Soils on a Warmer Earth

Developments in Soil Science 20

SOILS ON A WARMER EARTH

Further Titles in this Series 1. I. VALETON BAUXITES 2. IAHR FUNDAMENTALS OF TRANSPORT PHENOMENA IN POROUS MEDIA 3. F.E. ALLISON SOIL ORGANIC MATTER AND ITS ROLE IN CROP PRODUCTION 4. R. W. SIMONSON (Editor) NON-AGRICULTURAL APPLICATIONS OF SOIL SURVEYS

5A. G.H. BOLT and M.G.M. BRUGGENWERT (Editors) SOIL CHEMISTRY. A. BASIC ELEMENTS 5B. G.H. BOLT (Editor) SOIL CHEMISTRY. B. PHYSICO-CHEMICAL MODELS 6. H.E. DREGNE SOILS OF ARID REGIONS

7. H. AUBERT and M. PINTA TRACE ELEMENTS IN SOILS

8. M. SCHNITZER and S. U. KHAN (Editors) SOIL ORGANIC MATTER

9. B.K.G. THENG FORMATION AND PROPERTIES OF CLAY-POLYMER COMPLEXES 10. D. ZACHAR SOIL EROSION 11A. L.P. WILDING, N.E. SMECK and G.F. HALL (Editors) PEDOGENESIS AND SOIL TAXONOMY. I. CONCEPTS AND INTERACTIONS 11B. L.P. WILDING, N.E. SMECKand G.F. HALL (Editors) PEDOGENESIS AND SOIL TAXONOMY. 11. THE SOIL ORDERS 12. E.B.A. BISDOM and J. DUCLOUX (Editors) SUBMICROSCOPIC STUDIES OF SOILS 13. P. KOOREVAAR, G. MENELIK and C. DIRKSEN ELEMENTS OF SOIL PHYSICS

14. G.S. CAMPBELL SOIL PHYSICS WITH BASIC-TRANSPORT MODELS FOR SOIL-PLANT SYSTEMS 15. M A . MULDERS REMOTE SENSING IN SOIL SCIENCE 16. I.B. CAMPBELL and G.G.C. CLARIDGE ANTARCTICA: SOILS, WEATHERING PROCESSES AND ENVIRONMENT 17. K. KUMADA CHEMISTRY OF SOIL ORGANIC MATTER 18. V. VANCURA and F. KUNC (Editors) INTERRELATIONSHIPS BETWEEN MICROORGANISMS AND PLANTS IN SOIL 19. L.A. DOUGLAS (Editor) SOIL MICROMORPHOLOGY: A BASIC AND APPLIED SCIENCE

Developments in Soil Science 20

SOILS ON A WARMER EARTH effects of expected climate change on soil processes, with emphasis on the tropics and sub-tropics Edited by

H.W. SCHARPENSEEL Institute of Soil Science, University of Hamburg, Allende-Platz, 0-2000 Hamburg 13, F.R. Germany and

M. SCHOMAKER and A. AYOUB UNEP, P.O. Box 30552, Nairobi, Kenya Proceedings of a n International Workshop on Effects of Expected Climate Change on Soil Processes in the Tropics and Sub-tropics, 12-14 February 1990, Nairobi Organized by United Nations Environment Programme (UNEP) International Society of Soil Science (ISSS) Sponsored by United Nations Environment Programme (UNEP ) UNEP International Society of Soil Science (ISSS) EC Technical Centre for Agriculture and Rural Development (CTA)

(ISSS)

ELSEVIER Amsterdam - Oxford -New York - Tokyo

1990

ELSEVIER SCIENCE PUBLISHERS B.V. Sara Burgerhartstraat 25 P.O. Box 211,1000 AE Amsterdam, The Netherlands Distributors for the United States and Canada: ELSEVIER SCIENCE PUBLISHING COMPANY INC. 655, Avenue of the Americas New York, NY 10010, U S A .

ISBN 0-444-88838-1

0Elsevier Science Publishers B.V.. 1990 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, Elsevier Science Publishers B.V./ Physical Sciences & Engineering Division, P.O. Box 330,1000 AH Amsterdam, The Netherlands. Special regulations for readers in the U S A . - This publication has been registered with the Copyright Clearance Center Inc. (CCC ), Salem, Massachusetts. Information can be obtained from the CCC about conditions under which photocopies of parts of this publication may be made in the USA. All other copyright questions, including photocopying outside of the USA, should be referred to the copyright owner, Elsevier Science Publishers B.V., unless otherwise specified. 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. This issue is printed on acid-free paper. Printed in The Netherlands

V

CONTENTS Forewords Pn=face Conclusions and recommendations Chapter 1

Chapter 2

Chapter 3 Chapter 4

Chapter 5

Chapter 6

Chapter 7

Overview of the greenhouse effect. Global change syndrome, general outlook H .W . Scharpenseel and P . Becker-Heineman

Chapter 9

1

Inputs to climatic change by soil and agriculture related activities. Present status and possible future trends A.F. Bouwman and W.G. Sombroek

15

Processes that affect soil morphology R .W . Arnold

31

Influence of climatic change on soil moisture regime, texture, structure and erosion G.Y. Varallyay

39

Resilience against climate change? Soil minerals, transformations and surface properties, Eh, Ph R. Brinkman

51

Impact of climatic change on soil attributes. Influence on salinization and alkalinization I . Szabolcs

61

Soil organic matter and biology in relation to climate change P.B. Tinker and P . Ineson

Chapter 8

ix xiii xv

71

Influence of climatic change on development of problem soils, especially in the alluvial domains W.R. Fisher Addendum: Methane formation in waterlogged paddy soils and its controlling factors Y.Takai and E . Wada

101

Potential influence of climate change on soil organic matter and tropical agroforestry E.H. Franz

109

89

vi

Contents

Chapter 10 The use of models of soil pedogenic processes in understanding changing land use and climatic conditions 121 J.W.B. Stewart, D.W. Anderson, E.T. Elliott and C.V. Cole Addendum: Modeling nitrous oxide production by denitrification 133 J.R.M. Arah and K.A. Smith Chapter 11 Impacts of anthropogenic atmospheric pollution on soils, with special relevance to tropical and subtropical soils, and possible consequences of the greenhouse effect N . Van Breemen

137

Chapter 12 Changes in soil resources in response to a gradually rising sea-level H . Brammer and R . Brinkman

145

Chapter 13 Soils on a warmer earth: tropical and subtropical regions W.G. Sombroek

157

Chapter 14 Soils on a warmer earth: subtropical and Mediterranean regions 175 D. Yaalon Chapter 15 Impact of climate warming on arid region soils D.H. Dregne

177

Chapter 16 Soils of the subboreal region on a warmer earth B.G. Rozanov and E.M. Samoilova

185

Chapter 17 Climate induced changes of the boreal and subpolar soils S.V. Goryachkin and V.O. Targulian

191

Chapter 18 Approaches to mitigate tropical deforestation by sustainable soil management practices P.A. Sanchez, C.A. Palm and T.J. Smyth

21 1

Chapter 19 Managing global change by curtailing emission sources and creating new sinks R. Grantham

22 1

Chapter 20 Implications of the greenhouse effect for African agriculture R.S. Odingo

23 1

Contents

vii

Chapter 21 The agricultural environment of Latin America and the Caribbean and the greenhouse effect N . Ahmad

249

Chapter 22 The Asian agricultural environment and the greenhouse effect L. Venkutaratnam

267

This Page Intentionally Left Blank

ix

FOREWORDS The motivation for organizing the International Workshop ‘Effects of Expected Climate Change on Soil Processes in the Tropics and Subtropics” was threefold. Soil Science is heavily involved in the problems of trapped radiation in the atmosphere by greenhouse effect (GHE) promoting gases. The e€fect of soils on the GHE syndrome was analyzed in a preceding conference: the International Conference on Soils and the Greenhouse Effect, Wageningen, the Netherlands, August 1989, organized by the International Soil Reference and Information Centre (ERIC). An answer was needed regarding the effect of climate change on sustainable productivity of soils, with emphasis on tropical and subtropical regions. The confusing dispute in scientific and popular science journals regarding the predictions of effects of temperature and eustatic sea-level rise, as well as a wide array of possible advantages and disadvantages due to especially C02-rise, had to be thoroughly evaluated against the background of impacts on soil changes. A state of the art of predictive approaches, recognized by the majority of scientists, had to be elaborated. In this volume some 24 scientists contributed to such an evaluation and inventory, each in their own specific field. Though differences in opinion on assumptions, approaches and predictions still exist, these proceedings clearly bring the discussion a great step forward, and it is hoped that they will be of assistance to the soil science community in deciding on concepts for soil related core programmes of the forthcoming International Geosphere Biosphere Programme. The editors

x

Forewords

What once were local incidents of environmental damage, confined to one area and region now involve the whole world. Concern on the global environment has indeed become more and more pressing. All people concentrate their attention to this matter, regardless of their wealth. The United Nations Congress, the Alsh Summit and the Tokyo Conference on Global Environment and Human Response Towards Sustainable Development in 1989, issued the statements that the human race may be making our planet uninhabitable. They are concerned about the rapid decrease in land which should be set aside for forests and other vegetation, essential for maintaining the balance of atmospheric gases, and about the increasing deterioration of soil, water and air, that support the continued existence of all life on the earth. These threatening developments call for an increased emphasis on studies specifically aimed at problems of global change. In a broader perspective the need to expand the role of soil scientist's approaches to global environmental and resources utilization is evident. May these proceedings become an important milestone, which will amongst others contribute to a fruitful discussion during the 14th International Society of Soil Science Congress and its symposium "Global Soil Changes Under Influence of a Changing Environment" (Kyoto, Japan, August 1990). Professor Y. Takai Vice-president, ISSS

Fore words

xi

The theme of this book is a far reaching one and should attract great interest, both in industrialized as well as in less developed countries. Although the question of climate change and its effects on soil productivity is overshadowed by so many current events, it is almost certain that this subject will remain on the international agenda for the years to come; not only as science and research are concerned but also in relation to development policies, agricultural policies and others, especially since poor, rich, developed and developing countries are equally threatened by these problems. The Technical Centre for Agricultural and Rural Co-operation (CTA), provides ACP states with better access to information, research and innovations in the spheres of agricultural and rural development. CTA gives great attention to increasing agricultural production on a sustainable basis and CTA is prepared to strengthen and expand these activities in the years to come. Within this context the recommendations in these proceedings and the subsequent follow-up will be of great interest to us and we are certainly prepared to continue our collaboration in this respect with the United Nations Environment Programme (UNEP) and the International Society of Soil Science (ISS S ) . Dr. W. Treitz Deputy Director, CTA/ACP-EEC Lorn6 Convention

xii

Forewordr

Each day, we are getting a clearer understanding of the growing pressures that threaten our planetary biosphere. Each day, pressures are mounting, due to demographic momentum and rising industrialization. Though uncertainties remain about the magnitude of climate change, scientific evidence confirms human activity has undoubtedly altered, and continues to alter, the atmosphere. While the principle greenhouse gas is carbon dioxide from fossil fuels, agricultural practices are having important impacts on possible changes in global climatic regimes. Recent studies suggest burning of savannahs for agriculture contributes three times more carbon dioxide per hectare to the atmosphere as does burning of tropical forests. Savannah clearing and increase in cattle and rice paddies are just three greenhouse gas sources from agriculture. Studies suggest soil processes involving inputs and outputs of water, gases, soluble salt and organic matter are likely to be influenced by global mean temperature increases. Such changes could in turn reduce land productivity, further threaten biodiversity, exacerbate soil loss and disrupt sediment budgets. Productive cropping areas could turn into deserts, while coastal storms could inundate or degrade fertile lands. Restoring the balance between the sources and the sinks of greenhouse gases will likely not be achieved solely through carbon emission reductions. Conserving existing carbon sinks (oceans, rainforests, grasslands, mangroves, coral reefs, etc.) and significantly increasing the productivity on degraded soils would be cost effective measures to achieve both environmental and economical sustainability. Massive reforestation, regrassing and amelioration of degraded lands are needed. A growing challenge to soil experts and agriculturalists is to produce more food for our growing world-wide population (currently already at 5.2 billion) from finite and shrinking productive lands. Of the total world-wide area of potential arable land, nearly half (about 1.5 billion ha) is already cultivated. There are very few virgin lands easy to exploit for additional food production. With these proceedings, a serious effort has been made to: 1) discuss latest knowledge on the expected impact of climate change on soil processes, with emphasis on the tropical and sub-tropical regions; 2) agree on recommendations for meaningful and useful future research and monitoring programmes; and 3) formulate policy responses to the expected impact in the form of preventive and adaptive measures. I hope that with these proceedings, we have moved forward to a better understanding of soil processes and their relationship to climate change. Mostafa K. Tolba Executive Director, UNEP

...

Xlll

PREFACE The Plan of Action for the implementation of the World Soils Policy was endorsed by UNEP’s Governing Council in 1984. The document identified the international action required to promote the sound use of land and soil resources. It also highlighted the need for internationally acceptable methods for assessing and monitoring the existing status and risk of soil degradation. The importance of ensuring that practices to prevent soil degradation, improve land productivity and reclaim degraded areas are appropriate to Iocal physical, economic, social and cultural conditions had been recognized in the document. At the same time, it is also recognized that serious gaps exist in knowledge of the extent, mechanisms and economic consequences of soil degradation as well as in knowledge of costeffective means for controlling erosion and other forms of land degradation. Consequently, there is an equally urgent need for a long term commitment by international organizations and Governments to support research on mechanisms and effects of soil degradation on the development for combatting soil erosion, as well as for the identification of critical areas and for a standardized assessment of soil resources and of the seriousness of soil degradation, through surveys and inventories. Methods are required which can reliably detect significant changes in those soil and land characteristics which directly or indirectly affect the quality and quantity of the land and its liability to produce food, fibre and timber. The manipulation of the pedosphcre by man is substantial and is likely to increase strongly until at least the year 2050, whcn the world population is expected to peak. This manipulation certainly has an impact on global climate and should be taken into account in connection with the increasing greenhouse effect on climate. The present status and future trends concerning the effects of soils and land cover on the fluxes of greenhouse gases, the surface energy balance and the water balance are discusscd in Bouwman (1990). The current publication explores and assesses the concept, the trends and impacts of the possible global warming on soil processes in the tropics and sub-tropics. Hierarchical research requirements and response stratcgies nccessary to meet the ncgative impacts of climate change on soils are also outlined. UNEP sees that the great task for the world community is therefore three fold: 1) to reduce greenhouse gascs; 2) to prepare for the impacts of global warming; and 3) to provide technical and financial assistance to developing countries so that their strive towards legitimate economic expansion is not hampered by measures addressing global warming. UNEP welcomes thc concern expressed on global warming at the highest levels. Thcse and other priorities, such as protection of land resources by combatting dcsertification and

x iv

Preface

deforestation, have bcen constant pre-occupations of UNEP for well over a decade. The preparatory work for global conventions on climate change and biological diversity are but a few of the many actions of UNEP to safeguard our planet. A.T. Ayoub, Senior Programme Officer, UNEP

xv

CONCLUSIONS AND RECOMMENDATIONS OF THE INTERNATIONAL WORKSHOP ON THE EFFECTS OF EXPECTED CLIMATE CHANGE ON SOIL PROCESSES The workshop discussed how soils in specific ecosystems respond to changes of climate and the needfor soil scientists to work with other disciplines.It also listed research priorities and discussed how the use and management of soil resources can mitigate climate change and at the same time adapt to it.

ECOSYSTEM DYNAMICS, INTERACTIONS WITH THE ATMOSPHERE AND THEIR EFFECT ON SOIL AND LAND USE A scenario and some predictions

We assumed: over a time span of about 50 years a gradual warming of the atmosphere of about 3°C with 2°C in tropics and 5°C in sub-polar areas; 10% increase of precipitation, a sea level rise of 50 cm, and an equilibrium world population of 10 billion; by about the year 2100 a doubling of the atmospheric C 0 2 content. The results of geophysical, geochemical, and biogeochemical processes vary widely in current soil ecosystems. Within a decade changes of temperature and precipitation will influence soil temperature and moisture regimes, pH, base saturation, fertility status, surface litter and biological activity, and the presence of salic and fluvic soil properties where appropriate. Over a period of several decades changes of climate are manifested in soils through the depth and kind of humus in the topsoil, relative fertility, erosion, and in peatiness, swell-shrink features, degree of bleaching and calcareoumess. We believe that soil ecosystem responses to changes in land cover and climate differ mainly due to different degrees in sensitivity of soils to disturbance and modification. It is recognized that there are likely to be very large regional and local variations in the manifestation of warmer earth conditions. Nevertheless the following general statements can be made for some major ecosystems. In a Tropical Rain Forest ecosystem (hot, humid) one might expect increases of forest growth, nutrient cycling, and mineral weathering. Ferralsols are common in such biomes and with more percolation through the soil we would expect a decline of the fertility status, slight decreases of pH and base saturation, more litter production but also more biological activity with little change in humus content in the topsoil. Additional Weathering of minerals would occur, however, due to the great thickness of many saprolites there would be little change in relative fertility over time. Although soil processes would respond to the changes of climate and vegetation the overall impact would be minimal compared to the present ecosystcm. Thcse kinds of soils have low sensitivity to the assumed

xvi

Conclusions and recommendarions

climatic changes. In a Steppe or Grassland ecosystem (warm, semiarid) the increase of precipitation would increase biomass production and movement of water through and on the soil. In such ecosystems, Chernozemic soils are common; they are moisture sensitive and the humus content in the topsoil would increase, carbonates and other soluble salts would leach deeper and salinization might occur in adjacent lower areas. Soil pH and base saturation would decrease slightly but the nutrient supply would remain high. Increased erosion would occur on slopes where vegetation would become stressed. In a subpolar Boreal Forest ecosystem (cold, subhumid) we would expect an increase of biopmductivity and biogeochemical cycling. For the dominant soils in this ecosystem, Podzols, the humus content, base saturation, relative fertility status and pH would become higher. The litter and liistic horizon, if any, would become thinner, leaching would tend to intensify, and the spodic horizon would likely become thicker. The vegetative cover would minimize erosion of the soil. The soil processes in this ecosystem are fairly sensitive to the assumed climate changes, especially temperature. We conclude that the changes in soil processes will be substantially more pronounced in cold and temperate regions than in the tropics because the soils outside the tropics are more temperature sensitive and the temperature changes are expected to be greater there. As one approaches the transitional zones between ecosystems ("ecotones") it is more difficult to make general predictions about soil changes because of the strong interactions of climate and biota that often occur in such zones. In coastal swampy areas and tropical mountains a global warming can have dramatic effects on the ecosystems. Rising sea levels would generate a host of events such as flooding, salinization, and new erosion and sedimentation patterns that would affect many soil ecosystems. In mountainous areas the ecological zones respond to elevation, thus a warming would push the climatic belts up the mountains, creating new conditions. The essential contribution of soil science to climate change studies The impact of climate change will be extremely diverse and difficult to predict. The consequence for soils and their associated vegetation will be even more complex, and will have the most wide ranging and diverse consequences for agriculture, forestry, ecology, conservation, and all forms of land use. Progress in understanding and pedicting these effects can only be made by multidisciplinary studies. Soil science forms an essential component of such work. Soil is the basis for plant growth, and the full effects of climate change will be mediated by complex interaction between the two. There will also be direct effects of increase in atmospheric C02.At the highest level, full scale ecosystem studies are essential to understand changes resulting from broad shifts of vegetation bands and biomes. Major modeling programmes will be needed at

Conclusions and recommendarions

xvii

several different scales of size and complexity. Apart from these vital studies on impact, it must not be forgolten that soils have major sources and sinks for important greenhouse gasses: carbon dioxide, water vapor, methane, nitrous oxide. These fluxes require much more accurate measurement, and the understanding of all underlying processes. We also need to know how the expectcd climate changes will interact with these greenhouse gas processes in positive or negative feedbacks. Major research programmes are bcing dcvcloped on these themes in a number of countries, many with a strong international component. It is critically important that the central position of soil science in such studies is recognized at an early stage, so that these programmes are fully balanced and complete. This also demands from the soil science community that it is prepared to collabotate freely and widely with other essential disciplines having a bearing in this issue. These studies require new techniques, which are only just being introduced into soil science, such as molecular biology, natural abundance stable isotope studies, and the newest methods of remote sensing. A crucial group of soil processes arc tcrmed "anthropogenic", referring to those brought about by the activitics of man. Thcsc processes likely override all others. The implications of man's direct influence on soil processes, on ecosystems, and on the earth systcm as a whole are such that they tend to overshadow our attempts to deal with natural ecosystems and climate-induced changes. Such anthropogenic processes have the potential to change the course of civilization especially if they are ignored.

RESEARCH PRIORITIES ON EFFECTS OF EXPECTED CLIMATE CHANGE ON SOIL PROCESSES, WITH EMPHASIS ON THE TROPICS AND SUBTROPICS The soil research priorities identified during the workshop can be categorized in 6 main groups: Collection of baseline data; Study of soil processes; Study of land use dynamics; Modelling of soil processes; Long-term monitoring of global soil change; Management and mitigation measures of global soil change. Furthermore, to carry on many of thc soil studies, data have to be collected and processed in a whole hcarted collaborative effort with other disciplines, such as those dealing with climatology, ecology, and crop production. No priorities were identified for research on emission of greenhouse gases, because these were considered during the recent conference on "Soils and the

xviii

Conclusions and recommendafions

Greenhouse Effect" in August 1989 in Wageningen (see Bouwman 1990). The following specific subjects in each of the six categories were identified as priority research subjects. Base line data A large number of data bases pertaining to the nature and state of the worlds soil cover exist. To prcdict effects of climatic change on soil properties and on quantitative and qualitative aspects of ground water recharge and on surface water, it is necessary to identify gaps in these data bases. A first brief survey indicated the following gaps (some of which may overlap with each other): Soil information on a scale of 1:1 million, or even smaller (such as the FAOAJnesco Soil Map of the World at a 1 : 5 million schale), of large parts of Africa, South America, and Asia is not accurate enough for predictive and interpretative purposes; larger scales (up to 1:250,000) are needed. The state of soil/land degradation is generally poorly understood and quantified, although an overview will be available soon through the UNEPDSRIC Global Assessment of Soil Degradation (GLASOD) Project. Special attention should bc paid to mapping of (potentially) saline, alkaline and acid sulfate soils. Global data bases on actual land use are either out-of-date or too general. These baseline data should be organizcd and made available through the development of Land Information Systems (LIS) such as the ISSSDSRIC Soil and Terrain Digital Database (SOTER). Paleopedological maps of the world (1:2.5 million) of 2 or 3 climatic periods in the Pleistocene and Holocene will be very hclpful in analogue studies to forecast the effect of climatic change on present-day soils and for estimating the carbon cycle changes. As a first priority it would be sufficient to prepare such maps for a numbcr of selected, promising areas, whcre more background data are available. The Unesco-initiated project of mapping continental soil salinization and potential soil salinization should be cornpletcd. Soil processes and their dynamics In studying soil processes, priorities should be given to processes that operate on a time scale of 1 to 100 years, including those that may change suddenly or catastrophically. We bclieve that these are (in order of decreasing

Conclusions and recommendations

xix

potential rate of change): Changes in soil salinity and alkalinity (one month to 10 years); Changes in soil meso- and macro fauna important for bioturbation and homogenization (1 to 10 years); Changes in structural stability and moisture characteristics (1-10 yrs); Changes in amount and quality of organic C and N levels (1-10 yrs); Changes in nutrient status, acidity, redox regime (10 to 100 years); Changes in susceptibility to erosion (10 to 100 years); Changes in iron and amorphous minerals (10 to 100 years). A number of methods are available for this purpose: Analogue studies, using soil chronosequences or using situations where the soil climate has actually changed substantially (due to deforestation, artificial drainage, ponding, etc.) Manipulation studies in the field (small chamber, small watershed), greenhouse or laboratory. Field studies on soil processes should preferably be camed out in areas involved in a network for long-term monitoring (see below). An important problem to be addressed is the methodology of scaling down experimental (as well as modeling) results to values that are relevant at less detailed scales.

Changes in land use For a number of other disciplines (agronomy, animal husbandry, economics, general politics, medicine) it will be important to predict the changes in land use resulting from climatic change and related changes in soil properties. Soil scientists will be asked to provide data needed to help forecast optimal land use on the short (0-10 y) medium (10-30 y), and long (40-60 y) run. GIs- or LIS-based expert systems should be very helpful in this regard.

Mo dei i n g Process oriented simulation models describing changes in soil properties, as a function of temperature and precipitation (time scale of days, weeks, months, or years) are urgently needed. Research on soil processes and collection of longterm monitoring that should provide parameters and opportunities for calibration and validation of these models.

Long term monitoring To detect ecologically relevant changes related to climate change, and to provide opportunities for research, a network of stations for long-term

xx

Conclusions and recommendations

monitoring of vegetation-soil properties - topography-hydrology-hydrochemistryclimate of a number of globally representative,natural, rangeland and agricultural ecosystems should be set up. Priority should be given to: Transitional regions where rapid climate change is expected; Coastal areas where sea level changes are expected; Irrigated areas in semiarid regions; Deforested areas; Natural wetlands. Besides detailed monitoring at field/small catchment scale on the ground, detailed remote sensing monitoring of the general region, representative for the ecosystem, should be carried out. Remote sensing should be directed to estimating vegetation (cover), net primary production, evapotranspiration, surface temperature, IR radiation and, where applicable, surface soil properties. The monitoring research sites should be run by local organisations and coordinated internationally to promote the use of common methodologies and exchange of data. Data from other disciplines Climatologists

In addition to forecasts about mean annual temperature and precipitation, it will be very important to obtain estimates of the ranges of year to year and month to month variability, especially in rainfall, and the expected frequency of high intensity, highly erosive rainfall. Ecologists and crop production physiologists Effects of increased CO;!on organic matter return to the soil, estimated on the basis of crop production models. In addition to quantity, qualitative aspects of the organic matter (possible changes in biodegradability) should be considered.

ADAPTATION TO CLIMATE CHANGE: SOIL RESOURCES USE AND MANAGEMENT The workshop participants discussed strategies to counteract increases in greenhouse gases and defend present soils and their uses against adverse effects of climate change. The resulting recommendations are directed at policy makers and at the users of the soils. Management strategies were identified that are desirable in their own right as well as in the context of the greenhouse effect. Four sets of concerns with accompanying recommended management strategies are reported:

Conclwionr and recommenda~iom

xxi

Increasing C 0 2 sinks The following measures are advocated, recognizing that soil- and land management-related methods of C02 sequestering on-their-own would only remove a fraction of the increased atmospheric carbon. Besides removing CO2, each of these measures would provide direct production benefits to the land user. Conditions that favour C retention in soil and biomass include wetness, low temperatures, increased mineral nutrition (soil fertility), deep root development as well as biological activity. Management strategies should aim to preserve and, where possible, increase standing biomass and residues in soils. On many soils this will involve appropriate use of mineral fertilizers. On acid soils, selection of aluminium-tolerant crops or cultivars and liming of the acid subsoil to encourage deep root development are recommended. Irrigation in arid areas, either to produce grassland or annual field crops or to establish perennial vegetation, will increase standing biomass and soil organic matter. In the case of hardpans or other root barriers, subsoiling would enable deeper root development.

Reducing N2O and CH4 emissions Conditions favouring N2O emission from soils include high concentrations of mineral N species and alternation of reduction and oxidation. Conditions favouring methane emission include deep reduction in the absence of sufficient active iron oxides as well as of sulfate. Management measures to avoid such conditions include the following: Proper dosage. timing and placement of nitrogen fertilizers in accordance with crop growth stages, to minimize periodic excess N. Drainage or cultivation or other measures to avoid topsoil compaction and periodic reduction in cropland or pastures. In wetland soils, efficient water management so as to avoid periodic oxidation. Under rainfed conditions, this may involve land shaping, bunding, puddling, supplementary irrigation where feasible. Methane emission from wetland soils can be rninimizcd by avoiding incorporation of readily decomposable organic matter, and by addition of material rich in ferric iron, where practical, on soils low in active iron oxides. Also the use of fertilizers with the sulfate anion suppresses methanogenesis.

xxii

Conclusions and recommendations

Adapting to sea-level rise Sea-level rise causes a destabilized coastline, inland extension of areas affected by salinity and increased flooding depth and wetland conditions in inland parts of coastal plains. These problems can be combatted by one or more of the following measures: Sea defense including structural works such as embankments, as well as planting or preservation of protective mangrove forest belts. Embankment and pump drainage, or tidal drainage where feasible, of the impoldered land. Changing land use to fish farming where protection of the land against fresh-water or saline flooding is not practical. Changing to more salt-tolerant crops or cultivars in salt-affected areas. Selection of cultivars for higher yield under deep flooding or brackish water conditions.

Adapting to climatic variability The main problems to be considered are high-rainfall events and drought. Soil management measures to mitigate their effects in dryland soils should aim to maintain a complete soil cover, minimizing rainfall impact and non-beneficial evaporation; assure good macroporosity, infiltration and aeration; promote deep root penetration. Such measures include mulching, use of cover crops, relay cropping to improve cover and bioporosity; and subsoil liming where needed and appropriate deep ferlilization, especially with P, to increase rooting depth. Additionally, water harvesting methods and terracing and other recognized water conservation methods would increase amounts of water available to crops. Response farming techniques including, for example, late topdressing in amounts will capture the benefits of years with good rainfall or minimize costs in poor rainfall years. Finally, it should be mentioned that a rise in atmospheric C02 can have a positive effect on plant growth through increased photosynthesis ("CO2 fertilization"), and on the water use efficiency of plants due to reduced stomata1 openings. Global temperature rise will moreover result in higher precipitation due to greatly increased evaporation over the oceans - which in places will significantly enlarge the amount of fresh water that can be used for irrigation purposes.

1

Chapter 1

OVERVIEW OF THE GREENHOUSE EFFECT Global change syndrome; general outlook H.W. Schapenseel and P . Becker-Heidmann Institute of Soil Science, University of Hamburg Allende-Platzz, D-2000 Hamburg 13. Federal Republic of Germany

ABSTRACT Accumulation of cosmic dust and planetesimals was most likely the mechanism that created our planet. Due to dominance of hydrogen, the extruded gases produced a primordial reducing atmosphere. enriched with methane and ammonia. Then, after a slow start, continued oxidation with oxygen, released from photolysis of water, and the later development of life from photosynthesis caused the atmosphere to become dominated by COz, water vapor and N. The two former components were able to trap IR radiation and to produce a warming greenhouse effect of 33"C, shifting the surface temperature to +15"C. Oxygen from photosynthesis (at present yearly c a 330 bil t from terrestrial photosynthesis) was used over at least 2 billion years, for sustaining respiration of the various facets of life and for iron oxidation in marine and terrestrial sediments. During the last billion years oxygen began to enrich in the atmosphere, parallel to reducing CO2 concentration, due to its consumption by photosynthesis, chemical weathering and the carbonate precipitating pumping effect of the oceans. C02 replenishment occurs via volcanism and release from subduction zones. The faster biochemical cycle of smaller pool size (organic matter production, respiration, humification, kerogene formation, and biotic-abiotic-photochemical organic matter turnover) and over longer geological periods especially the slow but very large geochemical cycle (exchange of carbon between atmosphere, ocean, biosphere, and sediments), are decisive for CO2 concentration and its contribution to temperature. Some features of the biochemical cycle against the background of climate changes, including those due to Pangaea/Gondwana shifting, are discussed. Life is on a carbon trip. Wasteful consumption of fossil C based fuel, due to rising living standard and population explosion in conjunction with increasing release of greenhouse active (radiatively active) gases - which are fingerprinted - threatens to exert climate changes detrimental to our life conditions and civilization. Arguments to characterize the situation are assessed, also those expressing potential advantages of increasing C02 concentration for crop yields and expansion of the farmland area, doomed to shrinking at the present level of population explosion. The need for a change from the carbon trip to a mixed carbon - hydrogen trip is evident.

INTRODUCTION Environmental consciousness, especially watchfulness with focus on all anthropogenic activities causing pollution, tend to deviate our attention from tf-e dominant natural processes, underlying the whole web of contributing factors,

2

H.W. Schorpenseeiand P.Beckr-Heidmann

actions and feed-back systems in our unique earthly environment. A predictive analysis of the possible effects of a global climate change on soil processes and land degradation should be preceded by a short review of the scene as it existed, before a steadily growing human population created the syndrome of changing climate, basic to our worries. Its background is population explosion in conjunction with carbon-based energy sources and technologies, accompanied by steadily increasing release of nitrogen oxides (N20,NO), that absorb IRradiation or consume ozone in the stratosphere and produce ozone in the troposphere, as well as by increased infrared radiation trapping and stratospheric ozone destroying CFCs (chloro-flouro carbon compounds). These compounds, C02, CH4, N 2 0 , 0 3 , CFCs are expected by the majority of atmospheric chemists to lead to a further indirect temperature increase at the earth surface and decrease in the stratosphere. The estimated temperature rise of 3 to 5°C in the next 50 to 100 years may cause an eustatic sea level rise of 0.7 to 3 m due to water expansion and melting of polar ice masses. How did it all dcvelop and finally become a problem ?

OVERVIEW Basic facts, relating to this question in a nutshell expose the following tableau: Within the sun 700 mil t of hydrogen are fused per second into helium, i.e., ca 4.3 mil t of solar mass are converted into radiation energy, equal to 1.2 x 1015t per year (Wunderlich 1968). From the total solar mass of 2.2 x t , ca 1/40,000 has so far been consumed. The share of the solar radiation hitting the earth, the solar constant, amounts to 2.0 k 0.04 cm-2min-1.This solar radiation with wave lengths of less than 3,000 is absorbed in the ozone layer. Besides, light waves of 3,000 - 20,000A, till near-IR and radiowaves of 1 - lo3 cm wave length enter the atmosphere. The energy invested in the sun radiation is the origin and source of all important features of climate and environment, such as temperature, wind, clouds, precipitation and autotrophic organic matter production. The fact that the earth possesses an atmosphere, is taken as indication, that the origin of the earth is unlikely the result of a cosmic catastrophe, e.g. a collision of the sun with another cosmic body. I may have been formed however, by contracting dust and planetesimals with gaseous inclusions, giving rise after its compaction to extrusion of gases. Provided the gravitational forces are strong enough, those gases will be retained by the planet to form its atmosphere. The very light elements, such as H, He and Ne dissipated into space. This is revealed by comparison of the remaining atmospheric concentration with the share of these elements of the matter in the universe. Estimates are, that about 1 of 50 bil original Ne atoms in the primordial gas cloud is still left; He of the atmosphere is held 10 be almosr entirely radiogenic. The very wasteful atmospheric H, the major

Ilistorical overview of the greenhouse effect

3

cosmic element, may be representing 1 out of 5 mil H atoms in the original dust cloud; the even more reactive 0 about 1 of former 6 atoms; the less reactivc N about 1 of 800,000 N atoms (Asimov 1981). The high cosmic excess of the element H, also early earth, led to an initially reducing atmosphere of chiefly methane (CH4, carbon plus hydrogen) and Ammonia (NH3, nitrogen plus hydrogen). Depending on the amount of oxygen available, water (H20, oxygen plus hydrogen) was formed, which howcvcr was progressively precipitated and collected in depressions and marine basins together with the water vapor emitted by volcanic exhalations, thus leavingCH4, N H 3 , and water vapor as dominating atmospheric gases (Urey's work). Photodissociation of water (H2O + h.v = H+ + OH- ) led to slow oxidation of methane and ammonia into C02 and N2, producing an N2 and C02 atmosphere. With progressing integration of N2 into nitrates, C02 gained dominance till its rising conccntration increasingly blocked the photodissociation of water. Furthcrmorc, ozone formation from free 0 2 in the higher atmosphere absorbcd the UV-radiation and prevcntcd its penetration into the lower atmosphere and action of photolysis. As a rcsult, a stable C02 dominated atmosphere came into existcncc (scc also Habcr 1965). The high C02 concentration could have strongly promoted the greenhouse effect. Due to a rising temperature, water evaporation would have bcen further enhanced, with its additional promotion of the GHE and atmosphcric tcmpcrature rise until a hot earth would have emerged, envclopcd by a water vapor cloud and C02 dominated atmosphere. (For comparison, planct Venus built up a hot and stable COz atmosphere of ca 450" C). But planct earth took a completely different turn in the development of life, probably already slowly bcginning under the reduced CH4/NH3 atmosphere, whcre NH3 was decomposed, releasing N2 into thc aunosphcre, whilc excessive C 0 2 precipitated with Ca, Mg or Fc, which were dissolved by weathering procedures (without oxygen participation) in thc marine basins (not the least enclosed in phytoplancton). Thus, only a moderate GHE occurred due to water vapor and C02 built up, increasing carth's mean temperature by 33" C from - 1 8" C to +15" C (Arrhcnius 1896). This is a temperature level suitable for the liquid state of water and thc colloidal state within living organisms. Other conditions supported the sustainability of life as well, such as: 1 ) the Van Allen belt (the magnetosphere); 2) the shield against cosmic radiation; 3) similarly, the ozone shield for absorption of UV light; 4) the earth magnctic field, although changing its polarity rcpcatcdly in the course of earth history, giving furlher radiation protcction and orientation; 5 ) the high altitude cirrus clouds, heating the atmosphere; 6) the lower altitudc vapor saturated clouds exerting a cooling effect; 7) the inclined earth axis (23.5"), producing annual seasons of climate; 8) the earth rotation, causative for day and night change for regeneration of the metabolisms; and 9) the atmospheric currents for transportation of moisture, heat and dissipation of products of pollution.

4

H.W. Scharpenseel and P . Becker-Heidmann

Soil and humus formation, CO, and 0, trends

&M 0 Quaternary Ter t iar y

.

70

Cretaceous plus humus of Angiosperms

. 140

Jurassic

.

180

-

225

Triassic Permian . 275

first complete soil cover Carbonian plus humus of Gymnosperms

345

'

Devonian . 400

.

Silurian 440 Ordovician

. 490

Cambrian '

580

C02 trend decreasing 0 trend increasing

Fig. 1.1

Soil and humusformation in earth is history

Life on earth was slowly turning the N/C02 atmosphere into a N/O atmosphere. The 0 2 concentration increased by almost one order of magnitude (Fig. 1.1) during the last ca 600 million years, that is since the beginning of terrestrial plant growth (the Phanerozoic; flowering plants, the angiosperms since just ca 150 million years) . This is mainly the effect of oxygen release during the photosynthesis process of organic matter production from CO2, and 0 and H from H20, previously dissociated by sunlight energy (light reaction of photosynthesis). The slow development of our oxygen rich atmosphere after exhaustion of the enormous demand for marine and terrestrial Fe-oxidation is shown in Fig. 1.2. Meanwhile the total free oxygen pool in the atmosphere and dissolved in the oceans is estimated to amount to 1.3 x 1021 g. Most of the biologically produced oxygen, at present ca 3.3 x 1017g of 0 2 per year (corresponding with ca 1.2 x lOI7 g of C per year by terrestrial photosynthesis) plus ca 1.3 x 1017 g of 0 2 per year (corresponding with ca 5 x 10l6 g of C consumed by marine photosynthesis), is bound in the earth crust as metal oxide, sulfate, silicate, and carbonate and represents about 6 x g of oxygen (Chem. Ind. 1987). Considering the high reactivity of oxygen, its existence as free 02-gas in the atmosphere is possibly only the result of constant new 0 2 production and

5

Historical overview of the greenhouse effect

addition. Without replenishment by photosynthesis, our atmospheric oxygen may

be consumed in about 3000 years due to oxidation processes in the earth crust (Haber 1965, 1971). But also atmospheric C02 needs replacement outside the biochemical cycle of photosynthesis and respiration due to consumption of C02 by silicate weathering, where from 2 molecules of C02 involved in the bicarbonate reaction always only one is returned to the atmosphere, whereas the second one is precipitated as carbonate, which would use up the present atmospheric CO2-pool in about 10.000 years (2 COz + H 2 0 + CaSi03 t Ca2' + 2 HCO3- + Si02) (Berner and Lasaga 1990). Similarly the gas exchange pump of the oceans induce C02 intake to replace C02 of precipitated carbonate. After longtime involvement of the geochemical cycle, these carbonates may under high pressure and temperature be subjected to metamorphic processes and eventually release the C02 through volcanism or expulsion by subduction zones.

4 B I L L I O N Y E A R S AGO

3 TO

Primordial m l m o ~ p h e r e

Formallon 01

co2

S o l u l l o n O d 0, I " w a t e r

S O l U l l O " I " "11*,

4

B I L L I O N Y E A R S AGO

0, I n

water

CA

2 B I L L I O N Y E A R S AGO

Formillon 01

0, I n w a t e r

T B ~ l d S l I 1 ~l 1l x a l l o n 01

0,

......~

CA

0 5 B I L L I O N Y E A R S AGO

T e r r e $ t r i a l o r m ~ t i a n0 1

0, R I S E

0,

I N ATMOSPHERE

IIIIII 0 2

0 A C T E A IA

W E AT H E R I N G

CaCO, A

I

Fig. I .2

Fe*O 3

-

MUD

,

History of oxygen formation and dynamics

This gas exchange process would exhaust atmospheric C02 in about 300,000 years (Bemer and Lasaga 1990). Planet earth, its atmosphere and biosphere become vitally predetermined by the consequences of the vast but slow geochemical and faster biochemical cycle, which however represents a much smaller carbon compartment.

6

H.W. Scharpenseel and P . Becker-Heidmann

FEATURES OF THE BIOCHEMICAL CYCLE Only carbonaceous materials, produced by abiotic processes till ca 3 billion years ago are exclusively geochemical (Rankama 1948). All others also have a biochemical component. The organic matter residues of living organisms are preserved almost exclusively in aquatic sediments as carbonates or in contact with shales and clay minerals. The latter as clay domains provide also the matrix for organo-mineralcomplexation of younger or even today's terrestrial organic matter (Aylmore and Quirk 1960; Theng and Scharpenseel 1975; Theng 1979) (Fig. 1.3). 70-

60-

50I

P

E" 40-

1

H m i c acid concentration (mg/rnl)

Fig. 1.3

Isotherms at 20°C f o r the adsorption of 14 C-humid acid by montmorillonite saturated with diflerent cations (Theng and Scharpenseel 1975)

Most of recent as well as ancient sediment's organic matter stcms from phytoplancton and bacteria (Bordowsky 1965; Murphy et al. 1966); this forms the major sink of organic C and of CO2. Sediments, produced by precipitation, such as evaporites and carbonates, rarely contain large amounts of allochthonous organic matter. Detrital rocks like sandstone or shales engulf, usually diagenetically formed, relatively stable, secondary polymerized compounds, such as humic acids as oxidative or kerogene as reductive products (Tappan and Loeblich, ref. Welte 1963). Both together represent the major organic carbon sink

Historical overview of the greenhouse effect

7

at a level of ca 3.6 x loi5t, compared with petrol or coal with stocks of the order of 10l2- l O I 3 t only (Degens 1967). Finally, climatic and tectonic events have a great influence on the organic compound production and preservation, e.g. bituminous sequences often seem to be related to orogenic phases or epimgenic oscillations with corresponding eustatic lifts, trans- and regressions (Bitterli 1963). The C02 and 0 2 balance in ocean water and in the atmosphere changed with the organic matter production in the course of earth history, with its carbonate precipitation as well as the emergence of higher plants and animals (Tappan and Loeblich, ref. Welte 1969). During sediment diagenesis organic matter supports the microbial metabolism and it exerts influence on chemical reactions through pH and Eh changes, especially those involving C02 - SO2 CH4. After microbial activity terminates, chemical interactions with the inorganic matrix occur, leading to complexation and chelation, and reactive chemical groups like carboxyl, hydroxyl, and amino groups are released. The origin of life is believed to have occurred in an aquatic milieu, which provides more continuity due to less zoographic isolation than terrestrial life which developed later. According to Schilder (1956), 63 among 68 animal classes live in a marine environment, but, due to geographic/ecologic isolation in terrestrial environments, the differentiation in species is more pronounced, comprising ca 83 % of all known animal species. The transition from marine to land based life, that contributed most to an oxygenation of the atmosphere and that became a major sink for C02 excess in the biochemical cycle, must have begun preferably in marginal shelf fringes of the epipelagial, the euphotic zone, mostly under tidal influence, such as in marshy or mangrove environments and in shallow littorals. Since organisms sustaining the biochemical cycle by photosynthesis (C02 consumption) and respiration (C02 release) survive in evolutionary processes, due to their capability to adopt flexibly to environmental changes and to find ecological niches or refuges, the eco/geographical boundaries governing the distribution of species are constantly shifting due to climate changes as well as to tectonic effects. This applies particularly to stenothermic animal species requiring a narrow temperature regime. In biomes, animal life usually is more flexible and stretches further into critical environments than growth of vegetation (Wurmbach 1971). The spread of terrestrial life has also been largely influenced by plate tectonics in conjunction with the dissolution of Pangaea, the Gondwana and Europe-Angara drifting with the corresponding climate changes. In this context, the most striking process of soil degradation, a land fossilization and lateritic cuirass formation, occurred due to changed erosion and drainage patterns. These cuirasscs represent an extreme form of humid tropical weathering under the changing climate of the floating Gondwana subcontinents (mainly Jurassic to Oligocene) (Valeton 1984). A sporadic soil blanket (pers. comm. Dudal 1990)

8

H . W . Scharpenseel and P . Becker-Heidmann

existed probably since the end of thc Silurian, a soil continuum came into being some time after the cold spell during the Permian.

IMPACTS OF GEOCHEMICAL AND BIOCHEMICAL CARBON CYCLE A comparison of the magnitudes of the compartments of carbon in the (bio)geochcmical (exchange of C between sediments, atmosphere, biosphere and ocean) and biochemical (organic matter production, respiration, turnover) cycle shows the dominance in pool size of the former, which over geological time pcriods is all decisive with its enormous buffer capacity (Table 1.1). Shortcuts in the biochemical cycle and short-term excessive inputs may produce a flicker, strong enough though for consequences on the GHE, the earth temperature, the prccipitation and the circulation, which may damage or even exterminate species causativc of the disturbance. Thus, in our short historical span we can not rely on the buffcr capacity of the large but slow gcochcmical cycle to neutralize the consequcnces of our mistakes. Table 1.1

Comparison of carbon pool sizes in biochemical and geochemical cycle*

Componcnts CaC03 in sediments CaMg (CO& in sediments Organic sediments (kerogcne) HCOf and C032- dissolved in sea Fossil fuel (coal, gas, oil) Dead soil biomass (humus) C02 in atmosphere Living biomass (plant, animals) * Data slightly modified from Berner and

C-amount in 10'2 t of c 35,000 25,000 15,000

42 5

2 0.72 0.56 Lasaga 1990.

LIFE ON EARTH, A CARBON TRIP Lifc on earth is on a carbon trip. We must get aware of the need for changing course in time to avoid an erratic trip back into chaos, from which evolution made us ascend. Looking at the biochemical carbon cycle (Fig.1.4). about 115 to 120 x l O I 5 g of C are turned over annually in the terrestrial ecosystems by photosynthesis and inversely by respiration through the bio- and pedosphere. Worrysome is the surplus due to respiration of ca 1.5 x lOI5 g of C from annual land clearing (slash and bum) as well as the 5.5 x l O I 5 g of C from

9

Historical overview of the greenhouse effecr

combustion of fossil fuel, which are adding up at a ca 50% rate to the 720 x 10'5 g of C in the atmospheric carbon pool. However, recent results of Esser (1990) suggest that the C-sink due to C02 fertilization is already overcompenzating the C02 source from forest clearing (ca. 1.5 Gt C y - l from 10 to 15 mil. ha of clearing a year). Principally, carbon oxidation products, C02 and CH4, although being greenhouse active trace gases, are minor in importance compared with water vapor (1:5) in generating the temperature rising greenhouse effect of 33"C, lifting the surface temperature from -18°C to +15"C. According to Ramanathan (1989) as well as Raval and Ramanathan (1989), the total natural 33°C greenhouse effect generated by water vapor plus C02 equals 155 Wm-2 (ca 145 W at clear sky, 180 W at cloudy sky). Doubling of the C 0 2 concentration would add ca 4 Wm-2 only; human activities so far have enhanced the GHE of the atmosphere by ca 1.5 % only.

CARBON CYCLE

l

120

I

n

A

2'

B'OSPHERE + PEDOSPHERE

i...ca.?3P..!!.!v!na.! .......

$.

C6 2000 (dead)

ca 100

ca 100

ca 60

01

5"

,

DEPOSITS 5000 i ......COAL, . ...........OIL, .........GAS ............................................ SEDIMENTS/LITHOSPHERE 66 000 000 o,2 from l o r e s t a n d sol1 d e g r a d a t i o n

OCEAN SURFACE WATER c a 700 ................................................................... DEEPER OCEAN WATER c a 37000

l r o m c o a l , 011.

gas combusllon

a m o u n t s ~n G I ( I O " ~ ) , f l u x e s i n G I I ~

Fig. 1.4

Carbon compartments in biochemical and geochemical cycle

But the increments of C02, CH4, as well as the other radiatively active trace gases, such as the CFCs, N20 and tropospheric ozone, are wholly or to greatest extent the product of anthropogenic activities. Human civilization was in its earlier phases sustained by the direct products of photosynthetic C reduction and

10

H.W. Schorpenseel and P.Becker-Heidmann

photolysis of water (Calvin 1962). Since beginning craftsmanship and industrialization it progressively built up a dependency on minable and consumable forms of organic matter, produced mainly throughout the last 350 million years, such as coal, kerogene/petrol and methane. While the first scientists, expressing the vision of an expanding greenhouse effect with temperature and in consequence an eustatic sea level rise (Arrhenius 1896; Callendar 1938; Flohn 1941, cited in Lausch 1989) found little attention or enthusiastic response, at present an almost overheated scientific climate and preferably apocalyptic predictions thtcaten to misinterpret the real facts, needs and situative, curative potentials. We also should not ignore the main trend of decreasing C02 concentration over most of the geochemical and biochemical evolutionary phases of the past, with at least five times the present level still during Tertiary times and about 200 ppm after fluctuations during glacial and interglacial periods at the end of last Wisconsin-, Wurmian-, Weichselian glacial. Since even the present 350 ppm are at the lower end of the plant physiologically suitable C02 range, could there have been an acute C02 deficiency developing, and the C02 rise since begin of industrialization in reality being beneficial in our endeavour to increase food production for the growing world population ? The temperature rise of 0.7"C since 1860 and a sea level rise of about 17 cm are acknowledged by most climate research units, but a highly rated group of scientists from the G.C. Marshall Institute in Washington, D.C. draws attention to a better correlation between the observed trend in temperature and the solar activity/sun spot activity than the more commonly considered increasing C02 concentration (ref. Economist Vol. 313, No. 7633, 1989). R.S. Lindzen from M.I.T. Boston a.0. reported to be expecting nominal temperature increases only. Other critical observations focus in many different ways on the overestimation of the C02 concentration change and neglect of the importance of changes in atmospheric moisture level, which, as mentioned before, is undisputed the major promoter of the basic greenhouse effect. While the determination to reduce the release of the other greenhouse active trace gases, such as CH4, N 2 0 , CFCs, tropospheric ozone (see overview in Table 1.2) is apparently worldwide accepted, a production oriented pedologist/plant nutritionist can argue, whether the rise of C02, which is expected to enhance especially C3 photosynthetic efficiency and to improve the water economy of plants due to closing of stornatal aperture with increasing C02 concentration, which even under rising temperature and water evaporation may also conserve precious water (Schleser and Kirstein 1990), would be really all that bad. Fig. 1.5 shows the wide range of C02 concentration versus light intensity capable of increasing photosynthetic efficiency. The relative effect of increasing C02 on the major cultivated plants as well as the most obnoxious weeds requires thorough attention.

Historical overview of the greenhouse effect

Table 1.2

Greenhouse active and other (polluting)trace gases, basic pool sizes

Trace gases

% share of antrogenic

GHE co2 CH4 CFCs 0 3 (tropospheric)

co2

50 19 17 8 6 Residence time Concentration 100 y 350 ppm

co

1 - 6 months

CH4 CFCs (ClOx -radicals)

10 Y 50 - 150 y

100-150ppb N 40-80 ppb S (N, S. hemisphere) 1.7 ppm 0.2 - 0.3 ppb

N20

170 y few days only

0.31 ppm 0 - 100 ppb

N20

Components

NOx (NO, NO21 0 3 (stratospheric) 0 3 (tropospheric)

OH (atmospheric)

so2 Atmospheric C-pool Photosynthesis (terrestr.) Soil organic matter pool Ocean C-cycle flux

Fossil fuel C

11

few seconds only

10 ppm (35 km) 0.02 ppm (0.1 ppm max.) < O.oooO1 ppb

GHE-rising potential of trace gases rel. to c02 1 32 14 - 17,000 2000 150 Increase per year (1800->280, 1950>310); ca. 0.5% variable

1.1 % (18 ppb) 5% 4% 0.3 % (ca 1 ppb) 0.2 - 0.3 % (stratospheric)

0.5 %

50 ppb (max.) 740 bil t 115 bil t (ca 1/2 respired, ca. 1/2 Wmpo@ 1.8 x 10l2 t

38.5 x 10l2 t (3x 109tpery precipitated as carbonate) 5 - 10 x 1012 t

The vegetation belt, which may be particularly affected by increasing heat and drought, the inigation-dependent and responsive sub-tropics, may hopehlly

12

H.W. Scharpemeel and P . Becker-Heidmann

replenish water deficits by stepped up sea water distillation, which may become technically feasible in nearer future. Besides, the leading Russian climatologist from Leningrad, M.I. Budyko (ref. Spiegel 1990) expects even for the (semi)arid tropics more rainfall and a vegetation cover similar to that of the Pliocene. Increase of temperature and precipitation in the boreal belt could bear grave consequences due to enhancement of organic matter decomposition and consequent COs release, but may also hold unforeseeable opportunities to increase agricultural production in vast areas of so far low productive boreal lands, which may even become important carbon sinks (see for more details Chapters 16 and 17). The catalogue of pros and contras of effects of climate change is certain to expand. I

Net photosynthetic rate for different light intensities and C 0 2 concentrations Rp (mg C O P / q m

I Fig. 1.5

s)

I light Intens.. Rp net photosynth. r a t e

Influence of light intensity and CO;, on photosynthesis. (modified from Mengel and Kirkby 1979,acc. to Warren-Wilson 1969)

We should be critically alert towards the climate change syndrome, but as critically open to the more constructive and may be even optimistic arguments before destroying productive structures without having replacements in panic, through poorly conceived legal actions as well as preventive or curative measures, . The most important reaction, though, should be to prepare t k follow-up phase of a mixed hydrogen plus carbon trip. In geological times the geochemical cycle is absolutely dominating and decisive for CO2 concentration and its influence on temperature. However, the anthropogenic shortcut of the

H k l o r i c a l overview of the greenhouse effect

13

biochemical cycle that we practice with wasteful consumption of fossil fuel reserves, with rising population and living standards, may suffer under t k unsentimental regime of natural processes, such as temperature and sea-level rise, to destroy our species and civilization (only a fast forgotten flicker in earth history), We must therefore concentrate all our efforts on the development of the photovoltaic hydrogen technology for replacement of C-based fossil fuels, as well as on the methodology for use of sun energy in desalination of sea and brackish ground water to stabilize life and productivity in the (semi)arid lands.

REFERENCES Asimov, 1. (1984). Asimov’s new guide to science. Basic Books Publ., New York. Aylmore, L.A.G. and J.P. Quirk (1960). Domain or turbostratic structure of clays. Nature 187. 1046- 1048. Berner, R.A. and A.J. Lasaga (1990). Simulation des geochemischen Kreislaufs. Spektrum der Wissenschaft 5, 56. Bitterli, P. (1963). Aspects of the genesis of bituminous rock sequences. Geol. Mijnbouw. 42, 183-201. Bordovskij, O.K. (1965). Accumulation and transformation of organic substances in marine sediments. Marine Geology 3, 3-1 14. Calvin, M. and J.A. Bassham (1962). The photosynthesis of carbon compounds, W.A. Benjamin Inc., New York. Degens, E.T. (1967). Diagenesis of organic matter. In: Diagenesis in sediments, Larsen, G. and G.V. Chilingar, Eds., Elsevier, Amsterdam, chap. 7. Dudal, R. (1990). Global Soil Change, report of an IIASA-ISSS-UNEP Task Force Meet. on the Role of Soil Global Change, Chap. 3 (in print). Esser, G. (1990). Modeling global terrestrial sources and sinks of COz with special reference to soil organic matter. In: Soils and the greenhouse effect, A.F. Bouwman (Ed.) (1990). John Wiley and Sons, Chichester. Fond der Chemischen Industrie, Umweltbereich Luft (1987). Vol 22, p. 15. Haber. H. (1965. 1971). Die Entwicklungsgeschichte der Erde, Deutsche Verlags Anstalt, Stuttgart, p. 79, p. 208. Lausch, E. (1989). Treibhaus Erde. CEO (Gruner u. Jahr, Hamburg) 37,46-49. Mengel, K. and E.A. Kirkby (1979). Principles of Plant Nutrition. Int. Potash Institute, Bern, Switzerland, 233. Murphy, M, B. Nagy, G. Rouser, and G. Kritchevsky (1965). Analysis of sulphur compounds in lipid extracts from the Orguiel meteorite. J. A. Oil Chem. Soc. 43, 189.196. Ramanathan, V. (1989). Spurengase, Treibhauseffekt und weltweite Erwrmung, In: Das Ende des blauen Planeten ? Crutzen and Muller (Eds) Beck, Federal Republic of Germany, 6576. Rankama, K. (1948). New evidence of the origin of pre Cambrian carbon. Bull. Geol. Soc. Am. 59, 389-416. Raval. A. and V. Ramanathan (1989). Observational determination of the greenhouse effect. Nature 342, 758-761. Schilder, F.A. (1956). Lehrbuch der Allgemeinen Zoogeographie, Jena, German Democratic Republic. Schleser, G. and W. Kirstein (1989). Der Treibhauseffekt. Ursachen und Konsequenzen fur

14

H.W. Scharpenseel and P . Becker-Heidmonn

Klima und Biosphre. Seminar Technik und Gesellschaft, KFA Jlich, Federal Republic of Germany (preprint). Spiegel (1990). M.I. Budyko - Interview. Der Spiegel, Hamburg, 1, 143. Theng, B.K.G. (1979). Formation and Properties of Clay Polymer Complexes, part 3, Chap. 12, 283-314. Theng. B.K.G. and H.W. Scharpenseel (1975). The adsorption of 14-C labelled humic acid by montmorillonite. Proc. Internatl. Clay Conf., Mexico City, 643-653. Valeton, I. (1983). Klimaperioden, lateritische Verwitterung und ihr Abbild in den synchronen Sedimentationsrumen. Z. Dtsche Geol. Ges. 134, 2. Warren-Wilson, I. (1969) Maximum yield potential. In: Transition from extensive to intensive agriculture with fertilizers, Proc. 7th Coll. Intern. Potash Institute, Bern, Switzerland, 34-56. Welte, D.H. (1969) Organic matter in sediments. In: Organic Chemistry, Springer, Berlin, 262-264. Wunderlich, H.G. (1968). Einfhrung in dic Geologie, Vol. 1, Exogene Dynamik. Bibliographisches Institut, Mannheim. Federal Republic of Germany. Wurmbach, U. (1971). Zoologie. Vol. 2, G. Fischer Verlag, Stuttgart.

15

Chapter 2

INPUTS TO CLIMATIC CHANGE BY SOIL AND AGRICULTURE RELATED ACTIVITIES Present status and possible future trends A.F. Bouwman and W.G. Sombroek International Soil Reference and Information Centre P.O. Box 353, 6700 AJ Wageningen, the Netherlands

ABSTRACT The most important soil borne and land use related greenhouse gases are carbon dioxide

(COz), methane (CHd and nitrous oxide (N20). The present annual increase of atmospheric CO2 is 0.5%. The total emission of C 0 2 is 6.5 to 7.5 Gt C y-'. Fossil fuel combustion contributes 5.7 Gt C y-*. The present global rate of deforestation of 10 to 20 million ha y-l causes an emission of 1 to 2 Gt C y-' including the release of C 0 2 from soil organic matter oxidation. There is uncertainty about the sinks of C02. The oceanic uptake is less than 1 Gt C y-1, while the atmospheric accumulation accounts for approximately 3.5 Gt C. Increasing net primary production and other terrestrial sinks balance the budget. The emissions from fossil fuel use will probably increase, and projections of efficiency and magnitude of future energy use are rather uncertain. Deforestation will continue in the coming decades to satisfy the growing demand for agricultural land. The atmospheric concentration of CH4 is rising at a rate of 1% y-'. The major part of this fast increment is caused by increasing emissions, while a minor part can be attributed to decreasing destruction in the atmosphere. The total annual emission from all sources is 400 to 640 Tg CH4. Biotic sources make up about 80% of the total annual emission, the rest is from fossil sources. The biotic sources of CH4 are: wetland rice cultivation (20%of the total source), natural wetlands (20%). ruminating animals (15%). landfills (10%). oceans and lakes (5%) and biomass burning (15%). The contribution by termites is very uncertain. Most sources are increasing at present. Nitrous oxide is increasing at a rate of 0.2 to 0.3% and its sources are almost exclusively biogenic. Nitrous oxide is inert in the troposphere, but it destroys stratospheric ozone ( 0 3 ) . The causes of the increase in atmospheric N 2 0 are not well known. Fossil fuel combustion is a minor source of N20. Increasing use of N-fertilizers in agriculture is a growing source of N2O. The emissions from natural ecosystems are not well known at present .

INTRODUCTION Atmospheric gaseous constituents which are able to absorb thermal radiation and thus contributc to Lhe trapping of heat in the atmosphere are water vapor (H,O), carbon dioxidc (CO,), methane (CH,), nitrous oxide (N,O), ammonia (NH,), halocarbons (the most important are the chlorofluorocarbons, CFCs), and ozone ( 0 3 ) . Of Lhcse, CO;?, CH4, N 2 0 and NH3 are produced in natural and

16

A T . Bouwman and W . G . Sombroek

agroecosystems. To assemble a complete picture of the role of soils and land use in the greenhouse effect, it is necessary to attend to indirect effects of other gases as well. Indirect effects of a source change (i.e. soil change or land use change) are induced changes in a chemical species which affect a given valued atmospheric component through an intermediate influence on another chemical species. Atmospheric constituents with biotic sources having indirect effects on concentrations of the greenhouse gases listed above are carbon monoxide (CO), non-methane hydrocarbons (NMHC, isoprene, C5H8, and terpenes), nitric oxide (NO) and nitrogen dioxide (NO,; NO and NO, as a group are denoted by NO,), ammonia (NH3), oxides of sulphur (SO,), and organic S compounds. Ammonia has indirect effects as well: about 10% of atmospheric ammonia is oxidized to NO and NOz, which play a catalytic role in various photochemical reactions in which 03,CH,, CO and OH are involved. Carbon monoxide is oxidized to CO, thereby affecting many other atmospheric constituents, such as ozone (03). hydroxyl radicals (OH) and CH4. Non-methane hydrocarbons are oxidized by O3 to CO in the atmosphere.

Table 2.1 Residence time of the major greenhouse gases, their rise, atmospheric concentration, radiative adsorption potential and contribution to global warming Type

co2 co CH4 N20 03 CFCs'O

Residence time (y)

Annual

1985

rise (%)

concentration

0S3 0.6-1.07 15 0.256 2.02 3 .02

100' 0.24 8-123 100-2006 0.1-0.32 65-1102

Radiative Contribution absorption to global potential 9 warming 9 345 ppm3 1 50 90 ppbz n.a. n.a. 32 19 1.65 ppm7 300 ppb6 150 4 n.a.8 2,000 8 0.18-0.28 ppb > 10,000 15

-',

Total sources Q is 5-7 Gt C y and atmospheric burden B is 720 Gt. The following relation exists: B = Q x T (see Cicerone and Oremland 1988). The lifelime of C 0 2 must thus be f 100 y (strictly speaking the residence time of C 0 2 is shorter when exchanges between atmosphere and biosphere on the one hand and atmosphere and oceans are allowed for. In that case however, the true importance of C 0 2 is underestimated; Ramanathan et al. (1985) (data for 1980); Bolin (1986); The residence time of CO is not a single constant number, because spatial and temporal variations of sources and sinks are not identical. Total sources and average atmospheric concentration are not known accurately, bu1 accounting for the uncertainties Cicerone (1988) estimates the residence time of CO at 2-2.5 months; Cicerone and Oremland (1988); Crutzen and Graedel (1986); Bolle et al. (1986); 0, vanes from 25 ppb at surface to 7 0 ppb at 9 km (Ramanathan et al. 1985); Enquete Kommission des 11. Deutschen Bundestages "Vorsorge zum Schutz der Erdatmosphare" (1988); "Data presented are for the two major CFCs. Reprinted, with adaptations from Bouwman (1990). by permission from Wiley & Sons, Chichester.

D~

~, ....

-O.

. I . . I -

......

C"

I,

The role of water vapor will not be discussed here. By far the most important other atmospheric greenhouse gases from soil and agriculture related sources are C02, CH4 and N20. These will be discussed in more detail in this paper. We will discuss some chemical aspects of CO in the context of atmospheric chemical reactions of methane. For an overview of the sources and sinks of CO, the reader is referred to Cicerone (1988). A review of the other trace gases not discussed here is presented in Mooney et al. (1987). Due to their relatively long atmospheric residence times C02, CH4 and N20 have important effects on the radiative balance. Some of their characteristics are compared with a number of non-biogenic gases in Table 2.1. The information on the sources and sinks of C02, CH4 and N20 and the possible causes of their increase will be summarized below. In Table 2.3 the global sources and sinks of the three major gases discussed in this paper are summarized.

CARBON DIOXIDE Sources of CO,

The major sources of CO;! are fossil fuel combustion, gas flaring and cement manufacturing, which produced 5.3 Gt CO2-C y-1 in 1984 (Rotty 1987) (Gt = gigaton; 1 Gt = 1015 g), while the 1987 emission was 5.7 Gt C y - 1 (CDIAC 1989). The C02 emission from these sources will amount 2 to 20 Gt C y-l (Keepin et al. 1986), depending on hture efficiency of energy use and shifts to non-carbonaceous sources of energy. Past release of C02 from land use modifications has contributed significantly to the present atmospheric CO;? concentration. Today deforestation and increasing shifting cultivation are responsible for only 1-2 Gt C y-' (Table 2.2). The disagreement between the different suweys of deforestation shown in Table 2.2 is in the fate of fallow lands in shifting cultivation, which at present is being replaced by non-sustainable systems called "sedentary shifting cultivation" or "forest farming" (Houghton et al. 1985), whereby the forest is often not able to re-establish on fallow lands due to soil nutrient depletion. Data presented by INPE (1989) suggest that the deforestation in the Brazilian Amazon region is much slower than the rates presented by the authors quoted in Table 2.2. Part of the CO;?sources shown in Table 2.2 stems from soil organic matter oxidation. Detwiler and Hall (1988) estimated that this soil C 0 2 emission is 0.1 10.25 Gt C y-I for the tropics. Armentano and Menges (1986) estimated that the net release from wetlands as a consequence of drainage, calculated as the sum of loss of sink strength and gain of source strength as a result of wetland drainage, is 0.15 to 0.1 8 Gt C y-l. The role of weathering on fluxes of C02, which often is totally neglected, is discussed in Van Breemen and Feijtel (1990).

18

A P . Bouwman and W.G.Sombroek

Sinks of C O ,

The release of C02 from fossil sources and from land use modifications is being balanced by atmospheric accumulation uptake of C02 in the world's oceans and by the terrestrial biosphere. The increase of 0.5 % y - l in the C 0 2 concentration in the atmosphere accounts for about 3.5 Gt C y-l. The uptake of C02 by the oceans may be less than 1 Gt y - l (Tans et al. 1990). Increased atmospheric C02 concentrations stimulate net primary productivity considerably, and moreover plants will utilize water more efficiently (see Sombroek, Chapter 13 of this volume, for some details). Modelling results (Esser 1987, 1990) suggest that the C02 fertilizing effect is responsible for a growth in net primary productivity which since around 1970 would overcompensate the losses from clearings. The terrestrial biosphere was a small sink of CO;! of about 0.3 (Goudriaan and Ketner 1984) to 0.5 Gt C y-l (Esser 1990) in the early 1980s. The world's wetlands may sequester carbon, while formation of caliche (calcium carbonates) in desert areas is proposed to be a small but important sink. An additional sink of carbon may be the formation of charcoal or inert C , one of the products of biomass burning, which may even amount up to 0.5 - 1.7 Gt C (Seiler and Crutzen 1980). Table 2.2 Annual release of COZ for the early 1980sfrom terrestrial biota including soils caused by deforestation and increasing areas of shifting cultivation, demonstrating uncertainties in estimates. Release rates in Gt C y-I; areas cleared in 106 hay"

2 References

World

Tropics

Houghton et at. (1987)

1.0 - 2.6

0.9 -2.5

Detwiler and Hall (1988)

0.3 - 1.7

0.4

- 1.6

Area of permanent clearing

in the tropics low: 6.0 a high:15.2 low: 3.0 medium: 10.6 high: 15.2

F A 0 (1983); Myers in Houghton et al. (1985); Seiler and Crutzen (1980); Lanly (1982); ' Detwiler and Hall (1988) concluded from results in other studies that temperate forests contributed -0.1 to +0.1 Gt C y - ' in 1980; their own analysis considers only tropical deforestation. B o ~ hlow and high figures of forest biomass were used in the presented estimates. The low estimate for the tropics is based on forest volume figures: 90 t C h a - ' for closed and open forests, respectively (Brown and Lug0 (1984). The other estimate which is based on destructive sampling yields higher values of between 164 and 40 I ha-' for different types of tropical and subtropical forests (Brown and Lugo, 1982). Reprinted with adaptaoons form Bouwman (1990b). by permission of Butterwonh Scientific Ltd, UK.

