WORLD BANK ENVIRONMENT PAPER NUMBER 16 The Global Environmental Benefits of Land Degradation Control on Agricultural Land January1999 Global Overlays Program Stefano Pagiola The World Bank Washington, D.C. . B%,4t 'N World Bank Environment Papers Series No. 1 Cleaver, Munasinghe, Dyson, Egli, Peuker, and Wenc6lius, editors. Conservation of WVest and Central African Rainforests/Conservlation de lafor^t dense en Afrique centrale et Oe l Quest No. 2 Pezzey, Sustainable Development Concepts: An Economic Analysis No. 3 Munasinghe, Environmental Economics and Sustainable Developmnent No. 4 Dewees, Trees, Land, and Labor No. 5 English, Tiffin and Mortimore, Land Resource Management in AMach.)akos District, Kenya, 1930-1990 No. 6 Meier and Munasinghe, Incorporating Environmental Concerns into Power Sector Decisionmaking. A Case Study of Sri Lanka No. 7 Bates, Cofala, and Toman, Alternative Policies, Jfor the Control ofAir Pollution in Poland No. 8 Lutz, Pagiola, and Reiche, editors, Economic ,2ndiInstitutional Analyses of Soil Conservation Frojects in Central America and the Caribbean No. 9 Dasgupta and Maler, Poverty, Institutions, ancd the Environmental Resource Base No. 10 Munasinghe and Cruz, Economywide Policies, and the Environment. Lessons from Experience No. 11 Schneider. Government and the Economy on theAAmazon Frontier No. 12 Munasinghe, Global Climate Change. Economic and Policy Issues No. 13 Kramer, Sharma, and MNunasinghe, editors, Valuing Tropical Forests.' .ethoaology and Case Study of Mladagascar No. 14 Current, Lutz, Scherr, editors, Costs, Benefits and FarmerAdoption c AgroJirestr}: Project Experience in CentralAmerica and'the Caribbean No. 15 Pagiola, Kellenberg, Vidaeus, and Srivastava, L`ainstreaming Biodiversity inAgricultural Development. Toward Good Practice Related enviromuent publications Ambler and Marrow, Prioritiesfor Environmental Expendfi!ures in Industry--Eastern Europe,and the Former Soviet Union Somlyody and Shanahan, Municapal WaVsteuater Treatment in Central and Eastern Europe: Present Situation and Cost-Effective Development Strategies Environmental Action Programmefor Central and Eastern Europe: Setting Priorities WORLD BANK ENVIRONMENT PAPER NUMBER 16 The Global Environmental Benefits of Land Degradation Control on Agricultural Land Global Overlays Program Stefano Pagiola The World Bank Washington, D.C. Copyright © 1999 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First Printing January 1999 Environment Papers are pubLished to communicate the latest results of the Bank's environmental work to the development community with the least possible delay. The typescript of this paper therefore has not been prepared in accordance with the procedures for formal printed texts, and the World Bank accepts no responsibility for errors. Some sources cited in this text may be informal documents that are not readily available. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. The World Bank does not guarantee the accuracy of the data included in this publication and accepts no responsibility whatsoever for any consequence of their use. The material in this pubLication is copyrighted. The World Bank encourages dissemination of its work and will normal:ly grant permission promptly. Permission to photocopy items for intemal or personal use, for the internal or personal use of specific clients, or for educational classroom use is granted by the World Bank, provided that the appropriate fee is paid directly to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, U.S.A., telephone 978-750-8400, fax 978-750-4470. Please contact the Copyright Clearance Center before photocopying items. For permission to reprint individual chapters, please fax your request with complete information to the Republication Department, Copyright Clearance Center, fax 978-750-4470. All other queries on rights and licences should be addressed to the World Bank Office of the Publisher at the address above or faxed to 202-522-2422. Stefano Pagiola is an economist in the World Bank Environment Department. Library of Congress Cataloging-in-Publication Data Pagiola, Stefano. The global, environmental benefits of land degradation control on agricultural land: global overlays program / Stefano Pagiola. p. cm. -- (World Bank environment paper: no. 16) Includes bibliographical references (p.). ISBN 0-82:13-4421-8 1. Environmental degradation. 2. Land degradation--Environmental aspects. 3. Land capabiLity for agriculture. 4. Sustainable agriculture. I. Title. II. Series. GE14O.P34 1999 333.76' 15-dc21 98-53728 CIP Contents Foreword ......................................................... v Acknowledgments ......................................................... vi Executive Summary ......................................................... vii Acronyms and abbreviations ........................................................ xi 1. Introduction . ........................................................... 1 2. Global Dimensions of Land Degradation on Agricultural Land ....................... 6 Overview of Land Degradation Problems ........................................... 6 On-site Problems ....................................................... .... 7 Off-site Problems ....................................................... .... 8 Land Degradation's Effects on Global Problems ..................................... 10 Global Benefits of Land Degradation Control Activities ............................... 15 Summary ......................................................... 18 3. Integrating Global Dimensions into Land Degradation Control Projects ............... 19 Land Users' Motivation and Incentives ................. ............................ 19 Approaches to Land Degradation Control .............. ............................ 21 Integrating Global Dimensions: Current Practice ..................................... 24 Sharing the Costs: The Role of the GEF ................. ............................ 25 Implementation Issues ......................................................... 28 Pilot Project Concepts ......................................................... 28 Initial Lessons ......................................................... 34 Conclusion ......................................................... 37 4. Conclusions and Next Steps ...................................................... 38 Glossary ......................................................... 42 References ......................................................... 43 iii iv The Global Environmental Benefits of Land Degradation Control on Agricultural Land Figures Figure 1. Estimates of human-induced land degradation by region, 1945-1990 ..... ......... 1 Figure 2. Conceptual framework ............... ...................................... 4 Figure 3. Estimated relative contributions to global warming from agricultural and non-agricultural sectors . .......................................................... 10 Figure 4. Comparison of carbon stored in soil and in above-ground biomass ..... .......... 11 Figure 5. The soil carbon cycle and some effects of degradation .......................... 12 Figure 6. Application of GEF incremental cost principles ................................ 27 Tables Table 1. Estimated average above-ground biomass and carbon content for different agroforestry types in Sub-Saharan Africa ........................................... 16 Table 2. Pilot project concepts integrating global dimensions in land degradation control .... 29 Boxes Box 1. The Convention to Combat Desertification ...................................... 2 Box 2. Desertification ............................................................... 3 Box 3. How severe is land degradation? Country-level evidence ......................... 7 Box 4. Site-specificity of land degradation ............................................. 9 Box 5. Biodiversity in drylands . ...................................................... 14 Box 6. CAMPFIRE: Community-based wildlife management in Zimbabwe .................. 18 Box 7. Gestion des Terroirs... . ..................................................... 24 Box 8. GEF's COperational Programs .................................................. 26 Box 9. Land Quality Indicators . ...................................................... 39 Foreword This paper was prepared as part of the World practices for country planners and Bank task Bank's Global Overlay Program. The Global managers. The results will help guide national Overlays Program, launched by the World actions to reduce greenhouse gas emissions, Bank in partnership with bilateral donors and conserve biodiversity, and protect international NGOs, seeks to internalize global externalities waters. into national environmental planning and the Global overlays add a new dimension to tra- Bank's sector work, operations, and dialogue ditional sector economic planning by analyzing with governments and partners. It is an itera- environmental impacts and opportunities to tive process, combining conceptual studies, internalize global externalities. This analysis reviews of state-of-the-art techniques for mea- asks: How and at what cost would policies, in- suring and mitigating global externalities, and stitutions, and investment priorities change if testing these concepts and tools in country- global environmental objectives were added to level studies as a means of identifying good conventional sectoral objectives? v AcknoWledgments This paper was written by Stefano Pagiola of This paper has also benefitted from dis- the World Bank's Environment Department, cussions with Jan Boj6, Robert Clement-Jones, under the guidance of Lars Vidaeus, Chief of Jumana Farah, Isabel Valencia, Francisco the Global Environment Division. Pichon, Shiv Singh, and Bachir Souhlal (Africa This paper draws heavily on the existing Region); Douglas Graham and John Kellenberg literature and documentation of lessons learned (Latin American and Caribbean Region); John from Bank and other projects. Background Dixon, Kristin Elliott, Sam Fankhauser, Hassan papers on the linkages between land degrada- Hassan, and Robert Watson (Environment tion and climate change and biodiversity were Department); Doug Forno and Bill Magrath prepared by Susan Leloup (consultant). (Rural Development Department);John English (Operations Evaluation Department); Walter Donovan and Jeff Lewis (Africa Region); Robin Lusigi (GEF Secretariat); Stein Hansen (GEF Sci- Broadfield, Ncreen Beg, Kathy McKinnon, entific and Technical Advisory Panel); Nessim Ernst Lutz, ancd Natsuko Toba (Environment Ahmad, Willem Bettink, and Lorenz Petersen Department); Jiuilian Dumanski and Christian (IFAD); Timo Maukonen (UNEP); Pedro Pieri (Rural Development Department); and Sanchez and Meredith Soule (ICRAF); and Ken King and K anta Kumari (GEF Secretariat). Leonard Berry (Florida Atlantic University). vi Executive Summary Land degradation on agricultural land threat- global effects of land degradation, it is impor- ens the sustainability of growth and the welfare tant to remember that local and national effects of the many people who depend on agriculture are usually the most important. for their livelihoods-including many of the Significant gaps remain in our understand- poorest members of the world's population. It ing of land degradation. Even at the field and can also have adverse effects on problems of national levels, where problems are generally global significance, including climate change, well understood qualitatively, quantitative data biodiversity, and international waters. In some are often insufficient to allow specific problems cases, there may be important complement- to be analyzed. At the global level, the weak- arities between measures that address the na- ness of the quantitative data is compounded by tional and the global aspects of land degrada- a much poorer qualitative understanding. tion problems. The Convention to Combat Desertification Effects on Climate Change (CCD) defines land degradation as a "reduction or loss ... of the biological or economic pro- Emissions. Terrestrial ecosystems are an impor- ductivity and complexity of rainfed cropland, tant carbon sink. The stock of carbon stored in irrigated cropland, or range, pasture, forest and the upper lm of the world's soils is estimated to woodlands" (art.1). This note focuses on land be about 1.5 times the amount of carbon stored degradation in areas used for agricultural in biomass. Soil carbon is lower on agricul- production, including croplands and range- tural lands, but the amount stored is neverthe- lands, and focuses narrowly on instances in less much higher than might be expected from which the productive potential of the land is above-ground biomass alone. However, major reduced as a result of land use practices. The uncertainties remain with regard to the role of effects of land use change in and of itself (inclu- the terrestrial carbon pool. ding deforestation) will not be considered. The question of interest here is the extent to Land degradation can cause problems at which land degradation on agricultural land three levels: affects climate change. * At thefield level, land degradation can result * Does land degradation on agricultural land in reduced productivity. result in increased emissions of greenhouse * At the national level, land degradation can gases? cause problems such as flooding and sedi- * Does land degradation on agricultural land mentation. affect it's capacity to serve as a carbon sink? * At the global level, land degradation can * Can appropriate management enhance both contribute to climate change, and to damage land's productivity and its capacity to store to biodiversity and international waters. carbon? The relative importance, and specific nature, of Unfortunately, data to answer these questions problems at each level varies substantially from are scarce. Most attention has focused on emis- case to case. Although this note focuses on the sions under specific land uses, and on the vii viii The Global Environmental Benefits of Land Degradation Control on Agricultural Land impact of land use change (in particular, on the Below-ground biodiversity. The main direct impact of defo:restation). Very little work has adverse effect of cropland degradation on bio- been done on the effect of land degradation diversity is likely to be on below-ground biodi- within a given Land use on emissions. versity. Diverse and abundant organisms help Carbon cycle in soils. Because croplands and maintain soil fertility and productivity and are rangelands tenid to have relatively low above- fundamental to soil quality. Degradation of soil ground biomass, the linkages between land physical and chemical conditions can damage degradation and climate change are likely to this biodiversity, about which relatively little is come primarily from changes in soil carbon. known. * Some actions which cause land degradation Indirect effects. In most cases, the greatest can increase carbon emissions directly. For impact of cropland degradation on biodiversity example, burning crop residues causes both is likely to be indirect. By reducing productivity fertility loss, by preventing the return of on existing agricultural land, degradation can nutrients to the soil and reducing the build- force farmers to clear additional areas of natural up of soil organic matter, and increased habitat to maintain production. It should be emissions. remembered, however, that land degradation is * Some forms of degradation reduce soil car- not the only cause of agricultural expansion. bon. For example, erosion can carry away Rangeland. Rangelands tend to be less modi- soil organic matter. However, this does not fied from their natural state than cropland and always lead to increased emissions since car- to contain a much greater proportion of their bon carried away by erosion is often depos- original biodiversity. Livestock often shares ited under conditions where it may be well rangelands with considerable wildlife. Degra- preserved, such as in reservoirs. dation, therefore, can cause relatively more da- * The consequonces of land degradation also mage to biodiversity on rangelands than on adversely affect the soil carbon cycle. Lower croplands. production D)f crops and pasture, results in Many of the factors that cause pasture degra- lower carbon inputs in subsequent periods dation are also likely to have an adverse impact (less root material, less leaf litter, less crop on biodiversity. For example, both livestock residue), thus reducing carbon storage. and wildlife will suffer if access to areas that Land degradation on croplands and range- provide critical grazing or water at times of lands is thus likely to reduce the ability of soils stress are restricted. As both livestock and wild- to serve as a carbon sink and release carbon life are restricted to smaller, often less favorable currently stored in soils to the atmosphere. The areas, competition between them is likely to be magnitude of this effect is difficult to estimate, exacerbated. however. Effects on International Waters Effects on Biodiversity Many off-site consequences of land degradation Croplands. Croplands are substantially modified may be experienced beyond national borders. from their original, natural state, and their Sedimentation or flooding problems caused by levels of biodiversity are generally substantially degradation in an upstream watershed, for lower than those of natural habitats. Never- example, may affect a country downstream. theless, agricullural landscapes can contain con- Damage to international waters are a special siderable biodiversity. The concern here is case of the off-site damages. They are similar whether land degradation might further reduce to off-site effects felt nationally, except that na- the remaining biodiversity. tional policymakers have no incentives to take them into account. The problems of identifying Executive Summary ix and measuring cause-and-effect relationships tional or different measures may be required to are particularly severe in the case of inter- fully realize global benefits. national waters, since data collected in up- Any effort to incorporate global dimensions stream and downstream countries may not be into land degradation control must begin with compatible. a well-thought out strategy to address the local and national aspects of the problem. This requi- Global Benefits of Land Degradation Control res a clear understanding of the nature, extent, and severity of land degradation problems, Measures to control land degradation also have their causes, and their effects at both the farm effects at the field, national, and global levels. and national levels. It also requires a clear The specific range of benefits obtained depends understanding of the incentives and constraints on the measures used and the conditions under faced by land users. which they are applied. At this stage, it should be possible to deter- Land degradation control can help reduce or mine whether the measures already envisaged halt the adverse global effects of land degra- are sufficient to address the global problems dation. Some land degradation control practices originating at the site, or whether additional or can also have global benefits in and of them- different measures are required to do so. In selves, by stimulating additional carbon seques- such cases, it may be possible to obtain funding tration and/or biodiversity conservation over from the Global Environment Facility (GEF) to and above what might have occurred even in finance the incremental costs of the additional the absence of degradation. Practices likely to measures. It is important to realize, however, have positive effects on problems of global that not all activities that generate global bene- concern include agroforestry, which allows in- fits are eligible for GEF funding. The GEF's creased carbon sequestration while continuing Operational Programs set specific priorities as with crop production and which often provides well as specifying eligibility criteria. a more hospitable environment for biodiversity, The main difficulties likely to be encountered and community-based wildlife management, which by efforts to incorporate global considerations can provide an alternative to insustainable use in land degradation control activities are: of some marginal areas. Information. Information on the nature and magnitude of the adverse global effects of Integrating Global Dimensions into Land g .. Degradation Control Projects land degradation under specific conditions is Degradation Control Projects extremely scarce. * Implementation. To be successful, land degra- The primary reason for efforts to control land dation control programs need to obtain the degradation on agricultural land is to reduce, cooperation of land users. Insufficient atten- arrest, or reverse the field-level or national tion to the constraints and incentives land problems it is causing. Given the linkages users face has led to the failure of many land between land degradation and problems of deradationconthects. global concern, however, global benefits may degradation control projects. also be generated in some cases. Pilot Project Concepts When linkages between land degradation and problems of global concern exist, efforts to In view of the limited experience in the pre- control and reverse them can be mutually paration of projects that blend global environ- supportive. Land degradation control activities mental concerns with land degradation control, can be win-win in the sense that they reduce the World Bank and the International Fund for both the local and the global effects of land Agricultural Development (IFAD) have colla- degradation. In some cases, however, addi- borated in developing a pipeline of projects in x The Global Environmental Benefits of Land Degradation Control on Agricultural Land this nascent area of GEF operations. Project con- land degradation. The World Bank is also com- cepts were developed for Botswana, Mali, mitted to addressing global environmental Jordan, India (2 projects), Mongolia, Belize, and problems. Through its Global Overlays Pro- El Salvador. Although these pilot project gram, the World Bank is seeking to incorporate concepts are only illustrative of the potentials attention to global problems throughout its and pitfalls of attempting to integrate attention work. This includes seeking more syste- to global benefits into land degradation control matically to assess the impacts on biodiversity projects, they do provide some initial lessons. and other global externalities that land degra- The projects which seem to lend themselves dation and its control might generate, and in- best to integrating global dimensions are those corporating consideration of these issues in in which field activities are being carried out in policy and project responses. specified areas. It is far more difficult to identify potential globaLl dimensions of projects that Conclusion seek to combat land