British and foreign peats. Q. J. Eng. Geol. 19, 7–80. Holden, J., 2005. Controls of soil pipe frequency in upland blanket peat. J. Geophys. Res. 110, F010002, doi: 10.1029/2004JF000143. Holden, J., 2006 (this book, Chapter 14). Peatland hydrology. In: Martini, I.P., Matı´ nez Cortizas, A., and Chesworth, W. (Eds), Peatlands: Evolution and Records of Environmental and Climatic Changes. Elsevier, Amsterdam. Holden, J. and Burt, T.P., 2002a. Piping and pipeﬂow in a deep peat catchment. Catena 48, 163–199. Holden, J. and Burt, T.P., 2002b. Laboratory experiments on drought and runoff in blanket peat. Eur. J. Soil Sci. 53, 675–689. Holden, J. and Burt, T.P., 2003a. Hydrological studies on blanket peat: the signiﬁcance of the acrotelm–catotelm model. J. Ecol. 91, 86–102. Holden, J. and Burt, T.P., 2003b. Hydraulic conductivity in upland blanket peat: measurement and variability. Hydrol. Process. 17, 1227–1237. Holden, J. and Burt, T.P., 2003c. Runoff production in blanket peat covered catchments. Water Resour. Res. 39, 1191, doi: 10.1029/2003WR002067. Holden, J., Burt, T.P., and Cox, N.J., 2001. Macroporosity and inﬁltration in blanket peat: the implications of tension disc inﬁltrometer measurements. Hydrol. Process. 15, 289–303. Holden, J., Burt, T.P., and Vilas, M., 2002. Application of ground penetrating radar to the identiﬁcation of subsurface piping in blanket peat. Earth Surf. Proc. Land. 27, 235–249. Holden, J., Chapman, P.J., and Labadz, J.C., 2004. Artiﬁcial drainage of peatlands: hydrological and hydrochemical process and wetlands restoration. Prog. Phys. Geogr. 28, 95–123. Howe, G.M. and Rodda, J.C., 1960. An investigation of the hydrological cycle in the catchment area of the river Ystwyth during 1958. Water Water Eng 767, 10–16. Howe, G.M., Slaymaker, H.O., and Harding, D.M., 1967. Some aspects of the ﬂood hydrology of the upper catchments of the Severn and Wye. Trans. Inst. Brit. Geogr. 41, 33–58. Hudson, J.A. and Roberts, G., 1982. The effect of a tile drain on the soil moisture content of peat. J. Agric. Eng. Res. 27, 495–500. Huikari, O., 1968. Effect of distance between drains on the water economy and surface runoff of Sphagnum bogs. In: Robertson, R.A. (Ed.), Transactions of the Second International Peat Vongress, Leningrad, USSR, 1963. HMSO, London, pp. 739–742. Humphrey, W.D. and Pluth, D.J., 1996. Net nitrogen mineralization in natural and drained fen peatlands in Alberta, Canada. J. Soil Sci. Soc. Am. 60, 932–940. Institute of Hydrology, 1972. Peat hydrology. Institute of Hydrology Report 16, Institute of Hydrology, Wallingford. Ivanov, K.E., 1981. Water Movement in Mirelands. Academic Press, New York. Jones, J.A.A., 2004. Implications of natural soil piping for basin management in upland Britain. Land Degrad. Dev. 15, 325–349. Karsisto, K., 1970. Losses of fertilizers from peaty soils. Suo 21, 60–66. Kaunisto, S. and Mailanen, M., 1992. Potassium fertilization on peatland forests. Proceedings of the Ninth International Peat Congress, Uppsala, Sweden: Int. Peat J. 248, special issue. Laiho, R., Sallantaus, T., and Laine, J., 1998. The effect of forestry drainage on the vertical distribution of major plant nutrients in peat soils. Plant Soil 207, 169–181. Laine, A., 2001. Effects of peatland drainage on the size and diet of yearling salmon in a humic northern river. Archiv Fu¨r Hydrobiologie 151, 83–99. Laine, J. and Minkkinen, K., 1996. Effects of forest drainage on the carbon balance of a mire: a case study. Scandinavian J. Forest Res. 11, 307–312.

Impacts of artificial drainage of peatlands

527

Lane, S.N., 2001. More ﬂoods, less rain: changing hydrology in a Yorkshire context. The Yorkshire Humber Regional Rev. 11, 18–19. Lane, S.N., Brookes, C.J., Hardy, R.J., et al., 2003. Land management, ﬂooding and environmental risk: new approaches to a very old question. Proceedings of Chartered Institution of Water Environmental Management National Conference. The Environment – Visions, Values and Innovation. Harrogate, 10–12 September. Lane, S.N., Brookes, C.J., Kirkby, M.J., and Holden, J., 2004. A network-index-based version of TOPMODEL for use with high-resolution digital topographic data. Hydrol. Process. 18, 191–201. Lewis, W.K., 1957. An investigation or rainfall, runoff and yield in the Alwen and Brenig catchments. Proc. Inst. Civil Eng. 8, 17–52. Lundin, L., 1991. Retention or loss of N in forest wetlands. Vatten 47, 301–304. Maltby, E., 1997. Peatlands: the science case for conservation and sound management. In: Parkyn, L., Stoneman, R.E., and Ingram, H.A.P. (Eds), Conserving Peatlands. CAB International, Wallingford, pp. 121–131. Mawby, F.J., 1995. Effects of damming peat cuttings on Glasson Moss and Wedholme ﬂow, two lowland raised bogs in North-West England. In: Wheeler, B.D., Shaw, S.C., Fojt, W.J., and Robertson, R.A. (Eds), Restoration of Temperate Wetlands. Wiley, Chichester, pp. 349–358. Mayﬁeld, B. and Pearson, M.C., 1972. Human interference with the north Derbyshire blanket peat. East Midland Geogr. 12, 245–251. McDonald, A., 1973. Some views on the effects of peat drainage. Scottish Forestry 27, 315–327. McNeil, P. and Waddington, J.M., 2003. Moisture controls on Sphagnum growth and CO2 exchange on a cutover bog. J. Appl. Ecol. 40, 354–367. Miller, J.D., Anderson, H.A., Ray, D., and Anderson, A.R., 1996. Impact of some initial forestry practices on the drainage waters from blanket peatland. Forestry 69, 193–203. Ministry of Agriculture, Food and Fisheries, 1980. Land drainage service research and development. Field Drainage Experimental Unit Technical Report 3, MAFF, Trumpington. Minkkinen, K. and Laine, J., 1998. Long-term effect of forest drainage on the peat carbon stores of pine mires in Finland. Can. J. Forest Res. 28, 1267–1275. Mitchell, G. and McDonald, A.T., 1992. Discolouration of water by peat following induced drought and rainfall simulation. Water Resour. Res. 26, 321–326. Moklyak, V.I., Kubyshkin, G.P. and Karkutsiev, G.N., 1975. The effect of drainage works on streamﬂow. Hydrology of marsh-ridden areas, Minsk, USSR, 1972. IAHS Studies and Reports in Hydrology, 19, Unesco Press, pp. 439–446. Moore, T.R., 1987. A preliminary study of the effects of drainage and harvesting on water quality in an ombrotrophic bogs near Sept-Iles, Quebec. Water Res. Bull. 23, 785–791. Mustomen, S.E. and Seuna, P., 1971. Inﬂuence of forest drainage on the hydrology of peatlands. Water Research Institute Publication 2, National Board of Waters, Helsinki. (In Finnish with an English summary.) Mustonen, S., 1964. The inﬂuence of drainage on hydrology of peat land. Ratennustekniikka 3, 204–209. Newson, M.D., 1992. Conservation management of peatlands and the drainage threat: hydrology, politics and the ecologist in the UK. In: Bragg, O.M. (Ed.), Peatland Ecosystems and Man: An Impact Assessment, Proceedings. University of Dundee, Dundee, pp. 94–103. Newson, M.D. and Robinson, M., 1983. Effects of agricultural drainage on upland streamﬂow: case studies in mid-Wales. J. Environ. Manage. 17, 333–348. Oliver, J., 1958. The wetness of Wales. Rainfall as a factor in the geography of Wales. Geography 43, 151–163. Paavilainen, E. and Paivanen, J., 1995. Peatland Forestry: Ecology and Principles. Ecological Studies 111. Springer, Heidelberg. Prevost, M., Plamondon, A.P., and Belleau, P., 1999. Effects of drainage of a forested peatland on water quality and quantity. J. Hydrol. 214, 130–143. Price, J.S., 1997. Soil moisture, water tension, and water table relationships in a managed cutover bog. J. Hydrol. 202, 21–32. Price, J.S., Heathwaite, A.L., and Baird, A.J., 2003. Hydrological Processes in an abandoned and restored peatlands. An overview of management approaches. Wetland Ecol. Manage. 11, 65–83.

528

J. Holden et al.

Price, J.S., Rochefort, L., and Quinty, F., 1998. Energy and moisture considerations on cutover peatlands: surface micro-topography, mulch cover and Sphagnum regeneration. Ecol. Eng. 10, 293–312. Prus-Chacinski, T.M., 1962. Shrinkage of peat-lands due to drainage operations. J. Inst. Water Eng. 16, 436–448. Ratcliffe, D.A. and P.H. Oswald (Eds), 1988, The Flow Country: The Peatlands of Caithness and Sutherland. Nature Conservancy Council, Peterborough, . Reynolds, B. and Hughes, S., 1989. An ephemeral forest drainage ditch as a source of aluminium for surface waters. Sci. Total Environ. 80, 185–193. Robinson, M., 1980. The effect of pre-afforestation drainage on the streamﬂow and water quality of a small upland–upland catchment. Institute of Hydrology Report 73, Institute of Hydrology, Wallingford. Robinson, M., 1985. The hydrological effects of moorland gripping: a re-appraisal of the moor house research. J. Environ. Manage. 21, 205–211. Robinson, M., 1986. Changes in catchment runoff following drainage and afforestation. J. Hydrol. 86, 71–84. Robinson, M. and Armstrong, A.C., 1988. The extent of agricultural ﬁeld drainage in England and Wales, 1971–1980. Trans. Inst. Brit. Geogr. 13, 19–28. Rochefort, L., Quinty, F., Campeau, S., et al., 2003. North American approach to the restoration of Sphagnum dominated peatlands. Wetlands Ecol. Managem. 11, 3–20. Roulet, N., 1990. The hydrological role of peat-covered wetlands. Can. Geogr. 34, 82–83. Sallantaus, T., 1995. Leaching in the material balance of peatlands – preliminary results. Suo 43, 253–358. Silins, U. and Rothwell, R.L., 1998. Forest peatland drainage and subsidence affect soil water retention and transport properties in an Alberta Peatland. J. Soil Sci. Soc. Am. 62, 1048–1056. Stephens, N. and Symons, L.J., 1956. The Lough Erne drainage system. In: Suggate, L.S. (Ed.), This Changing World. Geography 41, 123–126. Stewart, A.J.A., 1963. Investigations into Migratory Fish Propagation in the Area of the Lancashire River Board. Lancashire River Board, Barber. Stewart, A.J.A. and Lance, A.N., 1983. Moor-draining: a review of impacts on land use. J. Environ. Manage. 17, 81–99. Stewart, A.J.A. and Lance, A.N., 1991. Effects of moor-draining on the hydrology and vegetation of northern Pennine blanket bog. J. Appl. Ecol. 28, 1105–1117. Sundstrom, E., Magnusson, T., and Hanell, B., 2000. Nutrient concentrations in drained peatlands along a north-south climatic gradient in Sweden. Forest Ecol. Manage. 216, 149–161. Tipping, E., Woof, C., Rigg, E., et al., 1999. Climatic inﬂuences on the leaching of dissolved organic matter from upland UK moorland soils, investigated by a ﬁeld manipulation experiment. Environ. Int. 25, 83–95. Tomlinson, R.W., 1981. The erosion of peat in the uplands of Northern Ireland. Irish Geogr. 14, 51–64. Updegraff, K., Pator, J., Bridgham, S.D., and Johnston, C.A., 1995. Environmental and substrate controls over carbon and nitrogen mineralization in northern wetlands. Ecol. Appl. 51, 151–163. Van Seters, T.E. and Price, J.S., 2001. The impact of peat harvesting and natural regeneration on the water balance of an abandoned, cut-over bog, Quebec. Hydrol. Process. 15, 233–248. Warburton, J., Holden, J., and Mills, A.J., 2004. Hydrological controls of surﬁcial mass movements in peat. Earth-Sci. Rev. 67, 139–156. Wells, E.D. and Williams, B.L., 1996. Effects of drainage, tilling and PK-fertilization on bulk density, total N, P, K, Ca and Fe and net N-mineralization in two peatland forestry sites in Newfoundland, Canada. Forest Ecol. Manag. 84, 97–108. Wilcock, D., 1979. The hydrology of a peatland catchment in Northern Ireland following channel clearance and land drainage. In: Hollis, G.E. (Ed.), Man’s Impact on the Hydrological Cycle in the United Kingdom. GeoAbstracts, Norwich, pp. 93–107. Williams, B.L., 1974. Effect of water table level on nitrogen mineralization in peat. Forestry 47, 195–201. Williams, B.L. and Wheatly, R.E., 1988. Nitrogen mineralization and water table height in oligotrophic deep peat. Biol. Fertil. Soils 6, 141–147. Wilson, P. and Hegarty, C., 1993. Morphology and causes of recent peat slides on Skerry Hill, County Antrim, Northern Ireland. Earth Surf. Proc. Land. 18, 593–601.

Peatlands: Evolution and Records of Environmental and Climate Changes I.P. Martini, A. Martı´ nez Cortizas, W. Chesworth, Editors r 2006 Elsevier B.V. All rights reserved.

529

Chapter 23

Peatland subsidence in the Venice watershed M. Camporese, G. Gambolati, M. Putti and P. Teatini

Introduction The draining of histosols to turn tidal marshland into fertile agricultural areas has been extensively accomplished worldwide. One widespread ecological hazard is the release of large quantities of nitrate as the result of organic matter decomposition in aerobic conditions that can enhance the eutrophication of the water bodies receiving the drainage water (Levin, 1970; Avnimelech, 1971). Another environmental consequence is the emission of carbon dioxide and methane to the atmosphere thus contributing to the accumulation of greenhouse gases (Armentano, 1980; Rojstaczer and Deverel, 1993) in an appreciable measure (Freibauer et al., 2004; Lal, 2004). A third hazard is land subsidence (Stephens et al., 1984) induced by microbial oxidation of peat soils enhanced by farming activity. The processes underlying ground settlement were investigated for the ﬁrst time in the Sacramento–San Joaquin Delta, California (Weir, 1938), a region where the occurrence has been much studied over the last 50 years (Weir, 1950; Procopovitch, 1985; Rojstaczer and Deverel, 1995; Deverel and Rojstaczer, 1996), and is still raising concerns for the stability of the levees and the risk of catastrophic ﬂooding (Mount and Twiss, 2005). Countless areas throughout the world have experienced land subsidence due to peat oxidation (Shoham and Levin, 1968; Prokopovitch, 1985; Nieuwenhuis and Schokking, 1997; Wo¨sten et al., 1997). Numerous studies have addressed the hydrologic properties of histosols that contribute to the control of the process (Boelter, 1964, 1965; Sturges, 1968; Dasberg and Neuman, 1977; Neuman and Dasberg, 1977; Weiss et al., 1998; Letts et al., 2000). This chapter is focused on the deformation properties of cropped peatlands under the drainage regime dictated by the agricultural practices in the Venice watershed. It is generally recognized that irreversible long-term land subsidence of drained peat soils is mainly related to biochemical aerobic oxidation of the organic matter (Stephens et al., 1984; Deverel and Rojstaczer, 1996). In an anoxic environment the action of anaerobic microorganisms breaks down the plant structure and creates peat that accumulates faster than it decomposes. Drainage for agricultural purposes leads the ISSN: 0928-2025

DOI: 10.1016/S0928-2025(06)09023-7

530

M. Camporese et al.

soil to aerobic conditions and microbial activity oxidizes the carbon of the peat causing carbon loss mainly in the form of gaseous CO2 ﬂux from the soil to the atmosphere (Rojstaczer and Deverel, 1993; Clair et al., 2002). Experimental ﬁeld and laboratory studies show that the magnitude of this process is primarily controlled by soil temperature and moisture content. Soil microbial activity becomes signiﬁcant when soil temperature is higher than 51C and generally doubles for each 101C increase in temperature (Stephens et al., 1984). CO2 ﬂuxes from the peat are thus inﬂuenced by depth of drainage and/or the soil–moisture content, the higher the water table the lower the loss of soil mass (Silvola et al., 1996). These relations are supported by the worldwide subsidence rates recorded in drained peat areas, which range from about 1–2 mm yr1 in the polders of the western Netherlands at a very shallow depth of drainage (0.1–0.2 m) (Nieuwenhuis and Schokking, 1997), to 2–3 cm yr1 in the temperate Sacramento–San Joaquin delta in California (Deverel and Rojstaczer, 1996), up to more than 5 cm yr1 in tropical peatlands, such as in Malaysia (Wo¨sten et al., 1997). The CO2 release, and hence the sinking rate, is enhanced by agricultural practices (plowing for example), that increase soil aeration and mechanically destroy the ﬁbrous peat structure (Andriesse, 1988). Recently a few investigations have shown that the peat surface may experience signiﬁcant, partly or totally recoverable, vertical displacement related to changes in the water-table depth. This peculiar peat characteristic (mire breathing) has usually been observed on a seasonal timescale and has been accounted for by changes in soil volume both above and below the water table. Peaks in bulk density in the unsaturated zone correspond to drier periods when the larger matrix suction results in a decrease of pore volume (shrinkage). Lowering of the water table induces saturated peat compression as the effective stress increases (Price, 2003). Deverel and Rojstaczer (1996) suggested that changes in the peat surface elevation related to a ﬂuctuating water table are probably due to buoyancy effects, as the bulk density of the organic soil is close to, if not less than that of water. Due to its high water content, peat is highly compressible with a volumetric change up to 10 times larger than in swelling clay soils (Hobbs, 1986). Peat displacements induced by water content changes are relatively small (1 cm) in highly mineralized and amorphous organic soils (Deverel and Rojstaczer, 1996), whereas ﬁbrous, poorly decomposed peatlands, with a very high water content, may experience seasonal movements up to 10 cm (Price and Schlotzhauer, 1999) or even 50 cm (Roulet, 1991). Note that the elevation change is also a function of the peat layer thickness. On the basis of daily measurements of water table and land surface elevation in a cutover peat ﬁeld in Que´bec, Schlotzhauer and Price (1999) found settlements of the peat surface that range between 11% and 23% of the lowering of the water table, with a considerable hysteresis, being the vertical displacement ﬁve times greater in response to water loss compared to rewetting. However, the measuring device employed by Schlotzhauer and Price (1999) was unable to capture the short time deformations and hence the reversible displacements. In fact, Price (2003) concluded that peat subsidence resulting from water table drawdown is largely recoverable when wetter conditions are restored. Peat swelling and shrinking signiﬁcantly affect also the soil hydraulic parameters related to pore-size volume, including water retention, hydraulic conductivity, and soil speciﬁc yield (Price and Schlotzhauer, 1999; Price, 2003).

Peatland subsidence in the Venice watershed

531

Large agricultural areas reclaimed from 1892 to 1967 south of the Venice Lagoon, Italy, are characterized by soils with high organic content. At present, these areas lie almost entirely below the mean sea level mainly because of land subsidence due to peat oxidation. The farmland is artiﬁcially kept drained by a complex network of ditches and pumping stations that discharge the drainage water into the Venice Lagoon or the Adriatic Sea. Flooding from the sea and the lagoon is prevented in normal conditions by levees. The maintenance of a given water table depth, as is dictated by agricultural requirements, has caused the lowering of the inlet shaft of the drainage pumping stations and the simultaneous increase of the pumping head. As a major consequence, the efﬁciency of the pumps has decreased with corresponding increase of the drainage costs. Moreover, the risk of ﬂooding during severe storms cannot be ruled out as well as the occurrence of adverse events such as saltwater contamination from the nearby rivers, the lagoon, and the sea (Rizzetto et al., 2003). Hence, the development of the area as a cereal farmland is becoming increasingly expensive, and is clearly unsustainable over the long term. In view of the above, the research project VOSS (Venice organic soil subsidence) was undertaken with the aim of characterizing the composition of the histosols, deﬁning the extent of the subsiding area, understanding the basic processes affecting the occurrence, quantifying the past and present land subsidence rate, and ﬁnally developing a predictive tool for helping to plan the most appropriate management strategies in relation to the dominant agricultural practices and the maintenance of an efﬁcient drainage network, which would be able to safely protect the farmland from exceptional ﬂoods. A ﬁeld site was instrumented at the end of 2001 to investigate the occurrence of land subsidence in the Zennare Basin (451 100 N latitude and 121 90 E longitude), a reclaimed agricultural area in the south catchment of the Venice Lagoon (Fig. 23.1). Drainage of outcropping peat soils has resulted in an overall settlement of 1.5–2 m since the 1930s. Continuous measurement of the hydrological regime and peat surface displacements by an ad hoc tool proposed by Deverel and Rojstaczer (1996) has allowed for an accurate estimate of both the peat reversible movement and irreversible subsidence. This chapter presents a review of the VOSS project that was conducted in close collaboration with the Land Reclamation Authority (Consorzio di Boniﬁca AdigeBacchiglione) and the farmland owners, and funded by Co.Ri.La. (Research program 2001–2003) and Sistema Informativo MAV-CVN. After a short description of the area of interest and the experimental site, the collected data are shown and discussed. The recoverable and unrecoverable components of the measured displacements are evaluated on the basis of a direct analysis of the recorded time series and by use of mathematical models relating peat porosity to moisture content, and peat oxidation to soil temperature and depth to the water table, respectively. Finally, some remarks are provided by way of conclusion.