Inputs to climatic chonge by soil and agriculture related activities

Table2.3

19

Global emissions and sinks of major greenhouse gases. G t = gigaton, 1 Gt = g, Tg = teragram, I Tg = 10l2 .

co2 Global annual 6.5- 7.5 Gt C emission* % Biotic 20-30

CH4

N20

400-640 Tg CHq

11-17 Tg N

70-90

90-100

Sources*

fossil fuel use (5.7 Gt) deforestation and shifting cultivation (1-2 Gt)

rice paddies (60-140 Tg) cultivated soils (S Tg?) wetlands (40-160 Tg) natural soils (?) ruminants (65-100 Tg) fossil fuel (?) termites (10-100 Tg) landfill sites (30-70Tg) oceansflakes(15-35) biomass burning (50100 Tg) fossil (50-95 Tg), coal mining/gas exploitation

Sinks*

atmospheric accumulation (3.5Gt) Oceans (<1 Gt) biosphere (? Gt)

atmospheric accumu- atmospheric accumulation (50Tg) lation (2.8) soil oxidation (32 Tg) atmospheric chemistry atmospheric chemistry (10.5) (300-650 Tg) soils (?)

charcoal formation (?)

* The total budget or total sources is not necessarily the s u m of the indicated individual sources. The total of sources can be derived independently from the atmospheric burden, residence time and measured increase (see also Table 2.1) or from models of atmospheric chemical processes. Reprinted with adaptations from Bouwman(l990b), by permission from Butterworth Scientific Ltd, UK. Future developments Possible future C 0 2 concentrations were analyzed using the Osnabriick Biosphere Model which includes C 0 2 fertilization (Esser 1990). The results suggest that depending on different energy scenarios (0.1% and 0.5% increase annually) and progressive clearing of 50% to 100% of the area of natural vegetation in the year 2400, the 2 x pre-industrial C 0 2 level (2 x k 285 ppmv = k 570 ppmv) will be reached in the year 209 1 (worst case) or 2 181 (best case). A similar analysis was camed out by Swart and Rotmans (1990), who used the dynamic biosphere model developed by Goudriaan and Ketner (1984). Based on a slow increase of 0.7% y-l of the energy use from 5.4 in 1985 to 12.5 Gt C y-l in the year 2100, the difference in 2100 between total deforestation and gradually

20

A.F. Bouwman and W . G . Sombroek

decreasing or even reversed deforestation was 60 ppm, being only 10% of the projected 2x pre-industrial CO, levels. This illustrates the modest role of the terrestrial biosphere. Actively growing trees accumulate carbon. Hence, forest plantations can be implemented to absorb part of the annual atmospheric C 0 2 injection. Approximately 465 x lo6 ha (which is equal to about the total land area of Europe) of forest plantations would sequester the present 3Gt C0,-C surplus of sources over sinks, based on very high accumulation rates of 6240 kg C ha-ly-l in new forests (Sedjo 1989). The future demand for different kinds of wood and the feasibility of afforestation and the actual productivity of planted forests was considered by Wiersum and Ketner (1989). On the basis of literature data these authors estimated that in the year 2000 there will be a worldwide need for 55 lOOx lo6 ha of afforestation to cover fuelwood deficits, 20-40x106 ha for industrial wood needs and 127-190x106 ha for environmental protection. The latter plantations would occur mainly in waste lands and regions in process of desertification. Part of the environmental plantations would also serve for fuelwood or industrial wood production, so that the authors arrive at their final estimate of 1 6 0 ~ 1 ha 0 ~for long rotation forest plantations with fast growing species and 4Ox1O6ha of plantations for fuelwood (mainly subsistence). It was assumed that the long rotation plantations (environmental + industrial wood) have an effective period of 30 years during which carbon accumulation is 3000 kg y - l (note that this accumulation is much lower than the rate reported by Sedjo 1989). Fuel wood forests sequester 2000 kg C ha-ly-' during an effective period of 5-10 y. The 1 6 0 ~ 1 ha 0 ~of long rotation forest and 4Ox1O6 ha of new forest for fuelwood would sequester 0.5 and 0.1 Gt C y-l, respectively. In addition an assessment was made of the feasibility of high-productive wood energy plantations, with rotation periods of 8 years and carbon fixation of 4500 kg C y-'. These plantations would replace areas of cropland, primarily in the USA and Europe, that may be taken out of production in future. The projected 30x lo6 ha of highly productive energy wood forest plantations wouId substitute 0.12 Gt C y-l of fossil fuel consumption. Wiersum and Ketner (1989) did not consider effects of such massive plantations on the partitioning of C 0 2 between atmosphere, oceans and biosphere, and this requires further analysis with dynamic models. Planting of trees in agroforestry systems, which are an alternative to shifting cultivation, can also contribute in this respect. A secondary effect of reduction of slash and bum practices is reduction of emissions of other gases and aerosols (see discussion of CH4 from biomass burning below). Currently the worldwide reforestation rate (on previously forested lands) is 1 4 . 5 ~ 1ha 0 ~(World Resources Institute 1988). Increasing tropical afforestation to achieve extents in the order of 2 0 0 ~ 1 0ha~ (half of the total land area of Europe) will meet many problems (Wiersum and Ketner 1989). This area of forest plantations would have to be increased by a factor 4 to 5 to sequester the

Inputs to climatic change by soil and agriculture relaled activities

21

total annual C 0 2 increase and this is only effective during the forest's active growth period. Large scale intensification of agriculture is therefore required to release land for afforestation and at the same time satisfy the growing demand for food. Populations in rural areas in the developing countries will not easily cooperate with long term afforestation programmes if their immediate needs are not fulfilled. Moreover, the major producers of C02 are the industrialized nations.

METHANE General The methane concentration in the atmosphere has showed an accelerated increase during the past 300 years (Khalil and Rasmussen 1985; Steele el al. 1987), and its current increase is 1% y - l or 40-90 Tg CH4 y-I (Tg = teragram; 1Tg = 1012g). Methane's atmospheric residence time is relatively short (Table 2.1). Its concentration in the atmosphere and its increase depend primarily on short term fluxes and removal processes. The major portion of the observed rise over the past 300 years is caused by increasing sources (Khalil and Rasmussen 1985). Hydroxyl radicals (OH)' are responsible for the removal of CH4 and CO. Parallel to increasing CH4 there is a strong increase of CO in the atmosphere, caused primarily by anthropogenic sources2 . This increase in part also depresses oxidation of CH4 by depleting OH radicals. Since CH4 is a source of CO itself, there is an instability in the CHJCOIOH system: increases of CH4 and CO lead to depletion of OH, and this leads to further perturbations of CH4 and CO. Khalil and Rasmussen (1985) estimated that 30% of the increase of CH4 is attributable to OH depletion, whereas Cicerone and Oremland (1988) suggested that the OH depletion induces an increase of atmospheric CH4 of only 0.5 to 3 Tg y-l (which

'

OH radicals are the cleaning agents of the atmosphere. They oxidize CH, and in that process CO is formed. What occurs during further oxidation of C O to CO, depends on the concentration of nitrogen monoxide (NO). With sufficient NO (generally in industrialized regions and tropical areas where biomass burning takes place), OH radicals and O3 F are formed. O3 is increasing at Northern mid latitudes and in the upper troposphere. In the absence of NO (e.g. in remote marine areas) O3 is consumed in the oxidation process of CO (Crutzen and Graedel 1986). CO has an atmospheric residence time of only 2-3 months (Cicerone 1988). Therefore determination of a secular trend is very difficult. CO has probably increased in the northern hemispheric troposhere since 1950. Its increase is not uniform or steady in space. No positive trend is observed in the Southern hemisphere.

22

A P . Bouwman and W.G. Sombroek

is much less than the observed total increase of 40-90 Tg y-I), illustrating the uncertainty of these atmospheric processes. Methane is produced during microbial decomposition of organic material under anaerobic conditions. Natural wetlands, wet rice cultivation and landfill sites for solid waste dumping are places where anaerobic conditions prevail. Methane is also formed in the digestive tract of ruminating animals and in the guts of various insects, the major species being termites. The a-biogenic process of biomass burning is a further source of CH4. Recent investigations showed that 20 to 30% of total sources is fossil or 'dead' CH4. This fraction includes approximately 30 Tg of dead CH4 from wetlands and old peat layers (Cicerone and Oremland 1988); emissions from fossil sources therefore amount to 50-95 Tg y-I (Table 2.3). A further source of methane is release of radioactive CH4 by nuclear power plants (see Grantham, Chapter 19 in this volume) which is a negligible source in amount, but with important effects on the isotopic composition of atmospheric CH4. Sinks for CH4 are accumulation in the atmosphere, oxidation by OH, and oxidation in soils. Their relative sizes are highly uncertain (Table 2.3). Land use related sources which will become increasingly important in the future are: wetland rice cultivation, natural wetlands, landfills, ruminants and biomass bu mi ng . Wet rice cultivation Paddy rice fields currently constitute approximately 20% of the total CH4 budget. The harvested area of paddy rice, about 10% of the world's total cultivated area, has shown a steady increase of more than 1% y-I during the last decades (Bouwman 1990a). By the year 2020 the global rice production may have to be increased by a factor 1.6 to meet the growing demand. Extrapolation of available flux measurement data to the global harvested area of paddy rice of each type (dryland, rainfed and irrigated) yields a global emission of 60-140 Tg CH4 y-* (ASehaM and Cmtzen 1989). The uncertainty in this estimate is caused by lack of flux measurements, gaps in the knowledge of the process of methanogenesis and lacunae in the geographic database of the harvested area of the various rice ecologies, soil types used, and soil, water and crop management practices. The use of straw and compost stimulate soil reduction and methanogenesis, while the application of sulphate containing fertilizers (such as ammoniumsulphate) may reduce methanogenesis. Specific agronomic practices such as adapting levels, application mode and timing of fertilization (including chemical as well as organic fertilizers), water management and new rice varieties may help reducing the methane emissions from flooded rice fields. Flooded rice soils with dry fallow periods between rice growing seasons, and fields that are likely to dry up during the cultivation period, are not prone to high CH4 production (Neue

lnpufs to climatic change by soil and agriculture related activifies

23

1990). Specific combinations of soil characteristics and agronomic practices are not likely to show high methane productions. Heavy textured soils with high content of montmorillonitic clays and high content of organic carbon, will probably show high rates of CH4 formation. In leached tropical soils and other soils with fenitic, gibbsitic, fermginous or oxidic mineralogy, the methane production will probably not be high if dry fallow periods are included in the crop rotation. It appears feasible to reduce CH4 emission rates at field level by 10 to 30% (Neue, pers. comm.). However, the global emissions may still increase due to intensification of cultivation and expansion of the cultivated area.

Natural wetlands The global wetland area is 530x106 ha (Matthews and Fung 1987) to 5 7 0 ~ 1 0ha~ (Aselmann and Crutzen 1989), but the various estimates show regional discrepancies and definitional differences. Most of these occur between 50 ON and 70 ON and between 10 ON and 10 "S. The global emission from wetlands ranges between 40 and 160 Tg y-l (Matthews and Fung 1987; Aselmann and Crutzen 1989). Better geographic information coupled with more flux measurements are therefore required to narrow the range of estimates. Methanogenesis is a temperature dependent process. It may even show a 5 fold increase with every 10 "C temperature increase (Burke et al. 1990). It is suggested that - based on past temperature anomalies and estimates of future warming - by the year 2080 CH4 emissions from wetlands will amount to 3 times the 1880 values and more than 2 times the present emissions (Burke et al. 1990). This estimate may be too high, because CH4 oxidation will also be stimulated, possibly even more than CH4 formation. Predictions of direction and magnitude of changes in the area of wetlands due to climatic change are hampered by the uncertainty about future climatic conditions at high latitudes and of temperature enhanced evaporation. In addition to the stimulation of methanogenesis, the possibility of destabilization of methane in hydrates by warming has been raised by various reports (see review by Cicerone and Oremland 1988). These hydrates are solid structures, composed of strong cages of water molecules surrounding CH4 molecules. Methane hydrates are most prevalent at depth in permafrost and in sea sediments. A tentative estimate of the CH4 release from hydrates is 5 Tg CH4 y-l (Cicerone and Oremland 1988). The total area of artificially drained wetlands in temperate regions in the period 1795-1980is 23x106 ha (being 6% of the temperate wetlands), of which 60% in Finland-USSR, 25% in Western and Central Europe and 7% in North America. Drainage of tropical wetlands during that period was only 4%. In the future large areas of natural wetlands will be affected by land reclamation. Large parts of Amazonian wetlands may vanish, while in other places new wetlands

24

A.F. B o u w m n and W . G . Sombroek

will be created in the realization of hydro-electric power schemes. Extension of cattle farming in seasonally flooded or submerged savannas and reclamation of wetlands is proceeding in many parts of South America, while in other tropical regions wet rice cultivation often occurs in reclaimed wetlands. Landfills Anaerobic decay of collected municipal and industrial organic matter that is dumped in landfills add 30-70 Tg CH4 y-l to the atmosphere globally (Bingemer and Crutzen 1987). By far the largest emission currently comes from landfills in the industrialized world. This source has been increasing in recent decades but at present its rise is stagnating. A growing urban population in the developing countries will produce more solid wastes. Projections for the year 2000 suggest a doubling of waste generation in developing countries (Bingemer and Crutzen 1987). Reduction of emissions from landfills may be achieved by gas harvesting or by recycling of solid wastes. Ruminants and termites The global CH4 emission by ruminant animals is 65-100 Tg y-1. Approximately 80% of this amount is produced by cattle and buffaloes, the rest is from sheep (7 Tg y-I), wild ruminants (2-6 Tg y-1) and others (Crutzen et al. 1986). About 40% of the CH4 produced by domestic animals is emitted in the developing countries, mainly in Asia, followed by South America and Africa. The major source regions for CH4 from cattle are North America (1 1 Tg y-'), Europe (8 Tg y - l ) and the USSR (7 Tg y - l ) (Crutzen et al. 1986). This source deserves priority for further research. Changing from animal protein to more plant proteins in human nutrition is a possible way of reducing these CH4 emissions. Modifying feeding strategies and rumen processes and development of CH4 inhibitors also merits attention (EPA 1989). The methane production by termites ranges from 10 to 100 Tg y-1 (Cicerone and Oremland 1988). Most measurements were carried out in vitro. Field measurements are required here, since living conditions and type of nutrition of the numerous families of termites are completely different. It is suggested that the total population of termites shows an increasing trend, caused by extensions of the global cultivated area, burning and forest clearing activities and possibly by increasing biomass production through CO, fertilization. Biomass burning A source of CH4 which is closely related with land use is the burning of biomass, such as burning of agricultural wastes, savannah fires, slash and bum agriculture (shifting cultivation) and buming of fuelwood. On the basis of the

lnpuls lo climatic change by soil and agriculture related oclivilies

25

increasing population of shifting cultivators, increasing production of industrialand fuelwood and of food production, Seiler (1984) estimated that the CH4 production from biomass burning increased from 41-74 Tg y-' in 1950, to 47-84 Tg y-l in 1960 and 53-97 Tg y-l in 1975. These estimates have recognized the large uncertainties and sources of variability, such as amounts of biomass burnt, types and conditions of burning. Growing populations will increase the demand for agricultural land and fuelwood and therefore this source will probably continue to increase in the future. Possible ways to reduce this source are replacement of slash and bum agriculture by agroforestry and alternative noncarbonaceous sources of energy (such as solar energy) to substitute fuelwood.

NITROUS OXIDE Nitrous oxide destroys ozone in the stratosphere (above 10-20 km height) but it is inert in the troposphere. This atmospheric photolysis is a sink of 10.5 Tg N20-N y-1, while atmospheric accumulation of 0.2-0.3% is equal to 2.8 Tg y-l (Table 2.3). Analysis of antarctic ice core samples suggests that the atmospheric N 2 0 concentration has risen from about 270 ppb 400 years ago to about 293 ppb in the beginning of the 20th century (Zardini et al. 1988). Records from longer periods indicate a slight increase in N 2 0 when climate was warming. Nitrous oxide is formed during nitrification and denitrification processes in soils. Nitrification is microbial transformation of ammonia (NH4+) to nitrate (NO3-) while denitrification is transformation of NO3- to molecular nitrogen (N2). Nitrification is dominant in aerobic soils. Denitrification occurs in anaerobic environments and in anaerobic microsites in well aerated soils. Nitrous oxide is an intermediate product of these microbial transformations and, being a gas, it may escape from the soil system to the atmosphere. The turnover rates of N in ecosystems are important for N20 losses. Factors influencing the speed of microbiological Nturnover are soil temperature, soil moisture, soil fertility, availability of organic substrate and soil drainage. N20 fluxes show an enormous spatial and temporal variability. The extrapolation of flux measurements to smaller scales is therefore fraught with potential errors. There are a number of possible explanations for the current 0.2-0.3% increase of the atmospheric N20 concentration. The expanding area of agricultural land and the increasing use of nitrogen fertilizers are identified as major causes. Part of the losses from N fertilizer in the form of N2O occurs directly from soils. Another pathway is through losses from N containing water leached from cultivated fields. The world's cultivated lands, about 1500 x 106 ha, may emit 2.4 to 3.6 Tg of N2O-N annually (Bouwman 1990a). The N20 emission can also be estimated as the fertilizer induced loss, which is the fraction of applied N-fertilizer lost as N20 exclusive of 'natural' emissions. This loss is in

26

A P . Bouwman and W.G. Sombroek

the range of 0.5 to 2.0% of applied N fertilizer (Bolle et al. 1986). In recent years the consumption of fertilizer increased from 67 Tg in 1983/84 to 72 Tg in 1986/87 (FA0 1988). Fertilizer induced N 2 0 emission was thus 0.4 to 1.4 Tg N20-N y-l in 1987. The fertilizer consumption figures suggest that this source is increasing. Literature data show that under certain conditions the timing of Nfertilizer application and choice of fertilizer type can strongly influence N20 emissions (Bouwman 1990a). Legumes may play an important part in the production of N2O. The symbiosis between leguminous host plants and Rhizobial bacteria can add nitrogen to the soil-vegetation system through fixation of atmospheric N2. Freeliving Rhizobia may be important denitrifiers. Free living Rhizobia can persist in infertile environments, thanks to a combination of properties such as ability to assume dormancy, slow growth, ability to denitrify and slow metabolism. It is also possible that symbiotically living Rhizobia bactemids in root nodules are able to denitrify. If symbiotic denitrification is important, the cultivation of leguminous crops is a potential increasing source of atmospheric N20. More on this subject can be found in O'Hara and Daniel (1985). Further potential causes of increasing N20 emissions are increasing amounts of N reaching the soil through wet and dry NO, deposition and NH4+in rainwater, which add to soil nitrogen and may in part be lost as N20, and N20 produced in surface waters and oceans. Tropical rainforests and tropical savannas emit high amounts of N20 per unit area. High rates of N20 emission in tropical savanna areas were measured by Hao et al. (1988). In the case of savannas the occurrence of alternating wet and dry conditions may enhance N20 fluxes. Leguminous plants occumng in these ecosystems contribute to soil-N through N-fixation. At the end of the dry season in savanna regions the availability of organic substrate in the soil is high. The first rains initiate a very fast mineralization of this organic matter and nitrification of N compounds. Since high temperatures favour fast turnover rates, under these conditions high N2O losses may be expected (see e.g. Keller et al. 1988). The emissions of N 2 0 may increase after replacing the original tropical forest vegetation by grassland (Matson, pen. comm.), possibly caused by an increased frequency of rewetting soils after periods of drought. Another source is N2O from fossil fuel combustion and from biomass burning, but their annual emissions are highly uncertain. Recently Muzio and Kramlich (1988) reported that estimates published so far of N20 fluxes by fossil fuel combustion are much lower than the 1 to 2 Tg y-l (Seiler and Conrad 1987) postulated in previous global budgets. On the basis of new data, Crutzen (pers. comm.) estimated that N 2 0 emission by biomass burning is less than half the 1.5 Tg y-l assumed by Seiler and Conrad (1987). Clearly, the global N20 budget needs further investigations.

Inputs

10 climaric

change by soil and agriculture related activities

27

CONCLUDING REMARKS With regard to COz the rate of deforestation still forms an uncertain factor. Definitional controversy on land use and vegetation types causes great disagreement between estimates of C02 emissions. A further cause of unreliability in the estimates is the biomass in tropical forests, but its effect on the net flux of C02 is small compared to the effect of uncertainties in the rate of deforestation. The accumulation in the atmosphere is well known, while the uptake in the Oceans is probably lower than assumed until recently. The increase of net primary production caused by increasing atmospheric C 0 2 concentrations needs further research to solve the uncertainties in the C02 budget. The production and oxidation of methane in soils and wetlands is extremely dependent on environmental conditions. As for the nitrous oxide, the variability of fluxes both in time and space of fluxes is extremely high. A number of specific fields require research attention: 1). the geographic distribution of rice ecologies, soil, crop and water management, and soil types used for wetland rice cultivation; 2) the soil processes responsible for the production and emission of methane; 3) the geographic distribution of the different types of salt and fresh water wetlands and the relation between the type of wetland and methane fluxes; and 4) fluxes of CH4 from landfill sites, especially quantities of organic waste which is decomposed anaerobically in landfill sites are virtually unknown. Methods to reduce the CH4 release from landfills merits research attention. The above investigations will also improve the capability to assess consequences of climate change for methane fluxes, especially those from natural wetlands. The allocation of N20 emissions to different sources is still very uncertain. As shown above, there is no good explanation for the atmospheric increase, which may in the future even be amplified by changing concentrations of other atmospheric constituents. Much more process oriented research parallel to flux measurements is needed, especially in heavily fertilized areas and in ecosystems where major changes occur. As for land use, an option which needs consideration is stimulation of agronomic practices to depress N 2 0 losses from agricultural lands and surface waters, stimulation of efficient N-fertilizer use and prevention of over-fertilization,and use of nitrification inhibitors. The currently used methods of measuring trace gas fluxes are point measurements. Extrapolation of such measurements to smaller scales is fraught with potential emrs. The development of methods to estimate trace gas fluxes for larger areas is needed to improve the quantification of the global and regional sources. A geographic database of present-day land cover and land use types, the latter including management practices, is urgently needed.

28

A . F . Bouwman and W.G. Sombroek

REFERENCES Armentano, T.V. and E.S. Menges (1986). Patterns of change in the carbon balance of organic soil wetlands of the temperate zone, Journal of Ecology 75:755-774. Aselmann, I. and P.J. Crutzen (1989). Freshwater wetlands: Global distribution of natural freshwater wetlands and rice paddies: their net primary productivity, seasonality and possible methane emissions, Journal of Atmospheric Chemistry 8:307-358. Bingemer, H.G. and P.J. Crutzen (1987). The production of methane from solid wastes, Journal of Geophysical Research 92:2181-2187. Bolin, B. (1986). How much C 0 2 will remain in the atmosphere? The carbon cycle and projections for the future. In: B. Bolin, B.R. Doos, J. Jager, and R.A. Warrick (Eds.) The greenhouse effect, climatic change and ecosystems, pp 93-156, SCOPE Vol. 29, Wiley and Sons, New York. Bolle, H.J., W. Seiler and B. Bolin (1986). Other greenhouse gases and aerosols. Assessing their role in atmospheric radiative transfer. In: B. Bolin, B.R. Doos. I. Jager and R.A. Warrick (Eds.), The greenhouse effect, climatic change and ecosystems, pp 157-203, SCOPE Vo1.29, Wiley and Sons, New York. Bouwman, A.F. (1990a). Exchange of Greenhouse Gases between terrestrial ecosystems and the atmosphere. In: A.F. Bouwman (Ed.), Soils and the Greenhouse effect, pp 61-128, Wiley and Sons, Chichester (1990). (Adapted version of ISRIC Working Paper 89/4, International Soil Reference and Information Center, Wageningen, the Netherlands). Bouwman, A.F. (1990b). Land use related sources and sinks of greenhouse gases. Present emissions and possible future trends, accepted in Land Use Policy. Brown, S . and A.E. Lug0 (1982). The storage and production of organic matter in tropical forests and their role in the global carbon cycle, Biotropica 14:161-187. Brown, S. and A.E. Lug0 (1984). Biomass of tropical forests: a new estimate based on forest volumes, Science 223:1290-1293. Burke, M.K., R.A. Houghton and G.M. Woodwell (1990). Progress toward predicting the potential for increased emissions of CH4 from wetlands as a consequence of global warming. In: A.F. Bouwman (Ed.), Soils and the greenhouse effect, pp 451-456, Wiley and Sons, Chichester (1990). CDIAC (1989). CDIAC Communications, CDIAC's corner, Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, Summer 1989. Cicerone, R.J. and R.S. Oremland (1988). Biogeochemical aspects of atmospheric methane, Global Biogeochemical Cycles 2299.327. Cicerone, K.J. (1988). How has the atmospheric concentration of CO changed ? In: F.S. Rowland and I.S.A. Isaksen, The changing atmosphere, Dahlem Workshop Reports, pp 49-61, Wiley and Sons, Chichester. Crutzen, P.J. and T.E. Graedel (1986). The role of atmospheric chemistry in environmentdevelopment interactions. In: W.C. Clark and R.E. Munn (Eds.), Sustainable development of the biosphere, pp 213-250, IIASA, Laxenburg, Austria, Cambridge University Press. Crutzen, P.J., I. Aselmann and W. Seiler (1986). Methane production by domestic animals, wild ruminants, other herbivorous fauna and humans, Tellus 38B:271-284. Detwiler, R.P. and C.A.S. Hall (1988). Tropical Forests and the Global Carbon Cycle, Science 239142-47. Enquete-Kommission Vorsorge zum Schutz der Erdatmosphare ( I 989). Schutz der Erdatmosphiire: eine internationale Herausforderung; Zwischenbericht der EnqueteKomm. des 11. Deutschen Bundestages Vorsorge zum Schutz der Erdatmosphae, Dt. Bundestag, Referat Offentlichkeitsarbeit, Bonn.

lnpuls to climatic change by soil and agriculture relared activities

29

EPA (1989). Reducing methane emissions from livestock: opportunities and issues (by M.J. Gibbs, L. Lewis and J.S. Hoffman), US Environmental Protection Agency, Washington, D.C., USA. Esser, G . (1987). Sensitivity of global carbon pools and fluxes to human and potential climatic impacts, Tellus 39B:245-260. Esser, G. (1990). Modelling global terrestrial sources and sinks of C 0 2 with special reference to soil organic matter. In: A.F. Bouwman (Ed.), Soils and the greenhouse effect, pp 247-262, Wiley and Sons, Chichester (1990). F A 0 (1988). Fertilizer Yearbook. F A 0 Statistics series No. 83. 123 p., FAO, Rome. F A 0 (1983). Production yearbook Vol. 37. F A 0 Statistics series No. 55, F A 0 Rome. Goudriaan, J. and P. Ketner (1984). A simulation study for the global carbon cycle, including man's impact on the biosphere, Climatic Change 6:167-192. Hao, W.M., D. Scharffe and P.J. Crutzen (1988). Production of N20, CH4 and C 0 2 from soils in the tropical savanna during the dry season, Journal of Atmospheric Chemistry 7:93105. Houghton, R.A., R.D. Boone, J.M. Melillo, C.A. Palm, G.M. Woodwell, N. Myers, B. Moore I11 and D.L. Skole (1985). Net flux of carbon dioxide from tropical forests in 1980, Nature 3 16:6 17-620. INPE (1989). Avaliacao de alteraqao da cobertura florestal na Amaz6nia legal utilizando sensoriamento remoto orbital, Instituto de Pesquisas Espaciais, Roberto Pereira da Cunha (coordinator), LNPE, Sao Jost dos Campos, May 1989. Keepin, W., I. Mintzer and L. Kristoferson (1986). Emission of C 0 2 into the atmosphere. The rate of release of C 0 2 as a function of future energy developments. In: B. Bolin, B.R. DOOs, J . Jager and R.A. Warrick (Eds.), The greenhouse effect, climatic change and ecosystems, pp 35-91, SCOPE Vo1.29, Wiley and Sons, New York. Keller, M., S.C. Wofsy and J.M. Da Costa (1988) Emission of N 2 0 from tropical soils: response to fertilization with NH4+, NO3- and P043-, Journal of Geophysical Research 93: 1600-1604. Khalil, M.A.K. and R.A. Rasmussen (1985). Causes of increasing atmospheric methane: depletion of hydroxyl radicals and the rise of emissions, Atmospheric Environment 191397-407. Lanly, J.P. (1982). Tropical forest resources. F A 0 Forestry Paper No. 30. 106 pp. FAO, Rome. Matthews, E. and I . Fung (1987). Methane emission from natural wetlands: Global distribution, area, and environmental characteristics of sources, Global Biogeochemical Cycles 1:61-86. Mooney, H.A., P.M. Vitousek and P.A. Matson (1987). Exchange of materials between terrestrial ecosystems and the atmosphere. Science 238:926-932. Muzio and Kramlich (1988). Artifact in the measurement of N 20 of combustion sources, Geophysical Research Letters 15:1369-1372. Neue, H.U., P. Becker-Heidemann and H.W. Scharpenseel (1990). Organic matter dynamics, soil properties and cultural practices in rice lands and their relationship to methane production. In: A.F. Bouwman (Ed.), Soils and the greenhouse effect, pp 457-466, Wiley and Sons, Chichester (1990). OHara, G.W. and R.M. Daniel (1985). Rhizobial denitrification: a review, Soil Biology and Biochemistry 17, 1985, pp 1-9. Ramanathan, V., R.J. Cicerone, H.B. Singh & J.T. Kiehl (1985). Trace gas trends and their potential role in climate change, Journal of Geophysical Research 90, pp 5547-5566. Rotty, R.M. (1987). A look at 1983 C 0 2 emissions from fossil fuels (with preliminary data for 1984). Tellus 39B:203-208.

30

A.F. Bouwman and W.G. Sornbroek

Sedjo. R.A. (1989). Forests to offset the greenhouse effect, Journal of Forestry, pp 12-15, July 1989. Seiler, W. (1984). Contribution of biological processes to the global budget of CH4 in the atmosphere In: M.J. Klug and C.A. Reddy (Eds.), pp 468-477, American Society for Microbiology, Washington, D.C.. Seiler, W. and R. Conrad (1987). Contribution of tropical ecosystems to the global budgets of trace gases, especially CH4, H2, CO and N 2 0 . In: R.E. Dickinson (Ed.), The geophysiology of Ammonia. Vegetation and climate interactions, pp 133-160, Wiley and Sons, New York. Seiler, W. and P.J. Crutzen (1980). Estimates of gross and net fluxes of carbon between the biosphere and the atmosphere from biomass burning, Climatic Change 2:207-247. Steele, L.P.. P.J. Fraser, R.A. Rasmussen, M.A.K. Khalil, T.J. Conway, A.J. Crawford, R.H. Gammon, K.A. Masarie and K.W. Thoning (1987). The global distribution of methane in the troposphere, Journal of Atmospheric Chemistry 5:127-171. Swart, R.J. and J. Rotmans (1990). Food or forest? Can the tropical forests survive along with continuing growth of population and economy? In: A.F. Bouwman (Ed.), Soils and the Greenhouse Effect, pp 431-440, Wiley and Sons, Chichester (1990). Tans, P.P., I. Y. Fung and T. Takahashi (1990). Observational constraints on the global atmospheric carbon dioxide budget, accepted in Science. Van Breemen, N, and T.C.J. Feijtel (1990). Soil processes and properties involved in the production of greenhouse gases, with special relevance to soil taxonomic systems. In: A.F. Bouwman (Ed.), Soils and the greenhouse effect, pp 195-224, Wiley and Sons, Chichester (1990). Wiersum, K.F. and P. Ketner (1989). Reforestation, a feasible contribution to reducing the atmospheric carbon dioxide content. In: P.A. Okken, R.J. Swart and S. Zwerver (Eds.), Climate and Energy, the feasibility of controlling C 0 2 emissions, pp 107-124, Kluwer Publishing Company, Dordrecht. World Resources Institute (1988). World Resources 1988-89. An assessment of the resource base that supports the global economy, with data tables for 146 countries, World resources Institute, International Institute for Environment and Development. Zardini, D., D. Raynaud, D. Scharffe and W. Seiler (1988). N 2 0 measurements of air extracted from Antarctic ice c b r e s : implication on atmospheric N 2 0 back to the last glacialinterglacial transition, Journal of Atmospheric Chemistry 8:189-201.

31

Chapter 3

PROCESSES THAT AFFECT SOIL MORPHOLOGY Richard W. Arnold USDA-Soil Conservation Service P. 0. Box 2890, Washington, D.C. 20013, USA

INTRODUCI'ION Soil morphology consists of the features and properties of soils that ~IC commonly visible when looking at a profile or handling samples of soil. The more usual features are associated with colors and their variation in kind, amount, location and distribution. The presence of organic matter, soluble salts, carbonates, iron compounds, and colors inherent in some parent materials are all examples of variations in color that we recognize as differences of morphology. In addition to color patterns there are aggregates of soil, peds, that give rise to patterns of organization and distribution throughout a profile. When a sample of soil is examined we can feel the differences in resistance to crushing which we call consistence and also feel the texture which is the combination of sand, silt and clay particles. The processes that give rise to variations in soil morphology can be grouped (Fridland 1976) into: Those that translocate the particles without significant alteration chemically; Those that translocate solutes, cause chemical alteration of compounds through reduction and oxidation, effect degrees of hydration, and so forth; Those that result in mixing of soil materials by animals or alter the soil through biotic influences; Those that are anthropically induced and may be planned such as imgation or terraces, or may be unplanned such as enhanced erosion due to grazing, cultivation, or other disturbances. It is also common and very likely that combinations of these processes are the major influences giving rise to variations of soil morphology.

GEOPHYSICAL PROCESSES Where the physical particles and the soil aggregates are moved without much alteration, the processes may be considered to be geophysical. Water

32

R . W . Arnold

erosion is an important geophysical process that gives rise to variations in soil morphology. Locally some soil materials are removed, transported across a slope and deposited elsewhere. The removal and additions of these particles and their subsequent arrangements are recognized as changes in the color and suucture patterns of a soil's morphology. Where the processes have been responsible for major accumulations of parent materials e.g. volcanic ash, the initial state of the profile is influenced, in contrast to the micro variations that occur in the surface layers of soils. Most landscapes in the world have been influenced markedly by the action of running water; thus landscape evolution and soil morphology tend to go hand in hand. Small amounts of sheet and rill erosion are usually obliterated by tillage operations and thus the impact on soil morphology tends to be transient in the minds of the observers. Wind deflation is a companion process that removes soil particles from one place which are transported and deposited in adjacent areas. Similar microstratification is observed in alluvial environments along streams and on terraces where sediments are deposited from the overflow waters. In places erosion also takes place. The alternating deposition and removal of sediments on alluvial fans gives rise locally to recognizable patterns of soil mophology. Within soil profiles the process of translocation of physical particles, often clay and/or organic matter, result in easily recognizable differences of morphology. In fact such features are usually identified as specific kinds of soil horizons, such as argillic for the clay enriched ones, and spodic (podzolic) for the humus enriched ones. Other processes that move soil particles include landslides and other forms of mass movement resulting in mixing and disturbance of the soil materials. The melting of ice and freezing of water in profiles and the variability of permafrost also moves soil materials, often giving rise to unique patterns of soil morphology. The churning of clayey soils due to wetting and drying also produces characteristic patterns of soil morphology. Fig. 3.1 illustrates an extreme case of vertic properties giving rise to short range variability in a Vertisol from Texas, U.S.A.

GEOCHEMICAL PROCESSES Where the major differences in morphology are related to the chemistry of compounds, whether translocated solutes or chemically altered materials, the processes can be thought of as being geochemical. In humid areas where relief gives rise to moisture differences across the landscape (catena) the patterns of soil morphology shift in definite patterns from drying on the lower slopes resulting in oxidation-reduction phenomena and the more freely draining uplands and upper slopes to the fluctuating wetting and

Processes

that

affecis soil morphology

33

influence on color patterns in soils. In drier areas the presence and absence of ground water at relatively shallow depths modifies the soil morphology as soluble salts are translocated upwards, downwards and laterally. Within the capillary fringe oxidation and reduction often occurs. Karst like suffusion alters minerals and removes some of the compounds, thus a sequence of changes of soil morphology may be observed in such environments. Variations in rocks that make up the parent rock of soils are often different in chemical and physical properties that influence how and where features will be observed in soils. In some environments there are paleo soils and parent materials that become exposed and blend into the current stages of soil evolution, consequently part of the observable soil morphology is relict. Where mineral transformations occur and the components are translocated, and then reorganized or synthesized into new compounds, the resulting accumulations are recognized as differences of morphology. Some argillic horizons and spodic horizons form this way as do some of the pans. Cementing materials of pans, like the petrocalcic, petrogypsic, placic and duripan are examples.

Fig. 3.1

"Bow structure" (Dudal and Eswaran 1989) in vertisols illustrating morphological microvariability

34

R.W. Arnold

BIOGEOCHEMICALPROCESSES Plants and animals interact with soils in some situations giving rise to mixing and chemical alteration of the materials. These processes can be thought of as biogeochemical. In and and semiarid areas the soils under certain shrubs or trees have characteristics that contrast with adjacent areas without the vegetation. It may show as more darkened aggregates or structures that differ due to increased moisture supply. In more humid areas there may occur distinct eluvial zones due to the chemical compounds released from trees or other types of vegetation. The mixing of soil is common where winds and storms uproot trees and micropits and mounds are produced with strikingly different morphologies. In cold, and many wet, environments there is an accumulation of biologic materials such as moss mounds or blanket bogs which modify or develop morphologies of their own. Zoogenic processes that mix and alter the soil materials are common in many parts of the world. In the tropics termites build mounds from the soil particles moved from the subsoil and substratum. Prairie dogs, marmots, moles and other animals build nests and burrows in soils altering the observable morphology. Crayfish build castle like tubes in some wet soils. Earthworms, beetles, cicada and many other types of fauna exert an influence on soils. Their excrement and gastric compounds significantly alter the chemistry on a microscale.

ANTHROPIC PROCESSES -AGRICULTURE Man seldom uses soils the way they occur naturally. Rather they become a medium for plant growth or an earthy construction material. Because of the almost limitless possibilities, there are many kinds of changes and alterations of soils that show up in morphology. Direct effects Tillage of land after the existing vegetation has been removed, mixes the surface materials together and tends to homogenize the morphology. This is common where the soils are disked or plowed, although changes of morphology will be noted even where the disturbance of the soil is less systematic. Due to mechanical alteration of the landscape such as hillside terraces and many rice paddies, there is a change in the moisture regime, which induces chemical and biological changes of the system, and in the resulting morphology. Fields that have been levelled and imgated have similar kinds of alterations. Land disturbance due to logging or other types of intermittent harvesting

Processes that affects soil morphology

35

also alter the morphology of soils. Gouging, slipping, and compacting of the surface may be noted within the soil.

Indirect effects As a consequence of overgrazing rangelands and of mismanaging cultivated lands, water and wind erosion are common. The processes are often accelerated relative to the same kinds of processes under natural conditions. The effects on morphology are about the same.

ANTHROPIC PROCESSES - INDUSTRY Some of the more striking disturbances of the soil landscape are those associated with industrial development and use of land. Cities cover the surface and bury pipes and other features under the surface. In the search for economic minerals the mining industries disturb soils to varying degrees depending on the procedures used to obtain the resources.

Direct effects The changes of soil morphology are associated with the kind and degree of disturbance of the natural soil. Soils under roads, streets, parking lots and other surface structures show disturbance and often have an increase of moisture and a decrease of biological activity. Park lands associated with cities often have been modified drastically to produce an externally pleasing view. Gardens and green spaces have anthropic soils of varying degrees of expression. Strip mining removes the soil and when the area is reclaimed the replacement usually is a mixture of new parent materials. The morphology often includes a random assortment of prior features. Dredging from stream channels that are deposited on land gives rise to new areas of artificial parent materials from which new soils can form. Industries that process agricultural products often dispose of their wastes on land. The wastes may vary widely in pH, have lots of organic debris, or contain caustic compounds which alter the chemistry and biology of the soils so treated. These additions are noted in changes of soil morphology. Archeology contains evidence of many kinds of uses and abuses of soils throughout man's history. Removal, burial, mixing or simply covering are all recorded in the morphology of the effected soils.

Indirect effects The burning of high sulfur content coal has contributed to an acidification of the precipitation falling in the down wind areas from major zones of such

36

R.W. Arnold

industries (see also Van Breemen, Chapter 11 of this volume). Other kinds of pollutants have been deposited on lands near smelting and refining industries. Some compounds deposited on soils, usually as aerosols, do not readily affect the existing morphology. For example, radiation nuclides are not easily detected in the gross features of morphology. They may eventually affect other processes that will result in visible changes.

TIME SCALE OF SOIL FEATURES A list of the estimated amount of time needed for processes to make a significant imprint on soil morphology has been proposed by Targulian (1989). The time scale is by powers of 10 and so those with characteristic times of 1-10 years and 10-100 years are of more immediate interest.

Characteristic time: 1

.

-

10 years

Root distribution of annual plants; Surface litter, Fluvic properties; Gleyic and stagnic properties; Slickensides; Permafrost presence; Inundic (temporary flooding); Salic features; Yermic features of fine earth (desert crust).

Characteristic time: 10

-

100 years

Tree root distribution; Humus content and thickness in topsoil; Salic properties; Calcareous properties; Sodic properties; Vertic properties; Histic epipedon, immature; Ochric epipedon; Gypsic horizon; Albic horizon in Spodosols; Natric horizon - immature; Spodic horizon - immature; Gilgai; Placic horizon - weak. It is interesting that many of the soil features that we associate with soil

Processes

affects soil morphology

that

37

morphology are thought to take more than 100 years and many others more than a 1000 years to develop their expression (Birkeland 1974). The concept of Birkeland (1974) is illustrated in Fig. 3.2. Profile DeveloPment . -

/I

A

/

/

10

6

0 0

1

3

2

4

5

Time (years as exp. of 10)

=

Oxisols

Fig. 3.2

Ultisols

0Mollisols

6

10) Spodosols

Schematic diagram to show time required by different kinds of soils to attain steady state (adaptedfrom Birkeland 1974)

Examples of the more than 100 years are high chroma colors, iron concretions, histic, mollic and umbric epipedons, calcic, natric, cambic, and spodic horizons, and desert pavement. In the latter group are argillic and oxic horizons, petrocalcic and petrogypsic horizons, duripans and fragipans. Carbonate accumulations in soils have been subject to dating and appreciating the errors associated with the dating procedure. Carbonate accumulation in soils is a good example of soil morphologic changes associated with time. Fig. 3.3 gives an example from New Mexico, USA, with data from Gile et al(1971). Even in a simple system such as the carbonate system, there are many processes (dissolutions, translocations, precipitations) acting with various intensities and in different parts of the profile, and these cumulatively contribute to the final profile morphology at any given time.

CONCLUDING REMARKS Pedology is based on a morphogenetic understanding of the pedons and profiles that exist in a landscape. It is not surprising therefore that most of the processes that are thought to affect the development of soils are also thought to be reflected in the observable features known as soil morphology. The grouping of processes into physical, chemical, biological, and anthropic is artificial in that most actions taking place in soils are complex combinations of processes.

38

R. W.Arnold

There has not been sufficient study of the time characteristics of processes that modify soil features and properties. If we wish to contribute to unravelling the earth's surface history, pedologists have much to offer and a long way to go.

SOIL CALCIUM CARBON

Om

0

10

Im

2m

> 1,000 < 2,500

> 10,000 < 15,000

>25,000

<50,000

APPROXIMATE SOIL AGE

Fig. 3.3

v

Evolution of depth function of carbonate profiles in an aridic environment (after Gile et al. 1971)

REFERENCES Birkeland, P.W. (1974). Pedology, weathering, and geomorphological research. Oxford University Press . New York. 285p. Dudal. R. and H. Eswaran (1989). Distribution. properties and classification of Vertisols. In Eds. L. P. Wilding, and R. Puentes. Vertisols: Their distribution, properties, classification, and management. Tech. Mono. 18.. Soil Management Support Services, SCS-USDA, Washington D.C., USA., I - 22. Fridland, V.M. (1972). Pattern of the soil cover. Translated from Russian, 1976. Keter Publ. House, Jerusalem. Gile, L.H., J.W. Hawley, and R.B. Grossman (1971). The identification, occurrence and genesis of soils in an arid region of Southern New Mexico. Training Bull. Soil Conservation Service, USDA. 177 pp. Targulian, V.O. (1989). Personal communication. IIASA, Laxenburg, Austria. influence on color patterns in soils.

39

Chapter 4

INFLUENCE OF CLIMATIC CHANGE ON SOIL MOISTURE REGIME, TEXTURE, STRUCTURE AND EROSION G.Y. Varallyay Research Institute for Soil Science and Agricultural Chemistry of the Hungarian Academy of Sciences Herman Orio UI 15, Budapest, Hungary

Human activities are currently leading to changes in the global environment challenging us to describe and understand the interactive physical, chemical and biological processes that regulate the entire earth system, the unique environment for life (IGBP 1988; Toward ...1988; Earth System Science 1988). Climate is a primary factor influencing these processes regarding direction, rate, spatial distribution, time variability and further consequences. In the last years several international scientific programmes were initiated for the registration, modelling and forecast of global climatic changes (Bolin et al. 1986; WCAP-4 1987; IGBP 1988; IGBP 1989). Due to natural processes and human activities such as increasing energy consumption, industrialization, agriculture, urban and rural development, considerable changes took place in the gas composition of the atmosphere. The C02 concentration, which was about 180200 ppm after the last glaciation and 270 ppm at early industrial times, increased to the present 350 ppm, a 50 % increase during the last 100-120 years. A doubling of the concentration of radiative gases in the atmosphere is expected to lead to a rise of global temperature of 3"C+/-1,5"C, which means a rapid, 0.103°C average temperature increase per decade. The spatial and temporal patterns of temperature increases will be heterogeneous and are expected to be largest in the northern mid-continent region of North America and Eurasia. The various Global Circulation Models (GCMs) predict a greater temperature increase in the polar regions than in the equatorial regions; and relatively higher temperature increase during summer, and lower during winter periods in both hemispheres. A temperature rise of 5 ° C for example, would imply northward displacements of isotherms in North America by 500 km, and such a change may occur within 100 years (Toward ... 1988). GCMs forecasting the temperature regime (WCAP-4 1987; WCRP-22 1988) are still rather uncertain, because in addition to solar radiation, the influences of circulation, changes in vapor content, cloudiness, albedo and surface roughness have to be evaluated more quantitatively. Warming up of mountain glaciers (resulting in changes in the waterflow dynamics, such as additional floodwaves), melting in the permafrost soil zone and the polar ice caps and increasing volume of the warmer oceans, will result in a rise of the eustatic sea-level. Its magnitude, assuming a 2-5°C temperature increase in the next 50-

40

C.Y. Varallyay

100 years, may vary between 0.20-1.40 m. This sea level rise is threatening low lying, man protected lands, including settlements, agricultural areas and extended sea shores with low slope angles. Another consequence will be the further extension of salt-affected areas under the direct temporal seawater inundation or indirect rise of the sea-level connected water table of saline or brackish groundwaters, as influences of the higher sea level (Szabolcs and Redly 1989). The change of sea-level will also raise the erosion basis in the affected catchment area which may result a non significant reduction of the water erosion potential. The forecast of precipitation characteristics (quantity of rain and snow; rainfall distribution pattern; intensity) is even more uncertain. Within the World Climate Programme, the World Meteorological Organization (WMO) takes serious efforts to improve the complex GCMs and quantify at least some basic plausible, hypothetical climate-change scenarios (Washington and Meehl 1983; Gleick 1986 and 1987; WCAP-4 1987; WCRP-5 1988). From the state-of-the-art of these GCMs it can be concluded that in the first period of the forecasted warming-up the average global precipitation will decrease, with high spatial variability and considerable regional redistribution. But later these tendencies will be counterbalanced by the increasing evaporation from water surfaces (first of all from the oceans), which leads to higher air humidity and probably more precipitation, again in a rather uneven spatial and time distribution, due to the changing circulation phenomena and vegetation pattern. The changing climate will result in considerable changes in the natural vegetation and in the land use practices (IGBP 1989; Bolin et al. 1986; Parry et al. 1990). The major vegetation belts will move northwards in the Northern and southwards in the Southern Hemisphere. Manabe and Wetherald (1 986) forecasted a rate of 25-200 km/IOO year northward "movement" of the coniferous forest zone on the North American continent. Vegetation in many cases cannot tolerate this "velocity", it will not be able to change its geographical distribution as fast as the changes in suitable habitat, there would be decades in length in the adjustment of ecological systems to rapidly changing climatic conditions. This leads to considerable changes in the species distribution, dynamics, diversity and production capacity of various ecosystems, consequently in their ecosystem functions. Similar changes are forecasted in agricultural (Blasing and Solomon 1984; Parry et al. 1990) and managed forest ecosystems, in spite of human activities, as land use, agrotechnics, amelioration, etc. The changes in the vegetation pattern result in feedback effects on climate, modifying albedo, surface roughness, micro-circulation processes, the heat and energy balance of the nearsurface atmosphere, the characteristics of both temperature and precipitation. Vegetation changes will considerably influence the field water cycle and soil formation proccsscs. The changes in tempcrature, precipitation and vegetation patterns (as soil forming Factors) will result in significant changes in soil conditions. These

41

Soil m i n u r e . [exlure, SIruclure and erosion

impacts and their relationships are summarized in Fig. 4.1.

J

I

--

S

E

T

wapo- Ironroiion Slaage ration spira -

Wilt-

*t

Fig. 4.1

tion

.

'7'

-

-

water

air

heat

r.

r.

r.

subsiancer.

- tl

The influence of climate change on soil moisture regime and soil erosion

It expresses in a selfexplanatory way why the quantitative evaluation of any climatic change on soil conditions is so difficult, and far from being satisfactory. The uncertainty of long-term global temperature and precipitation forecasts are combined here with the complex, integrated influences of changing vegetation and land use pattern and the changing hydrological cycle. Consequently, the global soil change prognosis can only be a rough, qualitative estimation and only

42

G.Y. Varallyay

allows to draw some rather general conclusions. Most of the forecasted soil changes are related to the field water balance (Vituki 1989; WCAP-4 1987). Its components are shown in Fig. 4.2. Because of the uncertainty of temperature and precipitation prognoses four plausible, potential climate change scenarios were selected to present their possible impacts on the elements of field water cycle and soil moisture regime. The moisture regime consequences of these four scenarios are also summarized in Fig. 4.2.

COLD

COLD DRY

Too cold for clay illuvis 1'ion

I

WET

-

frost heaving destroys any

texture differentiation Becoming too cold for clay illuviation

-

Maxirrwm clay illuviation

Minimum clay illuviation profiles high in sesauioxide c l a p

HOT DRY

'