degradation through policy reforms or support to research and extension. For many countries-and in particular for Conversely, whLen field activities are being car- many African countries-land degradation on ried out, identifying possible links to problems agricultural land is posing substantial threats to of global concern is substantially easier-parti- sustainability, economic growth, and the wel- cularly in the case of links to biodiversity. fare of the rural population. Strong efforts to The experience of the pilot projects also combat land degradation are justified on these suggests that establishing the nature and extent grounds alone. In some cases, reduction of of linkages betwxveen land degradation on agri- problems of global concern such as mitigation cultural land and biodiversity is simpler than of climate change or conservation of biodi- doing so for climate change. In the case of versity provide an additional reason to combat biodiversity, the main constraint is that the bio- degradation. At times this may require addi- diversity at risk from land degradation on agri- tional or different measures than if local and cultural land may not have been as well studied national considerations were the only ones as biodiversity in protected areas. In the case of involved. climate change, the main constraint is that very In cases where there are strong linkages be- few data exist on changes in emission resulting tween land degradation and problems of global from changes within a given land use (as oppo- concern, efforts to combat both can be mutually sed to changes in emissions resulting from supportive. It is important, however, to re- changes in land use itself, such as defores- member that the primary motivation for land tation). The data requirements are also more degradation control efforts will remain the local stringent in the case of climate change, due to and national benefits that can be derived the need to demonstrate cost-effectiveness. thereby. Linkages to global problems are not always present, or may not be sufficiently The Role of the World Bank strong to warrant specific attention. The World Bank is devoting considerable re- sources to assist its client countries to combat Acronyms and abbreviations ASR Agricultural Sector Review BSAP Biodiversity Strategy and Action Plan CBNRM Community-Based Natural Resource Management CBD Convention on Biological Diversity CCD United Nations Convention to Combat Desertification CESP Country Environnmental Strategy Paper CGIAR Consultative Group on International Agricultural Research CH4 methane CO2 carbon dioxide EIA Environmental Im,pact Assessment ESW Economic and Sector Work FAO Food and Agriculture Organization of the United Nations FCC United Nations Framework Convention on Climate Change GEF Global Environment Facility GHG greenhouse gas IFAD International Fund for Agricultural Development LQI Land Quality Indicators NAP National Action Program NARS National Agricultural Research System NEAP National Environmental Action Plan NGO non governmenta:l organization N20 nitrous oxide OP Operational ProgrLam PDF Project Development Facility SFI Soil Fertility Initiative UNDP United Nations Development Programme UNEP United Nations Environment Programme USDA United States Department of Agriculture ha hectares Mg megagram (1 x 106 grams = 1 metric tonne) Pg petagram (1 x 1015 grams = 1 billion metric tonnes) All monetary amounts are in U.S. dollars. xi 1. Introduction Since 1945, an estimated two billion hectares of many people who depend on agriculture for agricultural land-almost 18 percent of the their livelihoods-including many of the earth's vegetated land-have been degraded as poorest members of the world's population. a result of human activity [Oldeman and others, The urgency of addressing land degradation 1990]. Of these, an estimated 1.2 billion hect- problems has been reiterated by the adoption ares, or almost 11 percent of the earth's vege- and ratification of the Convention to Combat tated land, have been moderately or strongly Desertification (CCD), which came into force in degraded, implying that productivity has been December 1996. Under this convention, many significantly reduced (Figure 1). Land degra- countries have committed themselves to dation poses a substantial threat to the sustain- combating land degradation (Box 1). ability of development [World Bank, 1992]. In Although many of the consequences of land the drylands of Sub-Saharan Africa in parti- degradation are experienced either in the speci- cular, land degradation is thought to be an fic areas being degraded or in neighboring important impediment to agricultural growth areas, it is increasingly thought that land degra- [Cleaver and Schreiber, 1994]. dation can also have adverse effects on prob- Controlling land degradation on agricultural lems of global significance: land is important to the objectives of sustain- * Climate change. Land degradation can contri- able growth and increasing the welfare of the bute to global climate change by leading to emission of greenhouse gases (GHGs) or by reducing the ability of terrestrial ecosystems Figure 1 to act as a sink for these gases. Estimates of human-induced land * Biodiversity. Land degradation can result in degradation by region, 1945-1990 loss of biodiversity, both in the areas de- 800 graded and by inducing additional clearing of natural habitats. 600 * International waters. Land degradation can result in damages to shared international i 400 E S water bodies. In many cases, there are likely to be important 200 _ l ji complementarities between measures that 200 S Eaddress the national and the global aspects of o _ _ | l * land degradation problems. Africa Asia South and Other Under the terms of the Instrument Estab- Central lishing the Restructured Global Environment America Facility (GEF), the agreed incremental costs of 7 Strong degradation: Agricultural use impossible activities concerning land degradation, pri- Moderate degradation: Significant productivity decline atvte ocrigln erdto,pi Light degradation: Small productivity decline marily desertification and deforestation, as they Source: From data in Oldeman and others, 1990 relate to the four focal areas of the GEF (climate 1 2 The Global Enzironmental Benefits of Land Degradation Control on Agricultural Land Box 1. The Convention to Combat The World Bank is committed, through its Desertification environmental agenda, to addressing global environmental challenges. As one of the imple- The United Nations Convention to Combat menting agencies for the GEF, the World Bank Desertification (COCD) became effective in December . 1996, following its ratification by 50 signatories. To aovhas directressibl els cloent date, 150 countries have ratified the CCD. governments to address problems of global The objectives of the Convention are to combat concern. Although land degradation control has desertification (defined as land degradation occurring the potential of contributing to alleviation of in arid, semi-aricl, and dry sub-humid areas) and global problems in these areas, there is little mitigate the effects of drought in countries that experience with integrating global consider- experience them. The Convention gives priority to ations into land degradation control activities. Africa while not neglecting other regions (art.7). Parties to the convention commit themselves to Definitions. The term 'land degradation' has adopting an integrated approach to addressing the been used to describe a wide variety of prob- physical, biological, and socioeconomic aspects of lems. This study uses the definitions in the desertification and drought. Efforts to combat deser- Convention to Combat Desertification (art.1): tification are to be integrated with strategies for pO- verty eradication. To meet the objectives of the CCD, * Land means the terrestrial bio-productive parties are to be guided by three principles: system that comprises soil, vegetation, other * stakeholder participation; biota, and the ecological and hydrological * international coDperation; and processes that operate within the system; * consideration of the specific needs of affected and developing countries. Land degradation means reduction or loss ... Affected parties are to develop National Action Programs (NAPs) which will of the biological or economic productivity * define and promote preventive measures; and complexity of rainfed cropland, irri- * enhance climatologic, meteorological, and hydro- gated cropland, or range, pasture, forest and logical capabilities; woodlands resulting from land uses or from . strengthen institutional frameworks; a process or combination of processes, in- * provide for effective stakeholder participation; and cluding processes arising from human acti- * review implementation regularly. Subregional and Regional Action Programs (SRAPs), vities and habitation patterns, such as (i) soil which have the same basic features, will support im- erosion caused by wind and/or water; (ii) plementation at a regional level. deterioration of the physical, chemical and The Convention will not have any financing of its biological or economic properties of soil; and own. However, it establishes a Global Mechanism ... (GM) which airns to improve management, mobili- (111) long-term loss of natural vegetation. zation, and coordination of funds for combating deser- For the purpose of this study, these defini- tification. Developed country parties are committed to tions are interpreted narrowly to focus on in- providing assistance to affected countries to implement stances in which the productive potential of the their NAPs, to provide financing and other forms of support, and to mobilize new and additional funding unSerlying Lan reduced as a result of land from the GEF fcr the agreed incremental costs of use practces. Land use change i and of itself activities concering desertification that relate to the will not be considered land degradation. Only GEF's four focal areas. if the new use results in damage to the land's productive potential-and hence results in either falling yields or in the need for higher in- change, biodiversity, ozone depletion, and put levels to maintain yields-will it fall within international waters), are eligible for funding the scope of this study. (art.I, para 2). Lamd degradation control is also Land degradation problems occur in forest integrated in the GEF Operational Strategy land, in cropland, and in rangeland. This paper [GEF, 1996a; 1997]. focuses on problems encountered in cropland Introduction 3 and rangeland-the areas used for agricultural Box 2. Desertification production. A separate Global Overlay study addresses deforestation and forest management The term 'desertification' is avoided in this study, for problems [Kellenberg and Cassells, forth- two main reasons. First, it has been subject to a myriad coming]. Within agricultural land, only effects of definitions over the years, many of them vague and related to land degradation are consided. imprecise. It is not, therefore, a useful term from either related tolanddegradationareconsidered. an analytical or a descriptive perspective [Nelson, 19901. Many aspects of agricultural production can Second, considerable controversy has developed over affect problems of global concern even in the whether any phenomenon which might reasonably be absence of degradation. A previous Global described as desertification even exists. Analysis of Overlays study [Pagiola and others, 1997] has vegetation change in the Sahara by NASA, for example, already examined the interactions between clearly shows that vegetation fluctuates significantly agricultural development ,a,. biodiversity. [Tucker and others, 1991 ]. The current northem limits of vegetation are very similar to those preceding the severe Land degradation can cause problems at droughts of the 1970s and 1980s. The image of three levels: advancing sand dunes which gave desertification its * At the field level, land degradation results in emotional impact, therefore, is a poor representation of reduced productivity. reality. Even the Convention to Combat Desertification, despite * At the national level, land degradation results its name, moves away from the term by defining in a range of problems such as damage to desertification as "land degradation in arid, semi-arid downstream infrastructure through flooding and dry sub-humid areas" (art. 1). and sedimentation, reductions in water qua- lity, and changes in the timing and quantity of water flows. (including the opportunity cost of their family * At the global level, land degradation can labor), and by the characteristics of the farm contribute to climate change (through in- household. In turn, many of these elements are creased emissions of greenhouse gases and influenced, to varying degrees, by agriculture changes in the ability of terrestrial eco- and non-agriculture policies, institutions, and systems to serve as carbon sinks), to damage development programs. to biodiversity (both directly in degraded The resulting land use practices will, of areas and indirectly by inducing expansion course, affect the level of production, whether of cultivated areas), and to damage to inter- that production be a crop or pasture. Some land national waters (through sediment loads and use practices will also affect the quality of the changes in hydrological cycles). land; in particular, some land use practices can The relative importance, and specific nature, of lead to land degradation. For example, growing problems at each level varies substantially from crops that leave the soil exposed can result in case to case. erosion, and over-grazing can result in compac- Conceptualframework. The conceptual frame- tion of pasture lands and in a change in species work used in this paper is based on a number composition. Some-often most-of the effects of propositions that are embedded in Figure 2. of degradation are felt on-site, at the field level. Most land resource management decisions are To the extent that they are, they will affect the made by individual land users such as farmers future benefits that farmers can obtain from the and pastoralists, not by national planners. The land. Land-users, therefore, generally have a incentive structure under which land users direct incentive to take on-site effects into ac- make land resource management decisions is count in their management decisions. influenced by the technology they have avail- Degradation can also have off-site effects able, by the structure of marke-ts they operate in such as sedimentation. Land users generally do (including markets for the land itself), by the not have any incentive to take these effects into prices they face for their inputs and outputs account. Since these problems affect national 4 The Global Enmironmental Benefits of Land Degradation Control on Agricultural Land Figure 2 Conceptual framework Agricultural sector Non-agricultural policies, policies, institutions, and institutions, and development programs development programs iold ~Prices l l Technology l l Institutions l __,| La-nd resource management| r decisions Prdcto ofcos i Land quality | Off-site impacts | | Impacts on Off-site impacts global problems l Farm- level National Global ll cost and benefits cost and benefits costs and benefits I ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~L - - - - - - costs and benefits, national planners have an dies in the Global Overlay Program, this study incentive to modify policies or design project asks the question: How and at what cost would interventions to attempt to direct farmers' land policies, institutions, and investment priorities resource management decisions in socially opti- change if global environmental objectives were mal ways. added to conventional sectoral objectives? The Land quality changes can also affect prob- basic aspects of land degradation problems at lems of global c:oncern, through their effect on the field and national levels are relatively well carbon sequestration, on-site and off-site biodi- understood, at least in a qualitative sense. versity, and international waters. In the past, There is also a substantial body of work on how there has been no feedback from these prob- to design projects that address land degrada- lems, and so they have been ignored both by tion problems at these levels. In contrast, im- farmers and by national planners. portant gaps remain in our understanding of This study focuses on the missing link in the global dimensions of land degradation. Figure 2, indicated by the dotted line: on Moreover, there is little experience in preparing . . ... . , ~~land degradation control projects that blend measures to address the global dimensions of global environmental concerns with national land degradaticn and on how to integrate them globleevironment oncrnsw with measures designed to address farm-level sustainable development objectives and national-level problems. As with other stu- Introduction 5 Key issues. This paper examines the concep- tives and constraints faced by land users. It tual, methodological, and empirical issues then reviews previous efforts to address land involved in designing interventions to address degradation problems and the lessons which land degradation problems and global environ- have been learned from these efforts. If land mental benefits in developing countries. The degradation control efforts are modified to 'go issues to be addressed include: the extra mile' to generate additional global * The extent and nature of global benefits benefits, it may require land users and national resulting from control and mitigation of land authorities to undertake actions which do not degradation problems. bring direct local or national benefits. This may * The relationship between global and domes- require external compensation from agencies tic benefits of addressing land degradation such as the Global Environment Facility (GEF). problems. The chapter then reviews a number of pilot pro- * How consideration of global problems can ject concepts prepared by the International be incorporated in the design of land degra- Fund for Agricultural Development (IFAD) in dation control projects ancd how the costs of collaboration with the World Bank that in- such projects might be shared among the tegrate attention to attention to global problems beneficiaries. in land degradation control projects. Chapter 2 discusses the global dimensions Chapter 4 summarizes the discussions and of land degradation from a technical viewpoint. indicates the next steps required to advance the The chapter begins with a brief overview of the process. on-farm and national effects of land degra- Much of the discussion in this paper is dation, before examining the relationship of broadly applicable to land degradation prob- land degradation to problems of global concern lems worldwide. However, the emphasis is pri- -climate change, biodiversity, and interna- marily on conditions encountered in the dry- tional waters. As will be seen, significant gaps lands of Sub-Saharan Africa-the area where remain in our understanding of land degra- land degradation problems are thought to be dation problems. Even at the field and national most pressing. This emphasis is consistent with levels, where problems are generally well un- the CCD, which gives priority to Africa while derstood qualitatively, quantitative data are in not neglecting other regions (art.7). most instances insufficient to allow specific This aper should not be interpreted as a problems to be analyzed. At the global level, p p p . . . ~~~good practice handbook on incorporating glo- the weakness of the quantitative data is com- bal dimensions in land degradation control. pounded by a much poorer qualitative Too little work has been done in this field to understanding. The chapter then discusses n date, and too many gaps remain in our know- what ways land degradation control efforts ledge of the effects of land degradation on might generate global benefits, and examines global problems. Rather, this paper is a first some measures which appear particularly likely step in an intellectual journey that should ulti- to generate global benefits. mately lead to good practice. Because land Chapter 3 discusses land degradation con- degradation problems are so diverse, many trol programs and how attention to global generalizations and an abundant use of quali- problems might be incorporated into them. The fiers are inevitable. chapter begins with a discussion of the incen- 2. Global Dimensions of Land Degradation on Agricultural Land Land degradation on agricultural land causes a be particularly high in Africa, where about 320 wide array of problems, at many different million ha are moderately or strongly degraded levels. Problems at the field and national level (see Figure 1 on page 1 above). Stoorvogel and have been studied extensively and, although Smaling 11990] estimate that, over the last 30 quantitative data are often missing, are rela- years, an average of 660 kg/ha of nitrogen, 75 tively well understood. The global dimensions kg/ha of phosphorus, and 450 kg/ha of potas- have been much less studied, and our under- sium have been lost from cultivated land in standing is weak even at a qualitative level. Sub-Saharan Africa, primarily because of insuf- This chapter begins with an overview of the ficient use of nutrient inputs to offset nutrient field-level and national-level consequences of output through crop removals and other losses. land degradation problems. Because the differ- Considerable debate has arisen, however, ent aspects of land degradation problems are over the magnitude of the problem. Recent closely intertwined, any efforts to address the reviews of available evidence on land degra- global dimensions of land degradation require dation in several countries challenge the more an understanding of its local effects as well. The catastrophist views of its nature, extent, and chapter then describes current knowledge on severity (Box 3). What emerges in these and the possible effects of land degradation on other studies is a much more nuanced picture problems of global concern: climate change, of land degradation, in which site-specific prob- biodiversity conservation, and international lems dominate and in which a complex series of waters. causes lead to degradation. Data limitations. Any effort to examine land Overview of Land Degradation Problems degradation problems soon runs into formi- dable data limitations. Despite decades of work Agriculture is dependent upon natural bio- on land degradation, the available information physical environments; degradation of these remains highly fragmented, incomplete, and environments through misuse or over-use has often unreliable. In particular, much of the re- led to considerable concern that agricultural search on land degradation stops short of production may not be sustainable [World addressing the productivity issues which are of Bank, 1992; World Resources Institute, 1992]. A fundamental interest. Few of the many studies global assessment commissioned by the United which examine erosion problems, for example, Nations Environment Programme (UNEP), go beyond measuring erosion in tons of soil lost found that almost 11 percent of the earth's per hectare and also measure the resulting vegetated land has been moderately or strongly effect on crop production. Although a range of degraded, implying that productivity has been efforts are underway to remedy these problems significantly reduced [Oldeman and others, (such as the Land Quality Indicators Program, 1990]. The extent of degradation is estimated to see Box 9 on page 39 below), the gradual and 6 Global Dimensions of Land Degradation on Agricultural Land 7 Box 3. How severe is land degradation? growth in Sub-Saharan Africa [Woomer and Country-level evidence Swift, 1994; Sanchez and others, 1996b, 1997]. Other important forms of degradation include Recent reviews of available evidence on land salinization, which is an important threat to degradation in several countries challenge the more irrigated areas in semi-arid regions [Umali, catastrophist views of its nature, extent, and severity. 