Peat oxidation and geochemical land subsidence Land subsidence in cropped temperate and tropical peatlands is a major consequence of the microbial oxidation of the soil organic fraction in the upper aerated zone. The

532

M. Camporese et al.

Figure 23.1. Maps of the study area. (a) Map of the northeastern Italy. (b) Map of the Zennare Basin in the southern catchment of the Venice Lagoon. The major watercourses and the leveling lines used to measure land subsidence since 2000 are shown (after Gambolati et al., in press).

biochemical reaction of interest can be represented as follows: C6 H12 O6 þ 6O2 ) 6CO2 þ 6H2 O The release of carbon dioxide to the atmosphere causes a soil mass loss, which turns into an apparent layer compaction with a signiﬁcant land settlement. The reaction is controlled by temperature and is limited by the absence of oxygen. Therefore, the lower is the degree of water saturation in the subsoil and the higher the ambient temperature, the faster the reaction rate. The depth of the water table affects the soil–water content and the zone of aeration and hence the exposure of soil to oxygen. Since the soil moisture is sensitive to the amount of precipitation, we can conclude that dry and hot seasons are most favorable to the occurrence. By contrast, soil oxidation slows down to negligible values in winter. In light of the above we expect that this type of anthropogenic land subsidence in the future might increase should extreme climate events (hotter and dryer seasons) become more frequent, as the most recent meteorological records seem to indicate.

Description of the study area The Zennare Basin (Fig. 23.1) is an area of about 23 km2 reclaimed about 70 years ago, located directly south of the Venice Lagoon, approximately 10 km far from the

Peatland subsidence in the Venice watershed

533

Adriatic Sea. The organic soil in the study area is in the form of amorphous granular peat derived from the accumulation and decomposition of reeds (Phragmites australis) grown in the ancient marshes of the lagoon surroundings, where the oxidation reaction could not occur due to anaerobic conditions. After reclamation, aerobic conditions were established in the upper soil (few tens of centimeters). Moreover, the seasonal plowing for cereal cropping brought poorly decomposed peat to the ground surface. This contributed to the exposition of new organic material to the atmosphere hence enhancing the oxidation process and promoting new subsidence. Figure 23.2 shows the surface geomorphology of the Zennare Basin derived from aerial photographic interpretation, ﬁeld surveys, stratigraphic analyses, and altimetric investigations (Rizzetto et al., 2003). Note the large extent of the area where soils are rich in organic matter (peat) providing evidence of ancient swamps. Currently, the thickness of the remaining peat layer averages 1 m. Prevailing sand and silty sand are present in the northern part of the basin, whereas traces of a few paleorivers, most probably related to the ancient Adige River and with the main direction toward the Venice Lagoon margin (Fig. 23.2), are visible throughout the catchment. The Zennare Basin lies almost entirely below mean sea level, except for a small part in the

Figure 23.2. Geomorphologic map of the Zennare Basin (after Rizzetto et al., 2003). The trace of the DGPS survey performed in March 2002 is shown. The dots/numbers indicate the local estimated settlement (m) from 1964 to 1983. The circles/numbers represent the average displacement rate (centimeter per year, negative value means uplift) measured over the time period 1992–2000 using satellite radar Interferometric Point Target Analysis (IPTA) (after Gambolati et al., in press).

534

M. Camporese et al.

northern corner where mineral soils prevail. On the basis of an aerial photographic survey performed in 1983 (and never re-performed since then), elevation ranges mostly from 2 to 4 m below mean sea level. The areal extent of peatlands was investigated using satellite data (Nicoletti et al., 2003). Several images from the IKONOS, ASTER, and LANDSAT-7ETM+ satellites, which combine high geometric (1 m2 for IKONOS) and high spectral (6 bands for LANDSAT and 14 for ASTER) resolution, were analyzed and calibrated against the geomorphologic map of Figure 23.2 and a large dataset of peat spectral signatures collected in situ using a portable spectrometer. The best results of the spectral analysis have been obtained from a density slice of the synthetic Brightness band obtained from the Tasseled Cap analysis of the LANDSAT data. Scenes collected between February and May provide the best data source as the farmland is already plowed, so that no crop residues are present on the surface, and vegetation is only partially developed. The peat areas detected using the above modern technology well compare with an 1833 map of the local marshes drawn by government ofﬁcials of the Lombardo-Veneto kingdom (Fig. 23.3) (Gambolati et al., 2005). Although ad hoc campaigns to measure land subsidence in this part of the Venice territory were never made before 2000, a number of independent factors point to a long-term subsidence rate of 2–3 cm yr1. Land subsidence estimated from 1964 to 1983 by the reclamation authority, with the aid of elevation maps of the area, is provided in Figure 23.2. Note that over less than a 20-year period a maximum

Figure 23.3. Maps of the Venice area. (a) View of the Venice watershed with the map of the reedy and marshy areas drawn in 1833 during the Lombardo-Veneto kingdom and reclaimed in the past century. The dashed box identiﬁes the location of the map reported in (b). (b) Peatland map as derived from the spectral processing of the LANDSAT image of March 25, 2003. The boundary of the Zennare Basin is highlighted (after Gambolati et al., 2005).

Peatland subsidence in the Venice watershed

535

settlement of about 1 m was measured near the northwestern and the southwestern boundaries. As an example, Figure 23.4 shows the practical consequences of land settlement due to peat oxidation that has affected the area surrounding the hydraulic infrastructures. The protrusion above the groundwater level is nowadays a common feature of all the hydraulic installations that were built at the time of the basin reclamation in the central and southern part of the area where organic soils outcrop. In fact, peat oxidation is strongly reduced below the constructions due to the reduced oxygen availability. Furthermore, many of the infrastructures were founded on piles. The Zennare pumping station (Fig. 23.1b) provides further evidence, albeit indirect, of the magnitude of subsidence. The shaft level of the pumping station that drains off the surplus water from the study basin has been continuously lowered from the original 1930 elevation to conform with the subsiding ground elevation of the surrounding area (Fig. 23.5). The water level in the channels Magnana and Gorizia (Fig. 23.1b), which have ﬂowed together into the Zennare station since 1970, shows an overall average lowering of the order of 1.5 m (Gatti et al., 2002).

Figure 23.4. Evidence of the anthropogenic land subsidence in the reclaimed area. (a) A bridge has been turned into a useless structure: the left drainpipe helps convey the water of the channel originally ﬂowing through the protruding infrastructure. (b) An old masonry culvert presently above the water level and substituted by two lower concrete drainpipes, the higher of which already unusable. The approximate position of the ditch section in the original conﬁguration is sketched. (c) The protrusion of a sluice wall above the bed of an old disappeared channel. (d) An old bridge hanging over the canal bank that settled by 1.5 m.

536

M. Camporese et al.

Figure 23.5. Water level (asl) in the Magnana and Gorizia channels at the Zennare pumping station from 1930 to 2000. The channels have joined and ﬂow together into the station since 1970 (from Gambolati et al., in press).

Figure 23.6. Comparison of ground elevation along proﬁle A–I of Figures 23.2 and 23.3 as derived from the 1983 DEM and the 2002 DGPS survey (from Gambolati et al., in press).

Figure 23.6 provides documentary evidence that land subsidence is still occurring in the southwestern area of the basin. This ﬁgure compares the ground elevation along the AG and HI lines indicated in Figure 23.2 as derived from the DEM constructed with the 1983 data and a Differential Global Positioning System (DGPS) survey carried out in March 2002 using as a reference point a benchmark of the

Peatland subsidence in the Venice watershed

537

regional leveling/GPS network established at point I of Figure 23.2 and leveled in the year 2000 (Tosi et al., 2000). It can be noted in Figure 23.6 that some areas have subsided at a rate ranging from 2–3 cm yr1 (line AC) to 4–5 cm yr1 (line CD), whereas point C and segments EG and HI turn out to have been much more stable. This outcome is to some degree consistent with the areal peat distribution displayed in Figure 23.2. The stable point C corresponds to the intersection with a paved road along a sandy paleoriver, and the FG and HI lines run along two country roads across the basin. The stability of the northern part of the basin has recently been conﬁrmed by means of satellite radar interferometry. This technique, that interprets the radar phase signals of permanent structures scattered on the ground (Interferometric Point Target Analysis, IPTA) (Werner et al., 2003), has been implemented around the Venice Lagoon by processing 53 ERS-1/2 SAR scenes between 1992 and 2000 (Strozzi et al., 2003). A number of reﬂectors have been detected in the northern sandy–silty portion of the basin showing a displacement rate of less than 0.5 cm yr1, hence suggesting that the areas where peat is not present are subject only to natural subsidence, with a much smaller rate estimated at a few millimeters per year (Carbognin et al., 2004). The loss of mass from peat soils, as the major subsidence process affecting this area, is somehow conﬁrmed by the signiﬁcantly smaller displacement rates (less than 0.5 cm yr1) that were measured by high precision leveling surveys since the end of the 19th century by IGM (Military Geographic Institute) and CNR (National Research Council) along leveling lines adjacent to the study basin (Fig. 23.1b) and established on mineral soils or running along the embankments of the Venice Lagoon boundary where oxidation of organic matter does not occur because of lack of oxygen (Tosi et al., 2000).

Field measurements and model setup Environmental variables A test site was selected in the southern tip of the Zennare Basin (Fig. 23.1) on a portion of the peatland unplowed during the prior 10 years. The experimental ﬁeld was instrumented to monitor the land settlement to help to understand the overall process and predict its future occurrence. The following devices were installed and operated for more than two years (Fornasiero et al., 2003): (1) a tilting bucket pluviometer with a sensitivity of 0.2 mm; (2) a non-directional anemometer with an accuracy of 0.25 m s1; (3) two piezometers, one located within the test site and the other close to the adjacent ditch. Both were made from 3 m long PVC pipe of 5.08 cm diameter and instrumented with an atmospherecompensated pressure transducer characterized by a measuring range of 0–300 mbar and an accuracy of 71.5 mbar; (4) ﬁve tensiometers to measure the capillary pressure, inserted at a 451 slope so that the ceramic cups were all located along the same vertical line with a depth interval of 15 cm down to 75 cm; the measurement range of the electronic pressure sensor was from 1000 to 850 hPa with an accuracy of 70.2 hPa; (5) ﬁve three-wire time domain reﬂectometry (TDR) probes for soil

538

M. Camporese et al.

Figure 23.7. Steel tripod constructed to continuously monitor the elevation changes of the peat surface. The displacement transducer is shown enlarged on the right. The anchoring piles and the displacement transducers are highlighted.

moisture content measurement (accuracy 70.02 m3 m3), 15 cm long, inserted horizontally along the same vertical and at the same depth of the tensiometers, and connected to a multiplexer; and (6) ﬁve soil temperature sensors at 1, 5, 15, 30, and 100 cm depths with a measurement range between 15 and 501C and accuracy of 70.11C. As suggested by Deverel and Rojstaczer (1996), ground surface displacement was monitored by an ad hoc extensometer. Three displacement transducers, characterized by a measurement range of 0–25 mm and an accuracy of 70.125 mm, were attached at one end to a steel tripod anchored on three piles set into the ground to a depth of 11–12 m where an over-consolidated clay layer was located. The other end was connected to the land surface through a 0.5 cm thick, 10 10 cm aluminum plate resting on the soil (Fig. 23.7). The triangular steel structure, with sides of approximately 2 m, was designed to be as light as possible but with a negligible deformation with respect to the expected subsidence rate when loaded with the force exerted by the displacement transducers (2.5 kg each) and by a thermal excursion of 401C. All the sensors were connected to a datalogger and a 12 V, 260 Ah battery ensuring approximately one month of continuous functioning at an hourly data sampling rate.

CO2 fluxes A series of campaigns were performed in the Zennare Basin for the measurement of the CO2 released from the soil into the atmosphere. The ﬂuxes were measured with a portable Non Steady State (NSS) chamber (Hutchinson and Rochette, 2003). The

Peatland subsidence in the Venice watershed

539

chamber was a stainless-steel cylinder with a 20 cm base diameter and a height of 20 cm. It was open on the bottom face and it was coupled with a speciﬁc sharp-edge collar that was driven into the soil prior to the chamber installation. This procedure minimized both the soil disturbance and the pressure rise inside the chamber at installation time, while a shading cover was used to avoid large temperature ﬂuctuations due to the sun radiation. The evolution in time of the CO2 concentration is monitored inside the chamber by an infrared gas analyzer. The CO2 ﬂux f was calculated as qc/qt at the initial time t0, where c is the CO2 concentration, measured in ppm and then converted into milligram per liter (Camporese et al., 2004b). Records of the hydrological parameters controlling the oxidation process were carried out simultaneously to the ﬂux measurements. Soil and air temperature inside and outside the chamber was measured by thermocouple sensors and the soil volumetric water content was estimated by a portable TDR device. The CO2 ﬂuxes were converted into an estimate of land subsidence rate Z (cm yr1) by the formula (Deverel and Rojstaczer, 1996) Z¼

f c pc r po

(1)

where fc (g cm2 yr1) is the carbon ﬂux; r (g cm3) is the soil density (the peaty soil of the area has a r slightly larger than water); pc (dimensionless) is the percentage of carbon within the organic matter; and po (dimensionless) is the percentage of organic matter within the soil (approximately equal to pc). Detailed model of peat swelling/shrinking Peatlands respond to natural hydrologic cycles of precipitation and evapotranspiration with reversible deformations due to variations of water content in both the unsaturated and saturated zone. An original model for the simulation of the swelling/ shrinkage process in peat soil has been developed. Starting from the experimental observation that most of the deformations take place in the unsaturated zone (Price, 2003), the model takes into consideration the variation of porosity f with moisture ratio W (W ¼ V w =V s with Vw and Vs the volume of water and solid fraction, respectively, within the peat volume V). Using the developments proposed by Pyatt and John (1989) and Oleszczuk et al. (2003), the following swelling/shrinking constitutive relationship has been derived by Camporese et al. (2004a, 2006): ( ðW0 þ 1Þ1d ðW þ 1Þd 1 W W0 f ¼ (2) 1f W W4W0 where W0 (dimensionless) is the threshold value of the moisture ratio above which peat is totally saturated; d (dimensionless) is a soil parameter that increases with depth and ranges between 1/3 and 1 to account for a process of three-dimensional anisotropic peat deformation. Equation (2) has been implemented into a Richards equation-based numerical code through a suitable modiﬁcation of the general storage term. The calculation of the porosity evolution during the simulation is then used to obtain the changes in soil

M. Camporese et al.

540

surface elevation due to variations of the water content. The constitutive model has been shown in Camporese et al. (2006) to ﬁt well with the data of Oleszczuk et al. (2003). Stephens et al. (1984) model of peatland subsidence The data collected in the Zennare Basin allow for the use of the mathematical model proposed by Stephens et al. (1984). This model, developed for the Everglades marshlands in Florida, relates land subsidence due to peat oxidation to depth of drainage and soil temperature. By the use of Arrhenius’ law, stating that the logarithm of a chemical reaction rate is linearly dependent on the reciprocal of the absolute temperature, the authors provide the following basic subsidence equation: sT ¼ ða þ bhÞekðTT 0 Þ

(3)

1

where sT (cm yr ) is the land subsidence rate at temperature T; h (cm) is the depth to the water table from the peatland surface; k (1C1) is the reaction constant; T0 (1C) is the threshold soil temperature above which the biochemical reaction is activated; and a (cm yr1) and b (yr1) are constants to be estimated by ﬁeld or laboratory data. Denoting by Q10 the change of the reaction velocity for each 101C rise in temperature, k can be written as k ¼ 1/10 ln Q10 , and Equation (3) becomes TT 0 =10

sT ¼ ða þ bhÞQ10

(4)

From laboratory studies Stephens et al. (1984) derived Q10 ¼ 2 and T0 ¼ 51C. They calibrated a and b so as to reproduce the yearly subsidence rate sT at the Florida Everglades, with h and T the annual average depth to the water table and soil temperature at 10 cm depth, respectively.

Results Hydrology Hourly and cumulative precipitation, as well as soil temperature collected during 2002 and 2003, are shown in Figures 23.8 and 23.9, respectively. The ﬁgures show that very unusual climate features characterize the monitored time period. In fact, a very cold and dry winter occurred between 2001 and 2002, during which the top 10 cm of the peat soil was completely frozen for about six weeks and the cumulative precipitation totaled only 70 mm. This period was followed by a very rainy summer with a 400 mm cumulative rainfall (accounting for more than 50% of the overall 2002 rainfall). The entire area around the test sites was ﬂooded by 10 cm of water during the ﬁrst two weeks of August. On the contrary, total rainfall from January to September 2003 was only 280 mm, with very high and sustained temperatures that caused the upper 5 cm of peat to remain almost permanently above 401C for three months, with peaks well above the sensor FS (501C). The extremely dry conditions of

Peatland subsidence in the Venice watershed

541

Figure 23.8. Hourly and cumulative precipitation measured at the ﬁeld site over the 2002–2003 period.

Figure 23.9. Soil temperature recorded at the ﬁeld site over the last two years. Sensors at 5, 20, and 30 cm depth were established in November 20, 2001, the 1 and 100 cm deep ones in January 11, 2003.

summer 2003 are also proven by volumetric water content TDR measurements, which reached values below 30% in the upper 30 cm instead of the usual 50–70% range. Figure 23.9 indicates that, in general, daily temperature variations are large for the surface-most probes, small below 20 cm depth, and vanish at the 1 m deep sensor. It

542

M. Camporese et al.

Figure 23.10. Water table above mean sea level at the ﬁeld site and close to the adjacent ditch. The elevation of the ground surface is also shown.

is worth observing that also the deepest probe shows a seasonal ﬂuctuation of about 101C. Figure 23.10 shows the behavior of the groundwater table at the test site and at the two piezometers. The average depth ranges between 20 and 50 cm and 20 and 80 cm during 2002 and 2003, respectively. Signiﬁcant groundwater ﬂuctuations occur in connection with rainy events.

Recoverable peatland displacements Extensometer data collected during 2002 and 2003 (Fig. 23.11) have been analyzed to detect the recoverable and unrecoverable components of the measured peat surface movement. The continuous recording of the ground vertical displacement, soil temperature, precipitation, and depth to the water table reveals signiﬁcant recoverable vertical movements (swelling/shrinking) of the land surface as a response to drying/ wetting cycles. The elastic component of the displacement Dd can be easily detected on a short timescale of a few days or weeks during which the unrecoverable land settlement due to histosol oxidation is negligibly small. For example, hourly rainfall, depth to the water table, and land movements measured at the ﬁeld site from August 28 to November 11, 2002, are shown in Figure 23.12. The ﬁgure shows a good correspondence between the peaks of the rainfall records, the groundwater table depth, and the movement of the soil surface. Peat expands at each rainfall event. The swelling dynamics progresses very rapidly following the precipitation events in relation to the increase in soil moisture, whereas shrinking progresses at a slower rate, closely following the water table decline and exhibiting a timescale from a few hours

Peatland subsidence in the Venice watershed

543

Figure 23.11. Comparison between computed and measured land subsidence since February 2002 after ice melting.

Figure 23.12. Hourly precipitation, depth to the water table, and land subsidence measured between August 24 and November 11, 2002, at the Zennare Basin ﬁeld site.

544

M. Camporese et al.

to a few weeks. Note that the phenomenon is completely reversible. Looking at the single peaks, the ratio between Dd and the water table change Dh can be estimated at approximately 0.3 mm cm1 (Teatini et al., 2004). The dependence of Dd on h at larger timescales needs also to be evaluated. To this aim, the moving average method is employed for both Dd and h to reduce the complexity of the process and extract the main component of the ﬂuctuations. A window size of 60 days is used. The results of this calculation start after the ice melting in February 2002 and are displayed in Figure 23.13a. The hourly Dd and Dh computed by this approach are linearly regressed in Figure 23.13b to provide a proﬁle with a 0.3 mm cm1 slope, consistent with the value obtained over the short timescale. The modest correlation index R2 ¼ 0.54 indicates, however, that a more complex relationship exists between the two variables. The detailed model based on Equation (2) has been employed to provide a more accurate representation of the recorded swelling/shrinking peat deformations and the corresponding hydrological parameters. The ﬁnite element code based on the modiﬁed Richards’ equation has been applied for the simulation of the peat soil dynamics as measured in the Zennare Basin using the available hourly measures of rainfall and groundwater table depths as input data. The modeling results match very well with a large set of ﬁeld data and demonstrate that the model allows for an accurate reproduction of the elastic soil dynamics related to wetting/drying cycles (Fig. 23.14). Finally, signiﬁcant daily motions are observed in winter also in connection with temperature changes. Figure 23.15 compares the soil temperature at a 1 cm depth with Dd during February 2003. When the temperature approaches the freezing value

Figure 23.13. Water table displacement. (a) Measured surface displacement and water-table depth ﬁltered by the moving average method with a window size of 60 days. (b) Hourly displacements versus hourly water-table changes obtained with the moving average.