semiarid

HOT WET

sub-

humid

humid

r -

~

~~~~

The effect of climate on texture differentiation in soils

little vegetation growth

I

Humification is slow so organic matter accumulates throughout profile, impeding podzolisation

I

Some humus

in soils as cooler conditions slow the destruction of humus

Humification fairly intense so lulvic acids orrn resultiiig in illuviation of iron and humus

in their profiles tion

profiles

HOT

WET

The role of organic matter in soil formation in response to climate

Fig. 4.2

The effect of four potential climate scenarios on texture differentiation and organic matter (from IGBP Report No. 5,1989)

Soil moisiwe, iexiwe, s i r u i w e and erosion

43

A general conclusion can be drawn, that an increase in the average annual precipitation will be followed by an increase in:

Surface runoff (R) in hilly lands with undulating surfaces and without permanent and dense vegetation, if the infiltration rate, permeability and water storage capacity of soil is limited; Infiltration (I) and water storage (S) within the soil, if they are not limited, such as in flat lands; Groundwater recharge (G) if the soil profile has good vertical drainage, permeability is not limited, especially in low lying areas; Evaporation (E) if infiltration is limited; Transpiration (T) in the case of well developed plant canopies. The decrease in precipitation results in adverse changes. Increasing temperature will result in: Increases in the potential E and T, if the plant canopy is not suffering from limited water supply due to climatic or soil induced drought, e.g. low precipitation or limited water storage capacity; Decreases in R, I, S and G, especially if it is accompanied by low precipitation; Decreases in the intensity (depth) of permafrost, it will modify the geographical boundaries of permafrost, opening possibilities for increasing water storage and water movement, biological activity and soil formation processes within the unfrozen part of the soil. Decreasing temperature will result adverse changes. These general influences are modified with the impact of vegetation characteristics (type, density, dynamics, species composition, biomass production, litter and root characteristics), and depend greatly on the type, intensity, spatial and temporal distribution of atmospheric precipitation. Man's influence is still more complex. Land use, cropping pattern, agrotechnics, amelioration (including water and wind erosion control, chemical reclamation, imgation and drainage) and other activities sometimes radically modify the field water balance and its components. In regimes under agricultural utilization the influence of global climatic change on the soil moisture regime is partly affected through these human actions (Parry et al. 1990). Within the World Climate Programme of the WMO a series of international projects are dealing with the assessment, modelling and forecast of the impact of climate variations and climatic changes on the field water cycle. In these studies proper spatial and time scaling has special significance,determining the minimum and optimum data sets required, the applicable models and approaches as well as the interpretability of outputs for global and regional, long-term, yearly, seasonal

44

G.Y. Varallyay

or month to months applications (Solomon et al. 1987; WCAP-4 1987; WCRP-5 1987- 1988 ; WCW-22 1988). Based on general circulation models (GCMs) and hypothetical scenarios several prognoses were elaborated for the assessment of the influence of forecasted climatic changes on soil moisture regime. Washington and Meehl (1983) prepared a map on the geographical distribution of soil moisture differences for the periods December to February and June to August (see Fig. 13.1). These differences were calculated by subtracting the soil moisture content at the present atmospheric C02 concentration from the moisture content at an assumed double atmosphere CO;?concentration (Washington and Meehl 1983; Manabe and Holloway 1975). The map indicates a substantial drying in spring and summer in the middle to upper mid latitudes and significant high latitude increase in precipitation, and hence runoff. In Fig. 4.3 the monthly distribution of these soil moisture differences are indicated, with longitudinally averaged values. The plot shows increases in continental soil moisture between 30 ON and 60 ON with greatest increases in the winter and spring seasons. Decreased soil moisture is found for most seasons in tropical latitudes and the Southern Hemisphere. Manabe and Holloway prepared a map in 1975 indicating the estimated global distribution of the annual mean rate of runoff. SOIL MOISTURE DIFFERENCES

60N

30N

EQ

305

60s

J

F

M

A

M

J

J

A

S

O

N

0

J

Fig. 4.3 Monthly distribution of soil moisture differences in em (longitudinal averages)for 2 x C02 - 1 x C02 (afer Washington and Meehll984)

In addition to the global scale approaches, numerous authors analyzed the regional hydrological impacts of global climatic change. Gleick (1986) draws the conclusion that in North California in the case of the forecasted climatic changes

Soil moirture, texture, struclure and erosion

45

the soil moisture reserve will decrease with 8-44%, resulting in serious drought induced yield reduction. There will be a certain decrease in summer runoff but a dramatic increase of winter runoff, leading to serious environmental and socioeconomic consequences. The majority of the investigations has pointed out that in most cases 10% decrease of rainfall results in more than 10% reduction in surface runoff. Since the water storage capacity of the soil is relatively constant, most part of the reduced precipitation can be stored in the soil. However, a 10% increase of precipitation may result in more than 10% increase in surface runoff, because of the relatively constant, and limited water storage capacity of the soil. Spatial and temporal variation of surface runoff can be as high as (or even higher) the variation in precipitation, because of the modifying effects of soil cover and canopy structure. Climate, vegetation and water regime influence soil properties. It depends on the character and intensity of climatic, vegetation and moisture regime changes and the changeability of soil characteristics, which are summarized in Table 4.1. As can be seen from this table, soil texture (particle size distribution) is a rather constant soil parameter: characteristic response time is lo3 years. Physical and chemical weathering are slow processes. Climatic changes result in only slight and slow process changes, and only moderately influence the "biological weathering". The influence is more expressed in the texture differentiation within a soil profile and in the organic matter turnover. The effects of the four basic potential climate scenarios on these processes are illustrated in Fig. 4.4. The influence of climatic change on soil structure (type, spatial arrangement, stability) is a more complex process with numerous direct and indirect impacts. The most important direct impact is the aggregate destructing role of raindrops, surface runoff and filtrating water. The rate of structure damage depends on the intensity of the destroying factor and the stability of soil aggregates against these actions. The indirect influences act through the vegetation pattern and land use practices. The consequence of vegetation changes on soil structure can be both favorable (tundra-forest; forest-grassland) and unfavorable (desertification, waterlogging, salinity-alkalinity). The same is true for land use. The impact of overgrazing, irrational land use, misguided agricultural utilization (cropping pattern, crop rotation) and improper agrotechnics (heavy machinery, over tillage, over imgation) is unfavorable, practically non reversible and hardly correctable. However, rational land use, proper agrotechnics and amelioration practices may help the maintenance or restoration of good soil structure. Pedogenic inertia will cause different time-lags and response rates for different soil types. Soil changes will be more rapid and profound in the younger or less weathered sediments of the glaciated or desert fringe region of the Northern Hemisphere and slower or less profound on the stable, continental shields of the equatorial region.

Table 4.1

P

Time changebility of various soil characteristics

Time changeability categories and their characteristic time 1 < 10-l yr

Soil parameter

bulk density; total porosity; compaction moisture content; infiltration rate; permeability; composition of soil air; nitrate content

aeration; heat regime

2

10-l-lOo y r

microbiota

microbial activity; human controlled plant nutrient regime

3

10°-lO' yr

total water cap. field capacity; hydraulic conductivity; pH; nutrient status; composition of soil solution wilting percentage; soil acidity; cation exchange capacity; exchangeable cations; ion composition of extracts

type of soil stmcture; annual roots biota; meso-fauna; litter, fluvic, gleyic. stagnic properties; slickensides

sulphuric h o r i m ; gelundic, inundic, salic, yennic phases (fine earth properties only)

4

101-102 yr

specific surface clay mineral association; organic matter content

soil biota tree roots; salic, calcareous, sodic, vertic properties

5

102-103 y r

primary mineral composition; chemical composition of mineral

tree roots; colour (yellowish/ reddish); iron concretions; soil depth; cracking; soft powdered lime; indurated sub-soil parent material; depth; abrupt textural change

histic (<20cm), ochric, gypsic. albic. and immature natric and spodic horizons (Podsols); gilgai, placic, sodic. takyric phase histic, mollic, umbric, calcic, albic, natric, cambic, spodic, and nitic horizons; plinthite, placic, yermic phases (stone surfaces) argic, oxic. petrocalcic, petrogypsic horizons; duripan, fragipan, skeletic, petroferric. lithic,

Part

6 > 103yr

texture; particle-size distribution; SP. hy; particle density

Properties and characteristics

o\

Horizons and phases

Regimes

moisture; natural fertility; salinityalkalinity; permafrost

P

?

Soil moisture, texture. structure and erosion

47

= precipitation = irrigation water = surface runoff (in and out) = filtration in the unsaturated

zone (in and out) = groundwater flow (in and out)

= rise of water table = lowering of the water table = infiltration into the soils = infiltration into the groundwater = storage within the soil (recharge) = filtration to the plant roots,

uptake by plants

T E

= transpiration

C

=

= evaporation

capillary transport from groundwater

Climate Factors

Cold, wet

Cold, wet

Hot. wet

Hot, wet

P

I I

D dD

I

D

R

1

D

G

i

1

I

I

d d

1

i i

D

(i) (i)

D D

E E

D D D D I I

S

E T F Gr GS

d

1

E i (i)

1 1

I

i and I = slight and great increase; d and D = slight and strong decrease; E = no change (equilibrium).

Fig.4.4 The components of the field water balance and soil moisture regime and the influence of four potential climatic scenarios on thesefactors

48

G . Y . Varallyay

If temperature increases, a warming up of the Northern Eurasian and North American permafrost plains with their loamy to silty sediments, all the shallow, imperfectly to poorly drained soils of the tundra and the Northern boreal forest biomass, will be radically changed by the melting of huge amounts of ice. The peat soil of the polar and boreal zones will shrink and slowly disappear due to the increased rates of decomposition of the organic matter. The podzolized soils of the tundra and boreal forests, which are derived from (peri)glacial sands and coarse crystalline rocks will turn into more acid and more leached variants. If precipitation increases, the heavy textured soils of present day tundra, boreal/ and humid temperate regions (some Luvisols, Podzoluvisols) will develop gleyic features in their topsoil, turning them into pseudogleyic/stagnic variants. As was described before, there are no linear relationships between mean annual precipitation, surface runoff and the rate of denudation/erosion. The rate, type and extension of soil erosion depends on climate (primarily the quantity and intensity of rainfall), relief, vegetation (type, continuity, density), and soil erodibility characteristics (La1 1988). The starting point of a comprehensive erosion-risk assessment can be the rate of surface runoff (Washington and Meehl 1983) or the sediment losses in the various major river basins (Lal et al. 1989). Some general conclusions on the possible influences of climatic changes on soil erosion can be summarized as follows:

Higher precipitation may cause an increasing rate of erosion; higher runoff, if it is not balanced by improved soil conservation impacts of stronger vegetation due to better water supply; Lower precipitation generally reduces the rate of erosion, but it can be counterbalanced by the less intensive soil conservation influence of poor vegetation due to the non adequate water supply to plants; this can be the consequence of increasing temperature as well; Lower precipitation (or higher temperature) may intensify wind erosion; Increasing temperature may reduce the erosion hazard moderating the permafrost influence (limited infiltration rate and water storage capacity of the soil), but may considerably increase the erosion-risk reducing the snowhain ratio in the cold regions and in high mountains. More detailed studies are required for the quantitative characterization of these relationships and for the improvement of prediction models expressing the influence of climatic changes on soil erosion processes.

REFERENCES Blasting, T.J. and A.M. Solomon (1984). Response of the North American corn belt to climatic warming. Progress in Biorneteorology, 3:311-321.

Soil moirture. texture, structure and erosion

49

Bolin. B, B.R. Doos, J. Jager and R.A. Warrick (Eds.) (1986). The greenhouse effect, climatic change, and ecosystems. SCOPE 29.. Wiley, New York. Earth System Science. A closer view. NASA, Washington, 1988. Gleick, P.H.( 1986). Methods for evaluating the regional hydrologic impacts of global climatic change. Journal of Hydrology 88:97-116. Gleick, P.H.( 1987). Regional hydrologic consequences of increases in atmospheric C 0 2 and other trace gases. Climate Change 10: 137-161. IGBP (1988). A study of global change. A plan for action. Global Change IGBP Report No.4. IGBP (1989). Effects of atmospheric and climate change on terrestrial ecosystems. Global Change ICBP Report No.5. Lal, R. (Ed.) (1988). Soil erosion research methods. Soil and Water Conservation Society. Ankeny, Iowa. Manabe, S a n d J.L. Holloway Jr.(1975). The seasonal variation of the hydrologic cycle as simulated by a global model of the atmosphere. J. Geophys. Res. 80. 1617-1649. Manabe, S.and R.T. Wetherald (1986). Reduction in summer soil wetness induced by an increase in atmospheric carbon dioxide. Science 232626-628. Parry, M.1.. T.R. Carter and N.T. Konijn (1990). The impact of climatic variations on agriculture. Vol. 1.: Assessment in cool temperate and cool regions. V01.2. Assessment in semi-arid regions. Reidel, Dordrecht. Solomon, S.I., M. Beran and W. Hogg (Eds.) (1987). The influence of climate change and climatic variability on the hydrologic regime and water resources. IASH Publication, No. 168. Szabolcs, 1 and M. Redly (1989). State and prospects of soil salinity in Europe. Agroktmia 6s Talajtan, 38: 537-558. Toward an understanding of global change (1988). National Academy Press, Washington, D.C.. Vituki (1989). The influence of climate variability and potential climatic change on the hydrologic cycle and water regime. A Hungarian Case Study. Budapest. Washington, W.M. and G.A. Meehl (1983). General circulation model experiments on the climatic effects due to a doubling and quadrupling of carbon dioxide concentration. J. Geophys. Res. 88:6600-6610. Washington, W.M. and G.A. Meehl (1984). Seasonal cycle experiments on climate sensitivity due to a doubling of C 0 2 with an atmospheric general circulation model coupled to a simple mixed-layer ocean model. J. Geophys. Res. 89:9475-9505. WCAP-4 (1987). Water resources and climatic change: sensitivity of water-resource systems to climate change and variability, World Climate Programme Applications, WMO, Norwich, U.K. WCRP-5 (1988). Concept of the global energy and water cycle experiment. World Climate Programme Research. WMO, Montreal 1987 and Pasadena 1988. WCRP-22 (1988). The global water runoff data project. Workshop on the global runoff data set and grid estimation. World Climate Programme Research. WCRP-22. WMO. Koblenz.

This Page Intentionally Left Blank

51

Chapter 5 RESILIENCE AGAINST CLIMATE CHANGE? Soil minerals, transformations and surface properties, Eh, pH Robert Brinkman AGLS, FAO, Via delle Terme di Caracalla. Roma 00100, I d y

ABSTRACT The main soil changes directly resulting from climate change would be in soil temperature regimes and soil hydrology. The minor soil temperature increases in the tropics to moderate increases in temperate and cold climates would modify organic matter dynamics, and may increase soil reduction in high latitudes where part of the permafrost would disappear. Increased variability and greater incidence of high-intensity rainfall events would entail higher leaching rates in most soils, with more runoff on sloping soils (and increased erosion and sedimentation). There would also be greater extents of periodic soil reduction in humid climates and salinization in semiarid or arid climates. The more rapid leaching, mainly in the tropics and in high latitudes, could accelerate the processes of hydrolysis and cheluviation; larger soil areas becoming periodically reduced could also become subject to ferrolysis. The influence of climate change on soil reaction would be generally minor. Soil changes would be least in a temperate climate. Soils with maximum resilience against degradation by these processes have a high infiltration rate and high structural stability, at least moderate cation exchange capacity and anion sorption, and moderate or good external drainage. Most of these properties are found in certain Nitisols as found near Nairobi, which have oxidic surfaces of the clay fraction. Soils least resilient against such changes have amorphous siliceous material coating the surfaces of the clay fraction, as in certain thixotropic Andosols or in quick clays in Canada or Scandinavia. The majority of the world's soils has intermediate properties. Management measures to increase soil resilience against the effects of climate change should increase vegetative cover and soil faunal activity, the same measures as against several other perturbations and stresses caused by human exploits. These human impacts are potentially more damaging than the direct effects of climate change on soils.

INTRODUCTION The main potential changes in soil forming factors (forcing variables) directly resulting from climate change would be in soil temperature regimes and soil hydrology, the latter because of shifts in rainfall zones as well as changes in potential evapotranspiration. Soil changes because of a potential rise in sea-level resulting from a net reduction in Antarctic ice cap volume and ocean warming are discussed in Brammer and Brinkman (1990). The biggest single qualitative change in soils expected as a result of these postulated forcing changes would be the poleward shift of the permafrost boundary, discussed in Chapter 17 by Goryachkin and Targulian (1990). Other

52

R . Brinkman

changes would be in degree rather than in kind. The changes in temperature but particularly in rainfall to be expected as a result of global warming are subject to major uncertainties, because different global circulation models do not lead to mutually consistent results (an example for Europe in Santer 1985), they are not yet adequately verified, and the interaction with changes in location and intensity of major ocean currents and resultant possible modifications in sea surface temperatures still is most uncertain, as well as the interaction with possible major changes in cloudiness and land cover and the resulting changes in albedo and actual evapotranspiration. Indirect effects of climate change on soils through sea-level rise, decrease or increase in vegetative cover, or a change in human influence on soils because of the changes in options for the farmer, for example, may each well be greater than direct effects of higher temperatures or greater rainfall variability and larger or smaller rainfall totals.

POSSIBLE CHANGES IN FORCING VARIABLES With these caveats, one could stipulate the following changes in forcing variables as likely to materialize sometime during the next century. Minor increases in soil temperatures in tropics and subtropics; moderate increases and extended periods in which soils are warm enough for microbial activity (warmer than about 5°C) in temperate and cold climates, parallel to the changes in air temperatures and vegetation zones summarized by Emanuel et al. (1985); Minor increases in potential evapotranspiration in the tropics to major increases in high latitudes caused both by temperature increase and by extension of the growing period; Increases in amount and in variability of rainfall in the tropics; possible decrease in rainfall in a band in the subtropics poleward of the present deserts; minor increases in amount and variability in temperate and cold regions; Increased frequency and severity of cyclonic storms in the present cyclone belts and some poleward widening of these belts because of increased sea-surface temperatures, giving rise to greater frequencies of high-intensity rainfall events.

EFFECTS ON SOILS Tropical areas In the tropics, increased intensities of rainfall events and increased rainfall

Soil resilience against climate change?

53

totals would increase leaching rates in well drained soils with adequate infiltration rates, and would cause temporary flooding or water saturation in many soils in level or depressional sites. They would also give rise to greater amounts and frequency of runoff on soils in sloping terrain, with sedimentation down slope and, worse, downstream. There would be increased chances of mass movement in the form of landslides or mudflows in certain soft sedimentary materials, discussed below. Soils most resilient against such changes would have adequate cation exchange capacity and anion sorption to minimize nutrient loss during leaching flows, and have a high structural stability and a strongly heterogeneous system of continuous macropores to maximize infiltration and rapid bypass flow through the soil during high intensity rainfall.

Subtropical areas In subtropical areas, decreasing rainfall totals with increasing intra and interannual variability could lead to less dry matter production and hence in due course, lower soil organic matter contents. Periodic leaching during high intensity rainfall with less standing vegetation could desalinize some soils in well drained sites, cause increased runoff in others, and lead to soil salinization in depressional sites or where the groundwater table is high. Soils most resilient against the effects of such increasing aridity and rainfall variability would have a high structural stability and a strongly heterogeneous system of continuous macropores, hence a rapid infiltration rate, as well as a large available water capacity -but less than the total for a high rainfall event expected with a frequency of about once per year-; and a deep groundwater table.

Temperate climates In temperate climates, minor increases in rainfall totals would be expected to be largely taken up by increased evapotranspiration of vegetations or crops at the expected higher temperatures, so that net hydrologic or chemical effects on the soils might be small. The negative effect on soil organic matter contents of a temperature rise might be almost compensated by the greater organic matter supply from vegetations or crops growing more vigorously because of the greater potential evapotranspiration and the higher transpiration efficiency in a high CO2 atmosphere. The temperate zone would thus be likely to have the smallest changes in soils, even in poorly buffered ones, directly caused by the effects of global warming. A minor and probably slow, but very visible, change could be a reddening of presently brown soils where increased periods with high summer temperatures would coincide with dry conditions, so that haematite would be stable over the presently dominant goethite. This mineralogical change might decrease the intensity and amount of phosphate fixation. An overview of such changes, with emphasis on temperate climate zones, is given by Buol et al.

54

R. Brinkman

(1990).

Boreal climates In boreal climates, the gradual disappearance of large extents of permafrost and the reduction of frost periods in extensive belts adjoining former permafrost are expected to improve the internal drainage of soils in vast areas, with probable increases in leaching rates. The appreciable increase in period when the soil temperature is high enough for microbial activity would lead to lower organic matter contents, probably not fully compensated by increased primary production through somewhat higher net photosynthesis and a longer growing period. Paradoxically, the extent of soils subject to periodic reduction could well increase in level areas, in spite of the greater leaching, because of increased periods when the soils are water saturated but also sufficiently warm for microbial activity. Soils most resilient against such effects, including the leaching of nutrients and periodic soil reduction, would have similar characteristics as the most resilient ones in the tropics: adequate cation exchange capacity and anion sorption to minimize nutrient loss during leaching flows, a high structural stability and a strongly heterogeneous system of continuous macropores to maximize rapid bypass flow during periods with excess meltwater.

PROCESSES IN SOILS Loss of salts and nutrient cations is the most rapid process of the various chemical and mineralogical changes that can take place with increased leaching. Normally, the clay mineral composition of a soil and, particularly, the mineralogy of the coarser fractions, would change little even over periods measured in centuries. Only in rare cases would the bulk of the clay fraction change appreciably over periods of a few years, as for example, when the largely amorphous clay fraction in a perennially moist, young volcanic soil under rain forest is suddenly subjected to periodic drying after deforestation. Then, halloysite (a member of the kaolinite group) may be formed very rapidly, with a consequent decrease in available water holding capacity and cation exchange capacity. Another relatively rapid change would be the dehydration of goethite to haematite mentioned above. However, changes in the surface properties of the clay fraction of soils may be much faster than changes in its bulk composition or crystal structure. Such changes in surface properties leave the bulk composition substantially unchanged but have a dominant influence on the physical and chemical soil characteristics. These changes, as well as the much slower changes in the bulk clay mineralogy, are brought about by a small number of transformation processes (Brinkman 1982, 1985), summarized below in order of decreasing areal importance:

Soil resilience against climate change?

55

Hydrolysis by water containing carbon dioxide, which removes silica and basic cations; Cheluviation, which dissolves and removes especially aluminium and iron by chelating organic acids; Ferrolysis, a cyclic process of clay transformation and dissolution influenced by alternating Fe reduction and oxidation, which decreases the cation exchange capacity by A1 interlayering in swelling clay minerals; Dissolution by strong mineral acids, which attack all clay minerals, producing acid A1 salts and amorphous silica; Clay transformation under neutral to strongly alkaline conditions (reverse weathering), which creates minerals such as montmorillonite, or in extreme cases palygorskite or analcime. Each of these processes changes the buffer system of the soils. Except for the last process, they decrease the cation exchange capacity of the clay fraction; hydrolysis and acid dissolution tend to increase phosphate sorption capacity and intensity (fixation). Hydrolysis and cheluviation, in soils that were already subject to these processes, may be accelerated by increased leaching rates. Ferrolysis may occur wherever soils are seasonally reduced and oxidized while there is leaching during the reduced phase. This combination of circumstances may, in a warmer world, occupy larger areas than at present, especially in high latitudes and in humid monsoon climates. Each of these three processes may thus impoverish or degrade the soils in which it occurs - however, even at somewhat faster rates than at present, the effects would be appreciable only after centuries, or millennia for hydrolysis. Dissolution by strong acids would be expected if and as low lying coastal areas with pyrite rich clays would dry out more deeply during the dry season than at present - but this is not likely in view of the expected gradual sea-level rise concurrent with, or closely following, global warming. Reverse weathering may take place in areas drying out during global warming, and would continue in soils in most of the presently arid areas. This process is of limited practical importance, however, since the overriding limitation on use of the soil resources in these areas is lack of moisture rather than any chemical or mineralogical problem. Most of these processes occur both in aerated, oxidized and in reduced conditions. Ferrolysis is limited to periodically reduced soils. Cheluviation occurs in somewhat different forms: with accumulation of iron(II1) compounds in the illuvial horizon in well drained soils, but with virtually complete elimination of iron in poorly drained conditions. Similarly, hydrolysis eventually leads to soils dominated by iron(II1) and some aluminium oxides in well drained conditions, but to soils dominated by A1 oxides (bauxite) in conditions of poor drainage. The five processes mentioned above normally influence the clay mineral

56

R. B r i n h n

composition over long periods only, and modify surfaces of the clay fraction over intermediate periods of time (generally, centuries). Direct human action, rather than climate change, may vastly accelerate some of these processes, however. In parts of Europe, for example, acid rain has extracted the basic cations from large areas of sandy soils and, through acid dissolution, replaced these by amounts of exchangeable and soluble aluminium ions that are toxic to trees and undergrowth of the natural or planted forests, as discussed by Van Breemen (1990). A less extensive example is the extremely rapid ferrolysis in soils seasonally inundated by water level fluctuations in a large reservoir, described by Amatekpor (1989).

SOME PROPERTIES OF CLAY SURFACES Soils with a naturally high structural stability are relatively rare, even though they happen to be prevalent near the workshop site, Nairobi. The clay fraction in such soils generally has oxidic surfaces: mainly iron(II1) and A1 oxides or hydroxides, while the bulk of the clay fraction may have different compositions. The oxidic surfaces could form from parent materials with moderate or high iron contents under long-continued hydrolysis by water (containing carbon dioxide). At the other extreme are soils with a very low structural stability, or with a severe hazard of failure under load or shock (socalled quick clays). The surfaces of the clay minerals in these soils are generally covered by amorphous, gel-like material with a high silica content (McKyes et al. 1974). Such material may have originated in earlier periods when the soils were strongly saline and, presumably, subject to processes of reverse weathering. Examples are the quick clays of the Champlain Sea sediments in Ontario and Quebec and of parts of Scandinavia. Such soils are most likely to generate high proportions of runoff and suspended sediment, but also most liable to mudflows once sloping sites are water saturated to appreciable depth. Some Andosols are thixotropic and have similarly low stability because of their similar composition, derived from volcanic materials (tuff). Most soils fall somewhere between these extremes. Vertisols, for example, have moderate or low structural stability, and clay surfaces that are mainly silica, but with generally small amounts of amorphous coating. In Planosols, if formed by ferrolysis, the clay fraction in the upper, eluvial horizons has been partly decomposed with a residue of amorphous silica, but the remaining smectite or illite has been interlayered with aluminium hydroxide polymers, which has decreased the swell-shrink potential and the cation exchange capacity of the clay fraction. Concurrently,part of the free iron oxides have been reduced and leached out. The net effect of these changes generally is a decrease in structural stability.

Soil resilience againsf climafe change?

57

RESLIENCE AGAINST PHYSICAL SOIL DEGRADATION As discussed, most soils do not have a high intrinsic resilience against physical soil degradation by, e.g., high intensity rainfall. Under natural conditions in humid climates, it is the complete soil cover near ground level combined with the perforating activity of the soil fauna that makes the soilvegetation system resilient against physical degradation. In the Rhine river plain in the Netherlands, for example, most of the originally calcareous alluvial soils have been decalcified within a millennium or so. Only in small areas on the highest levees of that age, continually under forest, soils have remained calcareous, even with lime pseudomycelia (filaments) indicative of less humid soil conditions, and full of vertical macropores produced by earthworms. In these soils, faunal activity is high because of the adequate litter supply; the resulting macropores remain open, protected against rain impact by litter and undergrowth; and heavy rain passes to the substratum through the macropores without leaching lime from most of the soil mass. Soil and crop -including pasture- management that maintains soil cover and organic matter supply to soil biota, while minimizing mechanical disturbance by heavy traffic, cultivation or excessive grazing intensity, makes soils most resilient against the effects of climate change and other extreme events. Such physical management may also help conserve plant nutrients through decreased leaching, as illustrated in the natural example above. This does not imply that a single management recipe would be applicable in different conditions. Certain termite species harming crop performance may necessitate a period without residues on the soil; or crop residues may be needed for feed or fuel. Such factors, and others, are taken into account in designing an optimum management strategy for any specific natural and cultural environment.

RESILIENCE AGAINST SOIL REDUCTION Soil reduction, which would limit land suitability for dryland crops, or strong reduction which would be liable to produce toxins even for wetland crops, may take place once the soil is water-saturated long enough for microbial action to exhaust the oxygen remaining in the soil when water-saturation started. Another necessary condition is the presence of sufficient readily decomposable organic matter as an energy source for the microbial activity. In most soils, during reduction the redox status is stabilized at an Eh about 100-200 mV near neutrality by the Fez+ - Fe(OH)3 equilibrium, except where the content of readily decomposable organic matter is very high or the content of free iron(II1) oxides very low. In such cases, negative Eh values may occur, and toxic hydrogen sulfide or low molecular organic compounds -including methane- may be formed. Resilience against soil reduction in practice depends on the drainage

58

R. B r i n h n

conditions, since most soils have sufficient organic matter for reduction to start within about a week after water saturation. Soils most resilient against reduction in conditions of increased rainfall variability and incidence of high intensity rainfall have similar properties as those resilient against other effects of climate change or other perturbations: high infiltration rate, high structural stability and a permanent heterogeneous system of tubular macropores, good external drainage.

SOIL REACTION Most soils would not be subject to rapid pH changes resulting from climate change; exceptions might be found in potential acid sulfate soils subject to increasingly long dry seasons. Even though most of such soils are clays with moderate or high cation exchange capacity, the amounts of acid liberated in such soils upon oxidation generally exceed this rapid buffering capacity. Therefore, pH values may temporarily reach 2.5 to 3.5 and a small part of the clay fraction may be decomposed as indicated under 'processes', above. This then buffers the pH generally between 3.5 and 4 in the long run.Depending on the efficiency with which the excess acid formed can be leached out, the period of extreme acidity and aluminium toxicity may last between less than a year and several decades. In calcareous soils, soil reaction may range between about 8.5 and 7 depending on the C02 partial pressure in the soil; this range is maintained against leaching of basic cations by the different soil processes as long as a few percent of finely distributed lime remain. Buffering in noncalcareous soils is less strong, but depends on the cation exchange capacity at soil pH. In soils with variable charge surfaces of the clay fraction, this decreases with acidification. It should be noted that the simple modelling of accelerated CaC03 leaching under a doubled atmospheric C02 concentration generally does not hold true. In most soils, the ongoing decomposition of organic matter maintains C02 concentrations in the soil air far above atmospheric concentration even now, and C02 in soil air determines CaC03 solubility. Leaching of lime is thus positively related to rate of organic matter decomposition, negatively to gas diffusion rate, and positively to amount of water percolating through the soil. In conditions where leaching is accelerated by climate change, it would be possible to find relatively rapid soil acidification after a long period with little apparent change, as has been the case -but more rapidly- in some soils in Europe that have been subject to acid rain for several decades. The soil might in fact be steadily depleted of basic cations, but a pH change may start, or may become more rapid, once certain buffering pools are nearly exhausted. Such non-linear and time-delayed effects have been discussed in the context of soil and water pollution by Stigliani (1988); they are also expected to occur in various ways at different times after increased temperatures and changed rainfall patterns will have been operative.

Soil resilience against

climate change?

59

CONCLUSION The soil changes expected as a direct result of climate change are generally minor or relatively well buffered by the mineral composition, the organic matter content or the structural stability of most soils. Major soil changes would be expected where permafrost would disappear. Climate induced changes such as major decreases in cover by vegetation or annual or perennial crops, could degrade structure and decrease porosity in most soils, as well as increase runoff and erosion on sloping sites and aggravate the concomitant sedimentation. Changes in options available to land users because of climate change may have similar major effects. In most cases, changes in soils by direct human action, whether intentional, on site or off site (and mostly unintended), are far greater than the direct climateinduced effects. Soil management measures designed to optimize the soil's sustained productive capacity would therefore be generally adequate to counteract degradation of agricultural land by climate change. Soils of nature areas, or other land with a low intensity of management such as seminatural forests used for extraction of wood and other products, are less readily protected against the effects of climate change but such soils, too, are threatened less by climate change than by human actions - off site, such as pollution by acid deposition, or on site, such as excessive nutrient extraction under very low-input agriculture. To armour the world's soils against the effects of climate change, or other extremes in external circumstances such as nutrient depletion or excess (pollution), or drought or high intensity rains, the best that land users could do, would be: Managing their soils to give them maximum physical resilience through a stable, heterogeneous pore system by maintaining a closed ground cover as much as possible; Using an integrated plant nutrient management system to balance the input and offtake of nutrients over a cropping cycle or over the years, while maintaining soil nutrient levels low enough to minimize losses and high enough to buffer occasional high demands. An analogous philosophy, at lower levels of external inputs, could be formulated for extensive grazing land and production forest, whether planted or managed natural forest. Human action and management has been emphasized in these conclusions because most of the world's land is used and, to different degrees, managed rather than under natural conditions.

60

R. B r i n h n

ACKNOWLEDGEMENTS Critical and constructive comments on a drafi of this paper by J.R. Benites, G.M. Higgins, W. Klohn, F. Nachtergaele and D. Norse are gratefully acknowledged.

REFERENCES Amatekpor, J.K. (1989). The effect of seasonal flooding on the clay mineralogy of a soil series in the Volta lake drawdown area, Ghana. Land degradation and rehabilitation 1: 89- 100. Brammer, H. and R. Brinkman (1990). Changes in soil resources in response to a gradually rising sea-level. (Chapter 12, this volume). Brinkman, R. (1982). Clay transformations: aspects of equilibrium and kinetics. Ch. 12, p 433-458 In: Bolt, G.H. (ed.) 1982. Soil Chemistry, B. Physicochemical models. Developments in Soil Science 5B, 2d ed. Elsevier, Amsterdam. Brinkman, R. (1985). Mineralogy and surface properties of the clay fraction affecting soil behavior and management. p 161-182. In: Woodhead, T. (ed.) 1985. Soil physics and rice. International Rice Research Institute, Los Banos, Philippines. vi + 430 p. ISBN 97 1- 104-146-4. Buol, S.W., P.A. Sanchez, J.M. Kimble and S.B. Weed (1990). Predicted impact of climatic warming on soil properties and use. In: American SOC.Agron. Special Publ. (in press). Emanuel, W.R., H.H. Shugart and M.P. Stevenson (1985). Climatic change and the broadscale distribution of terrestrial ecosystem complexes. Climatic change 7: 29-43. Goryachkin, S.V. and V.O. Targulian (1990). Climate induced changes of the boreal and subpolar soils (Chapter 17, this volume). McKyes, E., A. Sethi and R.N. Yong (1974). Amorphous coatings on particles of sensitive clay soils. Clays and clay minerals 22: 427-433. Santer, B. (1985). The use of general circulation models in climate impact analysis - a preliminary study of the impacts of a COz-induced climatic change on West European agriculture. Climatic change 7: 71-93. Stigliani, W.M. (1988). Changes in valued "capacities" of soils and sediments as indicators of nonlinear and time-delayed environmental effects. Environmental monitoring and assessment 10: 245-307. Van Breemen, N. (1990). Impact of anthropogenic atmospheric pollution on soils (Chapter 11, this volume).

61

Chapter 6

IMPACT OF CLIMATIC CHANGE ON SOIL ATTRIBUTES Influence on salinization and alkalinization I. Szabolcs Research Institute for Soil Science and Agricultural Chemistry of the Hungarian Academy of Sciences Herman Otto ut 15. Budapest, Hungary.

INTRODUClTON If the predicted climatic changes due to an increase of atmospheric carbon dioxide and other "greenhouse gases" develop, they will definitely alter the entire environment, including the soils of our globe. The expected changes will surpass human impact on the soil in the past and will affect many soil attributes, both directly and indirectly. Part of the changes will manifests itself in soil morphology, in physical, chemical, physico-chemical and biological soil properties and eventually in the fertility and productivity of soils. Several changes are unpredictable, while others are more or less precisely predictable. Soil salinization and alkalinization, influenced to a high degree by the greenhouse effect and subsequent climate changes, belong to such predictable processes (Szabolcs 1989).

PRESENT GLOBAL STATUS OF SOIL SALINITY AND ALKALINITY About 10% of the total surface of the continents is covered by different kinds of salt affected soils, mostly by saline and alkali types. No continent is free from such soils; even in the Antarctic vast areas of saline soils have been discovered (Szabolcs 1979). In recent years salt affected soils have been described in many countries where they were formerly unknown. The global area of these soils is increasing partly due to the extension of irrigation, overgrazing, etc. Salt affected soils occur in different climatic conditions though their appearance is most frequent in arid and semiarid areas, and in bottomlands and river valleys (Polynov 1956; Szabolcs 1987) (see Table 6.1).

POSSIBLE INFLUENCE OF PREDICTED CLIMATIC CHANGES The predicted climatic changes will increase the areas with saline and alkali soils mainly for two reasons:

62

I . Szobolcs

The direct effect of climate on salt movement and the salt balance of soils by increasing the temperature and aridity, on the water and salt balance of the watershed and on the extent of coastal salinization due to the rise of the sea level. The indirect effect on salt dynamics by changing the land use pattern e.g. by extending the area of imgation as a consequence of the climate changes. Table 6.1 Salt affected soils on the continents and subcontinents (without the Antarctic) (Sub) Continent

North America Mexico and Central America South America Afnca South Asia North and Central Asia South East Asia Australia Europe Total

Thousand ha

15 755 1 965 129 163 80 538 87 608 211 686 19 983 357 330 50 804 954 832

It is well known that the cause of salinization and alkalinization is the accumulation of salts in soil horizons and, as its consequence, the development of soil properties unfavorable for production. Salt movement and the salt balance of a given territory is closely related to the geochemistry, hydrology and climate of the given area. As a rule, soil salinity is common in dry areas and seldom occurs in well-drained humid places (Fersman 1934; Kovda 1947). Where the climate gets hotter and/or drier, accumulation of salts can be expected and their leaching from the soils will decrease. The consequence of this phenomenon will be soil salinization and/or alkalinization in many places where they were unknown before (Dregne 1976). If the predicted climate changes indeed occur, substantial alterations will take place in the practice of land use, agro-, sylvi- and horticulture, etc. One of the main consequences of climatic changes will be the necessity to increase irrigated agriculture because the higher temperatures and subsequent increased aridity will necessitate the use of additional water for agricultural production (Alekseevsky 1971).

Influence on salinization ond alkalinization

63

It is also well known that one of the most dangerous side effects of imgation is secondary salinization and alkalinization of soils, due to the alteration of the natural salt balance of the environment. UN data show that more than half of the existing irrigation systems of the world are affected by this adverse process and every year several million hectares of irrigated land are abandoned from agricultural production because of salinization and alkalinization. It is evident that in case of the extension of irrigation the hazard of soil salinization and alkalinization will also increase (Kovda 1980). While it is difficult to predict quantitatively many of the possible adverse soil processes which can develop under the influence of climatic changes, we do have the knowledge to elaborate more or less reliable predictions on the possible occurrence and extension of soil salinization and alkalinization in most of the areas that may be affected. It is a pity that studies for such predictions started quite reluctantly and between very narrow limits. However, it is necessary to characterize as soon as possible the anticipated processes, together with their geographical distribution, extension and necessary measures. In the past few years we developed some methods for the prediction of the extension of soil salinization and alkalinization as a consequence of predicted climatic changes for a large part of Europe, though this continent is comparatively less affected by salt than any of the others (Szabolcs 1974).

SCENARIOS FOR THE CHARACTERIZATION OF POTENTIAL SOIL SALINIZATION IN EUROPE For the studies of potential soil salinity in Europe we assumed that, due to different factors both natural and man-made, several major processes can result in salinization in different parts of Europe in the next half century. We selected the following three scenarios which represent the most important processes of potential salinization in different parts of our continent: Potential soil salinity caused by climatic changes (Scenario 1); Potential soil salinity caused by sea-level rise (Scenario 2); Potential soil salinity caused by extension of imgation (Scenario 3). Different regions of Europe were selected to test the above scenarios. For Scenario 1 the Mediterranean Region of Europe was selected where possible changes in climate, such as increasing temperature and decreasing precipitation may result in secondary salinization. For Scenario 2 certain coastal areas of Northwestern Europe were chosen where possible sea-level rise may cause remarkable soil salinity. For Scenario 3 regions with continental and semiarid climates were chosen where the major hazards of secondary salinization are present and imgation is extending.

64

I . Szabolcs

The location of the areas of the three Scenarios is shown in Fig. 6.1. .It can be seen that the three Scenarios cover a large part of Europe, with potential salinity caused by different direct or indirect man-made salt accumulation processes.

Fig. 6.1

Location of Scenarios I 2 and 3 in Europe. ~

SCENARIO 1 Potential soil salinity caused by climatic changes For Scenario 1 we selected the major Mediterranean areas of Europe (see Fig. 6.1), where at present salt affected soils occur; mainly in the Iberian Peninsula and only to a smaller extent in Southern France, Italy, Sicily, Sardinia and Corsica, as well as in the Dalmatian coast of the Balkan Peninsula. The total area of this Scenario is 1,979,959 km2, with 885,826 km2 land surface. In comparison with the salinity conditions of Scenario 3, the extension of salt affected soils in Scenario 1 is much lower, it is nearly half of the percentage. We assumed that the possible climatic changes, due to C 0 2 accumulation and other causes, will increase the average annual temperature of the area by about 1°C in the next 50-70 years. As a consequence, the aridity index will also increase, which creates progressive salinity in those marginal areas where at

65

Influence on salinizaiion and alkulinizarion

present salinity does not exist or can be found only in a latent form in soils or in waters. The following assumptions were taken into consideration: The currently irrigated area will not change substantially; No natural disasters or tectonical changes will occur, The present riverbeds and major hydrological conditions will remain unchanged. Fig. 6.2 shows the area of existing salinity, and potential salinity as predicted in Scenario 1.

Mediterranean Sea

Iexisting

salinity potential salinity

Fig 6.2

0

200

400

kin

Area of salt affected soils and potential salinity as a consequence of climatic changes (Scenario I )

The area of potential salinity is estimated at about 122,000 km2 (13.8% of the total land surface), which is twice as much as the area of existing salinity. The dry areas of the Iberian Peninsula (like Castilia, the Ebro Valley), southwestern France and also several areas in the Italian and Balkan Peninsulas are particularly exposed to potential salinity due to the increasing aridity. It must be noted that if the increase in the average annual temperature will surpass IoC, the increase in salinity will not be linear, but rather exponential compared to the information in Fig. 6.2. The salinity in those areas threatens very fertile agricultural lands and must be predicted in order to instigate necessary preventive measures in good time, before the climatic changes occur. It should be noted that the potential salinity in Scenario 1 equally endangers river valleys and estuaries (Ebro, Neretva, Rhone, Guadalquiver, Tajo, etc.) as plains and plateaus (Castilia, Aragonia, Umbria).

66

I . Szabolcs

SCENARIO 2 Potential soil salinity caused by sea-level elevation For Scenario 2 we selected a part of North-Westem Europe (Fig. 6.1) where the following two conditions are favorable for the study: Measurable extension of present salt affected soils; Good probability of sea-level rise due to global climatic changes in the next 50 years. The total area of Scenario 2 is 344,942 km2, with 226,393 km2 land surface, including the southeastern part of England, the Western part of the Netherlands, Belgium and the Northeastern part of France bordering, the North Sea, the Atlantic Ocean and the English Channel. In this Scenario mainly coastal saline soils with high sodium chloride content occur in several seashores and adjacent areas. The area of existing and potential salinity in Scenario 2 is illustrated in Fig. 6.3. It was assumed that - due to global climatic changes - the expected sea-level rise will be 1 cm per year on average, due to which, the effect of seawater on land will provoke further salinization. Changes in aridity and humidity were not considered. However, the following points were taken into consideration: Irrigation will not be extended; No natural disasters or tectonical changes will occur, Present riverbeds and major hydrological conditions remain the same.

Mexisting salinity a potential salinity

Fig. 6.3

0

.

200

400

km

Area of salt affected soils and potential salinity as a consequence of sea-level rise (Scenario 2 )

As shown in Fig. 6.3, the area of potential salinity caused by sea-level rise again substantially surpasses that of existing salinity. This is also valid for the two other scenarios, proving that all over Europe potential salinity constitutes a much greater hazard than existing salinity.

Influence on salinizarion and alkaliniealion

67

For Scenario 2 - as it was mentioned above - those areas were selected where salinity caused by seawater is essential (10,520 km2) and constitutes 4.65% of the total land surface. The area of potential salinity is 24,977 km2, which is 11.03%of the total land surface. As observed in Scenario 1, potential salinity in Scenario 2 is predicted in areas surrounding the existing salt affected soils. It should also be noted that if sea-level rise surpasses the values taken for granted in this study, the extension of potential salinity will again not be linear, but will increase in an even higher degree.

SCENARIO 3 Existing and potential soil salinity caused by irrigation For Scenario 3 we selected a part of Europe (Fig. 6.1) where, due to both climatic and economic conditions, irrigation has been practised for a long time. More than half of all imgated areas in Europe are situated in the area of Scenario 3. As can be seen from Fig. 6.4, salt affected soils are rather extensive. A large part of existing salt affected soils is situated in the vicinity of or even within irrigation systems . In nearly all countries included in Scenario 3 further extension of imgation has been envisaged. In countries with less precipitation, larger increases in irrigated areas have been assumed (e.g. the USSR, Bulgaria, Rumania) than in countries which do not suffer from aridity (Austria, Poland). At present, irrigated lands cover less than 20%, in most cases less than lo%, of the total agricultural land. The envisaged increase in irrigated areas is different in the various countries, but in no case will it be more than 100% for the next fifty-sixty years. In most countries concerned one of the main limiting factors of the extension of irrigation in the future is the shortage of good quality irrigation water. In Scenario 3, taking the above described circumstances into consideration, we assumed that up to the middle of the 21st century the irrigated area will at the most double and calculations were made accordingly. There are two main processes of salinization caused by irrigation: Salt accumulation in the soil originating from the salt in irrigation water, Salt accumulation in the soil originating from the salt in rising salty groundwater. As far as soil salinization from irrigation water is concerned there m acceptable regulations for quality control of imgation water. Subsequently, only a small part of present secondary salinization occurs as a result of the use of poor quality irrigation water. In the case of soil salinization from groundwater, the

68

1. Szobolcs

table of the salty groundwater (which sharply rises in most irrigated areas), unfortunately results in a large part of the current soil salinity and this process can be predicted as the most hazardous one also for the future. Considering climatic, physico-geographic, hydrological, agricultural and pedological aspects, we have compiled a map of Scenario 3, using the following assumptions: Until 2050 the present irrigated areas will increase, but not more than double; There will be no substantial changes in climate and no elementary disasters will occur; The present quality requirements for applied irrigation water will remain the same. Fig. 6.4 illustrates the extension of existing and potential soil salinity, following the above described conditions and assumptions. The area indicated on Fig. 6.4 is about 20% of the total land area of Europe, and represents areas intensively affected by both present and potential salinization, including large areas of the European part of the USSR, Rumania, Hungary, Austria, Yugoslavia, Czechoslovakia an Poland.

Iexisting

salinity

0

200

400

km

apotential salinity Fig. 6.4

Areas of salt affected soils and potential soil salinity as a consequence of extended irrigation (Scenario3)

Although Scenario 3 covers only about 20% of the land area of Europe, about 50% of all salt affected soils of the continent are found here. As shown in

Influence on solinizorion and olkolinizotion

69

Fig. 6.4, the area of potential soil salinity exceeds that of existing salinization by almost two times. The three scenarios, which have been discussed and demonstrated above, represent the main processes leading to the hazard of secondary salinization in Europe. However, they do not exhaust all the possibilities of this phenomenon, which may develop due to other factors like: changes in cropping pattern, intensive use of chemicals, changing farm management, etc. In order to predict the adverse salinization processes in more detail and more accurately, more appropriate methods for special surveys need to be developed. However, we already are in the possession of a number of such methods, as well as of methods for the prevention of salinization mainly in imgated agriculture.

REFERENCES Alekseevsky, E.E. (1971). Irrigation and drainage of the world. Izd. Kolos, Moscow (In Russian). Dregne, H.E. (1976). Soils of arid regions. Elsevier, Amsterdam. Fersman, A. E. (1934). Geochemistry. Leningrad (In Russian). Kovda, V.A. (1947). Origin and regime of salt affected soils. Vols.1. and 11. Izd Akad. Nauk SSSR. Moscow (In Russian). Kovda, V.A. (1980). Problems of combating salinization of irrigated soils. UNEP. Nairobi. Polynov, B.B. (1956). Selected papers. Izd. Akad. Nauk, SSSR. Moscow. (In Russian). Szabolcs, I. (1974). Salt affected soils in Europe. Research Institute for Soil Science and Agricultural Chemistry of the Hungarian Academy of Sciences. Martinus Nijhoff, the Hague Szabolcs, I. (1979). Review on research of salt affected soils. UNESCO, Paris. Szabolcs, I. (1987). The global problems of salt affected soils. Acta Agron. Budapest, Hung. 36. (1-2). 159-172. Szabolcs, I. (1989). Salt affected soils. CRC. Boca Raton, Florida, U.S.A.

This Page Intentionally Left Blank

71

CHAPTER 7

SOIL ORGANIC MATTER AND BIOLOGY IN RELATION TO CLIMATE CHANCE P.B. Tinker and P . Ineson Natural Environment Research Council Polaris House, North Star Avenue, Swindon SN2 IEU,United Kingdom

ABSTRACT Climate change, a.0. due to high atmospheric C 0 2 concentrations, is expected to produce higher temperatures, and largely unpredictable changes in rainfall. The temperature changes are likely to be least in the tropical zones and largest at high latitudes. Soil carbon changes are implicated both as part of the global carbon cycle, and as a main area of impact of the climatic changes. Increased C 0 2 may lead to more rapid plant growth when this is the currently limiting factor, but it seems unlikely to be so to an important extent except under intensive agriculture. Primary productivity and hence the input of plant residues may then become larger. In addition, higher concentrations of CO2 may change the quality of residues, and the water use efficiency of vegetation. The C 0 2 increase is unlikely to have any significant direct effect within soils, where the COz level is usually enhanced already. Primary productivity will respond to temperature change only where the latter is limiting, which is least likely in the tropics, but will have a major effect at high latitudes. Rainfall is much more likely to be important in the tropics. The most pervasive effects on soils will be where the whole character of the vegetation changes. Increased temperature will usually lead to more rapid breakdown of soil organic matter, and these processes have been modeled with promising results. The relatively small changes expected in the tropics appear unlikely to induce key changes, but the effects in high latitudes are expected to be much larger. Soils with high organic matter content due to low temperature and waterlogging could be strongly affected due to both temperature and hydrology changes. The combined effect of changes in organic matter input and in rates of organic material breakdown will set the final new equilibrium level of soil organic matter. The effects that are most difficult to predict are those resulting from the effects of rainfall changes, and the behavior of soils which are not in equilibrium with the climate. Changes in ecosystems and migration of vegetation zones are likely in some areas, and the speed with which this happens will be of critical importance. Soil flora and fauna, including symbiotic organisms, may be seriously affected by any such changes, as their migration rates are likely to be small.

INTRODUCTION The general outline of processes leading to climate change is now well known. However the climate predictions from the Global Circulation Models (GCMs) are still uncertain; in particular there are serious doubts about the effects

12

P . B . Tinker and P . lneson

of cloudiness, and the way in which land-atmosphere interactions should be included in the modeling. There is a reasonable agreement about global temperature changes, but the impacts of climate change will be felt more locally, and the current predictions of regional or national climate change are extremely uncertain. We are, therefore, still very much in the situation of working out the consequences of scenarios that may or may not happen. Here we start from the most recent predictions of the GCMs (Fig. 7.1) (Dickinson 1986; Trabalka 1985), which predicts temperature increases averaging 1.5-4.5OC. Such changes are likely to be accompanied by major alterations in winds, storms and rainfall. Despite the fact that these proceedings focus most on tropical areas, we cannot ignore the fact that all predictions agree that the temperature effects will be largest at the poles and least at the equator. The higher latitudes must therefore also be considered. We take as a reasonable range for a possible scenario that rainfall may change by about 20% at any point, though some changes may well be larger than this (Department of the Environment 1988). A point of general agreement is that the increase in temperature will be most marked in winter, so that cold winters, sea, ice, and snow cover are all likely to diminish.

SOIL AS PART OF THE CARBON CYCLE Soil organic matter occupies a particularly critical position in this topic. The carbon cycle is affected by climate change, and it has an important impact itself upon the changes via the COz in the atmosphere. The carbon in soil organic matter is a significant component of the earth's carbon reservoirs, with around 1 . 5 ~ 1 g0C~ (Solomon ~ et al. 1985) (Table 7.1). This is small compared to the carbon stored in the deep ocean (ca. 38x10 I8 g C and in the deep earth, but most of this is only in extremely slow equilibrium with the atmosphere, and the amount in the more rapidly equilibrating surface ocean layer is only some 0 . 6 ~ 1 0g' ~C. The turnover time of carbon in the soilvegetation system is variable, ranging from less than a year for annual vegetation, to thousands of years for organic matter in the deeper parts of soil profiles. The loss of carbon from both terrestrial biomass and soil organic matter has been considerable over the last 200 years. The calculation of the net release, over this whole period or currently per year, is very difficult due to the heterogeneity of the land surface, the variation in treatment or cover, and the uncertainty of the areas of particular ecosystems.

The current net release of C from soils and plants is estimated to be around l O I 5 g C yr-l, but the uncertainties in the data and the variable assumptions and

definitions used by different authors provide a very large margin of error.

73

Impact on soil organic matter and biology

(0

1

(b)

SURFACE AIR TEMPERATURE DIFFERENCES. DJF

SURFACE AIR TEMPERATURE DIFFERENCES, JJA 180

9

Fig. 7.1

mw

9oE

60 9oN

Predicted air temperature changes ("c)for a steady-state doubling of C02, showing winter (a) and summer (b) surface air temperature @-omBolin et al. 1986)

Estimates of the net annual flux between the atmosphere and terrestrial ecosystems (Houghton et al. 1985) range from -2.0 to 20.0 x 1015g C yr - l , a range that clearly indicates the gross uncertainties that persist in this area. The net release is about 20% of the annual flux of C02 from fossil fuel burning, of ca 5 x 1015g C y r l . At present it is estimated that some 40% of the total increase in

74

PB . Tinker and P. lnesan

atmospheric C02 to date has arisen from changes in the soil-vegetation system (Trabalka 1985). The total release since 1880 is estimated at 120 x 1015g C (Solomon et al. 1985). Table 7.1

Estimated values and uncertainties of key parameters in the global carbon cycle (from Solomon et al. 1985)

parameter

Net carbon flux from land biosphere since 1800 Gross annual terrestrial plant Co;!uptake Net primary production Annual tropical forest area conversion (1970-1980) Annual net carbon flux from land conversion (1970-1980) Contemporary soil carbon Contemporary biomass

Value 120 1015 c

Uncertainty range 90-180 x 1015 g c

120 1015 g c

90-120 1015 g c

0.004

45-62 1015 g c 0.003-0.006

1.3 1015 g c

0.0-2.6 1015 g c

1.5 x 10l8 g C 560 1015 g c

1.2-1.8x 10l8 g C 420-660 1015 g c

60

1015 g c

During the last century the major component of this was probably the loss of soil organic carbon from the cultivation of virgin grasslands, but now it is from the felling of forests, mainly in the tropics. A major complication occurs in peatlands, where the storage of C can be regarded as part of a soil reservoir, or as a separate compartment (subfossil C). Indeed, the whole question of carbon changes in wetland ecosystems is extremely interesting (Olson et al. 1985), because of the potential changes in coastal wetlands due to sea-level rise, as well as to temperature change and the production of CH4, another greenhouse gas, as well as C02 from such systems. With an annual net flux from soil and plants of some 1 x loi5g C and a total soil plus plant biomass carbon pool of around 2 x 10l8g C, there is plenty of scope for continuation or even acceleration of the present release of C02 from the terrestrial biosphere. The prediction of such changes is a matter of urgency.

THE ANALOGY APPROACH There appear to be two ways of approaching the problem of predicting the impact of climate change on soils, which we define as the 'analogy' and the 'process' methods. The analogy approach depends upon identifying a soil elsewhere which has developed under a climate that is very close to that expected at the site considered. There are several assumptions involved; the first is that climate is the dominant

75

Impact on soil organic matter and biology

factor in determining both vegetation and soil type. This may not be true if there are marked differences in parent material. Secondly, it assumes a fairly rapid reestablishment of equilibrium between soil and climate after a perturbation of l i ~ latter. In fact, different soil characteristics will alter at different rates (Fig 7.2), and the full development of a zonal soil in equilibrium with a climate takes thousands of years. Thirdly, this approach tends to assume that there is a relatively sharp change from one climate to another. In fact it is likely that this will be progressive, so that climate changes gradually, with soil and vegetation changes lagging behind to varying degrees. Nevertheless, with all these weaknesses, it is clearly a step forward to identify a soil that is the 'equilibrium' soil for a given climate, and then estimate the rate of progression towards this. NEAR EQUILIBRIUM

1 steady slate

-

Slowly adjusting features SOIL-P

Initial stale

Id 5

Fig 7.2

Id 6

Generalized diagram of the time taken to achieve steady state in soil properties after the initiation of soil development (from Walker and Graetz 1989)

The vegetation classification of Holdridge (1947 1964) has been used to predict the alteration in vegetation zones that would follow climate change (Bolin et al. 1979). This analysis is based solely upon a projected temperature change scenario, and any additional changes, such as in rainfall, would certainly cause larger effects. However, given these vegetation changes, their impact upon soils can be estimated at least qualitatively. One of the most interesting extensions of this system is the one of Zinke ct al. (1984), who plotted the carbon content in 3400 soil profiles on the Holdridge triangular diagram (Fig. 7.3), to produce a set of contours. In principle, at least, the equilibrium amounts of organic matter in a soil with a defined climate can be determined, though it is hard to believe that other factors, such as the texture of the soil, are not involved. The predicted temperature changes will produce shifts on this diagram that imply a smaller total pool of soil carbon; it would be possible

76

P.B. Tinker and P . Ineson

to sum this change over the terrestrial surface. However, the structure of the diagram shows clearly that the soil carbon store is much more sensitive to changes in rainfall than to changes in temperature.

2

Fig. 7.3

L 6 8

10

1 L 18

22

Soil carbon content (kg rn-2) plotted as contours on a Holdridge (1967)life-zone chart (source Zinke et al. 1984)

It should be remembered that very large areas contain soils that cannot be appropriately delimited simply by the biome in which they are contained. In particular, cultivation, management and cropping change both the profile and the carbon content in all farmed soils.

THE SOIL ORGANIC MATERIAL BALANCE - THE PROCESS APPROACH The components of the carbon cycle that determine soil organic matter levels (including subfossil remains such as peat) are: Net primary productivity of organic matter (NF'P); Carbon partition above and below ground; Breakdown rate of organic debris;

Impact on soil organic matter and biology

77

Formation of humus; Breakdown rate of humus. The soil organic matter level in any soil will settle at an equilibrium level characteristic of its permanent characteristics, the climate and the land use or vegetation cover. This concept is most clearly shown in the studies of the soil organic matter levels in the long-term experiments at Rothamsted (Johnston 1987), in which any change in treatment caused a corresponding change in the progression to a new equilibrium level. We draw a clear distinction between the decomposition of litter or debris, in which plant or animal structures can still be recognized, and soil organic matter, or humus. The latter is ultimately bound up with the mineral soil, has no recognizable structure, and has a fairly constant C/N ratio.

NET PRIMARY PRODUCTIVITY - INPUTS In general, increased atmospheric C 0 2 is expected to increase net productivity by more rapid photosynthesis. The present C02 concentration in the atmosphere, of ca 350 ppm, is suboptimal for plant growth in certain circumstances (Goudriaan and Ajtay 1979). The benefit to plants with the C4 photosynthetic system is very small or zero, because of their lack of photorespiration. In C3-plants there is undoubtedly the potential for increased growth, of perhaps 30%, if the C02concentration rises to 600 ppm. For a canopy or stand smaller values may be found than for single plants, because selfshading is less in the latter. In practice, growth of most crops or vegetation stands are limited by factors other than C02, e.g. drought, nutrient deficiency or disease. Even when such constraints are not overtly present, only a small fraction of even high yielding crops appear to reach the optimum yield set by radiation income (Tinker 1984). Any increase in crop yields is therefore uncertain, and major increases in net primary productivity in natural vegetation seem unlikely. However, if the growth constraint is due to water shortage, there may be a benefit from the more efficient use of water in plants whose stomatal behavior is controlled by CO2 concentration. The increased C02 concentration in the atmosphere should then increase the ratio of the flux of C02 into the leaf to the flux of water out of the leaf, unless the relatively lower use of water raises the leaf temperature sufficiently to counteract most of the potential effect. For a doubling of the C02 concentration, the water use per unit of photosynthesis could possible be halved (Squire and Unsworth 1988). Where water is limiting, this interaction of CO2 level and water supply is likely to be the most important effect in practice. There are other effects of increased atmospheric C 0 2 which may be important in some circumstances. The morphology of plants may be changed, in

78

P B . Tinker and P. Ineson

particular the root-shoot ratio. There is also evidence that the quality and composition of vegetation, such as the C/N ratio (Van Cleve et al. 1983), and also mot exudation and mycorrhiza formation (Norby et al. 1987) is altered. The problem in making any forecast concerning the climatic variables is the unpredictability of the latter. In general, temperatures will increase, and this will normally give a higher productivity, but the reverse is true if rainfall is not adequate. It is therefore impossible to predict net primary productivity at a designated spot in response to climate change, except to conclude that on average dry matter formation is likely to increase. However, two general situations can be distinguished. Firstly, in most areas the general vegetation type will remain constant. There is a strong relationship between IWP and net annual evapotranspiration; the latter measures temperature, but with an inbuilt correction for any deficiency of water (Lieth 1975; Whittaker 1975) (see Fig 7.4). This can be used as a guide to the behavior of NPP in areas where the general vegetation type remains constant. Moderate changes in NPP in such conditions are unlikely to alter soil organic matter greatly.

Annual actual evaporation (rnrn)

Fig. 7.4

Relationship between actual evapotranspiration and productivity (-), decomposition of deciduous (--- j and coniferous litter (.-.----j

Secondly, the most striking effects on soils and vegetation will certainly occur near the boundaries of the present vegetation zones, where the whole character of the vegetation, and the type and amount of litter, may change (Table 7.2). Thus, Emanuel et al. (1985) calculated that the temperature rise from a doubling of CO2 could result in a decrease in the global area of boreal forest and tundra of 37% and 32% respectively, this area changing into grassland or cultivated land. Soils in such conditions will then have quite new and different equilibrium states, with very different organic matter levels. The organic cycle

79

lmpocl on soil organic mailer and biology

differs significantly in grassland and forest soils, the organic matter in the former being both greater in quantity relative to the plant weight, and more deeply distributed than in forest soils. The quality of the litter, and the type of soil fauna, also differ considerably (Anderson, in press). The 'analogy' approach must be used in predicting these. Table 7.2

Plant biomass, litter input, soil organic matter and microbial biomass of major global vegetation types.

Subject

Tropical forest

Tem-

Boreal Savannah Tem-

perate

forest

perate grass

forest Area x 1012 m2 Plant biomass g C m-2 Litter input g C m-2 Soil c g m-2 Soil N g m-2 Microbial biomass C

12.0 12.5 14,000 9,000 368 250

Tundra

15

land 9

8 250 1,440 667 75 23,000 22,000 2,100 1,125 215 20

24.5 18,000 710 13,000 816 50

640 110

1,800 360 15,000 5,400 1,100 333 60 35

2

14

2.5

8.7

51

1

0.07

0.30

0.14

0.17

0.32

0.27

9,000

g m-2

Microbial biomass N g m-2

Microbial turnover yrs

Plant biomass and net primary productivity from Whittaker and Likens (1973). Litter from Ajtay et al. (1979). Soil C and N from Zinke et al. (1984). Microbal biomass from E.A. Paul ( p e n comm.)

THE BREAKDOWN OF SOIL ORGANIC MATTER The Qlo coefficient for the response of microbes to temperature change is normally around 2, but may vary from 1.6 to 3.2 (Schlesinger 1977; Singh and Gupta 1979). However, such figures are not necessarily a good indication of what happens when soil temperature is permanently increased in the field. Microbial species, moisture content and resource quality will change, and all affect microbial reproduction and respiration rates. A laboratory measurement of the effects of a simple short-term temperature shift in a soil is therefore of rather little use, and only gives a very general indication of direction and magnitude of change. The rate of decomposition of organic materials is sensitive to both temperature and moisture, and can therefore be related to the actual evapotranspiration rate in that environment. Hence, both this and the effect on NPP can be compared as in Fig. 7.4. Generally the effect of evapotranspiration on decomposition is greater than on NPP, hence as evapotranspiration increases,

80

P B . Tinker and P. Ineson

there is a tendency for the ratio of vegetation biomass to litter to increase. This is very clear in a study of biomass and soil carbon in various biomes (Fig. 7.5).

t

4

1

I

I

I

300

200

100

0

1

400

I

I

I

Tropical forest

I

1

I

100

Temperate I forest

I

200

I

I

300

400

I 500

Carbon pool (t ha-’)

Fig. 7.5

Comparison of carbon pools in plant biomass and soil organic matter (including litter) across different ecosystems Cfrom Anderson in press)

The breakdown of organic material and humus has been successfully modeled. The clearest situation is in agricultural soils, in which the same crop is grown regularly year after year, so that input, in quality and quantity, remains constant. Jenkinson et al. (1987) have modelled this situation for the Rothamsted long-term experiments. The structure of the model is illustrated in Fig. 7.6.

Fig. 7.6

Diagram showing the flow of carbon through the soil model of Jenkinson et al. (1987). DPM = decomposable plant material; RPM = resistant plant material; BlOZ and BlOA = zymogenous and autochthonous soil microbial biomass, respectively; HUM = hwnijied organic matter

lmpaci on soil organic mafler and biology

81

The ratio of biomass to humus produced is always the same in all soils, but differs between A and B pathways. The decay rate is (l-e-k), where k is the rate constant, that varies with temperature and moisture (Van Veen and Paul 1981). The effect of soil texture is included by amendments of the ratio of CO2 to biomass plus humus produced at each decomposition step. It was also found necessary to postulate a separate class of 'biologically inert' organic matter. This model has been applied to Nigeria Alfisols and Ultisols (Jenkinson and Ayanaba 1977), where it has successfully modelled the breakdown of 14-C labelled maize and rye grass residues, using the original parameters derived in Rothamsted. A set of runs of this model are illustrated in Fig. 7.7 (Jenkinson, personal communication), with a direct comparison of the effect of two different temperatures, of 9.22"C (normal mean at Rothamsted) and 19.22"C. The higher temperature leads to more rapid decomposition of residues, as expected. Broadly, the amount remaining at any time at 19.2"C is roughly equivalent to what is left at the lower temperature, but after about 3 times the time. Using the effects of change of biome, of changes in C02 and of actual evapotranspiration on NPP, and the existing models of organic decomposition, it should be possible to make prediction on the soil carbon levels following climate change. The uncertainties and errors will however be very large.

SOIL ORGANISMS The general soil heterotrophic population will behave in a way largely characterized by the speed of decomposition of soil organic matter. Temperature changes and precipitation changes will have effects as discussed above. Species of micro-organisms will change, but the likely speed of change, and $e acceptable climatic limits for any one group pf organisms, are such that no major unexpected effects are likely. The temperature optima for soil organisms are usually quite broad, and a shift of one or two degrees will normally make little difference. The most interesting problem in this topic will be in the wetlands of the high latitudes of the northern hemisphere. Temperature changes will be largest here, perhaps 6-10°C, with sharp changes in water relations. The question of whether the microflora will change in line with the conditions, or will lag well behind these, does not appear to have been addressed yet. The long-distance transport of bacteria on migrating birds, animals or man will normally be fairly rapid; this is less certain with the larger soil organisms. Previous comparisons of microbial communities across different biomes have been largely mycological, and show that species composition is clearly related to biome type, even for the same biome in different continents (Kjoller and Struwe 1982). Simple biomass comparisons by these authors suggest that upper levels for fungal biomass are an order of magnitude higher in temperate grasslands than in tundra, and that turnover rates are similarly higher in the more

P . B . Tinker and P . lneson

82

productive systems. Bacterial populations have been less well studied in this respect, yet Sundman (1970) has demonstrated that the functional attributes of the soil bacterial population are influenced by plant cover type.

0

12

24

36

48

60

I 72

I 84

I

96

I 108

I 120

Time in Months.

Fig 7.7

Runs of the Jenkinson et al. (1987) model, showing a direct comparison of the predicted rates of decomposition at 9.22 "C(-) and 19.22 "C(-----)