1993]. In many cases, different forms of degra- * Kenya. A review of the experience of the Machakos dation are correlated. For example, soil com- district in Kenya, where land degradation problems have been widely noted beginning in the 1930s, has paction can result in increased runoff and, shown that increasing population pressure need not hence, in higher rates of erosion; conversely, lead to increased degradation. Despite a sixfold erosion can carry away nutrients and weaken increase in population, productivity increased both the soil's physical structure. An important char- in per hectare and per capita terrns, and degradation acteristic of such damage is that it is usually was controlled and even reversed [English and cumulative; its effects in any one year can be others, 1992; Tiffen and others, 1994]. * El Salvador. Pagiola and Dixon [1997] examined minor or insignificant, but become important as available data on land degradation in El Salvador, they accumulate over time [Lal, 1987]. where the common perception is that '75% of the Despite years of concern over the effects of country's surface is degraded', and found that only l about a third of farmers' fields experience erosion, land degradation on cropland, critical gaps and only a fraction of those appear likely to suffer remain in available data. In particular, much of productivity declines as a result. the data are purely qualitative. Land is vari- None of these studies suggests that land degradation ously classified as 'undegraded', 'degraded', or is not a problem. All, however, seriously question the 'severely degraded', but the meanings of these common perception of the severity and nature of land labels are seldom made precise. Data on the degradation.~~ael resldmmaePrcse at n h degradation. productivity effects of degradation are parti- cularly scarce. Examination of available data on crop yields cumulative nature of land degradation means do not provide strong evidence that degra- any effort to collect improved data willnecessa- dation is affecting productivity. On the con- trary, yields appear to have increased in many On-site Problems countries, including many in Sub-Saharan Africa. (In much of Africa, however, this in- crease has been slower than population growth, The on-slte effects of land degradation on so that per capita production has fallen.) Taken agricultural land are a major source of concern, alone, however, these data do not disprove the since they threaten the sustainability of agri- eitne, ho degradation. cultural production and the welfare of a sub- . National yield data are often suspect, both stantial portion of the world's population.intrsodaaqltyndntemofep in terms of data quality and in terms of rep- Cropl and resentativeness. If data collection tends to favor more prosperous farmers using im- Land degradation on croplIand can take a proved techniques, for example, it may miss variety of forms. Traditionally, erosion and its degradation problems experienced by the consequences have attracted the most attention bulk of farmers. [Brown and Wolf, 1986]. More recently, interest Both agricultural technology and input has focused on the effects of nutrient depletion. levels have been increasing (although again, The fertility constraints resulting from soil or- oregions), and then ma ganic matter and nutrient depletion are thought other regtons), and these emprovements may to be a major impediment to agricultural 8 The Global Environmental Benefits of Land Degradation Control on Agricultural Land dation. In the absence of degradation, yield Rangelands in arid and semi-arid systems are increases might have been faster. now recognized as being highly variable over * Expansion of agriculture into new, as yet un- both space and time (so that much apparent degraded, areas may mask the effects of degradation was actually the temporary effect degradation on existing agricultural land. of low rainfall). Their productivity at any given Continued expansion will be increasingly time is driven primarily by factors such as difficult, however, and will bring into use rainfall, rather than by stocking density. More- increasingly marginal land. over, considerable evidence now demonstrates * Data on average national yields may well that traditional pastoral strategies, which rely conceal significant regional variations. on extensive movement in response to stress, Although efforts are clearly needed to improve are not only well-suited to their environment, available data on the productivity effects of but can also outperform 'modern' ranching land degradation, it would be unwise to as- strategies [Scoones, 1995]. sume from the current weak evidence that such Here too, a realization that conventional wis- problems do not exist. dom may have exaggerated and mis-diagnosed problems does not imply that rangeland Ranrgelands degradation does not occur. The causes of deg- radation, however, lie more in encroachment To an even greater extent than cropland, range- into grazing land by settled agriculture, in land-particularly in Africa-has been per- obstacles to movement by livestock in response ceived as suffering from widespread degra- to climatic variations, and in the breakdown of dation. Indeed, conditions in semi-arid and arid traditional communal management arrange- rangelands played a major role in the popular ments than in excessive livestock numbers. Go- perception of desertification. Dregne and others vernment subsidies during times of low rain- [1992], for example, estimate that the majority fall, although they may alleviate short-term of the world's rangelands are moderately or se- hardship, may cause long-term problems by verely degraded. The main problem was diag- maintaining high stocking rates and preventing nosed as being overstocking of livestock by ecologically normal regeneration of vegetation pastoralists, resulting in overgrazing which re- after drought [Pratt and others, 1997]. moved vegetation, changed species compo- sition towards unpalatable species, and com- Off-site Problems pacted soils. This was thought to result in a collapse-perhaps irreversible-of rangeland In addition to affecting the productivity of the productivity, drastically reducing its carrying land directly affected (on-site effects), some capacity. forms of degradation can also cause damages As with cropland, however, there is a grow- elsewhere (off-site effects). In particular, erosion ing realization that this image may be mis- can cause economic damage to reservoirs and leading. Recent research has argued that the waterways and to aquatic life within them. In extent of dryland degradation is much lower some cases, off-site effects can be much more than has been believed and that rangelands are important than on-site effects. In the USA, for often more resilient than had been thought. example, changes in soil productivity have been Contrary to conventional wisdom, animal pro- estimated to be relatively minor compared to duction per hectare and meat production per off-site costs [Crosson and Stout, 1983; Clark head in the Sahel has increased over the last 30 and others, 1985]. In developing countries, on years [Steinfeld and others, 1997]. The diag- the other hand, on-site productivity concerns nosis of the problem has also been changing tend to be dominant. Magrath and Arens substantially [Behnke and Scoones, 1993]. [1989], for example, estimate that productivity Global Dimensions of Land Degradation on Agricultural Land 9 effects account for 95 percent of the costs of soil Box 4. Site-specificity of land degradation erosion in Java (although several categories of off-site costs could not be quantified). Repetto An important aspect of land degradation problems, at and Cruz [1991] obtain similair results in Costa all levels, is their site-specificity. Rica, albeit from weak and incomplete data. Field-level problems. Soil characteristics vary widely from place to place, depending on climate, parent Sedimentation. The off-site effect which has material, topography, biotic activity, and the length of led to the most concern is the sedimentation of time that soil formation has been underway. In the reservoirs and waterways resulting from up- United States alone, 13,000 distinct soil series have been stream erosion. In Morocco, for example, it is identified [Brady, 1986]. Climatic variations and crop estimated that up to 0.5 percent of national interactions add additional layers of complexity. Site- esmated sthrate uapaito 0 loct ofnnational specific variations in conditions can lead to significant water storage capacity iS lost annually to differences even within small geographical areas. sedimentation, a large amotnt for a country Finally, differences in economic conditions will mean with rising water shortages [Agro-Concept, that optimal land management practices and responses 1994]. Sedimentation can have a number of to land degradation will vary even within similar ego- adverse consequences: ecological areas. . Higher costs of hydroelectricity production National problems. The off-site effects of land degradation also vary substantially according to local and shorter reservoir lifetimes; conditions. The effect of sedimentation, for example, * Damage to irrigation systems (reductions in will depend on whether there is any valuable infra- irrigated area, higher costs of canal cleaning, structure downstream to be damaged. All watersheds damage to pumping equipment); do not feed the reservoirs of hydroelectric dams or irri- * Reduced ability to regulate streamflow, gation systems. .the danger of floodin; Attention to site-specificity is closely connected to increasing flooding, . participatory approaches to land degradation control. • Damage to fish stocks and other aquatic Only the land users themselves are likely to have lifeforms through increased water turbidity sufficient information on the agro-ecologic and socio- and burial of spawning grounds and coral economic conditions they face to be able to select the reefs. most appropriate management practice. These costs can be avoided by dredging re- servoirs and waterways to remove sediment or by building new infrastructure to replace that Only a small fraction makes its way into which has been silted up, but this is costly. streams. And once in streams, sediment tends However, the contribution of anthropogenic to be deposited on the streambed and mobi- land degradation tosedimenLationproblems is lized several times before it ever reaches a often not known. In many cases, it is implicitly reservoir or other vulnerable infrastructure The assumed that all sedimentation is anthro- me lag between erosion on a field and the con- pogenic. This ignores the often very high rates sequent effect, therefore, is often measured in of background erosion in many areas. In decades [Walling, 1988]. This has very im- Morocco, for example, the bulk of sediment is portant consequences for the evaluation of the now thought to originate 1irom streambank benefits of interventions designed to reduce erosion and not from erosion on agricultural sedimentation. land. Even among anthropogenic sources, agri- Hydrological changes. Another major category cultural land may not be the most important. of off-site effects which has caused concern are On a per hectare basis, roads are often the possible changes in hydrological patterns. If single biggest source of sediment. Even when land degradation reduces infiltration rates, cropland and rangeland contribute to sedi- more of the rainfall will run off, resulting in: mentation, it is difficult to establish the contri- * Reductions in dry season streamflow, re- bution of specific fields. Most soil eroded from ducing the availability of drinking water and a given field is simply re-deposited nearby. reducing water supplies to irrigation. 10 The Global Environmental Benefits of Land Degradation Control on Agricultural Land * Increases in stormflow, resulting in flooding Figure 3 downstream and increased riverbank and Estimated relative contributions to streambed scour, thus increasing down- global warming from agricultural and stream sedimentation. non-agricultural sectors * Reductions in aquifer recharge. 80 Changes in the timing, volume, and velocity of water flow and groundwater recharge, can also 70 - alter natural lake, riverine, estuarine, and t marine habitats, with adverse consequences for 60 aquatic and riparian ecosystems. Most work on 2 50_ hydrological changes has focused on the con- > , sequences of land-use change in forested water- 40 - Temperate Tropical sheds (see Chomitz and Kumari [1996] for a review). Information on the effects of land 30- degradation on agricultural land on hydro- c 20 logical patterns is limited. PI Water quality changes. In addition to changes 10 - in the quantity and timing of streamflows, land degradation can often have adverse conse- 0 quences on water quality. Sediment increases Non-Agriculture Agriculture the turbidity of streams, damaging fish and _ 02 < CH, II N0 other aquatic lifeforms. It also increases the m costs of potable water supply. In addition, nu- Source: Adapted from Duxbury, 1995 trients lost from agricultural land can cause problems when they collect in waterways, by mates of the impact of human activities on stimulating the growth of algae and other emissions of the main greenhouse gases- plants which deplete the available oxygen. carbon dioxide (CO2), methane (CH4), and ni- trous oxide (N20)-show a predominant influ- Land Degradation's Effects on ence of non-agricultural activities (Figure 3). Global Problems These represent mainly the combustion of fossil fuels, which is accompanied by high carbon Land degradation can contribute to global cli- dioxide emissions. Within the agricultural sec- mate change, to loss of biodiversity, and to tor, the main greenhouse gas emissions are of damaging international waters. Land degra- carbon dioxide and methane, and to a lesser dation can affect problems of global concern in extent nitrous oxide. two ways. * Agricultural sources are estimated to ac- * First, there can be direct effects of the land count for about 30 percent of total carbon degradation processes themselves. dioxide emissions; of this share, about 90 * Second, land degradation can cause indirect percent originates in tropical areas effects resulting from land users'responses to [Duxbury, 19951. Deforestation is a major land degradation problems. source of these emissions. * Methane emissions are mainly from paddy Climate Change rice production and livestock production. * Nitrous oxide emissions originate mainly Emissions. Though not as important as non- from fertilizer applications. agricultural activities, tropical agriculture is a Terrestrial carbon sinks. Terrestrial ecosystems significant source of greenhouse gases. Esti- are an important carbon sink. While much Global Dimensions of Land Degradation on Agricultural Land 11 Figure 4 have very high soil carbon, and indeed are Comparison of carbon stored in soil thought to account for a substantial share of and in above-ground biomass total carbon storage in the terrestrial pool. Soil carbon is lower on agricultural lands, but the .0 Above-ground amount stored is nevertheless much higher than might be expected from above-ground biomass alone. It should be stressed, however, U 5i that major uncertainties remain with regard to M4 - I I I _ 11 ithe role of the terrestrial carbon pool. Africa's contribution to greenhouse gas oJ l i a > 1 __ | emissions. Only 4 percent of Africa's arable land Tropical Temperate Grass- Desert, Agricul- is used for rice cultivation, indicating that Forest Forest land Tundra tural Africa's contribution to total methane emissions Land is likely to be minor, although this may be - * Gl i S S partly counterbalanced by the higher impor- - 8 tance of pasture and crop residue burning practices and relatively low fodder digestibility on the African continent, all of which tend to increase methane emissions. African nitrous oxide emissions are also likely to be relatively 10E low, given the low average level of fertilizer use - ~~~~~~~~~~(4.2 kg/ha versus 36 kg/ha in the tropics as a *- 4 whole [Lal and others, 1995]). The remainder of 152 !S _ this section, therefore, concentrates on carbon dioxide emissions. Land degradation and climate change. Although 20 agriculture as a whole is a significant contribu- tor to global climate change, the question of interest here is the extent to which land degra- Belowlgroi-ind dation on agricultural land affects climate change. Above-ground C Below-ground C * Does land degradation on agricultural land Leaf Root result in increased emissions of greenhouse Wood Humus gases? Litter Resistant Carbon * Does land degradation on agricultural land Source: Adapted from Goudriaan, 1993 affect it's capacity to serve as a carbon sink? * Can appropriate management enhance both attention has focused on carbon stored in land's productivity and its capacity to store above-ground biomass, storage in soils is sub- carbon? stantial (see Figure 4). Sombroek and others Unfortunately, data to answer these questions [1993] estimate the stock of carbon stord in . are scarce. Most attention has focused on emis- organesticmatteri the upp of the world's sions under specific land uses, and on the im- soils to be about 1,220 Pg, or about 1.5 times the pact that land use change is likely to have (in mousto of carbont1220 stre in abiomass. Ational the particular, on the impact of converting forests amount of carbon stored in biomass. Arcoli50Pg to other uses). Very little work has been done carbon at e pcr ona(720 Pg) on how land degradation within a given land and carbonate carbon (720 Pg). Grasslands can use affects emissions. 12 The Global Environmental Benefits of Land Degradation Control on Agricultural Land Figure 5 The soil carbon cycle and some effects of degradation C02 C02 Removal of crop residue A for fodder or fuel reduces C additions Some forms of tillage O Oxidation - accelerate oxidation i Lower yield reduces _ Carbon in litter and litter and root dead root material H Soil organic mater material Production by Hmfcto Vegetation I / I Erosion Erosion Leaching Erosion reduces /i crop yield /* Reductions ' V in soil organic matter reduce crop yield Carbon cycle in soils. Because the areas of in soils are complex, therefore, and can run in interest in this paper (cropland and rangelands) both directions. Some of the ways in which land tend to have relatively low above-ground bio- degradation can affect, and may be affected by, mass, the linkages between land degradation the carbon cycle are illustrated in Figure 5. and climate change are likely to come primarily * Some actions which cause land degradation from changes in soil carbon. Figure 5 provides can increase carbon emissions directly. For a simplified illustration of the carbon cycle in example, burning crop residues for fuel is soils. The carbon balance in a soil at a given thought to be an important contributor to time is a function of the gains and losses of soil fertility loss in many areas, since it prevents organic matter over time. Gains represent all the return of many nutrients to the soil and forms of organic matter supply (litter, roots, reduces the build-up of soil organic matter. crop residues, manure) while losses result from Some forms of tillage, particularly in arid erosion, leaching, and decomposition of soil and semi-arid environments, encourage oxi- organic material. Decomposition is brought dation of organic matter throughout the soil about by soil organisms using carbon com- profile [Pieri, 1992], resultingincarbonbeing pounds as a source of energy (oxidation), a pro- released to the atmosphere rather than cess accompanied by the release of carbon building up soil organic matter. dioxide. Decomposition rates depend on the * Some forms of degradation reduce soil characteristics of the organic material, of the carbon. Erosion carries away soil organic soil organisms, and of the soil itself (for matter-often preferentially so since organic example, aeration, temperature, and moisture). matter is highest in the upper soil layers Effects of degradation. Soil organic matter which are most subject to erosion. Although plays a key role in soil fertility. The linkages this leads to a loss of soil carbon, however, it between land degradation and carbon storage does not necessarily lead to increased emis- Global Dimensions of Land Degradation on Agricultural Land 13 sions. Much of the carbon carried away by necessarily rely heavily on rough approxi- erosion may be deposited under conditions mations. In particular, major uncertainties re- where it may be well preserved, such as in main over both the rates of oxidation and the riverbeds and reservoirs [Duxbury, 1995; magnitude of the feedback between organic Tinker and others, 1996]. matter and yields. * The consequences of land degradation also Albedo. By reducing vegetative cover, land affect the soil carbon cycle. Lower produc- degradation can also change an area's albedo, tion of crops and pasture, whether it results providing an additional possible link to climate from the damages of erosion, nutrient change. However, current research suggests depletion, or other forms of degradation, that while such changes might have an impact will result in lower carbon inputs in sub- on the local climate, they are unlikely to play a sequent periods (less root material, less leaf significant role in changing the global climate litter, less crop residue). This is an important [Dickinson and others, 1996]. linkage since it may contribute to a steady cycle of degradation. Biodiversity The links between degradation and soil carbon are thus numerous and complex. It Cropland. Croplands are substantially modified should be remembered, however, that reduc- from their original, natural state, and their tions in soil carbon and