Peatland subsidence in the Venice watershed

545

Figure 23.14. Measured and simulated reversible displacement of the peat surface over the period October 8to November 2, 2002 (after Camporese et al., 2004b).

Figure 23.15. Soil temperature at 1 cm depth and surface soil displacement measured from February 9 to March 1, 2003, at the Zennare basin ﬁeld site.

546

M. Camporese et al.

at night, down to 21C at a 1 cm depth, a ground surface uplift up to 1 cm occurs, generally from midnight to 8 am. This is entirely recoverable and quickly dissipates at a rate between 0.2 and 0.5 cm h1 over a 2–3-h period when morning temperature starts to increase. Unrecoverable peatland subsidence To predict land subsidence due to peat oxidation we use the model of Equation (4) with the available measurements. Instead of computing a mean annual value, Equation (4) is applied over an hourly timescale consistent with the sampling rates of the available hydrological information, and the cumulative subsidence is compared with the recorded ground displacement. The soil temperature at 10 cm depth is obtained with a linear interpolation of the 5 and 20 cm deep measurements and the gaps in the h and T time series are ﬁlled in by linear interpolation as well. The constants a and b are calibrated so as to reproduce the overall observed subsidence behavior, starting in February 2002 after ice melting. Using the constraint that all the hourly calculated sT rates can never be negative (to guarantee the physical meaning of Equation (4)) and assuming sT ¼ 0 whenever soil temperature is less than 51C, we obtain the values a ¼ 0.025 cm yr1 and b ¼ 0.0085 yr1. Figure 23.11 compares the measured and predicted land subsidence and shows that a cumulative anthropogenic subsidence on the order of 2–3 cm occurred over the period of two years. Small or negligible rates characterize the summer and winter periods of the year 2002, when persistent and intense rainfall events occurred. Most of the settlement occurred in the very dry and hot 2003 summer. Obviously, the model is not capable of capturing the dynamics of the process related to the elastic peat shrinkage/swelling, but in general the overall long timescale settlement is satisfactorily reproduced. The model outcome indicates that the hydrological conditions experienced between July 2002 and April 2003 have practically precluded the oxidation of the histosols. About 60% of the unrecoverable land subsidence occurred over the last 6–7 months. The computed settlement rates range from zero during the coldest season to 6 cm yr1 during the hottest days in August 2003. These values agree satisfactorily with the subsidence rates obtained from the NSS chamber records through Equation (1). Figure 23.16 shows the measured CO2 ﬂuxes versus soil temperature and moisture content. The relationship between emission rate and water content appears to be unclear whereas a seasonal variation of the CO2 ﬂux may be clearly recognized, with the variability at any given temperature likely due to heterogeneity of the organic matter fraction of the peat. The CO2 ﬂuxes range between 0.02 and 0.7 mg m2 s1; that is, minimum (winter) and maximum (summer) value, respectively (Camporese et al., 2004b). From these data we readily obtain an estimate of the current anthropogenic land subsidence ranging between 0.1 and 2 cm yr1 in winter and summer, respectively. Scenarios using the calibrated Stephens et al. (1984) model suggest that, if the 2003 temperatures are projected into the future, the remaining peat layer will completely disappear in approximately 65 years for a constant water table depth of 60 cm. On the other hand, about 200 years would be needed to oxidize the peat if a water table depth of 20 cm is constantly maintained (Fig. 23.17).

Peatland subsidence in the Venice watershed

547

Figure 23.16. CO2 ﬂux versus (a) soil temperature and (b) water content measured in the Zennare Basin.

Figure 23.17. Expected land subsidence over the next decade as computed by the Stephens et al. (1984) model calibrated on the 2003 measurements collected at the Zennare Basin. The three scenarios assume the 2003 temperature and a constant water table depth of 20, 40, and 60 cm. Based on these data a 1 m thick peat layer would vanish in about 180, 90, and 65 years, respectively.

Conclusions Field experiments, data analysis, and predictive modeling suggest the following conclusions. The reclaimed farmland south of the Venice Lagoon is experiencing a pronounced microbial peat oxidation that has caused a cumulative anthropogenic subsidence between 1.5 and 2 m over the last 70 years. Ground and remotely sensed records provide evidence that land settlement has progressed at the rate of 2 cm yr1 or more during the last 20 years. Installed extensometers show a present trend of

548

M. Camporese et al.

1.5 cm yr1 whereas direct CO2 emission measurements point to up to 2 cm yr1. It is therefore worth noting that three independent recording techniques agree quite satisfactorily upon the current subsidence rate. Elastic recoverable deformations due to soil freezing or rainfall may superimpose on the long trend ground motion making it difﬁcult to correctly interpret the local point-wise occurrence. If no remedial strategies are used in the near future with no changes in the present soil practice implementation, the entire peat layer might disappear in about half a century with an additional 75–100 cm of anthropogenic land subsidence expected and extremely negative impact on the environment and the economy of the area. Extensometer data collected in 2002 indicate that settlement could be effectively mitigated by keeping the water table at a lower depth. However, to become an effective management strategy, a shallower phreatic surface requires new agricultural practices and an accurate and timely control of the drainage system and the pumping stations, possibly with the aid of forecasting models. As an example, conservative tilling as a substitute to plowing may help decrease the exposure of unmineralized peat to atmosphere whereas the introduction of cover crops may partially counterbalance the loss of organic material yielding a reduction of the current anthropogenic land subsidence rate. References Andriesse, J.P., 1988. Nature and management of tropical peat soils. FAO Soils Bull. 59, Rome, pp. 178. Armentano, R.V., 1980. Drainage of organic soil as a factor in the world carbon cycle. BioScience 30, 825–830. Avnimelech, Y., 1971. Nitrate transformation in peat. Soil Sci. 111, 113–118. Boelter, D.H., 1964. Water storage characteristics of several peats in situ. Soil Sci. Soc. Am. Proc. 28, 433–435. Boelter, D.H., 1965. Hydraulic conductivity of peats. Soil Sci. 100, 227–231. Camporese, M., Ferraris, S., Putti, M., et al., in press. Hydrological modeling in swelling/shrinking peat soils. Environ. Geol. W06420, DOI: 10.1007/s00254-006-0174-6. Camporese, M., Putti, M., Salandin, P., and Teatini, P., 2004a. Modeling peatland hydrology and related elastic deformation. In: Miller, C.T., Farthing, M.W., Gray, W.G., and Pinder, G.F. (Eds), XV International Conference on Computational Methods in Water Resources. Elsevier, Amsterdam, Vol. 2, pp. 1453–1464. Camporese, M., Putti, M., Salandin, P., and Teatini, P., 2004b. Spatial and temporal variability of CO2 ﬂux from a peatland south of Venice. In: Pa¨iva¨nen, J. (Ed.), Wise Use of Peatlands. Internationla Peat Society Publication, Tampere, Finland, Vol. 1, pp. 117–123. Carbognin, L., Teatini, P., and Tosi, L., 2004. Relative land subsidence in the lagoon of Venice, Italy, at the beginning of the new millennium. J. Mar. Syst. 51, 345–353. Clair, T.A., Arp, P., Moore, T.R., et al., 2002. Gaseous carbon dioxide and methane, as well as dissolved organic carbon losses from a small temperate wetland under a changing climate. Environ. Pollut. 116, S143–S148. Dasberg, S. and Neuman, S.P., 1977. Peat hydrology in the Hula basin, Israel: I. Properties of peat. J. Hydrol. 32, 219–239. Deverel, S.J. and Rojstaczer, S., 1996. Subsidence of agricultural lands in the Sacramento–San Joaquin delta, California: role of aqueous and gaseous carbon ﬂuxes. Water Resour. Res. 32, 2359–2367. Fornasiero, A., Putti, M., Teatini, P., et al., 2003. Monitoring of hydrological parameters related to peat oxidation in a subsiding coastal basin south of Venice, Italy. In: Servat, E., Najem, W., Leduc, C., and Shakeel, A. (Eds), Hydrology of the Mediterranean and Semiarid Regions. IAHS Press, Oxfordshire, UK, Vol. 278, pp. 458–462.

Peatland subsidence in the Venice watershed

549

Freibauer, A., Rounsevell, M.D.A., Smith, P., and Verhagen, J., 2004. Carbon sequestration in the agricultural soils of Europe. Geoderma 122, 1–23. Gambolati, G., Putti, M., Teatini, P., and Gasparetto-Stori, G., (in press). Subsidence due to peat oxidation and impacts on drainage infrastructures in a farmland catchment south of the Venice Lagoon. Environ. Geol. Gambolati, G., Putti, M., Teatini, P., et al., 2005. Peatland oxidation enhances subsidence in the Venice watershed. EOS, Trans. American Geophysical Union 86 (23), 217–224. Gatti, P., Bonardi, M., Tosi, L., et al., 2002. The peat deposit of the subsiding Zennare Basin, south of the Venice Lagoon, Italy: geotechnical classiﬁcation and preliminary mineralogic characterization. In: Campostrini, P. (Ed.), Scientiﬁc Research and Safeguarding of Venice – Corila Research Program 2001 Results. Istituto Veneto di Scienze Lettere ed Arti, pp. 241–257. Hobbs, N.B., 1986. Mire morphology and the properties and behaviour of some British and foreign peats. Q. J. Eng. Geol. 19, 7–80. Hutchinson, G.L. and Rochette, P., 2003. Non-ﬂow-through steady-state chambers for measuring soil respiration: numerical evaluation of their performance. Soil Sci. Soc. Am. J. 67, 166–180. Lal, R., 2004. Soil carbon sequestration impacts on global climate change and food security. Science 304, 1623–1627. Letts, M., Roulet, N.T., Comer, N.T., et al., 2000. Parametrization of peatland hydraulic properties for the Canadian Land Surface Scheme. Atmos. Ocean 38, 141–160. Levin, J., 1970. Improvement of soils affected by high nitrate concentrations. Isr. J. Agric. Res. 20, 15–20. Mount, J. and Twiss, R., 2005. Subsidence, sea level rise, and seismicity in the Sacramento–San Joaquin Delta, San Francisco. Estuary and Watershed Sci., 3 (March), Article 5. Neuman, S.P. and Dasberg, S., 1977. Peat hydrology in the Hula basin, Israel: II. Subsurface ﬂow regime. J. Hydrol. 32, 241–256. Nicoletti, V., Silvestri, S., Rizzetto, F., et al., 2003. Use of remote sensing for the delineation of surface peat deposits south of the Venice Lagoon (Italy). International Geoscience and Remote Sensing Symposium (IGARSS), Toulouse, France, Vol. IV, pp. 2881–2883. Nieuwenhuis, H.S. and Schokking, F., 1997. Land subsidence in drained peat areas of the Province of Friesland, The Netherlands. Q. J. Eng. Geol. 30, 37–48. Oleszczuk, R., Bohne, K., Szatylowicz, J., et al., 2003. Inﬂuence of load on shrinkage behavior of peat soils. J. Plant. Nutr. Soil Sci. 166, 220–224. Price, J.S., 2003. Role and character of seasonal peat soil deformation on the hydrology of undisturbed and cutover peatlands. Water Resour. Res. 39 (9), 1241 doi:10.1029/2002WR001302. Price, J.S. and Schlotzhauer, S.M., 1999. Importance of shrinkage and compression in determining water storage changes in peat: the case of a mined peatland. Hydrol. Process. 13, 2591–2601. Prokopovitch, N.P., 1985. Subsidence of peat in California and Florida. Bull. Assoc. Eng. Geol. 22, 395–420. Pyatt, D.G. and John, A.L., 1989. Modelling volume changes in peat under conifer plantations. J. Soil Sci. 40, 695–706. Rizzetto, F., Tosi, L., Carbognin, L., et al., 2003. Geomorphic setting and related hydrogeological implications of the coastal plain south of the Venice Lagoon, Italy. In: Servat, E., Najem, W., Leduc, C., and Shakeel, A. (Eds), Hydrology of the Mediterranean and Semiarid Regions. IAHS Press, Oxfordshire, UK, Vol. 278, pp. 463–470. Rojstaczer, S. and Deverel, S.J., 1993. Time dependence in atmospheric carbon inputs from drainage of organic soils. Geophys. Res. Lett. 20, 1383–1386. Rojstaczer, S. and Deverel, S.J., 1995. Land subsidence in drained histosols and highly organic mineral soils of California. Soil Sci. Soc. Am. J. 59, 1162–1167. Roulet, N.T., 1991. Surface level and water table ﬂuctuations in a subartic fen. Artic and Alpine Res. 23, 303–310. Schlotzhauer, S.M. and Price, J.S., 1999. Soil water ﬂow dynamics in a managed cutover peat ﬁeld, Quebec: ﬁeld and laboratory investigations. Water Resour. Res. 35, 3675–3683. Shoham, D. and Levin, Y., 1968. Subsidence in the reclaimed Hula swam area in Israel. Isr. J. Agric. Res. 18, 15–18. Silvola, J., Alm, J., Ahlholm, U., et al., 1996. CO2 ﬂuxes from peat in boreal mires under varying temperature and soil moisture conditions. J. Ecol. 84, 219–228.

550

M. Camporese et al.

Stephens, J.C., Allen Jr., L.A., and Chen, E., 1984. Organic soil subsidence. In: Holzer, T. (Ed.), ManInduced Land Subidence. Review of Engineering Geology, Geological Society of America, Boulder, Colorado, Vol. VI, pp. 107–122. Strozzi, T., Tosi, L., Wegmu¨ller, U., et al., 2003. Land subsidence monitoring service in the Lagoon of Venice. IGARSS 2003, Vol. I, pp. 212–214. Sturges, D.L., 1968. Hydrologic properties of peat from Wyoming mountain bog. Soil Sci. 106, 262–265. Teatini, P., Putti, M., Gambolati, G., et al., 2004. Reversibile/irreversible peat surface displacements and hydrological regime in the Zennare basin, Venice. In: Campostrini, P. (Ed.), Scientiﬁc Research and Safeguarding of Venice – Research Programme, 2001–2003. CORILA, Venice, Italy, Vol. II, pp. 93–106. Tosi, L., Carbognin, L., Teatini, P., et al., 2000. The ISES Project subsidence monitoring of the catchment basin south of the Venice Lagoon (Italy). In: Carbognin, L., Gambolati, G., and Johnson, A.I. (Eds), Land Subsidence. La Garangola, Padova, Italy, Vol. II, pp. 113–126. Weir, W.W., 1938. Subsidence of peat land in the Sacramento–San Joaquin Delta of California. In: Fauser, O. (Ed.), Tran. Commiss. VI Int. Soc. Soil Sci. Zurich August 1937. Bern, Switzerland, Vol. B. ISSS, pp. 304–314. Weir, W.W., 1950. Subsidence of peat lands of the San Joaquin Sacramento Delta, California. Hilgardia 20, 37–56. Weiss, R., Alm, J., Laiho, R., and Laine, J., 1998. Modeling moisture retention in peat soils. Soil Sci. Soc. Am. J. 62, 305–313. Werner, C., Wegmu¨ller, U., Strozzi, T., and Wiesmann, A., 2003. Interferometric point target analysis for deformation mapping. IGARSS 2003, Vol. VII, pp. 4362–4364. Wo¨sten, J.H.M., Ismail, A.B., and van Wijk, A.L.M., 1997. Peat subsidence and its practical implications: a case study in Malaysia. Geoderma 78, 25–36.

Glossary C-CPMAS-NMR: Solid state 13C-NMR using cross polarization (CP) and magic angle spinning (MAS). 2 CO2: Term commonly used in climate change literature to mean a doubling of the present CO2 concentration in the atmosphere. Aapa, aapa mire: Aapa mire complexes are mainly minerotrophic peatlands that have a string (linear peat ridge) and ﬂark (linear pools or hollows) pattern where the strings are perpendicular to the slope. Abiotic methylation processes: Coupling of a methyl group to another chemical compound as an inorganic chemical process. Abiotic processes: Processes that are independent of the living component of an ecosystem. Acid bath technique: Immersion in a bath containing an acid solution used to separate non-resistant from resistant material. Acid lavas: Volcanic rocks with silica contents typically greater than 66%, such as rhyolites or rhyodacites. Acidophilous plant communities: An assemblage of plants thriving in a relatively acidic environment. Acid-sulfate soil: Soil developed on pyritic parent materials. In the solum, the pyrite hydrolizes and oxidizes to produce pH values of 3 or less. Thionic ﬂuvisols in the WRB system of soil cassiﬁcation. Acrotelm: Actively growing, upper layer of a bog consisting of the living parts of Sphagnum and into which vascular plants are rooted. The acrotelm lies above the average water table level. It is only periodically saturated. This allows air into the upper layer and some decomposition to occur. Acrotelm–catotelm model: A two-layered peat pedological model consisting of an upper acrotelm layer of living plant tissue and poorly decomposed organic matter, and a ﬂuctuating water table, in which water can readily ﬂow. The lower layer, or catotelm, consists of saturated peat consisting of well-decomposed plant matter. Actinomycete: Phylum of fungi producing ascospores. Synonym: Actinomycota. Aelotropic: Non-isotropic; that is, having properties with different values along different axes giving rise to banded or layered structures in peat. Aerenchyma: A particular tissue found in many aquatic vascular plants and characterized by large intercellular spaces favoring air transport. Aerenchymatous: Plants with large internal air spaces – aerenchymatous tissues. Aerosols: Colloidal particles dispersed in air. 13

552

Glossary

Afforestation: The action of converting an area of land into a wooded zone, often but not exclusively, by human intervention. Age of peat deposit: The age usually determined by radiocarbon dating. Alcohols: Aliphatic organic compounds with an –OH group. Algae: Uni- and multi-cellular eukaryotes, containing chlorophyll and therefore capable of using CO2 in the manufacture of sugars. Aliphatics: Organic compounds consisting of straight or branched –CH2– chains. Alkanes: Organic compounds formed by chains of –CH2– groups without double bonds. Alkanoic: Organic compounds related to alkanes. Alkenes: Organic compounds formed by chains of –CH2– groups with at least one double bond n-alkanes – alkene with a chain length of n ‘free’ n-alkanes – that are not part of large polymers. Allochthonous: A term applied to material that has been transported to the site of deposition and therefore not formed in situ (compare to autochthonous). Allogenic material inﬂux: Material transported into an area from the surroundings of the area. Ammoniﬁcation: The process by which ammonia is produced. Anoxic environments: Environments lacking free oxygen. Anthropogenic: Originating by human action. AOX (adsorbable organic halogens): Sum of organically bound chlorine, bromine and iodine, which can be adsorbed on activated charcoal. Aqua Regis or aqua regia: Literally ‘royal water’, a mixture of concentrated nitric and hydrochloric acids capable of dissolving the noble metals. Arcellinida: One of two groups of testate amoebae, with lobose (broad) pseudopodia. Archea: Microorganisms with a special protein structure that allows them to live in extreme environments such as those of high salinity or high temperature. Archeomagnetic dating: Dating by measuring the remnant magnetism of Earthsurface materials, wherein constituents such as magnetite have been reorientated to the ambient magnetic ﬁeld, during a burning episode. Arctic: The area extending north from the most northward extension of trees (north of the tree line). Aromatics: Organic coumpounds having a C6 benzene-ring structure. Ascomycota: Phylum of fungi producing ascospores. Synonym: Actinomycete. Ash content: Percentage of residue after combustion of organic matter. Ashing technique: Technique used to concentrate the inorganic material of peat by combustion. asl: Above mean sea level. Authigenic: A term implying a development in place during or after deposition (see also diagenesis). Autochthonous: Consisting of material within a geological deposit, formed in situ (compare to allochthonous). Autotrophic: Organism using carbon dioxide as a carbon source. Autotrophic protist: Any autotrophic eukaryotic organism that is not a plant. Basal moraine: Low gradient, subglacially originated landform made of till.