There is no direct evidence relating faunal biomass or productivity to organic matter quality; these seem to reflect responses to the environment directly. Thus, for example the distribution of termites appears to be limited by low minimum temperatures (Hams 1971), and the increases in temperature predicted as a consequence of climatic change can be expected to extend their range. Soil organisms with more specific characteristics may well be slower in adjusting to the new conditions. These are: Plant pathogens; Symbiotic organisms; Soil fauna. For plant pathogens and symbiotic organisms there is the additional complication that they are, to varying degrees, dependent upon specific types of vegetation. A vegetation zone shift caused by temperature/precipitation changes will only have reached full equilibrium when both the vegetation and the appropriate lower organisms have established themselves together.

Impact

on

soil organic mauer and biology

83

In terms of temperature only, the North-South displacement equivalent to the predicted climate change is up to a few hundred kms, depending upon latitude. With a C02 doubling time of 40-50 years, the rate of movement of organisms may need to be up to 10 km yr -1 to keep up with this shift. There is some information on the rates of spread of soil borne diseases and the recent rapid movement of the barley yellow mosaic virus across England provides an example of the time scales involved (Hill and Walpole 1989). In the seven years from the initial reports of the occurrence of this soil borne virus in England, the organism is now established over much of the country, and has spread at a rate which clearly exceeds those anticipated for climatic change. However, there are no precise analogies with the proposed scenario, in that all studies of organism dispersion have been within a broadly static environment, following an introduction of the organism. There has been no real analogy since the end of the last Ice Age, when the rate of spread and change was probably (though not certainly) slower. It will be essential for the appropriate symbionts to be available to the invading vegetation. Amongst the myconhizas, there is likely to be a decrease in the amounts of ectomycorrhizal fungi, as the large boreal forest belts move northwards towards the poles. Tropical and subtropical trees are almost wholly symbiotic with vesicular-arbuscular mycorrhizal fungi (Harley and Smith 1983). The rate of movement of mycorrhizal infection from root to root in grassland is slow, at ca 50 cm/yr (Sparling 1976). Fortunately these have low specificity, and some common species are found very widely distributed. Except in very high latitudes or in previously bare soils acceptable mycorrhizal symbionts are likely to be present, though the best equilibrium population of fungi may take a considerable period to establish. The N-fixing Rhizobia, actinomycetes or freeliving bacteria are more specific in their hosts, in particular the first two. There will certainly be situations in which there is a delay in introduction of these symbionts, and the spread of some leguminous plants may be delayed because of that. Heal and Ineson (1984) provide evidence to support the argument that the pattern of primary productivity in different biomes is reflected in the microflora and fauna, mainly through variation in resource quality, but also particularly for the fauna, through external climatic factors for the fauna. Again, it is when climatic change results in vegetation changes that associated shifts in below ground populations will become most pronounced. The time taken for the soil fauna to migrate into new areas is expected to be slower than for soil microorganism, and introductions of, for example, earthworms have shown rates of migration restricted to a few meters p e r year, in the absence of appropriate vectors (Van Rhee 1969). The net result is that the rate of change in climate may exceed the rate of colonization by the soil organisms usually associated with the new climate, and this lag will be greater for the larger components of the soil biota. The longer term picture is shown in Fig.7.2,

84

P.1)

T'inker and

P lneson

suggesting that equilibrium conditions will ultimately be reached; il is the predicted speed of change of climate which will result in imbalance.

LAND CAPABILITY EFFECTS Over most of the tropical and subtropical parts of the world, the restraint on plant production is water or fertility rather than temperature or radiation. As temperature will change relatively little with climatic change in these latitudes, the overwhelmingly important effect is likely to be in rainfall changes, which are the most difficult to predict. Any rise in temperature would be expected to produce more intense rain, even if the mean precipitation were unchanged or were lowered, with consequential effects on infiltration rates and erosion. In the higher latitudes in particular, there may be major changes in land use, with the movement of crop belts to the North (Warnck et al. 1986). If this involves the opening up of previously uncropped grassland or forest, the organic carbon contents of the soils will decrease, which has already contributed to global CO2 increase (Trabalka 1985). The greatest potential for catastrophe must be in impacts on the dryland areas of the world. These are soils which are already infertile, often with serious additional constraints of poor drainage, wind or water erosion, stoniness or shallowness and salinity (Dregne 1982). Their agriculture is already often barely viable, and dryland areas contain some of the most disastrous agroecological problems in the world (Steiner et al. 1988). Any decrease in mean precipitation, or any increase in rainfall irregularity, could be very damaging in many areas. Even now, the rainfall distribution is normally skewed, with many more years below the mean than above. The risks associated with this type of farming can easily become so large that the whole system loses its viability. The degradation that already occurs in many of these soils, will be accelerated by and associated with any decrease in organic matter (Lal 1987). The lowering of dry matter yields by lower rainfall will cause lower residue returns to the soil. A temperature increase will lead to some acceleration in organic matter decomposition, and therefore soil fertility (most probably nitrogen supply), water retention, infiltration and erosion resistance will gradually diminish. Changes of this type may be catastrophic in such areas, and the effects on soil organic matter will be strongly implicated.

CONCLUSIONS 1)

Climatic changes may have massive consequences for the total vegetationlsoil system , and it is quite unprofitable to consider soils alone in this context. The impacts will vary widely with the actual climatic change

lmpacl on soil organic matter and biology

85

experienced, and also with the sensitivity of the vegetation-soil system at any point. The changes are likely to result in a significant net release of carbon dioxide to the atmosphere, and it is important to be able to predict this as soon as possible. Hydrological and vegetation changes may be more important than simple temperature changes in this regard. Scientists dealing with the terrestrial part of the global system need to develop a more integrated approach, such as is particularly evident in the atmospheric sciences. Single site studies may be useful in defining principles, but are of little value on their own in developing a global picture. Soil scientists in particular need to emphasize the global nature of soil science, not so much in taxonomic pedology, but in process studies. There is a fundamental need to improve our methods of extending data from point measurements to global fluxes and processes. We need to look very carefully at our use of current classification systems, that are mostly based on semipermanent, slowly changing characteristics. We also need to improve our use of remote sensing technology. Where new technology or instrumentation an: relevant, we should have no hesitation in identifying this as an urgent need. This short review has shown quite clearly the weaknesses in soil ecology, and that additional effort and resources are needed if better predictions are to be made. The comment applies particularly to microbial ecology, in which the vast majority of soil microbes cannot even be identified, much less studied. It also applies to the ecology of soil fauna. More work should be done on soils that are at the boundaries of vegetation zones, and are thus sensitive to change; in general, soils not in equilibrium with their environment warrant more interest. Better methods of studying organic decomposition are needed including standard materials for testing rates (Anderson, in press). We must recognize the weaknesses of GCM predictions, but be ready to seize upon every advance as it appears.

ACKNOWLEDGEMENTS We thank Dr. I2 Jenkinson for the model runs in Fig. 7.7, and Dr. J. Anderson for allowing us to see a preprint of his paper in Functional Ecology.

P.B. Tinker and P. lneson

REFERENCES Ajtay, G.L, P. Ketner and P. Duvigneaud (1979). Terrestrial primary production and phytomass. In: The Global Carbon Cycle. Scope 13, Bolin B., E.T. Degens, S . Kempe and P. Ketner (Eds). The Global Carbon. Scope 13, 129-182. Wiley: New York. Anderson, J.M. (in press). The effects of climate change on decomposition processes in grassland and coniferous forest. Functional Ecology Bolin, B., E.T. Degens. S. Kempe and P. Ketner (Eds.) (1979). The global carbon cycle. SCOPE 13. Wiley: New York. Bolin, B., B.R. Doos, J. JSiger and R.A. Warrick (Eds.) (1986). The greenhouse effect, climate change and ecosystems. SCOPE 29. Wiley and Sons, Chichester, U.K.. Department of the Environment. (1988). Possible impacts of climate change on the natural environment in the United Kingdom. DOE: London. Dickinson. R.E. (1986). Evapotranspiration in global climate models. Paper presented atthe 26th COSPAR Meeting, July 2-10 1986, Toulouse, France. Dregne, H.E. (1982). Dryland soil resources. Science & Technology Agriculture Report. Washington: US Agency for International Development. Emanuel, W.R., Shugart, H.A. & Stevenson, M.P. (1985). Climatic change and the broadscale distribution in terrestrial ecosystem complexes. Climatic Change, 7, 29-43. Goudriaan, J. and G.L. Ajtay (1979). The possible effects of increased COz on photosynthesis. In: Bolin, B., E.T. Degens, S. Kempe and P. Ketner (Eds). The Global Cycle. Scope 13, 237-249. Wiley: New York. Harley, J.L. and S.E. Smith (1983). Mycorrhizal symbioses. Academic Press: London. Harris, W.V. (1971). Termites, their recognition and control. Tropical Agriculture Series, Longman: London. Heal, 0.W. and P. Ineson (1984). Carbon and energy flow in terrestrial ecosystems: relevance to microflora. In: M.J. Klug and C.A. Reddy (Eds). Current Perspectives in Microbial Ecology. 394-404. American Society for Microbiology: Washington. Hill. S.A. and B.J. Walpole (1989). National and local spread of barley yellow mosaic virus in the United Kingdom. Bulletin OEPPEPPO Bulletin, 19, 555-562. Holdridge, L.R.( 1947). Determination of world plant formations from simple climatic data. Science, 105, 367-368. Holdridge, L.R. (1964). Life Zone Ecology. Tropical Science Center: Costa Rica. Houghton, R.A., W.H. Schlesinger, S. Brown and J.F. Richards (1985). Carbon dioxide exchange between the atmosphere and terrestrial ecosystems. In: Trabalka, J.R. (Ed). Atmospheric carbondioxide and the global carbon cycle. (DOEER-0239). , 113-140. U S . Department of Energy: Washington. Jenkinson, D.S. and A. Ayanaba (1977). Decomposition of carbon-14 labelled plant material under tropical conditions. Soil Science Society of America Journal, 41, 9 12-915. Jenkinson , D.S., P.B.S. Hart, J.H. Rayner and L.C. Parry (1987). Modelling and turnover of organic matter in long-term experiments at Rothamsted. INTECOL Bulletin, 15, 1-8. Johnston, A.E.(1987). Effects of soil organic matter on yields of crops in long-term experiments at Rothamsted and Woburn INTECOL Bulletin, 15, 9-16. Kjoller, A. and S. Struwe (1982). Microfungi in ecosystems: fungal occurrence and activity in litter and soil. Oikos, 39, 389-422. Lal, R. (1987). Managing the soils of the sub-Saharan Africa. Science, 236, 1069-1076. Lieth, H. (1975). Modeling the primary productivity of the world. In: Lieth. H. and R.H.

Impaci on soil organic matier and biology

87

Whittaker (Eds). Primary Productivity of the Biosphere, 237-263. New York: Springer Verlag. Norby, R.J., E.G. O'Neill, W.G. Hood and R.J. Luxmore (1987). Carbon allocation, root exudation and mycorrhizal colonization of Pinus echinata seedlings under carbon dioxide enrichment. Tree Physiology 3, 203-210. Olson, J.S., R.M. Carrels, R.A. Berner, T.V. Armentano, M.I. Dyer and D.M. Yaalon (1985). The natural carbon cycle. In: Trabalka, J.R. (Ed). Atmospheric carbondioxide and the global carbon cycle. (DOE-0239). 175-213. U.S. Department of Energy: Washington. Schlesinger, W.H. (1977). Carbon balance in terrestrial detritus. Annual review of Ecological Systematics, 8. 51-81. Singh, J.S. and S.R. Gupta (1979). Plant decomposition and soil respiration in terrestrial ecosystems. Botanical review, 43,449-528. Solomon, A.M., J.R. Trabalka, D.E. Reichle and L.D. Voorhees (1985). The global cycle of carbon. In: Trabalba, J.R. (Ed). Atmospheric carbondioxide and the global carbon cycle.(DOE/ER-0239), 1-13. U.S. Department of energy: Washington. Sparling, G.P. (1976). Effects of vesicular-arbuscular mycorrhizas on pennine grassland vegetation. PhD Thesis, University of Leeds. Squire, G.R. M.H. Unsworth (1988). Effect of C 0 2 and climate change on agriculture. Contract report to U.K. Department of the Environment. Steiner, J.L., J.S. Day R.T. Papendick, R.E. Mayer and A.R. Bertrand (1988). Improving and sustaining productivity in dryland regions of developed countries. Advances in Soil Science, 8, 79-150. Sundman, V. (1970). Four bacterial soil populations characterized and compared by a factor analytical method. Canadian Journal of Microbiology, 16. 455-464. Tinker, P.B. (1984). Site specific yield potentials in relation to fertilizer use. 18th Coll. Int. Potash Institute, Bern 193-208. Trabalka, J.R. (Ed) (1985). Atmospheric carbondioxide and the global carbon cycle. (DOEER-239). US. Department of Energy: Washington. Van Veen, J.A. and E.A. Paul (1981). Organic carbon dynamics in grassland soils. 1. Background information and computer simulation. Canadian Journal of Soil Science, 61. 185-201. Van Cleve, K.. L. Oliver, R. Schlentner, L.A. Viereck and C.T. Dyrness (1983). Productivity and nutrient cycling in taiga forest systems. Canadian Journal of Forest Research, 13, 747-766. Walker, B.H. and R.D. Graetz (1989). Effects of atmospheric and climatic change on terrestrial ecosystems. IGBP Report No 5. Warrick, R.A.. R.M. Gifford and M.I. Parry (1986). Assessing the response of food crops to the direct effects of increased C 0 2 and climatic change. In: B. Bolin, B.R. Doos, I. JLger and R.A. Warwick. The greenhouse effect, climatic change, and ecosystems, SCOPE 29, 393-473. Wiley: New York. Whittaker, R.H. (1975) Communities and ecosystems. MacMillan: New York. Whittaker R.H. and G.E. Likens (1973). The primary production of the biosphere. Human Ecology, 1, 299-369. Zinke, P.J.. Strangenberger, A.G., Post, W.M., Emanuel. W.R. & Olson J.S. (1984). Worldwide organic soil carbon and nitrogen data. (ORNLnM-8857). Tennessee: Oak Ridge National Laboratory.

This Page Intentionally Left Blank

89

Chapter 8

INFLUENCE OF CLIMATIC CHANGE ON DEVELOPMENT OF PROBLEM SOILS, ESPECIALLY IN THE ALLUVIAL DOMAINS Walter R . Fisher Universitat Hohenheim Institut fur Bodenkunde und

Standortslehre

D-7000 Stuttgart 70, Fed. Rep. Germany

ABSTRACT Among the factors which change during a global climatic change, mean annual precipitation and increased temperature are supposed to have the strongest influence on the ecological properties of alluvial soils. Whereas increased precipitation would enforce leaching, or moderate salt problems, increased biological activity due to higher lernperature would accelerate microbial turnover in the soil and enlarge Lones of anaerobiosis.

THE DEFINITION OF "PROBLEM SOILS" - GENERAL CONSIDERATIONS The response of soil ecology on any changes in environmental conditions depends on the buffering ability of a given soil ecosystem. This is shown in Fig. 8.1, where a non linear relationship between the strength of the influencing factor and the respective soil properties is supposed. In this diagram, the buffering activity of the soil is documented by the inverse slope of the response curve. If one external factor is less strong compared to the buffering capacity of the soil system, it would be buffered, and the corresponding changes of the soil properties would be very low. At higher levels of external loads, a threshold may be exceeded, and irreversible changes of the ecological properties may occur. Depending on the actual land use of a given agricultural soil, problems can occur if such soil is in a condition where it is sensitive to external influences, and can be changed irreversibly. With respect to alluvial soils, such dominant factors may be acidification by rain or by soil organisms, the height of the permanent water table or the reduced horizon. In general, the water regime has a strong influence on the ecological soil properties, because nearly all nutrients or toxicants are transported by the soil water. Three environmental factors which are supposed to change during the global climatic change may influence the ecological properties of a given soil ecosystem: Composition of the atmosphere;

90

W.R. Fisher

Mean annual precipitation, and position of the phreatic groundwater table; Temperature of the atmosphere just above the soil surface. Influence of external factors on soil ecology

intensity of external factors (e.9. acid rain, temperature)

F i g . 8.1

Schemuiic drujl of the sensitivity of a soil 10 external factors

COMPOSITION OF THE ATMOSPHERE The CO, content of the atmosphere is about 0.03% by volume, whereas in biologically active topsoils 1 - 10% of CO, are typical concentrations. So the difference between the CO, concentration of the soil air and atmospheric air is relatively high, even if the atmospheric CO, concentration is increased by a factor of two. By consequence, the CO, concentration gradient is less influenced by the global C02 increase, and a biologically active soil is acting as a C 0 2 source. For the genera1 relations between CO, concentrations, light intensity, and photosynthetic efficiency, see Chapter 1 , Fig. 1.5 (Scharpenseel). This is not the case for soils which are very low in biological activity, or for rock surfaces not covered by soils. In these cases, especially for carbonaceous rocks, the weathering rate is increased due to the lower pH of the rain water. Among the consequences of this change is the accelerated release of nutrients from primary minerals as well as the intensified leaching by percolating water. With respect to acid soils, a marked impact of higher CO;? content on soil forming processes is not assumed. This is because the dissociation constant of the first step of H2C03:(into H+ + HC03‘) is very low (pk = 6.35) and should therefore not influence soils with pH values below 4. The two other contaminants which play a role in the global “greenhouse effect”,

Influence of climalic change on soils in the alluvial domains

91

methane (CHJ and nitrous oxide (N,O) are not assumed to have an influence on the more important soil processes (for NH,, see Addendum to this Chapter, by Takai).

MEAN ANNUAL PRECIPITATION After the most reliable models, the mean annual precipitation will increase globally. For estimating the appropriate influences upon a given soil, we have to distinguish between two hydrological situations (Fig. 8.2):

I PRECIPITATION

increasing

decreasing

A. groundwater level high:

-

zone of accumulation

6.groundwater level low: infiltration rate _ _

I1 - 1

Zoneofleaching (nutrients)

Fig. 8.2

a)

n

Soil ecological consequences due to groundwaterjluctuatiom

High groundwater level; with increasing precipitation, the level of the permanent groundwater table should increase slowly. As a consequence, the position of the capillary rise should also move upward in the soil. Problems

92

W.R. Fisher

could arise, if at high actual evapotranspiration,zones of accumulation (salt, iron oxides) would form within the rhizosphere of agricultural plants. This is shown in Table 8.1, where the Fe(II)/Fe(III) redox system is used as an example. Due to the higher mobility of the reduced phase, Fe can move along the redox gradient, and Fe oxide is accumulated in the contact zone. In contrast, no acidification occurs if the Fe(I1) comes from hydrolyzed species such as Fe(OH)2,because protons are released only if the degree of hydrolysis for Fe is changed during oxidation. As a result of the oxidation of Fe sulfides, acids (sulfuric acid) are formed. Prolon release during Fe and S oxidation

Table 8.I Reduced species Fe ion

Oxidized species Fe(ll1) oxide

Fe( 11) compound Fe( 11) sulfide

Reaction

Effect

4 Fe2++ 6H2O + 0 2 = 4 FeOOH + 8 H+

acidification due to Fe hydrolysis

Fe(l11) oxide

4 Fe (OH):! + 0 2 = 4 FeOOH + 2 H20

none

Fe(ll1) oxide sulfuric acid

4 FeS + 9 0 2 + 6 H 2 0 = 4 FeOOH + 4 H2SO4

acidification due to sulfur oxidation

Low groundwater level; in this case, increasing precipitation will enhance the rate of percolation and can therefore step up nutrient leaching or moderate salt problems, depending on whether it is a highly permeable or salt affected soil, respectively.

b)

Table 8.2

influence of changing soil humidity on pedological problems Humidity

Problem ~~

Increasing

Decreasing

moderated if water balance goes to positive increased in permeable soils increased in permeable soils (low buffering capacity) increased

increased

~

high salt concentration nutrient deficiency acidification reducing zones in topsoil

not influenced or increased due to low availability not affected moderated

93

Influence of climalic change on soils in [he alluvial domains

Table 8.2 summarizes briefly the expected changes in soil ecology due to changing annual precipitation. Not included in these considerations are soils which are going to be influenced directly by the (increasing) sea level, though this might be of great importance for large coastal areas.

TEMPERATURE OF THE ATMOSPHERE Possibly the most important impact on soil ecology (from a global point of view) would come from the increasing temperature of the air just above the soil surface. This may have three reasons: a)

Direct influence on chemical equilibria. Table 8.3 shows the temperature function of the equilibrium constant of a chemical reaction following the Van 't Hoff equation. As is demonstrated by the given example the variation of dk/dT with changing temperature is more or less negligible within the regarded temperature range.

Table 8.3

Theoretical influence of temperature on chemical equilibrium constants

Influence of temperature on chemical equilibria AGO = -RT In k [AGO = GIBBS energy of reaction, k = equilibrium constant, AH = free enthalpy of reaction] d(AG/T3 - -&

Crr T2 d(In k) - AHo R T ~

(Van't Hoff)

Example:

Assumption:

Increasing temperature favors endothermic reaction (for AHo/R = 10K) T[KI WdT 293 1.000116 298 1.000113

real effect negligible, because most reactions are not in a sensitive range Possible exceptions: formation of secondary minerals (oxides, clay minerals)

W . R . Fisher

The only exception from this model would occur if an equilibrium is extremely sensitive to temperature variations, e.g. the goethitehematite system in model experiments. However, under pedogenetic conditions, this equilibrium is superimposed by the strong influence of soil organic matter, and the direct influence of temperature variations is supposed to be very low. Increase of the reaction rate of most physical and chemical processes. Due to the relatively slowly moving water in the pores of a soil, this is supposingly of less importance (except some events of very high precipitation) because the chemical equilibrium between the solid surfaces and the soil solution is obtained in at least several hours. Table 8.4 indicates three different cases of initially varying soil humidity. Table 8.4

Ecological consequences of the acceleration of physical processes

Increased temperature makes most physical processes faster: evaporation - transpiration water transport in soil surface transport of dissolved salts to surface "dry case" "moist case" "wet case" increasing salinity increasingacidification increasing bioactivity decreasing biological first phase: increasing importance of activity increasing biomass reduced (anoxic) zones production (increasing osmotic potential) c)

second phase: toxic effects (of AI)

mot zone influenced

by anoxia

Increase of' biological activity. This is expected to be most sensitive to temperature variations; it possibly has the strongest influence on soil ecology. In the following paragraphs some examples are given on what the expected influence of increased biological activity might be.

The total biological activity, as documented by CO, production, depends strongly on temperature. As shown in Fig. 8.3, the slope of the cuwe increases with increasing temperature, so the expected influence of elevated temperature would be higher in the tropics than in moderate climates. This is also reflected by the electrolyte concentration of the pore water. As illustrated in Fig. 8.4, the electrical conductivity of the pore solution was increased dramatically during incubation at high temperature, whereas at 5°C it did not. By consequence, at elevated temperature, all chemical exchange

Influence of climairc change on soils in the alluvhl domains

95

processes are accelerated, and are continued to a higher extent. Biological

activity

(soil respiration)

5 t

0 0

10

20

30

-_

40

temperature PC

Fig. 8.3

Soil respiration as a function of temperature (after Anderson and Domsch 1986)

Electrical conductivity of soil solutions

0

I

I

I

2

4

6

8

period submerged 1 days

Fig. 8.4

Electrical conductivity of the soil solution at various temperatures (after Yu 198.5)

Under waterlogged conditions, the actual redox potential depends on two counteracting processes.

96

W . R . Fisher

Sulfide formation

Fig. 8.5

Sufide formation during incubation at various temperatures for two soils (after Uuo and Mikkelsen 1981)

The oxygen consumption by organisms, and the oxygen supply from the atmosphere, or from plant roots with aerenchyms. Whereas the latter process is less influenced by temperature variations, the increased biological activity t higher temperatures leads to a faster decrease of Eh, even at slightly permeable soils. This is shown in Fig. 8.5: in that case the lower diffusion rate of oxygen in the clay soil supports a rapid sulfate reduction at higher biological activity. Therefore the border line of the reduced horizon of a waterlogged or subaquatic soil is elevated at higher temperature (Fig. 8.6). low temDerature

n

l

I = Q)

8-

oxidized horizon

reduced horizon

Fig. 8.6

Schematic drafl of the height of the reduced horizon in a waterlogged soil due to varying temperature

InJluence

of climatic change on soils in the alluvial domains

97

Even if the permanent groundwater level would not be affected, a rise of the reduced layer could occur, and therefore, the zone of strong acidification could expand to the rhizosphere. This is shown in Fig. 8.7, which demonstrates that, at higher temperature, the zone of major Eh changes is both narrower and higher in the soil profile. So, the expected acidification concentrates on a narrower soil compartment and could therefore influence the soil ecology stronger.

-

with increasing temperature increase:

--

rate of biomass mineralization rate of oxygen consumption elevated zone of reduction

elevated zone of sutfide oxidation

redox

tential

predicted

5 a

a, -0

capillary

N

s (D

0, P, 9.

a =f? 0

%. 0 3

I Fig. 8.7

groundwater level

Schematic draft of the redox profile combined with soil acidification at increasing temperature (high groundwater level)

At permanently low redox potentials, the microbial metabolism is quite different compared to soils with air filled pores, and different compounds are formed. One of these compounds is acetaldehyde, the rate of formation depending strongly on temperature (Fig. 8.8). This, besides a direct impact on bacterial growth and selection, can influence the composition of the soil microflora community, due to toxic or competitive effects.

98

W.R. Fisher

I

Formation of acetaldehyde 100

,Q.

80-

-0' -

60-

63' :

: 20

.-

1

.*a. .a. .

'-....-.P I

.*w

'0

L2O"cJ

,

0.

Fig. 8.8

Formation of acetaldehyde in a soil during incubation at different temperatures (after Tsutsuki and Ponnamperuma 1987)

CONCLUSIONS The direct influence of a global climatic change upon soils in alluvial areas with a relatively high permanent groundwater level are presumably not very drastic. The increasing temperature as well as the increasing annual precipitation will lead to an increase in the biological activity in the soils. This, in general, is favorable but, at high water level or for submerged soils, zones of reduction might become more important and could influence the rhizosphere. For relatively dry soils, the increasing precipitation might moderate salt problems, but might also enhance nutrient leaching and soil acidification. So the ecological significance of the expected consequences of global climatic changes depends mainly on the actual ecology of the interesting soil system. Two other items which are not taken into consideration are the large area sedimentation of materials eroded upstream, and the salinization by seawater of flat coastal areas. These factors might be of greater importance than the direct influences mentioned above, especially in highly productive areas of the tropics.

REFERENCES Anderson, T. - H and Domsch, K.H. (1986). Carbon assimilation and microbial activity in the soil. Z. Pflanzenern&r. Bodenkd. 149: 457-468.

Influence of climatic change on soils in the alluvial domains

99

Kuo, S . and Mikkelsen, D.S.(1981). The effects of straw and sulfate amendments and temperature on sulfide production in two flooded soils. Soil Sci. 132: 353-357. Tsutsuki, K . and F.N. Ponnamperuma. (1987). Behavior of anaerobic decomposition products in submerged soils. Soil Sci. Plant Nutr. 33: 13-33. Yu, Tian-ren (1985). Physical chemistry of paddy soils. Science Press, Beijing.

This Page Intentionally Left Blank

101

Addendum to Chapter 8 METHANE FORMATION IN WATERLOGGED PADDY SOILS AND ITS CONTROLLING FACTORS Yasuo Takai* and Eitaro Wada**

* Tokyo University of Agriculture, Sakuragaoka. SeLagaya, Tokyo 156. Japan ** Mitsubishi Kasei Institute of Life Sciences, Minamiooya, Machida, Tokyo 194, Japan A distinguishing feature of waterlogged paddy soil lies in the creation of a reduced state in the soil due to the limited oxygen supply by the flood water. First, the authors show the successive reduction process in waterlogged paddy soils (Takai, Koyama and Kamura 1956a and 1956b; Takai and Kamura 1966; Takai 1984). As shown in Table 8.5, it could be concluded that each type of microbial metabolism proceeds successively according to the oxidationreduction condition from aerobic respiration using molecular oxygen, which is the most efficient energy- yielding reaction, to methane fermentation, which is a comparatively inefficient energy-yielding one. Table 8.6 is prepared on the basis of the idea of oxidative and reductive capacities in the soil using the results of incubation experiments under a closed condition. In the left column, eight paddy soils are listed in decreasing order of rice productivity. Total oxidative capacity is represented as 0 2 ml 100 g-’ of dry soil, after the conversion of N03-, bio-active Mn(1V) and Fe (111) contents to equivalents of molecular oxygen. The mineralized nitrogen content is assumed to be a reductive capacity. Both the relative values are normalized to the greatest yielding paddy soil Mimaki as 100%. The order of the ratio of oxidative to reductive capacities corresponds well to that of the ratio of COdCH4 formed during the incubation under waterlogging. This reveals that the content of bioactive ferric iron may be the most important controlling factor for methane fermentation. The mechanism of methane fermentation was investigated using a 14-C tracer technique in the 1960s (Takai 1970). Two metabolic pathways were found, firstly the transfer of the methyl radical in acetic acid and secondly the carbon dioxide reduction pathway. 14CH3COOH- l4Co4+Cq 14C02+4H2A- I4CH4+2H2+4A The first way was more predominant than the second way in waterlogged paddy soils. In the temperate paddy field and the rice plot experiment, mid-summer drainage increased soil Eh and suppressed methane formation (Takai and Koyama 1956).

Table 8.5

Succession of microbial metabolism in waterlogged paddy soils

Period of Stage of incubation reduction

Earlier

Later

1st stage

2nd stage

Chemical transformation

Initial Eh in soil V

Expected pattern of energy metabolism

Disappearance of molecular oxygen

+ 0.6

+ 0.5

Oxygen respiration

Disappearance of nitrate

+ 0.6 - + 0.5

Nitrate reduction

Formation of Mn2+

+ 0.6

Formation of Fe2+

+ 0.6 - + 0.3 (Ferric ion

-

-

+ 0.4

Y

i3 Formation of N Q - N

Formation of Co;!

Rapid progress

Rapid progress

(Manganate reduction)

Formation of organic acid

Hypothetical pattern of O.M. decomposition

Usually not Aerobic and accumulated semi-anaerobic without fresh decomposition organic process matter (0.M) Y

reduct.)

- -0.19

Sulphate reduction

-

Fermentation

Formation of sulphide

0

Formation of hydrogen

-0.15

Formation of methane

-0.15 - -0.19 Methane fermentation

-0.22

2 Early stage: rapid accumulation Slow progress

Slow progress

Advanced stage: rapid decrease

s D

K

Anaerobic decomposi tion process

tp;

5

Table 8.6

The oxidative and reductive capacities, and the ratio of CO2lCH4 formed during the incubation for 42 days under waterlogged conditions in various soils with different rice productivity

i

; I

Soils Nagano Mimaki Nagano Noushi Hitokuwata Aichi Noushi Aichi Arumi Nigata Yahiko Aichi Sanage Aichi Teruo

Rice Prductivity High

T

1 Low

Microbial Metabolism I stage Aerobic

T

I' stage Anaerobic

Oxidative capacity 0 2 m1/100 g Redu- Redu02 N Q cible cible Total Mn Fe ml % 2.4 0.6 4.3 56 63.3 100 2.4 1.0 0.7 63 67.1 106 3.8 1.2 0.3 63 68.5 108 1.1 1.2 0.2 20 22.5 36 3.8 1.4 2.1 35 42.3 67 3.4 1.9 0.9 70 76.2 120 1 4 0.8 0.2 4.2 6.6 10 217 1.0 0.4 7.0 11.1 18

Reductive capacity = Mineral N mg/lOOg 8.8 17.4 16.7 13.0 25.8 39.4 7.8 16.8

% 100 198 190 148 293 448 89 191

Ox.cap/ Red.cap

Cod

XlOO 100 54 57 24 23

% 100

27 11 9

k

CHq

30 37 19 17 14 11 7

$' 5' f

; 2 00

b h

0. i;-

104

Y.Takai and E. Wa&

In the tropical paddy field, the authors revealed that the content of methane gas in the rice plot was significantly lower than in the unplanted plot (Yoshida, Takai and Rosario 1975). It has been elucidated that rice planting suppressed methane formation due to the aeration by the diffusion of oxygen from the rice rhizosphere which was transported from the soil into the roots. At that time, a hydrophyte transport release of methane gas was unknown by the authors. Last decade, carbon isotope ratios of methane, C 0 2 and related organics in long-term fertilizer trials at Konosu, Japan were studied with emphasis on methane formation and emission (Nakamura, Takai and Wada 1990). Fig 8.9 summarizes carbon dynamics in a plot, supplied with organic manure (OM plot) at the above-mentioned field. The annual input of organic carbon accounts for 308 kg C/lOa/year with an average 613C value of -26./8%0, whereas the accumulation of organic matter in the plot is 40 Kg C/lOa/year. Consequently, the bulk 613Cvalue of decomposed organic matter (308-40= 268 K gC/lOa/year) is calculated through an isotope mass balance calculation to be 26.4%0.

Fig. 8.9

Summary of carbon cycling in the OM plot at Konosu paddy field

Assuming C02 (-18.5%0) and CH4 (-52.9%0)as the final products, the production rate of COdCH4 = f/( 1-0is calculated by the following mass balance: -26.4 = -18.5ft + (-52.9) (1-f)

f = 0.77

The ratio of COdCH4 in paddy soils under anaerobic condition is around 3.3 with an average 6I3C (CH4)value of -53 %o. The production rate of CH4 (268 x 0.23 Kg C/lOa year = 62 Kg C/lOa/year) in the OM plot is similar to that

Methane formalion in

water logged paddy soils

105

(60 Kg C/lOa/year) estimated from the incubation experiments of other paddy soils in Japan (Koyama 1964). The present soil methane 613C (-53 Ym) is significantly higher than those (from -58 to -67960) reported in the literature (Quay et al. 1988). However, an average 613C value for bubbles collected from Japanese paddy fields (-54O/w, unpubl.) is similar to the present data. The discrepancy might be explained as follows: recently the importance of methane transport through rice plants has been emphasized as opposed to difhsion and bubble transport across the water-air interface (Cicerone and Shetter 1981; Minami and Yagi 1988). Although contributions of both pathways to methane release from paddies have not yet been quantified, the relatively high occurrence of the C o r n 2 process near roots of rice plants might lower the values of methane transported via rice plants. Summing up current knowledge on the methane production in paddies a possible schema is presented in Fig. 8.10 with emphasis on the 613C variation of methane (Wada and Nakamura 1989). In Japan we are at present considering that bubble methane is important during the early stage (June - July) of flooding, while plant transporting methane becomes high from August to October.

A : 6°C ( ~ 0C ~ H ~- )

Fig. 8.10

Schematic illustration of methane production and its release. Possible

Sl3C values of methane at direrent sites are also included.

106

Y . Takai and E . Wada

Difference in 613Cvalues of substrata, reaction mechanism and its kinetics, and oxidation of methane in oxidized layers, are the three major factors that govern the variation of 6I3C values of methane in paddy fields. A difference in reaction mechanism may be the most probable factor for variations in methane 613C in bubbles evolved in paddies, since source organic matter in rice fields is mostly rice plants of which 613Cchanges within a very narrow ranges only. Stable isotope approaches will be promising not only for the study of methane emission but also nitrous oxide formation. Taking the above-mentioned data into consideration, factors for controlling methane formation in waterlogged paddy soils are: Contents of reducible iron and manganese compounds in soils; Water management such as mid-drainage, intermittent irrigation and water percolation; Enhancement of oxidative decomposition of organic debris such as rice straw, stubbles and weeds before waterlogging.

REFERENCES Cicerone, R.J., J.D. Shetter (1981). Sources of atmospheric methane: Measurements in rice paddies and a discussion. J. Geophys. Res., Vol. 86, 7203-7209. Koyama, E. (1964). Biogeochemical studies on lake sediments and paddy soils and the production of atmospheric methane and hydrogen. In: Recent researches in the fields of hydrosphere, Atmosphere and Nuclear Geochemistry, Maruzen Co. Ltd., Tokyo, pp. 143- 177. Minami, K., K. Yagi (1988). Method for measuring methane flux from rice paddies. Jpn. J. Soil Sci. Plant Nutr., vo1.59, 458-463. Nakamura, K., Y. Takai, E. Wada (1990). Carbon isotopes of soil gases and related organic matter in an agroecosystem with special reference to paddy field. In 'Geochemistry of gaseous elements and compounds', Theophrastus Publications, s.a., Athens, Greece, in press. Quay, P.D.. S.L. King, J.M. Lansdown and D.O. Wilbur (1988). Isotopic composition of methane released from Wetlands: Implications for the increase in atmospheric methane. global biogeochemical cycles, V01.2. 385-397. Takai. Y . (1970). The mechanism of methane fermentation in flooded paddy soil. Sci. Plant Nutr. m. Vol. 16, 238-244. Takai, Y. (1984). Microbial dynamics of paddy soil under waterlogged condition. Jour. Korean SOC.Soil Sci. & Ferti., Vol. 17, 187-199. Takai, Y., and T. Kamura (1966). The mechanism of reduction in waterlogged paddy soil. Folia Microbiologica, Vol. 11, 304-313. Takai, Y., T. Koyama (1956). Composition of gases and organic acids contained in soil of paddy field. Jour. Japanese Soil & Ferti., Vol. 26, 509-512. Takai. Y., T. Koyama and T. Kamura (1956a). Microbiological studies on the reduction of paddy soils. Trans. 6th Int. Congr. Soil Sci., Paris, 527-531.

Methane formation in water logged paddy soils

107

Takai, Y., T. Koyama and T. Kamura (1956b). Microbial metabolism in reduction of process of paddy soils, Part 1. Soil & Plant Food, V01.2. 63-66. Wada, E., K . Nakamura (1989) Carbon isotopic studies on global methane production with emphasis on paddy fields In: Proceedings of the fifth working meeting on isotope in nature, Leipzig. in press. Yoshida, T., Y. Takai, D.C.D. Rosario (1975). Molecular nitrogen content in a submerged rice field. Plant & Soil, Vol. 42, 653-660.

This Page Intentionally Left Blank

109

Chapter 9 POTENTIAL INFLUENCE OF CLIMATIC CHANGE ON SOIL ORGANIC MATTER AND TROPICAL AGROFORESTRY Eldon H . Franz Environmental Research Center and Program in Environmental Science and Regional Planning Washington State University, Pullman, Washington 99164-4430, USA

INTRODUCTION Soil organic matter (SOM) has a general range of "response times" on the order of 10' to lo2 years and is a sensitive indicator of climatic patterns (see Chapter 17 by Goryachkin and Targulian for general discussions of the "response time" concept). Thus, SOM can be expected to manifest measurable responses to global warming over the next 20 to 80 years, assuming that the global mean annual temperature will increase by 2 to 4°C at 0.5 to 1 "C per decade. This chapter introduces a simple model for estimating the temperature sensitivity of SOM and calculating the change in global SOM distribution that would be likely to accompany global warming. The implications of the predicted changes are outlined in terms of the global carbon cycle and the management of SOM in tropical agroforestry systems.

THE DYNAMICS AND GLOBAL PATTERN OF SOM The total amount of organic matter in the soil depends on the balance of biomass production and decomposition, and on the soils capacity for organic matter storage. It is, therefore, sensitive to temperature and precipitation and is quite variable at local scales. At continental scales, the highest levels of SOM occur in climates that are not at the extremes of either temperature or precipitation and decline toward hot/dry and cold/wet climates (Kononova 1975). At global scales, however, a more general pattern is evident: SOM increases along gradients of increasing precipitation and decreasing temperature (Post et al. 1982, 1985; Zinke et al. 1984). The role of temperature in determining the global pattern of SOM follows directly from its differential effects on production (additions to SOM) and decomposition (losses of SOM). Fig. 9.1 shows relative values for production and decomposition as a function of mean annual temperature from -15 to 30°C.

110

Fig. 9.1

EH.Franz

Relative rates of production and decomposition as a function of mean annual temperamre

The decomposition function was derived from data for cellulose humification (Sorenson 198 1) as described by Parton et al. (1987). It represents the temperature influence on decomposition in a complex simulation model (see Chapter 10 by Stewart et al.) that gives good predictions of steady state SOM on a broad regional basis (Parton et al., 1987). It agrees closely with the field observations that decomposition coefficients for IITA (Nigeria, mean annual temperature = 261 " C ) and Rothamsted (England, mean annual temperature = 8.9"C) differ by a factor of 4 (Jenkinson and Ayanaba 1977). The production function was calculated from Lieth's (1973) net primary productivity/mean annual temperature equation by scaling relative to the maximum value of 3,000 g m-2 y-l. Using it to represent relative rates of additions to SOM requires the assumption that the influence of temperature on SOM production parallels its influence on net primary productivity. For a soil at steady state with SOM assumed to comprise a single pool, dividing the annual additions to the SOM pool by the decomposition coefficient gives an estimate of the steady state SOM level. The basic first order model of this process using soil nitrogen as the measure of SOM (following Stevenson 1986) is:

Climatic change, soil organic matter and tropical agroforestry

111

N = A/r - (A/r - No)e-fi

where

N A r t

No A/r

= % of N in

the soil = annual rate of addition of N = decomposition coefficient = time = the initial level of N = the steady state level of N

The significant mathematical properties of function (1) relate to the dynamics of SOM change when SOM is not at steady state. Under such conditions, the gains or losses of SOM are determined by the difference between steady state and initial levels and constitute a negative exponential series for constant intervals of time. It is, therefore, possible to characterize the response of SOM to any perturbation involving a change in A and/or r by calculating the time required for a half-change in the SOM increment, the relaxation time or t 1,2. The relaxation time for SOM is an explicit measure of the "response time" of the system as SOM levels approach the steady state following a perturbation. It is an inverse power function of r and takes values between 10' to lo2 years for r ranging from 0.069 and 0.0069. This range of r values encompasses the majority of decomposition coefficients determined from field studies (Young, 1989) thus verifying that the general "response time" for SOM is on the order of 10' to lo2 years. Fig. 9.2 shows the relative steady state values of SOM with respect to temperature as estimated by taking the quotient of the two functions in Fig. 9.1. Since the decomposition function is 0 at OOC, the base line case calculation is from 1"C to 30°C. An increase in the mean annual temperature at any base case location has the effect of shifting the SOM curve 2 or 4 "C to the left as shown. As the model shows, on a warmer Earth, SOM levels decline. The curve suggests that SOM increases accelerate as temperature decreases, a finding that is consistent with the pattern revealed by aggregated data arrayed to show variations with both temperature and precipitation if precipitation is held constant (Post et al. 1982, 1985; Zinke et al. 1984). The similarity is particularly apparent for the precipitation interval of 650mm to 2000mm, approximately the range of values in the latitudinally averaged profile of the Earth's precipitation (see Fig. 7.3 for a visual comparison). Fig. 9.3 shows the percentage changes in SOM calculated from the shifts in Fig. 9.2. The largest changes are projected for mean annual temperatures below 10 "C at which point the changes increase exponentially with decreasing temperatures. Using the latitudinally averaged profile of the Earth's mean annual temperature (Fig. 9.4), the SOM/temperature curve is transformed to a SOMAatitude curve (Fig. 9.5). Although the temperature profile is approximately symmetrical around 5 ON, the "thermal equator", rather than 0 ", the differences

112

E H . Frau

between the profiles for the Northern and Southern hemispheres are not significant at the scale of the figure and only the Northern hemisphere curve is shown. The curve terminates just below 60"N where the temperaturecurve crosses 1 O C .

;

100

I

I

?

I 3

,

:.

I

..

.....

.

I

:

0

075

,

i

I

/I

, t

, I

'

,

'

Warmer

.

I1

.

.

'.

\

\

'.. \

0.25

7 D e g r e e s Wormer

. ~4 . Degrees . ._

,

.

\

0.50

I

- Bose Line

\

.... .. .-. .. . '....

-

-. _. ._ --:I-.. _,

- - -,:_.. . ..1:. .. . - - _ _---.-.

0 00

Figure 9.2

,

1

-__ ,

Relative SOM at steady state as a function of mean annual temperature for the base line case. For two degrees warmer and four degrees warmer at any base line temperature, steady state SOM decreases

The transformation from mean annual temperature to latitude (Fig. 9.4) is approximately linear at temperatures less than 20°C and latitudes greater than 30 degrees (0.75"C per degree latitude). Assuming that the shift in the SOM profile would be proportional to the temperature gradient, a 2OC warming would shift the SOM profile poleward by 2.67 degrees latitude (approximately 300 km) and a 4OC warming by 5.33 degrees (approximately 600 km). The predicted shift in steady state SOM in the Northern hemisphere is therefore on the order of 150 km per "C. The poleward migration of boreal forest during interglacial periods has been estimated to be 100 km per "C (Davis, 1989). In addition, the Holocene rates of expansion of hemlock, a particularly well documented species, appeared to have averaged 20 km per century (Davis 1989). Such rates are generally considered to be consistent with steady state tracking of glacial cycles by vegetation (Webb 1986; Prentice 1986). Since rates of interglacial tree expansion were probably limited by rates of soil development (Pennington 1986), estimates of the geographical extent of change for vegetation and soils should agree whenever climatic change is slow enough to allow for a steady state adjustment. Potential

CIimlic change, soil organic

matter

and tropical agroforesrry

113

rates of warming (between 0.5 and 1.0 degree per decade) over the next few decades suggest that trees at the higher latitudes may be displaced at 25 to 50 times the average rate for the Holocene.

100% 90% 80%

70% 60%

50% 40%

30% 20% 10% 0%

10

I

Base Line Mean Temperature (degrees C)

Percentage decrease in steady state SOM for 2 "c and 4 "C warmer as a function of base line mean annual temperature

Fig. 9.3 90N 80N 70N 60N 50N 40N

30N 20N 1 ON

0 10s

20s 305 40s

50s 60s

70s

aos 90s 4 0

-30

-20

-10

0

10

20

30

Mean Annual Temperature (degrees C)

Fig. 9.4

Latitudinal profile of mean annual temperature

114

E H.From

00%

I -

I

4 Degrees Warmer

._.

2 Degrees Warmer I

60%

40%

I

/

20%

, , ,

- -

_._ __ __ .__

0%

,

_ -. ___ - ---- - .-

___ I

I

The change in mean annual temperature per degree latitude is much less in the tropics (Fig. 9.4). This implies that the SOM profile would shift over a wider range but with much smaller absolute and percentage changes than at higher latitudes. Although the resultant distribution of changes in steady state SOM is generally the same as that predicted for the distribution of the temperature change by GCMs smaller increments in the tropics than at high latitudes, changes in the distribution of precipitation could produce shifts in vegetation zones that are at least a great as those in the north. For example, the steep gradient of summer precipitation recently located between 12 and 17'N was displaced 450 to 550 km to the North during the Holocene pluvial 5000 to 10,000 years BP (Ritchie and Haynes 1987). Vast improvements in GCMs will be required to adequately predict such patterns.

IMPLICATIONS FOR THE GLOBAL CARBON CYCLE The differential temperature sensitivities of production and decomposition processes, reflected in the SOM model, will apparently actuate decreases in the carbon density of the world's soils as global temperature increases. This implies a shift in global carbon inventories from SOM to atmospheric carbon dioxide and a potential positive feedback to greenhouse warming (Woodwell 1989; Houghton and Woodwell 1989).The loss of SOM has been occumng due to cultivation for at least the last 2000 years. Buringh (1984) estimated that the cumulative losses over that period now total 27% of the total SOM originally stored in soils. Due to

Climatic change. soil organic m i t e r and tropical agroforesiry

115

global warming, future losses will accrue from uncultivated soils as well. There is abundant evidence that the rate of decomposition changes more rapidly as a function of temperature than does productivity. However, it has been argued that the stimulation of photosynthesis, the "fertilization effect" of increasing carbon dioxide, could more than compensate for these temperature effects per se. This is proving to be a more complicated problem than it might appear at first glance. How the temperature response of net primary productivity might be modified by increased CO, concentrations is still difficult to predict. It is now clear, however, that the effects of the expected simultaneous increases in CO, concentrations and temperature will have interactive effects on plant photosynthesis and respiration, perhaps combining to lower net photosynthesis at temperatures below 18°C and increasing them above (Allen et al. 1990). If this proves to be a general pattern it may mean that tropical latitudes will benefit more from the fertilization effect than temperate and boreal zones. Experiments have been planned for the IGBP that would provide some definitive answers. In terms of the global atmospheric CO, pool however, it is the balance of effects on decomposition and production that are of concern. If the stimulation of photosynthesis by CO, is the predominant effect, then CO, concentrations should decrease as temperature increases and thus generate a negative feedback to CO,. If the stimulation of decomposition and loss of SOM is the predominant effect, then CO, concentrations should increase with temperature and thus generate a positive feedback to CO,. Several lines of evidence indicate that decomposition with positive feedback is the more likely of the two, and may have already been detected. First, temperature and CO, increase and decrease in tandem throughout the 160,000 year record provided by the Vostok core (Houghton and Woodwell 1989). While cause and effect cannot be established with certainty by such a correlation, the fact that CO, tracks temperature increases closely during deglaciation, but lags when the temperature trend reverses, strongly disconfirms a negative feedback effect. The pattern is consistent, however, with positive feedback increasing the rate of release of carbon dioxide from SOM during warming and decreasing it upon cooling. Second, the rate of accumulation of CO, in the atmosphere has recently increased. According to Houghton and Woodwell (1989) the rate has risen from about 1.5 ppm per year to 2.4 ppm year since the middle of 1987. They suggest that this reflects positive feedback from the loss of SOM in Northern hemisphere soils warmed during the decade of the 80s. Third, the latitudinal dependence of the seasonal amplitude of carbon dioxide peaks in the northern hemisphere. It rises sharply from the equator to a maximum between 50 and 60"N, suggesting that the ecosystems there are the source of carbon dioxide driving the seasonal amplitude increase (Gammon et al.

116

E H . From

1985). Taken together with the latitudinal dependence of change in SOM, which also shows a peak above 50"N (Fig. 9 . 3 , this observation gives additional weight to Houghton and Woodwell's attribution. It is also likely that much of the potential for loss of SOM has already been exhausted by cultivation of soils at latitudes below 50"N. Such losses are estimated to average approximately 30% over the first 20- 50 years of cultivation (Schlesinger 1986), an amount greater than the predicted loss due to a warming of 4°C at those latitudes.

IMPLICATIONS FOR TROPICAL AGROFORESTRY Management of SOM for productivity maintenance or enhancement is a unifying feature of tropical agricultural and agroforestry practices (Young 1976, 1989). Such practices rest on the knowledge that the availability of plant nutrients in tropical soils is largely a function of SOM (e.g. Sobulo and Jaiyeola 1977; Nair 1984). Climatically induced changes in SOM, therefore, have particular implications for the management of tropical soils. Although the percentage changes in SOM that are predicted for the tropics are small relative to those at higher latitudes, decreases in SOM by any amount will only increase the intrinsic challenges of agriculture in the tropics. Although temperate agriculture has performed satisfactorily in many locations for some time by returning only the annual crop residues to the soil, systems that also involve perennial woody plants are more generally practiced in the tropics. In fact, shifting cultivation, an indigenous form of agroforestry based on alternating periods of cultivation and fallow, is currently the largest category of land use in the tropics and subtropics (Detwiler and Hall 1988). In 1980 shifting cultivation was practiced by an estimated 500 million people on 21% of the total area of tropical forests (Lanly 1985). Each year 3.4 million hectares of closed forests and 1.7 million hectares of open forests are converted to shifting cultivation (Lanly 1982). Shifting cultivation is also the only form of land use that has been explicitly recognized as a source term in global cycles (Ha0 et al. 1989). Recent calculations based on the area estimates of Lady (1985) suggest that the annual conversion of primary closed forests to shifting cultivation releases 252541 x 10l2g C ha-* and conversion of open forests 38-60 x 10l2g C ha-l (Ha0 et al. 1989). Agroforestry systems are thus central to the assessment of effects on climate as well as of climate's effects. Shifting cultivation produces a mosaic of patches in the forest or savannah matrix: cultivated patches that have been recently cleared and burned, and fallow patches that are in different stages of succession. When practiced by populations at low densities, in its traditional and culturally integrated forms, the soil fertility that is lost during cultivation is regained during the fallow interval (Nye and Greenland 1960; Aweto 1981a, b; Young 1989).

117

Climatic change, soil organic matter and lropical agroforeslry

Nye and Greenland (1960) estimated that the decomposition coefficient under forest fallow in West Africa was generally on the order of 0.03 and under cultivation 0.04. The calculated relaxation time for a decomposition coefficient of 0.03 is 23 years, for 0.04, 17 years. It is therefore logical that the period of fallow must be longer than the period of cultivation for the SOM balance to be maintained. The effect of global warming on the decomposition coefficient will be to reduce the cultivation period even more and increase the time required for fallow, thus reducing the intensity of land use that can be sustained. This will increase the urgency of research to develop alternatives that can sustain more intensive use. A recent series of papers by La1 (1989a-e) provides a comprehensive evaluation of the changes in stocks of soil carbon and nitrogen that occur with different land use systems with the potential to replace shifting cultivation by providing a greater intensity of land use. The experiment included four land use systems which were studied over a period of four years: plow-till, no-till, and two agroforestry systems, alley cropping with Gliricidia at 4 m spacings, and alley cropping with Leuceana at 4 m spacings. The annual crops in each system were maize and beans. C and Nlevels declined in all cropping systems but increased, by comparison, in nine years of fallow (Table 9.1). Table 9.1

Differences in carbon and nitrogen densities between 1982 and 1986for the land use systems studied by Lal. Data comparing year 1 with year 10 from the study of Aweto (1981a. b ) are given for comparison. Both sets of data are for the surface 10 em of soil

Land use

Change in C (kg rn-2)

Change in N (kg m-2)

Plow-till

- 1.594

- 0.131

No-till

- 0.607

- 0.085

Leucaena 4m

- 0.653

- 0.167

Gliricidia 4m

- 1.208

- 0.145

9 Years of fallow

+- 0.657

+ 0.056

The data clearly indicate that these systems were not sustainable alternatives to the fallow systems they might replace. For the maintenance of SOM without fallow in the tropics, more research will be needed (see Chapter 18 by Sanchez el al.). The studies of Aweto and Lal exemplify research designed to assess soil changes. For detection of climatic effects, however, controls will be needed to differentiate effects of changes in land use from confounding changes due to climatic change. In addition, the intrinsic variability of soils means that sampling

118

EN. Frau

designs must be adequate to detect expected changes at predetermined levels of significance. For the purpose of monitoring climatic change, knowledge of soil changes during fallow could provide a more or less consistent basis for evaluating soil changes for other land uses. Much of the research in agroforestry in fact, is an explicit search for systems that improve on rotational tree fallows as a basis for soil conservation (Young 1989). Improvements may also be achieved through species selection. For example, the ratio of plant carbon bound in long lived tissue per year to plant carbon bound in short lived tissue per year, the wood to litter production ratio, is variable among species. To compensate for the effect of climatic change on the decomposition coefficient, tree species that allocate greater amounts to litter product and have lower wood to litter production ratios could be selected for planting in an "improved fallow" or on cropland. Selecting species for litter quality and timing of litter additions could be effective in synchronizing mineralization rates with the nutrient requirements of the growing crop in spatial agroforestry systems. The macroclimatic scenario is based on a deviation from the base line mean. Specific microclimates deviate substantially from that mean. Isolated mature trees, for example, have a pronounced effect on the microenvironment beneath the canopy. Compared to open areas, the canopies of Acacia rurtilis and Adansunia digirara in Tsavo National Park, Kenya, reduce incident solar radiation at the soil surface by 45-65%and soil temperatures by 5 1 1 ° C (Belsky et al. 1989). The soils under these trees are also higher in SOM than the surrounding grassland soils by a factor of 1.8 (3.6%compared to 2.0%), because additions tend to be greater and decomposition losses less than in the open (Cunningham 1963). One of the best known agroforestry systems in West Africa is based on the cultivation of crops under the canopy of Faidherbia (formerly Acacia) albida (Felker 1978). In addition to microclimatic and chemical effects associated with the subcanopy microsite, Faidherbia soils also have higher water holding capacities. With the potential €or microclimatic effects along with the other benefits of trees in cropland, planting trees could help to ameliorate the local effects of global warming and provide a sink for global atmospheric carbon dioxide as well (Franz 1989).

REFERENCES Allen, S.G., S.B. Idso, and B.A. Kimball (1990). Interactive effects of COz and environment on net photosynthesis of water lily. Agriculture, Ecosystems and Environment 30:8188. Aweto, A.O. (1 98 la). Secondary succession and soil fertility restoration in Southwestern Nigeria: I. Succession. Journal of Ecology 69:601-607.

C ’ l i m [ i c change, soil organic m u e r and fropical agroforeslry

119

Aweto, A.O. (1981b). Secondary succession and soil fertility restoration in Southwestern Nigeria: 11. Soil fertility restoration. Journal of Ecology 69:609-614. Belsky, A.J., R.G. Amundson, J.M. Duxbury, S.J. Riha, A.R. Ali, and S.M. Mwonga (1989). The effects of eees on their physical, chemical, and biological environments in a semiarid savanna in Kenya. Journal of Applied Ecology 26:1005-1024. Ruringh, P. (1984). Organic carbon in soils of the world. pp 91-109 IN: G. M. Woodwell (Ed). The Role of Terrestrial Vegetation in the Global Carbon Cycle: Measurement by Remote Sensing. John Wiley & Sons, Ltd., Chichester. Cunningham, R.K. (1963). The effect of clearing a tropical forest soil. Journal of Soil Science 14:334-345. Davis, M.B. (1989). Lags in vegetation response to greenhouse warming. Climatic Change 15:75-82. Detwiler, R.P. and C. A.S. Hall. (1988). Tropical forests and the global carbon cycle. Science 239:42-47. Felker. P. (1978). State of the art: Acacia albida as a complementary permanent intercrop with annual crops. Report to USAlD. University of California, Riverside. Franz, E.H. (1989). Tropical agroforestry. Background paper for the international workshop on the contribution of greenhouse gas emissions from agricultural systems to climate change, Intergovernmental panel on climate change. Gammon, R.H., E.T. Sundquist, and P.J. Fraser (1985). History of carbon dioxide in the atmosphere. pp. 25-62. IN: J. R. Trabalka (ed). Atmospheric carbon dioxide and the global carbon cycle. United States Department of Energy, DOE/ER-0. National Technical Information Service, Springfield, Virginia. Hao, W.M., M.H. Liu and P.J. Cruuen (1989). Estimates of annual and regional releases of C 0 2 and other trace gases to the atmosphere from fires in the tropics, based on the F A 0 statistics for the period 1975-1980. Paper presented at the Third International Symposium on Fire Ecology, Freiburg University. Federal Republic of Germany, 16-20 May 1989. Houghton, R.A. and G.M. Woodwell. (1989). Global climatic change. Scientific American 260(4):36-44. Jenkinson, D.S. and A. Ayanaba. (1977). Decomposition of C-14 labeled plant material under tropical conditions. Soil Science Society of America Journal 41:912-9 15. Kononova, M.M. (1975). Humus of virgin and cultivated soils. pp. 475-526. In: J.E. Gieseking (Ed). Soil Components Volume I: Organic Components. Springer-Verlag. Berlin. Lal, R . (1989a). Agroforestry systems and soil surface management of a tropical Alfisol: I. Soil moisture and crop yields. Agroforestry Systems 8:7-29. Lal, R. (1989b). Agroforesrry systems and soil surface management of a trop. Alfisol: 11. Lal, R . (198%). Agroforestry systems and soil surface management of a tropical Alfisol: 111 Changes in soil chemical properties. Agroforestry Systems 8:113-132. Lal, R. (1989d). Agroforestry systems and soil surface management of a tropical. Alfisol: IV. Effects on soil physical and mechanical properties. Agroforestry Systems 8:197-215. Lal, R. (1989e). Potential of agroforestry as a sustainable alternative to shifting cultivation: Concluding remarks. Agroforestry Systems 8:239-242. Lanly, J.-P. (1982). Tropical Forest Resources. F A 0 Forestry Paper 30. Food and Agriculture Organiz.ation of the United Nations. Rome. xii+lO6pp. Lanly. J.P. (1985). Defining and measuring shifting cultivation. Unasilva 37(147):17-21. Lieth, H. ( 1 973). Primary production: terrestrial ecosystems. Human Ecology 1:303-332. Nair, P.K.K. (1984). Soil productivity aspects of agroforestry. International Council for Research in Agroforestry, Nairobi.

120

E H . Franz

Nye, P.H. and D.J. Greenland (1960). The soil under shifting cultivation. Commonwealth Agricultural Bureaux. Farnham Royal, Bucks, England. viii+l56pp. Parton, W.J., D.S. Schimel, C. V. Cole and D. S. Ojima. (1987). Analysis of factors controlling soil organic matter levels in great plains grasslands. Soil Science Society of America Journal 51:1173-1179. Pennington, W. (1986). Lags in adjustment of vegetation to climate caused by the pace of soil development: Evidence from Britain. Vegetation 17: 105-118. Post, W.M., W.R. Emanuel, P.J. Zinke, and A.G. Stangenberger (1982). Soil carbon pools and world life zones. Nature 298:156-159. Post, W.M., J.Pastor, P.J. Zinke, and A.G. Stangenberger. (1985). Global patterns of soil nitrogen storage. Nature 317:613-616. Prentice, I.C. (1 986). Vegetation responses to past climatic variation. Vegetation 67: 131141. Ritchie, J.C. and C.V. Haynes. (1987). Holocene vegetation zonation in the Eastern Sahara. Nature 330:645-647. Schlesinger, W.H. (1986). Changes in soil carbon storage and associated properties with disturbance and recovery. pp 194-220. IN: J. R. Trabalka and D. E. Reichle (eds). The Changing Carbon Cycle: A Global Analysis. Springer Verlag, New York. Sobulo, R.A. and K.E. Jaiyeola (1977). Influence of soil organic matter on plant nutrition in Western Nigeria. pp. 105-1 15. In: Soil Organic Matter Studies, Volume I. International Atomic Energy Agency, Vienna. Sorenson. L.H. (1981). Carbon-nitrogen relationships during the humification of cellulose in soils containing different amounts of clay. Soil Biology and Biochemistry 13:313321. Stevenson, F.J. (1986). Cycles of Soil Carbon, Nitrogen, Phosphorus, Sulfur, Micronutrients. John Wiley & Sons, New York. Webb, T. 111. (1986). Is vegetation in equilibrium with climate? How to interpret latequaternary pollen data. Vegetation 67:75-9 1. Woodwell, G.M. (1989). The warming of the industrialized middle latitudes in 1985-2050: Causes and Consequences. Climatic Change 15:31-50. Young, A. (1976). Tropical Soils and Soil Survey. Cambridge University Press. Cambridge, UK. Young, A. (1989). Agroforestry for Soil Conservation. CAB International, Wallinford. UK. Zinke, P.J., A.G. Stangenberger, W.M. Post, W.R. Emmanuel, and J.S. Olson (1984). Worldwide Organic Soil Carbon and Nitrogen Data. ORNLflM-8857. Oak Ridge National Laboratory, Oak Ridge, Tennessee.

121

Chapter 10

THE USE OF MODELS OF SOIL PEDOGENIC PROCESSES IN UNDERSTANDING CHANGING LAND USE AND CLIMATIC CHANGE J.W.B. Stewart*, D.W. Anderson*, E.T. Elliott**, and C.V. Cole**

*

**

College of Agriculture, University of Saskatchewan, Saskatoon, Saskatchewan S7h' OWO, Canada NREL, Colorado State Univ., Fort Collins, CO. U.S.A.

INTRODUCTION This paper addresses the functioning of agroecosystems as one aspect of changing land use in varying atmospheric conditions. Agroecosystems are the most intensively managed ecosystems, located on the most productive land with the result that farmers have custody of more environment than any other group (Paarlberg 1980, Elliott and Cole 1989). Agroecosystems provide us low food and fiber costs but also have contributed to the degradation of the environment. Problems such as soil erosion, soil depletion, concurrent increased fertilizer use, and environmental pollution are actively being considered in areas of North America, Asia and Europe with quite different cultivation histories. The critical issue is: "Do we understand the functioning of the agroecosystems well enough to manage them for productive and sustainable agriculture given the cyclical nature of the weather patterns and the intensive use of the land?" An agroecosystem has been defined as an interactive group of biotic and abiotic components, some of which are under human control, that form a unified whole (ecosystem) for the purpose of producing food and fiber (Elliott and Cole 1989). The major driving variables are tillage, crop management and changing climate patterns. Our objective was to determine: 1) the current status (i.e.soi1 quality etc.) of the agroecosystems; 2) how this current state was reached; and 3) to predict long-term effects of existing and new management practices, changing weather patterns and impending climate change on soil quality and agricultural productivity. We illustrate this approach in the North American Great Plains, a fragile agroecosystem, which has already been subjected to considerable degradation and change as the result of a combination of management decisions and climatic change.

BACKGROUND Using concepts and procedures of quantitative pedology on a small number of soils selected along climo-, topo- and cultivation chronosequences we were able to expand on the use of driving variables or soil forming factors (cf. Jenny

122

J.W.B. Stewart

el a1

1980) and to integrate them (Fig. 10.1) with hierarchical perspectives used to describe ecosystems (Reiners 1986, Elliott and Cole 1989). We studied the processes and properties that characterize the highly dynamic nature of former grassland and bordering forest soils of the Great Plains of North America, especially those involved in organic matter stability, in nutrient cycling, and in the solution and movement of soluble components (Anderson 1987, Roberts et al. 19891. REGIONAL INFORMATION

-

climate change

- new management practices Impact of genetically altered organisms

GEOGRAPHIC

--

Experiment\ Field management treatments - Field manipulations - Greenhouse experiments Laboratory microcosms

I

Validations

I

Olhers

-

Process .Models - Photosynrhesis, plant growth - .Microbial decomposition - <;as fluxes - Water and solute flow - Orhers

Fig. 10.1

REGIONAL SITE NETWORKS

..U S D A - ~ S -

--

sites ,+r,cu~tural Canada sites State Experiment Station sites ( U S . ) Provincial research sites (Canadian) Private institute sites Olhers, worldwide

SCS-National Soil Survey Area Database * USDA-National Resource inventory * Climate databases * SCS-pedon databases Biological surveys * Other databases

Process level information is integrated and used to develop the agroecosystem model or modifr existing models. Sets of driving variables generated from geographic information and organized with CIS are used us model input. Model output is assigned to corresponding subregions using GlS to obtain regional prediction Cfrom Elliott and Cole 1989)

These studies emphasize that soils must not be viewed as relatively static bodies that are developed gradually over very long periods of time. True, they do reflect the properties of the original parent material, past climate and vegetation, topography and time, but more accurately within the time frame of global change concerns, soil development is presented (Fig. 10.2) as the result of basic processes of additions, removals, transformations and translocations (Anderson 1989). All of these processes can occur simultaneously, therefore the present

Use of soil models 10 undersfand land w e - and climale changes

123

characteristics of a soil represents an integration over the recent past of a large number of temporarily variable processes (e.g. Solonetz, Solodized-Solonetz and Solod soils occurring in landscapes of similar age usually in close juxtaposition) .

Fig. 10.2

Soil forming factors based on the concepts of Simonson (1959) as described by Anderson (1988)

Soils, in common with many ecosystems, have variables that operate at slow, intermediate and fast rates and it is important to recognize the nature of the variables studied. These may not always move soil propenies or development in the same direction. Thus, many soil properties are highly dynamic whereas others are static (see Table 10.1). Variables such as soluble salts are highly dynamic, varying over season and reaching tentative equilibrium in a few years, whereas organic matter levels have a time dimension of decades to centuries with carbonate and particularly clay weathering having a scale of millennia in semiarid climates (Anderson 1977). Interestingly, despite a medium time dimension for organic mattcr buildup in soils, at another level of detail one can differentiate fast (mainly microbial processes), intermediate components where turnover is dampened by physical sorption to clay and slow or chemically stabilized humus components in soils (Anderson 1979; Jenkinson and Rayner 1977).

124

J.W.B. Sfewarf el al.

Table 10.1

Grouping of soil related processes and components based on time

Highly Dynamic Soluble nutrients

Dynamic Adsorbed nutrients

More static, slow Nutrient reserves in minerals

Active or soluble organic matter

Labile organic matter adsorbed to clay

Chemically stabilized organic matter

Solution and movement of soluble components

Weathering of carbonate minerals

Weathering of silicates and clay minerals

Microbial growth

Micro-fauna and meso-fauna Plant growth

Vegetation, i.e. forest

Related to time is the concept of threshold controlled responses or mechanisms. Threshold controlled responses are considered general in the understanding of the development of above ground ecosystems and geomorphic surfaces, but have only recently been discussed in soil science (Stoner and Ugolini 1988). Episodic events that initiate a threshold response result in considerable change in properties over short time spans. In this aspect soils must be considered as being highly dynamic.

DATA BASES - SOURCES AND INTEGRATION A number of ongoing, long-term studies of cropping management systems studies at Agriculture Canada and the United States Department of Agriculture Agricultural Research Service research stations (Campbell et al. 1989, Haas et al. 1957) provide a valuable data base. They made it feasible to integrate and synthesize considerable information on long-term dynamics of agroecosystems without having to wait many years for research sites to mature. Soil inventories at 1:100,000 scale or better are available for the North Great Plains area within Canadian or U.S. soil survey data bases. For instance, the National Soil Survey Area Database (NSSAD) currently contains soils information for approximately 2500 of 3000 soil survey areas in the U.S. in a single database. NSSAD has two components of significance, the first is the map unit component which consists of site specific descriptions of individual soil types (series). These descriptions include information pertaining to the soil physical setting, horizonation, areal extent, predominant land use, crop types and crop yields. The second component is the soils interpretation record, which contains interpretive information for each individual soil type. A Soil Conservation Service database, which contains chemical and morphological attributes (organic C, N, sand/silt/clay percentage and bulk density are included, by horizon, for each soil) for approximately 300 cultivated and 500 uncultivated

Use of soil models

10

understand land me -and climate changes

125

soils of the Great Plains, was used in validation of models (Cole et al. 1989). Extensive climate data are available which includes mean annual and growing season (April through September) precipitation, temperature and potential evapotranspiration for approximately 560 weather stations in the U.S. Great Plains. A measure of site productivity (Sala et al. 1988) and decomposition (Parton et al. 1987) are included for each station. Similar data exist in Canada. The individual "map layers" described above have been used to form a composite GIS that integrates soils, climate, and management data to a single scale of resolution. Overlay procedures in ARC/INFO ((Environmental Systems Research Institute, Redlands, CA) are being used to combine the component map layers, producing a dense polygon map. The resulting composite polygon attributes can then be extracted and subjected to an external classification procedure, producing 100-500 classes of polygons, each with corresponding climate, management, and soils attributes. The ARC/INFO geographic information system seems well suited for the purpose of integrating these multiple data sets. More importantly, it should be possible to utilize a variety of existing data bases using this system.

PROCESSES, INTERACTIONS AND PREDICTIVE SIMULATION MODELS Progress in understanding the processes which are important in elemental interactions and which are important in soil quality considerations have recently been reviewed (Anderson 1988; Stewart and Cole 1989). Generalized agroecosystem models are used to integrate information on driving variables, processes and properties (e.g. Parton et al. 1989a). Processes already studied and integrated into models include above and below ground primary production, decomposition and nutrient cycling. Major driving variables include climate (temperature and water), parent material (soil texture), base status, total S and P, topography and management. The effects of these controlling factors are expressed over a wide range of resolutions, from the global down to regional landscape and field plot levels. We have been using the Century model (Parton et al. 1983), which was developed to simulate soil organic matter dynamics and plant production in grazed grasslands and agroecosystems. The model simulates the dynamics of C, N, and P in the soil-plant system using monthly time steps (Fig. 10.3a, 10.3b, 10.3~). The input data required for the model include soil texture, monthly precipitation, maximum and minimum air temperature, and plant lignin content. The Century model has been used to simulate regional patterns of soil C, N, and P and plant production for the U.S. central grasslands region (Parton et al. 1988, 1989a) and the impact of management practices on agroecosystems (Cole et al. 1989).

J . W . B . Slewart

126

el

al.

PLANT RCSIDUE

L O W

STRUCTURAL N (GIN = 150)

-

AtmOSpherE Fixatiun Deposition

METABOLIC N (GIN = 10-25)

WINERAL N (NO3

*

NH4)

Leaching Gaseous Erosion

PASS1 VE (VN

Fig. 10.3

=

9)

continued on next page

I = Immobilization = nineralizatlon

n

Use of soil models

10

understand land use

STRUCTURAL D (CIP 500)

- and

climale changes

127

OUllCl Tert~ I 1rer Depos I t i o n PLANT RESIDUC

flFTABOLlC P (C/P = 80 150

ACTlVt SOIL P (C/P = 30 8 0 )

LABILE P

tn)

Leaching Erosion Crop

OCCLUDtD P PASSIVE SOIL P ( C I P = 20-200)

Fig. 10.3

I = lmmoblii~ation M = Mineralization ontrols D = Temp 8. l(20 T = Texture h p~

Flow diagram for the a ) carbon submodel, b) nitrogen submodel and c) phosphorm submodel (adaptedfrom Parton et al. 1987, 19896)

This model has successfully simulated the impact of cultivation on agroecosystems in the United States (Cole et al. 1987), Canada (Parton et al. 1988, 1989a), and Sweden (Parton et al. 1983) over time periods ranging from 10 to 300 years. The model includes the direct effect of cultivation events on nutrient cycling in soil organic matter dynamics (Elliott 1986). The direct impacts include incorporation of standing and surface residues into soils of different texture, modification of soil temperature and soil water patterns, and increasing the turnover rate of soil organic matter. Comparison of simulated effects with detailed soil organic matter, nutrient cycling and plant production data from regional rcsearch sites validated the model. Larger areas are currently being investigated using this regional modeling approach as this procedure had been applied to Northeastern Colorado to produce maps of net primary production, net N mineralization and nitrous oxide production and to provide summed estimates of these output variables for the entire region (Burke et al. in press). Other more mechanistic models such as the Grassland Ecosystem Model (GEM) have also been used to predict the effects of climatic change on grasslands

128

J.W.B. Sfewort el 01

(Hunt et al. in press), and to represent important feedbacks among primary production, photosynthetic pathways, water use, decomposition, and nutrient cycling processes. For example, it includes mycorrhizae and fauna, and uses more mechanistic translocation, nutrient uptake, shoot and root death and decomposition processes. This model has been applied to analyzing the difference in primary production and nutrient cycling patterns between native shortgrass prairie and introduced crested wheatgrass in Wyoming. The GEM model is able to serve as an aid for interpreting differences in production and nutrient cycling between different crops and among sites. Another important cropping systems model is the nitrogen, tillage and residue management simulation model (NTRM) (Shaffer and Larson 1987) which can be used to quickly determine environmental impacts and provide direct management assistance to farmers in an interactive version, COFARM (Shaffer et al. 1984). I t has been used to simulate many processes such as the decay of crop residues and cycling of C and N. The crop growth submodel consists of a series of subroutines capable of simulating the growth and development of field maize, sweet corn, sorghum, soybean, spring and winter wheat, oat, barley, rye, sunflower, alfalfa, pasture grass, sugarbeet, cotton, peanut, tomato, field pea, sugarcane, sweet potato, and carrot. NTRM is capable of simulating any combination of these crops in rotational sequence, and several years of simulation can be run to estimate the impacts of crop rotation on various parts of the system. It can also be used in studies involving NO,-N loading of groundwater. NTRM model inputs include climate data such as daily maximum and minimum air temperatures, pan evaporation, precipitation, wind run, and solar radiation (optional). Values must be provided for soil physical, chemical, and biological properties of user specified soil horizons. The NTRM model contains crop submodels for growth of tops and roots as a function of climate and soil variables. Processes simulated include photosynthesis, respiration, leaf area growth, grain filling, transpiration, and N uptake. The root growth submodel (Shaffer and Larson 1987) includes root extension, branching, and death. The impacts of soil tillage on the physical, biological, and chemical properties of the soil are simulated using both tillage submodels and specific relationships within other subroutines (Shaffer 1985). Changes in certain soil "macro" properties such as bulk density, percent organic matter, and texture are translated into changes in other properties such as the soil water characteristic curves, C and N transformalion rates, soil water content, soil strength, soil aeration, and nutrient and salt concentrations.