increased emissions to levels of biodiversity are generally substantially the atmosphere are not necessarily synony- lower than those of natural habitats. Never- mous. It should also be remembered that some theless, agricultural landscapes can contain con- forms of degradation can increase soil carbon. siderable biodiversity [Pagiola and others, For example, pastures that are 'degraded' in the 1997]; indeed, biodiversity often plays a crucial sense of being dominated by unpalatable role in agricultural production [Srivastava and grasses or woody species can accumulate sub- others, 1996]. The concern here is whether land stantial carbon stocks. degradation might further reduce the remain- Estimating carbon losses. It does appear that ing biodiversity. land degradation is often correlated with Below-grouind biodiversity. The main direct increased carbon dioxide emissions and a adverse effect of cropland degradation on bio- reduced ability to store carbon. The magnitude diversity is likely to be on below-ground bio- of this effect is difficult to estimate, however. diversity. Diverse and abundant organisms Although some information exists on the stocks help maintain soil fertility and productivity. of carbon under different land uses, there are This diversity is fundamental to soil quality few data on how these stocks change as a result -often called 'soil health'. Small organisms, of degradation. Some models exist that allow such as insects and other invertebrates, play a carbon balances to be calculated, but all have vital role in developing and maintaining limitations. Complex process models such as healthy soils, and help to maintain nutrient the Erosion Productivity Impact Calculator cycling, soil structure, moisture balance, and (EPIC) [Williams and others, 1983] and CENTURY fertility. For example, mycorrhizae, which are [Parton and others, 1992] usually require sub- fungi that live in symbiosis with plant roots, are stantial data on soil and crop characteristics essential for nutrient and water uptake by which are often not available in developing plants. Degradation of soil physical and chemi- countries. Such models also usually require cal conditions can damage this biodiversity, extensive calibration to local conditions before about which relatively little is known. they can be used with confidence. Simpler mo- The consequences of cropland degradation dels such as Soil Changes Under Agroforestry for biodiversity are not necessarily entirely (SCUAF) [Young, 19891 are easier to use, but they negative. Abandonment of degraded lands may 14 The Global Environmental Benefits of Land Degradation Control on Agricultural Land result in their ultimately reverting to their Box 5. Biodiversity in drylands original condition, thus restoring to an extent the natural habitat. Whether this will occur When one thinks of biodiversity, the image that usually depends partly on the availability of nearby comes to mind is that of lush tropical rainforests. habitat from which the abandoned areas can be Biodiversity can also be quite high in drylands areas, recolonized, and partly on the condition of the however. Dryland ecosystems support a wide variety of fauna and flora, albeit at a relatively low density. abandoned fields. Moreover, plants and animals in drylands are often Indirect effects. In most cases, the greatest im- characterized by considerable variation within species pact of cropland degradation on biodiversity is due to the need to adapt to the variability of the likely to be indirect. By reducing productivity environment. Since dryland plants and animals have adapted in on existing agricultural land, degradation ways that enable them to survive under harsh climatic might force farmers to clear additional areas of conditions, they are a valuable source of genetic natural habitat to maintain production. The material to improve the tolerance of crops and livestock fundamental arithmetic of agricultural pro- to drought and disease [Hassan and Dregne, 1997]. duction is that total production equals mean yield times area cultivated. Increasing demand for agricultural products, therefore, can only be taining biodiversity [Mearns, 1996; de Haan met by increasing yields or expanding the area and others, 1997]. under cultivation. Houghton [1994] argues that Many of the aspects which are now thought much of the area being deforested is probably to be particularly likely to cause pasture degra- replacing land that is being abandoned after dation [Steinfeld and others, 1997] are also being degraded. This is a potentially serious likely to have an adverse impact on biodi- problem, since conversion of natural habitat is versity. The encroachment of settled agriculture the action that has the greatest impact on bio- into the more favorable areas will limit their use diversity [Pagiola and others, 1997]. It should by both livestock and wildlife. If access to areas be remembered, however, that land degra- that provide critical grazing or water at times of dation is only one of many possible causes of stress are restricted, for example, both will agricultural expansion. suffer. Agricultural encroachment and fencing Rangeland. Rangelands often tend to be less that restricts the movement of livestock in res- modified from their natural state than cropland. ponse to seasonal climatic variations is also They often contain a much greater proportion likely to impede the migration of wildlife. As of their original biodiversity. Livestock often both livestock and wildlife are restricted to shares rangelands with considerable wildlife. smaller, often less favorable areas, competition The potential exists, therefore, for degradation between them is likely to be exacerbated. to cause relatively more damage to biodiversity on rangelands than on cropland. There are International Waters many possible interactions between livestock and biodiversity in rangelands. Possible nega- Many of the off-site consequences of land tive effects include the disruption of migration degradation that affect watercourses may be patterns; the introduction and propagation of experienced beyond national borders. Sedi- diseases; competition for available food and mentation or flooding problems caused by water; and changes in forage species compo- degradation in an upstream watershed, for sition. Many of these adverse interactions are example, may affect a country downstream. not degradation per se, however; biodiversity From the point of view of land degradation, may be damaged even in a well-working, pro- damage to international waters are a special ductive livestock system. Conversely, there is case of the off-site effects previously discussed. evidence that livestock can contribute to main- Technically, they are indistinguishable from the Global Dimensions 'of Land Degradation on Agricultural Land 15 other off-site effects previously discussed. The dation is reduced or halted, the problems main way in which damage to international discussed in the previous section will not occur. waters differs from damage to waterways To the extent that land degradation is reversed, within the same country is that national policy- the global problems which had previously been makers have no incentives to take them into generated will also be reversed. Increases in soil account. The problems of identifying and mea- carbon, for example, are often a means of re- suring cause-and-effect relationships encoun- storing productivity; they are also an effect of tered in national off-site problems are likely to doing so. In either case, the soil's carbon sink be even more severe in the case of international function will no longer decline and may be en- waters, since the limited data collection under- hanced. To the extent that land degradation is taken in the upstream and downstream coun- halted or reversed, any resulting pressure on tries may not be compatible. natural habitats will also be alleviated. In addition, some land degradation control Global Benefits of Land Degradation practices can have additional benefits in and of Control Activities themselves, by stimulating additional carbon sequestration and/or biodiversity over and Land degradation problems can be addressed above what might have occurred even in the in many ways. A rough typology of land degra- absence of degradation. The following sections dation control measures might include the discuss some specific practices which seem following: particularly likely to have positive effects on * Changing production technology: for exam- problems of global concern. ple, by introducing practices such as mini- mum tillage or agroforestry. Agroforestry * Adding conservation techniques to pro- duction systems: for example, by intro- Reforestation could increase carbon seques- ducing terraces in cropland. tration substantially, but is unlikely to be fea- * Changing patterns of land use: for example, sible on a wide scale, since it would require for- by relocating cropland and pastures to lower going crop production from the reforested slopes and reforesting steeper slopes or by areas. Agroforestry offers a compromise solu- changing stocking rates on1L grazing land. tion, since it allows increased carbon seques- Such efforts are often urLdertaken on a tration while continuing with crop production watershed scale. [Unruh and others, 1993]. Research on agro- If degradation is far advanced, such measures forestry practices has mostly been focused on might have to be preceded by rehabilitation of the contribution they might make to the econo- the affected areas. mic and ecological characteristics of farming Just as land degradation on aLgricultural land systems. In addition to these benefits, agro- can have a range of effects, measures to control forestry also has the potential of producing it can also have a range of effects, at the field, global benefits in terms of both climate change national, and global levels. The specific range of and biodiversity. benefits obtained will depend on the measure On-site and national benefits. Agroforestry being used and on the conditions under which practices can provide a range of on-site benefits. it is applied. The trees can, of course, provide direct benefits As discussed in the previous section, land in the form of products such as fruit, fuelwood, degradation processes can have adverse global fodder, and timber. Combining trees with crop effects. A first benefit of land degradation con- production can, in some cases, also improve trol practices, therefore, is reducing or halting crop productivity: shade can help reduce eva- these adverse effects. To the extent that degra- poration and provide a more favorable micro- 16 The Global Environmental Benefits of Land Degradation Control on Agricultural Land Table 1. Estimated average above-ground biomass and carbon content for different agroforestry types in Sub-Saharan Africa Above-ground biomass (kg/ha) Carbon content (kg/ha) Agroforestry type Low density Medium density High density Low density Medium density High density Silvopastoral 0 - 400 mm rainfall 180 360 720 90 180 360 400 - 800 mm rainfall 175 350 700 88 175 350 > 800 mm rainfall 4,850 9,700 19,400 2,425 4,850 9,700 Fruit tree 3,000 7,500 15,000 1,500 3,750 7,500 Fuelwood 15,400 38,500 77,000 7,700 19,250 38,500 Shelterbelts 6,490 0 3,245 0 Timber trees 130,000 240,000 270,000 65,000 120,000 135,000 Source: Calculated from data in Unruh and others, 1993 climate for crops; trees can provide shelter from the carbon content of biomass is 50 percent wind and reduce the erosive impact of rainfall; [Schroeder, 1993; Unruh and others, 1993; leaf litter can be used as mulch and improve Young, 1989]. Table 1 provides estimates of soil quality by increasing soil organic matter. average above-ground biomass and carbon con- These are not, however, universal attributes: tent in a range of different types of agro- some trees compete with crops for nutrients forestry. Additional carbon would be stored and moisture, and some crops fare poorly below-ground, in the roots and in the soil. Little under shade. From the farmers'perspective, the information exists on the magnitude of this benefits of agroforestry derive from (i) addi- storage, although some estimates suggest it tional production from the tree component; (ii) might be substantial-perhaps as large as the maintaining and/or improving the productivity above-ground storage [Unruh and others, of the crop component; (iii) diversification of 1993]. Given the low wood yields and the low production; and (iv) contribution to the overall planting densities possible under semi-arid and farming system (for example, by providing sub-humid conditions, the increments in carbon fodder or income at a time when other sources storage achievable in dryland areas would be at do not) [Current and others, 1995]. Under some the lower end of the range shown in Table 1. conditions, successful practices can also help The net gain in carbon storage would depend abate off-site damages by reducing runoff and on the system that agroforestry replaces. erosion. Biodiversity. Agroforestry systems tend to Carbon sequestration. The tree component of provide a more hospitable environment for bio- an agroforestry system will sequester carbon diversity, both above- and below-ground. Some from the atmosphere. The magnitude of this agroforestry systems can contain as much as sequestration will depend on the specific tree half the species diversity found in neighboring species being planted, on their yield, and on the primary forest [Thiollay, 1995]. Substantial bio- density of trees in the agroforestry system. Al- diversity benefits are only likely, however, though there has been extensive research on when agroforestry systems cover a relatively agroforestry, little of it has measured the in- large area and are maintained for relatively crease in carbon storage. Estimates of carbon long periods [Sanchez and others, 1996a]. sequestration in above-ground biomass can be Secondary effects. In addition to its direct obtained, however, from available information effects on carbon storage on-site, agroforestry [Schroeder, 1993]. It is typically assumed that might have two secondary effects: Global Dimensions of Land Degradation on Agricultural Land 17 * To the extent that the tree component of the from which they will recover slowly if at all. agroforestry system provides fuelwood, it Even a few years' use might be attractive to would alleviate pressures on other fuelwood poor farmers or pastoralists, however, if the sources. In many cases, it will replace an un- alternative is to derive no benefit whatsoever sustainable fuelwood source with a sustain- from a given area. Developing alternative ways able one. Since emissions from burning fuel- for local communities to benefit from wildlands wood are offset by new growth, the net flux will thus both avoid degradation and preserve to the atmosphere is zero [Unruh and others, valuable biodiversity. 1993]. Various arrangements attempt to find ways * If agroforestry succeeds in stabilizing for local communities to increase the benefits and/or increasing yields on fields on which they receive from wildlife, so that preservation it is practiced, it will reduce pressures to of wildlands becomes an attractive land use clear additional areas for agricultural pro- alternative to cultivation or grazing. The prob- duction [Schroeder, 1993]. This would pre- lems encountered tend to be primarily insti- vent both the increase in emissions and the tutional and legal, rather than technical. In loss of biodiversity that would accompany many cases, issues of property rights dominate. such conversion. In others, new ways to derive benefits from wildlife need to be developed. Safari hunting Community-based Wildlife Management or viewing are popular mechanisms for animal species, while royalties for bioprospecting Rangelands contain considerable biodiversity, offers some promise in the cause of plant spe- most visibly in the form of wildlife. Land users cies. Zimbabwe's CAMPFIRE program (Box 6) is have often been prevented from deriving any perhaps the best-known example of an attempt benefit from this wildlife by hunting or other to develop community-based wildlife mana- means. In many countries, wildlife is legally gement, but other examples can be found in considered property of the state. Moreover, many African countries. wildlife is often felt to be competitive with pro- Multi-species systems. Recent research ductive uses such as grazing, by competing for indicates that livestock and wildlife are less forage or water or by propagating diseases. Not competitive than was previously thought; re- surprisingly, many land users actively seek to omeivetawspriulyhug;r- surprisingly, many land users actively seek to cent research indicates that wildlife and live- eliminate wildlife from areas used for grazing. stock can be complementary to each other, and Recent years have seen many efforts to at times even symbiotic [Mearns, 1996]. The reduce the conflicts between human and wild- grazing 'overlap' between livestock and many life use of the same land. These efforts can take wildlife species, for example, is now thought to two broad forms: be relatively limited [de Haan and others, 1997]. * Wildlife management as an alternative form Moreover, as discussed above, it is now of land use to grazing and/or cultivation; thought that many of the factors which tend to and result in rangeland degradation also tend to ad- * Integration of wildlife and livestock in multi- versely affect wildlife. This new understanding species systems. creates the potential for combining livestock Both are related to land degradation, although and wildlife into multi-species systems. in different ways. Multi-species systems can make a fuller use Wildlife management. Many of the remaining of the ecological potential of spatially and tem- areas with substantial wildlife tend to be in porally variable environments, and are often marginal areas unsuitable for intensive use more suited to conditions in drylands than [Kiss, 1990]. If used for cultivation or heavy mono-species ranching [Cumming, 1994]. For grazing, they may degrade rapidly, damage such systems to be viable, however, problems 18 The Global Environmental Benefits of Land Degradation Control on Agricultural Land such as encroachment of cultivation, fencing, Box 6. CAMPFIRE: Community-based and reduced access to waterpoints-which ad- wildlife management in Zimbabwe versely affect both wildlife and livestock-must be addressed. Zimbabwe's Community Areas Management Program for Indigenous Resources (CAMPFIRE) was established in 1989 in an effort to encourage land users to use wildlife sustainably by providing them with returns from doing so [Child, 1996]. Prior to the CAMPFIRE Land degradation on agricultural land is likely program, inhabitants of Zimbabwe's communal areas to affect problems of global concern in a num- had no rights to exploit the often rich wildlife resources ber of ways, although the linkages are probably in their land, since wildlife was deemed the property of not usually as significant as those encountered the state. Conversion of these fragile areas to in cases of deforestation. The main adverse glo- agricultural use was often the only option open these bal effects of land degradation on agricul.ural communities, often resulting in land degradation. Local bal effects of land degradation on agricultural communities also had little incentive to resist poaching land are likely to be: and, indeed, often participated in it themselves. * Climate change. Land degradation on crop- Under CAMPFIRE, communities can apply for the right land and rangeland is likely to reduce the to manage wildlife resources in their area. Management ability of soils to serve as a carbon sink and may involve both consumptive and non-consumptive release carbon currently stored in soils to the uses, such as hunting (by the community or by tourists paying for the privilege) or photographic safaris. atmosphere. Revenues from such activities are channelled back to the * Biodiversity. In rangelands, degradation may community through their elected Rural Development cause, or be related to through common Councils. In 1993, the program generated revenues of causes, loss of plant and animal biodiversity. US$1.4 million. At present, the bulk of revenues Degradation of both rangelands and crop- originate from trophy hunting, particularly of elephants, land can also indirectly result in damage to although efforts are underway to develop tourism as an land can also indirectly result in damage to additional source of revenue. biodiversity by increasing pressure to con- vert additional natural habitats to agricul- tural use. cases, this positive effect takes the form of * International waters. Land degradation can averted damages; in some, however, additional damage waterways in a number of ways, global benefits might be generated. including sedimentation and changes in the The following chapter turns to the issue of quantity, quality, and timing of waterflow. how attention to global problems might be Just as land degradation can result in ad- better incorporated into land degradation verse global effects, land degradation control control efforts. can result in positive global effects. In many 3. Integrating Global Dimensions into Land Degradation Control Projects The primary reason for efforts to control land be modified to 'go the extra mile' to generate degradation on agricultural land is to reduce, additional global benefits. Given the important arrest, or reverse the field-level or national role that the GEF plays in financing activities problems it is causing. Given the linkages be- that bring global benefits, the chapter then dis- tween land degradation and problems of global cusses how land degradation control activities concern detailed in the previous chapter, how- might fit in GEF's operational priorities and the ever, it is likely that global benefits will also be application of GEF's incremental cost criteria. A generated. This leads to two questions: number of pilot projects concepts which sought * How would land degradation control pro- to integrate global dimensions in land degra- jects which incorporate global concerns dation control activities are then reviewed. differ from projects that do not? * When land degradation control projects Land Users' Motivation and Incentives generate significant global benefits, how are costs to be shared between beneficiaries at Production of crops and livestock depends on different levels? In particular, since the GEF the decisions of individual producers. Farming is the principal contributor to financing requires constant, year-round attention by the activities that generate global benefits, how producer. Moreover, the site-specificity of far- are GEF's incremental costs guidelines to be ming operations makes them heavily depen- applied to land degradation control projects? dent on knowledge of the characteristics and This chapter examines the issues involved