Glossary

553

Base cations: Metallic cations such as calcium, magnesium, sodium and potassium that are plant nutrients and are involved in cation exchange processes. Baseﬂow: The stable portion of river discharge, contributed to by groundwater transfers. Basidiomycota: Phylum of fungi producing basidiospores. Synonym: Basidiomycete. Batholith: A large intrusive body with an areal outcrop greater than 50 km2. Most batholiths are granitoid in composition. Bdelloidea: A class of Rotifera. Biodiversity: The variety of life in all its forms, levels and combinations. Biological nitrogen ﬁxation (BNF): The conversion of molecular nitrogen to ammonia performed by procaryotes. Nitrogen ﬁxing bacteria live freely in the soil or in association and symbiosis with plants or animals. Biomagniﬁcation: The progressive build up of a given substance in rising from lower to higher trophic levels in an ecosystem. It relates to the concentration ratio in a tissue of an organism at the higher level compared to one at the lower level. Biosphere: The region close to the surface of the Earth, comprising elements of the lithosphere, hydrosphere and atmosphere, that constitutes the region occupied by living organisms. Biovolume: Volume of a living organism. It may be used for biomass estimates. Blackwater: Water rich in humic acids and with low nutrient concentrations. Blanket bogs: Bogs that form in areas with relatively high rainfall. In Europe they are found mainly on the lowlands in the westernmost countries and also on some coastal mountains. They are normally shallower than proper raised bogs forming a blanket-like layer of a few meters thick, over the underlying soil. Bloomery (iron bloomery): A furnace in which iron ore is smelted and metallic iron is produced. Bog: Gaelic word that refers to ombrotrophic mire. They are acid and poor in nutrients. Equivalent terms are: fagre (French), hochmoor (German), moor (English), mosse (Swedish) rahkasuo or ra¨me (Finnish), tremoal, borralleira, lamoso, chao (Galician), verxovoye (Russian), (see also blanket bog and raised bog). Boreal: It is a northern subarctic and humid biogeographic region in which the main plant life is coniferous evergreen spruces and ﬁrs, which are adapted to the cold climate. Peatlands are also common in this zone, which covers much of inland Canada and northern Russia. Boundary horizon: A sharp transition between two layers of peat with different properties and containing different plant remains. Below this horizon, the peat is usually darker, more humiﬁed, and burns better than the peat above it. Bowen ratio: The sensible heat ﬂux divided by the latent heat ﬂux. Bran˜a: From the Celtic term brakna, refering to cold, waterlogged and swampy areas with slow freshwater movement. In Galicia (NW Spain), its meaning is similar to hygro-peaty meadow or fen, being also termed poza, veiga or campa. Bronze Age: A diachronous time period marking the introduction of bronze tools and weapons. In the Balkans the earliest bronze implements date from as early as 4500 BC. In Britain, the Bronze Age is considered to have been the period from 2200 to 700 BC. In central Europe, the early Bronze Age has been located from 1800 to 1600 BC, the middle Bronze Age from 1600 to 1200 BC, and the late Bronze Age from 1300 BC to 700 BC.

554

Glossary

Bryales: Mosses having a hollow intercellular space between the spore case and the capsule wall. Burn-beating: The primitive form of agriculture where forests are burned and cultivation takes place in the ash-fertilized soils. Synonyms: slash and burn, swidden agriculture. Byelorussian peat corer: Chambered corer (also known as the Russian ﬂag peat borer) that collects samples that are not compressed or shortened during recovery. A common core size is 50 500 mm but both smaller and larger sizes occur. Capitulum: The top one-centimeter of a Sphagnum moss where growth occurs. Carbon Alkyl C: C atoms in –CH2– chains. O-alkyl C: C atoms in chains connected to O atoms (polysaccharides). Carboniferous: The period of geological time between 360 and 286 million years ago, named for the thick deposits of coal found in rocks of this age. Carboxyl group: Functional group that consists of a carbon atom joined to an oxygen atom by a double bond and to a hydroxyl group, OH, by a single bond, –COOH. Carr: Woodland growing on soils with permanently high water levels and usually dominated by alder or willow. Carr is a transitional type of wetland between fen and forest but there is a normal ground ﬂora of wetland species. Willow and alder are the dominant tree species, often with birch. Cation exchange capacity: Total content of exchangeable cations that a material may absorb, measured in centimoles of charge per kilogram of the absorbing material (usually soil). Catotelm: The water saturated, anaerobic/anoxic layer of the peat below the acrotelm. Cenozoic: An era of the geological time scale from 65 millions years ago to the present. Chalcophile: Applied to elements having a strong afﬁnity for sulfur and which tend to occur as sulﬁdes (Pb, Zn, Cu, Ag, for example). Chlor–alkali plant: Industrial plant where chlorine and sodium hydroxide are produced through electrolytic separation of brine. In this process elemental mercury is used as the cathode, and amalgamates with the sodium. Chloroperoxidases: Enzymes able to oxidize chloride to a more reactive form that can react with natural organic matter. C-horizon: Soil horizon immediately below the solum and taken to be the parent material of the soil above except in the case of polycyclic soils. Chroococcales: A group of cyanobacteria. Chrysmonad cysts: Protozoan silicate containers, sometimes reminiscent of ﬂasks. Thick-walled ﬂask-shaped cysts are thought to be resting cysts. Chrysophyceans: A group of unicellular algae also called golden algae. Cladocera – water ﬂeas: A group of micro-crustaceans. Climatic moisture balance: The balance between water inputs (precipitation) and outputs (evaporation) for a geographical location over a given period of time. Coefﬁcient of variation: A measure of the spread of the data around the average where a value of 30 or more indicates high variability (it is the standard deviation divided by the average and multiplied by 100).

Glossary

555

Condensation nuclei: Small particles or aerosols upon which water vapor attaches to initiate condensation. Dust particles, sea salt, sulfur and nitrogen oxide aerosols serve as common condensation nuclei. Conservative elements: Chemical elements which are assumed to be conserved in a natural material (soil, for example); that is, any process affecting that material does not change their absolute content. Continuous permafrost zone: Region with 490–100% of its area underlain by permafrost. Copepods: A group of micro-crustaceans. Cretaceous: Period of the geological time scale, between ca. 144–65 million years before present (BP). Critical load (pollution): The amount of a pollutant in an ecosystem, above which signiﬁcant harmful effects take place. Cryic horizon: A cryic horizon is deﬁned in the WRB classiﬁcation of soil as having (1) a (soil) temperature at or below 01C for two or more years in succession; (2) (in the presence of interstitial soil water) evidence of cryoturbation, frost heave, cryogenic sorting, thermal cracking or ice segregation; (3) or (in the absence of interstitial soil water) evidence of thermal contraction of frozen soil material; and (4) platy or blocky macro-structure resulting from vein ice development, and orbicular, conglomeratic and banded micro-structure resulting from sorting of coarse soil material. Cryosols: Soils having a cryic horizon within 100 cm of the surface (WRB classiﬁcation of soils) of very cold areas containing permafrost within 2 m of the soil surface. Curie-point pyrolyser: Apparatus in which organic matter is decomposed on a metal alloy rod that has a ﬁxed temperature of energy transfer (the Curie point). Cushioned meadow: Softened padded ﬁeld where grass grows. Cyanobacteria: A group of bacteria performing oxygenic photosynthesis. They were once thought to be algae and were called blue-green-algae. Cyperaceae: Pertaining to, or resembling, a large family of plants of which sedge is the prototype. Cutinite: (see liptinite). Decay constant (atomic): Constant (usually in 1010 years) representing the probability that an atom will decay in a given time period. Dendrochronology: Dating technique based on counting growth rings of living and dead wood. Desmids: A group of unicellular green algae taxonomically close to higher plants. Detrital: A term in geology applied to particles of minerals or rocks assumed to represent detritus worn away from an older body. Deuteromycota: Fungi that do not produce fruiting bodies but that, in most cases, belong either to the basidiomycota or actinomycota. Synonym: fungi imperfecti. Diagenetic (diagenesis): Post-depositional modiﬁcation to a sediment. It involves compaction, cementation, alteration of pre-existing minerals and formation of new ones. Diaspores: The dispersible reproductive units of fungi, mosses and ferns. Not to be confused with the mineral diaspore, AlO(OOH).

556

Glossary

Diatom frustules: Diatom shells formed by amorphous silica (opal). Some 100,000 varieties occur, all different, some with detailed geometric patterns. Diplotelmic: Two layered system. Dipterocarp: A member of the family Dipterocarpaceae, comprising approximately 600–700 species of mainly tropical lowland rainforest trees with two-winged fruits. Discontinuous permafrost zone: Region with 50–490% of its area underlain by permafrost. Ditch blocking: The process of damming or inﬁlling surface ditches; often involves installing plastic or wooden piling, or inserting peat or vegetation into the ditch to block it. Dome bog: A large bog (usually more than 500 m in diameter) with a convex surface, rising several meters above the surrounding terrain. Dopplerite: Humic gel complex comprising oxygen-reduced organic substances combined with iron, calcium and magnesium from bedrock, hardened to form a brittle dark brown material. Dry fall: Nutrient input to an ecosystem by dry deposition (as opposed to wet deposition in rainfall); includes dust, bird feces, insect droppings and nutrients with a gaseous phase, such as ammonia. Ecohydrology: The study of the behavior of water in an ecosystem. Ecotone area: An intermediate area between two adjacent ecosystems. Ecotoxicology: A branch of pathology or medicine concerned with the study of toxic substances in an ecosystem. El Nin˜o southern oscillation (ENSO): Warming of sea-surface temperatures in the equatorial Paciﬁc Ocean, which inﬂuences atmospheric circulation, and consequently rainfall and temperature in areas around the world. Eluvial horizon (E-horizon): Gray eluvial mineral soil horizon, below the surface organic horizon, and from which aluminum and iron are leached. End-member: One of two or more relatively simple (chemical) components used to deﬁne the composition of a mixture. Endolithic fungi: Fungi growing inside rocks. Endolithic fungi comprise a signiﬁcant component of the microﬂora in a wide range of rocks including silica, silicate and aluminosilicate types, limestone, marble and gypsum. Endo-symbiont: Symbiotic organism living within the cell of its host. Enrichment factors (EF): A ratio that indicates the enrichment of an element (M) compared to a lithogenic element (X), Sc, Ti, Zr for example, and normalized to a reference value for M/X. Enzymes: Proteins produced by living cells, which catalyze speciﬁc biochemical reactions. Chemical bonds broken by enzymatic reaction are said to be cleaved enzymatically. Epiﬂuorescence microscope: Light microscope to observe ﬂuorescence; that is, light of longer wavelength emitted by a specimen when it is excited by high intensity, shorter wavelength light (usually ultraviolet). Epipaleolithic age: The Epipaleolithic Era is the same time period as Mesolithic (9000–5500 BC), but is used in areas where glaciers did not exist. It was introduced to cover the period from the last ice age until the introduction of farming.

Glossary

557

Epyphite: Any organism growing on the surface of a living plant that remains independent of the plant except for physical support. Ergosterol: A crystalline sterol synthesized by yeast from sugars. Ericaceae: A family of plants of which Erica, or heather, is the common genus. Esters: Organic compounds having a –C–O–C– link. Euglenoids: Group of unicellular protists that includes both autotrophic and heterotrophic organisms. Synonyms: Euglenophyceae, Euglenophytes, euglenoid ﬂagellates. Euglyphida: Group containing most of the testate amoebae with ﬁlose (thread-like) pseudopodia. Eukaryotes: One of the three domains of life made up of cells containing nucluei; includes protists, animals, plants and fungi. Eutrophic: Designation applied to wetlands in which plants grow in nutrient-rich water. Eutrophication: The biological effects associated with an increase of nutrients, usually nitrogen and phosphorus, in aquatic ecosystems. Evapotranspiration: A collective term that includes the discharge of water to the atmosphere as a result of evaporation (a physical process) from soils and surfacewater bodies and as a result of plant transpiration (a biological process). Exchangeable cation content: More usually referred to as cation exchange capacity. Exoenzymes: Enzymes found in the environment and originating in the secretions of bacteria and fungi. Extensometer: Device to measure the displacements of the ground surface relative to a datum plane. Instrument for measuring deformations of ﬁne-grained beds under stress. Extra-Andean Patagonia: That part of Patagonia between the Andean ranges and the Atlantic coast. Fatty acid: An organic CH3–CH2– chain with a terminal carboxyl (acid) group, which occurs in oils and fats. Fen: Minerotrophic mire with a permanent groundwater table close to the surface. The term includes the quaking bogs, some transition mires, and, in NW Spain, the bran˜as. Equivalent terms are: flachmoor or niedermoor (German), karr (Swedish), letto or motasuo (Finnish), nyzymoye (Russian), veen (Dutch). Fibric: A qualiﬁer in the WRB system of soil classiﬁcation referring to histosols having more than two-thirds (by volume) of the organic soil material consisting of recognizable plant tissue. Fibric material: It refers to the least decomposed organic soil material containing large amounts of well-preserved ﬁbers that allow the identiﬁcation of the botanical origin. Flark: Elongated, wet depression in patterned peatlands, generally fens. Flat bog: A bog having a ﬂat or level surface. Floating index of tree-rings: A tree-ring index that is not related to a speciﬁc time period. FISH: Method for ﬂuorescence in situ hybridization. Fluorochrome: Fluorescent dye used to stain living organisms.

558

Glossary

Flushes: Shallow depressions on slopes containing ﬂowing water on the ground surface most of the time, characterized by dense hydrophytic vegetation but not having true stream channels. Fly-ash: Fine ash particles formed by the combustion of a solid fuel. Folic horizon: Characteristic of histosols. A folic horizon is (1) more than 20% (by weight) organic carbon (35% organic matter); (2) not saturated with water for more than one month in most years; and (3) thicker than 10 cm; if a folic horizon is less than 20 cm thick, the upper 20 cm of the soil (after mixing) must contain 20% or more organic carbon. Folic material: Organic matter associated with the accumulation of layers of leaves, twigs, branches and mosses in an upland ecosystem as opposed to a wetland ecosystem. The material is not water saturated for more than a few days a year. Formaldehyde: HCHO, the simplest of the aldehydes, used in aqueous solution as a preservative. Fractionation (isotopic): Separation of the isotopes of an element on the basis of mass difference. It may result from natural or anthropogenic processes. Frozen peatlands: Peatlands underlain by permafrost; permafrost-affected peatlands. Frustule: Siliceous shell of a diatom. Fuegian: Of or from Tierra del Fuego, Argentina. Funginite (previously called sclerotinite): Dark colored maceral derived from fungal spores, tests and stalks. Fusiform: Cigar-shaped or spindle-shaped. Gas chromatograph: Instrument that separates volatilized compounds according to molecular weight and charge. Gas ebullition: The formation of bubbles in a liquid, like that accompany boiling. Gastrotrichs: A phylum of freshwater and marine micro-metazoa. Geogenous peatland: A peatland with nutrients and water supply derived both from precipitation and groundwater. Unlike ombrogenous peatlands, these have been in contact with mineral soils and rocks (compare to ombrogenous). Glaciation: Period during which continental ice sheets grow and spread outward over large areas of Earth. Gley: A grayish or bluish-gray soil layer in which Fe and Mn have been reduced and mobilized from the soil by waterlogging. Gleysol: Soils that are characterized by gley in the subsurface, as a consequence of saturation with respect to groundwater at that level (WRB soil classiﬁcation). Where the water table has been lowered, naturally or artiﬁcially, ped surfaces may become oxidized and show reddish, brownish or yellowish colors, but the interior of peds remains the grayish/bluish colors indicative of reduced ions such as Fe2+ and Mn2+. Glutaraldehyde: OHC(CH2)3CHO – a water-soluble liquid used as a ﬁxative. Graminoids: Herbaceous grasses (that is, species belonging to the family of Poaceae) and other narrow-leaved, grass-like plant species. Greenland ice: The continental ice sheet covering the island of Greenland. Greenschists: Low-grade metamorphic rocks with schistosity (a planar metamorphic structure) and greenish color imparted by the mineral chlorite. Grenzhorizont: See boundary horizon.

Glossary

559

Ground-penetrating radar (GPR): A geophysical technique involving a measurement of the propagation of high-frequency electromagnetic waves into the ground. These are then reﬂected back to the surface from boundaries at which there are contrasts in electrical properties. The technique allows high-resolution mapping of subsurface features such as bedrock, pipes and other anomalies. Groundwater-evapotranspiration models: Models that predict the inﬂuence of groundwater level on evapotranspiration rates, or models which couple groundwater processes to evapotranspiration processes. Guaiacols: Lignin fragments containing a guaiacol (2-methoxyphenol) structure. Haber–Bosch process (industrial nitrogen ﬁxation): A chemical process using extremely high pressure and moderately high temperature to combine atmospheric nitrogen (N) with hydrogen (H) to directly synthesize ammonia (NH3). Halide: Binary chemical compounds containing halogens as ions with a charge of 1. See TOX. Halogenases: Enzymes that catalyze the incorporation of halogens into organic compounds. Halogenation: Processes of binding halogens to other chemical components. Halogens: Family of elements that includes ﬂuorine, chlorine, bromine and iodine. Chlorine: Most abundant halogen in natural environment. Dechlorination: Biotic or abiotic processes to detach chlorine from chemical compounds, for example chlorinated hydrocarbons. Dehalogenation processes: Biotic or abiotic processes to detach halogens from chemical compounds, for example chlorinated hydrocarbons. Halomethanes: Molecules of methane (CH4) with one or more of the hydrogen atoms replaced with halogen atoms. Haloperoxidases: A peroxidase that catalyzes the oxidative transformation of halides to XO– (X being Cl, Br or I), or organic halogen compounds. Harpacticoid: An order of Copepoda. Head-loading: Addition of a mass of material to the upper part of a slope, increasing the disturbing downslope stress on the slope, used with peatlands to refer to the tipping of excavated peat or other mineral spoil material by engineers onto in situ peat. Helophytic: Helophytes are marsh-loving plants adapted to having their roots in water or saturated soil but which have their leaves in the air. Hemic materials: They are intermediate in their degree of decomposition, between the less decomposed ﬁbric and the more decomposed sapric materials. Heterocysts: A specialized cell of Nostocales where N ﬁxation occurs. Heterotroph: An organism that uses pre-formed organic C as a C source (such as animals). High Arctic: Arctic areas with a sparse and discontinuous vegetation cover. Normal sites are typiﬁed by a sparse cover (up to 15%) of mixed communities of dwarf (less than 10 cm) herbs and shrubs. Annual precipitation is less than 200 mm. High-center lowland polygon: A landform characterized by a convex surface separated by trenches overlying ice wedges. These trenches form a polygonal pattern when viewed from above. High-moor: A peatland (mire) with an elevated surface (see also raised bog).

560

Glossary

Histosol: Soil containing at least 12 to 18% of organic carbon by weight and exceeding 30–40 cm in depth (WRB system of soil classiﬁcation). Most histosols are peats. HOC (halogenated organic compounds): Organic compounds containing halogens. Holocene: The current epoch of the Quaternary period, conventionally extending over the last 10,000 radiocarbon years. Humic substances: Organic compounds or components formed during humiﬁcation by biotic or abiotic processes. Humiﬁcation: The degree of decomposition of organic material creating humus. There are three levels of peat humiﬁcation, ﬁbric (lowest level), hemic and sapric. Hyaline cells: Empty, colorless or transparent cells, lacking chlorophyll, used for water storage and transport in the genus Sphagnum. Hydraulic conductivity: Rate at which water can pass through a porous material under unit gradient. Hydrochemical: Pertaining to the chemistry of aqueous systems. Hydroecological model: A model that links ecological and hydrological processes. Hydrogenotrophs: A group of as yet unculturable Archea that consume H2. Hypsithermal interval: The warmest segment the current interglacial period. It began immediately after the rapid warming that led to the retreat of the glaciers. Huminite: It designates a group of medium gray macerals having reﬂectances generally between those of the associated darker liptinites and the lighter inertinites. Ice cap: A large permanent body of ice (glacier) covering a tract of land from the scale of a single mountain (such as Kilimanjaro) to that of a whole region (Greenland, for example). Ignifact: Term intended to mean any item shaped, deformed, affected or produced by ﬁre. Includes ﬁre cracked stones, melted rock fragments or grains of minerals, pieces of wood deformed by ﬁre, charcoal and tar. Illuvial horizon: The horizon, below the eluvial horizon, where aluminum and iron hydroxyl compounds and organic complexes have accumulated (see spodic horizon). Incongruent weathering: Chemical reactions producing dissolved species together with new solids that are more stable in the weathering environment than the original reactant minerals. Indole: A heterocyclic N organic compound. Inductively coupled plasma mass spectrometer (ICPMS): An instrument for the analysis of most elements in the periodic table. Capable of a high sensitivity with detection limits in the parts per trillion (ppt) range for many elements in aqueous solutions. Industrial revolution: Name given to the period of rapid economic and technological change in the 18th and 19th centuries. Inertinite: It is a maceral with no UV ﬂuorescence, very high reﬂectance, formed in peat that has been oxidized early in its formation, derives from bark, stems, leaves and roots. Inﬁltration: The movement of a liquid, such as water, through the surface of a medium (the soil surface). Inosilicate: Chain silicates. Interferometric point target analysis: Technique for the measurement of the deformation of the Earth surface by analyzing the interferometric phases on single pixels, which are coherent over long-time intervals.