THRESHOLD CONTROLLED RESPONSE A threshold controlled response occurs when a low frequency, high intensity impact or combinations of impacts impinge on a system resulting in a

Use of soil models lo understand land use -and c l i m l e changes

129

new course of development for a previously stable system (Anderson 1989). Intense wind or precipitation events are examples. We can use our knowledge of wind and water erosion phenomena (e.g. Universal Soil Loss Equation) in conjunction with organic matter management to predict the change in soil properties such as erodibility, and to calculate the potential for erosion in a variety of climates and topography. Future long term studies of soil ecosystems must recognize the possibility of rare or intense threshold controlled responses, in that they may be critical to understanding the system.

USE OF THIS APPROACH IN MORE WEATHERED SYSTEMS We have tried, using this approach, to model organic matter dynamics in tropical soils (Parton et a1.1989b) without great success. In part, this is because several of the important processes and controls are not well known. Fewer data are available for tropical soils and soil organic matter dynamics are less well understood. The results from attempts to simulate soil organic matter dynamics in a tropical forest suggest some modifications are needed. Specific research is needed mainly including N and P compounds, to clarify: 1) the effect of soil texture, mineralogy and Fe and A1 chemistry on soil organic matter formation and decomposition; 2) the proportion of soil organic matter in active, slow and passive pools; 3) the dynamics of phosphorus forms in tropical soils, particularly organic P forms; and 4) the extent of the leaching of organics, including N and P compounds in tropical soils of varying mineralogy and texture. These processes are mainly the controlling factors of dynamic soil properties in weathered soils.

CONCLUSION This paper discusses a few of the models that have been developed to understand the functioning of agroecosystems in relatively young unweathered soils. These models or combinations of these models can be used to simulate the effects of a range of climatic conditions on crop productivity and soil quality. We plan to continue this approach concentrating on the technology transfer mechanisms that will unite theoretical approaches with practical action. Extrapolation of this approach to more weathered soils will depend on a better understanding of several major processes. These include the effect of soil inorganic constituents and selected minerals on abiotic and biotic formation and stabilization of organic matter, especially in weathered soils.

REFERENCES Anderson, D.W. (1977). Early stages of soil formation on glacial till mine spoils in a

130

J.W.B. Slewarl

el a1

semiarid climate. Geoderma 19:ll-19. Anderson, D.W. (1979). Processes of humus formation and transformations in soils of the Canadian Great Plains. J.Soil Sci. 30:77-84. Anderson, D.W. (1987). Pedogenesis in the grassland and adjacent forests of the Great Plains. Adv. Soil Sci.7:53-93 Anderson, D.W. (1988). The effect of parent material and soil development on nutrient cycling in temperate ecosystems. Biogeochem. 5:71-97. Anderson, D.W. (1989). Long term ecological research - a pedological perspective p 394-417 in Final Rep. of Int Workshop, Berchtesgaden on "Long Term Ecological Research - a Global Perspective" Publ. by the German Nat. Comm. for UNESCO/MAB Bonn, FRG. Burke, I.C., D.S. Schimel, C.M. Yonker. W.F. Parton and L.A. Joyce (1990). Regional modelling of grassland biogeochemistry using CIS. Landscape Ecology (in press) Campbell, C.A., et al. (1989). Crop rotations on the Canadian prairies - an Agriculture Canada perspective. Agric. Can. Bull.. Ottawa. Cole, C.V., J.W.B. Stewart, H.W. Hunt, and W.J. Parton (1987). Cycling of carbon, nitrogen, sulfur and phosphorus : Controls and interactions. Trans.XI1Ith Int. Society of Soil Sci., Hamburg Symposia and Joint Symposia, VI ; 636-643,. Cole, C.V., J.W.B. Stewart, W.J. Parton, and D. Ojima (1989). Modelling land use effects of organic matter dynamics in the Great Plains of N. America.p89-98 in M. Clarholm and L. Bergstrom (eds.) Ecology of Arable Land - Perspectives and Challenges. Martinus Nijhoff Publishers Dordrecht. Elliott, E.T. (1986). Aggregate structure and carbon, nitrogen and phosphorus in native and cultivated soils. Soil Sci. SOC..h e r . J. 50:627-633. Elliott, E.T., and C.V. Cole (1989). A perspective on agroecosystem science. Ecology 70:1597-1602. Haas, H.J., C.E. Evans, and E.F. Miles (1957). Nitrogen and carbon changes in Great Plains soils as influenced by cropping and soil treatments. USDA Tech. Bull. 1164. Hunt, H.W., M.J. Trlica, J.K. Detling, E.F. Redente, C.V. Cole. J.C. Moore, D.E. Walter, M.C. Fowler, D.A. Klein, and E.T. Elliott (in press). Simulation model for the ecosystem level effects of climate change in temperate grasslands. Ecol. Model. Jenkinson, D.S. and J.H.Rayner (1977). The turnover of soil organic matter in some Rothamsted classical experiments. Soil Sci. 123:298-305. Jenny, H. (1980). The soil resource. Ecol. Studies Vol. 37. Springer Verlag, New York. Paarlberg, D. (1980). Farm and Food Policy. Univ. of Nebraska Press, Lincoln, NE, USA. Parton, W.J., D.W. Anderson, C.V. Cole, and J.W.B. Stewart (1983). Simulation of soil organic matter formation and mineralization of semi-arid agroecosystems. In: R K. Lowrance,ed.,Nutrient cycling in agricultural ecosystems, pp. 533-550, University of Georgia, Col.of Agr., Spec. Pub. No. 23, Athens, GA. Parton, W.J., D.S. Schimel, C.V. Cole, and D. Ojima (1987). Analysis of factors controlling organic matter levels in grassland soils Soil Sci. SOC.Amer. J. 51:1173-1179. Parton, W.J., J.W.B. Stewart, and C.V. Cole (1988). Dynamics of carbon, nitrogen, phosphorus and sulfur in cultivated soils: a model. Biogeochem. 5:109-131. Parton, W.J.. C.V. Cole, J.W.B. Stewart, D.S. Schimel, and D. Ojima (1989a). Simulating the long-term dynamics of C, N and P in soils. pp. 99-108. In: M. Clarholm and L. Bergstrom (eds.) Ecology of Arable Land - Perspectives and Challenges. Martinus Nijhoff Publishers Dordrecht. Parton, W.J., R.L. Sanford, P.A. Sanchez and J.W.B. Stewart (1989b). Modelling soil organic matter dynamics in tropical soils. pp. 153.171. In D.C. Coleman, J.M. Oades and G. Uehara (eds.) Dynamics of Soil Organic Matter in the Tropics. Univ. Of Hawaii Press. Manoa.

Use of soil models 10 undersland land use - and climale changes

131

Reiners, W. A. (1986). Complementary Models for Ecosystems. Amer.Nat. 127: 59-73. Sala, O.E., W.J. Parton, L.A. Joyce, and W.K. Lauenroth (1988). Primary production of the central grassland region of the United States. Ecology 69:40-45. Shaffer, M.J. (1985). Simulation model for soil erosion-productivity relationships. J. Envir. Qual.. 14:144-150. Shaffer, M.J. and W.E. Larson ed. (1987). NTRM, a soil-crop simulation model for nitrogen tillage. and crop-residue management. U.S. Dept, Cons. Res. Rept. No. 34-1. Shaffer, M.J., J.B. Swan, and M.R. Johnson (1984). Coordinated farm and research management (COFARM) data system for soils and crops. 1. Soil and Water Cons. 39:320-324. Simonson, R.W. (1959). Outline of a generalized theory of soil genesis. Soil Sci. SOC.Amer. Proc. 23: 152-156. Stewart, J.W.B., and C.V. Cole (1989). Influence of elemental interactions and pedogenic processes on soil organic matter dynamics. Plant and Soi1,115:199-209. Stoner, M.G. and F.C. Ugolini (1 988). Arctic pedogenesis: Threshold controlled subsurface lcaching eposidcs. Soil Sci. 145:46-51. Roberts, T.L., J.K. Bettany and J.W.B. Stewart (1989). A hierarchical approach to the study of organic C, N, P and S in Western Canadian soils. Can. J. Soil Sci. 69:739-750.

This Page Intentionally Left Blank

133

Addendum to Chapter I0

MODELLING NITROUS OXIDE PRODUCTION BY DENITRIFICATION J.R.M. Arah and K.A. Smith The Edinburgh School of Agriculture West Mains Road, Edinburgh EH9 3JG. UK.

INTRODUCTION Developing rational responses to anticipated climate change depends on understanding the processes by which this change is brought about. Increasing concentrations of nitrous oxide (N20) in the atmosphere contribute both to the degradation of the stratospheric ozone layer and to the socalled "greenhouse effect" (Bouwman 1990). Much of this N 2 0 derives from soils and land use, a significant fraction of it from denitrification, which is the dominant source of N20 emission from poorly aerated, nitrate rich soils. Denitrification produces both N2O and molecular nitrogen (N2), in highly variable proportions; here the term '"20 fraction" is used to denote that fraction of gaseous products of denitrification which reaches the atmosphere as N20. Denitrification fluxes and N 2 0 fractions measured in the field using the acetylene inhibition technique (Ryden et al. 1979) show extreme spatial and temporal variability. High fluxes are usually associated with low N 2 0 fractions; beyond that it is difficult to see any pattern (Arah and Smith 1990). This indicates that the relationship depends on a number of parameters, the interaction of which must be modelled in order to predict current rates and N 2 0 fractions, together with any alterations which might occur in response to anticipated climatic changes.

THE MODEL A deterministic model, based on knowledge of the actual processes

occurring in the soil rather than on simple curve fitting, should apply to all soils under all climatic regimes. Arah and Smith (1989) and Arah (1990) describe such a model, developed for aggregated temperate soils but applicable in principle to all. Denitrification occurs when nitrate is present in anaerobic microsites, which develop wherever the microbial demand for oxygen ( 0 2 ) exceeds the diffusion controlled supply. This may occur where 0 2 diffusion is impeded by water, either at the centers of soil aggregates (Smith 1980), or in small saturated regions

134

JR.M. Arah and K A . Smith

within a structureless soil (Arah 1988), or wherever the local 0 2 demand (reduction potential) is exceptionally high (Parkin 1987). Whatever the precise mechanism causing local anaerobism is, denitrification in such a microsite may be modelled as a process of simultaneous diffusion and reaction of 0 2 , nitrate and N20. Enzyme mediated microbial reactions are assumed to follow dual substrata Michaelis Menten kinetics, nitrate reduction is inhibited by 0 2 , nitrate and N20 is inhibited both by O2 and by nitrate (McConnaughey and Bouldin 1985). Intramicrosite concentration profiles of 0 2 , nitrate and N 2 0 are calculated by an iterative method implemented on a microcomputer. The only requirement for convenient mathematical simulation is some degree of symmetry; linear, cylindrical or spherical. Linear symmetry is appropriate to homogeneous, structureless soils and sediments; the rhizosphere best approximates to a cylinder, and both organic "hot spots" and physical aggregates are approximately spherical. Denitrification rates and N20 fractions of larger volumes of soil (containing many microsites differing in size, moisture content and reduction potential) may be obtained by integrating microsite rates over appropriate probability distributions. The model is thus deterministic at a microsite scale, and stochastic in its macroscopic predictions.

MODEL RESULTS Analysis of the response times of microsite denitrification rates and N20 fractions to a major perturbation, such as instantaneous saturation suggests that spherical soil microsites under about 1 cm in diameter may usually be treated as if they were in a state of pseudo equilibrium (Arah 1990). Linear and cylindrical model microsites of similar dimensions respond equally quickly. This considerably simplifies the application of the model. A preliminary prediction is that, in soils where nitrate supply is not limiting, the denitrification rate of macroscopic soil samples should be proportional to the anaerobic fraction calculated according to the method of Smith (1980); in experiments conducted by Parkin and Tiedje (1984) this was found to be the case. Similar experiments performed on different soils in Scotland (unpublished data) support this finding. The calculated N 2 0 fraction of spherical model microsites is extremely sensitive to microsite radius and reaction potential, and considerably less so to the intra microsite gaseous diffusion constant and the external 0 2 concentration (Arah and Smith 1990).

CONCLUSION The model provides a means for rationalizing the otherwise extremely confusing interactive effects of the parameters known to influence N20 emission

DenilriJcation: model nilrow oxide production

135

by denitrification. The problem of relating microscopic parameters to macroscopic properties can be approached form either end: process modelers must attempt to "coarsen" their model parameters to fit in with the information available in soil mapping and GIS databases; compiles of the latter must be prepared to take account of internal inhomogeneity within their mapping units (Burrough 1989), especially where the phenomenon of interest is as variable as N20 emission. It is to be hoped that the relatively precise parameters of this model (microsite diffusion constants, size distributions and reduction potentials) can be related to more readily measured macroscopic properties, some relatively stable (such as soil texturc), and others likely to alter at various rates under the impact of climatic change (including soil organic matter content, pH, rainfall and temperature). If this can be achieved, thcn the model should permit large area predictions of N 2 0 emissions in a changing environment.

REFERENCES Arah J.K.M. (1988). In: Nitrogen efficiency in agricultural soils (D.S Jenkinson and K.A. Smilh, Eds), Elsevier. London pp. 433-444. Arah J.R.M. (1990). Biology and fertility of soils 9:71-77. Arah J.K.M. and Smith K.A. (1989). Steady state denitrification in aggregated soils: a mathematical model. Journal of Soil Science 40: 139-149. Arah J.R.M. and Smith K.A. (1990). Factors influencing the fraction of the gaseous products of soil denitrification evolved to the atmosphere as nitrous oxide. In: A.F. Bouwman (Ed) Soils and the greenhouse effect. Wiley. Chichesler, pp 475-480. Bouwman A.F. (Ed) (1990). Soils and the greenhouse effect, Wiley Chichester. Burrough P.A. (1989). Journal of Soil Science 40: 477-492. McConnaughey P.K. and Bouldin D.R. (1985). Transient microsite models of denitrification 1: Model development. Soil Science Society of America Journal 49:886-891. Parkin T.R. (1987). Soil microsites as a source of denitrification variability. Soil Science Society of America Journal 51:1194-1199. Parkin T.R. and Tiedje J.M. (1984). Soil biology and biochemistry 16:331-334. Ryden J.C., L.J. Lund, J. Letey and D.D. Focht (1979). Direct measurement of denitrification loss from soils 11: Development and application of field methods. Soil Science Society of America Journal 43: 110-118. Smith K.A. (1980). A model of the extent of anaerobic zones in aggregated soils, and its potential application to estimates of denitrification. Journal of Soil Science 31:263277.

This Page Intentionally Left Blank

137

Chapter I 1

IMPACT OF ANTHROPOGENIC ATMOSPHERIC POLLUTION ON SOILS, WITH SPECIAL RELEVANCE TO TROPICAL AND SUBTROPICAL SOILS, AND POSSIBLE CONSEQUENCES OF THE GREENHOUSE EFFECT N . van Breernen Department of Soil Science and Geology Agricultural University, P. 0. Box 37, 6700 A A Wageningen. the Netherlands

INTRODUCTION Widespread recognition of an appreciable effect of atmospheric pollutants on soils has come only recently, with most emphasis on the socalled acid rain problem. Hitherto, acid atmospheric deposition is a problem of the industrialized Northern countries, and there is little evidence of a serious occurrence in tropical and subtropical countries. However, with the expected increase in the use of fossil fuel for industrial purposes in a number of tropical countries, the problem could become greater in the future (Fig. 11.1). In addition to acid precipitation, I will discuss the related problem of nitrogen deposition, and will discuss possible consequences of the greenhouse effect for soil acidification caused by atmospheric deposition.

?/A Sensitive

SOIIS

Present emissions

Fig. I 1 .I

0Present problem areas

.-.

I.-;

Potential problem areas

Schematic map showing regions that currently have acidification problems, and regions where acidification might become severe in the 1988, by permission of SCOPE Publications) future (Rohde et d.,

138

N. van Breemen

RECOGNITION OF THE PROBLEM Acid atmospheric deposition was widely recognized as a problem in relation to acidification of lakes and streams in Scandinavia and Northeastern America in the early seventies (Likens and Bormann 1974). Decline in numbers of salmonid fish were soon found to be associated with low pH and high levels of dissolved aluminium (Schofield and Trofhar 1980). Although some argued that acidification in relation to changes in land use were more important (Rosenqvist 1983), it is now generally recognized that high acidity in precipitation caused by increased industrial emissions of SO2 was the main cause of water acidification and fish decline. Dramatic improvements of surface water quality by decreasing ambient atmospheric acid loads have been demonstrated in small watersheds in Southern Norway where precipitation was intercepted by transparent roofs, and replaced by preindustrial precipitation (Wright et al. 1988). Ulrich et a]. (1979) were among the first to call attention to acidification of forest soils from acid atrnosphcric deposition and its potentially harmful effects on forest ecosystems. The first comprehensive effort to make an inventory of acid rain problems in tropical countries was provided by Rohde and Herrera (1988).

NATURE OF ACID ATMOSPHERIC DEPOSITION AND EFFECTS O N SOILS IN NORTHERN TEMPERATE COUNTRIES Acid atmospheric deposition may consist of dissolved, gaseous and particulate matter that is cither acid or potentially acid. The source of the (potentially) acid substances include SO2 (from burning fossil fuel), NO, (from combustion processes), and NH3 (mainly from animal manure). The potential acidity of NH3 stems from its conversion to H N 0 3 - in soils by nitrifying microorganisms. Potentially acid components are also formed naturally: SO2 from volcanic aclivity, dimethylsulfide formed in marine environments, NO, in atmospheric processes involving other N gases as precursors, and NH3 from organisms and organic rcmains. However, the concentrations of most acidic precursors have increascd dramatically in many areas due to man’s activities. The prccursor gascs arc rather reactive, and have atmospheric residence times in the ordcr of hours to a weck or so. As a result, most of the S and N emitted is dcpositcd within several hundrcd to a few thousand kilometer of their source area. So, contrary to the greenhouse problem which is caused by more inert gascs, acid dcposition is a regional, not a global phenomenon. H2S04 or HN03in acrosol form or present in rainwater or snow is the main component of acid deposition in areas that arc relatively remote from the source areas. Closer to source arcas, dry deposition of the precursor gases is often more imponant. Due to higher rates of wet plus dry deposition of potentially acidic substanccs, prcsent day ratcs of soil acidification in most Western and Central

Impact of anrhropogenic atmospheric pollution on soils

139

European forests are much higher (2-6.5 kmol ha-1 yr-l) than the rates of natural soil acidification (generally 0.1 to 1 kmol ha-' y r l ) and in parts of the USA and Canada affected by acid deposition (1-1.5 kmol ha-' yr-') (van Breemen et al. 1984). pH Values in rain water do not differ much between the two continents: the main reason for the difference is higher dry deposition of sulfur and nitrogen in Europe (due to higher concentrations of gaseous N and S). Ammonia is an important potentially acid substance in the atmosphere in many European areas. Nitrification of deposited NH, may be intense even in acidic (pH 3.5-4.5) soils (De Boer 1989; Stams et al. submitted) and contributes to soil acidification. The high levels of atmospheric ammonium result from N H 3 volatilized from animal manure, produced in large quantities by widespread intensive stockbreeding activities, a pollution source that is of particular concern in the Netherlands and surrounding areas (Asman and Drukker 1987). The high acid loads in many European forest soils have resulted in ecologically relevant chemical changes in the soil such as decreased pH, decreased levels of exchangeable bases, and increased levels of dissolved Al, as shown by comparison of soil samples taken decades ago and taken in the 1970s and 1980s from the same sites in the FRG (Ulrich et al. 1980; Butzke 1988), Austria (Glatzel and Kazda 1985) and in South Sweden (Hallbacken and Tamm 1986; Fakengren-Grerup et al. 1987). These chemical changes are in accordance with those expected from various simulation models that take into account acid atmospheric deposition, soil hydrology, nutrient cycling, weathering, soil microbial processes etc. (Van Grinsven 1988). Sulfate sorption capacities of most European forest soils have been saturated, and mineral soil horizons now not only frequently have pH values (pH in aqueous extracts or in the soil solution) in the order of 3.3 to 4,and base saturations lower than 5 percent, but also soil solutions are often dominated by inorganic A13+ (which is potentially phytotoxic), with sulfate and nitrate as the accompanying anions (Mulder et al. 1987; Mulder 1988; Matzner and Ulrich 1987; Tyler et al. 1987). Concentrations of A13+ are usually in the order of 1-10 mg l-l, and seasonal peak concentrations of up to 50 mg 1-1 and 500 mg 1-1 have been reported in the most acidified soils in the FXG (Raben 1988) and in the Netherlands (Kleijn et al. 1988) respectively. In fact, present-day pH buffering of acid atmospheric deposition in many soils is mainly by dissolution of amorphous A1 compounds that have been accumulated in the soil by centuries to millennia of soil formation. The pools of these compounds in the surface soil are now being depleted rapidly (Mulder et al. 1989). As a consequence, continued acid atmospheric deposition at ambient levels, may cause soil pH values to drop to values closer to those of unbuffered dilute sulfuric-nitric acid mixtures at concentrations determined by the local acid atmospheric deposition and the degree of evaporative concentration of incoming meteoric water. At deposition rates as given above, and percolation rates between 200 and 500 mm per year, theoretical minimum pH values of the percolating water are

140

N.van Breemen

between 2.5 and 3.4. Atmospheric deposition of nitrogen in Western and Central European forest areas is generally between 10 to 70 kg ha-l yr-' (Moseholm et al. 1988). The net annual N uptake in growing timber usually varies from 6 kg ha-' yr-'(in slowly growing coniferous forests) to 15 kg ha-' y r l (in deciduous forests on productive sites). As a result soils tend to become enriched in N, as evidenced by elevated levels of dissolved nitrate in soil solutions, by elevated levels of exchangeable ammonium in surface soils, and by an increase in the proportion of nitrophilous ground vegetation in many forest areas (Van Brcemen and Van Dijk 1988). Where the capacity of the vegetation and soil microorganisms to tie N in organic form has decreased as a result of forest decline, mineralization and nitrification of soil organic N can further contribute to soil acidification and nitrate pollution of ground water. This so-called N saturation problem is of great concern at present in Western Europe. It is somewhat paradoxal that the effects of acid atmospheric deposition were discovered in the downstream parts of the hydrological cycle, i.e. in lakes and streams, before they were noticed higher up in the cycle, i.e. in soils. The main reasons are probably: 1) that certain highly visible and highly prized aquatic organisms (salmon, trout) are very sensitive to water acidification; and 2) that stream water acidification is particularly important when residence time of water in soils, and thus the possibility for buffering of acidity in soils, is limited. For instance, a relatively large fraction of atmospheric acidity is transfercd unbuffered to surface water during snow melt events, in areas with shallow, acid soils, and in higher, steeper reaches of watersheds. The findings discussed above refer mainly to to forested and heathland areas with naturally acidic soils: in agricultural soils and calcareous forest soils, soil acidification (or alkalinization) from other causes (COz, crop removal, fertilization, liming) generally overwhelm the effect of acid atmospheric deposition.

RISK OF SOIL AND WATER ACIDIFICATION FROM ACID DEPOSITION IN THE TROPICS AND SUBTROPICS Mean areal and per capita emissions and depositions of acidic and potentially acid substances are much lower in most countries outside Europe and North America, but exact data are relatively scarce. Although the composition of air and rainwater in the tropics is generally still close to natural background values, local higher emissions may be important in certain areas. At higher and lower latitudes, acidification can be a serious problem, as e.g. in southwestern China and in southeastern Brazil (Moreira-Nordemann et al. 1988; Zhao and Xiong 1988). A future projection of anthropogenic sulfur emissions by Graedel and

lmpacf of anfhrupugenic afmuspheric pollution un soils

141

Cmtzen (1989) suggests that in areas such as the Gangetic plain in India, with rapid ongoing industrialization and abundant supplies of (high sulfur) coal, acid atmospheric deposition could become an increasingly important problem in the future. In view of the fairly low inputs of N in agriculture, ammonium deposition is not likely to become a problem in most tropical countries. Because many soils in the wet tropics have a relatively low base status, increasing acid deposition could quickly result in ecologically harmful changes in the soils. However, because many tropical soils have relatively high sulfate adsorption capacities, the acidification front caused by sulfuric acid will move relatively slow, so that effects of acid deposition on aquatic systems will be delayed considerably.

POSSIBLE EFFECTS OF CLIMATE CHANGE ON SOIL ACIDITY FROM ATMOSPHERIC DEPOSITION A warming of the climate in areas where soils have been exposed to acid atmospheric deposition for a considerable time could aggravate the soil acidity status. First, N saturation and associated soil acidification and ground- and stream water eutrofication will be stimulated if decomposition of soil organic matter is accelerated by higher temperatures. The same may be true for mineralization of organic S. However, at higher temperatures, the sulfate sorption capacity of podzol Bs horizons may increase, while the solubility of (potentially toxic) aluminum may decrease (Slomp 1989), which may partly offset the aggravating effects noted above.

ESTIMATING CRITICAL LOADS OF ACID ATMOSPHERIC DEPOSITION Recently the concept of critical loads has been developed (Nilsson 1986; Nilsson and Grennfelt 1988). A critical load has been defined as "a quantitative estimate of an exposure to one or more pollutants below which significant harmful effects on specified sensitive elements of the environment do not occur according to present knowledge". The elements considered may be soils or soils plus vegetation, groundwater, or surface water. Criteria can be based on sensitivity of biota (vegetation, fish, man) to certain chemical aspects of their environment. Examples are pH, base saturation, Al/Ca ratio in soil solution, A1 concentration in surface water, NH4K ratio at the exchange complex, nitrate concentration in ground water used as drinking water, or combinations thereof. Critical loads have been set for sulfur and nitrogen deposition for specific ecosystems. While critical loads can be determined experimentally (Boxman et al. 1988), process oriented simulation models are being used increasingly for that purpose. This approach is very promising for pollution control strategies by

142

N . van Breemen

policy makers, if the models are designed for regional application (De Vnes 1987) and are combined with emission-atmospherictransport-depositionmodels for large areas (Kauppi et al. 1985; Alcamo et al. 1987).

REFERENCES Alcamo, 1.. M. Amman, J.P. Hetteling, M. Holmberg, L. Hordijk, L. K m a r i , L. Kauppi, P. Kauppi, G. Kosrnai and A. Miikela (1987). Acidification in Europe: A simulation model for evaluating control strategies. Ambio 16: 234-245. Asman, W.A.H., and B. Drukker (1987). Estimated historical concentrations and depositions of ammonia and ammonium in Europe and their origin (1870-1980). IMOU-Report R87-2. Boxrnan, D., H. Van Dijk and J. Roelofs (1988). Critical loads for nitrogen, with special emphasis on ammonium. In: Critical loads for sulphur and nitrogen. Nordic Council of Ministers, Milj~rapport1988: 15, 417 p. (Nilsson and Grennfeld eds.) pp. 295-322. Butzke, H. (1988). Zur zeitlichen und kleinraumigen Variabi1it;it des pH-Werten in Waldboden Nordrhein-Westfalens. Forst und Holz, 43(4): 81 -85. De Boer. W. (1989). Nitrification in Dutch heathland soils, PhD Thesis, Agr. Univ. Wageningen, 96 p. De Vries. W. (1987). The role of soil data in assessing the large-scale impact of atmospheric pollutants on the quality of soil water. Proc. Int. Conf. on the vulnerability of soil and groundwater to pollutants, Noordwijk aan Zee, the Netherlands. Falkengren-Grerup. U., N. Linnermark, G. Tyler (1987). Changes in acidity and cation pools of South Swedish soils between 1949 and 1985. Chemosphere. Vol. 16, Nos. 10-12, p. 2239-2248. Glatzel. G., and M. Kazda (1985). Wachstum und Mineralstoffernhung von Buche (Fagus s y l vatica) und Spitzahorn (Acer platanoides) auf versauertem und schwermetallbelastetem Bodenmaterial aus dern Einsickerungsbereich. Z. Pflanzenernahr. Bodenk. 148: 429-438. Graedel, T.E. and P.J. Crutzen (1989). The changing atmosphere. Scientific American, 261: 28-36. Hallbacken, L., C.O. Tamm (1986). Changes in soil acidity from 1927 to 1982-1984 in a forest area of Soulh-West Sweden. Scand. J. For. Res. 1 (2). 219-232. Kauppi, P., J. Kamari, M. Posch, L. Kauppi and E. Matzner (1985). Acidification of forest soils: a model for analyzing impacts of acidic deposition in Europe, version 11. Collaborative Paper 27 of the International Institute for Applied Systems Analysis, Laxenburg, Austria: 28p. Kleijn, C.E., G. Zuiderna and W. De Vries (1988). De indirecte effecten van atmosferische depositie op de vitaliteit van Nederlandse bossen. 2. De bodemvochtsamenstelling van 8 Douglas opsranden. Soil Survey Inst. Wageningen, the Netherlands, Rapt. nr. 2050, 67 p. Likens. G.E., and F.H. Bormann (1974). Acid rain: a serious regional environmental problem. Science 184: 1176-1179. Matzner, E. and B. Ulrich (1987). Results of studies on forest decline in Northwest Germany. In: Effects of atmospheric pollutants on forests, wetlands and agricultural ecosystems (T.C. Hutchinson, and K.M. Meema, Eds.), Springer Verlag, Berlin. Moreira-Nordemann, L.M., M.C. Forii, V.L. Di Lascio, C.M. do Espirito Santo and O.M. Danelon (1988). Acidification in Southeastern Brazil. p 197- 255 In: Rohde, H. and R.

lmpacl of anthropogenic almospheric p o i l u i o n on soils

143

Herrera (Eds.). Acidification in tropical countries. SCOPE 36, John Wiley & Sons, 405 P. Moseholm, L., B. Andersen and I. Johnson (1988). Acid deposition and novel forest decline in Central and Northern Europe. Nordic Council of Ministers, Milj~rapport,no. 9. Mulder, J., J.J.M. van Grinsven and N. van Breemen (1987). Impacts of acid atmospheric deposition on woodland soils in The Netherlands: 111. Aluminum Chemistry. Soil Science of Am. J., Vol. 51, no. 6 Nov.-Dec. Mulder, J. (1988). Impact of acid atmospheric deposition on soils: Field monitoring and aluminum chemistry. PhD thesis, Agric. Univ. Wageningen, the Netherlands, Department Soil Science and Geology,. Mulder, J., N. van Breemen and H.C. Eijck (1989). Depletion of soil aluminium by acid deposition and implications for acid neutralization. Nature 337: 247-249. Nilsson, J. (1986). Critical loads for nitrogen and sulphur. Nordisk ministerrld MiljOrapport 1986: 11. Nilsson, J., and P. Grennfelt (Eds.) (1988). Critical loads for sulphur and nitrogen. Nordic Council of Ministers, Milj~rapport1988: 15, 417 p. Raben, G.H. (1988). Untersuchungen zur raumzeitlichen Entwicklung boden- und wurzelchemischer Stressparameter und deren EinfluR auf die Feinwurzelentwicklung in bodensauren Waldgesellschaften des Hils. Berichte des Forschungszentrums Waldokosysteme/Waldsterben, Reihe A, Bd. 38. Rohde, H., and R. Herrera (Eds.) (1988). Acidification in tropical countries. SCOPE 36, John Wiley & Sons, 405 p. Rohde, H., E. Cowling, I. Galbally, J. Galloway and R. Herrera (1988). Acidification and regional air pollution in the tropics, p. 3-39. In: Rohde, H. and R. Herrera (Eds.). Acidification in tropical countries. SCOPE 36, Wiley & Sons, 405 p. Rosenqvist, I.T. (1983). Acidification of freshwaters in Europe. Water Quality Bulletin 8: 137-142. Schofield, C.L., and J.R. Trofnar (1980). Aluminum toxicity to Brook trout (Salvelinus fontinalis) in acidified waters. In: Polluted rain. (T.Y. Toribara, M.W. Miller and P.E. Morrow, Eds.). Environmental Science Research: vol. 17, 341-366. Slomp, C. ( I 989). The influence of temperature on aluminum precipitation and dissolution in Spodosol Bs horizons. Unpubl. report, Department of Civil and Environmental Engineering, Syracuse University, Syracuse NY, USA, 40 pp. Stams, A.J.M., H.W.G. Booltink, I.J. Lutke-Schipholt, B. Beemsterboer, J.R.W. Wottiez and N. van Breemen (in press). 15N field study on the fate of atmospheric ammonium in an acid forest soil. Submitted for publ. Biogeochemistry. Tyler, G., D. Berggren, B. Bergkvist. U. Falkengren-Grerup, L. Folkeson and A. Ruhling (1987). Soil acidification and metal solubility in forest of Southern Sweden. In: Effects of atmospheric pollutants on forests, wetlands and agricultural ecosystems (T.C. Hutchinson and K.M. Meema, Eds.). Springer Verlag, Berlin. Ulrich, B., R. Mayer and P.K. Khanna (1979). Deposition von Luftverunreinigungen und ihre Auswirkungen in Waldokosystemen im Solling. Schriften aus der Forstlichen Fakultat der Universitat Gottingen und der Niedersachsischen Forstlichen Versuchsanstalt. Band 58, 291 p. Ulrich. B.. R. Mayer and P.K. Khanna (1980). Chemical changes due to acid precipitation in a loess-derived soil in Central Europe. Soil Sci. 130: 193.199. Van Breemen, N., C.T. Driscoll and J. Mulder (1984). Acid deposition and internal proton sourccs in acidification of soils and waters. Nature 307: 599-604. Van Breemen, N., and H.F.G. van Dijk (1988). Ecosystem effects of atmospheric deposition of nitrogen in the Netherlands. Environmental pollution, Vol. 54: 249-274.

144

N.

van Breemen

Van Grinsven, J.J.M. (1988). Impact of acid atmospheric deposition on soils: Quantification of chemical and hydrologic processes. PhD thesis, Department of Soil Science and Geology, Agric. Univ. Wageningen, the Netherlands. Wright, R.F., E. Lotse and A. Semb (1988). Reversibility of acidification shown by wholecatchment experiments. Nature 25 August, vol 334: 670-675. Zhao, D. and J. Xiong (1988). Acidification in Southwestern China. p.317-345. In: Rohde, H. and R. Herrera (Eds.) Acidification in tropical countries. SCOPE 36, Wiley & Sons, 405 p.

145

Chapter 12 CHANGES IN SOIL RESOURCES IN RESPONSE TO A GRADUALLY RISING SEA-LEVEL H . Brammer* and R . Brinkman**

*

**

Formerly F A 0 Agricultural Development Adviser in Bangladesh. Present address: 37 Kingsway Court, Hove, East Sussex BN3 2LP, United Kingdom. Land and Water Development Division, FAO. Via delle Terme di Caracalla, Roma 00100, Italy.

ABSTRACT A gradual eustatic rise in sea-level of 0.5 to 1 m over the next century could erode some coastal land, impede soil drainage on coastal plains and salinize soils further inland than at present. However, the effects will vary considerably from place to place depending on such local or regional factors as climate, sediment supply to estuaries and coasts, concurrent land subsidence or elevation resulting from natural or human induced causes, and human interventions to protect land from the effects of a rising sea-level. Location specific monitoring and research are needed to understand environmental change processes, identify appropriate responses and estimate their technical, ecological and socioeconomic feasibility. They should include studies of recent and subrecent palaeo-environments which developed during past periods when the relative sea-level was rising gradually, world wide or locally.

INTRODUCTION This paper examines the probable effects of an eustatic rise in sea-level on coastal soils. It assumes, for the sake of analysis, that such a rise in sea-level will take place, will occur gradually and might amount to approximately 0.5 to 1 m over the next century. The paper ignores possible direct effects of climatic change on soils and land use that might be taking place concurrently with a rising sealevel. It also ignores the possible effects of a sudden, catastrophic rise in sea-level of up to 6 meters which could occur if the West Antarctica ice sheet were to separate and slide off into the sea (Hekstra 1989). A sea-level rise of 1 meter, which could affect some land even up to the present 5 m contour by storm surge flooding and salt water intrusion up river estuaries, could potentially affect a gross area of about 5 million km2, equivalent to about 3 percent of the world's total land area (Hekstra 1989). However, the coastal lowlands potentially at risk include about one-third of the world's total cropland and have a present population of about 1 billion - a figure which might be doubled within the next one hundred years. Thus, the potential effects of a 1 m rise in sea-level could be very much greater than would be suggested by the small proportion of the world's land and soils that might be affected.

146

H . Brammer and R . Brinkman

It is important to recognize at the outset that these potential figures grossly exaggerate the soil areas and population that might actually be at risk. 'The populist scenario that has been portrayed of sea-level rising contour by contour to overwhelm 20% of Bangladesh and the Nile delta (e.g. Broadus et al. 1986) is far too simplistic. Such a scenario ignores dynamic changes that will be taking place in coastal lowlands concurrently with or in response to a gradual rise in sealevel. Some of these changes are natural, such as continuing sediment deposition in some deltas and estuary areas which would reduce or totally offset the effects of a rise in sea-level. Others are the result of human action, such as the building of embankments to protect land and people from the effects of a rising sea-level. Little attention appears to have been given to date to soil changes that might take place with a climatically induced rise in sea-level. However, prediction of the likely consequences on soil properties and productivity need not be entirely speculative. There is abundant geological, historical and contemporary evidence of the effects of a rise in sea-level on adjoining land areas. Examples include: The effects of the 100 m eustatic rise in sea-level following the last glacial period on the coastal areas bordering the North Sea as described by Jelgersma (1988); Land subsidence in relation to sea-level of about 1 m during the last 2000 years in the Netherlands, attributed to tectonic subsidence of the North Sea basin and eustatic sea-level rise by geoidal changes following the last glacial period (Jelgersma 1988); Geological and archaeological evidence of land subsidence in the Mississippi, Nile, Rhine and other deltas resulting from tectonic warping or compaction of alluvial sediments (Morgan 1970; Stanley 1988; Jelgersma 1988); Recent evidence of local land subsidence occumng as a consequence of water, natural gas or oil extraction in the Northern Netherlands, Venice, Bangkok, Osaka and elsewhere (Jelgersma 1989; UNEP 1988; Volker 1989); and world-wide evidence of a 10 to 15 cm eustatic rise in sea-level during the lasl 100 years (Hekstra 1989). There is also evidence from a number of places that human interventions could aggravate the effects of a rising sea-level: e.g. by interrupting sediment supply to deltas by dams across rivers upstream, as has happened in the Nile delta following the construction of the Aswan high dam in 1964 (Coutellier and Stanley 1987), and by embanking rivers to channel flood flows directly to the

Effec! of euslalic sea-level rise on soil resources

147

sea, as in the Mississippi delta, thus depriving basin areas of sediment supplies which formerly offset land subsidence due to compaction of the deltaic sediments (Day and Templet 1989). In several localities, rates of land subsidence by natural causes or directly attributable to human activities are much greater than the rates of sea-level rise estimated to occur in response to global warming during the next century.

PROCESSES OF CHANGE IN A COASTAL AREA RESULTING FROM GRADUAL SEA-LEVEL RISE A gradual sea-level rise may cause a complex of changes in different parts of an estuary or coastal plain, depending on a range of external factors and on the nature of the area itself. These factors include the length of the dry season, the tidal range, the incidence of cyclonic storms and storm surges, the exposure of the coastline and the nature of the shelf offshore, and the supply of sandy or fine textured sediments to the area by longshore currents or from rivers. The nature and extent of the changes also depend on the nature of the sediments in the coastal area (clayey, sandy, peat), on the closed or open nature of the coast and the presence of coastal ridges or dunes, on the presence and density of any network of tidal tributaries, and on the presence of any coastal defences such as sea-walls, or other modifications to the natural hydrology by human engineering, such as polder embankments. Increased frequency of storm surges combined with rising mean sea-level may cause lateral erosion of exposed coasts, with landward movement of any sand, shell or clayey beach ridges. Changes inland in saline, brackish water and fresh water tidal zones and the lower parts of river plains will vary with the climatic conditions and are described separately below.

Changes in seasonally wet-dry climates In the tidal or estuarine floodplains, storm surges may bring intermittent salt water flooding, inducing soil salinization. Spring tide flooding during a rising sea-level may also cause seasonal salinization, particularly in the dry season or near equinoxes. In a broad band along saline or brackish tidewater, the natural rates of vertical accretion will be increased by a slow rise in sea-level, the rates also depending on the sediment supply, storm surge incidence and extent of spring tide flooding. In these areas, as well as in the clay plains further away from tidewater, the increased incidence and duration of brackish water flooding will also extend and intensify reducing conditions in soils. This increasing incidence of salinity would reduce the available length of growing period in the remainder, which would tend to decrease potential yields.

148

H . Brammer and R . Brinkman

In the widening belt influenced by brackish water flooding, fresh sediment and increased reduction, mangrove may begin to dominate again (generally Rhizophora species in the tropics; or reeds such as Phragmites in temperate climates). These vegetation types cause increased accumulation rates of soil organic matter and, in combination with sulphate from the brackish water, accumulation of pyrite and organic S compounds. Potential acid sulphate conditions thus may become more widespread. A rising sea-level will cause river beds to aggrade and levees to increase in height along lower river courses, thus impeding drainage from interior basins with consequent raising of water tables and increasing depth and duration of seasonal flooding. Soils subject to fresh water flooding or seasonal saturation will be reduced for longer periods and their organic matter contents will tend to increase. Land that becomes perennially wet will be liable to accumulation of a peat cover, or existing peat areas will increase in thickness and extent. Such changes in soil conditions, unless counteracted by protective embankments or artificial drainage, would generally reduce soil productivity or potential. Specifically: Land suitability of levee soils for perennial dryland crops would decrease; Areas suitable for annual dryland crops would decrease and this land might become better suited for wetland crops such as rice or jute; Areas presently shallowly flooded and well suited to high yielding wetland rice cultivars might only be suited for traditional, tall varieties; and Land now deeply flooded but dry for part of the dry season and used for floating rice and perhaps a short season oilseed crop, may become permanently flooded and unsuitable for cultivation. The aggradation of river courses and raising of levee levels along lower river courses associated with a rising sea-level will also tend to increase channel instability. This could reduce the intervals between potentially catastrophic shifts in river courses in major deltas such as that of the Ganges and Brahmaputra rivers in Bangladesh (Brammer 1990) or those of the Hoang Ho and Yangtze rivers in China. Changes in arid and semiarid climates Low lying coastal areas in tropical and or semiarid climates generally have narrow bands of mangrove vegetation along tidewater; in very arid environments, there may be as little as a single line of low trees. On the landward side, this is bordered by dry, hypersaline, barren or virtually barren land which is shallowly flooded by seawater during occasional spring tides or rare storm surges. Much of

Effecf of euslalic sea-level rise on soil resources

149

this water subsequently evaporates, adding further amounts of salts to the surface and the upper soil layers. A gradual rise in sea-level would appreciably increase the flooding frequency in such areas and somewhat increase the flooding depth and the proportion of the saline floodwater flowing back into tidewater. This would entail a partial desalinization of the outer parts of the hypersaline land and a broadening of the band suitable for the growth of mangroves (mostly Avicenniu in this saline to hypersaline environment). In this climate zone, therefore, mangroves would extend further inland, colonizing former barren hypersaline land rather than freshwater swamp or agricultural land. There would be some increase in organic matter content and some pyrite accumulation, but both to much lower levels than in more humid climates. Seawater intruding further inland than at present and reaching present rainfed wetland rice areas during the short wet season, would severely affect their suitability for this land use. Changes in perhumid climates In perhumid, mostly equatorial, climates without a significant dry season, large proportions of coastal and estuarine plains tend to be covered with a layer of eustatic peat (with its surface at about high tide level), except for relatively narrow, low levees along tidal channels and coastal ridges along exposed coasts. Natural vegetation grades from mangrove forest near saline or brackish tidewater to freshwater swamp forcst on the peat. Wetland rice, or oil palm or rubber on drained land (from which any shallow peat layer would have gradually disappeared) may extend to near tidewater, with a narrow protective belt of mangrove remaining. A gradual sea-level rise would tend to widen the strip influenced by brackish water flooding and vertical accretion of fine textured sediment. The latter would be colonized by mainly Rhizophora mangrove and subject to accumulation of pyrite. Much of the ncwly deposited material would thus become potential acid sulphatc soil. This would be at the expense of land presently suitable, or actually used, for wetland rice, coconuts, oil palm, sugarcane, pineapple or rubber. Further from tidewater, peat growth would tend to accelerate under natural conditions and the peat surface would thus keep pace with the rising groundwater level, linked to the rising mean sea-level. On cultivated peat soils, the maintenance of satisfactory drainage conditions would become increasingly difficult and costly with a rising sea-level, which might lead eventually to abandonment, especially of land under tree crops. The land would then revert to swamp vegetation under which peat growth would resume. Where the protcctive strip of mangrove forest is removed or is inadequate in width, higher storm surges associated with a higher sea-level could inject enormous volumes of seawater onto the peat land. This would kill the freshwater

150

H . Brammer and R . Brinkman

vegetation in the area affected. The peat would then become highly vulnerable to wave erosion and open water might form. This would generally be a shallow freshwater lake, subsequently invaded by a floating herbaceous vegetation and, in the long run, recolonized by freshwater swamp forest. A breach in the strip of land separating the lake from the coast or from a wide tidal channel would, however, convert it into a saline or brackish lagoon and thus change the coastline. Changes on coral islands Coral islands provide a special case. It seems to be popularly expected that a 1 meter rise in sea-level would submerge or destroy the homelands of several Pacific island nations (UNEP 1988). Such islands are formed of coral rubble thrown up by typhoon storm waves to a height of a few meters above high-tide level. Apart from any artificial defences that might be built, the extent to which islands would be submerged or destroyed by a rise in sea-level will depend on whether or not the growth of the living coral at individual locations can keep pace with a rising sea-level and continue to supply sufficient coral debris to maintain storm ridges at the same relative height above sea-level as at present. Living corals can grow vertically at rates up to ca. 1 cm annually, except where retarded by muddy or polluted water.

HUMAN ACTIVITIES AGGRAVATING OR COUNTERACTING EFFECTS OF SEA-LEVEL RISE Human activities may influence the response of low lying coastal areas to sea-level rise in different ways. The interventions may be remote, as in the case of dam construction, or local, as in the case of subsidence because of drainage.

Effects of upstream interventions The construction of dams in upstream areas, for hydroelectric power, irrigation supplies or protection of land downstream from river floods, blocks sediment supply downstream and to coastal areas. This prevents or slows down vertical accretion, thus further aggravating salt water intrusion and impairing drainage conditions in riverine, deltaic or estuarine areas. It also diminishes or blocks sediment supply to the coast itself, which may give rise to retreat of the coastline by wave erosion and longshore transport of the eroded material, as has occurred in the Nile delta following the completion (1964) of the Aswan high dam (Coutellier and Stanley 1987). In the Northeastern Nile delta, the problem of sea-level rise would be compounded by the ongoing, probably neotectonic, subsidence of about 0.5 m per century estimated by Stanley (1988); until 1964, this subsidence was compensated by the deposition of sediments from the annual

Effect of eustatic sea-level rise on soil resources

151

Nile floods. Ongoing natural, geological erosion, or human induced erosion caused by large scale deforestation, constitute major sources of sediment for low lying coastal areas. In the Netherlands, the upper 4 m of mainly clayey sediment in coastal areas, deposited over the last 4000 years, largely originated from central European sources. This sediment supply prevented major changes in the coastline during the slow sea-level rise of about 4 meters over the same period. Major soil conservation activities over large upland areas could therefore have effects on coastal lowlands similar to those of large dams. River training works, such as are under consideration for the Brahmaputra river in Bangladesh, can increase the flow of water and sediment through the main channel at high river stages. This tends to increase the sediment supply and deposition downstream, which could offset the effects of a rising sea-level, partly or wholly. Increased water abstraction for irrigation development in upstream areas, particularly from rivers during the dry season, tends to increase the distance of sea-water penetration in water-courses in coastal areas. This would aggravate the effects of sea-level rise in cutting off existing irrigation inlets from fresh water supplies and by extending the area of coastal saline soils. Effects of local interventions Pumping of groundwater or natural gas or oil may accelerate land subsidence to rates up to several times those envisaged for the sea-level rise over the next century. Natural gas extraction, for example, has lowered part of the Northern provinces in the Netherlands by about 0.6 m over about 30 years, and parts of the Mississippi delta are subsiding more rapidly than in the 19th century, probably in part because of oil and gas extraction. Parts of Bangkok city have subsided by more than 1 m, and are currently subsiding by about 0.12 m per year, by large scale pumping of water for urban and industrial supply. Parts of Taipei have subsided more than 2 m during the last 20 years for the same reasons (Poland 1984). This subsidence is creating serious land drainage and saline intrusion problems in affected areas. Groundwater extraction from coastal aquifers may also induce sea water intrusion into these aquifers at rates far exceeding those envisaged as a result of sea-level rise alone, particularly in arid and semiarid climates. Sea-level rise would therefore compound an already growing problem. The fresh groundwater in coastal dunes and coral islands often occurs in the form of a lens, in dynamic equilibrium with approximate vertical isostasy at the boundary between fresh and saline water. This requires that the groundwater level be above mean sea-level by an amount of about one-fortieth of the total thickness of the fresh water lens. If the sea-level rises, the volume of the fresh water lens would remain dynamically stable only if its surface would also rise by

152

H . Brammer and R . Brinkman

a similar amount. If the groundwater surface were to remain constant, the depth of fresh water storage would be reduced in due course by up to 40 times the sealevel rise, which could drastically reduce or eliminate local water supplies in coral islands and coastal dune areas. A narrowing of dunes or coral islands by coastal erosion would entail similarly serious consequences for the capacity of a fresh water lens. Sediments in low lying coastal areas normally subside by slow compaction, which can be compensated by vertical accretion of fresh sediment. The Mississippi delta, for example, was subject to compaction and sedimentation at a rate of about 1 m per century before human influence became dominant (Titus 1987). Embankment against sea water intrusion or fresh water flooding stops vertical accretion in the embanked area, but subsidence by compaction or by oxidation of peat continues, so that drainage conditions tend to deteriorate. Eventually, this leads to loss of land to shallow fresh water lakes in basin areas. Artificial drainage of coastal soils by ditches or subsurface drains, with evacuation of the drainage water through sluice gates at low tide or by low lift pumping, will accelerate land subsidence. This might be very rapid in peat soils, which are compacted as well as oxidized at high rates after drainage and air entry, but it can occur also in clayey soils. This entails a need for progressively higher embankments and greater pumping heads of the drainage water, with consequently increasing capital and running costs, aggravating the effects of a sea-level rise. On and around coral islands, as in other cases, human activities tend to compound the dangers from sea-level rise: e.g. by effluents, intensive fishing and tourism damaging or slowing the growth of the coral, and thus decreasing the source of material essential for the continued existence of coral islands in a dynamic equilibrium with the sea.

LESSONS FROM THE PAST About two thousand years ago, extensive areas in the southwestern and Central parts of the Netherlands were peat swamps, protected from the sea by levees, coastal ridges and, locally, dunes, as well as by their own elevation: the peat had grown up somewhat above high tide level (Zagwijn 1986). During the post Roman transgression, the sea breached the coast in several places, and the peat was naturally drained and subsided to near sea-level along the tidal inlets. Since about 1000 A.D., farmers colonized the peat areas by cutting shallow ditches for gravity drainage and by repeated burning of the soil surface to accumulate enough nutrients for a crop. By the time the general land level had gradually subsided through these causes to about mean sea-level, and locally lower, a series of major storm surges broke through the protective bands of mineral soils, generally at the sites of estuaries or inlets. The sea-water destroyed

Effecr of eurlalic sea-level rise on soil resources

153

crops and natural vegetation, leaving the peat soils an easy prey to wave action. Large areas in the Northcentral and Southwestern Netherlands thus became part of the North Sea. It took nature and the people centuries to repair the damage. The present IJsselmeer (the former Zuiderzee) is a gash that remained - albeit presently as a fresh water lake embanked against the sea and with polders reclaimed from part of the lake bottom. Coastal peat areas in the tropics that are used for agriculture need protection against such catastrophes in case of a rise in sea-level. Since the end of the last glacial, about 12 to 10 thousand years ago depending on the criteria adopted, there has been a eustatic rise of sea-level of about 100 m , very rapid initially (almost all of the rise took place before 6000 years B.P.), diminishing to rates of the order of 1 m per century around that time and essentially zero since then. Sedimentary landscapes from the time of the gradual sea-level rise, about 6000 to 5000 years B.P. (mid Holocene) have been recognized in different parts of the world. In the coastal plain of the Guyanas, for example, as well as in the Mekong delta (Brinkman and Pons 1968; Brinkman et al. 1990), there are extensive areas of soft clay soils which contain much finely distributed mangrove root material and high proportions of pyrite. These soils occur inland from younger clay plains where soils have less organic matter and much lower pyrite contents. The two landscapes are separated from each other by a beach ridge or major levee landform; in other places, they grade into each other. The soil material in the older, inland part of the coastal plain extends down for several meters, or down to still older, generally very firm,Pleistocene sediments. The most inland parts of this landscape, furthest away from the rivers, are covered by deep peat which grades laterally into the soft clay. The soft, pyrite rich, peaty clays or clays with large amounts of fine root material have been deposited from brackish water in a very broad band of mainly Rhizophora mangrove forest while the sea-level rose at rates similar to those speculatively estimated for the next century. Where 14-C dates are known, they indicate ages between about 6000 and 5000 years B.P., as for a shell ridge separating the older landscape from the younger one, and for the upper soil layers of the older landscape. The younger landscape has a sediment sequence clearly indicating its genesis under conditions of essentially constant sea-level. The upper clay layers are underlain by silty clay material with very thin silt layers at depths ranging between 1 and 3 m; and broad or narrower clay plains are interrupted by sand (or locally shell) ridges running roughly parallel to the present coast or to present estuary levees. Unless there is human intervention, for example in the form of embankment, it would be expected that this young Holocene landscape would again be buried by a very broad band of brackish water swamp forest, slowly

154

H . Brammer and R. Brinkman

rising by vertical accretion with a stipulated slowly rising sea-level and accumulating soft, pyrite-rich clay with high organic matter contents. Landward from this broad landscape, renewed peat accumulation would be expected to occur. The primary productivity of a swamp forest would be ample for raw peat accumulation to keep pace with a sea-level rising at a rate of 1 m per century. Because dead plant material is decomposed much more rapidly above groundwater or floodwater level than when submerged, present day peat accumulation rates are generally smaller.

MONITORING AND RESEARCH REQUIRED Monitoring in carefully selected type locations with very good baseline information will be needed to quantify changes in soils resulting from sea-level rise, including: Disappearance (by lateral erosion); Rejuvenation by reworking (lateral transport); Rejuvenation by vertical accretion; Salinization; Desalinization (from hypersaline conditions); Reduction (increasing period and intensity); Increasing organic matter content; Peat formation; Accumulation of pyrite and organic S compounds. Monitoring will be needed in natural environments as well as where t k present day nature and rates of these changes are influenced by human actions: for example, by embankment or extraction of deep groundwater, natural gas or oil. Such monitoring will not only increase our understanding of the processes themselves, but also aid in understanding and modelling changes by a stipulated sea-level rise. Any corrective action that may need to be taken can then be identified and its technical, ecological and economic feasibility assessed. Research does not start at zero as far as the response of low lying coastal areas to a sea-level rise is concerned. There have been many instances of gradual sea-levcl rise in the recent and more distant past as well as, locally, ongoing at present. Some of these cases have been studied and described or monitored in some detail, as indicated above. Only in a few cases, however, have soils been taken explicitly into account. Further studies are needed on subrecent and recent palaeo-environments where the sea-level was rising, eustatically or for tectonic or other reasons, at rates of about 0.2 to 1 meter per century. Cases could include: the North Sea coast between Denmark and Belgium about 4000 years B.P.; the shallowly

EfJeci of euslaiic sea-level rise on soil resources

155

buried mid Holocene landscape South of Dhaka, Bangladesh, recognized by Brammer and Brinkman in 1969; the Nile delta (Stanley 1988); and the Musi delta area North of Palembang, South Sumatra, Indonesia (Brinkman 1987).

ACKNOWLEDGEMENTS We thank S. Jelgersma, D. Norse and G.M. Higgins for critical and helpful comments and suggestions on a draft of this paper.

REFERENCES Brammer, H . (1990). Geographical complexities of detailed impact assessment for the Ganges-Brahmaputra-Meghna delta of Bangladesh. In: Wanick, R.A. and T.M. Wigley (Eds). Climate and sea-level change: observations, projections and implications. Cambridge University Press (in press). Brinkman, R. (1987). Sediments and soils in the Karang Agung area. Ch. 4, p 12-22. In: Best, R., R. Brinkman and J.J. van Roon. Some aspects of tidal swamp development with special reference to the Karang Agung area, South Sumatra Province. Mimeo, World Bank, Jakarta, Indonesia. Brinkman, R., and L.J. Pons (1968). A pedo-geomorphological classification and map of the Holocene sediments in the coastal plain of the three Guyanas. Soil Survey Paper No. 4. 40 p. separate map. Soil Survey Institute (Staring Center), Wageningen. Brinkman, R., Nguyen Bao Ve, Tran Kim Tinh, Do Phuoc Hau and M.E.F. Van Mensvoort (1990). Sulfidic materials in the Western Mekong delta, Viet Nam. In press. Paper presented at the Dakar Symposium on Acid Sulphate Soils, January 1986. Broadus, J . , J. Milliman, S . Edwards, D. Aubrey and F. Gable (1986). Rising sea-level and damming of rivers: possible effects in Egypt and Bangladesh. p 165-189. In: Titus, J.G. (Ed.). Effects of changes in stratospheric ozone and global climate. Vol. 4: Sea level rise. UNEP/U.S.EPA. Coutellier, V., and D.J. Stanley (1987). Late Quaternary stratigraphy and paleogeography of the Eastern Nile delta, Egypt. Marine Geology 77: 257-275. Day, J.W. and P.H. Templet (1989). Consequences of sea level rise: implications from the Mississippi delta. Coastal Management 17: 241-257. Hekstra, G.P. (1989). Global warming and rising sea levels: the policy implications. Ecologist 19: 4-15. Jelgersma, S . (1988). A future sea-level rise: its impacts on coastal lowlands. p 61-81 in: Geology and urban development. Atlas of urban geology, Vol. 1 . UN-ESCAP. Morgan, J.P. (1970). Depositional processes and products in the deltaic environment. p. 3147. In: Morgan, J.P. and R.H. Shaver (Eds). Deltaic sedimentation, modem and ancient. Special Publ. No. 15, SOC.of economic paleontologists and mineralogists, Tulsa, Oklahoma. Poland, J.F. (Ed.) (1984). Guidebook to studies of land subsidence due to groundwater withdrawal. Studies and Reports in Hydrology No. 40. Unesco, Paris. Stanley, D.J. (1988). Subsidence in the Northeastern Nile delta: rapid rates, possible causes, and consequences. Science 240: 497-500. Titus, J.G. (1987). Causes and effects of sea level rise. p 125-139. In: Preparing for climate change. Proc. of the first North American conference on preparing for climate change:

156

H . Brammer and R . Brinkman

a cooperative approach. October 27-29 1987, Climate Institute, Washington, D.C. Government Institutes, Inc., Rockville, MD. UNEP (1988). Report of the joint meeting of the task team on implications of climatic changes in the Mediterranean and the co-ordinators of task teams for the Caribbean, Southeast Pacific. South Pacific, East Asian seas and South Asian seas regions, Split, 3-8 October 1988. UNEP(OCA)/WG.2/25. Volker, A. (1989). Impacts of climatic changes on hydrology and water resources of coastal zones. Vol. 1. p 114-127. In: Conference on climate and water, Helsinki 11-15 Sept. 1989. Publ. Acad. Finland, Govt. Printing Centre, P.O.Box 516, SF 00101 Helsinki. Zagwijn, W.H. (1986). The Netherlands in the Holocene. (Dutch; Nederland in het Holoceen. Geologie van Nederland, Dee1 1). Geological Service of the Netherlands, Staatsuitgeverij, 's Gravenhage. 46 p, separate map sheet.

157

Chapter 13

SOILS ON A WARMER EARTH: THE TROPICAL REGIONS W.G. Sombroek International Soil Reference and Information Center, P. 0. Box 353. 6700 AJ Wageningen. the Netherlands.

ABSTRACT After discussing the vagary of the trend in global warming during the past hundred years, an overview is given of the several predictive models of climatic change in the forthcoming century at a doubling of the atmospheric C 0 2 content, with special reference to the tropics and subtropics. The positive effects of rising atmosphere CO, and near surface temperatures on plant growth are outlined; more freshwater for irrigation due to higher evaporation over the oceans; more biomass production, especially of C3-plants; more efficient water consumptive use of plants because of lower stomata1 conductance. Then a discussion follows on the effects of climatic change on soil development and conditions in various upland regions of the tropics and subtropics, taking the outcome of the NCAR modelling of 1984 as starting point (2°C temperature increase; varying seasonal increase or decrease of soil moisture availability), superimposing that on present day climatic conditions and soil patterns. The coastal lowlands and the mountainous cases, where impact of climatic change may be largest, are treated separately.

INTRODUCTION The greenhouse function of the earth's atmosphere can be described as the trapping of part of the thermal emission from the earth's surface, because of the presence of a number of gases. The most important greenhouse gases are water vapor (H20), carbon dioxide (COz), methane (CH4), nitrous oxide (NzO), ozone ( 0 3 ) and chlorofluorocarbons (CFC's). The greenhouse function is an essential part of the life support system of planet Earth. In its absence there would be no life as we know it, because mean global annual temperatures would be minus 18°C. An enhanced greenhouse function of the atmosphere, caused by anthropogenic increase of the above greenhouse gases, is commonly known as the GreenHouse Effect (GHE), or the Arrhenius Effect after its first postulator in 1896. A definite increase since 1850 has been measured in the atmospheric concentration of COz (from 290 to 345 ppm) and CH4 (from 0.85 to 1.7 ppm). This has coincided with a trend towards global warming of about 0.5"C (global mean annual near-surface temperature) in the period 1880-1990 (Hansen et al.

158

W . C . Sombroek

1981 and 1987). Part of this gradual increase, viz. 0.2"C, has recently been ascribed to "urban warming" at some key observation points (Karl et al. 1988). The series of temperature measurements may also be misleading due to changes in the methods of ocean surface temperature measurements in the past hundred years (for instance a change-over from "bucket" to "engine" measurement of ocean surface temperature). Long-term measurements in parts of Europe (Scandinavia, Holland) show no increase at all or even a slight decrease in local temperatures over the period considered. At global level there is a peculiar interruption in the period 1940-1975 when a slight downward trend occurred. This may well be caused by changed solar activity (sun spots) as already mentioned by Scharpenseel in Chapter 1; during the period concerned many researchers speculated on the imminence of a new ice age. Notwithstanding the observed increase in atmospheric C02 and CH4, the rise in global temperatures over the past 100 years is slow and irregular and even non significant statistically; it would be "noise" rather than a "signal". Hansen et al. (1988) estimated that only by the mid nineties a definite signal on temperature rise from the combined effect of all anthropogenic greenhouse gases can be expected.

CLIMATIC CHANGE MODELS There is little doubt that the concentration of anthropogenic greenhouse gases will increase substantially in the forthcoming 50-100 years. A doubling of the atmospheric preindustrial C02 content to about 580 ppm by the year 2050 is not unlikely. Therefore, atmospheric scientists and climate modelers have combined the 0.5"C overall temperature rise in the past 100 years, uncertain as it may be, with the measured atmospheric C02 rise of 55 ppm in the same period, using algorithms of atmospheric processes, to make projections of future climates. A doubling of the atmospheric C02 has been taken as scenario for the development of a number of General climatic Circulation Models (GCMs, see Table 13.1 for the results of five of them). The temperature rises predicted by these models vary between 2.8 and 5.2"C worldwide. However, there is much discrepancy between the various modelling results. Even the present day climate is not simulated satisfactorily (Mitchell 1989). The scientists concerned use varying estimates of the influence of feedback mechanisms by oceans (surface phenomena and deep water upwelling), by cloudiness, by ice coverage, by land cover and land use, by soils, etc. Terms used in this respect are: "negative feedback" which is mitigation of the GHE, and "positive feedback" which is strengthening of the GHE. Most models are of the "steady state" type, which simulate a new equilibrium situation after a sudden increase of atmospheric C02 from 300 ppm to 600 ppm. Models of the "transient response" type which include a year-by-year cumulative time path, with due account of temporal lag

Soils on a warmer earth: tropical regions

159

effect (such as delayed ocean warming) as well as short term effects (such as increasing plant assimilation), are more complicated than steady state models. Comprehensive reviews of the GCM results and their drawbacks have been given by Dickinson (1986) and Mitchell (1989). Table 13.1 Global m a n changes infive recent C02-doubling studies

Study GISS NCAR GFDL

UKMO

osu

Source Hansen et al. (1984) Washington and Meehl(l984) Wetherald and Manabe (1986) Wilson and Mitchell (1987a) Schlesinger and Zhao (1987)

Surface Temperature Change, K 4.2

Precipitation Change, % 11.0 7.1 8.7

4.0 4.0 5.2

15.0

2.8

7.8

Goddard Institute for Space Studies (Hansen et al. 1984). Seasonal data on a 7.83" latitude x 10" longitude grid: mean surface air temperature and precipitation anomalies for (2 x C 0 2 - 1 x C02). GFDL Geophysical Fluid Dynamics Laboratory (Wetherald and Manabe 1986). A 4.5" x 7.5" grid: mean surface air temperature and precipitation anomalies for (2 x C 0 2 - 1 x C02). NCAR: National Center for Atmospheric Research (Washington and Meehl 1984). Seasonal data on a 4.5" x 7.5" grid: mean surface air temperature and precipitation anomalies for (2 x C 0 2 - 1 x C02). OSU: Oregon State University (Schlesinger et al. 1985). Seasonal data on a 4" x 5" grid: mean surface air temperature and precipitation anomalies for (2 x C 0 2 - 1 x C02). UKMO: UK Meteorological Office (Wilson and Mitchell 1987a). Monthly data on a 5" x 7.5" grid: mean surface air temperature and precipitation for control run (1 x C@) and perturbed run (2 x C02). from: Hulme and Mitchell (1989). GISS:

All GCMs predict a much stronger temperature rise at higher latitudes in the respective summers than near the equator. The maximum steady state responses at C02 doubling are upward from 5°C annual mean surface air warming for the high latitudes, but only 2°C for the equatorial regions. The latest models, presumably including not only C02 doubling but also the longer and stronger action of increases of trace gases such as methane and nitrous oxide, are said to predict a global rise in temperature of 5°C. with increases up to 12°C in polar regions (Press 1989). The magnitude of such a change would be comparable to the change from a glacial to an interglacial period in the Pleistocene. On the other hand, within the Intergovernmental Panel on Climate Change (IPCC) a consensus emerged at a recent meeting that mean global temperatures will have gone up by only 2°C by the year 2050 (Anderson 1990).

160

W.G. Sombroek

Associated with such a global warming would be a sea level rise of 0.5-1 m, caused by temperature induced expansion of surface water volume and strongly increased melting of ice in polar regions. These predictions are also fraught with uncertainties, as discussed by Pirazzoli (1989). A rise of only about 0.3 m would be the best estimate according to a recent review in Nature (Meier 1990). The global precipitation would increase by 7-15% (Mitchell 1989), though the effect on freshwater availability would be largely offset by increased evapo(transpi)ration. The precipitation pattern would also change: a smng increase is predicted for the Northern high latitudes; non change or decrease in the mid latitudes and minor increases in the tropics. Estimates on the geography of annual and seasonal precipitation vary widely between the GCMs. Often the predicted changes would not be more than "noise" (Schlesinger and Mitchell 1985), because of problems in modelling of the seemingly chaotic regime of the weather, particularly in the Northern higher latitudes (North America, Western Europe). For the tropics the scarcity of reliable long term weather records is a serious drawback. Some models give the net effect per region of changing seasonal rainfall patterns and increased evapo(trans)piration as expressed in soil moisture availability, though not taking into account changed runoff conditions. An example is the NCAR model (see Fig. 13.1) described in Washington and MeeN (1984). It gives a degree of winter and strong summer drying in the Northern mid

latitudes (Southern USA, Indochina) but more soil moisture in the Northern high latitudes, especially in the winter. In the tropics, the model outputs on seasonal soil moisture changes are complicated: Little or no change in the year-round soil moisture conditions in the monsoon climate of Southeast Asia; Up to 30 mm increase in soil moisture in December-January-February (DJF) for Eastern Amazonia but up to 30 mm decrease for Western Amazonia and Central Brazil; a 10-20 mm increase in eastern Amazonia and central Brazil in June-July-August (JJA) which is at present a dry season in that region, but no change in Westem Amazonia (no dry season at present); Little change in DJF for Westcentral Africa but a decrease of up to 40 mm in JJA, nowadays the middle of the wet season; and Up to 40 mm decrease in available soil moisture in Eastern Africa for DJF (at present already the "long" dry season), and 20-30 mm increase in JJA (at present the "short" dry season). The implication of this is the attenuation of the bimodal pattern of rainfall effectiveness in this region ("long rains" moving more smoothly into the "short rains") but a more pronounced (long) dry season - not considering effects of

Soils on a warmer earth: tropical regions

161

local orography; in Southeastern Africa however, with a single rainy season in DJF, this rainy season would be less pronounced.

SOIL MOISTURE DIFFERENCES, JJA

_,._ -,\.,-.-..---’

________*

Fig. 13.1

Geographical distribution of soil moisture differences in cm for DecemberFebruary and June-August obtained by Washington and Meehl(l984)for 2 x CO2 - I x C02 (NCAR model)

Whatever the merit of the NCAR model, on its own and in comparison with

162

W . G . Sombroek

the other models, - for instance the one by Manabe and Wetherald (1986), as reproduced in Chapter 4,- the above predicted regional changes in soil moisture will be taken as starting point for the discussion of the effects on tropical soil formation and conditions in this paper. First, however, the effect of changing atmospheric C02 and temperature on tropical plant growth will be discussed.

EFFECTS OF RISING TEMPERATURE, ATMOSPHERIC CO2 AND SEA-LEVELON PLANT GROWTH IN THE TROPICS AND SUBTROPICS An increase of global mean annual near surface temperatures will stimulate plant growth in the world-at-large, other conditions remaining the same. A 1°C increase would result in 5 1 0 % extra biomass production worldwide (Idso et al. 1987). The model-surmised higher temperatures will be reflected in the Holdridge Classification of biotemperatures, resulting in changed patterns and acreages of its "life zones" or schematized biomes. The modeling results of Emanuel et al. (1985) show only minor shifts in the main tropical life zone belts ("tropical wet forests", "tropical dry forests", etc.) but it should be noted that any changing rainfall patterns have not been taken into account in that study. In the tropics and subtropics, where temperatures are normally not a limiting factor, the increase in annual biomass production will be modest and grain yields of some major agricultural crops may decline if temperatures happen to surpass 24"-27°C (see Chapter 22). Only in mountain areas such as the Andean Cordillera and the highlands of Eastern Africa the change will be spectacular: the successive natural vegetation or crop growing belts will shift upwards by 1000 feet at an increase of 2°C. In these areas more land would be vegetation covered, though the net acreage of individual altitudinal crop-belts, such as coffee and tea, will become smaller (Ackland 1971 for East Africa, and Braun's report on the Agroclimatic Zone map of Kenya, in Sombroek et al. 1982). A preliminary case study by UNEP-GEMS-GRID for Uganda has shown that suitable conditions for Robusta coffee in Uganda would decrease in acreage to near zero (GRID 1987) assuming no gradual adaptation by introduction of new crop cultivars. The changing acreages are exemplified in Fig. 13.2. Increase in near surface temperatures over oceans will increase evaporation substantially. The present-day annual evaporation of all oceans is 4.4 x 1020 cm3y-', at a global mean annual surface temperature of I1.5"C (Holland 1978). An increase in temperature of only 1°C worldwide would result in about 20% extra evaporation (Handbook of Chemistry and Physics, 1987 edition) or 1.0 x 1020cm3y-', (10,OOO km3 water; compare with 85 km3 annual discharge of the Nile river). The resulting extra water vapor will induce more cloud formation and ultimately more rain- and snowfall. Part of the increased rain- or snowfall will be on the land surface, enlarging the present body of fresh water in rivers and lakes

Soils on a warmer earth: tropical regions

163

and the availability of water for plant growth. Models of the world hydrologic cycle are, however, not yet able to predict how much of this increased fresh water body will accumulate in subhumid and semiarid tropical land areas where it could contribute to increased crop production through irrigation. The effect of a rising sea level will be dramatic, especially in the tropics. Many parts of tropical coastal plains and river deltas such as those of the Mekong, the Sunderbans, the Congo, the Nile, the Amazon and the Orinoco will become flooded to such a depth that natural vegetation and crop growth will be impossible - unless huge empoldering works are executed (see Brammer and Brinkman, Chapter 12, for details). This loss of landbased plant growth will only partially be offset by plankton and algae growth in coastal waters. Interior parts of deltas will have poorer drainage conditions.

Fig. 13.2

Upward changes in altitudinal cropping c.q. vegetation belts in the tropics at a 2 "C increase of annual temperature (a = mountains of Eastern Africa: relatively small decrease in acreage; b = high lands of Uganda: near disappearance of the belt concerned)

In contrast, the positive effect of rising atmospheric C02 levels on landbased biomass production ('TO2fertilization") will be significant. This effect has not yet been duly incorporated in the GCMs. As already mentioned by Scharpenseel (Chapter l), laboratory experiments on a range of crops at doubled atmospheric C 0 2 level have shown an average increase of about 30% in vegetative productivity (Kimball 1983; Goudriaan and Unsworth 1988; Schleser and Kirstein 1989). In natural environments, light and C02 are normally suboptimally present. Hence, photosynthesis of green plants can be stimulated by an increase of ambient C02, by suppression of photorespiration particularly in C3-plants. These plants include nearly all leguminosae and woody plants, and many crops such as cotton, rice, wheat, soya beans, barley, sunflower, and tuber crops (cassava, potatoes). The reaction to increased CO2 is less strong, about lo%,in C4-plants, such as many tall tropical grasses, halophytes and the crops maize, sugar cane, sorghum and millets. CAM-plants such as pineapple, Opuntia and agaves show an intermediate reaction. Another effect of higher atmospheric C02 concentrations is a reduction of

164

W . G . Sombroek

plant stomatal openings, hence smaller water vapor diffusion (higher stomatal resistance). This will increase the plant's water use efficiency, and the process would be particularly strong in some C4-plants such as maize (37% smaller stomatal conductance, leading to a 26% lower transpiration -Cure 1988). The combined result of modest extra assimilation and much smaller transpiration of this crop would also be in the order of 30%, comparable to the pure assimilation effect on C3-plants (see also Schleser and Kirstein 1989). The likelihood of increased dark respiration (dissimilation) because of higher night temperatures may, however, have a mitigating effect. In this context, it is likely that the measured gradual increase in atmospheric C 0 2 over the past 100 years of about 20% is responsible for part of the approximately doubling of the agricultural crop productivity worldwide over that period (Allen et al. 1985, estimated the C02-induced increase to be 15% for soya). Such a CO2-induced increase of terrestrial net primary production would also account for the "missing" sink in the global carbon cycle, resulting from the smaller reported CO;! uptake in ocean surface waters of the Northern hemisphere than assumed previously (Tans et al. 1990). Doubling of C02 in combination with a 3°C mean global air temperahm rise would result in another 25% extra biomass production (Idso et al. 1987). For the tropics the combined effect would be smaller because of:

Less than average predicted temperature rise; The chance that plants become overheated because of too limited evapotranspiration - especially in dry season regions; The likelihood that daylight will become a limiting factor - especially in the below canopy growth environment of standing vegetation and crops; and The likelihood that soil nutrients become limiting - especially in the humid tropics where strongly weathered soils are widespread. Higher C02 concentrations would, however, also stimulate root growth which, combined with more organic matter production and its even stronger decomposition, would intensify rock weathering (see below) - hence deeper soils with extra availability of some nutrients such as potassium from the freshly weathered substratum. There is yet another factor: higher COz concentrations would induce shorter growing periods to maturity for many annual plants (Oecher and Strain 1985), which implies increased possibilities for plant growth in (sub)tropical regions with a short rainy season (plus less sensitivity to soil salinity), and more crops per year in humid areas. Finally, several indirect effects should be mentioned: 9

Increased biomass production leads to better soil cover, hence less erosion (unless the seasonal or annual ratio between rainfall and

Soils on a warmer earlh: tropical regions

165

evaporation changes for the worse); The competitive force of perennial grassy weeds - predominantly C4plants - relative to many crops would become smaller at higher C 0 2 levels, hence weed control would be less costly (the situation becomes more complex in case the crops concerned are C4-plants too, such as sorghum and millets in the Sudano-Sahelian zone of West Africa or semiarid India). In summary, atmospheric C02 doubling and related temperature rise on their own are likely to have a definite positive effect on plant biomass production in tropical forests (C3-plants), less so in tropical savannas (C4-plants), and a significant increase of production potential of most agricultural crops. The demand for fertilizers, especially the nitrogen and phosphorus based ones will, however, strongly increase. It also remains to be established whether other biological influences such as the activity of pathogens, pests, decomposers and symbionts will change at the same or higher rate as biomass productive capacity per se. In contrast with C02, the rising concentrations of atmosphere trace gases CH4 and N20 have no effect on plant growth. In view of their stronger and longer term forcing of the surmised global temperature rise, and their chemical reactivity, they may even turn out to be the most worrisome eventually. One can however imagine systems of "harvesting" CH4 gas in paddy fields and garbage piles for local fuel supply.

EFFECTS OF AN ENHANCED GREENHOUSE EFFECT ON SOIL CONDITIONS AND SOIL DEVELOPMENT IN THE TROPICAL AND SUB-TROPICAL UPLANDS A distinction should be made between soil features that have a short "response time" (RS) and those that change only very gradually (rapidly adjusting soil features vs. slowly adjusting ones). Walker and Graetz (1989) give three RS classes: short ( 4 0 years), long (50-1000 years) and very long (>lo00 years). A more detailed scheme is given in Chapter 4 by Varallyay. It is obvious that in soils the moisture, temperature and the organic matter and nutrient/salts dynamics are easily changed. Within the time frame of Global Change, say 50-100 years, this would not normally affect pedological soil classification, since the current systems are based on more stable/static soil characteristics (a partial exception is the US Soil Taxonomy system, where soil moisture and partly also soil temperature regimes figure at a high categoric level). The modest surface air temperature rise surmised for the tropics (maximum of 2°C) is likely to be even smaller in the soil, because of the denser vegetative covers (higher biomass production because of higher atmospheric CO;?, see

166

W.G. Sornbroek

above). With regard to soil moisture and salinity, much depends on the regional changes in the rainfall and evaporation patterns, but here the models are vague and contradictory.

Changes in dominant soil forming processes Soil forming factors acting slowly, dominant soil forming processes within the Global Change time frame of 50-100 years would not change significantly. There are, however, some fragile ("ecotonal" or "threshold") soil situations, where a slight change in one of the factors would induce a major change in the dominant process ("ferralitization" vs. "podzolization"; "illuviation" vs. "homogenization"; "salinization" vs. "leaching"), without even considering landscape processes such as wind and water erosion. Examples of ecotonal soil situations where a transition to a different genetic soil group may occur within a time span of tens of years, upon only slight changes of climatic conditions or somewhat higher frequency of extreme events are the following: a)

The lighter textured yellowish ferralitic soils of eastern Amazonia, the Congo basin and Kalimantan (xanthic Ferralsols and ferralic Arenosols in the FA0 terminology) may become rapidly podzolized upon small changes in precipitation (total annual amount or more seasonal concentration) and/or increasing acidic organic matter production, resulting in Podzols or albic Arenosols ("Giant Podzols") of which gleyic variants are already present in large areas in the Rio Negro-Rio Branco area of Northwestern Amazonia. The fragility of the present ferralitization process in the Amazon region has been amply demonstrated by detailed studies on short distance sequences of ferralitic soils-Podzols by French pedologists (for instance Turenne (1975) for French Guyana; Dubroeucq and Volkoff (1988) for the Brazilian Rio Negro area; Lucas et al. (1987) for the Amazon in general). The predicted increase of 30 mm effective soil moisture in the NCAR model during the dry season of Eastern and Southern Amazonia may be just enough to induce podzolization over large areas of sandy Pleistocene terraces - where up till now Podzols are of patchy occurrence only (Sombroek 1966; Camargo 1981).

b)

The imperfectly drained loamy soils on the flat water divide areas of the Western Amazon with their plinthitic subsoil (the Madeira-Purus area Southwest of Manaus) may dry out, resulting in an irreversible hardening of this subsoil into "laterite" (change from Plinthosol or plinthic Acrisol into Acrisol, petroplinthic phase - see Camargo 1981; Sombroek 1984) (Fig. 13.3). The decrease in soil moisture availability in DJF predicted by NCAR may induce such drying out, especially if the overall river hydrology would become more erratic.

167

Soils on a warmer earth: tropical regionr

Norlhweslern region (Rlo Negro) Southwesternregion (Rlo Purus)

Eastern region (Lower Amarmas)

(A)

t + t

......, (i'..

----

--a-

_-........, ---

grsnlle oulcrop

LP

sandy sediments

FRX

xanfhlc Ferralsols a 0

loamy to clayey sedimenls

ARa

albic Arenosols

pllnlhlle

PZh

deep humic Podzols

---

pelropllnlhlle

Leplosols

I

'Giant' Podzols

or15IeIn

PZg

gleylc Podzols

hlgh (ground) water level

PTd

dystrk Pllnlhosols

GL

Gleysols undiHerenlialed

seasonally low (ground) water level

--

expanslon of area at cllmallc change

(c)

Fig. 13.3

area wlth palchy occwenca 01 'Gianr Pcdrols

Schematic presentation of some Amazonian ecotonal soil situations

c)

The deeply weathered reddish, loamy to clayey and porous soils of the forest-savannah transition zones of Eastern Africa, which are stable under their present day natural vegetation, may be leached to an extent that a relatively dense clay illuviated horizon develops below an unstable topsoil low in organic matter (changing the soil from rhodic or orthic Ferralsol into an orthic Acrisol or Lixisol). This in fact is happening already at present day land clearing in parts of the region, because of the sudden diminution of the homogenizing action of soil biologic life. The NCAR predicted changes in soil moisture conditions for this area would therefore stimulate the replacement of ferralitization by illuviation as a dominant soil forming process, especially in the present bimodal rainfall areas (Sombroek et al. 1982).

(d)

Certain silty sedimentary deposits in the wide riverine valleys of the Sudano-Sahelian zone of West Africa ("fadama's"; the interior delta of t k Niger river) ma) develop from Fluvisols into saline and/or sodified soils at an even minimal change in precipitation and flooding regimes - as exemplified by current human actions with the same soil-hydrological implications (Sombroek and Zonneveld 1971).

168

W . G . Sombroek

The NCAR model for this region predicts a decrease in precipitation during the rainy season, hence less suppression of salinity.

Changes in soil weathering rates In non ecotonal tropical upland soil situations, the pathway of soil forming processes would not change, but the weathering potential and rate may increase somewhat. The weathering potential at a doubling of the atmospheric C02 per se, as measured by the increase of the solubility of calcite will be discussed first. The present C02 concentration of the atmosphere is taken as 300 ppm or 0.03%. The C02 concentration in the soil gaseous phase is 10 to 100 times higher, because of root respiration. The low values would occur under dry and cool climatic conditions, the high values in continuously warm and moist soils such as in tropical rain forests where root activity is very high (when the soil condition is anaerobic, such as in paddy fields the value of soil C02 may be up to 300 times the air C02; Novozamski and Beek 1976). At a 10 times higher CO2 concentration in the topsoil ( ~ 0 . 3 % )the equilibrium pH of water would be 5.22 (in the absence of any mineral component). With a doubling of the C02 content such as in the GCM scenarios, the C02 concentration in the topsoil would become: 0.30 (original) + 0.10 (33% increase of biomass production) + 0.03 (increase of atmospheric C02) = 0.43%. The corresponding equilibrium pH would be 5.14 l , meaning 20% more H-ions available for protolytic weathering (Fig. 13.4). This can be illustrated by the increased solubility of calcite. In the present situation 113 mg CaC03 1-l can be dissolved by soil water whereas in the future this would be 128 mg 1-l (Fig. 13.5) which means an increase of about 15%.In practice, however, the increase may be smaller than lo%, because the 33% increase in biomass production may not be reached, due to likely limitations of nutrients and light - as discussed before. When the C02 content of the topsoil is taken to be 100 times the atmospheric content (e.g. tropical humid forest) then the solubility of CaC03 increases from 251 mg 1.' to 277 mg 1-I, which is an increase of ca. 10%. Thus, taking into account a slight increase of the effect of rate limiting factors at a higher rate of reaction (e.g. diffusion of components), an average increase of 10% in the rate of weathering due to a doubling of atmospheric CO;!can be expected. The increase in temperature will also lead to a higher rate of weathering.

'

The lowering of pH to 5.14 includes the slight effect of a 2°C temperature increase. If buffering

minerals are present (which is usually the case in soils) then the equilibrium pH will of course be higher.

On h e other hand, this will

be

counteracted by an increase of h e content of organic acids.

Soils

on

a warmer earih: lropical regions

169

%

co,

Fig. 13.4 p l l of water as a function of 3' 6 C02 in air

% co,

Fig. 13.5 Solubility of CaC03 in water of 25 "c as a function of % C02 in air

There are indications that a temperature rise of 10°C results in an increase of velocity of mineral weathering reactions by a factor of approx. 1.5 (Qlo or "Van 't Hoffs factor"; Loughnan (1969), p.69; Zivkovic et al. 1983). A 2°C temperature increase would thus result in a 10% higher rate of weathering.

170

W . G . Sombroek

It follows that the combined effect of the increase of both the C02 content and the temperature on the rate and therefore the depth of Weathering can be estimated at a minimum increase of 20%. Of paramount importance, however, is the influence of the change in climate on the effective rainfall i.e. the water available for the reactions and transport of components. The estimated minimum rate increase of 20% will only occur if the effective rainfall does not decrease. The higher biomass production inevitably leads to a higher amount of organic decomposition products. In view of the somewhat higher soil temperatures, this decomposition of fresh organic matter is likely to be more complete, hence there may in fact be a smaller rather than a higher production of stable humus (see also Tinker in Chapter 7). At the same time the deeper soil profiles offer extra space in the subsoil for fixation of the humus on the newly formed silicate clay minerals. The net result may be a slow change in sesquioxide rich clayey tropical soils such as Nitisols and rhodic Ferralsols with humic topsoils (rnollic or umbric A horizons) into units with the stable humus distributed more evenly in the whole soil profile. A more prominent presence of organic acids by decomposition may favour organo-metallic complexing. This will particularly be the case where nutrient supply cannot keep pace with the increase in biomass production and a poorer type of organic matter (with a higher C/N ratio) is produced. The result may be a degree of podzolization where the soil is sandy and of relatively low sesquioxide content (sandy xanthic Ferralsols; see above). All above considerations refer to the (sub)humid tropics. In the semi-arid tropical and sub-tropical uplands the weathering potential, its rate and direction, depend very much on the seasonal and annual changes in soil moisture availability - for which the GCMs, as stipulated above, give unreliable predictions. Higher surface temperatures in open savannah like conditions (predominantly Cq grasses hence less increase in biomass production) would lead to faster combustion of soil humus already present, while soil fauna such as termites would carry more litter to their cool underground nests (which may lead to more methane emissions as a positive feedback to the Arrhenius effect). Whcre free salts are available at some depth in the subsoil these may be transported upwards due to stronger evaporation and evapotranspiration, leading to a degree of soil salinization. The likely combined result of a greenhouse warming on the semiarid tropics is a strengthened hazard of desertification. The statements given apply to whole continents or regions. The local pattern of soils in the tropics, often a reflection of former climatic conditions different from the present ones, coupled with the local implications of any anthropogenic climatic change in the near future, offer a large and intricate scala of possibilities of future soil development. This cannot be discussed in the framework of this general review on tropical soils on a warmer earth. The modeling of soil genetic

Soils on a warmer earlh: lropical regions

171

processes as proposed by Stewart in Chapter 10 may, however, be put to good use for individual soil profiles and their present day or surmised future environmental conditions.

EFFECTS ON SOIL CONDITIONS OF TROPICAL COASTAL LOWLANDS AND OF TROPICAL MOUNTAIN AREAS The tropical coastal lowlands and wetlands, as well as the tropical mountain areas (roughly all areas above 2000 m altitude), merit special attention. Changes are likely to be substantially more dramatic than in tropical uplands. The coastal wetlands will be affected principally by the expected rise of sea level of 0.5 to 1 m. Brackish water will penetrate in many coastal marsh lands and swamps. The brackish water, in combination with the high organic matter accumulation of these lands, will result in soil forming processes favouring the formation of potential Acid Sulphate Soils (thionic Fluvisols, - Gleysols, and Histosols). These are real problem soils, because they turn very compact and extremely acid upon artificial drainage, becoming unsuitable for most plant growth. Coastal lands at present only a few meters above flooding would have shallower levels of ground water, often saline, resulting in salinization and/or sodification of soil profiles especially where the climate is subhumid or semiarid. For details, also as regards low level islands with coral reefs such as prevalent in the Pacific, see Brammer and Brinkman in Chapter 12. Tropical mountain areas, as discussed before, will see a substantial shift of natural or agroecological zones in an upward direction, because of the predicted 2°C temperature increase. The intensity of rock weathering will increase and soils will become deeper. This implies more agricultural potential where the adjoining uplands remain humid, such as is the case on the Eastern slopes of the Andean Cordillera and the Northeastern slopes of the Sumatran mountain ridge. If however, the associated uplands become dryer, as may be the case in Eastern Africa, then the net agricultural area in the region may diminish. Whatever be the case, the surface and subsurface hydrology on the mountain slopes will change substantially, with a potentially strong effect on the drainage conditions of the lands and soils of the lower parts of the catchment areas concerned. The details of soil changes in mountain areas depend not only on differences in parent materials, but also on past shifts, upwards and downwards, of the belts of climate conditions and associated plant communities. Systematic and detailed paleoecological studies, such as those already carried out on several altitudinal transects in the Colombian Andes (Van der Hammen et al. 1983, 1984 and 1989) will provide important clues as to the future soil development in tropical mountain regions under anthropogenic climatic change.

172

W . G . Sombroek

ACKNOWLEDGEMENTS The author is indebted to Dr. Pieter van Reeuwijk, soil chemist of ISRIC, Wageningen, for his contribution and constructive comments.

REFERENCES Ackland, J.D. (1971). East African Crops. F A 0 - Longman. Allen, L.H., K.J. Boote, P.H. Jones et al. (1985). Response of vegetation to rising C02: Soybean Photosynthesis and Yield. In: Direct effects of increasing carbon dioxide on vegetation. DOE/ER-0238 Report p.175. US Dept. of Energy. Anderson, G.C. (1990). Climate Change; towards a 2°C consensus. Nature 343:401. Arnold. R.A., I. Szabolcs and V. Targulian (1990). Concept of Global Soil Science. Report of 1989 rIASA/ISSS/UNEP task force meetings. IIASA, Vienna (in press). Bolin, B., B.R. DOOs. J. Jager and R.A. Warrick (eds)(1986). The greenhouse effect, Climatic Change and Ecosystems. SCOPE 29. John Wiley & Sons, Chichester, 541p. Camargo, M.N. et al. (1981). Mapa de solos do Brasil, escala 1:5.000.000. EMBRAPASNLCS, Rio de Janeiro. Cure, J.D. (1988). Carbon dioxide doubling responses. In: The use of statistical climate-crop models for simulating yield to project the impacts of C 0 2 induced climate change. DOE/ER/60444-H1, July 1988, p.31. Dickinson, R.E. (1986). How will climate change? The climate system and modelling of future climate. In: Bolin et al. op.cit., 207-270. Dubroeucq, D. and B. Volkoff (1988). Evolution des couvertures pddologiques sableuses i podzols giants d'Amazonie (Bassin du Haut Rio Negro). Cah. ORSTOM, Strie P6dologie 24(3):191-214. Emanuel, W.R., H.H. Shugart and M.P. Stevenson (1985). Climatic change and the broad scale distribution of terrestrial ecosystem complexes. Climatic Change 7:29-43. Esser, G. (1990). Modeling global terrestrial sources and sinks of CO, with special reference to soil organic matter. In: A.F. Bouwman (Ed), Soils and the greenhouse effect, pp. 247-262, Wiley and Sons, Chichester (1990). F A 0 (1988). FAO/Unesco soil map of the world, Revised legend. World soil resources report 60, FAO, Rome. Goudriaan, J., and M.H. Unsworth (1988). Implications of increasing carbon dioxide and climatic change for agricultural production and water resources. Proceedings annual ASAE meeting, Los Angeles, USA (in press). GRID (1987). Uganda Case Study: A sampler atlas of environmental resource data sets within GRID. UNEP-GEMS, Nairobi. Handbook Chemistry and Physics (1987). 68th ed., CRC Press. Hansen, J., D. Johnson, A. Lacis, S. Lebedeff, P. Lee, D. Rind and G. Russell (1981). Climatic impact of increasing atmospheric carbon dioxide. Science 213:957-966. Hansen. J . and S. Lededeff (1987). Global trends of measured surface in temperature. J. Geophys.Res. 92:13345-13372, 1987. Hansen, J., I. Fung. A. Lacis, D. Rind, S. Lededeff, R. Ruedy and G. Russell (1988). Global climate as forecast by Goddard Institute for Space Studies; three dimensional model. J. Geophys. Res. 93:9341-9364. Holland, H.D. (1978). The chemistry of the atmosphere and oceans. Wiley & Sons, New York. p. 56-63.

Soils on a warmer earth: tropical regions

173

Hulme, M., and P.D. Jones (1988). Climatic change scenarios for the UK. (in press). Idso, J.B., B.A. Kimball, M.G. Anderson and J.R. Mauny (1987). Effects of atmospheric CO2 enrichment on plant growth: the interactive role of air temperature. Agric. Ecosys. and Environ. 20:1-10. Karl, T.R., H.F. Diaz and G.Kukla (1988). Urbanization: its detection and effect in the United States. Climate Record. J. Climate 1:1099-1123. Kimball, B.A. (1983). Carbon dioxide and agricultural yield: An assemblage and analysis of 430 prior observations. Agronomy Journal 75:79-88. Loughnan, F.C. (1969). Chemical weathering of the silicate minerals. American Elsevier Publ.Co. New York. Lucas Y., R. Boulaine, A. Chauvel and L. Veillon (1987). Systbmes sols ferralitiques-podzols en rtgion amazonienne. In: Podzols et podzolisation. D. Righi and A. Chauvel (Eds). AFES-INRA, Paris. 53-65. Manabe, S. and R.T. Wetherald (1986). Reduction in summer soil wetness induced by an increase in atmospheric carbondioxide. Science 232:626-628. Meier, M.F. (1990). Reduced rise in sea-level. Nature 343:115-116. Mitchell, J.F.B. (1989). The "Greenhouse" effect and climate change. Review of Geophysics 27(1): 115-139. Mooney, H.A., P.M. Vitousek and P.A. Matson (1987). Exchange of materials between terrestrial ecosystems and the atmosphere. Science 238:926-932. Novozamski I. and J. Beek (1976). Common solubility equilibria in soils. In: Bolt, G.H. and M.G.M. Bruggenwerth (Eds). Soil chemistry. a basic element. Developments in soil science 5A. Elsevier Amsterdam. p.69-125. Oechel, W.C. and B.R. Strain (1985). Native species responses to increased atmospheric carbon dioxide concentration. In: Direct effects of increasing carbon dioxide on vegetation. DOE/ER/0238 Report, p. 117.154. US Dept. of Energy Pirazzoli, P.A. (1989). Present and near future global sea-level changes. Global and planetary change 1(4), Elsevier Science Publishers, Amsterdam. Press, E. (1989). What I would advise a head of state about global change. Earthquest 3(2):13. UCAR, Bouldcr, U.S.A. Schleser, G and W. Kirstein (1989). Der Treibhaus Effect - Ursachen und Consequenzen fur Klima und Biosph5re. Seminar Technik und Gesellschaft fiir Journalisten, K.F.A Jiilich, 26-4- 1989. Schlesinger M.E., and J.F.B. Mitchell (1985). Model projections of the equilibrium climatic response to increased carbon dioxide. In: The potential climatic effort of increasing carbon dioxide, J.F.B. Mitchell, U.S. Dept. of Energy. Sombroek, W.G. (1966). Amazon Soils. A reconnaissance of the soils of the Brazilian Amazon, 292p. Pudoc, Wageningen. Sombroek, W.G. (1984). Soils of the Amazon Region. In: Sioli H., The Amazon, limnology and landscape ecology of a mighty tropical river and its basin. Dr.W. Junk Publishers. Dordrecht/Boston. p. 521 -536. Sombroek, W.G., H.M.H. Braun and B.J.H. van der Pouw (1982). Exploratory soil map and agroclimatic zone map of Kenya 1980. Exploratory soil survey report E l . Kenya Soil Survey, Nairobi. Sombroek, W.G., and I S . Zonneveld (1971). Ancient dune fields and fluviatile deposits in the Rima-Sokoto river basin (N.W. Nigeria). Stiboka, Soil Survey Papers 5. Wageningen. p.109. Tans, P.P., I.Y. Fung and T. Takahashi (1990). Observationed constraints on the global carbon dioxide budget. Accepted by Science. Turenne, P. (1975). Mode dhumification et de diffkrentiation podzolique dam deux stquences

174

W.G. Sombroek

guyanaises. Mem. ORSTOM 84. Paris. pp.173. Van der Hammen. Th., A.P. Preciado and P. Pento (1983). La Cordillera Central Colombiana, transecto parque Los Nevados (Introduccion y datos iniciales). Studies on tropical Andean ecosystems Vol.1. A.R. Gantner Verlag K.G., Berlin-Stuttgart. Van der Hammen, Th., and P.M. Ruiz (1984). La Cordillera Central Colombiana, transecto parque Los Nevados (Introduccion y datos iniciales). Studies on tropical Andean ecosystems Vol.11. A.R. Gantner Verlag K.G.. Berlin-Stuttgart. Van der Hammen, Th., S. Diza-Piedrahita and V.J. Alvarez (1989). La Cordillera Central Colombiana, transecto parque Los Nevados (Introduccion y datos iniciales). Studies on tropical Andean ecosystems Vol.III. A.R. Gantner Verlag K.G., Berlin-Stuttgart. Walker, R.H., and R.D. Graetz (1989). Effects of atmosphere and climate change on terrestrial ecosystems. Global Change Report no 5. Stockholm. Washington, W.M., and G.A. Meehl (1984). Seasonal cycle experiment on the climate sensitivity due to a doubling of C 0 2 with an atmospheric general circulation model coupled to a simple mixed-layer ocean model. J. Geophys.Res. 89:9475-9505. Zivkovic, D.Z..J.J. Crnko and G. Sretenovic (1983). Kinetics and mechanism of the acid kaolin leaching process. In: Augusthitis, S.S. (ed.), Leaching and diffusion in rocks and their weathering products. Theophrastus Publications. Athens, Greece. p.63-77.

175

Chapter 14

SOILS ON A WARMER EARTH Projecting the effect of increased CO, and gaseous emissions on soils in Mediterranean and subtropical regions

Dan H. Yaalon Institute of Earth Sciences, Hebrew University. Givat Ram Campus Jerusalem 91904, Israel

ABSTRACT The subtropical region is defined here as roughly the zone of 10' to 35" N and S of the equator, with a pronounced seasonal rainfall, with about 2 to 4.5 months in which precipitation exceeds evapotranspiration by 75%. Rain falls mainly in summer in contrast to the Mediterranean Region with a winter rainfall season. The area covered by the subtropics is around 18 million km2, mainly in Africa South of Sahara and smaller regions in Asia (India), Australia and South and North America. The area of the Mediterranean region cover close to 9 million km2 mostly in Southern Europe, North Africa and the Near East. General circulation models for the condition of a double C02 concentration are not consistent in predicting the moisture conditions. Some models predict a decrease in precipitation and soil moisture, especially in the summer months, which would result in a considerably change in the moisture regime both on cotenary slopes and in individual pedons. In general these are very fragile environments, sensitive to climatic change, partly because of the high variability and seasonality of the climatic parameters. The major soil groups in the two regions are Luvisols, Acrisols, Vertisols, Cambisols, Arenosols and Fluvisols. Six major soil responses can become evident within a short time of less than 50 years: 1) organic matter reduction; 2) carbonate and salt regime changes; 3) erosion; 4) crusting; 5 ) waterlogging; and 6) fire frequency. Among these the reduction of organic matter due to increased temperature should be most easily modelled and is of special importance because of a positive feed back effect on additional temperature increase. The soil carbonate regime and salt balance may be affected by both decreasing and increasing precipitation. Salinization on footslopes may be caused by increased rainfall upslope. Changes in intensity and seasonality of rainfall will result in increased runoff and erosion, in other cases surface crusting. Reduced infiltration in clayey soils may cause waterlogging downslope. Prolonged drought periods may cause an increased fire

176

D.H . Yaalon

frequency and result in subsequent severe erosion. The response action will depend not only on more precise prediction of the climatic parameters but also on the nature of the soils, their texture, structure, depth and spatial variability. Much detailed research on soil distribution and sensitivity to changes is needed. The basic soil processes are well known and understood, but their actual rates and possible feedbacks are known to a much lesser extent.

Note from the editors: Unfortunately Dr. Yaalon could not deliver the final paper in time due to unexpected hospitalization. The reader may wish to contact the author directly for more detailed information.

177

Chapter 15 IMPACT OF CLIMATE WARMING ON ARID REGION SOILS H.E. Dregne Texas Tech University, Lubbock Texas 79409. USA

INTRODUCTION Global atmospheric warming is generally expected to occur as a consequence of the increases in carbon dioxide and other greenhouse gases such as nitrous oxide, methane, chlorofluorocarbons, and tropospheric ozone. However, the imprecision of general circulation models, the uncertainty of the effect warming will have on ocean evaporation and global precipitation, and doubts about the validity of numerous assumptions that must be made, have raised questions about what is going to happen and where it will happen. Nevertheless, it is prudent to consider the impact of climate warming on natural resources, of which soil is an essential component. For the purpose of this discussion, the maximum global temperature increase is set at 3°C by 2050 and the mean annual precipitation change is set at plus or minus 10% of the current mean. The climate change is expected to be least in equatorial zones and greatest at high latitudes. Arid regions are commonly divided into three climatic zones: hyperarid, arid, and semiarid (Fig 15.1). Together, the three zones cover about one-third of the global land surface. Not shown on the map are the dry regions of the Arctic and Antarctic. In the latter regions, surface soils generally have properties similar to those of surface soils in the hot arid regions, but subsoils differ markedly. Changes in properties of soils in the hyperarid regions, due to climate warming, will not be discussed since any changes are expected to be insignificant due to the very low rainfall and near absence of natural vegetation. Just as profile development increases, when time is held constant, as one goes from the dry end of the arid climatic zone to the end of the semiarid climatic zone, so would the impact of climate change be expected to vary with mean annual rainfall as well as with temperature.

SOILS And region soils belong to six of the 11 orders of the U.S. Soil Taxonomy

178

2

I i 3 'tI PJ? . A

H . E . Dregne

2 0

3

2 ,

e

K

Fig. 15.1

Distribution of arid region climates

Impact of climale warming on arid region soils

179

(Aridisols, Alfisols, Entisols, Inceptisols, Mollisols, and Vertisols) and 14 of the 26 orders of the F A 0 classification system (Arenosols, Cambisols, Chemozems, Fluvisols, Kastanozems, Lithosols, Luvisols, Regosols, Solonchaks, Solonetz, Vertisols, Xerosols, and Yermosols). Most arid region soils are calcareous somewhere within the top 2 meter of soil; alkaline, neutral or slightly acid in the surface horizon; coarse textured sandy or gravelly on upland sites, sometimes covered with a desert pavement of gravels and stones (gobi, reg); fine texture in closed basins that frequently are saline as in the chotts and sebkhas of North Africa and the Middle East; variably textured in stream flood plains; and usually are only weakly developed morphologically. Within the dry regions large areas occur of gypsiferous soils, such as in Iraq, the Soviet Union, Syria and the United States. The most notable exception to the generality about arid region soils having calcareous horizons and an alkaline or slightly acid pH occurs in the semiarid African Sahel. There, where upland soils are permeable and rain is concentrated in a short period during the hot season, soils tend to be strongly acid, particularly in the subsoil. Similar conditions are said to exist in the Eastern part of Rajasthan State in India. One characteristic arid region soils have in common is a low level of organic matter. Organic matter content varies with rainfall and vegetative cover as well as with texture. Clay soils tend to be considerably higher in organic matter (humus) than sandy soils in the same climatic zone and organic matter usually is higher when the vegetative cover is relatively heavy. This means that upland sandy soils in the hot arid climatic zones are very low in organic carbon, typically containing less than 0.4%, whereas upland clay soils in the semiarid zones commonly have more than 1 % organic carbon and may contain 2% or more.

MEASURING SOIL CHANGES The principal problems with demonstrating that atmospheric temperature increases alter soil properties are sampling errors and difficulties in measuring soil erosion. Sampling errors in rangeland soils poses a special problem due to differences in density of vegetative cover during the year and the kind of vegetation present. Sampling a uniform cover of annual or perennial grasses during the growing season is quite different from sampling hummocky land or soils under shrubs and trees. Analytical methods are at least two or three orders of magnitude more accurate than sample collection methods. Independent sampling of one hectare for soil organic matter, for example, would not be expected to give results that are reproducible to within 20%. Salinity measurements are even harder to reproduce. In the case of water erosion, fairly elaborate and costly experimental plots must be constructed to provide reasonable accurate measures of soil loss from a

180

H.E. Dregne

small area. Determining water erosion on large fields with a good degree of accuracy is nearly impossible. Gully erosion commonly is estimated by measuring gully length as shown on aerial photographs. No one has yet devised a satisfactory technique for measuring removal and deposition of wind eroded soil on a field scale.

DATA COLLECI'ION METHODOLOGY A logical way to estimate the effect a 3°C rise in air temperature would have on soils would be to compare two areas where the physiography, parent materials, and rainfall were similar but there was a 3°C difference in the atmospheric temperature. Many such areas can be found in upland plains. One of them is the Texas High Plains, where the difference in mean annual temperature between Big Spring and Amarillo is 33°C. The two cities are approximately 400 km apart. Mean annual precipitation is 450 mm in Big Spring and 485 mm in Amarillo, the end of the semiarid climatic zone. If a 3°C increase in air temperature has readily detectable effects on soils, they should be observable here. A thick, fairly uniform eolian mantle covers most of the Texas High Plains. Immediately below the mantle is the Ogallala Formation, of Pliocene age. The Ogallala consists of calcareous fluviatile material composed of sand and gravel. The High Plains between Big Spring and Amarillo give the appearance of being absolutely flat. They are broken only by numerous closed basins known as playas. Soils are so similar that several belong to the same series and textural class. An extensive group of soils consists of Paleustalfs (Luvisols) of the Amarillo soil series. A comparison of those soils would seem to be ideal for determining the effect a 3°C temperature change would have on soil properties. Unfortunately, virtually all of the land in this part of Texas is cultivated. Half is rainfed and half is irrigated. Although the principal crops are only two, cotton and grain sorghum, many different management systems are practiced. Some farmers rotate their crops, some fertilize, some employ minimum tillage systems, some use terraces, some do not plant on the contour, some apply cotton gin trash (organic waste) to their fields, a few apply feedlot manure, etc. With all that management variability, it is almost impossible to find fields where everything is nearly the same except temperature. A critical observer could find several plausible reasons why soil organic matter differences, for example, were due to something other than higher or lower temperatures. Unger and Pringle ( 1 98 1 ) made intensive comparisons of the variability in one soil, Pullman clay loam, a Paleustalf, in the vicinity of Amarillo, Texas. Among the five samples from adjoining countries, organic matter content in the cultivated surface horizon ranged from 1.4% to 2.2%. Comparable differences were measured in the subsoils.

Impacl of climate warming

OR

arid region soils

181

Similar problems in lack of uniformity in land management can be expected in most semiarid regions, whether they are cropped or grazed. If topography, exposure, erosion, and other factors also vary, field comparisons may be of little value in arriving at definitive conclusions about climate change impacts until after many decades have passed. Small experimental plots will have to be monitored.

SOIL CHANGES AND LAND USE Approximately 90%of the arid and semiarid climatic zones of the world are uscd for grazing. Only about 6% arc utilized for growing rainfed crops. The remainder is used for irrigated agriculture, urban and rural settlements, and a number of other activities (Dregne 1983). Climate changes should have little effect on imgated soils. A slight effect on hastening organic matter decomposition may occur. It should not significantly alter soil properties, so those soils will not be discussed. According to Emanuel et al. (1985), a doubling of atmospheric carbon dioxide leads to increasing the area of global deserts from 20.6% to 23.8% of the land surface and a large increase in grasslands from 17.7% of the land to 28.9%. The estimates of changed area are based upon temperature increases that one of the general circulation models predicts from the doubling of carbon dioxide. Temperature increases predicted from the model are about 1.25"C at the equator and 3.75"C or more near the poles. In the Holdridge Life-Zone maps employed in estimating the desert and grassland area changes, a desert corresponds roughly to the arid climatic zone and the grasslands are similar to semiarid lands. In the Holdridge classification, desert precipitation is less than 250 mm and grassland precipitation is between 250 and 500 mm. The increase in grasslands would be at the expense of forcsts. The accuracy of that prediction is highly questionable.

Grazing lands Pastoralism is by far the dominant land use in the dry regions. It is practically the sole land use in the arid climatic zone and is dominant in the dry part of the semiarid zone, which it shares with rainfed cropping (dryland farming). Plant cover consists of a mixture of annual grasses, perennial grasses, forbs, shrubs, and trees. An estimated 80% of the world rangelands are at least moderately overgrazed (Dregne 1983). Of that 80%, threefifths is moderately degraded and two-fifths is severely degraded. Along with a reduction in perennial grasses by overgrazing may go a slight reduction in annual grasses, a large reduction in tree cover, an increase in unpalatable shrubs, and an increase in bare land. While total green biomass production may not decrease, the organic matter content of the soil very likely would decrease due to the lower production of perennial grasses and the increased bare space.

182

H.E. Dregne

An atmospheric temperature change of 3°C without a change in rainfall probably would have effects similar to overgrazing. Biomass production would be likely to decline significantly over the area as a whole, as the combined result of overgrazing and increased temperatures. Soil organic matter levels would also drop considerably on a percentage basis. The organic matter decline should occur nearly everywhere grazing is practiced. The decline would be difficult to detect in a few dccades because of the wide natural variability in daily and seasonal temperatures in the dry regions. A simultaneous 10%increase in mean annual precipitation would change things very little because the higher temperature would increase organic matter decomposition rates. A 10% decline would aggravate the organic matter loss. Runoff would increase, bringing more water erosion, probably in the form of gullies, on most soils as the vegetative cover at the soil surface became less. Wind erosion also would increase on upland sandy soils. Salinity should increase, as- well, although the effect on the salt content of chotts, sebkhas, salinas, and salares would be too small to detect against the large amount already present. Surface soil crusting should increase as the result of lower soil organic matter and more rapid surface soil drying. All of these effects would be minimal in the drier part of the arid climatic zone and greatest in the wetter grazing lands. This probably means that any soil changes would be undetectable in the drier arid regions as air temperature gradually became 3°C higher by the year 2050. In the semiarid rangelands, detectable declines in soil organic matter may occur. Increase in gully formation and wind erosion in drought years are likely. However, validating the increases would be difficult, given the normal high temperatures and dry conditions during droughts. Salinity changes and increases in surface crusting would be undetectablc. Rainfed cropland

Global warming, in the absence of any change in mean annual precipitation, would be expected to increase evaporation and reduce soil moisture (Manabe and Wetherald 1986), according to one scenario. That scenario also predicts enhancement of winter soil moisture. Changes would be greatest in certain regions in the middle and high latitudes. These regions include the Great Plains of Canada and the United States, Western Europe, Central Asia, and North Africa. Rainfed croplands commonly have lower organic matter contents than their uncultivated counterparts. The reason is less surface insulation by plant cover, summer fallow, harvesting of crops, little or no addition of fertilizers or manures, and the absence of a soil building crop in a rotation. A temperature rise in the cool or cold climatic zones in areas such as the Northern Great Plains, however, should increase production of cultivated crops and of native plants in rangelands

Impact of climate warming on arid region soils

183

(Williams et al. 1987). Increased plant production on rangelands should increase soil organic matter in these semiarid climatic zones. There probably would be a less pronounced increase, if any, in the rainfed croplands because organic matter decomposition rates would also increase. Of the other soil properties, dryland salinity may decrease slightly, the pH would change very little, and there probably would be no measurable change in base saturation and in calcium carbonate content or the depth at which the calcic horizon appears. There should be a slight decrease in surface soil crusting on both croplands and rangelands. Wind erosion would be substantially worse. Water erosion may increase slightly. In the hot semiarid regions, temperature rises will reduce soil organic matter levels, whether or not the rainfall changes upward or downward by 10%. There would be more surface crusting but no significant change in other characteristics. Polar arid regions Climatic conditions on the Antarctic continent probably are the most severe of any arid land area in the world. They are more severe than conditions in the Arctic, where most of the area is ocean. A major difference between Antarctic and Arctic soils is the virtual absence of organic matter in Antarctic soils. In fact, Tedrow and Ugolini (1966) note that, by common soil definitions which emphasize the biotic element, there are no soils in Antarctica. Other than the absence of organic matter, the soils of the dry (ice-free) valleys resemble temperate and region soils in the upper part of the profile. They are alkaline, are well supplied with basic elements, have very low amounts of nitrogen and phosphorus, generally are saline, and contain free calcium carbonate. A global 3°C rise in air temperature could increase the organic matter level by some small amount, in contrast to the decrease expected in temperate zone soils. In the warmer Arctic, soils commonly are alkaline, have a high base saturation, contain free calcium carbonate, are sometimes saline, and are very low in organic matter content (Tedrow 1974). They are underlain by permafrost at about 1 m. Global warming, which is supposed to be greater at the high latitudes than in the tropics, should increase plant growth and soil organic matter levels by a significant amount. Soils probably would have a somewhat lower salinity, base saturation, and calcium carbonate content. The soils would be thawed to a greater depth during the summer. Williams et al. (1987) estimated from one of the general circulation models (GISS) that a doubling of carbon dioxide would raise mean annual air temperature in Southern Saskatchewan, Canada (50"N latitude) by 4.7"C, a large amount. Presumably, the rise would be even greater in the Arctic.

184

H.E. Dregne

CONCLUSIONS In the tropical and temperate and climatic zone, where mean annual precipitation is less than about 250 mm and grazing is the dominant land use, a global warming of 3°C by the year 2050 should cause a modest decline in soil organic matter, whether precipitation increases or decreases by 10%. There should be a slight to moderate increase in surface soil crusting and, perhaps, a slight increase in salinity. Water erosion may increase somewhat but wind erosion should increase significantly. Conditions in the polar and regions will be different in the Arctic than in the Antarctic. In the latter region, warming may bring a slight increase in soil organic matter. By contrast, there should be a sharp increase in organic matter in Arctic soils, as well as a lowering of the permafrost layer. Semiarid grazing lands should experience a decrease in organic matter. Hot rainfed croplands will show less soil organic matter and no change in salinity or other properties. Cold rainfed croplands will have a modest increase in organic matter but no change in other properties. Wind erosion will increase more in the cold regions than in the hot regions. Water erosion may not change.

REFERENCES Dregne, H.E. (1983). Desertification of arid lands. Harwood Academic Publishers. New York. 242 p. Emanuel, R., H.H. Shugart, and M.P. Stevenson (1985). Climatic change and the broad-scale distribution of terrestrial ecosystem complexes. In Martin L. Parry (Ed.), The sensitivity of natural ecosystems and agriculture to climatic change, United Nations Environment Programme reprint, Nairobi, Kenya, p. 29-43. Manabe, S.. and R.T. Wetherald (1986). Reduction in summer soil wetness induced by increase in atmospheric carbon dioxide. Science 232626-628. Tedrow, J.C.F., and F.C. Ugolini (1966). Antarctic soils. In: J.C.F.Tedrow (Ed.), Antarctic soils and soil forming processes, American Geophysical Union, Washington, D.C., p. 161-177. Tedrow, J.C.F. (1974). Soils of the high arctic landscapes. In: T.L. Smiley and J.H. Zumberge (Eds.), Polar deserts and modern man, The University of Arizona Press. Tucson, Arizona, p. 63-69. Unger, P. W., and B. Pringle (1981). Pullman soils: Distribution, importance, variability, and management. Texas Agricultural Experiment Station B-1372, College Station, Texas. 23 p. Willliams, G.D.V., R.A. Fautley, K.H. Jones, R.B. Stewart, and E.E. Wheaton (1987). Estimating effects of climate change on agriculture in Saskatchewan, Canada. International Institute for Applied Systems Analysis, Laxenburg, Austria. 147 p.

185

Chapter 16

SOILS OF THE SUBBOREAL REGION ON A WARMER EARTH Boris G. Rozanov and Elena M . Samoilova Faculty of Soil Sciences, Moscow State University, 117234 Moscow, USSR

GENERAL ASPECTS AND ASSUMPTIONS Generally, four factors could be responsible for major changes in soils within their natural zones: 1 ) climate change; 2) geomorphological development of land surface; 3) human impact; and 4) self evolution of natural ecosystems. The main difficulty in the analysis of ongoing soil changes is presented by the fact that all above four factors, or driving forces of soil evolution, act simultaneously, and their specific impacts and consequences are almost impossible to distinguish and separate. However, the future trends can be more or less reliably predicted on the basis of the analysis of past and present soil forming processes, which were reflected or are being expressed nowadays in certain measurable or observable soil features. This is based on the assumption that soil is a natural body that preserves the results of soil forming processes in its specific properties or attributes, which could bc correlated with each other on the basis of h e cstablishcd soil science inherent laws and regularities. Among the factors of soil formation and evolution climate plays a very important role in determining a general soil geography of the world, the global pattern of soil zonality. Broadly speaking, there is a very close relationship bctwecn the type of climate and the type of major zonal soil. Naturally, certain geographic deviations from the general law of soil zonality do occur due to local variations in soil age, topography, parent materials, ground waters, etc., but when we are considering broad natural soil zones their predetermination remains indisputable. That is why, analyzing soil changes of major importance, we pay first attention to probable climate changes at a global or regional scale. Within the soil zones two different although interrelated aspects should be considered: 1) changes of soils composing the zonal soil cover; and 2) changes of the structure of zonal soil cover, e.g. ratio of different soil types in its spectrum. As the preliminary analysis shows, both aspects are of equal importance.

THE SUBBOREAL REGION Scvcral climate-related soil zones are distinguished within the broad

186

B.G. Rozanov and E . M . Samoilova

subboreal region of the world, starting from the coldest and the most humid: zone of deciduous or mixed forests with Cambisols and Luvisols; - zone of forest-steppes with Greyzems, Phaeozems, and Chemozems; - zone of steppes (prairies) with Chernozems and Phaeozems; - zone of dry steppes with Kastanozems and Solonetz; - zone of semideserts with Calcisols, Gypsisols, and Solonchaks; - zone of deserts with Arenosols, Regosols, Solonchaks, and nonsoil formations (desert detritus). All of the above subboreal soil zones have a rather complicated but specific structure of the soil cover, related, first of all, to specific topography determined redistribution of surface and underground watering with certain modifications due to the diversity of parent materials, including their calcareousness and salinity. Taking into account the great natural soil diversity of the subboreal region, which is considerably magnified by the diversity of past and present land use, we do not expect any uniform or one way directed soil change of the region under the impact of more or less even climatic change at a global scale, e.g. a global warming of 2-4°C. On the contrary, each of the soil zones, and each of the components of the structure of their soil cover, may experience their own soil changes in full conformity with their specific characteristics. At present, bearing in mind the magnitude of the task and the availability of the data required for its solution, we have to limit ourselves to some examples for a detailed analysis, leaving the region as a whole for future considerations and drawing only certain very general conclusions, if any, at this stage.

ZONES OF STEPPES AND PRAIRIES Steppes with Chemozems and Phaeozems as well as dry steppes with Kastanozems and associated soils might be selected as good examples, being most intensively studied and occupying the central and most agriculturally utilized part of the subboreal region.This is a broad latitudinal belt stretching through loess-covered plains of North America and Eurasia a the total area of Chemozems and Phaeozems of ca 2.60 million km2 and of Kastanozems of 2.62 million km2. Certain areas of Chemozems and Kastanozems are scattered in the surrounding mountains at appropriate altitudes. The present climate of this belt was gradually formed during the Holocene and is characterized as temperate, from subhumid to semiarid in different provinces, mostly continental, with annual precipitation ranging from 300 to 700 mm, and soil temperature regimes ranging from long and deep winter freezing (e.g. Eastern Siberia) to non freezing (e.g. Danube Lowland). The topography is mostly level or slightly undulating. The terrain is usually well drained due to a more or less dense network of gullies, balkas and river valleys. The parent materials are mostly represented by calcareous loesses or loess like sediments in periglacial uplands, while on riverine or marine old terraces they are sometimes

Soils on a warmer earth: subboreal regiom

187

saline, in the entire profile or in buried layers, with soda and sodium sulfate dominating among the soluble salts. The original vegetation was represented by grass associations of varying physiognomy and composition representative for various steppes or prairies in most humid and arid areas. At present the area is almost entirely under cultivation with patches of grazing land among the arable fields, roads, shelterbelts, and settlements. The age of Chernozems is estimated to be 8 to 10 thousand years, or of Holocene age.

ZONAL SOIL CHANGE VERSUS CLIMATE CHANGE Soil is a rather conservative natural system experiencing very slow and gradual changes under the influence of changing factors of soil formation or in the course of self development if not subjected to destructive external influences. Only certain soil features, such as water and temperature regimes and related to them salt, air, redox and similar regimes are more or less quickly changing under the impact of a changing environment. Furthermore, relatively labile soil properties have rather high spatial and seasonal variability, due to which it is very difficult to observe any specific trend in soil change in a changing environment. Another point which has to be taken into consideration, is the fact that during the last hundred years, when gradual climate warming has occurred with some fluctuations and delays in the Northern hemisphere, the soils of steppes and prairies have experienced very strong direct impact of industrializing agriculture: development of large new areas for cultivation, growing pressure of agricultural machinery, increased use of chemicals and, as a consequence, an increase of soil (de)structuring, (de)humification, (de)acidification, alkalinization, water and wind erosion. All above factors have considerably masked possible climate induced soil changes, if any. Thus, by the end of the 1930s soil scientists did not notice any change in soils of the region connected with the climate warming which started in this area at the end of the last century. Either these changes were negligible in comparison with those induced by human impacts, or they remained unnoticed because of high spatial and seasonal variability or because they did not come out of the criteria frameworks established for the lowest taxonomic units, on which the changes might be fixed. Assuming that the climate of steppes and prairies will be similar to that of the Holocene optimum 5-6 thousand years ago (about 1°C wanner than at present as an average) by the year 2000, of the Mikulinsk (Sangamonian in the USA) Interglacial 125 thousand years ago (about 2°C wanner than present) by the year 2025, and of the Pliocene 1 million years ago (3-4°C warmer than present) by the year 2050, it does not automatically mean that the soil cover will correspond adequately to the climatic environments of those periods. Studying soils of the corresponding periods by means of paleopedology, we are dealing with the developed soils which were formed in wanner climates during the millennia. While at present, expected climatic changes

188

B.G. Rozanov and E . M . Samoilova

due to atmospheric accumulation of COz and trace gases are going to occur very quickly, within 20-50-75 years. As a result, soils will not come into equilibrium with their new climatic environment. Only certain labile soil features will change, but these changes will hardly be sufficient for any marked shift of the boundaries of soil zones. There will be a change in water-salt regime, in the ratio between humus accumulation and mineralization, but the main soil compositional, constitutional and morphological attributes will remain unchanged. This conclusion is particularly supported by the fact that under the expected climate warming the water regime will not be changing uniformly throughout &he region. As the comparison with past climate changes shows, the warming of global temperature by 1°C will cause aridification in a considerable part of the region, while the consequent rise of the temperature by 2-3°C will be accompanied by a reverse process of humidification with an increase of precipitation.

POSSIBLE SOIL CHANGE IN THE CHERNOZEMIC ZONE If we assume a global warming of 1"C by the year 2000 in comparison with the present mid century conditions, the following soil changes within the zone might be expected. It is possible to expect certain favorable changes in steppe and forest-steppe zones of East Asia. The climatic conditions of the most continental regions will become more comparable to those of the regions with a more soft climate. In the steppes of East Siberia the amount of precipitation may increase from 200-300 mm up to 250-400 mm and thus will be similar to that of the Pre-Altai province. Certain similarity between these provinces will also occur in the temperature regime. As a consequence, the soil productivity in East Siberia will increase, may be by 2-3 times. The expected shift could be rather substantial. If the analogy with the climatic optimum of Holocene is correct, certain aridification of the climate is to be expected after a 1°C temperature increase in the chernozemic steppes and prairies of Eastern Europe, Western Siberia, Kazakhstan and Central United States. The general direction of soil evolution in these regions will be from Phaeozems to Chemozems and from Chernozems to Kastanozems, more pronounced in the warmest parts of the zone and in its the least continental parts, particularly in Central Europe, Lower Danube Lowland, Southern Ukraine, and Pre-Caucasia. It is possible to expect some weakening of humificalion with a corresponding decrease of soil humus content and thickness of humus horizons, increase of alkalinization, decrease in area of the most productive meadow-chernozemic and meadow soils with high ground waters. Water supply for agricultural crops will go down followed by a decrease in productivity, possibly by an order of two times. However, certain improvements in bioclimatic conditions could be expected in the narrow forest-steppe strip North of 55"N and in the whole steppe belt of East Asia under the impact of

Soils on a warmer earlh: subboreal regions

189

global warming of 1 O C . At the same time, the steppes and prairies of Central and Eastern Europe, as well as of North America, where the presently most productive Chernozems and Phaeozems occur, will experience some aridification and a drop of biological productivity. To a less extent aridification will occur in Western Siberia and Kazakhstan. Substantial changes could be expected in the water-salt regime of soils and less substantial in the process of humification. However, by the year 2000 these changes will not cause any marked shift in zonal soil boundaries. Within the zones, there could be a change in ratio between automorphic, semihydromorphic and hydromorphic soils connected with different groundwater levels. In general, there will be a tendency to some levelling of the climatic conditions and corresponding soil conditions for crop production within the chemozemic zone. The changes, expected by the year 2025 and conditioned by a global warming of 2"C, might be predicted analog with the conditions characteristic for the Mikulinsk Interglacial (Sangamonian Interglacial in the USA). In this case the warming of 2°C will bring about an increase of precipitation by ca 50 mm within the whole belt of steppes and prairies of Eurasia and North America. In soil evolution there will be a tendency to shift towards the warmer and more humid soils, towards the more productive Chemozems and Phaeozems, which are now characteristic for more humid and warmer parts of the zone. The soil productivity will go up throughout the zone. This trend will persist with a further warming to 3°C by the year 2050.

POSSIBLE S O L CHANGE IN THE KASTANOZEMIC ZONE In the zone of dry steppe Kastanozems of Eastern Europe and North Amcrica the annual precipitation will decrease under the impact of global warming by 1°C up lo the year 2000; in Western Siberia and Kazakhstan they will remain unchanged, while in Eastern Asia they will increase by 50-100 mm. In the dry steppes of Eastem Europe and North America there will be further aridification of the landscapes; the evolution will go in a direction of a semidesert. It is not expected that up to the year 2000 these changes would cause any marked change in soil properties with the exception of water-salt regimes. Some reduction of the area of semihydromorphic and hydromorphic soils is anticipated, and an increase of soil salinity and alkalinity due to a drop of the water level in inland water basins and blowing of salts from their drying bottoms. Some decrease of soil humus content is also possible. The water supply conditions for natural and cultivated vegetation will drastically worsen. Hence, a drop of productivity of pastures and croplands will occur. In Western Siberia and Kazakhstan a warming of 1°C with an unchanged amount of precipitation will lead to an increase of evaporation and general aridification, but it will be expressed less explicitly than in Eastern Europe. and

190

B.G. Rozonov and E M . Somoilovo

North America. In addition, the warming will prolong the frost free period and lower the frequency of spring and autumn frosts, thus inducing a more favorable impact upon vegetation grown and soil formation. In the dry steppes of Eastern Siberia, Mongolia and China the warming will be accompanied by an increase of precipitation. For the regions of Eastern Asia, where deep and continuously frozen soils occur, this warming will be extremely favorable. Here an intensification of humification, desalinization, and dealkalinization of soils can be expected. The general trend of soil evolution will be in a direction from Kastanozems to Chernozems. In weakly drained lowlands the area of semihydromorphic and hydromorphic soils will expand. However, it is hardly possible that these changes will lead to a marked shift of soil zonal boundaries or to a change of taxonomic positions of major soils of the zone. Agriculture will be in a more favorable position with an increase in productivity.

CONCLUSION The above considerations concerning the expected soil changes in the subboreal region on a warmer earth were based on the climate change analyses in different parts of the region (Changing Climate 1983; Vinilov 1986; Budyko and Israel 1987; Budyko and Groysman 1989) Naturally, these considerations can only be very general at this stage. However, soil science has accumulated a large amount of results of rather detailed and accurate studies of soil-climate relationships in different soil zones. These data can be used as a starting point in the analysis of possible soil changes related to observed and predicted climatic changes. Together with the existing data of paleopedology they provide a reliable basis for this endeavor. If climatologists, in future, can provide us with a series of global or regional maps containing different scenarios of possible climatic changes in different time frames, we shall be able to compose adequate maps of corresponding soil changes. However, this is not a simple task, it requires time and international cooperation of many scientists and institutions. It can be done within the framework of the International Geosphere - Biosphere Programme.

REFERENCES Budyko M.I., Groysman P.J. (1989). Warming of the 80s. Meteorologia i Hydrologia, No 3, 5-10 (in Russian). Budyko M.I., Israel Y.A. (Eds.) (1987). Anthropogenic climate changes. Hydrometeoizdat. Leningrad (in Russian). Changing Climate (1983). Nat. Acad. Press, Washington, D.C. Vinnikov K . Y . (1986). Sensibility of Climate. Hydrometeoizdat, Leningrad (in Russian).

191

Chapter I7

CLIMATE-INDUCED CHANGES OF THE BOREAL AND SUBPOLAR SOILS S.V.Goryachkin and V.O.Targulian Institute of Geography, USSR Academy of Science 2g Staromonetny Line, 109017 Moscow, USSR.

ABSTRACT Human induced change may be most significant in the North, therefore the subpolar and boreal soils may be affected by it. The evaluation of possible soil changes in the future is the main task of this paper. We try to assess the potential changeability of soil using the following prepositions, operations and concepts: 1) four quantitative change scenarios (warming, cooling and aridization); 2) environmental parameters for main subpolar and boreal soil units; 3) grouping of the soil attributes according to their characteristic response time (CRT); 4) climatic soil evolution and chronosequences concept; soil reflectivity and sensitivity concept. The climatic changeability of 12 subunits of subpolar and boreal soils (Cambisols, Podzols, Histosols, Rendzinas, Gleysols) was considered. It was concluded that individual soil features may be subdivided into 13 groups of changes according to their CRT and occurrence frequency (from very rapid changes occurring everywhere, to very slow changes occurring once). Since within each soil body the unstable and stable features are combined, there would be no equal response by different soil features or by whole soil bodies to even one climate change scenario. Soil bodies (units, subunits) change was assessed as a soil shift within the F A 0 World Soil Map legend. The main part of subpolar and boreal soils would have no classification shifts at all or only shifts within each unit. Some have shifts from one sub-unit to the other, and a few have shifts from unit to unit. The most changeable soils are Podzoluvisols. Cambisols, Greyzems, and the less changeable soils are Greysols. Histosols, Podzols, Leptisols. The spatial and temporal changeability of the soil is controlled mainly by: 1) the CRT of each individual soil feature and soil process; 2) "sensitivity" of the parent material and soil solid phase LO heat and moisture regime changes; and 3) the degree of coincidence or contradiction between the existing soil features and processes and the future climate induced soil changing factors and processes.

INTRODUCTION The world scientific community is challenged by the problem of climate change caused by man induced C 0 2 increases in the atmosphere. The biggest climate change is suspected to occur in the subpolar and boreal belts. The main purpose of this paper is to analyze possible alterations of the subpolar and boreal soils due to climate change. This is a rather new aspect in pedology, because the

Table 17.1

Environmental ranges of boreal and subpolar soils

Soils in different classifications FAO/UNESCO 1977 Glazovskaya and Fridland 1982 USSR Orthic Podzols Fern-humic permafrost (Po) Podzols Humic Podmls Fern-humic, frequently

Textural class F A 0 1977

coarse

Ferric Podzols Ferric Podzols (Pf) Dystric P o d ~ o Typical ~. podzolic soils luvisols (Dd) Dystric Cambisols (Bd) Acid Burozems Rendzinas Raw humus calcareous (El) permafrost soils Dystric Gleysols (Gdl) Gley tundra soils Dystric Gleysols (Gd2) Gley taiga soils Eutric Gleysols

Gley forest, meadowforest soils Dystric Histosols (Odl) Tundra dry-peat soil Dystric Histosols (Od2) Peat-gley, peat oligotrophic soils Eutric Histosols Peat-gley, peat eutrophic Joe) soils Gley tundra permafrost Gelic Gleysols (Gxl) soils Gelic Gleysols (Gx2) Gley taiga permafrost soils Gelic Cambisols-leptic Dry-permafrost podburs Podzols (Bx-PI) Dystric and gelic Raw humus forest acid Cambisols (Bdx) soils

(W

Hydrological regime @c C' percolative, July SI - stagnant) pc +8+16

300-1000

PC

+10+20

coarse medium

PC PC

coarse, medium coarse, medium medium, fine fine, medium, coarse fine, medium, coarse coarse organic or fine organic or fine medium, fine coarse, medium, fine coarse coarse

coarse

Climate XT >1WC

C '

300-2500

Period with Annual soil temp. < precipitation O", months (P),mm -24-36 >8pf (per 250-800 mafrost) +1-26 0-8 400-800

100-1000

+10+20 +15+18

500-3500 1000-1600

+6-28 -12-28

0-8 5-8

450-2200 500-800

100-1000 180-300

PC

+12+22

1500-3500

+6-20

0-8

600-2500

150-1700

PC

+12+16

500-1000

-36-40

>8pf

300-500

50-200

st

+10+12

500

-6-16

5-8

250-400

100-200

st

+14+20

800-2000

-6-25

5-8

500-700

100-350

January

P-ETo, mm

0-300

. s d '/1 e

f:aR a

it

5

st

+16+20

1500-2500

+6-20

0-5

600-1000

50-450

9

St

st

+4+10 +12+20

300 500-2200

-8-38 +I-25

>8 pf 0-8

200-600 500- 1000

100-400 50-450

Do

st

+16+20

1500-3500

+5-20

0-8

500-1000

50-450

st

+4+12

500

-17-38

>8 pf

300-500

100-300

St

+8+18

300-1000

-24-50

>8pf

200-600

0-300

PC

+8+16

300-1000

-24-36

>8pf

300-800

0-400

PC

+12+18

1000-2000

-1-34

2->8 pf

500-1000

200-600

T t

9'

3.PV 19-

Soils on a wormer earth: boreal and subpolar regions

3 o S P I PZ-

3 O Z I + I p+

.-U

3052-I I+

309I+I ZI+

d

3.0Z+1OI+

SLEP

.000I

SLEP OP

SLEP OSI-09

.00zz-o001

001-OP .00zz-o00 I

193

194

S.V. Goryachkin and V.O. Targulian

major part of soils alteration was so far studied by reconstructing the rather slow soil evolution, connecting it with long-term climatic changes, but seldom as forecasts of fast changes in the future. Pedologists proposed some ideas which can be used for predicting soil change (Jenny 1941; Rode 1947; Nikiforoff 1949; Hekstra 1988; Walker and Graetz 1989).

PRESENT CONCEPTS AND THEORETICAL BACKGROUND Some existing pedological concepts can be used as an approach to this problem: 1) The concept of soil evolution as proposed by Rode (1947) has been further developed in many publications. Climate induced soil changes can be considered to be analogous to natural soil evolution caused by long-term climate change; 2) The concept of “soil sensitivity” (Sokolov and Targulian 1976; Fedoroff 1987; Yassoglou 1987) supposes that individual soils have their own ways of responding to climate changes; 3) The concept of characteristic response time of soil properties or processes (Armand and Targulian 1974) assumes that soil processes need varying periods of time to come into quasiequilibrium with the environment after its change; and 4) The concept of soil chronosequence (Jenny 1941; Yaalon 1971) supposes that some soil changes in space could be assumed as time changes if the soils have similar environments but different ages. Pedologists have experience in substituting soil change in space for temporal soil change. This experience can be used as the basis for predicting future climate induced changes of soils. Above theoretical concepts serve as a starting point for our study. The most important sources of world and boreal soil information are the world soil maps, particularly the most comprehensive ones. Soils of the World (USDA 1972), Soil Map of the World (FAOKJNESCO 1977) and Soil Map of the World (Glazovskaya and Fridland 1982). One can obtain information about the: 1) distribution of the main soil units, including the distribution of diagnostic soil horizons (USDA and FAO) and soil temperature and hydrological regimes (partially USDA, Glazovskaya and Fridland); 2) distribution of the main texture classes (FAO, Glazovskaya and Fridland); 3) soil phases (FAO); 4) slopes (FAO); and 5) soil cover patterns (Glazovskaya and Fridland). This information is mainly static, whereas in order to predict soil changes we nced some facts about soil dynamics. Unfortunately, there is no temporal data base for world or boreal soils. This knowledge is disseminated in many publications on soil chronosequence investigations and on monitoring station data. One of the most serious obstacles in predicting climate induced soil changes is the lack of data on quantitative relationships between labile soil features and climatic characteristics. The main approach to predict climate induced rapid soil changes is the consecutive expert evaluation of soil properties, processes, bodies and soil cover changes.

Soils on a warmer earth: boreal and subpolar regiom

195

BOREAL SOILS: THEIR ENVIRONMENTAL CHARACIERISTICS AND "ECOLOGICAL GROUPS" Based on soil map analysis, 26 boreal and 5 subboreal and subpolar soil units were selected that represent more than 90% of the soil cover. Every soil unit area was compared with some maps of the World Agroclimatic Atlas (1972) and the climatic range for every soil unit was determined. The information on textural classes was taken from FAO/UNESCO (1977), the hydrological regimes from Glazovskaya and Fridland (1982). The result of this analysis is shown in Table 17.1. As shown in the table, soil units which developed on calcareous and coarse parent material, and soil units with stagnant hydrological regimes, have t k largest climatic amplitudes. This analysis allows us to create a matrix table in which soil units are distributed according to their environmental characteristics. Using this table, t k main environmental soil groupings were made. The result of this exercise is shown in Table 17.2. Some cells of this table are empty due to the lack of such combinations of climate, relief and substrata. Using the knowledge about soil property dynamics gathered from monitoring stations and based on soil chronosequence investigations, we compiled the ranged sequence of soil properties in accordance to their characteristic response times or time changeabilities. These soil properties are used for soil unit and subunit diagnostics in the revised legend of the Soil Map of the World (FAOAJNESCO 1988). We added a wide set of soil regimes, biotic and other characteristics to those properties (see Table 17.3).

QUALITATIVE EVALUATION OF THE MAIN BOREAL AND SUBPOLAR SOILS AND CHANGES IN THEIR PROPERTIES DUE TO CLIMATE CHANGES. Nine subpolar and boreal soil units, including 14 soil subunits, were analyzed on probable future changes. Other soil subunits have much in common with their ecological neighbors, so we evaluated only their possible classification position change. We did not analyze all the soil properties listed in Table 17.3. For example, we left out the properties with characteristic times of more than 1000 years, because it is not reliable to analyze relatively stable properties. Orthic, humic and ferric Podzols

These 3 soil subunits (Table 17.4) are distributed widely in boreal, subboreal and even polar belts, both in their humid and subhumid segments, and are closely connected to coarse acid parent materials. The temperature regime and moisture content of Podzols change rapidly, in close correspondence to climate

196

S.V. Goryachkin and V.O.Targulian

change: less precipitation - lower moisture content, more heat - warmer temperature regime, etc: Table 17.3 Soil features and properties with differeni characreristic time

Characteristic Time in Years

Soil Features and Properties

10'

heat regime, aeration regime, composition of gases, composition of solutions, moisture content, microbiota moisture regime, fertility regime, annual biota, litter properties, pH, base saturation, salinity/alkalinity,fluvic, gleyic and stagnic properties, permafrost, gelundic phase, inundic phase, salic phase humus content and composition (topsoil), relative fertility status, salic, calcareous, sodic, vertic properties, histic (< 20 cm), ochric A, gypsic, albic E (in podzols), spodic (immature) horizons, placic phase

n.10' - 10'

n.10' - lo2

(Weak)

n-102 - lo3

tree roots, color (yellowish,reddish), Fe-concretions, depth (in loose parent material), clay cutans, interfingering,smeary consistence, andic properties, histic (> 20 cm), mollic, umbric, calcic, albic E (in soils of medium and fine texture), cambic, spodic (mature) horizons, placic phase

The waterhydrological regime remains percolative because of the enormous permeability of the coarse textured substrata. Due to increasing temperature, th? intensity of bioproductivity and biogeochemical cycles would escalate, and characteristics such as base saturation, humus content, relative fertility status and pH would increase. The litter and histic horizons, if any, would become thinner, and at the same time an ochric A horizon would occur in the upper part of the Podzol profile. Due to humidity and temperature increase, the leaching would intensify and there could be some discrepancy between processes of accumulation connected with increased bioproductivity, and leaching processes that could result in changes of pH and humus content. Gleyic properties are correlated with humidity - if it becomes less humid, there will be less or even no gleyic properties. In the case of increased temperature and moisture, an albic E horizon could remain the same or become thicker, and it could be replaced by an ochric A horizon due to warming and aridization. A spodic B horizon would remain the same or become thicker in accordance with future warming and humidization. So, the sequence of Podzol alteration after climate change would be: < 10 years - alterations of some properties within the same sub-units; 10-100years transformation of some humic Podzols into orthic ones and some orthic Podzols into ferric ones (warming + aridization ) .

197

Soils on a warmer earth: boreal and subpolar regions

Table 17.4 Orthic, humic andferric Podzols

Characteristic Time in Years

Properties

<1

heat regime moisture content water regime PH base saturation litter thickness gleyic properties humus content re]. fert. stat. histic horizon ochric A horizon albic E horizon spodic B horizon placic phase solum depth cam bic B horizon

1 -10'

n.10' - lo2

n.1o2 - lo3

Result

Climatic Scenarios Humidization Aridmtion Warming Warming WXIII€X

warmer

more

less Same higher

same

same, higher higher thinner Same

same, more higher thinner

occurrence same, thicker same, thicker same same, more no orthic and ferric Podzols

higher thinner loss

more higher thinner occmw same, loss lighter same Same

no orthic and femc Podzols

Dystric Podzoluvisols

In comparison with Podzols, dystric Podzoluvisols have much narrower ecological amplitudes - they occur only in the humid segment of the boreal belt and only on fine or medium acid substrates (see Table 17.5). Some changes are similar to Podzols, but some properties have another character of change. The water regime will be the same (percolative) only in case of warm humidization; in the case of warm aridization a period of impercolative water regime may occur. Humidization results in more distinct gleyic properties and aridization results in their loss. Humus content would be higher due to warming (because of higher bioproductivity). Histic H horizon is not a characteristic feature of Podzoluvisols, but it could occur in the case of humidization. Aridization would decrease the intensity of the podzolization processes and increase the accumulation of humus in the upper horizon; clay cutans and interfingering would be inherited and at first, an ochric A horizon would occur. It would then even be transformed into a mollic A horizon (warming + aridization). This is why the total thickness of the albic E horizon may be reduced. Due to humidization the processes of podzolization could be reinforced and such soil features as interfingering and

198

S . V . Goryachkin and V.O. Targulion

thickness of the albic E horizon would increase; only humus accumulation processcs due to warming could cover up the albic E horizon enlargement. So, the sequence of dystric Podzoluvisol alteration is as follows: < 10 years - alteration of some properties predominantly within the same subunit; 10-100 years - transformation of some dystric Podzoluvisols into gleyic (humidization) and eutric (aridization) Podzoluvisols; > 100 years - transformation of some Podzoluvisols into orthic and gleyic Luvisols (warm humidization) or into orthic Greyzems (warming + aridization). Table 17.5

Dystric Podzoluvisols

Characteristic Time in Years

Properties

< 1

heat regime moisture content water regime PH base saturation litter thickness gleyic properties humus content rel. fert. stat. histic horizon ochric A horizon clay cutans interfingering mollic horizon albic E horizon

1-10'

n.10' - lo2

n.1o2 - lo3

Result

Climatic Scenarios H um idization Aridization Warming Warming warmer warmer more less Same same, impercolative same, higher higher same, higher higher same, thinner thinner same, more distinct loss more more higher higher, same no, Occurrence no no, Occurrence occurrem more same more distinct same no occurrence same, thinner thinner orthic and orthic Greyzems gleyic Luvisols

D yst ric Cambisols Dystric Cambisols occur in the humid segments of the boreal belt on basic, intermediate and even acid paEnt material of coarse and partially medium texture. These soils would have some alterations similar to those of Podzols (Table 17.6). The percolative hydrological regime of some dystric Cambisols could change more distinctly in comparison with Podzols - it could become periodically impercolative in the case of aridization because of its partially finer texture. Humus content change is similar to that of Podzoluvisols, as there is no complicated combination of accumulative and illuvial humus (as in Podsols) in

199

Soils on a warmer earlh: boreal and subpolar regions

dystric Cambisols. Cambic B horizon would remain the same (aridization) or even become thicker (warm humidization) because of the increase in weathering intensity. The umbric A horizon would become thicker (warming + humidization) or even transform into mollic A horizon (warming + aridization). Thus, we suppose the following succession of dystric Cambisols change: c 100 years appearance of humic Cambisols among eutric ones (in the case of aridization); > 100 years - transformation of some humic Cambisols into haplic Phaeozems (warming + aridization). Table 17.6

Dyszric Cambisols

Climatic Scenarios Characteristic Time in Years

Properties

< 1

heat regime moisture content water regime

1-10]

PH n.10' - lo2

n.1O2 - lo3

Result

base saturation litter thickness humus content rel. fert. stat. histic horizon spodic B horizon mollic horizon umbric A horizon cambic B horizon

Humidization

Warming warmer more same same, higher same, higher thinner more

higher thinner, loss no more thicker

thicker dysmc and eutric Cambisols

Aridization Warming warmer

less same, impercolative higher higher thinner more higher loss no occurrence trans. to mollic same

eutric and humic Cambisols, haplic Phaeozems

Rendzinas Rendzinas develop on calcareous substrates of different textures (predominantly coarse) in both humid and subhumid segments of boreal, subboreal and polar belts. These soils have a very wide ecological amplitude (Table 17.7). The hydrological regime of these soils would remain percolative in spite of any climate alteration because of the high permeability of coarse or wellstructured soil material. There could be some alterations of pH but it would still remain alkaline or neutral and weakly acid only in the upper horizon. They would be base-saturated even in spite of humidization. The litter thickness would decline due to warming because of more intensive transformation of litter. Permafrost

200

S.V. Goryachkin and V . O . Targulian