idiosyncrasies of each location-knowledge in integrating global dimensions into land which is typically only available to locally- degradation control projects. It begins by exa- based agents. Understanding the motivations of mining the motivation and incentives of land users is critical, therefore, if patterns of re- individual land users. To be successful, any source use are to be understood and if appro- land degradation control program will need to priate policies to deal with any problems that obtain the cooperation of land users, and so an might be identified are to be formulated. understanding of the constraints and incentives That land users should adopt degrading they face is necessary. Insufficient attention to practices or fail to adopt conservation practices this has meant that many land degradation con- that prevent degradation in the face of negative trol projects have failed to achieve their objec- production effects has long puzzled and frus- tives. Previous efforts to address land degra- trated conservationists. But cultivation practices dation problems and the lessons which have that damage the soil can also have beneficial been learned are reviewed in the following sec- aspects in terms of crop production, at least in tion. The chapter then discusses how attention the short term. Moreover, action to slow or to global problems can be integrated into land arrest degradation through changes in crop and degradation control efforts-how they might management practices or through adoption of 19 20 The Global Environmental Benefits of Land Degradation Control on Agricultural Land conservation techniques can be costly, both di- than the benefits of averting degradation. Even rectly in terms of investment requirements or cheaper conservation measures might not be indirectly in terms of forgone production. worth undertaking, however, depending on Though sustainable practices may bring long- their relative costs and benefits. Recent analyses term benefits, they often have short-term costs. of the economics of land degradation and con- The critical question faced by land users is servation in many countries throughout the whether the long-term benefits of adopting sus- world have demonstrated that not adopting tainable practices make these costs worth conservation measures can often be econo- bearing. mically rational for farmers, even from a long- Land users generally have a wide range of term perspective [Lutz and others, 1994; possible responses to degradation. They might Pagiola, 1994; Pagiola and Bendaoud, 1995]. It increase their use of fertilizer or other inputs, or is important to note that the relevant costs are change their land use practices to less dama- those perceived by the land users, including the ging ones. A variety of reduced tillage practices prices they actually pay for inputs and receive exist, for example, which seek to reduce the da- for outputs (which might be distorted by mage to soil that conventional tillage can cause. government policies), the opportunity cost of Land users can also attempt to mitigate the their own labor (frequently underestimated by effects of damaging practices by using off- analysts), and their own rate of time preference setting conservation practices. Some of these [Pagiola, 1993]. conservation measures can themselves be pro- Policy-induced problems. Decisions to adopt ductive; others might interfere with production, conservation practices can often be heavily in- by reducing the effective area available for fluenced by government policies, both by in- production or limiting the use of machinery. fluencing the costs and benefits of conservation Since land users will experience the effects of and by introducing other constraints: any on-site problems directly, they generally * Until relatively recently, most developing have a direct incentive to respond to them. And countries had policies that discriminated in many cases, they do. Almost all farmers in heavily against agriculture. Resources were the Machakos and Kitui districts of Kenya, for extracted from agriculture in a variety of example, have adopted some form of soil and ways: over-valued exchange rates, pro- water conservation practice [English and tection of competing sectors, price controls, others, 1992; Tiffen and others, 1994; Pagiola, and high direct taxation. A sample of 18 19941. Similarly, over half the fields farmed by developing countries found that transfers small farmers in El Salvador have some form of out of agriculture during 1960-84 averaged conservation [Pagiola and Dixon, 1997]. When 46 percent of agricultural GDP [Schiff and farmers do not respond to degradation prob- Valdes, 1992]. These policies made invest- lems, this might be the result of several reasons. ments in agriculture, including conservation Costs and benefits of conservation. The investments, less attractive. available conservation options may simply not * A variety of rules and regulations also affect be cost-effective. For many years, research on behavior, although not always in the in- conservation concentrated on physical struc- tended way. Under Mali's forest code, for tures such as terraces and drainage ditches. example, any land left fallow for more than Such measures, while effective, are expensive five years is considered part of the forest both to build and to maintain. Moreover, since domain and hence government property; they often reduce the area available to culti- this clearly discourages land users from vation, they incur an opportunity cost from for- allowing productivity to regenerate natu- gone crop production in addition to their direct rally. Even rules intended to protect the costs. In many cases, these costs are greater environment can have adverse effects. Some Integrating Global Dimensions into Land Degradation Control Projects 21 countries have rules forbidding the cutting undermined by social and technical changes of trees [Current and others, 1995]. Although and-sometimes deliberately-by government intended to prevent deforestation, such rules policies. also discourage agroforestry and the plan- It should be noted that neither individual ting of fuelwood lots. In many cases, rules land users nor communities will take into ac- and regulations often prove unenforceable, count externalities in their decision-making. but can affect behavior by forcing farmers What is an externality from the perspective of and others to avoid them (for example, by an individual land user will differ, however, bribing enforcers to overlook infractions). from the perspective of a community. Since Tenure. When tenure security is uncertain, communities cover a wider area than their land users may be less likely to avoid activities individual members, some effects which are that cause long-term damage or to undertake external from the perspective of the individual investments which bring long-term benefits will be internal from the perspective of the because they are not sure that they will be able community. To that extent, we might expect to enjoy these benefits. Here too, it is important communities to respond to a broader range of to look at the problem from the perspective of environmental side-effects than individuals individual land users. Absence of title, for would. example, does not necessarily imply insecurity In theory, national policymakers take both of tenure. Few of the farmers in Kenya's Kitui on-site effects of land degradation and off-site district who had terraced their fields had titles problems whose effects are felt nationally into to their land, for example [Pagiola, 1994]. consideration when formulating policies. (They Other problems. A variety of other problems may, however, value these problems differently can also constrain adoption of conservation than private agents.) They are unlikely to consi- measures. Undertaking conservation measures der global effects, however, for two reasons. often requires investments, so lack of access to First, the bulk of the information they have credit can prove to be a constraint. Poverty and available is likely to say nothing about any the need to meet subsistence requirements is possible global impact of land degradation often thought to prevent poor farmers from problems. Second, even when they understand undertaking conservation measures [Pagiola, the linkages between land degradation and glo- 19951. bal problems, national policymakers have no For a time, collectively-managed lands were incentive to take global effects into account thought to be peculiarly at risk of degradation, unless they are either committed to doing so by following Garrett Hardin's classic 'tragedy of treaty obligations or compensated for doing so the commons' argument. Research in recent by the global community. years has shown, however, that communities can and do manage land resources successfully. Approaches to Land Degradation Control When mismanagement of communal resources adversely affects the community itself, there Diagnosis are strong incentives to improve management. The problem Hardin was describing exists not Any land degradation control effort must begin so much on collectively-managed land but on with an assessment of the extent and severity of free access land. Which is not to say that land degradation within a country and a diag- collectively-managed land is not more prone to nosis of its main causes. Much remains to be problems, since coordination problems arise done in this area. Too often, perceptions of land that are not present in the case of individually- degradation problems-both by the govern- managed resources. Moreover, traditional com- ments and by the World Bank-remain rooted munal management regimes have often been primarily in anecdotal evidence rather than in 22 The Global Environmental Benefits of Land Degradation Control on Agricultural Land a thorough review of available evidence. This is thought to be 'lost' to land degradation. These extremely worrying, since in several cases figures are generally uninformative. It is rare where a more careful analysis was attempted, for an area to be degraded to the point that it is the conventional wisdom was challenged sub- abandoned or otherwise 'lost'; it is much more stantially. common for an area to have its productivity In-country. Countries affected by land degra- slightly reduced by degradation. The impact of dation have undertaken a variety of assess- this degradation is less severe than on any'lost' ments of the problems they face. Some particu- areas, but a far greater area is affected. larly vulnerable countries have long-standing The analysis of the causes of land degra- plans to combat land degradation. Mali and dation is often particularly weak. That land Niger, for example, both produced National users were ignorant of both land degradation Plans to Combat Desertification in 1985 and problems and possible solutions was a common have since updated them. For most countries, implicit-and often explicit- assumption. however, the first attempt at a comprehensive World Bank. The World Bank has several assessment of environmental problems, inclu- instruments to aid its client countries in diag- ding land degradation, was undertaken during nosing land degradation problems, including preparation of their National Environmental Country Environmental Strategy Papers Action Plans (NEAPs). NEAPs are intended to (CESPs), Agricultural Sector Reviews (ASRs), provide a framework to integrate environ- Economic and Sector Work (ESW) in the mental considerations into country economic agricultural sector, and Environmental Impact and social development efforts. Signatories of Assessments (EIAs). A selection of these reports the CCD affected by land degradation are to for African countries were reviewed for this prepare a National Action Program (NAP), pro- study. Both the attention given to land degra- viding a new opportunity to review land degra- dation and the quality of analysis varied consi- dation problems and prioritize interventions. derably. In many cases, discussion of land NEAPs often mention land degradation and degradation has been relegated to an environ- -especially in Sub-Saharan Africa-often mental section of sectoral work and CASs single it out as one of the most pressing envi- rather than being integrated into agriculture ronmental issues. As with other aspects of sector strategies. For example, the 1997 Morocco NEAPs, however, the analytical rigor tends to Rural Sector Strategy, although it reviewed other be low tLampietti and Subramanian, 1995]. In aspects of Morocco's rural sector in consi- many cases, they simply repeat the con- derable detail, made almost no mention of land ventional wisdom regarding land degradation, degradation problems, even though such prob- with very little evidence to back it. The focus lems are thought to jeopardize the future of tends to be primarily on the easily-identifiable irrigated areas because of reservoir sedimen- consequences of land-use change rather than on tation and to threaten productivity on rainfed the more difficult to detect problems within lands-productivity which the Strategy hopes land use systems. Very few NEAPs attempt to to increase substantially. Rather, a completely quantify the problems. Figures for rates of ero- separate exercise was carried out later, in the sion or supposed productivity declines are context of an environmental review. The qua- sometimes cited, but upon examination either lity of analysis of land degradation problems prove impossible to confirm or are found to be also varied. At times, they do little more than derived from single observations. Almost al- repeat the conventional wisdom. Even where ways, there is insufficient attention to the site- a more thorough analysis is attempted, it is specificity of problems, with the analysis being often stymied by the lack of data. Yet in relati- carried out in terms of national averages. A vely few cases have follow-up efforts been common approach is to report the area that is made to fill the data gaps. Nevertheless, several Integrating Global Dimensions into Land Degradation Control Projects 23 examples of good practice can be found, obligations to either undertake certain mea- including: sures or to not undertake others. It is symptom- * Ethiopia. As part of the preparation of the atic of this approach that project success was CESP for Ethiopia, a comprehensive review usually measured in terms of indicators such as of land degradation problems was under- 'linear meters of terrace constructed' and that taken. This review pointed to the need to almost no effort was made to measure either revise previous perceptions of the nature productivity or income effects. and extent of land degradation [Bojo and It is fair to say that this approach has failed. Cassells, 19951. In many cases, adoption of the recommended * El Salvador. As part of a comprehensive practices was low; in others, the practices were review of prospects for and obstacles to adopted temporarily but were soon abandoned growth in El Salvador, undertaken in coop- once the project ceased. Seldom did adoption eration with a Salvadoran NGO, a review of spread spontaneously to other land users who evidence of land degradation was under- did not receive the subsidies. taken and supplemented by analysis of a * Many recommended conservation measures, newly executed farm household survey. while technically sound in terms of pre- Here too, the results led to a substantial revi- serving land, were inappropriate to farmers' sion in the understanding of land degra- conditions. Many relied on the supposed dation problems [Pagiola and Dixon, 1997]. existence of low- or zero-opportunity cost family labor for their implementation, for Designing Solutions example, while others interfered with farmers' cultivation practices. Given the top- Even where land degradation problems are down, purely technical basis on which mea- thought to be the most important problem sures were selected, this is not surprising. facing the country, attention to these problems * Insufficient attention was paid to the policy tends to have a low priority within agriculture environment. ministries. Land degradation, for all its sup- * Insufficient attention was paid to the con- posed threats to agricultural productivity, is straints land users faced and the causes of often relegated to environment ministries, their use of degrading land use practices. which tend to be weaker institutionally and to Indeed, far from achieving sustainable adoption have fewer resources and political clout. More- of conservation measures, many projects seem over, many only have oversight responsibilities to have created perverse incentives, leading to and cannot intervene directly. extensive construction of unnecessary con- Historically, the approach taken by most servation measures which were then rapidly land degradation control projects has been to abandoned once the project ended. In some encourage land users to adopt some specific cases, projects seem to have their incentive conservation measures. The proposed measures structure backwards. In El Salvador, for were generally selected centrally, on the advice example, provision of subsidized credit was of technical specialists. They were almost in- justified as allowing farmers to overcome the variably assumed to be beneficial to land users, investment constraints associated with con- who were expected to readily adopt them once servation investments. In practice, however, their benefits were demonstrated. Nevertheless, farmers appear to have been undertaking con- various forms of encouragement were often servation measures so as to gain access to included, ranging from access to subsidized credit, making conservation a cost of obtaining credit or inputs to outright subsidies of some or credit rather than a benefit of doing so [Pagiola even all of the costs of implementation. At and Dixon, 1997]. times, these efforts were further backed by legal 24 The Global Environmental Benefits of Land Degradation Control on Agricultural Land Within the last decade, the emphasis of land Box 7. Gestion des Terroirs degradation control projects has changed in two important ways. The Gestion des Terroirs approach, also known as * Much greater attention is now paid to Community-Based Natural Resource Management changing the incentive framework within (CBNRM), has been increasingly used in recent years, which land resource management decisions especially in Western Africa, where it was first deve- loped. In this approach, communities design and imple- are made. This includes reform of the policy ment a management plan for the area they regularly use structure, with measures such as changes in (the terroir) with the help of a multidisciplinary team of price policy and liberalization of markets technicians. The plan includes rules governing access once dominated by government parastatals. to and exploitation of common resources such as pas- It also includes efforts to improve the institu ture, forests, and water, and specific land improvement tional framework, for example through works, mainly on common lands but also on individual reform. holdings. tenure The principles of the approach include: * Land degradation control efforts have been * Management plans must be site-specific--there are moving toward a much more participative no 'blueprints'; approach, in which both the selection of * Even where one activity, such as grazing, dominates solutions and their implementation are deci the production system of a terroir, a multidisciplinary approach is needed; ded upon and executed in cooperation with * Management units must be based not only on social beneficiary groups. The Gestion des Terroirs units (such as villages) but also on natural resources projects underway in several West African units (such as watersheds) that need to be treated countries are a good example of this new together for management purposes; Box 7). * When a resource is shared by several communities, approach (see the management unit needs to include all the users. Implementation of Gestion des Terroirs projects often Integrating Global Dimensions: Current requires substantial changes in government practices: Practice * Technicians and extension agents often have diffi- culty in internalizing the terroir concept and Diagnosis. Given the low overall level of anal- approach; of land degradation problems, it is not sur- * Government departments tend to be poorly orga- ysis nized to deliver the required concerted and conver- prising that their relationship to problems of ging actions at the community level. global concern has received even less attention. Although Gestion des Terroirs projects initially focused Biodiversity Strategy and Action Plans (BSAPs) solely on natural resource management activities, they are one vehicle through which consideration of have gradually been broadened to include a wider the biodiversity aspects of land degradation range of activities, including social investments (which had been explicitly excluded in most early project problems might be achieved. BSAPs are a key designs). Recent projects typically provide for: instrument in countries' implementation of the * Investments providing long-term economic and en- Convention on Biological Diversity. One of vironmental benefits, such as soil erosion control, their objectives is to identify threats to biodi- water-harvesting, soil fertility improvement, pasture versity and propose priorities for action. By and improvement, natural forest management, sand dune large, however, BSAPs have tended to focus fixation, tree planting, bush fire prevention, and windbreaks. primarily on biodiversity in relatively un- * Economic investments providing quick financial disturbed areas, and have paid little attention to returns, such as wells, boreholes, and irrigation, biodiversity in agricultural areas, degraded or construction of storage facilities for grain and pro- otherwise [Pagiola and others, 1997]. Several duction inputs, cereal banks, rural markets, animal mention agricultural encroachment as a pres health services, and processing units for food and sure factor threatening biodiversity, but do not animal feed. beyond that. * Social infrastructure investments, such as schools and go health care centers. Integrating Global Dimenisions into Land Degradation Control Projects 25 Current projects. Relatively few current pro- GEFfinancing criteria. GEF financing of land jects attempt to explicitly incorporate global degradation control activities is subject to the dimensions. Of these, most take biodiversity as following principles: their entrypoint. Many protected areas projects, * Land degradation control activities are not for example, have components that seek to im- eligible for GEF financing in and of them- prove productivity and/or reduce degradation selves; they are only eligible for financing in surrounding areas so as to alleviate en- insofar as they relate to one or more of the croachment into biodiversity-rich natural habi- GEF's focal areas (the relevant ones being tats. While such efforts will, if successful, un- climate change, biodiversity, and interna- doubtedly have positive results on global prob- tional waters). This is explicitly recognized lems, they are unlikely to play an important in the CCD (art.20). role in addressing a country's land degradation * The project should be consistent with the concerns since they only operate in very re- GEF's priorities, as described in its opera- stricted areas. The Burkina Faso Environmental tional programs [GEF, 1997]. Management Project is an example of a project * As with other GEF projects, financing must that addresses biodiversity protection from a conform to incremental cost criteria. GEF fi- land degradation entrypoint. In the south- nancing is not intended to finance activities western provinces of Houet and Bougouriba, which bring direct national benefits. This too terroir management is combined with forest is explicitly recognized in the CCD (art.20). management. In these areas, forests are under * The project must have a high probability of severe pressure from massive land clearing by success and be a cost-effective means of spontaneous settlers and fuelwood demand, generating the expected global benefits (as threatening the remaining forest and wildlife. defined in the relevant GEF Operational Integrated forest management and conservation Program or, in the case of climate change, plans for the protected forests were to be com- Short-Term Measure). bined with terroir management plans for the * The project must be country-driven. adjacent community lands. The areas protected through the project include the Marc gazetted GEF's Priorities forest, the Hippo Pool forest (a UNESCO Bio- sphere Reserve), and the Nab6re gazetted forest Operational Programs. The GEF has developed and provisional wildlife reserve. ten Operational Programs (OPs) that describe the activities it is prepared to support in each of Sharing the Costs: The Role of the GEF its focal areas [GEF, 1997]. Box 8 summarizes these Operational Programs and their relation- Calls have often been made for grant financing ship to land degradation control activities. It is of land degradation control activities by inter- important to realize that not all activities that national agencies, based on the presumed glo- generate global benefits are eligible for GEF bal benefits of such activities (for example, see financing. The GEF's Operational Programs set Sanchez and others [1996b]). Since the Global specific priorities and describe the activities the Environment Facility (GEF) is the principal GEF supports in each focal area. source of such financing, in practice the main This point is particularly important in the issue is whether, and under what conditions, case of measures to control climate change. GEF grant financing might be obtained to assist Existing GEF Operational Programs on climate in land degradation control projects. Both the change focus on the introduction of new tech- Instrument Establishing the Restructured nologies and on the removal of obstacles to Global Environment Facility and the CCD their adoption. As such, their relevance for land assign the GEF an explicit role in this regard. degradation control activities is limited. An 26 The Global Environmental Benefits of Land Degradation Control on Agricultural Land operational program on carbon sequestration is Box 8. GEF's Operational Programs under preparation. Until this program is adopt- ed by the GEF Council, the GEF's Short-Term Biodiversity. Four of GEF's Operational Programs (OPs) Measures are the main mechanism by which are based on specific ecosystems: land degradation-related climate change miti- * OPI - Arid and Semi-Arid Zone Ecosystems * OP2 - Coastal, Marine, and Freshwater Ecosystems gation measures might be eligible for GEF fi- * OP3 - Forest Ecosystems nancing. The criteria for funding under this * OP4 - Mountain Ecosystems mechanism include a requirement that projects The objectives of these OPs are the conservation and meet a cost-effectiveness criterion of sequeste- sustainable use of biodiversity and biological resources ring carbon at a cost to the GEF of $10 per ton in the relevant ecosystem. Land degradation control or less. This requirement is hard to meet in the efforts in these ecosystems are eligible for funding insofar as they address threats to the viability of the case of land degradation in agricultural areas ecosystem, redress the damages of past degradation in (as opposed to cases of deforestation) because biologically-sensitive areas, or develop sustainable use of the lack of data on the magnitude of changes practices. in carbon stocks caused by degradation. Climate change. The emphasis of the OPs on climate change activities (OP5, OP6, and OP7) is on the Application of incremental cost principles introduction of new technologies and the removal of barriers to their adoption, so they have limited relevance for land degradation problems. An OP on The application of incremental cost principles carbon sequestration is under preparation. Until such caused considerable confusion in the early an OP has been adopted, the main eligibility criteria for years of GEF's work. As experience with their GEF financing for land degradation activities that application has grown, the methodology for mitigate climate change are those of the short-term doing so has become clearer [GEF, 1996b; measures. Under these measures, among other criteria, projects must meet cost-effectiveness criteria. That is, Hansen, 1997]. However, although efforts have they must mitigate a specified amount of greenhouse been made to develop hypothetical examples of gas emissions for a given cost, typically a low unit how incremental cost principles might apply to abatement cost (approximately less than or equal to land degradation control projects [King and US$10 per ton of carbon). Kumari, 1997], specific experience with their International waters. Two other OPs with direct relevance to land degradation control activities are: application in this area is extremely limited. * OP8 - Waterbody-based Operational Program Problems in defining an appropriate baseline * OP9-IntegratedLandandWaterMultipleFocalArea against which to measure incremental costs, Operational Program already one of the most difficult issues in GEF OP8 focuses on seriously threatened waterbodies and projects, are likely to aggravated in land degra- on the most imminent threats to their ecosystems, while dation control projects,because 0f the presence OP9 focuses more broadly on integrated management dation control projects because of the presence of land and water resources and on preventive rather of benefits at multiple levels and the scarcity of than remedial measures. data. An example can help to make the issues in- volved concrete. Consider a project to revege- tate a degraded hillside. If left degraded, the local residents. Some benefits may also accrue hillside will generate few on-site benefits to to downstream residents, through a diminution land users, since its productivity will be low, of problems such as sedimentation or flooding. and it may generate off-site damages to down- Revegetation might also generate some global stream areas. Revegetation would restore some benefits by sequestering carbon and providing of the area's productivity. It might make it once a more hospitable habitat for biodiversity. again usable for pasture, for example. It might Would such revegetation be eligible for GEF also produce fuelwood and other valued pro- financing? ducts. Many of these benefits will accrue to Integrating Global Dimensions into Land Degradation Control Projects 27 Figure 6 a biodiversity perspective. Such modifications Application of GEF incremental of the original revegetation plan fall squarely cost principles within the incremental cost guidelines. As long as the global benefits being generated are suffi- Baseline GEF Alternative cient and fall within the Operation Program Project Project guidelines, GEF financing would be appro- $ priate. GEF financing would be available for the incremental costs of implementing the GEF alternative relative to the costs of implementing Increm+ ental the baseline project (that is, the project the coun- Costs try would have undertaken based on national ---- _ ---- considerations alone). This is illustrated in Figure 6. If the GEF alternative project gene- rates some additional national benefits compa- red to the baseline project, the incremental costs might be adjusted by subtracting the avoided Costs Benefits Costs Benefits costs to the country of generating these benefits. In some cases, local and national benefits Global Benefits alone are insufficient by themselves to justify Costs National Off-site Benefits revegetation but the addition of global benefits Field-level Benefits would make the intervention profitable. The appropriate baseline for this project would be If revegetation is profitable based solely on to do nothing. Under these circumstances, the the local and/or national off-site benefits that same principles would apply: as long as the will be generated, it would not be eligible for project is a cost-effective way of generating the GEF financing even if some global benefits are expected global benefits in the context of an generated. Suppose, however, that the propo- operational program, GEF would in general be sed revegetation project was modified to in- willing to finance the incremental cost of the crease its contribution to alleviating global pro- activity (once again, adjusted for the avoided blems. For example, the hillside might be reve- costs of generating the local and national off- getated with indigenous species rather than site benefits). with a monoculture of quick-growing exotics. GEF financing is also available for activities This might result in greater biodiversity bene- which remove obstacles to adoption of practices fits by more closely recreating the original which are expected to generate significant glo- habitat. It might prove more costly than reve- bal benefits. For example, there might be a need getating with a monoculture, however, both in for reforms to create a more enabling policy terms of direct costs and perhaps also in terms environment. Reform of the tenure system of lower production of goods such as fuelwood. might be necessary to ensure that degradation Alternatively, a larger area might be revege- problems do not recur following revegetation. tated than originally envisaged. Some areas Targeted research might be needed to either might have been left out of a revegetation pro- design appropriate revegetation schemes or to ject predicated solely on local and national be- identify areas to be revegetated. Since these nefits because they are unsuitable for grazing or activities would also generate substantial local too distant from population to make fuelwood and national benefits, however, establishing an production attractive. Yet benefits such as pro- appropriate baseline is likely to prove difficult viding corridors between areas of natural habi- in such cases. tat may make revegetating them attractive from 28 The Global Environmental Benefits of Land Degradation Control on Agricultural Land Implementation Issues investments in social infrastructure. For example, in the GEF-financed West Africa Pilot Land degradation control projects are likely to Community-Based Natural Resource and Wild- face difficult implementation problems as they life Management Project in Burkina Faso and attempt to address problems of global concern. C6te d'Ivoire, progress on habitat and wildlife In particular, two problems will pose them- management is contractually linked to the pro- selves: vision of social infrastructure and other socio- * How to design projects that take global con- economic benefits from the project. Several of cerns into consideration in a participative the project concepts discussed below have way; and adopted this approach. * How to secure the cooperation of land users in cases where the activities that generate Pilot Project Concepts global benefits are not in their direct in- terests. In view of the limited experience in the pre- As discussed previously, land degradation paration of projects that blend global environ- control projects no longer prescribe 'appro- mental concerns with land degradation control, priate' conservation measures for land users to the World Bank and the International Fund for adopt. Rather, land users are offered a large Agricultural Development (IFAD) have colla- choice of possible activities to undertake. In borated in developing a pipeline of projects in many instances, even the activities themselves this nascent area of GEF operations. This effort, are developed in cooperation with land users, which was partly financed by a GEF Project through adaptive research. This approach is Development Facility (PDF) Block B Grant of proving much more successful and sustainable US$334,000, sought to clarify how global envi- than previous approaches, but it does pose pro- ronmental benefits can be generated while ad- blems for any efforts to integrate global con- dressing poverty and land degradation at the cerns. Since the actual mix of activities that will community level by developing a series of pro- be undertaken cannot be known in advance, ject concepts based on IFAD's pipeline of rural identifying the likely global benefits will be development projects. This section summarizes difficult ex ante. Adapting the chosen activities the initial lessons of this effort. so they generate additional global benefits (for Selection Criteria. Table 2 summarizes the example, by choosing species to be planted project concepts developed by IFAD. These based on carbon-sequestration or biodiversity- project concepts were selected from within enhancing criteria) will also be difficult. IFAD's pipeline of rural development projects From the perspective of individual land using the following criteria: users, global benefits are as much an externality * That the project directly address land degra- as national off-site effects. The same problems dation trends or address the physical pro- which have plagued efforts to undertake con- duction base of the rural poor in areas affec- servation measures to abate national off-site ted by or at risk of land degradation, such as damages are likely to arise. Of course, if the arid, semi-arid, or dry sub-humid areas. conservation measures being envisaged are in * That there appear to be links between land the land users' own interest, then adoption and degradation processes in the project area and long-term maintenance will not be an issue. If problems of global concern such as biodi- this is the case, however, the measures may not versity protection. meet GEF eligibility criteria. The approach * That potential GEF-financed activities be taken by projects such as the gestion des terroirs compatible with the proposed baseline pro- projects in West Africa is to link certain conser- ject activities. vation objectives to receipt of project-financed Integrating Global Dimensions into Land Degradation Control Projects 29 Table 2. Pilot project concepts integrating global dimensions in land degradation control Land degradation problem Baseline project Global dimension Possible response Botswana: Community Based Natural Resources Management Project Rangeland degradation Help local communities Degradation of one of the Emergency measures to develop sustainable use last large-scale migratory rehabilitate the integrity of the practices systems in the world southwestem wildlife system Mali: Sahelian Areas Development Programme (SADeF) Unsustainable land use in the Community-based natural Rich agro-biodiversity Targeted measures to conserve, Interior Delta of the Niger resource management threatened by encroachment rehabilitate, and promote the River and unsustainable use sustainable use of biodiversity Jordan: National Program for Rangelands Rehabilitation and Development Rangeland degradation Improved forage production Damage to biodiversity; Enhancing pasture regeneration to and promotion of other viable greenhouse gas emissions slow GHG emissions; measures to productive activities conserve arid zone biodiversity Belize: Community-Initiated Agriculture and Resource Management Project Unsustainable land use and Community-based natural Encroachment into protected Community-based natural deforestation resource management in areas resource management in agricultural areas non-agricultural areas El Salvador: Rural Development in the North-Eastern Region Erosion and deforestation in Improved access to labour Encroachment in remaining Reforestation of hillsides; hilly areas in the Bay of and credit markets; improved natural habitats Conservation measures in the Fonseca watershed farming practices buffer zones of areas with rich India - Madya Pradesh: Madhya Pradesh Tribal Development Project Unsustainable land use and Improved farming systems; Encroachment into protected CBNRM and targeted activities to deforestation small-scale irrigation; forestry areas (11 national parks and conserve unique plant and animal activities; marketing of 32 wildlife sanctuaries) biodiversity non-wood forest products India - North East Region: Community Resource Management Project for Upland Areas Unsustainable land use and Intensification of production Encroachment in natural Improvements to system of deforestation in more favorable and habitats (area is designated a protected areas; creation of buffer accessible areas world ecological 'hot spot' zones; targeted research on natural due to its unique regeneration biodiversity) Mongolia: Arhangai Rural Poverty Alleviation Project Lake Baikal watershed is Improved livestock and Lake Baikal and its Afforestation; improved energy threatened by salinization due vegetable production; surrounding area form a efficiency; development of a to poorly-managed schemes, comprehensive rangeland unique habitat for a range of watershed management plan; overgrazing, and monitoring aquatic and animal species strengthened management of deforestation protected areas In addition, some projects were rejected for a Botswana: Community Based Natural Resources variety of operational reasons, such as the stage Management Project of preparation making it impractical to add new components. It should be stressed that Wildlife in Southwestern Botswana has reached since project concepts were selected from a critically low point, and its recovery is hin- IFAD's current pipeline of projects, they do not dered by loss of critical wet season breeding represent a systematic review of the various range and dry season grazing range, impedi- possible types of projects. ments to migration, and high offtake. A number of water-dependent species have already been lost from the system. If current trends continue, large ungulates will be reduced to small popu- 30 The Global Environmental Benefits of Land Degradation Control on Agricultural Land lations in the protected areas. This would result Mali: Sahelian Areas Development Program in the loss of one of the last large scale migrato- ry systems in the world, as well as of within- The Sahelian Areas Development Programme species genetic diversity. The proposed baseline (SADeF) will follow an approach similar to the project was to help local communities plan, pre- Gestion des Terroirs method (see Box 7 on page pare, and execute income-generating activities 24 above) by supporting local communities' ef- such as sale of hunting permits, wildlife and forts to plan, implement, and administer a vari- ecocultural tourism, and use and marketing of ety of natural resource management initiatives. 