Glossary

561

Inter-tropical convergence zone: An area of low pressure that forms in the atmosphere near the equator, and to which winds converge and moist air is forced upward, resulting in heavy precipitation. Iron age: The period when the working of iron came into general use, replacing bronze in tools and weapons. Iron working was introduced to Europe around 1000 BC, and Iron technology had spread throughout the classical world by about 500 BC. Isotopes: Two or more nuclides of a single element differing in the number of neutrons. Atomic numbers are the same but mass numbers differ. Julian day: The day in the year as numbered consecutively from 1 to 365, with January 1 being day one; not to be confused with Julian calendar day. Jurassic: Period of geological history between ca. 200–140 million years BP. Kerapah: Inundated tropical heath forest (kerangas) that can accumulate peat. Kettle: Ground depression generated by the melting of stagnant ice blocks surrounded or covered by sediments. Knoll: Small rounded hill or mound. In glaciated terrain it is usually associated with kettles. Lagg: The marginal edge of a peatland, containing many natural drainage channels. Lagoa: Galician word that refers to a wetland type with high water content, developed in a narrow permanent water layer, in contrast to the waterlogged wetlands with stagnant water named bran˜as. Depending on the degree of inﬁlling of terrestrialization of the lagoa, there is a gradient or sequence: lagoa – bran˜a – other fens. La-ICPMS (laser ablation inductively coupled plasma mass spectrometry): ICPMS in which the elements analyzed for, are excited from the surface of a sample by laser. Late Glacial: It refers to approximately the last 5000 years of the Pleistocene epoch, between 15,000 and 10,000 14C BP. LGM (last glacial maximum): A period of time during which the most recent glaciation cycle was at its peak; that is, from about 22,000 to 18,000 BP. LiDAR (light detection and ranging): The time taken for a highly focused beam of light (laser) to travel from a scanner to the ground surface and back to the scanner. It is used as a measure of distance from the scanner to the ground and allows highresolution remote measurements of topography to be collected very quickly. Ligand: (see organic ligand). Lignin moieties: Fragments of lignin molecules within the organic matter of soil or peat. Lipid stratigraphy: Stratigraphy based on the occurrence of speciﬁc lipids. Liptinite (exinite): It is a maceral group with hydrogen content and includes alginite, cutinite, resinite, and sporinite, depending on its origin. It has strong yellow to green UV ﬂuorescence, low reﬂectance. In type I kerogen derives from waxy, lipid-rich and resinous part of plants; in type II kerogen, from green algae and blue-green algae. Lithogenic element: Element preferentially incorporated into the lithosphere. Little Ice Age: A time period from between about 1350 and 1850 A.D. In particular there were three spells of intensely cold winters beginning about 1650, 1770 and 1850, each separated by slight warming intervals. The phenomenon was recorded in documents and by the advance of mountain glaciers, in North America, Europe and Asia. Surface temperatures in many regions were at least 1–21C lower than those of today. This cooling brought an end to an unusually warm era known as the medieval warm period.

562

Glossary

Loess: An accumulation of wind-blown silt. The silt derives either from material produced by the grinding action of glaciers or from desert areas. Loess was ﬁrst described in 1821 from the Rhine area, and takes its name from a village in Alsace. LORCA (long-term, apparent rate of carbon accumulation): Rate at which carbon has been sequestered in a peat deposit. This apparent rate does not take account of long-term peat decay and the ‘actual’ rate of carbon accumulation (ARCA). Low Arctic: Arctic areas with sites mostly characterized by a nearly continuous cover of high shrub tundra vegetation (usually less than 30 cm tall). Annual precipitation is usually less than 300 mm. Low-center lowland polygons: A landform on perennially frozen lowlands where the intense winter cold causes the formation of cracks and ice wedges of polygonal cross section in the horizontal plane. They occur at the raised edges of shallow, wide depressions. Low-grade metamorphism: Low temperature, low-pressure metamorphism characterized by zeolitic and/or chloritic minerals. Low moor: A mire with a near-level surface (see also fen). Macerals: They are phytogenetic organic substances (optically homogeneous aggregates of phytogenetic substances) possessing distinctive chemical and physical properties. They are the equivalent in coal to what minerals are in rocks. Macronutrient(s): Nutrient(s) needed in large amounts by plants. They generally include nitrogen, phosphorous, potassium, magnesium, calcium and sulfur. Macropore: Infrequent opening or void (generally greater than 0.1 mm in diameter) in the soil that can promote rapid, preferential transport of water and solutes. They are formed by structural cracking or ﬁssuring brought about by physical processes or by biological activity by earthworms, burrowing creatures and plant roots, for example. Macropore ﬂow: Groundwater ﬂow through large pores. Macroporosity: The density of macropores (by number or volume) in a given volume of soil. Maﬁc: Applied to any igneous rock that has a high proportion of ferromagnesian minerals such as pyroxene and olivine. Magellan transcurrent fault: A large, regional transcurrent fault separating the South American Plate from the Scotia Plate. Marine aerosols: Aerosols originating as sea spray mainly through the action of wind. Mainly contain inorganic components of seawater but organic components may also be present. Mass spectrometry: A method of detection (in the case of peat and humic materials) of molecular fragments on the basis of mass. Mearing ditch: Boundary ditch between lands of different ownership in Ireland. Mercury Dimethylmercury: Highly toxic, methylated organic mercury compound (CH3HgCH3), volatile at low temperatures. Methylmercury (MeHg) (CH3Hg+): Highly toxic methylated organic mercury compound. Exists as soluble methylmercury chloride or incorporated into natural organic matter. Methylmercury is frequently found in lakes and streams naturally or elevated as a result of industrial pollution. Methylmercury is bioaccumulated in ﬁsh

Glossary

563

and can cause great health hazards (nervous system) in those who eat the ﬁsh. Methylmercury is frequently found in lakes and streams as a result of industrial pollution. Mesocosm: An experimental apparatus, often indoor, designed to approximate natural conditions and in which environmental factors can be controlled and manipulated. Mesozoic: One of the major eras of geologic time, extending after the Paleozoic and before the Cenozoic, between about 250 and 65 million years ago. Meteorite ablation spherules: The extraterrestrial spherules are thought to have formed by melting of meteoritic metal or rocks during hypervelocity entry of meteoroids into the atmosphere. During entry the surface of larger fragments ablates, thus forming the sparkling lines that can be observed during falling star episodes. Methanogenic bacteria: A group of primitive microorganisms that produce methane gas as a byproduct of their metabolism. Methanomicrobiales: A group of anaerobic, highly specialized methanogen organisms. Methanotrophic bacteria: Aerobic microorganisms that oxidize methane as an energy source and carbon source through an enzymatic reaction. Methyl esters: Structures containing and –O–CH3 group. Methyl ketones (methylketone): –CH2– chains containing a –C ¼ O group. Microbial loop: Food web based on organic carbon assimilated by bacteria and fungi instead of CO2 ﬁxed by autotrophs. Mid Arctic: Arctic areas with normal sites characterized by a 40–60% cover of low shrub tundra vegetation. Annual precipitation is less than 300 mm. Mine adits: Small tunnels where ore has been excavated into a rock by man to extract the ore. Minerogenic mire: Mire formed under minerotrophic conditions. Minerotrophic: Receiving nutrients from surface runoff and groundwater containing dissolved minerals. Miocene: One of the epochs of the Cenozoic era, extending after the Oligocene and before the Pliocene, between about 23 and 5 million years BP. Mire: Variation of the Swedish word myr that includes all peatland habitats with active peat formation and associated characteristics. Its meaning is similar to active peatland. Mire breathing: The sequence of swelling and shrinkage events in a peatland. Mire macrotope: Mire complex representing the highest level of peatland hierarchy. It is a system containing several different mire types or mesotope hydrologically connected. The macrotope is regarded as a self-contained unit in terms of hydrology and vegetation. Mixotrophic species: A species that uses both CO2 and organic C as carbon sources. Moieties: General term for components with speciﬁc characteristics. Monogononta: Class of Rotifera. Moor: A term applied to any area of deep peat, whether acid or alkaline. In England, the word is applied to high-lying country covered with heather and other ericaceous dwarf shrubs.

564

Glossary

Moorland: Region where large areas are covered by peat. Mor: The mor is an organic soil horizon where there is an abrupt transition from the surface organic matter to the underlying mineral soil due to practically no mixing of organic matter with mineral soil. It is the most widespread organic horizon (rawhumus) in podzols and podzolized soils, which form the dominant soil type in largely coniferous-forested areas of Sweden. The mor is divided into a litter layer (L-layer), a fermentation layer (F-layer) and a humose layer (H-layer). MRT (mean residence time): Average time of residence of a mobile element within a system: the ratio of total content of the element in the system, to the net ﬂux (ﬂux out minus ﬂux in). NEE (net ecosystems exchange): The difference between the amount of CO2 ﬁxed by photosynthesis and the amount emitted by total community respiration. If NEE is positive then the peatland acts as a C sink, but if aerobic respiration dominates over photosynthesis, then NEE is negative and the peatland is a C source. Nematodes: Organisms in the phylum Nematoda; also referred to as roundworms. Nemoral: Adjective applied to broad-leaved forests. Neoglacial: Glaciation events developed during the Holocene epoch, much more restricted than those of the Pleistocene epoch. Neolithic period: Literally the ‘New Stone Age’. It refers to the period of human prehistory when the hunting and gathering lifestyle gave way to an agricultural one. The change is sometimes referred to as the Neolithic Revolution and is conventionally taken as beginning around 10,000 years BP, though recent work extends this to about 13,000 years BP. It was characterized less by the technology of stone grinding (the old justiﬁcation for the name), and more by the domestication of animals and plants, the cultivation of soil, the invention of metallurgy, the development of writing systems, and eventually (by about 7000–8000 years BP) the construction of cities (compare to Paleolithic). ng g1: The concentration of a substance or element in nanograms per gram (SI unit). NMLSD (northern and mid-latitudes soil database): A digital soil database for the mid and northern latitudes. NMR (nuclear magnetic resonance): A method used to determine the environment of C atoms (see 13C-CPMAS-NMR). Northern circumpolar area: The area encompassing those portions of the oceanic, boreal, subarctic and arctic regions lying north of approximately 401 N latitude in North America and approximately 501 N latitude in Eurasia. Northern circumpolar peatlands: Peatlands occurring in the northern circumpolar area. Nostocales: Order of cyanobacteria in which the organisms form colonies and produce heterocysts. Nuclide: An atom characterized by its atomic number and the number of neutrons in its nucleus. OIS (oxygen isotope stages): Geologic time periods that are characterized by a speciﬁc oxygen isotopic signal, that is, relatively richer or poorer in the stable, heavy 18 O isotope.

Glossary

565

Older Dryas: Short-lived cooling period between about 12,300 and 11,800 years BP in the Northern Hemisphere characterized by advances of alpine glaciers. Its simultaneous occurrence in the Southern Hemisphere has been a matter of debate during almost two decades. Oligochaeta: Order of Annelida; includes earthworms. OM: Organic matter. Ombrogenous peatland: A peatland whose nutrients and water supply are derived entirely from precipitation (compare to geogenous). Ombrotrophic: Receiving nutrients exclusively from precipitation or other windblown particles. Ombrotrophic mire: Wetland conditions and its nutrient supply are derived from direct bulk atmospheric deposition (rain, snow, fog, aerosol, dust). These rain-fed mires have been generally termed bogs. Organic ligands: A molecule or ion that forms a complex with a metal by donating an electron pair. Citrate ion is a simple example, and humate ions are complex varieties. Organochlorine: Chlorine bound in organic compounds. Organohalogen: Halogens bound in organic compounds. Organoiodine: Iodine bound in organic compounds. Oribatid mites: Order of minute Arachnida, acting mostly as decomposers of organic matter. Orographic effect: Forced ascent of a moist air mass over a mountain barrier producing precipitation. Orthosilicate (or nesosilicate): Silicate having SiO4 tetrahedra that are isolated from each other, and connected by octahedrally coordinated cations. Oxfordian (geological time): Part of the late Jurassic (ca. 157–156 million years BP). Oxic conditions: Conditions where free oxygen is available. Padang (open) community: An open, savannah-like, tropical, peat swamp forest community with stunted trees. Usually located on thick, waterlogged peat at the center of the peatland dome. Paleolithic period: Literally the ‘Old Stone Age’, period older than about 15,000 years ago, characterized by a human lifestyle based on hunting and gathering, and the use of unpolished stone tools derived by the chipping and ﬂaking of materials such as ﬂint and obsidian. Paleozoic: One of the major eras of geologic time, between 554 and 253 million years BP. Palsa: Dome-shaped landform that develops in permafrost areas. Composed of frozen peat and mineral soils, and generally up to 5m high with a maximum diameter up to 100 m. Palsa mires: Peatlands with developed palsas. Paludiﬁcation: Process of peatland formation on land due to waterlogging and the development of reducing conditions (compare to terrestrialization). Palynomorphs: Parts of plants, mainly pollen and spores, and animals that have undergone rapid evolution and therefore are characteristic or relatively short time ranges. Non-pollen varieties include fungal spores, testate amoebae, rhizopods and cynobacteria.

566

Glossary

Paralic: Pertaining to near shore environments, such as lagoonal, littoral, shallow neritic and similar situations. Particulate-phase mercury: Mercury attached to particles. Peat bog: General term at time used to mean peatland, commonly for those with acidophilic moss (peat moss or Sphagnum spp.) (compare to bog). Peatland: All peaty areas, under natural conditions, artiﬁcially drained or drastically transformed. A wetland with a vertical minimum peat thickness of 30 (in the Canadian classiﬁcation) or 40 cm (in the Russian classiﬁcation). Equivalent terms are: boloto (Russian), mase´cage or tourbie´re (French), mire (English), moor (German), muskeg (Canadian), suo (Finnish), turfeira, turbeira or tremoal (Galician), turbera or tremedal (Spanish). Peat plateau: A landform composed of perennially frozen peat, rising abruptly to about 1 m above the surrounding unfrozen fen peatland. The surface is relatively ﬂat and often covers a very large area (compare to plateau bog). Peat swamp: Forested peatland occurring in tropical regions with high rainfall, primarily in Southeast Asia. Includes both rain- and groundwater-fed peatlands. Peat swamp forest: Natural forest-vegetation cover of most lowland tropical peatlands. Ped: A small structure in soil, acting as a coherent unit or crumb, when the soil is broken apart. The smallest unit of soil lacking internal shrinkage cracks. Peepers technique: Peepers are instruments used to sample interstitial waters in sediments. They have a number of chambers, which allow section-wise sampling of porewater through a semipermeable membrane. Penman–Montieth model: A model to describe evapotranspiration incorporating a consideration of the inﬂuence of aerodynamic resistance to vapor transport. Perennially frozen peat: The portion of the peat that is perennially frozen in a permafrost-affected peatland. Permafrost: Ground (soil or rock) that remains at or below 01C for at least two years. Its distribution is subdivided into continuous permafrost areas where the terrains are perennially frozen, discontinuous permafrost areas where there are discontinuous thawed terrains, and sporadic permafrost areas where there are discontinuous permanently frozen terrains. Permafrost table: The upper boundary of the permafrost. Phenols: Organic compounds composed of one or more benzene rings, with at least one OH group attached to one of the rings; also called phenolic compounds. Phospholipids: Esters of glycerol incorporating a phosphate group with two fatty acids. Phytoliths: They are microscopic particles composed most commonly of silica that are made by many plants. Phytophagous: Feeding on plants. Picocyanobacteria: Cyanobacteria with diameter o2 mm. Pipeﬂow: Fluid ﬂow through soil pipes. Pipes (soil pipes): Tubelike subsurface cavities within the soil, usually deﬁned as being macropores larger than 1 mm in diameter. They may transmit water, sediment and solute through their length. Piping: Pertaining to the presence and form of soil pipes in a given soil.

Glossary

567

PIXE (particle induced X-ray emission): Technique used to determine the elemental make-up of a material. Plant litter: The uppermost soil layer consisting mainly of fallen leaves and other decaying organic matter. Plateau bog: A raised, unfrozen bog elevated from 0.5 to 1 m above the surrounding fen. The surface is generally even, with only small, wet depressions (compare to peat plateau). Platyhelminthes: A phylum of aquatic animals; ﬂatworms. Pneumatophores: Roots rising above the level of water or soil and acting as a respiratory organ in some peat swamp forest trees (mangroves, for example). Podzols: Acid soils with an ash-gray subsurface horizon above a dark horizon enriched in humic and/or iron compounds. Podzols occur in humid areas, in particular in the boreal and temperate zones, but locally also in the tropics. Podzolization: Soil forming process characteristic of, though not conﬁned to, humid, temperate to cold climate, coniferous forests and heathlands. Transfer of dissolved organic matter, and soluble organic complexes of Fe and Al out of the A horizon into the B is the formative process. It leads typically to the formation of a bleached lower part (the albic horizon) to the A, overlying a dark, complementary B (the spodic horizon). Polar front: Low-pressure atmospheric ﬁeld located between 401 and 601 latitude. Pollen ﬂux studies: A study of the actual pollen rain in an area, at a certain time, by introducing a known amount of pollen (exotic pollen) into a system. The number of pollen grains (arboreal, grass, shrub or total pollen) per surface unit and per time unit is tracked in order to calibrate movement of the pollen, and is used to estimate past pollen rain. Polyaromatics: Organic compounds having more than one benzene ring within the structure. Polysaccharides: A general term for structures built of sugar units. Precambrian shield: Complex of Precambrian metamorphic and plutonic rocks exposed over a large continental area. Pre-industrial: The period before the beginning of the industrial revolution; that is, prior to the late 17th to mid-18th century. Priestly–Taylor model: A simple evapotranspiration model that, unlike the Penman–Montieth model, does not simulate the inﬂuence of aerodynamic resistance to vapor transport. Primary production: Production of organic matter by autotrophs. Primulin: Fluorochrome used for observing protozoa by epiﬂuorescence microscopy. Prokaryotes: Unicellular organisms lacking a nucleus in the cell. They include bacteria and Archea. Protista or protists: They are eukaryotes, commonly single celled, belonging to the kingdom Protoctista, and including protozoa and algae. Protozoa: Polyphyletic group of heterotrophic protists including amoebae, ciliates and ﬂagellates. Proxy materials: Biological, physical or chemical characteristics of an Earthsurface material, used to obtain approximate values for paleoenvironmental variables such as ambient temperature, rainfall, wind direction and others.

568

Glossary

Pseudopodia: Extension of the cytoplasm in amoebae, used for feeding and locomotion. Pyridine: A heterocyclic N compound. Pyrogenic: Due to ﬁre. Pyrolysis: The disruption for analytical purposes of larger organic molecular structures, without oxidation, by application of a certain amount of energy or a speciﬁc temperature. Quaking bog: Floating peat deposit covering the surface of a pond shaking when walked on. Radioactive decay: Process by which a radioactive ‘parent’ element loses elementary particles from its nucleus and in doing so becomes a ‘daughter’ element that may decay further until a stable nuclide is reached. Radiocarbon date: Age of organic material determined by radiocarbon dating. Radiogenic isotope: Isotope produced by the process of radioactive decay. Raised bog: Acid, dome-shaped bog with low mineral nutrient content, sustained mainly by rainwater. Its development is the ﬁnal stage of a large evolutionary process, from an initial minerogenic mire phase, with a progressive centrifugal raising and thickness increase, to an ombrotrophic ﬁnal stage. Other terms used to describe these mires are: hochmoor (German), ho¨gmosse (Swedish) and raised moss (English). Rate of peat accumulation: Thickness of peat accumulated in a given period of time. Reclamation: The conversion of wetland by artiﬁcial drainage into land suitable for use as habitation or for cultivation. Red clay: Synonymous with brown clay and pelagic clay. The red color is due to ferric iron compounds, and indicates formation in an oxidizing environment. It is derived mostly from wind transport of dust weathered on land, especially in deserts. It generally contains extraneous components such as fallout from space, Mn nodules and fossils. The sedimentation rate of red clays is ca. 1–5 mm per 1000 years. Redox: Refers to chemical processes involving the transfer of electrons from a reducing agent such as organic matter, to an oxidizing agent such as oxygen. REE (rare earth elements): The series of elements with atomic numbers lying between 57 (lanthanum) and 71 (lutetium) inclusive, in the periodic table; also called lanthanides. Regmaglypts (or thumb-prints): Surface structures that appear on meteorites and micro-meteorites. The forms are reminiscent of smoothly rounded cavities. They originate by the process of surface ablation during entry through the atmosphere at very high speeds. Restoration: The process of returning a landscape, or elements thereof such as the water table, to a former state. This may not be possible and, in any case, may be inappropriate where natural changes have affected the former state targeted. Rheological properties: Physical properties of matter that relate to its behavior (such as deformation, ﬂow, shearing) in response to intrinsic and applied stresses. Ribbed fen: A fen peatland with parallel, low, elongated peat ridges alternating with wet hollows or shallow pools (ﬂarks), oriented across the major slope, at right angles to water movement. Rice cluster I: A group of as yet unculturable methanogenic Archea.

Glossary

569

Rifﬂe: A rocky shallow reach of river, in which the water moves quickly and becomes aerated. RNA: Ribonucleic acid. Roman lead peak: A characteristic peak in lead accumulation sometimes observed in peat, lake sediments, and glacial ice, in many parts of Europe as well as the Greenland ice sheet. It coincides with an established maximum in lead production during the late Roman period. Rotifers: Phylum of micro-metazoans with rotatory organs used for locomotion and feeding. Russian-type sampler: A manual peat sampler with a cutting edge designed to obtain a subsurface, 5 cm wide sample at a particular depth (see also Byelorussian peat corer). Sapric material: The most highly decomposed kind of organic matter in soils. Consequently they have the smallest amount of plant ﬁber, the highest bulk density, and the lowest water content on a dry-weight basis at saturation of any OM component. Sapric soil materials are commonly very dark gray to black in color. Scotia Plate: A small, very active, tectonic plate, between the major South American and Antarctic Plates, and moving eastwards from the Paciﬁc Ocean domain into the Atlantic Ocean domain. Sedges: Members of the Carex family (see also Cyperaceae). SEM: Scanning electron microscope. SEM-EDX: Scanning electron microscope–energy dispersive X-ray instrument used in elemental analysis. It is also referred to as EDS or EDAX analysis. Siberian Tunguska event 1908: About 2150 km2 of taiga were devastated and 80 million trees were overthrown near the Stony Tunguska River in Siberia on June 30th, 1908. An explosion about 8 km above the ground was the proximate cause, though whether this was due to a comet, an asteroid or something else, is still debated. Silane: Organic compound with a –C–Si– group. Siloxane: Organic compound with a –C–O–Si– group. Sipping technique: A method of extracting porewater from soil or sediment by allowing the water to ﬂow directly into a sample container. Slope bog: A bog occurring in areas of high rainfall on appreciably sloping land surfaces. Snipes: Wadding bird with a straight long beak (genus Gallinago). Soligenous: Refers to mires (fens) that are formed where sloping terrain provides a continuous supply of (ﬂowing) water containing local mineral matter. The localized ﬂow of water supplies more nutrients than are found in plain areas (topogenous mires). Where the water is particularly nutrient-rich, the vegetation is dominated by small-sedge communities and brown moss species. Solum: The upper part of a soil proﬁle, the A- and B-horizons, where the soil forming processes principally occur. Sorption: General term for the interaction (binding or association) of a solute ion or molecule with a solid. South American Plate: One of the major, continental tectonic plates, comprising most of the South American continent.