could change or disappear due to various climatic alterations but it would never play a role in the formation of these soils because of its dryness. Humus content and composition will remain the same or change slightly due to calcareous soil mass, but the store of organic carbon could increase with climate warming because of the upper horizon thickness alteration. In spite of the expected change of leaching intensity, calcareous properties would remain the same. Perhaps long leaching after the climate has become warmer and more humid would result in a notable decline of CaC03 content. In this case even a cambic B horizon could occur in a few Rendzinas. Mollic horizons could become thicker or occur in Rendzinas due to higher bioproductivity in a warmer climate. Thus, Rendzinas have only slight changes after climate changes; only after more than n"100 years some Rendzinas could become calcaric cambisols (warming + humidity) Table 17.7 ~

~~

~~

Rendzinas ~

Characteristic Time in Years < I 1-10'

n-10' - lo2

n.10~- lo3 Result

~

Climatic Scenarios Humidization Aridization Warming Warming heat regime warmer Warnla moisture content more less water regime same same lower same PH base saturation same same litter thickness thinner thinner gleyic properties no no permafrost no, deeper no, deeper humus content same same rel. fert. stat. same same calcareousproperties less distinct same histic horizon no no mollic horizon thicker, Occurrence thicker, Occurrence cambic B horizon no, Occurrence no Rendzinas, rarely Rendzinas calcic Cambisols Properties

Eutric and dystric Gleysols

Eutric and dystric Gleysols have poor or very poor drainage caused by their relief and/or substrata conditions. They occur less in semihumid than in humid segments of boreal and subboreal belts, on acid, coarse, medium and fine parent materials. Their hydrological regime is stagnant. Some parts of these soils get

Soils on a warmer earlh: boreal and subpolar regions

20 1

their moisture not only from the atmosphere directly, but also from neighboring soils occupying higher positions. That is why expected changes of these soils are connected not only with climate changes directly but also with future changes of their neighbors (Table 17.8). The water regime of Gleysols of the present humid climate would remain the same in spite of different humidity alterations, but in some places in semihumid regions, aridization could cause the presence of some periods of water pulsation. Base saturation and pH would depend on climate change as well as on their initial values: in the case of warming, the bioproductivity would increase but it could result in a contrast change of these two properties only in dystric Gleysols, which would become more eutrophic. Gleyic properties are closely Connected with humidization: less humid - less distinct gleyic properties. Histic H horizon would increase in the case of humidization and decrease or even become replaced by an ochric A and then a mollic A horizons due to warm aridization. In some Gleysols of the semihumid regions (in hydromorphic relief positions), aridization could result in fundamental soil profile changes - the occurrence of mollic A, albic E, cambic B or Bt and spodic B (coarse substrata) horizons. The possible succession of Gleysoils may be the following: c 10 years - in accordance with climatic change some eutric Gleysols would be transformed into dystric ones; 10-100 years - some Gleysols would be transformed into Histosols (humidization) and some of them into gleyic Cambisols (aridization); > 100 years - in the case of warm aridization, some Gleysols and gleyic Cambisols would become mollic Planosols (or Luvisols) and gleyic Podzols (on coarse substrates). Eutric and dystric Histosols Eutric and dystric Histosols are soils of very poor drainage occurring in both humid and semihumid segments of boreal, subboreal and subpolar belts. These bog soils change very little in spite of different climate changes because of the extreme water content in the peat and because they occupy relief depressions which always get more water than any other relief position. This is why eutric and dystric Histosols are the only soils in our analysis which almost never change their moisture content due to humidization or aridization (Table 17.9). The majority of their properties remain the same. Only in some cases due to aridization and subsequent increase of ground and soil water mineralization, some Histosols get higher pH and base saturations. Only one property would certainly change wiLh climatic and bioproductivity changes - the thickness of a histic H horizon. In the case of warm humidization it would become thicker. In the case of warm aridization, the processes of peat growth due to higher bioproductivity would be hindered by peat mineralization processes due to less humidity, and the histic H horizon thickness could remain the same.

202

S.V. Goryachkin and V.O. Targulian

All possible changes, supposed to be very slight, could change a Histosol classification in 10 or n 10 years. Some dystric Histosols could become eutric (aridization) and vice versa (humidization). Table 17.8 Eulric and dysrric Gleysols

Characteristic Time in Years

Properties

<1

heat regime moisture content water regime

1-10'

n.10' - lo2

n.10' - lo3

Result

PH base saturation litter thickness gleyic properties humus content rel. fen. stat. histic horizon ochric A horizon placic phase mollic horizon albic E horizon cambic B horizon spodic B horizon

Climatic Scenarios Humidization Aridization Warming Warming WaflIler WNITlH same less same same, pulsation same, higher same, higher same, higher same, higher thinner thicker same, more distinct less distinct more more same, higher higher thinner thicker no, occurrence no no, occurrence no no no, occurrence no no, occurrence no no, occurrence no no, occurrence eutric Gleysols, eutric and dystric Gleysols and Histosols Planosols on fine and Podzols on coarse substrata

Gelic Gleyols Gelic Gleysols are the permafrost soils of the humid boreal and subpolar cold continental climate. We analyze those which occupy well drained relief positions, but still have imperfect drainage because of shallow permafrost and fine or medium texture of parent material. The major part of gelic Gleysol property changes are similar to those of some other boreal soils (see Tables 17.5, 17.6 and 17.10). Still permafrost causes peculiarities in some change processes. As permafrost degrades and becomes deeper due to warming, the water regime can be changed from stagnant to periodically percolative (aridization) or remain the same (humidization) because

203

Soils on a warmer earth: boreal and subpolar regions

of increased water content. Many phenomena of permafrost degradation will influence the soil body; pedoturbution, solifluction, surface subsidence, creep, etc. As these soils are saturated with water for long periods, humidization will not change gelic Gleysols prominently, contrary to aridization. Aridization would cause the occurrence of Fe-concretions and cambic B horizons, (which would replace the horizon with distinct gleyic properties), the occurrence of an ochric and later umbric horizon, and the loss of histic H horizon thickness. So, < 10 years after climate change, soil alterations would concern only non-diagnostic properties, except perhaps permafrost degradation which causes transformation of gelic Gleysols into dystric ones. In n 100-1ooO years, the soil property changes would result in transformation of gelic Gleysols into gleyic Regosols in the case of permafrost degradation, or into gleyic, gelic, dystric and even eutnc Cambisols in the case of warming + aridization. Humidization would not cause enough soil property alterations for transformation of gelic (or dystric) Gleysols into another subunit. It must be noted here that a rapid and strong warming could cause intensive permafrost degradation and subsequent intensification of such exogenic processes as solifluction and creep, which could completely destroy these soils. Table 17.9

Eutric and dystric Histosols

Characteristic Time in Years

Properties

<1

heat regime moisture content water regime

1-10'

PH

n.10' - lo2 n.1o2 - lo3 Result

base saturation rel. fert. stat. histic horizon depth

Climatic Scenarios Humidization Aridization Warming

Warming

warmer

warmer same, less

Same

same Same same

Same

same, higher

thicker

same, higher same same

more

same

eumc and dystric Histosols

eumc Histosols

Same

Gelic Cambisols Gelic Cambisols are the soils of semihumid (and partially humid) segments of boreal and subpolar permafrost regions with subpercolative hydrological regime, developed on all kinds of substrata. Warm humidization would result in the occurrence of some Podzol soil features on coarse substratum and in some

204

S.V. Goryachkin and V . O . Targulian

Podzoluvisol features on fine and medium substrata (Table 17.1l), but many properties would change similar to those already described above. Warm humidization and subsequent deepening or loss of permafrost would transform this regime into a percolative one and could cause the appearance of albic E, spodic B (on coarse substrata) and some features of Bt (cracking clay, interfngering on fine and medium substrata) horizons. Aridization would cause the gelic Cambisol subpercolative water regime to change to a nonpercolative one. Hence, it would result in more distinct calcareous properties, thicker calcic horizons and appearance of mollic A horizons. All these horizon alterations would change or transform the cambic B horizon. Thus, gelic Cambisols are supposed to have the most pronounced classification changes. On coarse substrata in < 100 years after the climate becomes more humid and warmer, gelic Cambisols would be transformed into leptic Podzols (acid rocks) or dystric Cambisols (basic rocks). In n 100-1OOO years after the same climate alteration, gelic Cambisols on fine and medium substrates would be transformed in the direction of dystric or eutric Podzoluvisols. Warm aridization in n 100-1000 years would result in the transformation of gelic Cambisols into calcaric Phaeozems.

Diversity of soil properties, soil and soil cover type changes due to climate changes The main conclusion of our paper is that soils would show great diversity of changes due to climate changes. They differ in kind and intensity. The different soil properties have their own characteristic response time, and this causes soil properties to change differently in rate and in frequency. Soil profiles would show a great diversity of changes corresponding to the concept of soil sensitivity. Some soils have only slight changes (if any) in spite of the different climate changes and others would change such that they would be classified differently. The results of the previous soil subunit analysis are assembled in Table 17.12, where the soils shift degrees in respect to their classification (FAO) following different climate changes scenarios and for different periods of time after climate changes started. It becomes clear that many boreal and subpolar soils would have no classification shifts or would only shift within soil units (from one subunit to another). Only 4-8 soils would shift from unit to unit. The soils of the lowest changeability are those developed on peats (Histosols), coarse (Podzols) and calcareous (Rendzinas) substrata, those of impeded drainage (Gleysols, Histosols), and those forming under the control of litho-gcomorphological processes (Lithosols, Fluvisols, Regosols, Andosols). The highest soil changeabilities occur mainly in soils developed on medium textured substrata (Podzoluvisols, some Cambisols). As a rule, soil needs 10-100 years to change within a unit and > 100 years to change into another.

205

Soils on a warmer earth: boreal and subpolar regions

Warm aridization would cause more soil shifts than warm humidization, since the first scenario shows a greater impact on soil change than the second one. Table 17.10 Gelic Gleysols ~~

Characteristic Time in Years <1 1-10'

n.10' - IO*

n.1o2 - 103

Rcsult

Properties heat regime moisture content water regime PH base saturation litter thickness gleyic properties permafrost gelundic phase humus content rel. fert. stat. histic horizon ochric A horizon Fe-concretions umbric A horizons cambic B horizon

~~

Climatic Scenarios Humidization Aridization Warming Warming warmer Warme€ more less same percolative same, lower higher same, lower higher same, thicker thinner less distinct same deeper deeper less distinct, loss less distinct, loss more more same, higher higher thinner thinner occurrence occurrence no occurrence no occurrence no occurrence gleyic, dystric and dystric Gleysols or eutric Cam bisols Regosols in area margins

What do we need to transform the expert evaluation of soil change into a soil change model?

This paper is mainly a qualitative expert evaluation of soil change, based on soil genesis and geography experience. Unfortunately, we could not describe the full algorithm of such an evaluation. We realize that our evaluation includes a lot of discussion opinions and depends strongly on both our personal approaches and the obvious uncertainty of soil changes in different climate change scenarios.

206

S . V . Goryachkin and V . O . Targulian

Table 17.11 Gelic Cambisols

Climatic Scenarios Climatic Characteristic Time in Year < I 1-10'

n.10' - lo2

n.1O2 - lo3

Result

scenarios

Properties H um idization Warming heat regime warmer moisture content more water regime percolative lower PH base saturation lower litter thickness thicker gleyic properties no, Occurrence permafrost deeper, loss gelundic phase no, loss humus content same rel. fert. stat. same calcareous properties loss hi stic horizon occurrence ochric A horizon same, thinner albic E horizon occurrence spodic B horizon no, occurrence Feconcretions occurrem clay cutans M), occurrence in terfingering no, occurrence mollic horizon no loss calcic horizon cambic B horizon trans. to S ~ or C argic, same dystric Cambisols, leptic Podzols dysmc and eutric Podzoluvisols

Aridization Warming warmer less impercolative higher higher thinner no deeper, loss no, loss more higher more distinct no thicker no no no no no occurrem thicker same, thicker calcaric Phaeozems

This study was done lacking a generalized data bases on the major boreal and subpolar soil units and is an example of non formalized brainstorming. We hope it presents a challenge for more exact collection, analysis, and generalization of the empiric data. Further needs are given below.

Table 17.12

Degree of shifts of boral and subpolar soils due to dferent climatic scenarios Climatic scenarios

Degree of shifts

Time

Humidization Warming

No classification (FA0)change

0-loo0 years

Aridization Cooling

Humidization Cooling

Oe, Od, I, Jt, Jd, Oe, I, Jt, Jc, Je, Je, Ge, Gd, Rd, Pf, Mg, Re, E, Pf Bd

Ox, I, Jt, Jc, Je, Gx, To, Re, E, Pf, Mo,

Ox, I, Jt, Jd, Gx, To, Rd, E, Ph, Mg


Gx-Gd, Bd-Be

Jd-Je, Gx-Gd, Gd-Ge, Bd-Be

Jd-Je,

Gd-Ge

Je-Jd, Ge-Gd

10-100

Ox-Od, Jc-Je, Gd-Od, Re-Rd, Ph-Po, Po-Pf, Bx-Bd

Ox-Od, Od-Oe, Rd-Re, Ph-Po, Po-Pf, Pg-Ph, Mg-Mo, Dd-De, Be-Bh, Bd-Be, Bx-Be

Od-Ox, Rd-Re, Po-Pf, Mg-Mo,

Oe-Ox, Ph-Po, Pg-Ph, Dd-De

Od-Ox, Oe-Ox, Jc-Je, Gd-Od, Re-Rd, Pf-PI, Po-Ph, Dd-Dg, De-Dd

Shift within units (FAO)

Aridization Warming

l k

Shift

<1

from unit

10-100

Gx-Bg, Ge-Bg

Gx-Rx, Ge-Wx

Bx-PI, Bd-Gx

to unit

100-1OOO EmBk, Pg-Gd,

Ge-Wm, Gd-Pg, Mo-HI, CI, De-Mo Bh-Ho, Bx-Hc

Po-Bx, Bx-Xh

Pg-Gd, Mo-De

Mo-Lo, Dd-Lo, De-Lo

2

P 0

a

208

S.V. Goryochkin and V . O . Torgulion

1)

Soil science has accumulated a huge amount of soil data that has been generalized in the 3 main world soil maps (FAOKJNESCO 1977; USDA 1972; Glazovskaya and Fridland 1982), which are used as the main sources of soil information on a global scale. Soil science has also theoretically elaborated the concept of different global soil functions: atmospheric, hydrospheric, lithospheric and biospheric. Soil characteristics, explicated on global soil maps, deal mainly with the substantial and genetic stable features of the soil solid phase. This information is not sufficient for global modeling of biosphere-geosphere changes. We need data which characterize the atmo-, hydro-, litho- and biospheric functions of the pedosphere, e.g. gaseous, moisture, temperature, macro- and microbiota regimes of the main soil units. Unfortunately, this kind of data has not yet been systematized nor indicated on global soil maps. Furthermore, for many soil types of the world no data about their functional characteristics are available at all. For quantitative modeling of the global and macroregional soil changes we need short and medium term data of changeable soil parameters.

2)

The next problem for modeling is the lack of systematized and generalized data on temporal changeability of the whole set of soil features. Such information is dissipated in many studies on soil chronosequences and monitoring stations data. There are many gaps in the knowledge of temporal changeabilities of some soil features in many soil units. In order to elaborate a quantitative soil change model we need to know the rates and characteristic times of the main soil processes and properties (including both the most changeable and the most stable ones) within the major soil units.

3)

Perhaps the most important problem for a global soil change model is the shortage or lack of quantitative information on linkages between spatial and temporal changes of the environment and of soils. First we need to determine the environmental parameters which cause the spatial and temporal soil changes. Then it is necessary to find quantitative correlations between changes of these parameters and changes of different soil properties within the main soil units of the world.

We should stress that if global data on the three above-mentioned subjects were available, it would be possible to elaborate reliable quantitative global soil change models and incorporate them into broader geosphere-biosphere change models.

REFERENCES Armand, A.D., and V.O.Targulian (1974). Some principal limitations of experimenting and

Soils on a warmer earth: boreal and subpolar regions

209

modcling in geography (The principle of feasibility and characteristic time). Report of the USSR Academy of Science, Geography Series 4:129-138 (in Russian). FAOAJNESCO (1977). Soil Map of the World. Volumes 1-10. Paris. FAO/UNESCO (1988). Soil Map of the World. Revised Legend. Rome. (not mentioned in text). Fedoroff, N. (1987). The production potential of soils: Part I - Sensitivity of principal soil types to the intensive agriculture of northwest Europe. Pages 65-87 in Soil Protection in the European Community. Elsevier. Glazovskaya, M.A., and V.M. Fridland (1982). Soil Map of the World. Moscow (in Russian). Hekstra, G.P. (1988). Effects of future climatic changes. Ministry of Housing, Physical Planning and Environment, Leidschendam, the Netherlands (unpublished manuscript). Jenny, H. (1941). Factors of Soil Formations. McGraw-Hill, New York. Nikiforoff, C.C. (1949). Weathering and soil evolution. Soil Science 67:219-230. Rode, A . A . (1047). The Soil-Forming Process and the Evolution of Soil. MOSCOW(in Russian) . Sokolov, 1.A.. and V.O. Targulian (1976). The interaction of soil and environment soilmemory and soil-moment. Pages 3 2 4 8 in Environmental Studies. Moscow. USDA. (1972). Soil of the World. Distribution of orders and principal suborders. United States Department of Agriculture, Washington, D.C. Walker, R.H., and R.D. Graetz (1989). Effects of atmospheric and climate change on terrestrial ecosystems. Global Change Report No. . World Agroclimatic Atlas (1972). Moscow (in Russian). Yaalon, D.H. (1971). Soil forming processes in time and space, pp. 29-39. In D.H. Yaalon, (Ed.) Paleopedology - origin, nature and dating of Paleosols. Israel University he ss, Jerusalem. Yassoglou, N.J. (1987). The production potential of soils: Part I1 - Sensitivity of the soil systems in Southern Europe to degrading influxes, pp. 87.123. In Soil protection in the European Community. Elsevier.

This Page Intentionally Left Blank

21 1

Chapter 18

APPROACHES TO MITIGATE TROPICAL DEFORESTATION BY SUSTAINABLE SOIL MANAGEMENT PRACTICES Pedro A. Sanchez, Cheryl A. Palm and Thomas Jot Smyth Soil Science Department, Tropical Soils Research Programme North Carolina State University, Box 7619, Raleigh, N.C. 276957619, USA

ABSTRACT Deforestation and subsequent changes in land use in the tropics are responsible for 18% of global warming during the 1980s. with expected rates during the 1990s to increase. Third world population growth forces landless rural populations to migrate and over exploit tropical rainforests. The resulting agriculture is unsustainable and leads to further deforestation and migration to urban centers. Research has shown that deforestation rates can be decreased, by eliminating the proximate cause. An integrated approach consisting of development and application of sustainable management technologies for tropical soils and appropriate government policies will eliminate the pressure for further deforestation. The principal management technologies are based on low input systems for acid soils of the humid tropics which lead to agroforestry, legume based pastures or continuous crop cultivation. Some management technologies are available while others are still evolving. For every hectare put under sustainable agriculture five to ten hectares of rain forests are saved each year.

THE PROBLEM Recent estimates indicate that about 18% of the global warming is due to the clearing of tropical rainforests (EPA 1990), which is occumng now at a rate of 7 to 10 million hectares of primary forest per year (Melillo et al. 1985). Deforestation is also decimating the world's largest depository of plant and animal diversity (Wilson 1988). Tropical deforestation is driven by a complex set of demographic, biological, social and economic forces described in Fig. 18.1. Population growth in developing countries continues at a high rate, while most of the fertile and accessible lands are already intensively utilized. Government policies ofien exacerbate land scarcity by allowing gross inequities in land tenure. These L'actors result in an increasing landless rural population which essentially has three choices: stagnate where they are, migrate to the cities, or migrate to the rainforests that constitute the frontier of many developing countries. Although urban migrations are spontaneous, national policies in key countries include the occupation of their tropical rainforests, notably Brazil, Peru and Indonesia via colonization programmes.

212

P.A. Sanchez, C.A. Palm and T J . Smyth

THIRD WORLD POPULATION GROWTH

1

LIMlTED FERTILE LAND

URBAN CARRYING CAPACITY EXCEEDED

Rural Unemployment

UNSUSTAINABLE AGRICULTURE

Economic

Failure

FURTHER DEFORESTATION

UNEMPLOYMENT

Revolution, Social

Upheaval

LANDTEW INEQUITES

Crime Inflation

Resource Deterioration

Loss of

Accelerated

Genetic

Greenhouse

Diversity

Effed

Densely populated rural environments such as the Andean valleys, Northeast Brazil, and Java suffer from an ever decreasing farm size and the overuse of steepland areas. This results in widespread soil erosion, siltation of reservoirs and other adverse off site effects to urban centers. Migration to the cities in search of a better life results in bitter disappointments, and coupled with limited urban infrastructure, produces unmanageable cities with populations far exceeding their carrying capacity and infrastructure. Migration to the humid tropics seldom results in a bountiful cornucopia (with few notable exceptions). An equilibrium between the rainforest and shifting cultivation by traditional humid tropical societies is broken by the colonists, and

Soil managemenl practices

10 mifigate

tropical deforestation

213

in some countries by land speculators as well. The result is shifting cultivation in disequilibrium which quickly turns into various forms of unsustainable agriculture. Traditional societies are disrupted, economic failures abound and migration to urban centers follows. The two end points are urban unemployment and further deforestation. The consequence of the former is abject urban poverty which leads to widespread crime, poor health and in many cases social upheaval. Deforestation depletes the ecosystems' limited nutrient capital, decimates plant and animal genetic diversity and accelerates global warming due to carbon dioxide, methane and nitrous oxide emissions.

STRATEGIES TOWARDS A SOLUTION Several strategies have been proposed to mitigate tropical deforestation. These include: 1) economic development and more equitable land tenure in densely populated areas; 2) encouraging migration to less fragile areas such as the Cerrado of Brazil; 3) preserving the remaining forests by a vast network of wellprotected national parks; and 4) sustainable use of the forests as extractive reserves. While policies that promote these four strategies should be pursued, all are necessary but insufficient to stop deforestation. The first strategy requires long periods of real economic growth which are too slow to offset current deforestation rates. The second strategy is clearly insufficient in deflecting migration; the third one is unrealistic in preventing hungry people from clearing land; and the fourth one, extractive reserves, is likely to support a very small segment of the humid tropical population. A fifth strategy is to stop deforestation in situ by eliminating the need to abandon cleared land. Land use management options are urgently needed that improve the economic status of subsistence farmers, maintain agricultural productivity on deforested lands and recuperate productivity of degraded lands. Such options will provide sustainable development of the Amazon and other humid tropical regions in a way that satisfies human needs and preserves the ecosystem. These options must be compatible with the various socioeconomic needs in the region so that they are readily and widely adopted. Is such an approach possible in the predominant acid, low fertility soils of the humid tropics? Our answer, based on long-term research, is an emphatic yes, with the use of alternatives to slash and bum. The key is an integrated approach consisting of: 1) development and application of sustainable management technologies for tropical soils; 2) appropriate government policies that provide incentives against further deforestation; and 3) effective, economically sound rainforest conservation methods. Farmers do not cut tropical rainforests because they like to; they clear them

2 14

P.A. Sanchez. C.A. Palm and

TJ.Smyth

out of sheer necessity to grow more food. Deforestation, therefore, can be reduced by the widespread adoption of sustainable management practices that permit the use of cleared land on a continued basis. Sustainable management options for acid soils of the humid tropics have been developed at Yurimaguas, Peru, and elsewhere to fit different landscape positions, soils and levels of socioeconomic infrastructure development (Sanchez et al. 1987). The principal sustainable management options and alternatives to slash and bum are: paddy rice production on alluvial soils, low input cropping, continuous cultivation, legume-based pastures, and agroforestry. Their place in the landscape is shown in Fig. 18.2.

maay

ACC

contmuour Cmpplng ,sew

~nll..,rYC,"r.l

Lor Input croppong ,POI,

I"lrlllrUC1"I.J

R*I",.I

4roiorrriry

ForeruFarmmg Mosaic R i g a w r n l m g S19pos A l l U Y l l l Sollr

Fig 18.2

Acld Solls

Some soil management oplions for humid tropical landscapes dominated by Oxisols and Ultisols (Sanchez 1988)

Additional options developed by other research institutes include perennial crop production (rubber, oil palm), plantation forestry and alley cropping for the higher fertility soils (Kang et al. 1981; Alvim 1982). They are also part of the solution. Most of the options are based on low input systems which serve as transition technology to agroforestry systems, legume-based pastures and continuous crop rotations. In spite of the continuing need for both strategic and applied research, there is no question that the technological basis for sustainable management options for acid soils of the humid tropics is available now. The following describe some of the main management options.

Land clearing that does not damage the soil All the options are bascd on slash and bum clearing, using axes, machetes and chain saws as opposed to mechanized bulldozer clearing. We found that the

Soil management praclices lo mifigale lropical deforesfafion

215

slash and bum system is better because the bum provides a liming and fertilizing effect of the ash and does not disturb the topsoil while conventional bulldozer clearing often compacts the soil and removes valuable topsoil (Seubert et al. 1977). Whcn these results were transmitted to colleagues in Brazil and Indonesia, policies to discourage bulldozer clearing were instituted. We also found little difference in fertilizer content of the ash after burning primary or secondary forests (Smyth and Bastos 1984) giving a further impetus to clear secondary forests, instead of primary ones. Paddy rice in alluvial soils Alluvial soils encompass approximately 1 1% of the Amazon region. In addition to accessible water and river transportation to markets, large proportions of these soils are characterized by high fertility status. Adaptation of Asian flooded rice technology to Yurimaguas alluvial soils led to the establishment of the following components: 1) land is cleared by slash and bum and land leveling is performed only to ensure gravity flow of water along natural contours; 2) supplemental irrigation, by either gravity flow or pumping from rivers, increases yields by about 50% as compared to crops dependent on rainfall; 3) transplanting provides higher yields during initial crops but may be replaced by broadcasting pregerminated seeds in subsequent crops after paddies are adequately leveled; 4) fertilizer responses are thus far limited to nitrogen deficiencies with continuous use; and 5 ) with yields averaging 5.5 tons ha-1 crop-1, at the farmer level, one hectare of paddy rice provides the equivalent to 14 hectares of upland rice under shifting cultivation, since two crops are grown per year ( A r k d o et al. 1983). This technology has been adapted rapidly by settlers migrating into the Amazon Basin from rice producing areas at the coast of Peru. Approximately 75,000 hectares of new imgated rice land was put into production within five years in the Peruvian Amazon (INIPA 1984), making it one of the major rice producing regions of Peru. In Brazil, the State of Amazonas alone contains about 25 million hectares of such soils, underscoring the vast rice production potential of the Western Amazon. Low input cropping

This management option has evolved as a transition technology between shifting cultivation and several sustainable options. It enables the farmers to drastically increase short-term crop production while preparing themselves and their land for sustained land use alternatives. This option is targeted for farmers on acid, infertile soils in rural areas with limited capital and marketing infrastructure and is therefore applicable to large areas of the humid tropics. Its principal features are: 1) clearing secondary forest fallows by slash and bum; 2)

216

P.A. Sanchez, C.A. Palm and T J . Smylh

using of acid-tolerant upland rice and cowpea cultivars in rotation, with only grain removal to minimize nutrient export; 3) no use of fertilizers, lime, or external organic inputs; 4) establishment of legume fallows when weed competition and nutrient deficiencies make cropping unfeasible; and 5) elimination of the fallows by slash and bum after one year, shifting to other management options such as grass-legume pastures, agroforestry, or mechanized continuous cropping (Sanchez and Benites 1987). Current results indicate that the initial cropping cycle lasts two or three years and that there is a progressive reduction in cycle length after each legume fallow. The system is presently only considered as transitional because of two major constraints: nutrient depletion and weed encroachment. Ongoing investigations seek to prolong the duration of low input cropping by: 1) broadening the base of acid tolerant cultivars and species; 2) increasing knowledge on components of the nutrient depletion process; and 3) improving weed management through crop rotations, plant density, and frequency and time of legume cover crop fallows.

Agroforest ry Research on agroforestry began in Yurimaguas in 1983. It was the first and remains the most comprehensive agroforestry research program for acid soils of the humid tropics. Agroforestry is an attractive management option for much of the humid tropics because it can be adapted to a wide range of socioeconomic and soil-landscape conditions. Research has focused on: 1) selection and management of acid tolerant leguminous trees; 2) food-tree production systems; 3) managed fallows; and 4) alley-cropping (Szott et al. 1990). Legume selection trials have identified Inga edulis, Cassia rericulata, Erythrina sp. and certain varieties of Gliricidia sepium as promising species (Salazar and Palm 1987). These species are fast growing and respond well to pruning, making them good candidates as multipurpose trees for acid soils. Work with native fruit trees has concentrated on peach palm (Bactris gmipaes) for fruit and heart of palm production but work has also begun on arazA (Eugenia stipitata), achiote (Sixa orellana),and carambola (Averrhoa carambola). Managed leguminous fallows performed well compared to natural secondary fallows. Some of the managed fallows restored stocks of nitrogen, phosphorus, and potassium and suppressed weeds faster than the natural fallow (Szott et al. 1990). Current work in managed fallows is looking for the optimum crop-tofallow period ratios. Alley cropping trials, however, have had limited success on acid soils. Soil phosphorus levels soon became limiting to crop production because competition between trees and crops is severe and reduces crop yields considerably. Current research is aimed at overcoming these problems and testing alley cropping on slopes, where it is obvious soil conservation advantages could be maximized.

Soil management practices to mitigale tropical deforestation

217

G r ass/legume pastures In contrast to the bad reputation associated with extensive pastures for beef production in the Amazon, our soil-plant-animal research has focused on the development of pastures for dual purpose (beef-dairy) production in small land holdings where farmers will also grow crops and trees. Initial investigations sought to adapt technology from CIAT's Tropical Pastures Programme, developed primarily in savanna ecosystems, to humid tropical conditions. Legume and grass ecotypes were screened for their performance under acid soil conditions and, subsequently evaluated for their persistence and compatibility when subjected to various grazing intensities. An ongoing 11 year old grazing trial in Yurimaguas is the longest running replicated trial testing acid tolerant grass-legume mixture in the humid tropics (Ayarza et al. 1987). If current persistence of legume-dominated pastures prove to be sustainable, then a new concept for cattle production may emerge in the humid tropics. New studies are also underway to gain further insight on nutrient cycling and to refine management practices used for the transition to and from pastures to crop and/or tree production. Mechanized continuous cropping Initial investigations seeking to quantify the production potential and primary soil constraints of humid tropical ecosystems, focused on this management option. It is targeted for farmers near urban areas where favorable marketing infrastructure ensures that fertilizer-based continuous food crop production can be used. Large Amazonian cities currently import most of their food from other areas. There is a potential comparative advantage in growing such food crops near the cities where the infrastructure is available. Sustained crop yields have been obtained with continuous cropping trials for 41 crops (17 years) in Yurimaguas Ultisols (USDA)/Acrisols (FAO) and 17 crops (8 years) in Manaus Oxisols (USDA)/Ferralsols (FAO) (Sanchez et al. 1983; Alegre and Sanchez 1989; Smyth and Cravo 1989). The key to continuous production is effective crop rotations and the judicious application of lime and fertilizers. If vigorous crop growth is insured by adequate liming, fertilization, and mechanization, soil chemical properties improve with continuous cultivation. Primary research thrusts in this soil management option are currently directed towards improving nutrient use efficiency, weed control, and optimizing tillage pmctices. Nutrient cycling efficiency Nutrient cycling must be maximized in all systems in order to minimize the need for external nutrient inputs. The management of crop and root residues is

218

P.A. Sanchez, C.A. Palm and T J . Smyth

crucial in this regard. Approaches proposed by TSBF (Tropical Soil Biology and Fertility Programme) on quantifying the nutrient release of organic inputs and the management of soil organic carbon, nitrogen and phosphorus (Swift 1986; Ingram and Swift 1988) are major components of low-input cropping, agroforestry and pasture research. The promising results at Yunmaguas in predicting the rate of nutrients released from leguminous sources (Palm and Sanchez 1990) provide for the first time an opportunity for the quantitative management of organic inputs in a manner comparable to the management of chemical fertilizers.

DEFORESTATION REDUCTION POTENTIAL. For every hectare put into these sustainable soil management technologies by farmers, five to ten hectares per year of tropical rainforests will be saved from the shifting cultivator's ax, because of their higher productivity. Estimates at Yurimaguas for the various management options are given in Table 18.1. These estimates will vary with climate and soils. Such technologies are particularly applicable to secondary forest fallows, where clearing does not contribute significantly to global warming because of the small tree biomass. Nevertheless the use of secondary forest fallows is of very high priority, because in many areas they are a viable alternative to primary forest clearing. Many of the degraded or unproductive pastures or croplands resulting from poor management practices can also be reclaimed using some, but not all of these available technologies. Table 18.1

Hectares saved from deforestation for various management options; estimates for Yurimaguas in Peru

1 hectare in sustainable management options

equals hectares saved from deforestation annually

Flooded rice Low input cropping (transitional) High input cropping

11.0 4.6 8.8

Legume based pastures Agroforestry systems

10.5

'XI

not determined

Such technologies, however, are useless without effective government policies that encourage, support and regulate them. Likewise, well conceived policies will fail without sustainable technologies. Therefore, the hope lies on a joint policy-technology approach; a worldwide deforestation reduction initiative (Sanchez 1988).

Soil monagemeru practices to mitigale tropical deforestation

219

The present situation is analogous to when the world technical assistance community launched the Green Revolution in the late 1960s. At that time sustainable technologies for high yielding rice and wheat production were sufficiently developed to be tested at a large scale. Key government officials were convinced of their importance by leading scientists, and instituted the necessary policies to make massive farmer adoption possible in India, Pakistan, Philippines and other countries. The Green Revolution became a worldwide success during the next twenty years and the goal of arresting worldwide famine was definitely achieved. Unlike the Green Revolution, however, the impact of a Deforestation Reduction Initiative will be gradual and less spectacular. This is because we are focusing on marginal ecosystems, and more complex technological and policy problems.

CONCLUSION A worldwide Deforestation Reduction Initiative will help directly to improve the livelihood of both developing and developed countries. Sustainable agricultural options for the humid tropics are necessary, but not sufficient conditions to stop tropical deforestation. Coupled with appropriate, conservation oriented government policies the following objectives can be achieved at the same time:

Increase food and fiber production by farmers now practicing shifting cultivation; Reverse the pattern of degradation of many of the lands already cleared of tropical rainforest; Preserve much of the remaining tropical rainforests with their rich genetic diversity; Alleviate emission of greenhouse gases by as much as 18%.

REFERENCES Alegre, J.C.. and P.A. Sanchez (1989). Central continuous cropping experiment. In: TropSoils Technical Report, 1986-1987, pp. 86-87. Alvim, P.T. (1982). An appraisal of perennial crops in the Amazon Basin, In: S.B. Hecht (Ed.): Arnazonia: agriculture and land use research. Centro lnternacional de Agricultura Tropical, Cali, Colombia pp 31 1-328. ArCvalo, L.A., J.R. Benites and D.E. Bandy (1983). Paddy rice in the alluvial soils of the Peruvian Amazon Basin. INIPA, Yurimaguas, Peru, 25 p. Ayarza, M.A., R. Dextre, M. Ara, R. Schaus, K. ReLtegui and P.A. Sanchez (1987). Producci6n animal y cambios en la fertilidad de suelos en asociaciones bajo pastoreo en un Ultisol de Yurirnaguas, Peru. Suelos Ecuatoriales 18: 204-208.

220

P . A . Sanchez, C.A. Palm and T J . Smyth

EPA (Environmental Protection Agency) (1990). Policy options for stabilizing global climate. Draft report to Congress. Office of Policy Analysis, EPA, Washington D.C, 45 PP. INIPA (1984). Fundamentos para el cambio tecnol6gico en la agricultura Peruana. Instituto Nacional de Investigaci6n y Promocih Agropecuaria, Lima. Peru. 18p. Ingram. J.S.I., and M.J. Swift (Eds.) (1988). Tropical soil biology and fertility (TSBF) Programme: Report of the Fourth TSBF Interregional Workshop. Biological International Special Issue 20. International Union of Biological Sciences, Paris. Kang. B.J.. G.F. Wilson and L. Pinkens (1981). Alley cropping of maize and leucanena in Southern Nigeria. Plant and Soil 63: 165-179. Melillo, J.M.. C.A. Palm, R.A. Houghton and G.M. Woodwell (1985). A comparison of two recent estimates of disturbance in tropical forests. Envrionmental Conservation 12: 3740. Palm, C.A., and P.A. Sanchez (1990). Decomposition and nutrient release patterns of the leaves of three tropical legumes. Biotropica (in press). Salazar, A., and C.A. Palm (1987). Screening of leguminous trees for alley cropping on acid soils of the humid tropics. In: Gliricidia sepium: Management and improvement. Nitrogen Fixing Tree Association Special Publication 87-01, Honolulu, Hawaii,pp. 6167 Sanchez, P.A., J.H. Villachica and D.E. Bandy (1983). Soil fertility dynamics after clearing a tropical rainforest in Peru. Soil Science Society of America Journal 47:1171-1178. Sanchez, P.A., and J.R. Benites (1987). Low-input cropping for acid soils of the humid tropics. Science 238: 1521-1527. Sanchez, P.A. and E.R. Stoner and E. Pushparajah (Eds) (1987). Management of acid tropical soils for sustainable agriculture. IBSRAh4 Proceedings No.2, IBSRAM, Bangkok, 299 PP. Sanchez, P.A. (1988). Deforestation reduction initiative: an imperative for world sustainability in the twenty-first century. Paper presented at the U.S. Agency for International Development, Washington, July 22, 1988. 11 p. Seubert, C.E., P.A. Sanchez and C. Valverde (1977). Effects of land clearing methods on soil properties of an Ultisol and crop performance in the Amazon Jungle of Peru. Tropical Agriculture 54:307-321. Smyth, T.J., and J.B. Bastos (1984). Alteraqoes na fertilidade em um latossolo amarelo rilico pela queima da vegetaqao. Revista Brasileria de Cihcia do Solo 8:127-132. Smyth, T.J., and M.S. Cravo (1989). Nutrient dynamics. In: TropSoils Technical Report 1986-1987. p. 148. Swift, M.J. (Ed.) (1986). Tropical soil biology and fertility: Inter-Regional Research Planning Workshop. International Union of Biological Sciences, Paris. Szott, L.T., C.A. Palm, P.A. Sanchez, J.M. Perez, E.C.M. Fernandes, A.Salazar, R.J. Scholes, B. Pashanasi and C.B. Davey (1990). Agroforestry systems for acid soils in the humid tropics. (in press). Wilson, E.O. (Ed.) (1 988). Biodiversity. National Academy Press, Washington. 521 p.

Note: This chapter is an adapted version of "Deforestation reduction iunitiative: an imperative for world sustainability in the 21 st ceentury", by P.A. Sanchez. In: A.F. Bouwman (Ed.) (1990). Soils and the greenhouse effect, Wiley and Sons, Chichester. Included in this publication by permission of Wiley and Sons, Chichester.

22 1

Chapter 19

MANAGING GLOBAL CHANGE BY CURTAILING EMISSION SOURCES AND CREATING NEW SINKS Richard Grantham Institut d'Evolution Moleculaire. Universitt Claude Bernard Lyon I 69622 Villeurbanne cedex, France

INTRODUCTION The greenhouse drift, or buildup of trace gases in the atmosphere (Grantham 1989b) results in increased earth warming and must be corrected to avoid environmental degradation. Such a correction means controlling and reversing the accumulation of trace gases and the deterioration of our major ecosystems. A scientific consensus is becoming firmer that correction is required without further delay, otherwise too much risk for society and the biosphere is implied. We have been building up our military "defense" and neglecting our ecosystems. Our survival depends on planning our future. There is planning for research as well as longer range planning to do for geotherapy. We know there is a global disequilibrium, that society is producing too much pollution and trace gases for a healthy physiology of the biosphere. We know that every year photosynthesis falls behind combustion and respiration by several billion tons of C02. We know that the other trace gases combined are growing even faster in atmospheric concentration than C02 is. Furthermore, each of these other trace gases (methane, nitrous oxide, tropospheric ozone, the CFCs) is a stronger greenhouse gas than CO;?, that is, each absorbs more infra-red energy per molecule. Thanks to UNEPs effort a correction has started on the CFCs. But little has been done on a global scale to control emissions of carbon dioxide, methane, tropospheric ozone or nitrous oxide. In fact, the major producers of COz have so far refused a concerted curtailment. Scientists continue to gather data, study models and make predictions. All these studies are needed. They are, however, not the only necessities. They are, more and more, telling us the same thing: it is time to begin using the data to avoid the consequences they predict. There are indeed uncertainties and differences between predictions. Ordinarily this calls for increasing the protection, not procrastination. Although, mainly due to differing estimates of the effect of clouds, there is a threefold variation among the 14 General climatic Circulation Models (GCMs) compared (Cess et al.1989), all the predictions go in the same direction, they all indicate warming. At the present rate of progress it will be several years before

222

R . Granlham

reliable predictions of global and regional climate change are available from the models (Sling0 1989). Nevertheless, we have evidence that the temperature and the sea level are rising. Most predictors say these tendencies will accelerate within the next 50 years in the absence of other strong influences. Above predictions are taken as basic assumptions for the deductions made in the course of this paper.

BACKGROUND FOR DEVELOPING CORRECTION PROPOSITIONS Predictions of the consequences of the greenhouse drift have, however, hardened in the last year or so, largely due to the groups of Hansen (Hansen et al. 1988), Schncider (1989) and Ramanathan (a.0. Ramanathan et al. 1985). These researchers and others now believe that as a direct result of trace gas buildup, the Earth can warm by as much as a half degree per decade (Strong 1989).Thus, in 50 years the global temperature could be 2.5"C higher than at present. At the same time the sea level is expected to rise by perhaps a meter (Schneider 1989). (About the only thing that is decreasing in the atmosphere is OH, hydroxyl radical, whose average tropospheric concentration is about 9 x lo5 ~ m -or~ 0.03 , ppt (Prinn and Golumbek 1990). OH is the main sink for oxidizable trace gases, especially methane). These parameters are taking on new values not seen before for over a 100,000 years (Faure 1989), although in much earlier geological periods CO2 and methane concentrations were higher (see Chapter 1). The envelope of parameter values representing a relative steady state of the Earth-system is being violated. Conceptually, our planet now lies outside this "geophysiological envelope", and is departing faster and faster from it. The uncertainty associated to this excursion is equated with higher risk and the need for some kind of geotherapy, that is, global measures leading to greater environmental security and well being, in short, correction. In addition, Raval and Ramanthan (1989) strengthen previous suggestions of a positive feedback in earth heating due to the way the greenhouse effect increases with sea surface temperature (SST). The higher temperature resulting from trace gas buildup in the air evaporates more water and the increased water vapor absorbs more infrared energy, adding to the heat captured by the trace gases. Thus, we experience a strong positive feedback due to having a large surfacc of liquid water on the planet. But the added heat is more than previously anticipated, cspecially for SST above 25°C. This extra feedback could accelerate the warming; therefore even higher temperatures over the globe than foreseen may occur in 50 years. Contrary to some earlier predictions (Mitchell et al. 1989; Reynolds et al. 1989; Robock 1989), the effect should be most prominent in the tropics (Raval and Ramanathan 1989). The tropics thus appear as a potential gigantic radiative oven (Granthan 1989b). Before discussing some possible corrective measures I want to complete

Curtailing emissions and creating sinks

223

this background by a few more points. First, I see a possibility that the climate will get drier, at least at low latitudes, in spite of the higher evaporation rate. By this I mean that the relative humidity, especially over land, could drop as the surface and atmosphere warm. This is a consequence of the greater carrying capacity of the air for water vapor as the temperature rises: For example, in going from 286K to 298K the vapor pressure of water doubles (Handbook 1987). Thus the temperature changes Raval and Ramanathan (1989) believe possible in 50 years either mean much more precipitation (if the relative humidity is conserved) or drier air. Dry air cannot be supposed to favor the productivity of the biosphere. It is more apt to favour dust bowls and desertification. On the other hand, drier air may represent an optimistic hypothesis, for if the relative humidity docs not drop with higher temperatures the increased precipitation could be hard to deal with. Another concern is the recent claim of Keeling and others that the annual rate of increase of C02 in the air is going up sharply (Pearce 1989). This, of course, makes it more urgent that curtailment be started. Some researchers do not believe much in the capacity of plants for short-term adaptation by increasing primary productivity as CO;?and temperature rise, or by rapid migration to either higher altitudes or higher latitudes (see Roberts 1989). Some banks and political entities want to industrialize the tropics. I believe industrialization of the tropics is an impractical goal that would accelerate the process of sterilizing the soil. Indeed, installation of an industrial economy in the tropics within, say, 50 years, appears impossible, especially for a nuclear-based economy. The nuclear is the most expensive energy route being considered (electricity plants fueled by coal, petroleum or biomass cost much less to build). Pressurized water reactors are becoming more costly. A fast breeder reactor of 1500 M W costs over 2.5 billion U.S. dollars (McGourty 1989). More than a thousand such plants would be needed for the nuclear to be an effective answer to C02 buildup. Where would the 2-3 trillion dollars (or pounds) come from for their installation? The USA alone would need 120 to 220 new large nuclear plants by the turn of the century (Swinbanks 1989). The USA can afford them; tropical countries cannot. The USA spends each year in subsidies to the nuclear industry: 15 billion dollars (Byrne 1988). As for England, British Nuclear Fuels has just announced in a paid advertisement that they are spending 1.5 million pounds a day at Sellafield (British Nuclear Fuels 1990). Finally, the cost of the still unresolved waste disposal problem is incalculable. At the same time, a 5°C temperature rise, say, in Kenya or any other tropical country is not livable! Half of the world’s population lives in the tropics. Unless a hidden bias has deceived the hundreds of researchers studying global change, it appears likely that tropical peoples, and possibly all of us, will depend on corrective measures for survival in the 21st century. Immediate conversion to solar geothermal and other sustainable energies is necessary in the tropics.

224

R . Grantham

THE SIZE OF THE PROBLEMS The concentration of C 0 2 in the air has gone from about 275 ppm two centuries ago to slightly more than 350 ppm at present. This means that the atmosphere has accumulated: 75 x

x 12/29 x 52 x lom = 161 x 1015g or, say, 160 Gt

of carbon, mostly from fossil fuels, since the industrial revolution. The amount accumulated is going up by 3 Gt per year. We can be relatively sure of these amounts; the 160 Gt C gained in the last 2 centuries and the 3 Gt net addition of carbon as C02 to the atmosphere each year. However, a third quantity has so far escaped us: how much extra C02 has been taken up by the ocean during this time? We can now define three linked problems in C02 correction: Compensating for the net annual anthropogenic increment, that is, the 3 Gt of atmospheric C accumulating each year. Removing part of the total industrial era buildup of 160 Gt, which is over 50 times as much as in problem 1). Total correction on a time scale of a century is unthinkable, but we may need to reduce this amount to bring the atmospheric COz load back into the Quaternary envelope, in order to insure the long-term stability and health of the biosphere. Possible correction for the amount of carbon added to the sea - which could belch back out at us once we start taking C02 out of the air (Grantham 1989a). The ocean is estimated to contain 60 times more dissolved carbon than the atmosphere (Broecker and Denton 1990), but what fraction of this is "anthropogenic" has not been determined. In the context of problem 1). consider the wood harvested over the globe annually, which is estimated at 12 Gt (Emsley 1987). This is roughly double the amount of carbon in the anthropogenic yearly net increment to the atmosphere. How much new forest would we need to replace the fossil fuels used to produce that 3 Gt atmospheric excess of carbon? Sedjo, Myers and Goreau, and I have calculated the area needed and we agree that the answer is a few million square kilometers; my estimate was 8 million k m 2 (Grantham 1987 and 1989a; Myers and Goreau 1990; Sedjo 1989). In fact we can state this in terms of replacement biomass. The idea is to substitute biomass for fossil fuels for two reasons: i) we are running out of fossil fuels and new energy sources are clearly needed for the 21st century (MacNeill 1989); and ii) new forest sequesters carbon, evcn though part of the carbon is subsequently burned (Sedjo 1989). On average a tree grows for 30-40 years and during that time it sequesters several kilograms of carbon per year. Replacing fossil fuels by biomass implies, as a first cut, doubling the biomass fuel production area. But how can we do this without encroaching on land needed for food production? An answer must be sought.

Curlailing emirsions and crealing sinks

225

Problems 2) and 3) imply the possible long-term necessity of reconverting most of the fossil fuel C 0 2 that remains in the atmosphere and sea, by photosynthesis or other processes.

RECOMMENDED CORRECTIVE MEASURES The two actions of emission curtailment and new sink creation are joined in the conservation of natural ecosystems, which is the best proven way of correcting CO;! buildup. Conserving and extending an ecosystem avoids its degradation into C02. At the same time this can maintain or enlarge a sink. For variety sake I would like to change the order of the gases to be considered. This because the other greenhouse gases are increasing faster than C02 in contribution to earth warming. Also, UNEP has launched CFC correction, which I hope all governments can be brought to support actively; so we won't talk about CFCs herc. Likewise, the hydrological cycle and water sequestering are treated elsewherc (Grantham 1987, 1989 a and b). The measures I propose here are suggestions to get correction on the other trace gases started. They are for discussion and development with a view to international agreement and application as soon as possible. Methane Sevcral steps can be taken to bring methane emissions more under control: Stop natural gas leaks, which are estimated at 1-10% of production. Clearly, one immediate thing to do is to refine these estimates. Researchers and engineers can tackle this. Ricc growing, a major source of methane, should be studied to identify possible biological aids, for example rice strain selection and methanogenic bacteria characteristics - also growing conditions to minimize methane releases. Some mussels consume CH4 (Cary et al. 1988), and might be incorporated into the paddies. Nitrogen fertilization reduced the CH4 sink of forest soils (Steudler et al. 1989). Fertilizer choice and application quantity need to be better related to production of methane and other gases. Also acid rain can stimulate methanogenesis. Sulfate fertilizer, however, may suppress methane production. The cattle population should be reduced. Ruminants (cattle, sheep, goats, buffalo and camels) have been said to number 3.3 billion (Stevens 1989). Feed changes to lower methane eructation are possible. The methane

226

R . Granlham

producing physiology needs quantitative study in the different species. People can be encouraged to eat more poultry, rabbit and fish to replace beef (which is a much less efficient human energy source than the first three foods); cereals other than rice can substitute for some of the meat. 5)

A better global budget for methane is desirable; releases of methane containing 14-C from nuclear reactors should be controlled (Wahlen et al. 1989). Wetlands, tundra and nuclear reactors require more analysis as methane sources.

Nitrous oxide 1)

Sub-surface sea water is supersaturated in N 2 0 , which is mostly made by denitrification of nitrate (Yoshida et al. 1989). One of the sources of marine nitrate is runoff, which can be controlled. Methane and nitrous oxide emissions may be linked through fertilization and runoff, as well as ocean temperature.

2)

Combustion of biomass, or coal or petroleum, creates NzO as well as CH4 and CO. These trace gases interact with other constituents of the atmosphere. We should not seek to eliminate them but to delimit and achieve their optimum concentrations.

Tropospheric ozone

Gases in the troposphere react to produce or destroy ozone according to conditions of sunlight, clouds and time of day. Ozone may be destroyed by the hydroxyl radical OH or the hydroperoxyl radical H02, and it photolyzes to make OH. Under natural conditions this astonishing chemistry produces a balance, somewhat different in each hemisphere, between the amounts of 0 3 , OH, formaldehyde and nitrogen oxides present, thereby maintaining the cleansing efficiency of the troposphere and concentrations appropriate to comfort, health and greenhouse contribution (Lelieveld and Crutzen 1990). We are, however, polluting the air and disturbing this balance. 1)

Energy economies, which must be sought, can go in the same direction as reducing ground-level ozone, a health menace. Fossil hels are probably the main reason for tropospheric ozone increase; lowering their use will therefore cut down on both C 0 2 and 0 3 production. Here again, we are not seeking to completely eliminate tropospheric ozone. The work of Lelieveld and Crutzen (1990) indicates that a small amount helps to maintain the OH level necessary for a clean atmosphere.

2)

Contributors to ozone production (peroxides, nitrogen oxides) result from

Curtailing emissions and creating sinks

227

burning biomass, coal or petroleum products. Combustion standards and counsel could be made available to lower concentrations of these compounds.

Carbon dioxide After water vapor, C02 is the most important single greenhouse gas. As seen above a huge amount of it has been accumulated. What can be done? We are continuing to burn fossil fuels and inject C02 into the atmosphere. This net loss of sequestered carbon will stop in the 2 1 s century because fossil fuels will be largely exhausted. If we do not anticipate this exhaustion, widespread suffering and economic perturbation will result. Photosynthesis is the only proven process foreseeable for resequestering carbon to restore atmospheric balance. All major ecosystems, which perform natural photosynthesis, should be cared for and extended (Grantham 198 9b) . Examples of replacement energies for coal and petroleum are: Hydraulic; Thermal; Eolian; Solar; Biomass; Hydrogenated fuels (H2, biogas and alcohols); Nuclear. Only the last of these possible replacements has had generous research and development funding. However, because of the enormous installation and decommissioning costs, the danger of weapons proliferation and the waste disposal dilemma, nuclear power appears unattractive in developing countries. The other energies merit development. The mass of soil carbon is twice that of atmospheric carbon (De Groot 1990). The chemical state and the respiration, decomposition and other reactions of soil carbon should be studied for a global synthesis. Many soil organisms, as well as plant roots, store carbon for varying periods of time. In general, there is wide agreement that afforestation is the surest route to CO2 correction (Grantham 1989 a and b; Myers and Goreau 1990; Sedjo 1989). New mixed species forests, from which individual trees would be selectively harvested, are what most of us prescribe. That way the forest is never destroyed; mature trees are felled individually and new young trees replace them, closing the

228

R. Grantham

canopy again. I would add that savannas and grasslands, due to their underground carbon mass, are already a large scale sequestering mechanism. They, like corals and mangroves, require care, that is conservation, and extension to assure a strong role in bringing the atmosphere in equilibrium by reducing its COZload. Of course the other approaches will still be needed since these natural ecosystems cannot possibly absorb all the COz load within the next century (Grantham 1989a).

CONCLUDING REMARKS It is folly to believe that the minimum predictions will be those realized but some humans have this tendency. Even if only the low predictions are realized, controlling the greenhouse drift will still be necessary. As for starting correction, does i t really matter whether the temperature will rise by 2.5"C or 5°C or 7.5"C upon doubling of the C02? The answer is no, we are overdue in undertaking correction in any case. Some people also fear that the consequences of correction may be worse than what the modelers are predicting. I have chosen measures that would be desirable to improve the state of the biosphere even if no greenhouse imbalance existed (Grantham 1989b). Nevertheless, these measures would require wide agreement, they have not been sufficiently discussed. This is because correction has not been taken seriously enough. Soil degradation, desertification and pollution all need to be corrected with the greenhouse drift. We will be correcting from here on, it is not a temporary task. Correction could, and 1 believe should, lead us into developing a conscious evolutionary strategy - for our species and for the biosphere. I believe such a strategy is necessary to insure the durability of the earth system with its humans! To finish, here are the most urgent things to do in my opinion: 1)

Notwithstanding remaining differences on details in any assessment, a short draft "Correction Request", or "Geotherapy Mandate" should be formulated, to be sent to all governments and international organizations. This document could be signed by responsible persons of a special task group from UNEP, the ISSS and the Geotherapeutic Approaches Workgroup of the INQUA Committee on Global Change.

2)

Try to get some immediate corrections moving: leak study and budget clarification on methane, conservation of corals and mangroves, massive afforestation, energy economy steps, etc. Identify countries and organizations friendly to remedial action and encourage joint efforts.

3)

Follow the Correction Request by a more developed list of recommended measures and create a Regional Action Plan, which would suggest: what can be done in each world region. To whom the Plan would be formally

Curlailing emissions and crealing sinks

229

addressed is something which can be discussed if we agree on its general desirability. 4)

If the three above points are accepted, the very first step is to set up and seek financing for the Special Effort on Geotherapy.

REFERENCES Broeker. W.S. and G.H. Denton (1990). What drives glacial cycles? Sci. Am, Jan, p.49 British Nuclear Fuels (1990). Just how green are you about nuclear power? New Scientist 13 Jan, p. 41. Byme, G. (1988). The energy index. Science 242, 1639. C a y , S.C., C.R. Fisher and H. Felbeck (1988). Mussel growth supported by methane as sole carbon and energy source. Science 240.78. Cess, R.D. et al. (1989). Interpretation of cloud-climate feedback as produced by 14 atmospheric general circulation models. Science 245, 513. De Groot, P. (1990). Are we missing the grass for the trees? New Scientist 6 Jan, p.29. Emsley, J. (1987). Plant a tree for chemistry. New Scientist 8 Oct. Faure, H. (1989). Le cycle du carbone, le climat et I'homme, unpublished conference. Lyon, 1 1 Oct Grantham, R. (1987). Castles in the Saharan air. Nature 325, 384. Grantham, R. (1989a). Cogene. IGBP Report 7:2, p.259-264. Grantham, R . (1989b). Approaches to correcting the global greenhouse drift by managing tropical ecosystems. Tropical Ecology 30, no 2, 157-174. Handbook Chemistry and Physics (1987). 68th ed, CRC Press, p. D-190 Hansen, J. et al. (1988). Global climate changes as forecast by Goddard Institute for Space Studies three-dimensional model. J. Geophys Res 93, D8, 9341. Lelieveld, J. and P.J. Crutzen (1990). Influences of cloud photochemical processes on tropospheric ozone. Nature 343, 227 MacNeill, J. (1989). Strategies for sustainable economic development. Sci Am Sept, p.155. McGourty. C. (1989). Europe's fast reactor plan inches forward. Nature 337, 680. Mitchell, J.F.B., C.A. Senior and W.J. Ingram (1989). C 0 2 and climate: a missing feedback? Nature 341, 132. Myers, N. and T.J. Goreau (1990). Tropical forests and the greenhouse effect. Climatic Change (in press). Pearce, F. (1989). Felled trees deal double blow to global warming. New Scientist 16 Sept, p.25 Prinn. R.G. and A. Golumbek (1990). Global atmospheric chemistry of CFC-123. Nature 344, 47. Ramanathal. V. et al. (1985). Trace gas trends and their potential role in climate change. J. Geophys. Res. 905547-5566. Raval, A. and V. Ramanathan (1989). Observational determination of the greenhouse effect. Nature 342, 758. Reynolds, R.W.. C.K. Folland and D.E. Parker (1989). Biases in satellite-derived sea-surfacetemperature data. Nature 341, 728. Roberts, L. (1989). How fast can trees migrate? Science 243, 745. Robock, A. (1 989). Satellite data contamination. Nature 341, 695. Schneider, S.H. (1989). The changing climate. Sci Am Sept, p.70.

230

R . Granlham

Sedjo, R.A. (1989). Forests a tool to moderate global warming? Environment 31, no 1, 14. Slingo, A. (1989). Wetter clouds dampen global greenhouse warming. Nature 341, 104. Steudler. P.A. et al. (1989). Influence of nitrogen fertilization on methane uptake in temperate forest soils. Nature 341. 314. Stevens, W.K. (1989). Methane from guts of livestock is new focus in global warming. New York Times 21 Nov, Environment. Strong, A.E. (1989). Satellite data contamination. Nature 341, 695. Swinbanks, D. (1989). Nuclear power industry struggles on, despite opposition. Nature 338, 190. Wahlen, W. et al. (1989). Carbon-14 in methane sources and in atmospheric methane: the contribution from fossil carbon. Science 245, 286. Yoshida, N. et a]. (1989). Nitrification rates and 15N abundances of N20 and NO3 in the Western North Pacific. Nature 342, 895.

23 1

Chapter 20 IMPLICATIONS FOR AFRICAN AGRICULTURE OF THE GREENHOUSE EFFECT Richard S. Odingo Department of Geography University of Nairobi Nairobi, Kenya.

ABSTRACT Africa is the one continent which still relies very heavily on agriculture to feed its fast growing population because industry, though significant in a few patches in the North and in the South, is still in its infancy. African agriculture is still largely traditional although important facets of it are gradually coming under increased scientific management. As it is, agriculture is very sensitive to climate which is marked by fluctuations and incessant variability. Frequent and prolonged droughts and dessication are already a threat to agricultural production. Thus the risk of COz induced global warming as part of the greenhouse effect would naturally be an added burden to be coped with. At present the projections of global warming produced by the various GCMs suggest that much of tropical Africa will remain warm and may be up to l 0 C warmer, but that the subtropical parts of the continent could experience a more significant warming of up to 1.5"C. Contrary to the expectations, global warming in Africa may be accompanied by a Northward shift in the rainbelts bringing more rainfall to the hitherto parched desert lands of the Sahara in the North and the Kalahari in the South, making it possible to carry out some form of agriculture in these regions. But not enough is known of the actual likely distribution of rainfall on a seasonal and annual basis and how agriculture in these lands will respond to the changed conditions. Certain crops like wheat and corn associated with the subtropical latitudes may suffer a drop in yield due to increased temperature on the one hand and rice may may disappear due to higher temperatures in the tropics. Elsewhere agriculture is expected to survive and even become stronger especially where mixed cropping is currently practised and where tree crops are predominant. The high altitude farming districts may have their altitudinal zonation wiped out and be forced to find new forms of agriculture. However, methods of adjustment could be adopted to cope with climate change including the use of new seeds, drought resistant crop varieties and the greater use of irrigation where water availability becomes a constraint to agricultural production. In general Africa should be in a position to survive a global warming by introducing such adjustments and, by making agricultural land use more flexible.

232

R.S. Odingo

INTRODUCTION The African continent relies heavily on agriculture to support its rapidly growing population. Industry is still relatively insignificant except in South Africa, and restricted parts of the Mediterranean Region such as Egypt and some of the Maghreb states, but the total contribution to the greenhouse effect in Africa as a whole is very small. Forests cover large parts of tropical Africa, but in almost all cases they are under severe pressure leading to large scale felling for timber as well as clearing for agriculture, and this is thought to be adding to the global warming through the massive release of greenhouse gases, chiefly (202. As is well-known, large parts of the continent are either arid (as in the Sahara and Kalahari Deserts), semiarid (including all the "Sahelian" lands - with the "Sahel" as a generic term) or subhumid. The continent has a long chequered history of droughts and dessication, which have been marked by severe famines since time immemorial, and these should be considered a good dress reheanal for the global warming if it ever comes. All the same, it is fair to say that much of Africa will be peripheral to the greenhouse induced global warming because it is already either tropical or subtropical whereas the greatest impacts of such warming will be experienced in the temperate and polar latitudes. What is known about African agriculture, if indeed it could be generalised, is that it depends heavily on rainfall, hence on the significance of climatic fluctuations as well as the possibility of climate change because these could exacerbate the already severe conditions of drought which have been typical of the Sahelian lands of West Africa and other semiarid and sub-humid regions throughout the continent. In this context the subtropical parts of the continent both in the North as well as in the South are different because these are the regions where crops sensitive to significant temperature variations such as wheat, barley and oats are grown, and which will be affected by large temperature shifts as predicted in global warming models. The highland areas in tropical and subtropical Africa will fall into the same category as the subtropics so that significant temperature shifts could disrupt existing agricultural patterns which are currently strictly zoned according to altitude induced temperature differences. African agriculture as a whole is arranged according to agrcecological zones (FAO, 1982) which are climate sensitive. Any drastic changes as those likely to be brought about by global warming could be extremely disruptive in terms of the crops which could no longer be grown, and other farming systems which could no longer be sustained. Areas of the continent which rely on imgation, though not very extensive, are significant, and cover the densely populated areas such as the Upper and Lower Nile Valley in the Sudan and Egypt, and large tracts of the Mediterranean lands of Northern Africa (the Maghreb lands including Morocco, Algeria, Tunisia and Libya). Greenhouse induced global warming could bring about important changes in the agricultural potential of such areas, and it is therefore interesting to study the likely nature of climate changes to be expected so

African agriculfure and the greenhouse eflecf

233

that appropriate adjustments can be made. In the event of climate change, conditions suitable for continued agricultural production could change in two directions, either adverse or more conducive, depending on the particular circumstances of a given region or subregion. Changes in precipitation will largely be expressed in terms of changes in soil moisture; similarly changes in temperature will be reflected in seasonal changes in evapotranspiration rates and ultimately in soil moisture and changed conditions for crop growth. Unfortunately most of the projected changes are better understood at the experimental level (i.e. in the laboratory) than at the country level, let alone, at the continental level. Nevertheless, it is instructive to study the projected climatic changes even if they remain theoretical, because with these projections it is possible to learn how African agriculture might respond to the new conditions.

WHAT THE GCMS SAY ABOUT GREENHOUSE WARMING IN AFRICA In a short article such as this one it is not possible to go into too much detail about the various climate models and what they say or do not say about the expected conditions for Africa. Nevertheless, it is possible to summarize the main arguments put forward, and their relevance to the African scene. To begin with most GCMs are generally vague and even more vague about the tropics. Part of the reason for the vagueness about the tropics, and about the tropical Africa in particular, is due to the paucity of climatological data upon which to base their analyses. Secondly, it is true to say that the greatest sources of C02 emission (industrial) are located in the temperate latitudes of the Northern Hemisphere, and that simulations of global warming indicate that it is these same latitudes which will also be the most affected, and that the effects will spread polewards (see Hansen et al. 1981 and 1988; Manabe and Whetherald 1986). Over a 50-100 year period, global mean annual temperatures are expected to rise by from 1.5"C to 5°C assuming that by that time global C 0 2 levels will have doubled (2 x C02). Practically all the models agree on the fact that the warming will be more pronounced at higher than lower latitudes, and that the warming will probably be greater in winter than in summer (Parry et al. 1986 and 1990). If one looks at the conclusions of the main GCMs, namely the CCM, GFDL, GISS and OSU, medium USA temperature increases are of the order of 3.OoC, 5.6"C, 3.8"C and 3.5"C respectively for a doubling of CO;! and accompanying rainfall scenarios are more uncertain than temperature change scenarios and may vary considerably over regions. For example, under some scenarios there is a likely decrease rather than an increase in precipitation with increased temperatures. If additional GCMs are considered for example NCAR (Washington and MeeN 1984) and UKMO, there is an even more categorical statement that expected global mean temperature rises will be around 2°C in the tropics and 5°C in higher latitudes. Inevitably, all

234

R.S. Odingo

the models are more refined when commenting about temperature changes than they are about rainfall and other climatic parameters which are relevant to agriculture such as cloud cover, radiation, soil moisture and air movement. So far few categorical statements are found in the models about what temperature rises will look like in the tropics and what the rainfall distribution will look like in specific regions such as tropical Africa. As a result of this vagueness in the GCMs it may be of some assistance to look at analogue data or historical evidence in areas with poor or broken records such as in Africa.

HISTORICAL EVIDENCE OF CLIMATE CHANGE IN AFRICA The most comprehensive evidence of climate change in Africa over the last 4,000years has been put together by Sharon Nicholson (1976, 1978, 1979, 1980 and 1981). It is not possible to summarize all her contributions in a short paper, but her work contains a more detailed record of rainfall variability and general climate variations in various parts of tropical Africa over the period covered. A detailed record is provided of droughts, enhanced or depleted lake levels and river gaugings, which leave no doubt about the dramatic changes that have been experienced. The historical information on famines, droughts, lake and river levels in Africa has been supplemented by dendrochronological evidence from other parts of the continent, and they all confirm the variability as well as longterm climatic fluctuations which are continuing into the present era. Indeed, if the C02 induced global warming occurs in large parts of and and semiarid Africa, one will not be taken by surprise because the droughts, famines and dessication have already provided a dress rehearsal for what may be expected and the new conditions may not necessarily be much worse than the experience of for example the 20 year drought in the Sahel. Other scholars have equally elucidated the historical evidence of climate change and climatic fluctuations in Africa. Most authors agree on the frequency of prolonged droughts and dessication some of which may have a return period of 30 to 50 years. From the point of view of agricultural performance it is clear from the available evidence, in particular of droughts, that much of tropical Africa has already had to cope with so much rainfall variations that even the projected changes, projected under C02 induced global warming will not necessarily be worse. If anything, the arrival of the presumed global warming might bring wetter conditions with it for many currently semiarid areas like the Sahel in West Africa and even for parts of the Sahara and Kalahari deserts which have not experienced wet conditions for centuries. Kellogg (1977) and Kellogg and Schware (1981), two of the earlier researchers on C02 induced global warming, collected some interesting analogue evidence for Africa from more than 100 sources which were used to construct a map showing what the conditions looked like during the Altithermal Period

African agricullure and the greenhome effect

235

(4000-6000 BP) in terms of areas which were then "wetter" or "drier" than at present. For tropical Africa, and similar areas like the Rajasthan Desert in India, conditions then were decidedly wetter than they are today (see Fig. 20.1).

Fig. 20.1

Rainfall distribution in summer during the Altithermal period of 4000 to 6000 BP (after Kellogg and Schware (1981)

If the Altithemal experience could be reproduced for Africa during the C02 induced global warming of the next 50-100 years, even large areas currently unable to support agriculture because they are too arid might become very attractive for the first time in several centuries. The map in Fig. 20.1 is best interpreted if compared with Fig. 20.2 which portrays the current mean annual rainfall1 on the continent, and the map showing the co-efficient of the variability of annual rainfall for Africa given in Fig. 20.3. The rainfall in many parts of Africa has been marked by violent fluctuations, interannual variability and general lack of dependability. There is nothing to suggest, even with changed climate as is postulated in global warming, that the characteristics of the climate of Africa will be any different. Thus, even if global warming were to bring more rainfall to the hitherto parched lands of the Sahara Desert and its fringes, agricultural planners will still have to cope with both seasonal and interannual variability of rainfall. In the context of the current climate, researchers in Africa still have very little to go on since the signals from the GCMs seem to be much clearer for areas outside the tropics, more towards the polar latitudes, especially on the Northern Hemisphere. What is known and fairly well established is the fact that the effect of increased CO;! levels on the atmospheric circulation pattern and hence on the temperature and precipitation distribution, will result in a general northward movement of agroclimatic zones (Bach 1978). If this shift will occur, there is

236

R.S. Odingo

some general agreement that tropical agriculture may suffer the least (Bach 1978), but increased drought risk on the fringes will occur.

Fig. 20.2

Annual mean raigall. Unit: 100 mm. Dark shaded area: > 1600 mm. Light shaded area: 200-800mm. Cross marks: rainfall observation stations. Thin broken line: 1OOOmm contour (derived from Nicholson African data set, in Makomura (Ed.)1989)

African ogriculiwe and ihe greenhowe effecl

237

CLIMATIC CHANGE AND LAND USE AND AGRICULTURE IN AFRICA Before tackling the more specific impacts of climate change on the African scene in general, and global warming in particular, it should be instructive to work at the general picture of land use on the continent. The current land use is very much governed by water availability whether this be in the form of rainfall, or in the form of water available for irrigation.

0

Fig. 20.3

UOE

Coefficient of the variability of annual rainfall (standard deviation of annual totalslannual mean). Unit: %. Shaded area: 20-40%. (derived from the Nicholson African data set, in Kadomura (Ed.) 1989)

238

R.S. Odingo

Fig. 20.4 is a generalised map showing the extent of water resources which can support agriculture on the continent, including irrigated areas, areas with adequate rainfall and the semiarid, and and extremely arid areas where water for crops becomes a problem progressively. In this map one cannot fail to notice the pcrennial problem areas where inadequacy of rainfall coupled with the lack of imgation water have resulted in the impossibility of any meaningful form of agriculture. Fig. 20.5 illustrates the expected response to the water situation which has evolved over centuries. I

I

I

Water ulitization for agriculture ze%zydams,

@

canals