'veld' products. The program also aimed to in- The Interior Delta of the Niger River is one of duce private sector involvement in tourism- SADeF's areas of operation. This area provides related infrastructure and investment in the the habitat for a variety of natural and human communities' capacity to manage wildlife re- ecosystems, including a large diversity of wild sources and gazetting of protected areas. and domesticated animal and plant genetic re- The GEF component would have comple- sources. Both the natural ecosystems and those mented the activities of the baseline by underta- modified by human influence have been degra- king emergency measures aimed at reinstating ded or are under severe pressure because of de- essential components of the ecosystem, parti- mographic growth, the progressive expansion cularly through access to identified dry season of cultivated land, ever shorter fallow periods, range and water, so as to rehabilitate the inte- soil salinization in irrigated areas, the increase grity of the southwestern wildlife system. The of livestock herds, the reduction of pas-ture benefits of these measures would in turn have lands and pastoral corridors, and deforestation. been sustained by the CBNRM activities being The baseline activities under SADeF will promoted under the baseline program. already result in global benefits by reducing the The proposed activities were consistent with pressure on the remaining natural habitats of the objectives of the GEF's operational program the Delta. However, preserving and restoring on Arid and Semi-Arid Zone Ecosystems (OP1), the region's rich and globally significant biodi- in that they provided for in situ conservation of versity, including animal and plant genetic biological resources threatened by pressure resources, will require additional efforts and from more intensified use and drought, and en- resources. The GEF alternative would comple- couraged sustainable use by helping to develop ment SADeF activities with targeted measures multiple use practices. to conserve, rehabilitate, and promote the sus- The program had strong links between base- tainable use of the rich biodiversity in the Niger line and GEF alternative activities. While the Delta. Specific activities would include: baseline focused on measures which would * Integrating biodiversity conservation in bring direct benefits to participating local com- SADeF's CBNRM activities. munities and to the national economy, the GEF * In situ conservation of domesticated and altemnative would have undertaken comple- wild animal and plant genetic resources, by alternativmeawoul havmeda undeten compn- supporting local cultivators' production, dis- mertntr e. a .aime at biovsi con- tribution, and exchange of seeds of tradi- baselineand GEFaltenz ative were strongksince tional landraces and by helping pastoral both were developed in parallel. communities conserve traditional animal bothough t hweresdeveopedei parallexpel. edtobreeds through the establishment of Natural Although this concept was expected to have Traditional Breeds Centers. strong potential of obtaining GEF financing, . Targeted research onbiodiversity resources, Botswana's Ministry of Finance ultimately deci- which will feed directly into the design and ded not to borrow for the baseline project. implementation of conservation activities, extension, and training. Integrating Global Dimensions into Land Degradation Control Projects 31 The proposed activities are consistent with too costly to be viable as an effective way of the objectives of the GEF's operational pro- achieving global climate change benefits, grams for Arid and Semi-arid Zone Ecosystems and that it would be unlikely to meet the (OPI) and Coastal, Marine, and Freshwater short-term response criteria of sequestering Ecosystems (OP2), as well as on Integrated carbon for US$10 per ton or less. Land and Water Multiple Focal Area (OP9). * Efforts to re-establish the equilibrium bet- Preparation of this project is underway. A ween wildlife and domestic animals might PDF B grant is being requested from the GEF to have helped conserve biodiversity in this finance preparatory activities, including: arid ecosystem. However, the global benefits * Identification of gaps in the environmental were not well established. Although there is ,nfrm.,,,available, valuable biodiversity in the area, it was not i Preparnservation pro- clear to what extent land degradation trends gram, including prioritization of wild and threaten it. In addition, the synergy between domesticated species to be included, deve- the proposed biodiversity conservation ef- lopment of gender-specific incentive pro- ak. grams for use and preservation of indige- pw arn nu koleg tha suprsarbo Improving water harvesting practices and nous knowledge that supports agrobio- *I diversity conservation, and definition of management could have helped decrease training requirements. unsustainable use and aquifer depletion, but - Detailed design of the CBNRM Program, in- insufficient information was available to cluding development of a detailed 'menu' adequately evaluate the possible contribu- of options eligible for funding. tion of degradation control measures on i Policy analysis of the underlying causes of international waters in the region. land degradation. The baseline project thus proceeded without a GEF component. Had the possibility of incor- Jordan: National Program for Rangeland porating global aspects been raised earlier in Rehabilitation and Development project preparation, it might have been possible to fill some of the data gaps. This possibility The project area is undergoing accelerating will be re-examined when the next phase of the degradation because of high grazing pressure. baseline project is prepared. Rangeland productivity is thought to have de- Belize: Community-Initiated Agriculture and clined by up to 50 percent over the last 30 years, Resource Management Project and is expected to decline further at a rate of 3 to 5 percent per year. The baseline project will address the poor socio-economic conditions of Land degradation is a serious threat in Belize maddess by .rehabiitatngmrage ondveion, but is largely not yet a fact. However, the pres- wt a emp . rhasis pti ond o rgefor, sures on the natural environment are increasing livt and promo tion of oh ra vi e pro- rapidly: The rate of deforestation, for example, ductive activities, r has accelerated considerably in the last few years and is expected to continue to do so. Pres- Several ways of supplementing the planned sures are particularly great in the south of the baseline activities with measures to enhance country. Existing agricultural practices only global benefits were examined, but none ap- permit cultivation for short periods, and increa- peared promising: sing population pressure is leading to shorter * Enhancing pasture regeneration could have fallow periods. stimulated carbon fixation. However, the The baseline roject to be implemented in limited data available suggested that the The stann Creek toibe i nte in project area was too small and regeneration Toledo and Stann Creek Districts in the south 32 The Global Environmental Benefits of Land Degradation Control on Agricultural Land of the country, will address land degradation remainder are under pressure from fuelwood by developing sustainable production practi- collection and clearing for agricultural use. ces on land suitable for agricultural use, thus Cropping areas are threatened by erosion due avoiding further declines in soil fertility or to steep slopes and lack of cover at the begin- further expansion of the agricultural frontier ning of the rainy season, while pastures are into zones unsuited for agriculture. The project subject to compaction and baring of soils due to will also seek to improve farmers access to overgrazing. The baseline project seeks to im- financial services and to technical and marke- prove the living conditions of the rural poor, ting services. by improving their access to labor markets, The GEF component would complement the supporting local financial institutions, and pro- baseline activities by approaching the same moting non-degrading farming practices. problem from the non-agricultural side: assis- Possible activities considered for a GEF alter- ting communities in defining the lands that native included: should not be used for agriculture, and develo- * Measures to protect biodiversity, such as re- ping ways to conserve them. GEF-financed habilitation of micro-watersheds and conser- activities would focus on the Sarstoon-Temash vation measures in the buffer zones of areas National Park, which includes the most pristine with specific biodiversity needs, such as the wetlands in the country. A CBNRM approach Cacahuatique Mountains; and will be followed, with activities being drawn * Testing and promoting more efficient cook- from a 'menu' of options based on Belize's bio- ing stoves as a means to prevent further de- diversity strategy. The problems at Sarstoon- forestation caused by fuelwood use. Temash are very similar to those encountered The strongest need for intervention concerns in other protected areas in the countries, and biodiversity conservation. There is clear evi- the project is also intended as a pilot project dence of biodiversity loss caused by loss of which the approach proves successful, might forest cover. However, there was only limited later be replicated (the period of implemen- synergy between activities designed to protect tation of the GEF-funded activities is shorter the remaining biodiversity and the planned than that of the baseline project to allow for baseline activities, which mainly address areas such replication). already converted to cultivation. A targeted Because of the small-scale, targeted nature of conservation program would probably provide the proposed activities, the GEF's Medium- a better vehicle to achieve biodiversity conser- Sized Grant program is being used. The pro- vation objectives in this case. This is especially posed activities are consistent with the ob- true since the baseline land degradation control jectives of the GEF operational programs on project is already quite complex, with a dispa- Coastal, Marine, and Freshwater Ecosystems rate array of participating organizations. The (OP2) and on Forest Ecosystems (OP3). The case for activities designed to mitigate climate project area lies within the Meso-American change is not as strong. Although deforestation Biological Corridor, a priority area for GEF bio- results in release of carbon, most of the area's diversity conservation activities. forests have already been lost. The link between land degradation in cultivated areas and cli- El Salvador: Rural Development in the mate change is not well established; nor is the North-Eastern Region extent to which promotion of energy-efficient cooking stoves would contribute to reductions The project area, in the Departments of La in fuelwood consumption, or to reductions in Union and Morazan in the Bay of Fonseca zone, greenhouse gas emissions. For these reasons, it is the driest area in El Salvador. Much of the was decided not to proceed with preparation area's forests have already been lost, and the of a GEF component at this time. Integrating Global Dimensions into Land Degradation Control Projects 33 India - Madya Pradesh: Madhya Pradesh Tribal in land area clearing for agricultural use and a Development Project shortening of the fallow period, eroding the sustainability of indigenous shifting cultivation Madhya Pradesh has the largest area of pro- system (ihum). The baseline project focuses on tected areas in India, including 11 national addressing the needs of the most vulnerable parks and 32 wildlife sanctuaries. Forests in the households through an integrated community- project area are being lost at a rate of 1.5-2.0 based approach for land use and productive percent per annum. The main causes of the loss support measures. These measures are desi- are increased demand for commercial and non- gned to intensify production in the more favo- commercial forest products, including bamboo, rable and more accessible areas. In addition, the and the expansion of cropping and livestock project is supporting conservation activities for grazing. The baseline project had been designed unique tribal cultural and religious areas of pri- to improve the quality of life of tribal groups mary forest known as 'Sacred Groves'. and to prevent degradation of the environ- The GEF alternative would integrate biodi- ment by supporting a range of economic live- versity conservation with the baseline activities lihood activities, including small scale irri- by enlarging existing protected areas, estab- gation, soil and water conservation, farming lishing new ones, and creating buffer zones to systems improvements, forestry activities, and conserve the integrity of the parks. Targeted support for marketing of minor non-wood research and field testing of approaches for en- forest products. hancing natural regeneration after jhum and to The GEF alternative would have aimed at reduce the need for burning would also be strengthening biodiversity protection in the undertaken. Efforts would also be made to inte- project area by investing in participatory insti- grate biodiversity considerations into areas of tutional mechanisms for conservation and im- intensified agricultural production-for ex- proved biodiversity management and by imple- ample by maintaining forest corridors between menting targeted activities to conserve unique protected areas. forest food and medicinal plant genetic re- Even without the GEF component, the base- sources and charismatic fauna in selected pro- line project is likely to induce a reduction of tected areas. pressure on existing protected areas, thanks to Development of a GEF alternative was ham- its agricultural intensification activities. In addi- pered by uncertainty over the baseline project. tion, the lifting of jhum pressure on land could Problems in the design of the baseline led it to result in longer fallow periods, during which be reformulated in 1997 and subsequently re- significant biodiversity regeneration might take jected in 1998, at which point preparation of the place. However, this would not address already GEF alternative also ceased. Had work pro- existing damage, nor would it ensure adequate ceeded, it would have been necessary to better protection of natural habitats. By addressing establish the global importance of the biodi- these issues, the GEF alternative would sub- versity that the project would have protected. stantially enhance the global benefits generated by the project. The proposed activities are con- India - North East Region: Community Resource sistent with GEF's Operational Program 4, on Management Projectfor Upland Areas Mountain Ecosystems. Although the proposed project and its GEF The project area, in the states of Meghalaya, component appeared quite promising, both had Assam, and Manipur, in Northeastern India, to be put on hold because of a change in the na- has been designated as a world ecological 'hot tional government, tensions between Central spot' because of its unique biological diversity. and North-East Region administrations, and Population growth has resulted in an increase civil strife in the region. 34 The Global Environmental Benefits of Land Degradation Control on Agricultural Land Mongolia: Arhangai Rural Poverty Alleviation Initial Lessons Project Efforts to integrate global dimensions into land Lake Baikal is the world's deepest trough lake degradation control programs are still in their and holds one-fifth of the world's fresh water. infancy. Nevertheless, some lessons are begin- The lake and its surrounding area form a ning to emerge. unique habitat for a range of aquatic and ani- mal species. Approximately two-thirds of the Need to understand the underlying land degradation water flowing into Lake Baikal originates in processes Mongolia from the Selenge River watershed, which is threatened by salinization caused by The need for a clear understanding of the land poorly-maintained and inefficient irrigation degradation problems being confronted and of schemes, overgrazing and trampling, and de- their causes was reinforced. One of the obsta- forestation. The baseline project covers about cles to the India Madhya Pradesh project was one-third of the watershed and aims at raising precisely that land degradation problems were the income of the poor herders through live- not fully understood, leading to repeated re- stock development and vegetable production. designs of the baseline project, which in turn It also introduces comprehensive rangeland made it very difficult to design appropriate monitoring and efforts to ensure that herds are complementary activities to generate global kept to sustainable levels. benefits. In many of the projects reviewed for The GEF alternative would address the possible inclusion in the sample, the underlying causes and effects of land degradation in the land degradation problems were too ill-defi- watershed by: (i) promoting energy efficiency ned to allow consideration of the possible links in cooking and heating and afforestation; (ii) to problems of global concern. developing a water management plan for the watershed; and (iii) strengthening management Establishing links to problems of global concern of protected areas, so as to ensure quality and quantity of waterflows into Lake Baikal, thus Establishing whether there are links between helping to conserve its biodiversity. land degradation and problems of global con- Although the proposed GEF alternative ap- cern, and the exact nature and magnitude of peared attractive from the perspective of pro- those links, proved to be a significant obstacle tectinganinternationally-importantwaterbody in all cases. Development of all the project con- it suffered from several problems. Most impor- cepts was hampered by a paucity of relevant tant, Lake Baikal is affected by extensive indu- data. strial pollution; land degradation only creates a Climate change. With the exception of cases small portion of the threats to the Lake's eco- in which deforestation was taking place, the system. Although some efforts to address pollu- prospects for activities designed to generate cli- tion are already underway, devoting substan- mate change benefits appeared to be very limi- tial efforts to the more diffuse threats posed by ted. In almost all cases, it appeared plausible land degradation is unlikely to be justified at that the proposed land degradation control acti- this time. vities would result in some reductions in car- bon emissions, either by reducing losses from carbon stocks or by enhancing sequestration. In every case, however, the data were insufficient to establish the magnitude of this benefit and to demonstrate that it met the cost-effectiveness guidelines of the GEF's short-term climate Integrating Global Dimensions into Land Degradation Control Projects 35 change measures. The exception were cases in- which addresses the problems of Lake Baikal. volving deforestation, for which more data In this instance, however, it was judged that tended to be available. None of the project land degradation problems represented a relati- concepts examined involved agroforestry, so vely minor part of the waterway's problems, so the extent to which activities which promote it an intervention on this basis was unlikely to be might satisfy GEF climate change financing cri- justified. teria remain unexplored. Biodiversity. Linkages between land degrada- Integrating measures to address global problems tion and biodiversity proved much easier to with measures to address national problems establish in the sample of project concepts. These linkages generally took the form either of Even when the links between the global and encroachment onto natural habitats adjacent to the national and local dimensions of land cultivated areas being degraded or of damage degradation have been established, the ques- to wildlife that shares rangeland being degra- tion arises whether combining activities to com- ded with livestock. There was relatively little bat them is desirable. The answer is not always information on possible losses of biodiversity yes. Two aspects need to be considered: within cultivated land. The SADeF program in * The compatibility of the actual measures Mali was the only exception; in this case, many being proposed to combat the national and of the diverse traditional landraces and cattle local problems and those required to address breeds found in the interior delta of the Niger the problems of global concern. River were found to be at risk from degradation ' The compatibility of these activities from an processes. This is a reflection of the extraor- organizational perspective. dinary agro-ecological diversity of the area, Several projects showed considerable syn- however, so it would be unrealistic to expect ergy between activities targeted at the local and similar benefits in the majority of areas under- national problems and those targeted at the glo- going degradation. bal problems. Perhaps the best example of this Using land degradation as the entry point synergy is that of the proposed Botswana means that the biodiversity being conserved CBNRM project. Sustainable use of rangelands may not be the highest priority biodiversity. In is a precondition for implementation of the pro- cases such as the interior delta of the Niger posed GEF-financed wildlife conservation acti- River in Mali, the rangelands of Botswana, or vities-without the planned baseline activi- southern Belize, land degradation problems ties, the GEF-financed activities would be very coincided with areas of very rich biodiversity. unlikely to succeed, since the underlying pro- In several other cases, however, the global im- blems would not have been addressed. In turn, portance of the biodiversity at risk was less implementation of the GEF-financed activities obvious. In the India Madhya Pradesh project, will bring benefits for the wildlife-based sus- for example, the degree of endemism among tainable use activities. Likewise, in the Belize species in the project area was low even though project, the proposed GEF-financed activities there were several national parks in the area, complemented those of the baseline by ad- and the area had not been identified as a prio- dressing the issue of appropriate management rity in the national biodiversity strategy. This of community lands that fell within protected does not mean that efforts to protect this biodi- areas, while the baseline activities did the same versity would not be justified, but it does mean for the surrounding agricultural areas. Con- that a case for doing so must be prepared. versely, in the El Salvador project the synergy International waters. Only one of the project was limited. The biodiversity loss problems concepts had a significant international waters being experienced were largely outside the dimension: the Arhangai Project in Mongolia, agricultural areas targeted by the baseline pro- 36 The Global Environmental Benefits of Land Degradation Control on Agricultural Land ject. Moreover, the activities required to protect CBNRM Fund with approval procedures simi- this biodiversity had little in common with lar to those for baseline activities. The key fea- most measures contemplated under the base- tures of this plan involve the highly participa- line project, such as efforts to improve rural tory nature of the choice of activities, and the credit and labor markets. use of explicit quid pro quos when conservation The organizational synergy between mea- requires farmers to undertake activities which sures to address national and global problems are not in their direct interests. How well these also varied. The synergy was high in the case of arrangements will work out in practice has, of the Botswana project, since the same agencies course, yet to be determined. were involved in both aspects of the problem. One drawback of this approach at the time Conversely, the synergy was low in the El of project preparation is that it makes it difficult Salvador project, where the agencies that would to establish ex ante the extent of global benefits have undertaken the biodiversity conservation that might be generated and their cost, since the activities were not the same as those involved exact activities that will be undertaken is not in the baseline activities-indeed, adding more known. As long as the rationale for activities to implementing agencies might actually have generate global benefits was strong and their been harmful, since the project was already relationship to the baseline activities clear, how- quite complex. ever, these problems proved less severe than When either the substantive or the organi- initially anticipated. The main difficulties seem- zational synergy between measures to address ed to lie in ensuring participation rather than in the national and the global aspects of land the nature of the activities to be implemented degradation problems are low, it is likely that themselves. For the purposes of estimating in- self-standing, targeted projects to address each cremental costs, representative activities can be problem separately would be more effective used. than attempting to integrate the two. Determination of incremental costs Implementation issues Incremental cost analyses were only carried out Most project concepts addressed the problem of for some of the project concepts. As in other securing farmer participation by adopting vari- GEF projects, the distinction between costs that ations of the now well-established CBNRM were eligible for financing and those that were approach. The application of this approach to not was the major point of discussion. In many GEF-financed activities was most fully fleshed- cases, controversy arose primarily because of out in the Mali project. Since CBNRM is a pro- the mistaken expectation that any activity cess rather than a blueprint, the specific acti- which generated global environmental benefits vities will be defined during implementation, would be eligible. In several projects, including drawing from a 'menu' of options based on those in Botswana, Belize, Mali, and India's Mali's biodiversity strategy. Training for local Northeast Region, important