570

Glossary

Spodic horizon (dark illuvial horizon) Bs: Dark colored subsurface horizon, characteristic of podzols (WRB soil classiﬁcation), called spodosols in the American System of Soil Taxonomy. Contains illuviated amorphous substances composed of organic matter and aluminum, with or without iron. Sponge spicules: Microscopic needle-like and multi-rayed skeletal elements secreted by sponge cells. Siliceous sponge spicules consist of amorphous biogenic silica (opal). Stemwood: Part of a tree stem with the branches removed. Sterol: A solid cyclic alcohol found in both plants and animals. Stigonematales: Order of cyanobacteria producing heterocysts. Subarctic: The area where open-canopied coniferous woodlands are the dominant vegetation form, with or without outliers of treeless tundra. Suberin: An aliphatic macromolecule occurring in cork structures (such as plant roots). Swamp: This term has been used as synonymous with peatland. In this book it is used speciﬁcally for a minerotrophic peatland with more than 25% tree-cover. Syringol: Lignin fragments with a 2,6-dimethoxyphenol basic structure. Taiga: Northern or boreal forest. Tar foam: Tar is formed during the dry distillation of wood during forest wild ﬁres. The tar foam is formed by tar pressed out from the burning wood by heat. The bubbles in the tar foam are ﬁlled by combustion gases. Once solidiﬁed the foam is a relatively hard but very light product reminiscent of polyurethane foam. Tar foam fragments are easily transported in air. Tardigrads: Aquatic micro-metazoans of debated phylogenetic position; also known as water bears. Temperate forest: Forest occurring in a temperate climate, and characterized by moderate to high levels of precipitation, mild winters and warm summers. Tension inﬁltrometer: A device for controlling water-ﬂux at different pore water tensions. This allows comparisons of water ﬂow to be made for a column of soil where inﬁltration may occur for all pore sizes, or be restricted to the smallest pores. Thus, by comparison, the proportion of water ﬂowing through macropores or the soil matrix may be determined. An inﬁltrometer is most commonly used to measure the hydraulic conductivity. Tephra: General term for fragments of volcanic rock regardless of size that are blasted into the air by explosions or carried upward by hot gases in eruption columns or lava fountains. Tephra includes large dense blocks and bombs (particles larger than 64 mm), and small light rock debris such as lapilli (2–64 mm particle size), scoria, pumice, reticulite, and ash (o2 mm particle size). Tephrostratigraphy: Use of volcanic material deposited in sediments and soils to correlate and date them. Terpenoids: Organic compounds with multiple C5-isoprene units. Terrestrialization: The process of partial to total inﬁlling of a pond or shallow lake through the accumulation of peat. Open waters become peatlands (fens and bogs) (compare to paludiﬁcation). Tertiary: Period of the geological time scale, between approximately 65 and 1.8 million years BP. Testate amoebae: Polyphyletic group of protozoa that produce a shell (or test).

Glossary

571

Textinite: A dark gray, commonly with a brownish tinge and abundant orange to red–brown internal reﬂections maceral, with yellow–brown to red–brown UV ﬂuorescence. It derives from cell walls of both herbaceous and arborescent plants. It can be distinguished from ulminate because it shows separate cell walls, whereas ulminite consists of clearly recognizable but compressed and geliﬁed cell walls. Till: Nonsorted sediment deposited directly from a glacier. Time domain reﬂectometry: The analysis of a conductor (peat soil in the present case) by sending a pulsed signal into the conductor, and then examining the reﬂection of that pulse. TOF-ICP-MS: Electro spray time of ﬂight, inductively coupled plasma mass spectrometry. Topographic index: A measure of upslope area draining to a given point and expressed in units of volume per unit contour length divided by the slope angle. Total ion current (TIC): Total intensity of a mass spectrum or of a number of selected peaks. TOX (total organic halide concentration): Total concentration of organically bound ﬂuorine, chlorine, bromine and iodine. Transition mires: Mires with peat-forming plant communities whose characteristics are intermediate between those of minerogenic and ombrogenic types, both in a spatial and biogeochemical succession. Trophic status: Nutrient status; availability of nutrients in an ecosystem (oligotrophic to eutropic). Tundra: A level to undulating, treeless, frozen plain, characteristic of arctic regions supporting grasses and stunted shrubs. Ulminite: It is a maceral that denotes geliﬁed plant material in which cell structures can still be recognized (compare to textinite). Undrained shear strength: Strength, or resistance to failure, of material in response to shear stresses (that is, adjacent forces acting in opposite directions within materials or hillslopes) when drainage of water contained in pore spaces within the material is impeded during (rapid) shearing. This has the effect of increasing pore water pressure and therefore reduced the strength of the material. Unfrozen peatlands: Peatlands not affected by permafrost. Utermo¨hl method: Method of counting phytoplankton using plankton-settling chambers. Variable source area concept: The idea that different parts of a catchment may contribute to overland ﬂow and that these source areas will change over time. Varve: A general term for a feature in sediments that represents deposition during the course of one year. A varve contains deﬁnable seasonal layers of varying color and particle size. Vitrinite: It is a maceral with no or poor UV ﬂuorescence, moderate reﬂectance. It is the most common organic compound of most humic coals; it derives from remains of cell walls, woody tissues, leaves and roots. von Post method for determining humiﬁcation: A qualitative ﬁeld test to assess the state of peat decomposition that involves squeezing a handful of peat through the ﬁngers, recording what happens and matching this outcome to the corresponding

572

Glossary

description in a standard reference table. The scale runs from H1 (no decomposition) to H10 (complete decomposition). Water content of peat: It is determined as the ‘mass of water per unit mass of ovendry solids’, normally multiplied by 100 and reported as a percentage. For example, a 220 g sample of wet peat might be found to comprise 200 g of water and 20 g of solid matter. The water content is therefore 10 g of water for each gram of dry mass, i.e. ( 100) 1000%. Water contents of low-humiﬁcation fen and bog peats commonly exceed 1000%. Water tracks: Small ponded areas of a peatland where water ﬂows preferentially at and very near the surface. Wisconsinan glaciation: The last glacial stage of the Pleistocene epoch in North America, which ended around 10,000 BC. It is equivalent to the Weichsel (in Scandinavia), Devensian (in the British Isles), Midlandian (in Ireland), and Wu¨rm glaciation (in the Alps). WRB: Acronym for world resource base, and referring to the WRB system of soil classiﬁcation, a system developed from the earlier FAO soil classiﬁcation. It was designed to serve as a basis for the FAO map of the world soils. Wu¨stite: Mineral form of ferrous oxide (FeO) found with meteorites and native iron. Xenobiotics: Chemical compounds of anthropogenic origin, which would not otherwise occur in nature, for example DDT. XRD: X-ray diffraction. XRF: X-ray ﬂuorescence spectrometry. Younger Dryas: A very short, cool-climate stage during most of the last two millennia of the Late Glacial in Scandinavia and the Northern Hemisphere, characterized by relative abundance of Dryas sp. pollen in peat or lacustrine sediments. Its simultaneous occurrence in the Southern Hemisphere has been a matter of debate during almost two decades. Zygomycota: Phylum of fungi producing zygospores. Synonym: Zygomycetes.

Subject index

Localities in italics aapa, 19, 27, 28, 41 ablation, 247, 252 acetic acid, 221, 230, 236 acid bath technique, 201 acid lavas, 131 acid rock, 88, 89 acid sulphate soils, 146 acidophilus plant communities, 281 acrotelm–catotelm model, 8, 320, 330, 331, 336, 337 actinomycetes, 309 Acton Park II metalwork production period, 420 Adige River, Italy, 533 Adriatic Sea, 531, 533 Aeduan civilization, 416 aelotropic, 391 aerenchyma (of vascular plants), 276 aerenchymatous, 330 aerenchymous tissues, 276 aerenchymous transport, 276 aerobic decay, 6, 9, 176, 218, 219, 276, 469 aerosol, 7, 71, 101, 149, 152, 157, 188, 203, 247, 248, 260, 263, 265, 436, 441, 451, 454, 457, 483, 488, 492, 494 afforestation, 501, 503, 516, 523 Africa, 145, 146, 246, 257 age of peat deposits, 35 agriculture, 10, 148, 192, 244, 247, 248, 250, 309, 412, 413, 415, 490, 502, 516 Aguilera Volcano, South America, 135 Alaska, 35 Albany Forks, Canada, 350 Albany River, Canada, 348, 349, 353 alcohols, 221, 229–31 algae, 230, 280, 291, 292, 300, 304, 514 brown algae, 298 green algae, 298, 299 microalgae, 8, 294, 298–9, 302, 307–9, 311

aliphatics, 6, 218, 221, 229, 230, 233, 234, 237 alkanes, 6, 221, 230, 234, 237 n-alkanes, 217, 230, 231, 233, 235, 237 ‘free’ n-alkanes, 230 alkanoic, 221 alkenes, 6, 221, 230–2, 234, 237 allochthonous, 252, 256 allogenic material inﬂux, 157 Alps, 244, 277, 278 aluminosilicate, 175, 181, 182, 207–9, 211, 488 aluminum, 101, 207, 242, 440, 467, 513, 538 ammonia (NH3), 271 ammoniﬁcation, 512 ammonifying bacteria, 512 ammonium (NH4+), 271, 276, 512 amoebae naked amoebae, 299, 308 testate amoebae, 299 Anacardiaceae, 154 anaerobic decay, 6, 9, 218, 219, 276 Andean Glacial Valley, South America, 342 andesite, 210 Andorra Valley, Tierra del Fuego, 139 anglesite (PbSO4), 480 Anglian Fens, UK, 502 annite, 206 Annonaceae, 154 Antarctic, 112, 142, 218, 242, 244, 255, 265 Antarctic Peninsula, 129 Antarctica, 112, 116, 118, 119, 132, 322 anthropogenic emission, 455, 467, 475 apatite, 480 aqua regia, 244 Arachnida, 304 arboreal pollen, 64, 134, 413, 415, 416 Arcellinida, 302 Archea, 294 archeological stones (in bog), 210 archeomagnetically dated, 418

574 Argentina, 4, 241, 242, 253 aromatics, 218, 221, 229, 234 Arrhenius law, 540 Arvidsjaur, Sweden, 433, 443, 444 Ascomycota, 296 ash content, 27, 67, 105, 151, 152, 200, 201, 203, 204, 231, 241, 244, 247, 249, 250, 252, 264, 389, 391 ashing technique, 201 Asia, 4, 145–50, 154, 155, 159 Askja, Iceland, 252 ASTER, 534 Asteroidaceae, 415 Astrakhan State Reserve, 41 Atlantic, 21, 45, 88, 92, 105, 137, 256, 257, 325 atmospheric deposition, 188, 192, 202, 244, 433, 437, 458, 466 atmospheric dust, 188, 192, 198, 211, 244, 423–5, 479 Australia, 67, 377, 385, 388, 491, 492 authigenic, 181, 247, 250, 259–61 autoacidiﬁcation, 99 autochthonous, 256 autoﬂuorescence, 292 autogenic, 157 autotrophic micro-organisms, 293, 304, 309 protists, 291, 298 Auvergne, France, 304

Bacho swamp, Thailand, 160 bacteria cyanobacteria (with diameter >2 mm), 292 picocyanobacteria (with diameter o2 mm), 292 bacterivorous, 304, 308 Bahı´a Aguirre, Tierra del Fuego, 134 Bahı´a Inu´til, Tierra del Fuego, 141 Bahı´a Lapataia, Tierra del Fuego, 134, 136 Bakong River, Sarawak, 161 Baram River, Sarawak, 160, 161 barite (Ba4SO4), 409 Barney volcano, South America, 135 Barreiras do Lago, Serras Septentrionais, Galicia, 90 Bartington MS2E magnetic susceptibility probe, 423 basalt, 205

Subject index base cations, 182, 350, 513 baseﬂow, 324–6, 342 Basidiomycota, 296 basin or valley peatlands, 150 Batang Hari River, Sumatra, 110 batholith, 131 Bdelloidea, 302 Beagle Channel, Tierra del Fuego, Argentina – Chile, 131, 133, 138, 141 Beagle Glacier, Argentina, 132, 134 Beaker grave, southeast Scotland, 420 Beaufort Sea coast, Northwest Territories, Canada, 34 beidellite, 180 Belgium, 192, 298 Beltrami Arm of Lake Agassiz, Minnesota, 352 Bengkalis Island, Sumatra, 161 Berbak, Sumatra, 161 biodiversity, 88, 147, 152, 155, 166, 272, 517 biogeophysical, 166 bioindicators, 309 biomonitoring, 309–10, 465 biotite, 209, 210 biovolume, 292, 293 Black Forest, Germany, 209, 210 Blackbear Island, Canada, 350 blackwater, 155 blanket bog, 8, 88, 90, 92, 105, 188, 190, 198, 236, 242, 249, 320, 321, 324, 358, 383, 387, 394, 396, 398, 403 bloom smithing, 421 bloomery, 424, 425 bog burst, 383, 386, 388, 392, 394, 515 bogﬂow, 383–90, 392–4, 398, 402, 403 bog iron ore, 422 bog pools, 288, 298, 341 bog slide, 383, 384, 390, 392, 39, 397, 398, 402 bog water chemistry, 199 Bohemia, 262 Bolton Fell Moss, UK, 230, 233 Borneo, 62, 148, 154, 155, 159, 162 boro (ﬁbrist; hemist; saprist), 101 Borralleiras de Cal Grande, Spain, 486, 487 botanical composition, 26, 198–200, 238, 389 Bottnaryd, Sweden, 433, 441 boundary horizon, 94 Bowen ratio, 323, 324 Bran˜a de Lamelas, Serra dos Ancares, Galicia, 102

Subject index British Columbia, Canada, 26, 62 British Isles, 10, 90, 96, 246, 378, 386, 403, 409, 411, 412, 417, 420, 426 Broken Hill Mine, Australia, 483, 492 Bronze Age, 93, 412–4, 418–20, 422 brown algae, 298 Brunei, 147, 148 Bryales, 122, 249 Bryn y Castell, southern Snowdonia, northwest Wales, 413 bryophytes, 88, 152, 154, 272 bulk density (BD), 99, 105, 151, 152, 340, 349, 391, 508, 511, 530 burn-beating, 257 Burseraceae, 154 Byeloruss, 19 Byelorussian peat corer, 242 Bytownite, 210

C-horizon, 438, 440, 441 Cabo San Pablo, Tierra del Fuego, 138, 139 Caithness, Scotland, 340, 503 calcite, 5, 176, 181, 182, 187, 193, 210, 211, 260, 261 calcium, 156, 157, 246, 249, 260, 302, 320, 349, 357, 358, 360 Caleta Ro´balo bog, Tierra del Fuego, Argentina, 141 Cambodia, 151 Cambrian, 409 Campbell Island, south of New Zealand, 198 Canadian Craton, 197 Canadian Shield, North America, 352, 355, 356, 493 Capitulum (top 1 cm of Sphagnum mosses), 199, 291 Carbajal, Ushuia, Tierra del Fuego, Argentina, 139 carbon alkyl C, 218, 220, 237 13 C-CPMAS-NMR, 218, 219, 237 O-alkyl C, 218, 220 carbon dioxide (CO2), 159, 275, 309, 323, 529, 532 carbon stocks in northern peatlands, 43 carbon storage, 2, 159, 162, 166, 342 carbon/nitrogen ratios, 461

575 carbonate, 53, 101, 181, 182, 184, 187, 193, 203, 210, 243, 260, 261, 355 carbonatic rocks, 461 Carboniferous, 67, 164 carboxyl groups, 199 carboxylic functional groups, 181, 185 Carex, 88, 95, 113–6, 118, 119, 122, 125, 126, 18, 217, 249, 37 Carey Islands, Greenland, 198 Caribbean, 145 Carpathian Mountains, Eastern Europe, 299 Caspian Sea, 41 catalyst, 209, 494 cation exchange capacity (CEC), 101, 105, 176 Cd, 192, 198, 422 cellulolytic exoenzymes, 305 Cenozoic, 131, 132 Central America, 145, 146 cereal-type pollen, 413, 415, 416 cerussite (PbCO3), 480 chalcophile element, 479 chalcopyrite, 419, 422, 424 Charterhouse on Mendip, England, 420 Chenopodiaceae, 415 Chile, 130–2, 135, 139, 141, 450, 452, 453, 456, 460, 469, 483 China, 246, 249 chlor-alkali plants, 465 chloride, 355, 451, 453, 460 chlorite, 210 chlorophyll-a, 299, 309 chondrites, 253, 263 chronostratigraphic, 132 Chroococcales, 294 chrysophyceans, 310 ciliates, 299, 309 Circum-Polar Current, 129, 309 Cladocera (water ﬂeas), 304 cladocerans, 310 climate climatic moisture balance, 145 climatic variations, 265, 302 Clusiaceae, 154 coastal peatlands, 45, 114, 150, 162 coefﬁcient of variation (CV), 488 Comet Lode, Ystwyth Valley, Wales, UK, 422, 423 complexants, 175

576 composition and properties, 96–101, 380 Conaghra, Ireland, 393 condensation nuclei, 241 conservative elements, 209, 449, 467, 484, 488 Copa Hill, Cwmystwyth, Wales, UK, 413, 414 Copepoda, 304 copepods, 310 harpacticoid copepods, 304 copper (Cu), 261, 262, 409, 419, 420, 422, 491 copper mining, 412–4 copperworks, 416 Cornwall, UK, 426 cosmic dust, 202, 247, 252 cosmic materials, 248, 265 cosmic winter, 250 County Antrim, Ireland, 383 County Cavan, Ireland, 384, 390, 392–4 County Clare, Ireland, 385 County Cork, Ireland, 413 County Fermanagh, Ireland, 394 County Galway, Ireland, 377, 385 County Kerry, Ireland, 377, 386 County Laois, Ireland, 389, 402 County Limerick, Ireland, 383 County Mayo, Ireland, 377, 384, 387–9, 392 County Sligo, Ireland, 386, 388 County Tipperary, Ireland, 383 CP-Sil 51b, 220 Cretaceous, 112, 131, 164 Crustacea, 304 crustaceans, 308 Crymlyn Bog, UK, 322 cryosol, 18, 61 Cuilcagh Mountain, Ireland, 384, 385, 387, 390, 392–7, 400, 401, 403 cultural landscape, 10, 412, 425 Cumbria, England, 262, 521 Curie-point pyrolyser, 220 current ecological functions, 523 cushioned meadow, 119, 121 cutinite, 72 Cwmystwyth, Wales, UK, 413, 414 cyanobacteria (>2 mm), 292 Cyperaceas, 115 Cyperaceae, 121, 122, 124, 126, 236 cysts, 259 Czech Republic, 274, 492

Subject index Dalat sago plantation, 164 Dalat, Sarawak, 164 Danau Sentarum, Kalimantan, Borneo, 159 Danby Low Moor, England, 392 Dartmoor, UK, 409 Darwin Cordillera, South America, 132, 133 dendrochronology, 139–140 Denmark, 198, 274, 472 Derbyshire, England, 262 Derrybrien, Ireland, 377, 385, 402 Des Moines, Minnesota, 348, 355 desiccation, 386, 391, 402, 510, 522 desmids, 298 detrital, 7, 247, 249, 261 detritivorous, 308 Deuteromycota, 296 DGPS, 533, 536 diagenesis, 11, 455, 46, 473 diagenetic, 192, 218, 247, 259, 456, 469, 470 diaspores, 517 diatom frustules, 204, 259 dimethylmercury, 467 DIN, 273 diplotelmic, 330, 331, 336 Dipterocarpaceae, 154 dipterocarp, 154, 156 discharge, 8, 9, 132, 200, 324–6, 337, 350, 359–66, 370, 434, 506, 518, 531 dissolved organic carbon (DOC), 10, 191, 459, 460 dissolved organic matter (DOM), 191, 308, 417, 450 ditch blocking/drain blocking, 517, 519, 521, 522 DNA, 293 Dnepr River, 41 dolomite, 187 dome bogs, 29 DON, 273, 274, 513 Don River, 41 Dooncarton Mountain, Ireland, 384, 387, 388 dopplerite, 388 drumlin, 133 dry fall, 156 Dumme Mosse, Sweden, 433, 439, 469, 470, 474 Dundrum, Ireland, 383 Dunmor, Ireland, 385 dustfalls, 262