...........

Main irrigation schemer

Water resources in relation to present economic requirements Areas with adequate rain

Semiarid areas

Fig. 20.4

Water availability for agriculture

239

African agriculture and the greenhouse effect

I

Fig. 20.5

I

Lund use in Africa

This land use map shows areas where arable land is in abundance (largely because water for agriculture is available), pasture lands, tropical grazing (usually associated with the savannahs with adequate rainfall), dry steppe and semidesert grazing (where the shortage of moisture keeps out agriculture), the deserts , and the closed forests. The foregoing section has briefly sketched the current climatic background to agriculture in Africa. It is also possible to go into detailed agro-climatological zoning of the whole continent, as has been carried out by the United Nations

240

R.S. Odingo

Food and Agriculture Organization (FAO) and the International Institute of Applied Systems Analysis (IIASA) who prepared an Agro Climatological Map of Africa and followed this with an analysis of the potential population supporting capacities of the various agroecological zones for tropical Africa (FAO/UNFPA/IIASA 1986). The point that needs emphasising is that currently Africa is relying rather heavily on rainfed agriculture. However, in the projected climate change scenarios, there is likely to be increased availability of rainfall, hence soil moisture even if this is not evenly distributed regionally. It therefore now only remains to comment on the likely impacts on agriculture of the changed conditions which are expected to accompany global warming.

TYPES OF AGRICULTURE IN AFRICA AND THE LIKELY EFFECTS OF GLOBAL, WARMING Thc likely effects of global warming on African agriculture are expected to be uneven even if one were to assume a generally increased rainfall because the current agriculture is already sensitive to temperature in two particular subregions and subzones, namely: The subtropical parts of Africa in the "Mediterranean North" and "Mediterranean South" are currently referred to as areas of Mediterranean agriculture with cereals like wheat, orchards, vineyards and specialized crops which are all sensitive to even minor temperature changes and therefore likely to be impacted by the envisaged temperature increases ( I S O C ? ) under a global warming. High altitude areas within the tropics where currently temperate crops (including cereals like barley and wheat) are grown because of the reduced temperatures brought about by altitude.

For the rest of Africa and especially for tropical Africa current agricultural practices are likely to shield the agriculture from the adverse effects of a changed climate, in particular a warmer climate. Fig. 20.6 which shows the main types of agriculture in Africa brings out certain important features. There is for instance a preponderance of mixed cropping which is thought to be advantageous even in a changed climate because some crops will benefit from temperature increases where others suffer. In general, areas dominated by tropical crops are expected to bcnefit rather than suffer from increased temperatures. Even more important is the fact that the GCMs do not envisage very large temperature increases (1S"C) for these tropical areas. It is therefore somewhat hypothetical to try and guess how the agriculture is likely to respond to a changed climate where the possible change is expected to remain within the ranges which have always been experienced, at

African agriculture and the greenhouse effect

24 1

least in the last 200 years. In the final analysis each crop will have its own unique response to changed or changing climatic conditions and it is here where some instructive information may be sought. Unfortunately, the bulk of the studies on these aspects has been carried out in temperate latitudes, so one can only infer what is likely to transpire in the tropics and subtropics where similar crops are found, as will be summarised briefly in the next section.

Fig. 20.6

Types of agriculture in Africa

242

R.S. Odingo

CLIMATIC EFFECTS ON SELECTED CROPS RELEVANT TO THE AFRICAN SITUATION Bach (1978) briefly analysed the effects of rainfall and temperature changes on corn (maize in Africa), and these are summarized in Figs. 20.7 and 20.8. In these illustrations it is shown that corn is very sensitive to both temperature and rainfall changes particularly during those months (July and August) where the combincd effects of temperature and rainfall are critical to their growth. The accompanyingprojected yield changes linked to temperature and precipitation are also summarised in Table 20.1 obtained from Bach (1978). According to this table expected changes would be significant enough to depress corn yields by up to 11%. 85

-

70-

I

I

I

I

I

I

-

I

*NORMAL I

I

I

66 68

I

70 72

I

I

I

76 78 80 82

74

Degrees Fduenhcit

Response of corn yields to temperature in the US corn belt (from Biggs a d Bartholic 1973, as cited in Bach 1978)

Fig. 20.7

70

I

0

I

I

I

1

1

I

I

1

2

3

4

5

6

7

I n c h or R.inrau

Fig. 20.8

I

8

hch hionih

Response of corn yields to rainfall during diferent stages of growth in the US corn belt Cfrrom Biggs and Bartholic 1973, as cited in Bach 1978)

African agriculture and the greenhouse effect

243

Table 20.1. Estimatedpercent change in corn yield due to changes in temperature and precipitation

Temperature Change in precipitation (% of normal *) change,"C -2 -1 0 +1.0 +2.0

-20 19.8 8.4 -2.9 -14.2 -25.6

-10 21.2 9.8 -1.5 -12.8 -24.2

0 22.7 11.3 0 -9.8 -22.7

+10 24.2 12.8 1.5 -8.4 -21.2

+20 25.6 14.2 2.9 -8.4 -19.8

Source: Benci et al. (1975) quoted in Bach (1978) p.155 * Normal = 85+16 bu/acre. 1901-1972 average for selected stations in Missouri, Dlinois, Indiana, Nebraska, Iowa and Kansas.

Two other crops which are also grown abundant in Africa and for which there are research results from other parts of the world are wheat and rice. Fig. 20.9 and Fig. 20.10, obtained from Yoshida (1978) and Asama (1976) respectively, show the intimate link between temperature changes and rice productivity, and between day/night temperature changes and wheat productivity. There is no doubt whatsoever that changed temperatures such as those exptected under a global warming will be highly significant for continued rice and wheat production with available evidence pointing clearly in the direction of possible drastic declines in the yields of these two key crops.

21

20.9

1

I

27

29

Effect of increasing temperature on (he productivity of rice at different rates of radiation Cfrom Yoshida 1978)

244

15

R.S. Odingo

-

r

V0-

d

P40-

f

0 35-

Q$

30

-

b DIADEM

DAY/NIQHt TEMPERATURE ('C)

Fig. 20.10

Effect of increasing temperature on grain development in wheat (from Asana 1976)

As far as the rice crop is concerned, it has hitherto been assumed that it should thrive under conditions of high temperatures, so that global warming could even be conducive to a higher rice productivity. Unfortunately, this is untrue. Sinha et al. (1988) have shown that rice yields tend to decrease markedly as temperature rises above 28°C (see also Fig.20.9). Table 20.2 based on Stansel and Huke (1975) and quoted in Bach (1978) gives some indication of expected percent deviation from world rice production when influenced by changes in temperature and precipiation, and from this it can be seen that global warming will bring with it many conditions, some of which will be clearly undesirable, even for Africa where rice cultivation though still insignificant has been increasing in the last few decades. Table 20.2 Percent deviation from world rice production, as influenced by changes in temperature and precipitation* P change Temperaturechange Total P in % in "C change in % -2" -lo -0.5" +0.5" +lo +2" % -15 -19 -13 -8 4 0 3 -8 -10 -17 - 1 1 -6 -2 2 5 -6 -5 -13 -7 -2 2 6 9 -2 +5 -9 -3 2 6 10 13 2 +10 -5 1 6 10 14 17 6 +15 -3 3 8 12 16 19 8 Temp. -11 -5 0 4 8 11 change

* Based on world production of 300 Mt.

Source: Stansel and Huke (1975)

African agricullure and the greenhouse effecr

245

As far as agriculture in Africa is concerned, it is important to underline the fact that it is not the absolute magnitude of a climate change that will determine the nature of the effects as much as the change in climate relative to the existing or baseline conditions. For example, in tropical Africa, in theory, there will be a lengthening of the potential growing seasons in areas where it is now short due to the latitudinal as well as the altitudinal position. But in other cases there might even be a shortening of the required growing period because of local soil moisture and altitudinal circumstances. Such changes will also affect the maturation period of certian crops differently. As far as changes in crop yields are concerned, moisture, and especially soil moisture has been paramount in tropical and subtropical Africa. Unfortunately, a look at the GCMs does not provide consistent or confident estimates (Parry et al. 1986 and 1990) of regional precipiation changes. Second to rainfall (and hence soil moisture) is temperature and higher temperatures may ultimately rule out some crops alltogether. The work by Sinha et al. (1988) already quoted shows that beyond 28OC, rice yields for instance will be depleted. In the final analysis, if the climate changes projected are severe enough, the present crop cultivars may cease to be suitable for the changed circumstances. Fortunately for Africa, according to current GCMs, no conditions envisaged should be so severe as to rule out many crops. Much of tropical Africa has tended to rely on crops such as cocoa, rubber, oil palm, coffee and tea. Some of these crops should be able to absorb increased temperature if there is a corresponding increase in rainfall. As a result, higher yields may even be anticipated, but one is not always sure how the crops will react because of the lack of specificity in the projected conditions. What is known for certain is that climate change will first and foremost be felt in soils, and in turn in plant growth through changed photosynthesis. For those areas in Africa where there is currently high altitude farming (a.0. tea and coffee), global warming may bring drastic changes. It is possible that the current altitudinal zoning (agroclimatological zoning) may be seriously disoriented by increased temperatures, so that crops which now grow well at the higher altitudes (temperate crops) may totally disappear. Experimental results in Africa as to what to expect in crop plant performance under the changed soil moisture and temperature conditions are few and it may be safer to make some inferences from work in the tropics in other parts of the world. For example, it is fair to assume that under increased C02 there should be an increase in leaf photosynthesis, but it is not clear how this will affect African plants. In general carbon dioxide fertilization effects can partially offset the adverse climate, particularly for C3-plants (see also Chapter 13 by Sombroek), but i t would be unwise to rush to conclusions before field and laboratory experiments are carried out in Africa to establish the expected scientific results. According to Sinha et al. (1988) the higher yields in C3-crops are obtained

246

H.S. Odingo

around a daily mean temperature of 15°C and in C4-crops around 30"C, but it is necessary to accompany such generalisations with more detailed information on physiological requirements of each individual crop.

AGRICULTURAL ADJUSTMENTS TO COPE WITH GREENHOUSE WARMING Much of the agriculture practised in Africa is operated at a low to medium scientific level, so that even weather advises are rare. Because of this it would be unwise, if not impossible, to suggest very precise adjustments which could be made to cope with expected changes in climate. Among the existing options for adjustment, which are already being practised and some of which could be brought into full practice in the event of a climate change, would be the following: Germplasm improvements for drought-tolerance (this is already being implemented in Kenya with Katumani Composite Maize and other dryland maize composites like Taboran Composite in Tanzania), or for heat stress resistence could be undertaken; Appropriate research should now be commenced on simulated situations such as crop production responses to high temperatures to establish suitable germplasm requirements for the changed conditions; There is already the possibility for the substitution of crop species e.g. pigeon peas (Cujunus cujun) for soy beans (pigeon peas are already a very successful crop in the dryland areas of Kenya); Agricultural management changes could be introduced including multiple cropping already much practised in much of tropical Africa, and early planting, already a successful weather advisory practice in some countries like Kenya and Uganda; Africa (tropical) already has good experience with tree crops (e.g. coffee, cocoa, tea, oil palm etc.) and this should be the focus of intensified research to understand their place in a changed climate.

SUMMARY AND CONCLUSIONS The likely impacts of a changed climate due to increased levels of (maily) C02 are still difficult to project, largely because the models of change are still rather vague. That there will be significant effects on African agriculture cannot be doubted. There is evidence of some existing resilience in African agriculture which could enable it to cope wilh a changcd climatic situation, but it is quite clear that more intensive research will be required at national, regional, and global levels. to establish more accurate scientific facts which can form the basis for

African agriculture and the greenhome effect

247

more informed policy formulation.

REFERENCES Asama, R. (1976). Plupiological approach to breeding of drought resistant Crops. ICAR Technical Bulletion, New Dehli. Bach, W. (1978). The potential consequences of increasing C 0 2 Levels in the atmosphere. 1. Williams J (Ed) (1978). pp. 141-167. F A 0 (1982). Methodology of agroecological studies in Africa. Rome. F A 0 FAO/UNFPA/IIASA (1986). Potential Population Supporting Capacities of Lands in the Developing World. Rome. FA0 Hansen, J. et al. (1981). Climate impact of increasing atmospheric carbondioxide. Science 213:957-66. Hansen, J., A. Lasis, D. Rind, S. Lebedeff, R. Ruedy and G. Russel (1988). The greenhouse effect: projections of global climate change. In: Effects of changes in stratropheric ozone and global climate. Nairobi, UNEP and EPA Kellogg, W.W. (1977). Effects of human activities on global climate. WMO Technical Note No. 156. WMO. Geneva. Kellogg, W.W. and R. Schware (1981). Climate change and society -consequences of increasing carbondioxide. Boulder Colorado, Westview Press. Manabe, S. and R.T. Wetherald (1986). Reduction in summer soil wetness induced by an increase in atmospheric carbondioxide. In: Effects of changes in stratrospheric ozone and global climate. Nairobi, UNEP/EPA. Nicholson, S.E. (1976). A climatic chronology for Africa: synthesis of geological, historical, and meteorological information and data. Unpublished Ph.D. dissertation, University of Wisconsin, Madison, 324 pp. Nicholson, S.E. (1978). Climatic variations in the Sahel and other African Regions during the past five centuries. Journal of And Environments. 1:3-24. Nicholson, S.E. (1979). The methodology of historical climate reconstruction and its application to Africa. Journal of African History. 20:3 1-49. Nicholson, S.E. (1980). Saharan climates in historic times.In: The Sahara and the Nile. M.A.J. Williams and H. Faure (Eds). Balkema Rotterdam. Nicholson, S.E. (1981). The historical climatology of Africa. In: Climate and history. T. Wigley. M. Ingram, and G. Farmer (Eds). Cambridge University Press, Cambridge. Pany, M.L. and T.R. Carter (1986). Effects of climatic changes on agriculture and forestry. In: Effects of changes in stratropheric ozone and global climate. Nairobi, UNEP. Parry, M.L., T.R. Carter and N.T. Konijn (Eds) (1990). The impact of climatic variations on agriculture. Vols 1 & 2, Reidel, Dordecht.. Sinha S.K., N.H. Rao and M.S. Swaminathan (1988). Food security in the changing global climate. In: Conference Proceedings on the Changing atmosphere. WMO, Geneva, pp. 167- 191. Stansel, J. and R.E. Huke (1975). Rice. In: Impacts of climatic change on the biosphere, CIAP Monograph 5 . pt. 2, Climatic Effects 4-90 amd 4-132. Washington, W.M. and G.A. Meehl (1984). Seasonal cycle experiment on the climate sensitivity due to a doubling of C 0 2 with an atmospheric general circulation model coupled to a simple mixed-layer ocean model. J. Geophys. Res. 89:9475-9505.

248

R.S. Odingo

Williams. J. (Ed). (1978). Carbondioxide, climate and society. Oxford, New York, Toronto, Sydney, Paris, Frankfurt, Pergamon Press. Yoshida. S. (1978). Tropical climate and its influence on rice. IRRI Research Paper Science 20. IRRI. Manilla.

249

Chapter 2 I

THE AGRICULTURAL ENVIRONMENT IN LATIN AMERICA AND THE CARIBBEAN AND THE GREENHOUSE EFFECT N . Ahmad Department of Soil Science The University of the West Indies St. Augustine. Trinidad, West.Indies

ABSTRACT The Latin American region is diverse, representing a land mass of over 20,340 million km2. The area is not densely populated and only about 7% of it can be considered too dry for agriculture. There is a wide range of soils with all the orders represented, the six major ones are Ferralsol/Oxisols. Acrisols/Ultisols, Luvisols/Alfisols, Cambisols/ Inceptisols, Fluvisols/Entisols, and Phaeozems/Mollisols. The region is well known for very high levels of management for several perennial crops such as coffee, banana, citrus, cacao, oil palm and sugar cane. These individual crops probably approach forest conditions in terms of total photosynthetic activity and no serious fires are involved. The projected climatic change would lead to changes in rainfall distribution in different parts of the region which would necessitate a high level of technology in agriculture. The effect of the projected sea-level rise could be disastrous in several areas. Subsistence agriculture in the region is based on slash and bum and shifting cultivation with little capital inputs. Depending upon the soil nutrient reserves, the effect of this treatment on the soil is very variable. On the Oxisols, Ultisols and Alfisols of the Amazon basin with a total estimated area of nearly 7 million km2, soil deterioration is extremely rapid due to low reserves of fertility; shifting cultivation leads to large annual clearings of virgin forest, estimated at present at about 10 million ha. The burning which takes place probably adds about 625 million tons of CO, to the atmosphere annually besides reducing the capacity of the vegetation to absorb CO, in photosynthesis. The total contribution can be quite important in the global greenhouse effect. The cultivation of rice which is elsewhere associated with increased concentrations of methane does not have this objection in Latin America since over 75% of the crop is grown in upland conditions. On the other hand, the continually rising pasture fed cattle population throughout the region with subsequent methane production could be an important factor in the overall greenhouse effect. Ways to improve land use and crop production without increasing the concentration of greenhouse gases and adverse climatic change have been suggested.

INTRODUCTION The countries of South and Central America cannot be regarded as densely populated and the area has the highest percentage of land still under forest (63%) compared to the world average of 58% (Espig 1989). From areas with a suitable climate for crop production, only 6% are actually cultivated, 23% is under pasture

250

N . Ahmad

and 46% under forests. Potentially, 35% could be cultivated, 40% can be used for pastures and/or forests and 25% should be reserved. The Caribbean area is an outstanding example within the region of an area having a high population density and hardly any unused land resources. In this paper, the possible effects of climatic change within the region on agriculture and land use are discussed. The present agricultural environment, the ways in which it contributes to the amount of greenhouse gases and global warming and what can be done to alleviate the problem are presented.

POSSIBLE EFFECTS OF GLOBAL WARMING TO AGRICULTURE IN LATIN AMERICA Changes in precipitation and temperature According to the predictions of Washington and Meehi (1984) and the Commonwealth Secretariat (1989), there are likely to be important rainfall distribution changes in at least the tropical and subtropical parts of Latin America. The wet seasons are likely to get wetter and the dry seasons drier. Sombroek in Chapter 13 summarizes the predicted changes with up to 30 mm increase in soil moisture during the period December - February for Eastern Amazonia but up to 30 mm decrease for Western Amazonia and Central Brazil. There is also expected to be a 10 - 20 mm increase in Eastern Amazonia and in Central Brazil in JuneAugust which is at present a dry season in that region. No change is predicted in Western Amazonia during this period which is normally wet at this time. The models also predict a temperature increase of 5°C for the higher latitudes but only 2°C for thc equatorial region. Sombrock also discusses the possible direct effect of increased air temperature on tropical crop plants. The expected increase in atmosphere CO2 production and associated temperature increase could result in a 25% increase in biomass production world wide although for the tropics the combined effect might be smaller. No precipitation changes are predicted for the rest of Latin America except that Uruguay, Northern Argentine and Chile are expected to be somewhat wetter in June-August. Such a change is likely to increase agricultural production in this part of the region which normally has a deficiency of rainfall. The predicted changes in precipitation would probably affect the water balance of the Amazonian region and the discharge of water by the Amazon River, especially if this is associated with increased clearing of natural vegetation with a resulting increased runoff. The full effects of this are likely to be felt not only in the Amazon basin but also on the South American coast. In Central Amazonia, the expected precipitation changes may lead to a better rainfall distribution for crops. The projected decrease in the rest of the area in DecemberFebruary would make cropping more risky than it is at present, this period being

.!An

American and Caribbean agriculture and the greenhouse effect

25 1

presently the short wet season for a good part of this region with already unpredictable rainfall. This moisture supply problem is likely to be exacerbated by the tendency for higher biomass production and higher yields due to increased atmospheric COz. Changes in crops and land use may be involved in which more drought tolerant species may have to be grown. These trends are also likely to encourage more extensive cattle rearing in areas with low population densities which is characteristic for the whole region. There are also likely to be wider annual fluctuations of water levels in the Amazon River system which could lead to great flooding in the wet season. With appropriate soil management to cope with wider soil moisture fluctuations, the effects on agriculture need not be traumatic. If the climatic changes are accompanied by a higher level of farming especially in soil and water management and including greater use of irrigation, the adverse effects can be minimized. In association with the climatic changes, changes in the crops grown, especially the commercial crops must be viewed as a possibility. New technologies would have to be adopted. Thc increase in temperature could have an important beneficial effect in extcnding the cropping area and widening the range of crops produced in Southern Argentina and other cold areas of Latin America. However, in this event there would be greater evapotranspiration demands with implications for appropriate crop management. Rise in sea-level The projected rise in sea levels due to global warming has been estimated to be about 0.5 to 1.0 m over the next hundred years (Brammer and Brinkman in Chapter 12, Commonwealth Secretariat 1989, Sombroek in Chapter 13; Washington and Meehl 1984). The consequences of this are frightening for several Latin American and Caribbean countries. Some present land masses would be inundated, particularly the flat, low lying coastal fringes which are so important in Guyana and Surinam. The present coastal plain in those countries which are now below the level of the highest tides would be completely submergcd; so would several areas in the region that have been formed from coral and marl. This would lead to obvious catastrophic situations in which parts of countrics may disappear. Another result would be salinization of groundwater. As an example, in Barbados and Jamaica the groundwater of low lying limestone areas is being used for domestic purposes and for irrigation. A significant rise in sea-level will disturb thc existing delicate balance between fresh and saline waters in these aqui fcrs, leading to salinization. Any risc in sea-levels would cause greater encroachment of salt water in the rivers of the regions which would lead to reduced potential for irrigation throughout and would create the need to replace, modify or expand domestic

252

N.Ahmad

water supply systems. Damage from saline intrusions into fresh water wells is already a problem in parts of the region. Many low lying coastal areas in the Caribbean and South America are protected either by coral reefs or mangroves. Both are under pressure from human activity such as pollution and sedimentation as a result of bad land use and construction processes and excessive cutting of mangrove for poles and fuel. A rising sea-level would tend to narrow the band of mangrove between the sea and human occupation and it might outstrip the growth capacity of some coral reefs. The vulnerability of the coasts to erosion and flooding would increase. These events would also greatly upset marine life in coastal and estuarine areas which could lead to the disappearance of presently valuable fishing grounds. Even if the full expected rise in sea-level is not attained, any further rise would pose problems to several areas in the region. There is already much coastal erosion occumng along the coastal plains of Surinam and Guyana. While the full causes for this are not known, one factor could be greater runoff in the Amazon Basin and greater discharge by the Amazon River due to clearings for agricultural purposes. This phenomenon is likely to increase due to greater precipitation in parts of the the Basin and from runoff from further agricultural clearings. As can be seen from the above, the consequences of any appreciable global warming to agriculture in Latin America and the Caribbean can have traumatic results. Therefore, it would be very germane to this problem if the present agricultural pursuits in the region can be considered in the light of possible contributions to atmospheric greenhouse gases and climate change.

THE AGRICULTURAL ENVIRONMENT Climate Much of the land area lies in equatorial latitudes and it has the earth’s greatest continental extent of humid tropical climates - areas in which plant growth is restricted by moisture stress only for very short periods of the year. These areas, about 42% of Tropical America, have a udic moisture regime. Another 43% has an ustic moisture regime. In these areas, there may be several dry months in any year. Examples are Northeast Mexico. Only 7% of the area is aridic. Although there are few areas in lowland Latin America where low temperatures may inhibit plant growth (mainly in temperate Brazil) low temperatures caused by high elevation are common in much of the Andean Region and to a lesser extent in parts of Central America and the Caribbean.

Lorin American and Caribbean agriculrure and rhe greenhouse eflecr

253

Soils All the soil orders are represented in the region. Soil classification in this text is according to the USDA Soil Taxonomy System (Soil Survey Staff 1975); between brackets the approximately equivalent names are given according to the legend terminology of the FAO-Unesco Soil Map of the World (FA0 1974). Oxisols and Ultisols (Ferralsols) cover the largest area of tropical Latin America followed by Inceptisols, Entisols and Mollisols with minor occurrence of the other orders. Oxisols are the dominant well drained soils of the areas affected by the Guyana and Brazilian Shields, including the Cerrado, the Llanos and the Eastern Amazon basin. According to Sanchez and Buol(l974) there are about 5 13 million ha. Oxisols occur on many of the old erosion surfaces and also on the more recent depositional plains. An important area of Oxisols is in the Greater Antilles where they have developed on hard limestone. Here the soils are commonly bauxitic and are mined as aluminium ore. Ultisols (Acrisols; most Nitosols) are extensive in Latin America, covering some 371 million ha (Sanchez and Isbell 1978). They are fairly common in the higher rainfall areas of Central America but are more widespread on gently sloping outwash plains of the Amazon and Orinoco basins. Typically, they occur in South and Central America on an old coastal plain which was dissected at the end of the Quaternary Period and upon which the new coastal plain has been built. Alfisols (Luvisols; some Nitosols) occur throughout the region. In South America they occur in Northeast Brazil, the North coast of Colombia and Western Venezuela and occupy approximately 160 million ha. In Central America and the Caribbean there are about 32 million ha, derived generally from basic materials. In the Caribbean, Alfisols occur on old volcanic materials and on intermediate and basic rocks of igneous and metamorphic origins. Inccptisols (Cambisols) are widespread in Andean South America and mountainous areas of Central America and the Caribbean where they are associated with volcanic materials (Andepts/Andosols). Other Inceptisols are also common in these upland areas and in Eastern Brazil. Large areas of Aquepts (Gleysols) are widespread in the seasonally flooded areas on the coast of South America and along the major river systems. Vertisols (Vertisols) are widespread through Latin America and the Caribbean and may be more common than previously thought. They are distributed from Mexico to Argentina and occur in practically every country. They are common in the less humid regions where they are developed on basic igneous and mctarnorphic rocks and on volcanic materials and calcareous rocks. Vertisols are also common in high rainfall areas on very fine marine, lacustrine or riverine sediments which have since been uplifted. In some of these areas, Vertisols occur on fairly steep slopes and they are naturally unstable. Due to the nature of the parent materials, acid Vertisols are frequent in parts of Coastal South America and

254

N.Ahmad

the Caribbcan where they form important agricultural soils. Entisols (Arenosols; Regosols and Lithosols; Fluvisols) are another soil order which may be more important in Latin America than previously thought. In Central and Eastern Brazil, the chief forms are deep red or yellowish sands formed from siliceous parent materials on old erosion surfaces. In the Amazon basin, the Guyanas and Venezuela they are widespread and occur on whjte sand deposits. There are very large areas of quartz sand in Brazil, Guyana, Surinam, French Guyana and Venezuela. In Brazil alone 70.9 million ha of these soils are distributed (Bowen and Lobato, 1989). Many of these have spodic subgroups. Entisols occur throughout the steeper parts of the Andean system and other mountainous areas. In similar locations they are important in Central America and the Caribbcan. In this area exploitive use over a long period of time results in the occurrence of these soils where they should not normally be. Entisols are also very important on deltaic, estuarine, coastal and marshy areas where they can have sulphiric properties and intergrade with Histosols. Mollisols (Chemozems; Phaeozems; Kastanozems) are productive soils and are locally important in subhumid parts of Latin America. Examples are in Central Mexico and the Yucatan Peninsula, in some drier inter-Andean valleys of Peru and Colombia, in Northwest Argentina and in Northern Paraguay. They are also important in the Caribbean where they are associated with less humid climates on calcareous and other basic rocks and on volcanic parent materials. Aridisols (Xerosols and Yermosols; Solonchaks) are distributed on the coastlines of Surinam, Guyana and Venezuela and Northern Brazil where salinity is due to the constant influence of the sea. They also occur in dry areas of desert or near desert climates in Mexico and Peru. Histosols (Histosols) are widespread throughout the region but do not occur in large expanses in any one location. Commonly they are associated with brackish inland swamps along coastlines and may be complicated with both salinity and sulphiric features. Spodosols (Podsols) are important in the Amazon (Klinge 1965; Bleakley and Khan 1963) and are also known to occur in Venezuela. They were first described by Richards (1941) in Guyana as giant Podzols. A most characteristic feature is the dark colored chemically pure drainage water from these Podzol soil areas from which the Rio Negro derives its name (Herrera et al. 1978); the same dark colored water is characteristic of the rivers of the Guyanas and the Orinoco. These soils are infertile but can be managed for crop production (Ahmad 1989). The region’s soil resources have been very little studied. Some soil situations have been studied with nonproportional intensity compared to their area of occurrence (i.e Spodosols) due to special interest of visiting scientists. The detailed distribution and classification of the soils is not known with the exception of certain restricted areas, so that their management for agriculture is based more on sociological conditions than soil properties. Soil surveying is an essential,

Lalin American and Caribbean

agricullwe and [he greenhouse effeci

255

initial step. A notable exception is the Commonwealth Caribbean where the soils of the entire region have been mapped to at least semidetailed level and the soils have been studied to a good extent.

LAND USE Cropland and cropping systems Only 1 1 % of the cultivable area is tilled at present in Latin America (Schaffer 1980). There is much unused land and the population densities are relatively low. The exception is the Caribbean area which has long been colonized and settled with labour intensive enterprises and therefore the population density is now high and there is literally no untilled land which is suitable for farming. Agricultural land use can be divided into areas with farming systems where the land is occupied by permanent crops and those in which short term food crops are produced largely on a subsistence level. The permanently farmed areas produce a number of important crops such as coffee, banana, sugar cane, cacao, rubber, oil palm and citrus; and the levels of management and productivity of thesc crops are very high. When fully established, the extent of ground cover can approach forest conditions, Therefore, this form of agriculture may not contribute to the greenhouse effect. Exotic tropical fruits are not yet produced in large quantities for export. The other important aspect of permanent land use is pasture and livestock production. Land use for short term crops includes cultivation of the important food crops of the region such as rice, beans, maize, other temperate cereals, manioc, sweet potato or white potato. The distribution of land for these uses is not exactly known; probably pasture and livestock production would be the overall dominant land user which has increased consistently in recent years throughout the region (Kohlhepp 1986). There are no land zoning policies, at least none that are effectively applied. Even in the Commonwealth Caribbean whett land resources are extremely scarce and valuable, here are no effective land zoning laws. Throughout the Caribbean, land use is largely determined by the landowner. In effect, the flatter areas and better soils are generally in occupation as large blocks and are in permanent plantation agriculture. The small, landless cultivator is forced to squat on steeper lands with fragile soils, cultivating short term food crops. Since the farmer does not own the land and in any case for economic reasons, hardly any inputs except the farmer's labour are made; hence, there is rapid deterioration. This level of farming exists alongside highly efficient, capital intensive production of major export crops. Land use for food crop and livestock production is generally based on shifting cultivation. The traditional form is cutting of the vegetation, allowing

256

N. Ahmad

time for it to dry, burning to clear the land, and begin with alternate cropping; this is a cyclical land use technique in which short term cropping alternates with long periods of forest fallow. In the system, soil ameliorants or crop rotations are not used. It begins when trees, bushes and lianes of a forest are felled or slashed and after a certain amount of drying, burned (Walters 1971; Goodland and Irwin 1975; Neugebauer 1988). The rain forest contains up to 500 tons ha-1 of biomass and a considerable part of the mineral nutrients present in the soiVplant system. Fire destroys this biomass with much of the N, S and P while the other nutrients are mineralized to suffer loss through surface runoff and leaching. With the beginning of the favorable rainy season, seed is sown in the burned untilled land. In the first year there is a high yield and very little labour input, but even for the second harvest, considerably more labour is required because of weeds and the deterioration of soil structure. This is due to the accelerated loss of fertility caused by heat, moisture and extreme insolation and soil compaction. By the third year competing weeds can become uncontrollable. As soon as the labour requirements become too high, or the harvests too low, the shifting farmers desert their area. Using the same technique as before, they open a new area by the same process with at least 25 years of fallow. Profound soil deterioration both physically and chemically results from the practice. The changes have been well documented both in South America (Sanchez 1981) and in West Africa (Nye and Greenland 1960). If there are large areas available with a low population density shifting farming allows continuous high yields. However, where cultivation and regeneration are practised without the necessary care, or where population density has risen so high that the land is not left fallow for long enough, soil erosion and reduced soil productivity results. This induces successive changes in the vegetation structure and in climate and edaphic conditions, and finally brings about destruction of the tropical forest. At a population density of 25 people per km*, the carrying capacity in this farming system has been reached (Neugebauer 1988). The rate at which the deterioration of the vegetation takes place is of course dependent on the reserves of fertility. In areas of volcanic soils or soils developed on calcareous materials, an initially cleared area can be used for a long time for agricultural production and such areas can support a higher population density. The resistance of volcanic soils to lapid deterioration can be seen in some pans of Indonesia for instance, where the population density which lives on subsistence agriculture is over 400 per km2. The same also occurs in Latin America and the Caribbean. In such areas probably total deterioration never results. In Central America for instance land initially cleared by the Mayas, cultivated and then deserted, is today in a fairly productive secondary forest vegetation. Traditionally, the method of managing land in these areas is somewhat different from where the soil environment is poorer. The "milpa" and the "conuco"

L d i n American and Caribbean agriculture and the greenhouse effect

257

systems of land use are examples in which the land is semipermanently used for agriculture. Livestock and production systems The ultimate result of shifting agriculture in Latin America, especially in South America, is the wellknown man made savannahs or the replacement of magnificent forests with coarse grass scrub. The resulting savannahs can be beneficially used for cattle rearing but it forms such poor pasturage that unless more productive plants can be adapted, the productivity is extremely low and without improvement it is an unproductive form of land use. Besides, annual fire is a component of management since burning of the savannah stimulates young growth so that for a short time at least after the burn, the stock can have a more nutritious forage. (Medina 1982 and Medina et al. 1978). This adds to the greenhouse effect. An estimate of the area of these man made savannahs throughout the region is not readily known but it is very large. Throughout the region, there are also examples in most agroecological zones of high production levels of pasture management and livestock production. These areas are not burned and they remain in production on a permanent basis. This type of land use presents the other extreme in livestock production and is at least an alternative to the extensive grazing on man made savannahs. Although not without exception, livestock production in Latin America has been at some to the environment. The area is a net exporter of beef and has a high per capita consumption of meat products. Production has been increasing in many of the countries during the last 10 years while most other agricultural commodities have decreased during that period. Throughout the tropical area of Latin America the clearing of land for livestock production occurs mainly through the subsistence farmers' slash and bum with shifting agriculture, so that pastures result from a degradation process with atmospheric COz enrichment. Further, the animals are all grass fed and therefore livestock production adds to atmospheric methane. Due to the abundance of land and principal lack of inputs, livestock production is extensive and stocking rates are usually quite moderate. Where the nutrient reserves in the soil are very low, such as in the Amazonian region of South America, the deterioration of the vegetation is very rapid and manmade savannahs can be formed probably after only two cycles of land clearing for shifting cultivation. There are, however, examples in the region where the quality and quantity of vegetation is greatly improved by agriculture. The Cerrado of Brazil is the best example of this situation, where the typical scrub vegetation is cleared, all the needed inputs of soil ameliorants and fertilizers are made and the land is being transformed into agricultural land either for cultivation or improved pastures with high production. This could be an example of a positive effect of human interference on the environment.

258

N. Ahmad

AMMONIA

The environment and its exploitation In recent times the rate of deterioration of the environment has accelerated in the Amazon Region, caused by the traditional slash and bum shifting cultivation and other activities such as mining, forest exploitation and cattle rearing. 'Ihe possible global effect on the environment this ecological transformation is likely to exen, has the world's population sensitized to what is happening in this vast region. Thc drainage basin of the Amazon River is still by far the world's largest area of continuous forest. The whole catchment area exceeds 7 million km2, and while i t comprises 40% of the area of Brazil, it extends into French Guyana, Surinam, Guyana, Venezuela, Colombia, Peru and Bolivia. The natural vegetation is like non other anywhere else in the world with over 100,OOO species and an equally varied fauna. It has an abundance of water - about 20% of all water flowing on the earth's surface runs down the Amazon to the sea. The typical rainforest vegetation has developed, in which the biomass exceeds 500 tons per ha and the annual production is 25 tons per ha. The area is as yet only sparsely populated - about one person per km2. Characteristic for the vegetation, apart from the great variety of species, are its vertical division into 2-3 levels of trees of different heights, the highest storey being up to 50 m, its fewer appendages of epiphytes hanging from the trees as compared to the tropical mountain forest and its sparse ground vegetation due to lack of light. Only from the river banks, largely concealed by dense vegetation, is the impression of impenetrability created. At the present time, the forest is being cleared at the rate of 10 million ha per year. It was estimated that in 1987 up to 25 million ha of land in the basin was burned but most of this area consisted of already deforested pasture and plantations. Up to 1980 only 2.45% of the Brazilian Amazon forest region was cleared. Even if this is doubled in the last decade, less than 5% of the region would have bcen cleared so far, but there is a rapid rate of increase. During the 1970s and even later, the Brazilian Government positively encouraged settlement in the Amazon Basin. Incentives to cattle ranchers exceeded one billion dollars over the last decade but this activity brought little success and much damage to the fragile environment. It also embarked on the construction of the Trans-Amazon Highway, a system of roads that runs West from Recife toward the Peruvian border. The idea was to prompt a land rush. To encourage settlers, the Government offered transportation and other incentives, allowing them to claim the land that they had improved by deforesting. However, most of the 8000 families that responded at least between 1970 and 1974 have bcen disappointed due to rapid deterioration of crop yields. A later development is the Grandc Carajas Programme to develop mineral resources. The principal iron

Latin American and Caribbean agricdfure and the greenhouse effeci

259

ore mine began production in 1985 but the ore melters are powered by charcoal which is produced from the wood of the surrounding vegetation; it is feared that this venture would be a repeat of the experience of the state of Minas Gerais in Central Brazil, where pigiron production consumed nearly two-third of states forests. In another huge project, Polonoroeste, the Government is trying to develop the State of Rodonia. Instead of model settlements, the project has produced impoverished itinerants. Settlers grew rice, maize, coffee and manioc on a shifting basis in which the farming and burning became a perpetual cycle of degradation. Since 1980, the percentage of the State covered by native forest has dropped from 97 to 80%. Earlier in the 1920s Henry Ford was unsuccessful in a bold attempt to promotc large scale rubber cultivation in a part of the Amazon Basin and in the late 1960s and 1970s Ludwig was also unsuccessful in a project to produce wood pulp and rice along the Jan River. Apparently, the only reasonably successful agricultural production in the region has been the production of black pepper, a crop which was introduced by Japanese scttlers in 1933. In the culture of this crop, the forest vegetation is not completely cleared and the soil never fully exposed. Besides, the crop is managed at a high agronomic level. The fires during a five to six months dry season can release up to 620 million tons of C02 into the atmosphere, an amount equivalent to 10% of all the C 0 2in the world's atmosphere. This result is caused by the activities of relatively few pcoplc. Instead of contributing to atmospheric C02 levels, the forest can in fact serve as a filter for up to 25% of the 5 billion metric tons of C02 released by the burning of fossil fuels each year. The burning of the forest therefore, not only releascs more C 0 2 but it also destroys a unique system for purification of the atmosphere. It must be pointed out that the rapidly increasing use of alcohol as fuel in Brazil is ecologically better than the use of only fossil fuels and in this respect this country is pioneering. Clearly, the results of human activity in the Amazon so far has tragic results overall. Not only is the forest being destroyed and the whole environment deteriorated, but the people involved in the process have been leading miserable lives as well. Cutting of the forest and burning is obviously too harsh an initial treatmen1 for this fragile ecosystem, posing a great shock to the environment from which it cannot rccovcr. I t is imperative that more is known about the ecological conditions of this uniquc area so that man's activities can be adaptable. One outlook is ihat enough land has already been cleared of its forest cover and what is nceded now is to develop management techniques and the necessary input to make these areas agriculturally productive on a permanent basis. There is already some information about how this can be done from the work of Sanchez and his colleagucs in the Peruvian Amazon area (Sanchez 1981; Bandy and Sanchez

260

N.A

h d

1982). This approach would obviously lead to a more settled population and eventually a much better quality of life. The other approach is practice of more ecologically adapted farming and more and better use of the natural products of the forest. This approach would not have such negative impacts on the environment, and the hazards of an increasing global greenhouse effect would be diminished. It would require much more research, product development and promotion, a greater variety in the diet of the population and adoption and creation of appropriate technology for mining systems for the Amazon Region. The actual and potential agricultural systems in the region can now be examincd in the light of the above. Annual crops The main annual crops are rice, maize, cassava and peanut. Valverde and Bandy (1982) concluded that the growing of annual food crops in Amazonia requires the use of modem technology, to be developed in situ and to be sensitive to the local socioeconomic conditions. If crops are grown in rotation or interplanted, the overall performance is better. Also, the native method of land clearing is superior compared to bulldozing. The crop variety is too small to provide a wide enough range for intercultivation and crop rotations. More legumes should be involved and an obvious crop is pigeon pea (Cajanus cajan) which can be grown as a perennial. This applies LO Latin America as a whole. According to Valverde and Bandy (1982), soy bean has shown promise and winged bean and yam have been tried. Carbohydrate crops, aroids and some yams are well adapted to grow with a certain amount of shade and may be adaptable to Amazonian conditions as they are to forest conditions in West Africa. However, the initial problem with these new food crops is the existing dietary habit of the population, with traditional food consisting of maize, rice, beans and manioc. As far as the greenhouse effect is concerned, the cultivation of annual crops leads to atmospheric enrichment with C 0 2 due to frequent burning for land clearing. Rice production is a notable exception. Although in Latin America a lot of rice is grown, over 75% of it is produced in upland conditions and this does not lead to methane emissions. However, about 2 million ha flat lands capable of producing irrigated rice exist in the Amazon Basin and its development would contribute to the methane problem (Alvim 1982). Perennial crops Alvim (1 982) recommended certain potentially suitable perennial crops for

the Amazon Region. Such crops, replacing forests, would have the potential of largely substituting the beneficial ecological effect of the forest. In this

L o h American and Caribbean agriculiure and the greenhouse effect

26 1

connection, it is interesting to note that two of the most important tropical perennial crops i.e. rubber (Hevea brasiliensis) and cacao (Theobromacacao) evolved in the Amazon, but until recently Amazonian countries paid little attention to research and development of these crops. In the case of rubber only 1% of the world's crop is grown in Brazil, the country producing only about 25% of its requirements. With improved techniques of controlling the South America Leaf Disease (using fungicides), in areas with a marked dry season and where suitable production techniques are propagated, rubber production can be greatly increased. Brazil produces only about 0.2% of the world's supply of cacao. It was believed that the soils of the Amazon were not productive enough for commercial production of this crop but according to Alvim (1982) about 10 million ha of land with fcrtility comparable to the currently best cacao soils elsewhere in Brazil have been identified in the Amazon Region. Experiments have demonstrated that very high yields of cacao can be obtained in these soils without the use of fertilizers or liming. Oil palm (Elaies guineenis) is a crop which may have important prospects in the region. As far as climate and soils are concerned, the environment seems ideal and in fact its adaptability has been proved. Crossing of the African oil palm with an Amazonian species (E. Melanoccoca) has produced very interesting material for use in the Amazon (Alvim 1982). Some Latin American countries already produce palm oil but the potential for expansion is great. Sugar cane (Saceharum officinarum) especially for alcohol production is a semiperennial crop, which is suitable in large areas of the Amazon Basin. This crop is widely grown in other parts of Latin America which is the world's largest producing area. It requires a fairly high level of management for efficient production. However, the annual burning of the crop prior to harvest destroys up to 5 tons per ha of biomass and together with the use of nitrogen fertilizers, have certain environmental implications. Black pepper (Piper nigrum) is the most popular crop in the Brazilian Amazon at present and Brazil is now the third largest producer. The crop lasts for about 10 years after which it usually succumbs to a root disease and it is replaced with crops such as cacao, rubber, papaya or passion fruit, while a new area is cleared for pepper growing. These crops benefit from the residual fertility left in the soil from the well managed pepper crop. Potential perennial crops These include cashew (Anacardium oflicinale - Amazon native), achiote (Bixa oreffan- Amazon native), Brazil nut (Berrholleria excelsa - Amazon native), and several palms such as pejibaya (Guilielma gasipaes) or peach palm, palmito (Euterpe oleracea), seje (Jessenia s p p ) , piquia (Caryocar villosum), buriti (Mauritia flexuosa) and copaiba (Copaifera spp), all of which produce either

262

N.Ahmad

edible fruit or palm hearts. Breadfruit (Arrocarpus aftifis) should also be mentioned as being a prolific, high yielding tree, the fruit of which is an excellent source of carbohydrates. There are several exotic tropical fruits which have possibilities as fresh fruit, nectars or other preserves and among the most adaptable species may be avocado (Persea arnericana),citrus (Citrus spp.). carambola (Averrhoa carambola),guava (Psidium guajava), Malacca apple (Eugenia malaccensis), mammey apple (Mummeaamericana),mammey sapote (Calocarpum mammosum), sweet and sour sops (Anonu spp.) and several spices. Systems of cultivation would have to be developed for trees and would require genetic improvement, product development and international promotion.

Pa st 11 re s What is needed, not only in the Amazon but also elsewhere in Latin America, is an improvement of the existing pastures to increase the carrying capacity rather than clearing more land and expand the current extensive system. Research on pasture plants adaptable to poor soils has been in progress at CIAT for years, and this research could now support pasture improvement in the region. Agraforestry This is a farming system which is being projected in the tropics as a substitute for shifting cultivation; it combines the protective and soil improvement roles of spccific highly desirable trees and crop production in various geometrical arrangements. It has most appeal in overpopulated areas of the world, where there is almost an equal shortage of food as of forest products such as fire wood, poles and livestock fodder. The forest component contributes soil protection and improvement, and wood products and fodder which are all in high demand, while the arable crops grown in association provide human food. In Latin America and especially in Amazonia there is as yet not a great appeal for agroforestry. In the Caribbean area a form of agroforestry incorporating forest food plants as a top storey cropping and crops at the ground surface with an almost random distribution of the various components of the system has been practised for at least 150 years. This system is popular as it offers many advantages for the farmer and it protects and enriches the soil. This form of agroforestry may well be applicable to the Amazon Region.

Forest improvement

Tropical rain forests of the Amazon Region are unique as far as flora and fauna species diversity is concerned. However, the growing stock and

263

Lo& American and Caribbean agriculture and Ihe greenhome effect

commercial volumes are not as high as generally assumed and it is commonly stated that these forests are more impressive than useful (Von Maydell 1984). In Table 2 1.1 an analysis is given of the tree species of forests of Latin America, compared to those of forests elsewhere. Table 21 .I

Comparison of characteristicfigures of closed tropical lowland rainforests a

Wood charactcristics Atiica Total nurnbcr of stems 60 - 80 Numbcr of species 40 - 50 Numbcr of commercial 5 - 20 species 3 Volurnc per stcm, all species (in m3) Total growing stock in rn3 180 - 250 Growing stock of commer- 20 - 70 cia1 specics (m3) Net timbcr rcrnovals 10 - 20 a

Latin America Asia 70 - 90 120- 150 50 - 70 70 -90 3 - 10 10 - 30

Maxima 200 150 10 - 30

2

3

4

140 - 200 15 - 40

200 - 400 80 - 200

450 - 800 450

2 - 10

4 0 - 100

150

All data are averages per ha for trees over 30 cm diameter growing stock; data refer to sternwood (from Von Maydell 1984) above buttress

Table 21.2 presents an indication of the relative productivity of Latin American forests. There is much scope for improving the productivity of the forests for actually harvesting more from much less land. In Trinidad, the shelterwood system of forest regeneration has improved forest from 6 to 10 productive trees to 60 to 80 productive trees per ha in 30 to 40 years. In this case, the new forest itself can be maintained, while the harvest is useful timber (Bruenig 1986). Table 21.2 Slatistics of individual wood production in the tropics and subtropics, 1980 ~~

Type of wood

World

Latin

Africa

America

Roundwood mill m3 Lumber mill m3 Woodbasepanels mill m3 Pulp mill t Papx,cxdboard mill t

*

excluding USSR and Japan

1393.5 428.7 102.0 126.8 174.2

76.9 24.2 4.2 5.1 7.1

50.4 7.9 1.2 1.2

1.6

Asia* Tropics and Australia Subtropics in Oceania %ofworld production 173.6 21.6 46.4 18.3 8.4 13.5 3.5 7.7 14.2 13.1 (from Von Maydell 1984)

264

N . Ahmad

REFERENCES Ahmad, N. (1989). Acid sandy soils of the tropics with particular reference to the Guyanas. In: Farming systems for low fertility acid sandy soils. D. Walmsley (Ed.). CTA Seminar Proceeding, Georgetown, Guyana. Alvim, P. de T. (1978). Agricultural production potential of the Amazon Region. In: Pasture production in acid soil of the tropic. P.A. Sanchez and L.E. Tergas (Eds.). C U T , Colombia. 13-23. Alvim, P. de T. (1982). A perspective appraisal of perennial crops in the Amazon Basin. In Amazonia: agriculture and land use research. B. Hecht (Ed.). C I A T 311-328. Bandy, D.E. and P.A.Sanchez (1982). Continuous crop cultivation in acid soils of the Amazon Basin of Peru. In: Management of low fertility acid soils of the American humid tropics. J.F. Wienk and H.A. de Witt (Eds.). IICA, San Jose, Costa Rica; 153174. Bleakley, D. and E.J.A. Khan (1963). Observations on the white sand areas of the Berbice Formation, British Guyana. J. Soil Sc. 14: 44-51. Bowen, W.T. and E. Lobato (1989). Possibilities and constraints for crop production on acid sandy soils (quartz sands) in Brazil. In: Farming system for low fertility acid sandy soils. D. Walmsley (Ed.). CTA Seminar Proceedings. Georgetown, Guyana. Bruenig. E. (1986). The tropical rain forest as an ecosystem. Plant Res. & Dev. 24: 18-30. Commonwealth Secretariat (1989). Climatic changes: Meeting the challenge. Report by a Commonwealth Group of Experts. Commonwealth Secretariat, Marlborough House, London. Espig, C. ( 1 989). Ecological problems and environmental stresses caused by agricultural production in the tropics. Natural Resources and Development. Institute for Scientific Cooperation, Tubingen. 55-68. F A 0 (1974). FAO-Unesco Soil Map of the World 1:5,000,000 Volume I Legend. UNESCOParis. Glauner. H.J. and H. Keil (1988). The approaches, goal and methods of eco-farming in tropical and subtropical regions of developing countries. Plant Res. and Dev. 28: 4762. Goodland, K.J.A. and J.S. Irwin (1975). Amazon jungle: green hall to red desert. Elsevier, Amsterdam. Hecht, S . B . (Ed) (1982). Amazonia: agriculture and land use research. CIAT. Herrera, K., C.F. Jordan, H. Klinge, and E. Medina (1978). Amazon ecosystems; Their structure and functioning with particular emphasis on nutrients. Interciencia 3: 223232. Klinge, H. (1965). Podzol soils in the Amazon Basin. J. Soil Sci. 16: 95-103. Kohlhepp, G. (1986). Problems of agriculture in Latin America. Production of food crops versus production of energy plants and export. Appl. Geog. and Dev. 27:60-92. Medina, E. (1982). Nitrogen balance in the Trachypogon grasslands of Central Venezuela. Plant and Soil 67:305-314. Medina, E.. A . Mendoza and R. Montes (1978). Nutrient balance and organic matter production in the Trachypogon savannahs of Venezuela. Trop. agric. (Trin.) 55: 243254. Neugebauer, €3. (1988). Starving in the forest. Plant Res. and Dev. 28: 7-31. Nye. P. and D.G. Greenland (1960). The soils under shifting cultivation. Tech. Comm.51. Cornmonwealch Bureau of Soils, Harpenden. Richards, P.W. (1941). Lowland tropical Podzols and their vegetation. Nature, London, 148 : 129-131.

Lotin American and Caribbean agriculture and [he greenhouse effect

265

Sanchez, P.A. (1981). Soil management in the Oxisol savannahs and Ultisol jungles of tropical South America. In: Characterization of soil. D.J. Greenland (Ed.). Clarendo Press, oxford. Sanchez, P.A. and S.W. Buol (1974). Properties for some soils of the upper Amazon Basin in Peru. Soil Sci. SOC.of Amer. Proc. 36: 117-121. Sanchez, P.A. and R.F. Isbell (1978). A comparison of the soils in tropical Latin America and tropical Australia. In: Pasture production in the acid soils of the tropics. P.A. Sanchez, and L.E. Tergas (Eds.). CIAT. Colombia: 25-53. Schaffer, G. (1980). Ensuring man's food supplies by developing new land and preserving cultivated land. Applied Geog. and Dev. 16: 7-27. Soil Survey Staff (1975). Soil Taxonomy; a basic system of soil classification for making and interpreting soil surveys. SCS-USDA. Agricultural Hand book 436. Washington DC . Valverde, C. and D.E. Bandy (1982). Production of annual crops in the Amazon. In: Amazonian: Agriculture and Land Use Research. S.B. Hecht (Ed.). CIAT: 243-280. Von Maydell, H. (1984). The role of forestry in rural development. Plant Res. and Dev. 19:52-70. Washington, W.M. and G.A. Meehl (1984). Seasonal cycle experiment on the climate sensitivity due to a doubling of CO2 with an atmospheric general circulation model coupled to a simple mixed layer ocean model. J. Geophys. Res. 89: 9475-9505. Walters. R.F. (1971). Shifting cultivation in Latin America. FAO, Rome.

This Page Intentionally Left Blank

267

Chapter 22

THE ASIAN AGRICULTURAL ENVIRONMENT AND THE GREENHOUSE EFFECT L. Venkataratnam National Remote Sensing Agency Balanagar, Hyderabad - 500 037. India

Modem agriculture depends heavily on four technologies: mechanization, irrigation, fertilization and chemical control of weeds and insect diseases. Each of these technologies has made an important contribution to the earth’s increased capacity for sustaining human population and each has perturbed the cycles of the biosphere. In the technologically advanced agriculture of today, even in some of the Asian countries, the expenditure of the fossil fuel energy per acre is often substantially greater than the energy yield embodied in the food produced. In the poorer countries, man’s expanding need for fuel has forced him to cut forest far in excess of their ability to renew themselves, making the soils vulnerable to soil erosion. The inorganic nitrates and phosphates discharged in to the lakes or ponds provide a rich medium for algae. The massive growth of alae in tum depletes the water of oxygen and thus endangers fish life. In the end, the eutrophication or overfenilization of lakes slowly brings about its death as a body of the fresh water converting into a swamp. In spite of rapid advances achieved in the agricultural sector during this decade, enabling developing countries to increase their agricultural production at the rate of 3.6% per year. Over 200 million people in Asia alone wills till continue to be undernourished. Even assuming a linear growth in agricultural output, the annual global cereal production in AD 2000 is estimated to fall short of demand by almost 140 million tonnes. Year to year variations in the food production due to climatic changes, drought and flood damages, pests and diseases further compounds the problem leaving a gap of over 20% between production and demand (Rao 1989). For an accurate prediction of the climatic changes, we have to clearly understand three fundamental issues: 1) the forces such as solar radiation, atmospheric dynamics, and chemistry which initiate global changes; 2) the response of the earth to these forcing functions involving complicated interactions between climate, ocean, land, biota, and the atmosphere; and 3) accurate modelling and reliable long range prediction of global changes (Rao 1989). In addition to h e problems created by increased emissions of greenhouse gases (see Bouwman and Sombroek in Chapter 2), the progressive destruction of the ozone layer in the upper atmosphere (1 5 to 30 km altitude) has been observed

268

L. Venkatorotnom

particularly during spring season in Antarctica. This is attributed to an increase in the use of CFCs and fluorocarbons for purpose of refrigeration, air-conditioning, propellants for aerosols, expanded foam for packaging etc. Thus the world is faced with a multidimensional problem directly related to the patterns of energy consumption, life styles, industrial and urban expansion and agricultural technology. As of today, out of the estimated total emission of CO, (6.5 to 7.5 Gt C y-') only half is contributed by the developing countries having 75% of world population. The present contribution of all developing countries of CFC emission is less than 15% of the total global estimates of 1.2 m tons y-l, the per capita annual contribution ranging from 1.22 kg in USA and 0.9 kg in Europe and Japan to just 0.006 kg in India and China (Rao 1989). National Environmental Engineering Research Institute (NEERI) in India has been collecting air quality data in and around Agra, India since 1976 and its findings are given in Table 22.1. Table 22.1 Air quality data for Agra

Component Sulphur Dioxide

Data Annual average :

24 hrs average : 2 hrs average : Nitric Oxide

15 to 20 pg m-3 7 to 42 pg m-3 20 to 160 pg m-3

Negligible

Suspended particle matter 24 hrs average : 2 hrs maximum :

66 to 448 pg m-3 106 to 803 pg m-3

In general, the levels of different pollutants are well within the permissible limits laid down by International Agencies, and also recommended by Indian Standards Institute, the optimum level to cause a atmospheric hazards being 80 pg m-3 (0.03 ppm). It is however, observed that the two hourly maximum value of SO2 crosses 100 pg m-3) at times and this increase even for shorter duration is harmful (Tripathi and Prasad 1984).

GLOBAL WARMING DUE TO AGRICULTURE The crucial role played by vegetation (crops and forests) in the global cycling of C02 is well known. The vegetation and the soil of our earth together hold more than two trillion tons of carbon, three times the amount stored in the atmosphere. When the trees are felled or the crops are harvested, the carbon they

Asian agriculture and the greenhouse effecl

269

contain along with soil carbon is oxidized and released to the air by decay or by burning. Since 1860, deforestation has contributed 90- 180 billion tons of carbon to the atmosphere compared with at least 150-190 billion tons from the burning of fossil fuel and natural gas. Agricultural practices lead to the emissions of several gases as well. The burning of forests and savanna grasses in tropical and sub-tropical regions to create pastures and crop land yields additional large amount of carbon monoxide, methane and nitrogen oxides. Moreover soils exposed after forest clearing emits nitrous oxide, as do nitmgen rich fertilizers spread over fields. The concentrations of many gases, given in parts per billion (ppb) are expected to be significantly higher 40 years from now (see Bouwman and Sombroek in Chapter 2) if anthropogenic emissions continue to increase. The concentrations of NO2 and SO2 over highly industrial sites may not rise much in 40 year, but the number of polluted sites can be expected to grow, particularly in the developing nations as in Asia (Graedel and Crutzen 1989). The reduction in the rate of deforestation and initiation of afforestation programmes can delay the onset of global climatic changes in the world. India has more than 50 million hectares of land which is lying wasted. The National Wasteland Development Board established in 1985 was entrusted with the task of transforming the degraded land every year into viable plantations to supply fuel wood as well as fodder in addition to acting as C02 sink. Increased carbon dioxide in the atmosphere could have a favorable impact on crop yield provided it is not simultaneously accompanied by higher temperatures. A study carried out by Sinha, Rao and Swaminathan indicated higher mean temperature will adversely affect the wheat production (Swaminalhan 1989). This is because, for every 0 . 5 " ~increase in temperature and subsequent limited moisture reserves there could be a reduction in crop duration by seven days leading to a loss in yield of 4.0 quintals per hectare.

DEFORESTATION It is a fact that C02 fixed by the forest vegetation from the atmosphere would return when the plants are cut and burnt or when they decay. However, forests account for about two thirds of the photosynthesis taking place on the land and since the forests are long lived, they tend to spread the replenishment of COz to the atmosphere over a long period of time. Deforestation, has far reaching direct and indirect consequences and is one of the most serious human impacts on environment. Deforestation as a cause of ecological degradation has been recognized since long (Thomas 1956) and the current awareness of this problem in developing countries demonstrate the seventy of environmental damage and wood shortage attributable to deforestation (Allen and Bamess 1985). According to the estimates of F A 0 (1981) developing countries will loose 40% of their

270

L. Venkalaralnam

existing forests by the year 2000 at the present rate of exploitation. Available studies indicate that countries like Bangladesh, Nepal, Sri Lanka and Thailand record annual deforestation at a rate of 4%, 3%, 1% and 2% respectively, in 1960s and 1970s (USAID 1979; Revelle 1980; Eriksson 1979). The US InterAgency Task Force on Tropical Forests (1980) concluded, that with the present trend of deforestation, the worlds' tropical forests outside Central Africa and the Amazon Basin, would be nothing but scattered remnants by the year 2025. In India, official estimates show that the country has lost about 4.04 million ha of forest land or about 12% of its total geographical area under forest between 1951-52 and 1975-76 (Puri et al. 1983). According to NRSA (1983) forest cover in India has been reduced from 17.05% to 14.10% of the total geographical area of the country in a span of 7 years (Table 22.2). Table 22.2

Forest cover in India (km2)(NRSA 1983)

Forest type

Closed forest Open forest Mangroves Total Percentage forest area to the total geographical area

1972-75

1980-82

4,64,226 87,683 3,28 1 5,55,180

3,60,229 1,00,592 2,649 4,63,470

16.89%

14.10%

HTMALAYAS The Himalayan mountainous chain, about 2,400 km long and 240-400 km wide rising from low lying Indian plains to well over 8,000 m has a complex physiography situated between the latitudes of 27"-37"N and longitudes 73"-97"E. It is thus a mountainous region cutting across many latitudes, covering the entire tenitorics of Nepal, Bhutan and all or parts of five states of India namely Jammu & Kashmir, Himachal Pradesh, Uttar Pradesh, West Bengal, Sikkim and Arunachal Pradesh. The Himalayas ranges run from West to East. In h e longitudinal plain, South to North, the Himalayas can be viewed as four broad physiographically distinct zones parallel to each other. The outer Himalayas confirm to the climatic zones based on the attitudes as follows: W a n tropical Warm subtropical Cool temperate Alpine Arctic

below 800 m; 800 m -1200 m; 1200 m-1400 m; 2400 m - 3600 m; Above 3600 m.

27 1

Asian agriculture and the greenhouse effecl

In each of the above zones, there is a great variation in precipitation, temperature, wind pattern, humidity, radiation and sunshine which determine the local climates. Kawasa (1988) has prepared vegetation maps of the Himalaya region using Landsat imagery (MSS) of the years 1972-75. The modelling of the Himalayan system has indicated its biological extinction within a matter of a few decades if the present rate of disturbances continues. Forest area lost for various purposes in the Himalayan region including Uttar Pradesh, Jammu & Kashmir, Himachal Pradesh, Arunachal Pradesh, Sikkim and West Bengal as well as Nepal and Bhutan are given in Table 22.3. Table 22.3

Forest area lost for various purposes in the Himalayan region from 19521976

Agriculvalley tural projects purposes River

Country/statcs in India

Jamrnu & Kashrnir Himachal Pradesh Uttar Pradmh Nepal Sikkim & W. Bengal Bhutan Arunachal Pradesh Source: India's Forest

(1980)

10% 0.1 7.7 16.2

10% 0.3 12.2 14.54

Roads

Industries

Miscellanews purposes

Total

10% 0.2 1.2 0.79

10% Nil Nil 3.36

10% 90.2 5.7 3.56

10% 90.8 26.8 38.45

6.1

40.0

Negligible

-

0. I

26.3

0.4

7.1 - :

No information available

EFFECT OF GLOBAL WARMING ON COASTAL AND LOW LYING AREAS Coastal erosion is one of the major environmental hazards faced by maritime states. It assumes great importance for a country like India which has a long coast line of nearly 7000 km with numerous fishing villages, settlements and other establishments right on the shore. Fluvial transport of materials and erosion processes become a focus of attention. Periodic become a focus of attention. Periodic surveys of river basins can provide a reasonable estimates of sediment yields from a drainage basin. Continental fluvial processes and their impact on the ocean require detailed studies of medium size rivers in Asia. Of the total sediment yield to the world oceans, the Indian subcontinent alone contributes about 35% (Milliman and Meade 1983). This will result in silting of the reservoirs reducing their capacity to hold rain water year by year resulting in the reduction of the commandable area.

272

L. Venkataratnam

This would further result in a geographical shifting of the cropping patterns. Conservative scientific estimates revealed that the likely rise in global temperature over the next few decades would result in an eustatic sea-level rise of about 0.5 to 1 m, threatening the islands, low lying areas of deltas and coastal areas. The worst affected areas in Asia would be the Maldives, Indian islands, Bangladesh, and few other countries with long coast lines.

REMOTE SENSING OF LAND DEGRADATION Systematic earth orbital observations began in the year 1960 with the launch of TIROS-1, the first meteorological satellite and since than more than 40 meteorological and environmental satellites with steadily improving sensor data collection capabilities and resolution of the data have been sensing the land surface. In addition to the Landsat, the first satellite designed specifically to collect the data of the earth’s surface, other satellites withe earth resources experiments packages launched by USA, are Skylab, Seasat, HCMM, and space shuttles. Other present and planned satellites around the earth include ERTS & SPOT (Europe), MOS (Japan), IRS (India), Chinasat (China), Radarsat (Canada) and TERS (Netherlands) in addition to the satellites from USSR. In spite of the great strides taken in the machine and manual processing of these data in recent years, so far no uniform methodology or an agreed global albedo data set exists which could lay the basis for current and future research in this field. In order to arrive at comparable interpretations, many scientific groups are mobilizing the scientific community into using satellite data with a prime objective of developing a uniform methodology and albedo data sets for various surface features to ensure a wise usage of the data and their usefulness and repeatability. Twelve July integrations were made with the GLAS (Goddard Laboratory for Atmosphere) GCM (General Circulation Model) to investigate the influence of the surface albedo, surface roughness and evapotranspiration over the Indian subcontinent on the monsoon circulation and rainfall (Sud and Smith 1985). The surface albedo and the surface roughness are governed primarily by vegetation. Any reduction of vegetation increases the surface albedo and decreases the surface roughness. For example, changing forests to agricultural or bare land may imply about a 6- 10% increase in the surface drag coefficient. Their studies also demonstrated that major modifications of the biosphere in the Indian subcontinent might be expected to influence its monsoon circulation and rainfall by altering: 1 ) the surface energy balance; 2) the Planetary Boundary Layer (PBL) motion fields and the moisture convergence; and 3) the hydrological cycle. The results further suggested that the excessive land use via destruction of vegetation would reduce the rainfall and weaken the monsoon. The authors inferred that the feedback effects of changes in land, surface albedo and roughness produced by

273

Asian agriculture and the greenhouse effect

deforestation of the Indian subcontinent, may in turn be partial contributors to the progression of the Thar desert into neighboring arable regions. Robinove et al. (1981) used multitemporal Landsat generated albedo images for studies of indicators of desertification. Comprehensive investigations of atmospheric influence on satellite imagery and derivation of surface albedo from digital Landsat data were presented by Otterman and Fraser (1976) and Otterman et al. (1980). A method for derivation of surface albedo from Landsat data only, without any atmospheric models or ancillary data on the atmosphere, was presented by Robinov et al. (1981). Satellite data have a unique capability in monitoring and mapping natural resources such as crops, forests, water, soils and thus whole environments and provide timely information on the agricultural situation, droughts and floods etc. In India, forest vegetation has been mapped for the years 1972-75 and 1980-83. Wastelands in India have also been mapped for the whole country on a 1:l million scale with village boundaries superimposed on the maps. Similar work is in progress to map the soils affected by salinity and alkalinity at a 1:250,000 scale for the whole country. Once such information is made available, using the remote sensing data, necessary planning can be done for the development of wastelands, reclamation of saline/dkaline soils and conservation of other degraded lands for their optimum utilization. In densely populated South and Southeast Asia, where both food and arable land are scarce, about 90 million hectares of land, climatically, physiographically and hydrologically suited for rice are uncultivated, largely because of land degradation (Table 22.4) (Ponnamperuma 1984). Greening of such vast lands, after proper reclamation measures, by growing crops, forests and grasses would contribute to the creation of new C02 sinks by fixing a significant portion of C02. Table 22.4 Country

Bangladesh Burma India

Indonesia Kampuchea Malaysia

Pakistan Philippines Thailand Vietnam Total

Distribution and extent of problem rice lands, South and Southeast Asia.

Saline soils

Alkali soils

2.5 0.6 23.2 13.2 1.3 4.6 10.5 0.4

0.5 2.5

0.7 0.2 0.4 2.0 0.2 0.2

Peat soils

Total

0.8

4.3 0.8 26.1 31.2 1.5 7.2 14.5 0.4 2.3 3.5 91.8

16.0 2.4

4.0

1.5 1 .o 58.8

Extent, lo6 ha Acid sulfate soils

0.6 1

7.0

.o

5.3

0.2 1.5 20.9

274

L . Venkataratnam

REFERENCES Allen, J.C. and D.F. Bamess (1985). The causes of deforestation in developing countries. Annals of Association of American Geographers. 75: 164-184. Eriksson, J.R. (1979). Energy, environment, and forestry in Sri Lanka: Some major issues. Paper presented in U S AID Asia Bureau Conference on Energy, Forestry and Environment Manila. F A 0 (1981). F A 0 Production Year Book, FAO, UNO, Rome. Graedel, T.E and P.J. Crutzen (1989). The changing atmosphere. Scientific American, 261: 58-68. India's Forests 1980 Compiled by Central Forestry Commission, Ministry of Agriculture (Forestry Division) Govt. of India. Kawasa, M.A. (1988). Remote Sensing of the Himalaya. Natraj Publishers, Dehra Dun, India. Milliman, J.D and R. Meade (1983). World wide delivery of river sediment to the oceans. I. Geology. 91: pp 1-21. NRSA (1983). Mapping of forest cover in India from satellite imagery, 1972-75 and 198082. Summary Report. National Remote Sensing Agency, Dept. of Space, Govt. of India. Otterman, J and R.S. Fraser (1974). Earth-atmosphere system and surface reflectivities in arid regions from Landsat MSS data. Remote Sensing of Environment 5: 247-266. Otterman, J., S. Ungar, Y. Kaufman and M. Podolak (1980). Atmospheric effects on radiometric imaging from satellites under low optical thickness conditions. Remote Sensing of Environment 9: 115-129. Ponnamperuma, F.N. (1984). Adverse soil conditions: their diagnosis and delineation by remote sensing in Applications of Remote Sensing for rice production Edited by Deepak A. and K.R. Rao (1984). A. Deepak Publishing, Hampton, Virginia, USA. Puri, G.S., V.M. Meher - Homji, R.K. Gupta and S. Puri (1983). Forest ecology Phytogeography and forest conservation. Oxford & IBH Publishing Co., New Delhi. Rao. U.R. (1989). The next 4 0 years in space - A view point of developing countries. Presented at the Theme Session. The 40th congress of the International Astronautical Federation, Malaga, Spain 7-13 Oct. 1989. ISRO Special publication No. ISRO-SP4689, India Space Research Organization, Bangalore, India. Revelle, R. (1980). Energy dilemmas in Asia: the need for research and development. Science: 209, pp 169-174. Robinove. C.J., P.S. Chavez (Jr), D. Gehjring, and R. Homgren (1981). Arid land monitoring using Landsat Albedo different images. Remote Sensing of Environment, 11: 133-156. Sud, V.C and W.E. Smith (1985). Influence of local land - surface processes on the Indian monsoon: A numerical study. J. of Climate and Applied Meterology, 24: 1015-1036. Swaminathan, M.S. (1989). Danger of greenhouse gases. Published in Hindu, 5 June 1989. Thomas, W.L. (1956). Man's role in changing the face of the earth. University of Chicago Press, Chicago, USA. Tripathi, C and U. Prasad (1984). Environmental degradation in the marble rocks of Tajmahal, Agra, Uttar Pradesh, India. Proc. Seminar on India's environment, problems and perspectives, Trivandrum, Kerala, India. USAID (1979). Considerations for Bangladesh. Paper presented at the US AID Asia Bureau Conference on Energy, Forestry and Environment, Manila. US Inter Agency Task Force on Tropical Forest (1980). The worlds' tropical forest: A policy, strategy, and program for the Untied States. Report to the President, Washington D.C. U.S Govt. Printing Office.