Subject index East Anglia, UK, 322 East Asia, 145, 250 East Cuilcagh, Ireland, 395, 402 Eastern Europe, 349, 359, 502 ecohydrology, 342 ecotone area, 138 EDX, 242, 253, 254 effective cation exchange capacity (eCEC), 101 El Nin˜o southern oscillation (ENSO), 146 electrical conductivity, 152, 199 electron pumps, 182–4 eluvial horizon (E-horizon), 440 endo-symbionts, 302 England, 204, 27, 262, 293, 325, 337, 367, 384, 392, 420, 491, 503, 505, 508, 514, 516 Enniskillen, Ireland, 521 enrichment factor (EF), 11, 103, 479, 483 environmental changes, 3, 9, 78, 102, 141, 142, 377, 402, 431 enzymes, 271, 277, 450, 452, 462 epiﬂuorescence, 292 epiﬂuorescence microscopy, 291, 292, 294 Epipaleolithic Age, 93 epiphytic algae, 280 ergosterol extraction method (for fungi), 293 Ericaceae, 64, 217, 235–7, 323 erosion, 30, 89, 93, 94, 103, 122, 192, 202, 203, 208, 332, 338, 340, 361, 490, 494, 501, 514, 516, 521, 523 esters, 221, 229 Estonia, 96, 203, 252, 254 E´tang de la Grue`re, Switzerland, 200, 205, 437, 485, 486 euglenoids, 298, 299 euglenes, 309 Euglenophyceae, 298 Euglyphida, 302 eukaryotes, 298 Euphorbiaceae, 154 Eurasia, 17–9, 43–5, 47, 349, 358 Europe East Europe, 28, 246 European Russia, 35, 45, 46 European Union (EU), 19, 89, 494 europium, 264 eutrophic, 26 eutrophication, 7, 272, 275, 278, 280–2, 529 evapotranspiration, 45, 67, 272, 320, 323–4, 342, 458, 460, 503, 539

577 Everglades, Florida, 95, 540 exchangeable cation content, 511 Exmoor, UK, 412 exoenzymes, 302 extensometer, 538, 542, 547, 548 Extra-Andean Patagonia, 132

Fagnano Glacier, Tierra del Fuego, Argentina, 132 Fairloch Moss, Ireland, 383, 392 Falkland Islands, 198, 379, 451 Fallahogy Bog, Northern Ireland, 243, 250, 252, 256, 264 Faroe Islands, 190 fatty acids, 6, 218, 221, 230, 231, 233, 234, 293, 294 fecal pellets, 433 Fe-hydroxides dissolution, 209 Fe-hydroxy phases, 180 feldspar, 5, 181, 187, 203, 206, 207, 209–11, 250, 480 fen cluster, 294 Fennoscandia, 356, 436, 439 ferrihydrite, 181, 189 fertilization, 7, 274–9, 309, 310, 514, 523 fertilizers, 7, 272, 296, 309, 513 ﬁbric, 67, 96, 101 ﬁbrists, 101 ﬁnite elements, 544 Finland, 19, 22–6, 90, 95, 96, 192, 257, 502, 503, 514, 515 Finnish fen, 294 ﬁresetting, 422 FISH (method for ﬂuorescence in situ hybridization), 293 ﬂagellates, 291, 292, 299 ﬂarks, 29 ﬂat bog, 21, 29 ﬂatworms, 304 Platyhelminthes, 304 ﬂoating index tree-ring, 140 ﬂood peak, 505–7, 509, 523 ﬂooding, 45, 122, 125, 149, 152, 319, 321, 501, 502, 529, 531 Florida, USA, 540 Florida Everglades, 95, 540 ﬂuorine, 452 ﬂuorochrome, 292, 294 ﬂuospar (CaF2), 409

578

Subject index

ﬂy ash, 204, 248, 483, 492 folic materials, 18 Forest of Dean, UK, 412 Forest-Tundra, 17, 44–6 formaldehyde, 292 Fort Simpson area, Northwest Territories, Canada, 25 France, 90, 96, 298, 304, 307, 413, 422, 440, 491 Franches Montagnes, Swiss Jura, 199 freeze-thaw, 87, 210 French Bretagne, 90 freshet, 363 frozen peatlands, 2, 19, 34, 43–5 frustules, 204, 259 Fuegian forest, 132 Fuegian, Tierra del Fuego, Argentina Fuegian Andes, 131, 134, 138 Fuegian Archipelago, 129, 135 fungal hyphae, 292, 308 fungi, 296 endolithic fungi, 260 fungivorous, 304 fusiforms, 257, 258

Gobi, China, 247 goethite, 181, 182, 191 Gola di Lago, Switzerland, 485, 486 gold, 262, 409 Gorizia Channel, Zennare Basin, Italy, 536 graminoids, 275 granite, 89, 200, 209 Great Bear Lake, Canada, 355 Great Patagonian Glaciation (GPG), 132 Great Slave Lake, Canada, 355 Greece, 262 greenhouse effect, 288 Greenland, 17, 198, 421, 436, 437, 440, 472 Greenland ice, 421, 436, 437, 440 greenschists, 131 grenzhorizont, 94 ground penetrating radar (GPR), 339, 386 groundwater groundwater-evapotranspiration models, 324 Penman–Montieth model, 324 Priestly–Taylor model, 324 guaiacols, 233 Gulf of Mexico, 356

galena (PbS), 262, 419, 480 Galician, 85, 88, 92–4, 96, 102, 105 Ga¨llseredsmossen Bog, Sweden, 248, 251, 255, 259, 263 gas analyzer, 539 gas chromatograph, 217, 220 gas ebullition, 332 gas ﬂuxes, 6, 275–7 gastrotrichs, 304 geoarcheological tool, 417 geochemical archives, 102, 412, 479 geogenous peatlands, 145 geological substrate, 199–200 Germany, 198, 200, 202, 379, 415, 491 gibbsite, 181, 182, 207 GIS, 113, 518 glaciation, 35, 88, 114, 132, 134, 449 glassmaking, 413 Glenamoy, Ireland, 504, 505, 507, 508 Glencastle Hill, Ireland, 389 gleys, 502 global warming, 2, 46, 67, 73, 78, 80, 272, 275, 309, 370 glutaraldehyde, 292

H2O2, 201 Haber–Bosch process, 272 Hadden Catchment, England, 514 Hagenmoos, Switzerland, 485 halogen halogenated organic compounds (HOC), 449 halomethanes, 449, 455 organohalogens, 11, 449, 450, 452, 453, 455, 458–60, 462, 463 organoiodine, 450, 453, 454, 458 Hamsterley, UK, 420 Harberton, Tierra del Fuego, Argentina Estancia Harberton, 136 Harberton Bog, 243, 253, 255, 256 Puerto Harberton, 141 hard pan, podzol, 151 Harpacticoid, 304 Hart Hope, England, 384 Harz Mountains, Germany, 262, 491 Ha¨sthult, Sweden, 433, 441 hazel, 413 heavy metals, 102, 187, 191, 261, 449, 479, 484

Subject index Hekla, Iceland, 247, 252 helophytic, 330 hemic, 67, 96 heterocysts, 294 heterotrophic bacteria, 294, 296, 304, 308–11 heterotrophic ﬂagellates, 291, 292, 299 heterotrophic production, 287 heterotrophic protists, 288, 299–302, 308–11 hierarchic classiﬁcation system, 88, 89 High Arctic, 33, 35, 39, 198, 199 high moor, 26, 27, 218 high, interior or watershed peatlands, 150, 151 high-center lowland polygon, 33, 34 histosol, 5, 18, 101, 193, 529, 531, 542, 546 Holme Fen Post, East Anglia, UK, 322 Hongyuan Plateau, China, 249 Hudson Strait, Canada, 53, 355 Hudson volcano, South America, 135 humic substances, 209, 450, 466 humiﬁcation, 9, 99, 125, 139, 151, 217, 229, 231, 389, 390, 402, 403, 450, 455, 459, 473 huminite, 67 hummocks, 113, 118, 119, 124, 154, 198, 277, 298, 319, 341, 358 hyaline cells, 204, 291 hybridization, 293 hydrochemical, 329, 501, 517, 523 hydrogenotrophs, 294 hydrograph, 324–6, 333, 337, 504, 518, 519 hydrologic cycle, 539 hypsithermal interval, 33, 65

Iberian Peninsula, Spain, 85, 95, 97, 103, 490 ice cap, 133, 242, 255, 262 ice cores, 10, 265, 409, 417, 421, 466, 474 ignifacts, 248, 255–7, 265 IKONOS, 534 illuvial horizon, 440 impact ejecta, 252 incongruent weathering, 207 indole, 230 inductively coupled plasma mass spectrometer (ICP-MS), 244

579 Industrial Revolution, 10, 192, 271, 417, 441, 465, 491 inertinite, 67 inﬁltration inﬁltration-excess overland ﬂow, 326, 328, 329, 336 inorganic compounds, 191, 200, 203 inosilicate, 205 interferometry, 537 inter-tropical convergence zone, 147 iodine, 10, 246, 249, 449–55, 458–60 iodine deﬁciency disorders, 449 IPTA, 533, 537 Ireland Northern Ireland (NI), 264, 377, 502 Republic of Ireland, 377 Irish blanket bog, 242, 394 Iron Age, 3, 12, 93, 103, 412, 413, 415, 416, 421, 422 iron meteorites, 252 Isla de los Estados, Tierra del Fuego, Argentina, 131 Isla Gable, Tierra del Fuego, Argentina, 134 Isla Grande de Tierra del Fuego, Argentina, 129 isotopic signature of N, 274 Italian Alps, 277 Italy, 13, 90, 274, 298, 304, 531

James Lobe, Minnesota, USA, 348, 355 Java, 151 Johnsbach Valley, Steiermark, Austria, 416 Joldelund, Nordfriesland, Germany, 415 Jura Mountains, Switzerland, 298, 489

Kaali Impact Crater, Estonia, 252 Kalimantan, Indonesia, 147–59, 161–3, 200 Kalven, Sweden, 433–6 kaolinite, 81, 181, 182, 203, 207 Kapanihane Bog, Ireland, 383 Kapuas River, Kalimantan, 161 Kempen, Belgium, 192 kerapah, 151 Kilmaleady, Ireland, 385 King’s County, Ireland, 385 Knocknageeha Bog, Ireland, 377 Kolhu¨ttenmoor, Germany, 200, 202, 209 Koltja¨rn, Sweden, 433, 434, 436

580 La Correntina, Tierra del Fuego, Argentina, 126 La Misio´n Bog, Rı´o Grande, Tierra del Fuego, Argentina, 136 LA-ICP-MS, 211 Labrador, Canada, 53, 352 Lagoa de Lucenza, Serra do Caurel, Galicia, 91 Laguna de Tagua-Tagua, Tierra del Fuego, Argentina, 141 Lake Agassiz, North America, 347, 349, 363, 367, 522 Lake Fagnano, Tierra del Fuego, Argentina, 111, 131, 133, 136, 139 Lake McConnell, Canada, 355 Lake Sentarum, Kalimantan, Borneo, 151 LANDSAT, 534 lanthanides, 263, 264 Lapataia, Tierra del Fuego, Argentina, 134, 136, 141 Las Lengas, Tierra del Fuego, Argentina, 127 Last Glacial Maximum (LGM), 3, 91, 111, 132, 165, 241, 246 Late Glacial, 35, 111, 126, 134, 135, 137, 141, 142 Lauraceae, 154 Laurentide Ice Sheet, 53, 348, 355 Lautaro volcano, South America, 135 Lawas River, Sarawak, 161 lead enrichment factors, 11, 479 history, 489 isotopes, 421, 437, 440, 441, 479, 480, 483 mineralogy, 480 mining, 262, 420 pollution, 12, 418, 426, 436 radiogenic, 480, 483 sources, 12, 438, 483, 492 three-isotope plot, 491 used for tracing, 193, 483 Leguminosae, 154 Lerwick, Scotland, 377 ligand, 207, 208, 210 light detection and ranging (LiDAR), 518 lignin moieties, 233, 235–7 lime burning, 413, 416 limestone, 181 lipid stratigraphy, 218 liptinite, 67 lithogenic, 104, 437, 484–6, 488, 489

Subject index litterfall, 156 Little Ice Age, 104, 142, 490 Llwyn Du, Wales, UK, 424, 425 loess, 246, 249, 250, 263, 265 long-term (apparent) rates of carbon accumulation (LORCA), 162 low ﬂows, 507 low-center lowland polygons, 31 low-grade metamorphism, 131 lowland tropical peatlands, 145–72 low-moor, 27 Lune, England, 514 Lyngmossen Bog, Sweden, 250, 251, 253

Mackenzie River Valley, Northwest Territories, Canada, 24 Macquarie Island, Australia, 384, 388 macromolecules, 450 macropore, 336, 33 macropore ﬂow, 329, 330, 509 macroporosity, 336 Maesnant Catchment, UK, 338, 339, 380, 509 Magellan Glacier, South America, 132 Magellan Strait, South America, 135, 141 Magellan transcurrent fault, 130 Magellanic Moorlands, Chile, 450, 455, 460, 469, 474 maﬁc minerals, 480 magnesium, 157, 246, 249, 320 Magnana Channel, Zennare Basin, Italy, 535, 536 magnetite (Fe3O4), 189, 204, 206, 250, 254, 260 Malay Peninsula, 155 Malaysia, 147, 148, 150–2, 157, 160, 163, 165, 379, 385, 530 Malaysia Sabah, Sarawak and Southeast Thailand, 147 Malthusian, 183 Manchester, UK, 204, 422 manganese, 157, 512 mangrove, 150 Manitoba, Canada, 356 marine terrace, 136 Marudi, Sarawak, 160 mass spectrometry, 217, 220 Massif Central, France, 413 matrix ﬂow, 323, 386 Mauntschas, Switzerland, 485

Subject index mean residence time (MRT), 442 medieval, 103, 412, 416, 417, 419, 422, 424, 426, 441, 491 Medieval Warm Period (Europe), 104, 139 mediﬁbrists, 102 medihemists, 102 medisaprists, 102 Mediterranean (Sea), 490 Mendips, Shropshire, UK, 409 mercury mercury accumulation rate, 471–6 mercury concentration, 11, 467–71, 473–5 mercury toxicity, 465 mesotrophic, 26 Mesozoic, 53, 78, 131 metal metal enrichment, 415, 417, 422, 486 metal production, 415, 434 metalworking, 10, 412, 415–7, 421, 425, 426 metazoa, 7, 287, 288, 292, 302, 304, 308, 311 meteorites meteorite ablation spherules, 247 methane (CH4), 2, 275, 276, 294, 305, 311, 323, 366–8, 529 methanogenic bacteria, 276, 288, 293 methanomicrobiales, 294 methanotrophic, 276, 288, 294 methyl esters, 221, 229 methyl ketones (methylketone), 6, 221, 230, 231, 233, 234, 237 methylmercury (MeHg), 465, 467–9 microalgae, 8, 294, 298–9, 302, 307–9, 311 microbial microbial biomass, 179, 293, 296, 304, 311 microbial ecology, 7, 288 microbial loop, 307–8, 311 microbial respiration, 305, 440 micro-metazoa, 7, 287, 288, 291, 292, 302–4, 308, 311 micrometeorite, 247, 252, 253 micro-regmaglyptic structure, 253 micro-tektites, 252 minerogenic, 88, 91, 95, 96, 99, 101, 105, 191, 434, 451, 462, 467, 470 minerotrophy, 247, 263, 264 mining oreﬁeld, 420, 412 prehistoric mining, 412, 415

581 Minoan Santorini eruption, 247, 251 Miocene, 132, 164 Mire mire macrotopes, 88 mountain mires, 85–109 ombrogenic mires, 88, 89, 95, 96, 105 minerogenic, 88, 95, 96, 99, 101, 105, 191, 462, 470 mixotrophic species, 302 Moat Moraines, Tierra del Fuego, Argentina, 134 modeling peat instability, 399–401 moieties, 233–7 moisture ratio, 539 mollusks, 308 Monogononta species, 302 monophyletic, 299 monsoon, 147, 247, 265 Monte Gallinero, Ushuaia, Tierra del Fuego, Argentina, 139 montmorillonite, 180 Moor House, north Pennines, England, 504, 505 moraine, 53, 91, 114, 115, 122, 125, 126, 133, 134, 352 Morvan, Massif Central, France, 413 Mount Gabriel, County Cork, Ireland, 413, 415 Mt Beuvray, Morvan, Massif Central, France, 413, 422 Muka, Sarawak, 164 Myristicaceae, 154 Myrtaceae, 154

NEE, 277, 278 Nee Soon, Singapore, 160 nematodes, 304 neoglacial, 134, 138, 142 neoglaciation, 104 Neolithic, 93, 192, 417 Nepenthaceae, 154 net ecosystem exchange (NEE), 277, 278 Netherlands, 90, 274, 502, 530 New Zealand, 198, 209, 257, 502 Newfoundland, Canada, 62, 321, 324 nickel, 253 Nipa palm swamps, 150 nitrate (NO3-), 271, 276, 512, 529 nitric acid (HNO3), 271

582 nitrifying bacteria, 512 nitrogen oxide (NOx), 271 nitrous oxide (N2O), 271 nomadism, 418 Nong Thale Song Hong, Thailand, 160 non-heterocystic, 294 non-hydrolyzable aliphatic biopolymers, 230, 234 non-pollen palynomorphs, 475 Nordfriesland, Germany, 415 Norra Kvill, Sweden, 433, 441 North Atlantic, 137, 325 North Pennines, England, 384, 392, 409, 416, 418, 420, 421, 504, 509 North Sea, 246 North Yorkshire Moors, UK, 392 northern and mid latitudes soil database (NMLSD), 18 northern circumpolar area, 17, 18 northern circumpolar peatlands, 18, 47 Northern Hemisphere, 199, 241, 321, 440, 466, 468, 472 Northwest Territories, Canada, 23–5, 34 Norway, 6, 244, 274, 443 Norwegian, 257 Nostocales, 294 Nothofagus Forest, Tierra del Fuego, Argentina, 4, 139 NSS chamber, 538, 546 nucleic acids, 271 Nuorittajoki River, Finland, 515 nutrient cycling, 147, 157, 305, 311 nutrient limitation, 271, 273, 294, 307

oak, 413 Older Dryas, 137 Oligochaeta, 308 olivine, 205 ombrogenic, 88, 89, 91, 95, 96, 105, 217, 462, 468 ombrotrophy, 247, 263, 264 Onamonte, Tierra del Fuego, Chile, 139 O¨nneby Mosse, Sweden, 433 Ontario, Canada, 62, 65, 67, 326, 349, 350, 355 opaline silica, 259 orangutan, 155 oreﬁeld, 409, 412, 416, 418, 420

Subject index organic acids, 99, 175, 176, 186, 197, 199, 207–10, 217, 349, 350, 359, 360 organic molecules, 209, 305, 459 organohalogens, 11, 449, 450, 452, 453, 455, 457–60, 462, 463 organoiodine, 450, 453, 454, 458 oribatid mites, 304 orthosilicate, 205 Oughtershaw, UK, 322 overland ﬂow, 319, 326, 328–30, 332–7, 341, 342, 504, 509, 518, 519, 521 Owenduff Bog, Ireland, 243, 250–2 Oxfordian, 199 oxic conditions, 294, 453 oxygen, 99, 132, 176, 183, 185, 186, 199, 218, 254, 261, 271, 319, 433, 511, 512, 532, 535, 537 oxygen isotope stages (OIS), 132 oxy-hydroxides, 92

Pacific, 21, 45, 112, 138, 146, 379 Padang (open) community, 154 Palangka Raya, Kalimantan, 161 paleoecology, 17, 310 paleoindian, 141 paleopollution, 417 paleoriver, 533, 537 paleotemperatures, 104 Paleozoic, 53, 81, 131 palsas, 30 paludiﬁcation, 1, 3, 56, 64, 87, 88, 92, 263, 349 palynology, 53, 64, 85, 418 palynological, 4, 10, 64, 126, 134, 142 palynomorphs, 235, 475 Pandanaceae, 154 Papua New Guinea, 147, 148 paralic, 247 partial contributing area concept, 329 particulate mercury, 466 Patagonia, South America, 111, 113, 130, 132, 134, 135, 140, 141, 452, 459, 460, 470, 474 Patagonian Bog, Argentina, 459 pE, 205 Peak District, Shropshire, UK, 409 peat accumulation, 39, 134, 281, 455 peat bog, 209, 409, 434, 442, 443, 450, 454, 460

Subject index peat corer, 242 peat cutting (turf cutting), 210, 377, 383, 386, 388, 392 peat decomposition, 10, 11, 101, 281, 455, 456, 458, 459, 465, 469 peat domes, 149, 154, 156 peat extraction, 377, 378, 386, 388, 502 peat failure, 9, 377, 378, 380, 381, 383, 385–8, 395, 396, 399, 401–3 peat ﬂow, 383, 385, 392, 394 peat plateau, 21, 23, 24, 30, 37, 61 peat slides, 377, 383–5, 388, 389, 392, 393, 398–401, 403, 515 peat soil matrix, 152 peatland classiﬁcations, 22, 26, 30, 31, 319 peatland development, 8, 27, 30, 35, 61, 120, 131, 323, 364, 370 peat swamp forest (in tropics), 145, 150–8, 163 pedogenetic, 92 Peepers techniques, 456 Pekan Nanas, Malaysia, 160 Pena da Cadela (PDC), Galicia, Spain, 102, 104, 105, 235, 486, 487 Penido Vello, Spain, 102–104, 218–20, 486, 487, 488 Pennines, UK, 381, 382, 384, 392, 409, 412, 416, 418, 420, 421 perennially frozen peat, 19, 30, 43–5 periglacial environments, 132 processes, 133 permafrost continuous permafrost zone, 19, 30, 43, 45, 55, 61 discontinuous permafrost zone, 19, 29, 43, 44, 55, 58 permafrost table, 19 Peru´ Peruvian Andes, 385 petrol, 421 phasic communities, 154 phenolic compounds, 281 phenolic groups, 450 phenols, 6, 221, 229, 230, 234, 235, 281, 282 Philippines, 148, 502 Phoenicians, 3, 103 phospholipid fatty acids, 293 phosphorus, 157, 271, 277, 511 phytoliths, 204, 241, 242, 259

583 phytophagous nematodes, 304 picocyanobacteria (with diameter piezometer, 537, 542 pipeﬂow, 319, 337, 338 pipes (soil pipes), 320, 337, 339, 340, 342, 509 piping, 8, 338–41, 509, 510, 514, 522 Pista de Ski moraine, Ushuaia, Tierra del Fuego, Argentina, 134 PIXE, 211 plagioclase, 209, 210, 264 plant competition, 244 plant growth, 1, 76, 78, 341 plateau bogs, 21, 23, 24, 29, 244 plattnerite (PbO2), 480 Platyhelminthes, 304 Pleistocene, 53, 55, 57, 58, 78, 88, 91, 114, 126, 129, 131, 133, 136, 141, 159, 247 pluviometer, 537 Pneumatophores, 154 podzol, 59, 61, 151, 330, 338, 439 Polar Desert, 17 Polar Front, 112 Polesie, Southeastern Europe-Russia, 349 Polistovo-Lovatsky Mire, Russia, 19 pollen exotic pollen, 131 original pollen, 132 pollen ﬂux, 131 polyaromatics, 221, 229 polycyclic soils, 92 polyphyletic, 298, 302 polysaccharides, 6, 221, 229, 230, 233, 234, 236, 237 porewater, 322, 386, 456, 460, 468, 470 porosity, 151, 152, 322, 391, 511, 531, 539 potteries, 413 Praz Rodet, Switzerland, 485, 486 pre-anthropogenic aerosols (PAA), 483, 492 Preboreal (geologic time), 105 Precambrian Shield, 64, 436 preferential ﬂow, 320, 336, 341, 342, 522 prehistoric mining, 412, 415 pre-industrial, 11, 12, 244, 250, 252, 434, 450, 452, 456, 467, 472, 474, 491 primary production, 305, 307, 309, 367, 370 primulin, 291, 292 prokaryotes, 293 protists, 7, 287, 288, 291–3, 298, 299, 302, 304, 308–11 proton-promoted dissolution, 206

584 protozoa, 8, 291, 299 protrusion, 535 Pseudopodia, 302 Puerto del Hambre, Magellan Strait, Chile, 135, 141 Puerto Harberton, Tierra del Fuego, Argentina, 134, 141 Punta Pingu¨inos, Ushuaia, Tierra del Fuego, Argentina, 134 pyridine, 230 pyrite (FeS2), 260 pyrogenic, 202, 203 pyrolysis, 6, 217, 220 pyroxenes, 205, 209

quaking bogs, 89, 91 quartz, 5, 81, 152, 175, 182, 203, 206, 208, 210, 211, 241, 246, 250 Quaternary, 104, 105, 132, 135, 141, 251 Que´bec, Canada, 37, 38, 513 Queen Charlotte Island, British Columbia, Canada, 26

radiocarbon dates, 39, 91, 95, 139, 419, 435 radiocarbon dating, 129, 131, 134, 136, 418, 419, 424, 434, 436 Radmer Valleys, Steiermark, Austria, 416 rain-fed mire, 217 rainwater, 9, 176, 199, 325, 329, 386, 388, 402, 449, 451, 522 raised bog, 115, 117, 241, 247, 296, 332, 357–9, 362, 363, 366, 394 Rajang Delta, Sarawak, 150, 162 Rajang River, Sarawak, 161 rare earth elements (REE), 7, 262, 490 rate of peat accumulation, 39, 41, 164, 242, 356 Reclus Volcano, South America, 135 red clays, 250 redox, 5, 8, 92, 175–8, 187, 191–4, 205, 292, 319, 417, 433, 512 reductive dissolution, 198, 205–7 refugia, 132 regmaglyptic, 253 rehabilitation/peatland rehabilitation, 12, 517, 522 resorcinols, 217

Subject index restoration/peatland restoration, 12, 165, 288, 307, 320, 324, 501, 511, 516, 517, 522, 523 reversible displacements, 530 rheological properties, 378 Riau Region, Sumatra, 147 ribbed fen, 21, 25 Ribble Catchment, England, 514 Rice cluster I, 294 Richards’ equation, 544 rifﬂe, 515 Rı´o Grande, Tierra del Fuego, Argentina, 136 Rı´o Varela, Estancia Harberton, Tierra del Fuego, Argentina, 136 RNA, 293 Roman Roman Iron Age, 415 Roman lead peak, 434, 436 Romano-British ironworking, 413 Rome, 262 Rookhope, UK, 421 rotifers, 291 Rubiaceae, 154 Russia, 2, 17, 19, 20, 26, 35, 41, 43, 45–7, 96, 197, 203, 439, 502 Russian-type sampler, 131

Sabah, Thailand, 147 Sacramento, California, 529, 530 Sacramento-San Joaquin Delta, California, 529, 530 Sahara, 203, 489, 490 sal/sodic soils, 175 San Joaquin Delta, California, 529, 530 San Pablo, Tierra del Fuego, Argentina, 126, 138, 139 Santorini, Greece, 247, 251, 252 sapric, 99, 101, 105, 161 Sarawak, Sumatra, 147, 149–51, 154, 155, 162, 165 Saskatchewan, Canada, 355 saturation-excess overland ﬂow, 328, 329, 333, 336, 337, 341, 342, 504 Scandinavia, 113, 244, 247, 440, 468, 502, 514 Scandinavian Shield, 489 scanning electron microscope, 7, 210, 242 Schleswig-Holstein, Germany, 415 Scho¨pfenwaldmoor, Switzerland, 485 Scotia Plate, 131

Subject index Scotland, 90, 198, 257, 340, 377, 388, 409, 420, 421, 436, 492, 493, 503, 513, 514 Scottish Highlands, 416 Sebangau Catchment, Kalimantan, 159 sedges, 30, 152, 218, 236, 276, 323, 520 selenium, 456, 474, 475 SEM, 201, 242, 251, 254, 255 SEM-EDS, 209, 242 Serra do Caurel, Galicia, Spain, 91, 102 Serra do Xistral, Spain, 102, 218 Setia Alam, Sebangau, Kalimantan, 161 settlement, 29, 141, 356, 415, 420, 529–32, 535, 537, 542, 546–8 Sg. Enok, Riau, Sumatra, 153 Sg. Nipah, Singapore, 160 Sg. Sebangau, Kalimantan, 149, 153 Shetland, Scotland, 377, 402 Sheuy, Co. Tipperary, Ireland, 420 shrinkage, 163, 394, 396, 402, 502, 503, 508, 509, 530, 539, 546 Shropshire, UK, 409 Siak Kanan, Sumatra, 161 Siberia, 6, 19, 20, 2 West Siberia, 19, 45, 53 Siberian Craton, 197 Sichuan Province, China, 2598, 46, 96, 254 siderite, 205 silanes, 220 silicate, 175, 176, 187, 203, 204, 206, 210, 246, 250, 251, 480 silicic, 208 siloxanes, 220 silver, 262, 409, 490, 491 Singapore, 160 sipping techniques, 456 Situ Bayongbong, Java, 161 Skerry Hill, Ireland, 392 slag pits, 415 Slieve Bearnach, Ireland, 385 Slieve Bloom, Ireland, 389, 402 Slieve Rushen, Ireland, 392 Slieve-an-Orra, Ireland, 392 Slovenia, 274, 468 smectite, 180, 182 smelting, 103, 192, 204, 412, 413, 415–22, 426, 492 snipes (bird), 299 Snowdonia, Wales, UK, 413, 422, 425 sodicity, 183 soil pipes, 320, 337–40, 342, 509

585 soligenous peatlands, 249 solum, 179, 184 Somerset, UK, 332, 513 soot, 248 sorption, 204, 454 South America, 3, 4, 81, 111, 129, 135, 141, 145, 251, 256 South American Plate, 131 South Pennines, England, 381, 384 South Pole, 252 Southeast Asia, 4, 145–172 Southern Hemisphere, 132, 137, 139–42, 199, 466, 472 Southern Pacific, 112, 138 Southernmost South America, 4, 129, 141, 256 spectral analysis, 534 spherules, 6, 7, 204, 206, 242, 247, 252–8 spicules, 259 spodic Bs-horizon, 440 spore, 131, 132, 139, 292, 296, 308, 310, 416 spring fens, 9, 299, 359–61, 363, 364, 370 St. Columb, Cornwall, UK, 420 Staaten Island (Isla de los Estados), Tierra del Fuego, Argentina, 131 stability analysis, 399 Steiermark, Austria, 416 steppe, 4, 41, 46, 113, 132, 136, 138, 139, 141 steroids, 221 Stigonematales, 294 stony meteorite, 252 Store Mosse, Sweden, 243, 433–6, 438, 439 Stormossen Mire, Sweden, 323 Straduff Townland, Ireland, 386 streamﬂow, 321, 324, 326, 328, 330, 333, 337, 505, 518 Subantarctic deciduous forest, 112 Sub-Atlantic, 92, 257 Sub-Boreal period, 92 suberin, 230 subsurface ﬂow, 8, 319, 323, 328, 329, 511, 518, 519 subtropical anticyclonic cells, 138 Sulawesi, 151 sulﬁ (hemist; saprist), 101, 102 sulphate soils, 146 sulphide, 198, 206, 422 sulphide deposits, 480 sulphidic, 146 sulphur, 157, 164, 204, 249, 250 Sumatra, 147–9, 155–7, 162, 163, 165

586 Suossa, Switzerland, 485 supercooling, 265 swelling, 530, 539, 542, 544, 546 Swiss Jura, 199 syringol, 225, 230, 233, 236

Taiga, 17, 19, 35, 41, 44–6 Taklamakan, China, 247 Tambille Valley, Peru´, 384 Tambora, Indonesia, 250 tar, 248, 255–8 Tardigrada, 304 tardigrads, 304 Tasek Bera, Malaysia, 151 TDR, 537, 539, 541 temperate forest, 17 temperature sensor, 538 tensiometer, 537, 538 tension inﬁltrometer, 336 tephra, 41, 42, 113, 135, 203, 208, 209, 248, 250–2, 436, 470 tephrostratigraphy, 203 terpenoids, 217, 221, 237 terrestrialization, 1, 56, 88, 91, 349 terric, 101, 102 tertiary, 111, 131, 164, 199 Testaceans, 299 tetraethyl lead (TEL), 492, 494 textinite, 67, 72 Thailand, 147, 148, 151, 483 The Netherlands, 90, 274, 502 thermal lability, 103 thionic, 101, 102 thorium (Th), 436, 480, 481, 483 Tibet, 250 till, 53, 57, 58, 115, 352, 355, 392, 440 tin, 409, 413, 420, 421 Tipperary, UK, 383, 420 To Daeng Swamp, Thailand, 160 TOF-ICP-MS, 211 TOPMODEL, 518, 520 topographic index, 340, 518, 519 total ion current (TIC), 221, 229 total organic carbon (TOC), 44, 45, 78, 513 Tourbiere de Genevez, Switzerland, 485, 486 trace element, 412, 417, 422, 423 Tremoal do Penido Vello, Serras Septentrionais, Galicia, Spain, 103, 104

Subject index Tres Arroyos, Tierra del Fuego, Argentina, 140 Trollsmosse, Sweden, 433 trophic status, 302 tropical peat soils, 145, 152 tropics, 4, 5, 145, 166, 198, 199, 241 Trout Beck, 325, 326, 334 Tullynascreen Townland, Ireland, 388 Tundra, 17, 19, 35, 41, 44–7, 65, 131, 132, 136, 137 Tunguska, Russia, 244, 252, 254 typology, 88–91, 94 Tyrrell Sea, Canada, 53, 57, 58, 355

Ukraine, 19, 505 ulminite, 72 unfrozen peatlands, 18, 19 Ungava, Canada, 352 United Kingdom (UK), 92, 274, 280, 377, 409 United States of America (USA), 8, 96, 349, 442, 451, 466, 467, 491 Upper Continental Crust (UCC), 197, 437, 480 Upper Tees Catchment, UK, 324, 325, 333, 337 Upper Wharfedale, UK, 321, 322, 331, 518, 519 Ural River, 41 uranium, 480 urea, 271 Ushuaia, Tierra del Fuego, Argentina, 4, 134, 137, 139, 141 U.S.S.R., 20, 22, 26–8, 33, 43 Utermo¨hl method, 292

Valley Bog, UK, 377, 398, 421 variable source area concept, 329 varve, 433–6 vascular plants, 88, 198, 202, 272, 273, 275, 276, 278, 280, 282, 307, 309, 323 Vega Varela, Tierra del Fuego, Argentina, 122 Venice Lagoon, Italy, 531–3, 537, 547 Venice, Italy, 13, 529–50 Vesuvian eruptions, 247 Vietnam, 148 volcanic glass shards, 242, 251, 258 Volga River, 41

Subject index VOSS, 531 Vraconne, Switzerland, 379, 388

Wales, UK, 62, 330, 338, 409, 412–4, 419, 421, 422, 424, 513 water content, 287, 288, 292, 308, 330, 369, 391, 393, 530, 539–41, 546 water storage capacity, 326, 508, 509 water tracks, 336, 350, 352, 357, 358, 360, 361, 363, 364, 522 Weald, UK, 412, 421 Welsh Bronze Age, 422 Welsh, UK, 412, 422 Wessenden Head Moor, England, 384 West Papua, 147 Western Europe, 92, 247, 249, 251, 332, 490 Western Russia, 246 wet fens, 288 Wharfedale, Yorkshire, UK, 321, 322, 331, 518–20 Wilburton Phase metalwork, 420 Wingecarribee Swamp, Australia, 379, 388, 402

587 Wisconsinan glaciation, 35 Wolsingham South Moor, Weardale, UK, 420 Wu¨rm glaciation, 88 Wu¨stite, 254

xenobiotics, 449 Xistral Mountains, Spain, 190 XRD, 210, 211 XRF, 211

Yellow River, China, 383 Yorkshire, UK, 321, 392 Younger Dryas, 137, 142, 490 Yukon Territory, Canada, 35

Zennare Basin, Italy, 531–4, 537, 538, 540, 544, 547 zeolites, 181 zinc, 104, 261, 409, 418, 422 Zygomycota, 296

This page intentionally left blank

Environmental Geomorphology (Developments in Earth Surface Processes 4)

Mountains: Witnesses of Global Changes, Volume 10: Research in the Himalaya and Karakoram: SHARE-Asia Project (Developments in Earth Surface Processes) (Developments in Earth Surface Processes)

Evolution of Earth and its Climate: Birth, Life and Death of Earth (Developments in Earth and Environmental Sciences 10)

Antarctic Climate Evolution, Volume 8 (Developments in Earth and Environmental Sciences)

Global Warming and Global Cooling, Volume 5: Evolution of Climate on Earth (Developments in Earth and Environmental Sciences)

Catchment Dynamics and River processes, Volume 7 (Developments in Earth Surface Processes)

Climate, Environment, and Society in the Pacific during the Last Millennium, Volume 6 (Developments in Earth and Environmental Sciences) (Developments in Earth and Environmental Sciences)

Earth Surface Processes, Landforms and Sediment Deposits

Developments in Earth Surface Processes, Volume 13: Special volumes of geomorphology (Natural Hazards and Human-Exacerbated Disasters in Latin America)

Peatlands, Volume 9: Evolution and Records of Environmental and Climate Changes (Developments in Earth Surface Processes)

Environmental Geomorphology (Developments in Earth Surface Processes 4)

Mountains: Witnesses of Global Changes, Volume 10: Research in the Himalaya and Karakoram: SHARE-Asia Project (Developments in Earth Surface Processes) (Developments in Earth Surface Processes)

Evolution of Earth and its Climate: Birth, Life and Death of Earth (Developments in Earth and Environmental Sciences 10)

Antarctic Climate Evolution, Volume 8 (Developments in Earth and Environmental Sciences)

Global Warming and Global Cooling, Volume 5: Evolution of Climate on Earth (Developments in Earth and Environmental Sciences)

Global Warming and Global Cooling, Volume 5: Evolution of Climate on Earth (Developments in Earth and Environmental Sciences)

Catchment Dynamics and River processes, Volume 7 (Developments in Earth Surface Processes)

Physical Processes in Earth and Environmental Sciences

Rock Coatings (Developments in Earth Surface Processes 6)

Rock Coatings (Developments in Earth Surface Processes 6)

Rock Coatings (Developments in Earth Surface Processes 6)

Climate, Environment, and Society in the Pacific during the Last Millennium, Volume 6 (Developments in Earth and Environmental Sciences) (Developments in Earth and Environmental Sciences)

Earth Surface Processes, Landforms and Sediment Deposits

Earth surface processes, landforms and sediment deposits

Earth Surface Processes, Landforms and Sediment Deposits

Developments in Earth & Environmental Sciences, 7

Snow and Climate Physical Processes Surface Energy Exchange and Modeling

Quantitative Modeling of Earth Surface Processes

Developments in Earth Surface Processes, Volume 13: Special volumes of geomorphology (Natural Hazards and Human-Exacerbated Disasters in Latin America)

Advanced Biological Treatment Processes: Volume 9 (Handbook of Environmental Engineering)

Developments in Surface Contamination and Cleaning

Geosciences, Environment and Man (Developments in Earth and Environmental Sciences, Volume 1)

Evolution of Earth and Its Climate: Birth, Life and Death of Earth

Climate and Climate Change (The Living Earth)

Oil Pollution and its Environmental Impact in the Arabian Gulf Region, Volume 3 (Developments in Earth and Environmental Sciences)

Surface Chemistry in Biomedical and Environmental Science

Cultural Landscapes and Environmental Changes (Key Issues in Environmental Change)

Earth Paleoenvironments: Records preserved in Mid- and Low-Latitude Glaciers (Developments in Paleoenvironmental Research)

Unit Operations and Processes in Environmental Engineering

Environmental Foresight and Models Developments in Environmental Modelling

Peatlands, Volume 9: Evolution and Records of Environmental and Climate Changes (Developments in Earth Surface Processes)

Environmental Geomorphology (Developments in Earth Surface Processes 4)

Mountains: Witnesses of Global Changes, Volume 10: Research in the Himalaya and Karakoram: SHARE-Asia Project (Developments in Earth Surface Processes) (Developments in Earth Surface Processes)

Evolution of Earth and its Climate: Birth, Life and Death of Earth (Developments in Earth and Environmental Sciences 10)

Antarctic Climate Evolution, Volume 8 (Developments in Earth and Environmental Sciences)

Global Warming and Global Cooling, Volume 5: Evolution of Climate on Earth (Developments in Earth and Environmental Sciences)

Global Warming and Global Cooling, Volume 5: Evolution of Climate on Earth (Developments in Earth and Environmental Sciences)

Catchment Dynamics and River processes, Volume 7 (Developments in Earth Surface Processes)

Physical Processes in Earth and Environmental Sciences

Rock Coatings (Developments in Earth Surface Processes 6)

Rock Coatings (Developments in Earth Surface Processes 6)

Rock Coatings (Developments in Earth Surface Processes 6)

Climate, Environment, and Society in the Pacific during the Last Millennium, Volume 6 (Developments in Earth and Environmental Sciences) (Developments in Earth and Environmental Sciences)

Earth Surface Processes, Landforms and Sediment Deposits

Earth surface processes, landforms and sediment deposits

Earth Surface Processes, Landforms and Sediment Deposits

Developments in Earth & Environmental Sciences, 7

Snow and Climate Physical Processes Surface Energy Exchange and Modeling

Quantitative Modeling of Earth Surface Processes

Developments in Earth Surface Processes, Volume 13: Special volumes of geomorphology (Natural Hazards and Human-Exacerbated Disasters in Latin America)

Advanced Biological Treatment Processes: Volume 9 (Handbook of Environmental Engineering)

Developments in Surface Contamination and Cleaning

Geosciences, Environment and Man (Developments in Earth and Environmental Sciences, Volume 1)

Evolution of Earth and Its Climate: Birth, Life and Death of Earth

Climate and Climate Change (The Living Earth)

Oil Pollution and its Environmental Impact in the Arabian Gulf Region, Volume 3 (Developments in Earth and Environmental Sciences)

Surface Chemistry in Biomedical and Environmental Science

Cultural Landscapes and Environmental Changes (Key Issues in Environmental Change)

Earth Paleoenvironments: Records preserved in Mid- and Low-Latitude Glaciers (Developments in Paleoenvironmental Research)

Unit Operations and Processes in Environmental Engineering

Environmental Foresight and Models Developments in Environmental Modelling

Recommend Documents