biodiversity con- communities will help ensure adequate local servation benefits are likely to be generated by capacity to participate in developing natural re- the baseline activities, since they will tend to re- source management plans in and around pro- duce pressures on natural habitats. Since these tected areas and enter into 'contractual' agree- are incidental benefits of activities that are ments over the adequate conservation of areas justified on the basis of the national benefits with particular biodiversity value. Develop- they generate, they are not eligible for GEF ment of natural resources management plans funding. and resulting project proposals and conserva- Once understanding was reached on this tion measures would be financed out of a point, determination of incremental costs was Integrating Global Dimensions into Land Degradation Control Projects 37 relatively straightforward. Two factors, which ried out, identifying possible links to problems may not always be present, eased the deter- of global concern is substantially easier-parti- mination of incremental costs in the project con- cularly in the case of links to biodiversity. cepts. First, since the project concepts were The experience of the project concepts also drawn from projects already under preparation, suggests that establishing the nature and extent the baseline was already well defined. Second, of linkages between land degradation on agri- most of the activities specifically designed to cultural land and biodiversity is simpler than generate global benefits were additional to those doing so for climate change. In the case of foreseen in the baseline projects, making the biodiversity, the main constraint is that the bio- incrementality obvious. If generating additional diversity at risk from land degradation on agri- global benefits had involved modifying the base- cultural land may not have been as well studied line activities, apportioning costs would have as biodiversity in protected areas. In the case of been more complicated. climate change, the main constraint is that very few data exist on changes in emission resulting Summary from changes within a given land use (as oppo- sed to changes in emissions resulting from It should be stressed that these project concepts changes in land use itself. The data require- are only illustrative of the potentials and pitfalls ments are also more stringent in the case of of attempting to integrate attention to global climate change, due to the need to demonstrate benefits into land degradation control projects. cost-effectiveness. Moreover, even the most advanced of the pro- The design of a GEF dimension for projects jects has only reached the advanced prepa- benefits from starting as early as possible in the ration stage; no doubt additional lessons will be project cycle-ideally at the identification stage. learnt during implementation. This will allow time for any data necessary for The projects which seem to lend themselves project development to be collected. best to integrating global dimensions are those in which field activities are being carried out in Conclusion specified areas. It is far more difficult to identify potential global dimensions of projects that The land degradation question as a whole has seek to combat land degradation through policy not been marked by substantial success. More reforms or support to research and extension. recent approaches, based on improved analyses Because of the site-specificity of land degra- of the causes and incentive structures within dation problems, it is very difficult to determine which land degradation occurs and employing the nature and extent of links to global prob- a participative approach, appear to have a mar- lems when only policy-level measures are con- kedly improved chance of success. However, templated. The synergy between measures to integrating global dimensions into such efforts generate global benefits and the activities un- will not be an easy task. dertaken in these projects also tends to be low. Conversely, when field activities are being car- 4. Conclusions and Next Steps There are clearly important direct and indirect lead to adverse global effects, specific infor- linkages, in many instances, between land mation on the nature and magnitude of degradation on agricultural land and problems these effects is extremely scarce. of global concern. Although these linkages are * Implementation. Land degradation control not likely to be as strong as those encountered programs have proven extremely difficult in cases of deforestation, land degradation on to implement unless on-site benefits are agricultural land can: high. When local and global benefits are * Reduce the ability of soils to serve as a highly correlated, this does not pose a carbon sink and release carbon currently problem for the achievement of global stored in soils to the atmosphere. benefits, although it does imply that there * Lead to loss of plant and animal biodiversity may be little need for outside financing in rangelands; (either from national or global sources). * result in increased pressure to convert natu- When the correlation is low, so that different ral habitats to agricultural use. or additional interventions are required to * Damage international waterways. secure global benefits, however, the prob- The strength of these linkages will vary sub- lems might be substantial. Just as it has pro- stantially from case to case, however. In cases ven extremely difficult to undertake land where land degradation results in adverse glo- degradation control efforts that mitigate na- bal effects, land degradation control can result tional off-site damages in a sustainable way, in positive global effects. it will be difficult to mitigate global off-site These linkages, when they are present, open damages in a sustainable way. the possibility that efforts to control and reverse Given these difficulties, the next steps in the land degradation and efforts to mitigate global development of effective programs to address problems can be mutually supportive. Equally land degradation and global concerns need to clearly, however, explicitly incorporating global be undertaken on two parallel tracks. First, considerations into land degradation control efforts need to be made to plug some the gaps efforts will not be easy. in the data-on both local and national prob- The two main difficulties likely to be en- lems and on any resulting global effects. countered by efforts to incorporate global consi- Second, there needs to be experimentation with derations into land degradation control different approaches to incorporating global activities are: concerns in land degradation control projects. * Information. If land degradation processes are to be better addressed, there is a clear Research Needs need for an improved understanding of their extent, nature, and severity. In terms of link- Any effort to develop projects that address both ages to global problems, although sufficient land degradation and global concerns must be information is available to indicate that land based on a clear understanding of the under- degradation can, under many conditions, lying land degradation problems, their causes, 38 Conclusions and Next Steps 39 and their effects at both the farm and national Box 9. Land Quality Indicators levels, as well as on an understanding of their linkages to problems of global concern. Yet Land quality is the condition or health of the land rela- accurate and reliable information on both of tive to its capacity for sustainable land use and en- these aspects is often not available. vironmental management. At present, however, few indicators are available that allow changes in land re- National research systems. Given the site- sources to be monitored and evaluated. The Land specificity of land degradation problems, Quality Indicators (LQI) Program being developed by National Agricultural Research Systems a coalition of international agencies including the World (NARS) clearly have an important role to play Bank, FAO, UNDP, and UNEP, aims to help fill this gap in understanding its effects on problems of both [Pieri and others, 1995]. national and global concern. However, NARS LQIs are needed to address major land-related issues have traditionally been oriented almost solely such as land use pressures, land degradation, and soil towavetraditionally beenort-teriinted lonstler and water conservation. The LQI program is deve- towards short-term productivity consider- loping a core set of indicators to achieve this. Once ations, an approach that many maintain to this developed, LQls will be used for policy and program day. In addition, many NARS are institutionally formulation for district, national, and global assess- weak and short of resources. Their links to ex- ment, environmental impact monitoring, and to pro- tension have also often been weak. mote technologies, policies and programs to ensure better use of natural resources and sustainable land International agricultural research. The whole management. focus of the international agricultural research The LQIs being developed for immediate application centers in the CGIAR system was also tradi- include: tionally solely on productivity enhancement; * Nutrient balance: describes nutrient stocks and flows under specific land uses. environmental and natural resources concerns * Yield gap: compares current with potential yields. were not addressed and farmers' participation * Land use intensity: describes the impact of intensifi- was not sought. Although productivity growth cation on land quality. is still the main objective, environmental con- * Land cover: describes the extent, duration, and timning cerns are now also being considered, for of vegetative cover. Indicator guidelines for nutrient balance and yield gap example through farming systems research. In have been prepared; guidelines for the other indicators 1996, the CGIAR centers spent about US$49 are in various stages of development. Additional LQIs million (about 17 percent of overall expen- that attempt to capture soil quality, land degradation, ditures) on soil and water conservation re- and agrobiodiversity are still in the research stage. search. About 35-40 percent of soil and water Work is also underway on developing indicators of the conservation research is being carried out ini extent of carbon storage (above- and below-ground) under different land uses. high potential areas (irrigated areas and rainfed under_different_land_uses. lowlands). Funding for soil conservation re- search in the CGIAR system is 4 times that for water conservation. About 70 percent of CGIAR is also required on the global dimensions of research focuses on on-site effects at the field or land degradation. This is likely to prove more farm levels; about 30 percent looks at effects at difficult to organize on a systematic basis, since the community and watershed levels. national research organizations have no direct In an effortto improve the availability and incentive to undertake such research and, Inanity efforta tonland improve tion, the avaad indeed, often have insufficient resources even qalkity coflaataong land degradation,ther orld for their national research objectives. Mindful Bank is collaborating with severalor oran of this, the GEF has already financed targeted zBox 8iI research in areas where the potential contri- (Box 8). bution to global problems is significant. For In addition to the research required to example, the Alternatives to Slash and Burn develop appropriate responses to land degra- program examines the greenhouse gas emis- dation at the local and national level, research 40 The Global Environmental Benefits of Land Degradation Control on Agricultural Land sions and biodiversity effects of slash-and- since land degradation problems, and their glo- burn practices and alternative land use systems bal dimensions, tend to be highly site-specific, in tropical forests [Tinker and others, 1996]. it is not usually necessary to have complete Global dimensions in World Bank Agriculture information on conditions in the entire country. Sector Work. As noted earlier, attention to land What is needed is information on the specific degradation problems has not always been area where interventions are being envisaged. fully integrated into the Bank's agriculture sector work. In countries threatened by degra- Operationalizing the Global Dimensions of dation, greater efforts need to be made to un- Land Degradation derstand the nature and severity of degradation problems and the threats they might pose to The World Bank has already devoted consi- agricultural sustainability and to integrate derable resources to assist client countries un- efforts to address them into agriculture sector dertake land degradation control activities strategies. [World Bank, 1996, 1998]. During 1990-95, the The need to address global dimensions, Bank financed 108 projects (for a total of where relevant, adds an additional layer. US$13.4 billion) that aimed at improving land Through the Global Overlay Program, the management in dryland areas; of these, 34 World Bank is working to ensure that attention projects (for a total of US$2.2 billion) dealt to global problems becomes a regular part of primarily with land degradation Additional mainstream Bank work. Guidelines for Climate projects are in the pipeline. Change Overlays have already been prepared Soil fertility initiative. In the Africa region, [World Bank, 1997], and best practice guide- where land degradation problems are most lines for biodiversity overlays are under prepa- severe, a Soil Fertility Initiative (SFI) has been ration [Pagiola and others, 1997]. The Bank launched by the World Bank, in partnership needs to strengthen its agriculture sector work with a number of international, public, and pri- to effectively help developing country partners vate organizations. The objectives of the initia- mainstream climate change mitigation and tive are to help reverse the cycle of soil degra- biodiversity conservation in planning for this dation in Sub-Saharan Africa and increase the sector. sustainability of agricultural production. Soil In a local context, Bank staff interacting with fertility improvement action plans will be deve- country sector planners need to be able to loped and incorporated in country assistance address several questions: strategies, and soil fertility components will be * What are the extent, severity, and causes of included in relevant Bank-financed operations. the land degradation problems facing the This work program began with several pilot country? countries, in which the activities are being * What are the effects of degradation at the tested and refined. These include Guinea, farm and national levels? Madagascar, Malawi, Mali, and Uganda. The * How do the specific land degradation pro- FAO is participating in the design and super- cesses at work affect problems of global sion of specific soil fertility project compo- concern? Of course, full and complete answers to these It is quite likely that many of the land degra- questions will not be possible in most cases, dation control projects already planned or given the current state of knowledge. Even pre- underway will generate global benefits. Efforts liminary efforts to answer them can often pro- will be made to also incorporate explicit atten- vide considerable information and help target tion to global problems in forthcoming land research efforts towards the most important degradation projects, with financial support gaps. It is also important to bear in mind that from the GEF when appropriate. Conclusions and Next Steps 41 The point of departure of any effort to incor- * Whether the proposed activities are consis- porate global dimensions into land degradation tent with GEF's Operational Programs; and control efforts, must be a well-thought out * What the expected incremental costs from strategy to address the local and national the proposed activities are. aspects of the problem. This requires a clear The project concepts developed by IFAD in understanding of the nature, extent, and seve- collaboration with the World Bank provide rity of land degradation problems, their causes, some initial examples of what is possible in this and their effects at both the farm and national field. levels. It also requires, as discussed previously, a clear understanding of the incentives and Conclusion constraints faced by land users. At this stage, it should be possible to deter- For many countries-and in particular for mine whether the measures already envisaged many African countries-land degradation on will be sufficient to address most global prob- agricultural land is posing substantial threats lems originating at the site, or whether addi- to sustainabilityh economic growth and the tional or different measures would be required welfare of the rural population. Strong efforts to do so. Here too, project designers need to to combat land degradation are justified on ensure that the measures being contemplated of prounds alobal I n s uch as miti- are consistent with land users' incentives and of problems of global concern such as miti- constraints. Since such measures will usually gation of climate change or conservation of bio- inolve what is, from the perspective of land diversity provide an additional reason to com- ioe wthebatis,fromenthfe perspective sof land bat degradation. At times this may require users, the abatement of externalities, some me- additional or different measures than if local chanism will often be required to compensate and national considerations were the only ones them for undertaking the proposed measures. involved. As discussed in Chapter 3, this will not be an easy task. In cases where there are strong linkages In cases where an application for GEF finan- between land degradation and problems of cing for the incremental cost of activities de- global concern, efforts to combat both can be signed to address problems of global concern is mutually supportive. It is important, however, being envisaged, it will then also be necessary to avoid having the tail wag the dog. The pri- to determine mary motivation for land degradation control * How the proposed measures are likely gene- efforts will remain the local and national bene- rate benefits in the GEF's focal areas (climate fits that can be derived thereby. Linkages to change, biodiversity conservation, and pro- global problems are not always present, or may tection of international waters); not be sufficiently strong to warrant specific attention. Glossary Agroforestry - A land use in which trees or livestock production activities and/or other other woody perennials are incorporated renewable resources activities (the "terroir"). into fields used for crop or animal pro- Greenhouse Gas - Chlorofluorocarbons (CFCs), duction. carbon dioxide (CO2), methane (CH4), and Biodiversity - Short for biological diversity; it nitrous oxide (N20) are the main greenhouse encompasses the variability among living gases which are significantly increased by organisms from all sources, including, human activity. among others, terrestrial, marine and other Intensification - Increasing the use of inputs aquatic ecosystems and the ecological com- and/or changing land use so as to increase plexes of which they are part; this includes productivity (output per unit of land). diversity within species, between species, Natural habitat - Land and water areas where and of ecosystems. (i) the ecosystems' biological communities Extensification - Increasing agricultural pro- are formed largely by native plants and ani- duction by expanding the area under culti- mal species, and (ii) human activity has not vation. essentially modified the area's primary Gestion des Terroirs - Approach to commu- ecological functions. nity-based land resource management Off-site effects - Effects of a land use change developed in West Africa. Under this that are felt outside the area on which the approach, communities design and imple- land use change is carried out. ment, with the assistance of a multidisci- On-site effects - Effects of a land use change plinary team of technicians, a management that are felt within the specific area on which plan for the territory on which they live per- the land use change is carried out. manently or seasonally and on which they regularly carry out agricultural and/or 42 References Agro-Concept. 1995. Plan National d'Amenagement Crosson, P.R., and A.T. Stout. 1983. Productivity des Bassins Versants. Rabat: Ministere de Effects of Cropland Erosion in the United States. l'Agriculture et de la Mise en Valeur Agricole. Washington: Resources for the Future. Behnke, R.H. Jr, and I. Scoones. 1993. "Rethinking Crutzen, P.J., and M.O. Andrea. 1990. "Biomass Range Ecology: Implications for Rangeland Burning in the Tropicas: Impact on Management in Africa." In R.H. Behnke Jr, I Atmospheric Chemistry and Biogeochemical Scoones, and C. Kerven (eds), Range Ecology at Cycles." Science, No.250, pp.1669-1678. Disequilibrium. London: Overseas Development Cumming, D.H.M. 1994. "Are Multispecies Institute. Systems a Viable Landuse Option for Southern Bojo, J., and D. Cassells. 1995. "Land Degradation African Savannas?" Paper presented at the and Rehabilitation in Ethiopia: A International Symposium on Wild and Reassessment." AFTES Working Paper No.17. Domestic Ruminants in Extensive Land Use Washington: World Bank. Systems, Berlin, 3-4 October 1994. Brady, N.C. 1986. The Nature and Properties of Soils. Current, D., E. Lutz, and S.J. Scherr. 1995. "The Ninth Edition. New York: MacMillan. Costs and Benefits of Agroforestry to Farmers." World Bank Research Observer, Vol.10 No.2, Brown, L.R., and E.C. Wolf. 1984. Soil Erosion: pp.151-180 Quiet Crisis in the World Economy. Worldwatch Paper No.60. Washington: Worldwatch de Haan, C. H. Steinfeld, and H. Blackburn. 1997. Institute. Livestock-Environment Interactions: Finding a Balance. Brussels: European Commission Chile, B. 1996. "Zimbabwe." In E. Lutz and J. D)irectorate-General for Development. Caldecott (eds), Decentralization and Biodiversity Dick orae-GVnelehor D.eRand . Conservation. Washington: World Bank. Dickinson, R.E., V. Meleshko, D. Randall, E. Sarachik, P. Silva-Dias, and A. Slingo. 1996. Chomitz, K., and D.A. Gray. 1996. "Roads, Land "Climate Processes." In Intergovernmental Use, and Deforestation: A Spatial Model Panel on Climate Change, Climate Change 1995: Applied to Belize." World Bank Economic Review, The Science of Climate Change. Cambridge: Vol.10 No.3, pp.487-512 Cambridge University Press. Chomitz, K., and K. Kumari. 1996. "The Domestic Dregne, H.E., M. Kassas, and B. Rozanov. 1992. Benefits of Tropical Forests: A Critical Review." "Status of Desertification and Implementation Working Paper GEC96-19. Norwich: CSERGE. of the United Nations Plan to Combat Cleaver, K.M., and G.A. Schreiber. 1994. Reversing Desertification." Desertification Control Bulletin, the Spiral: The Population, Agriculture, and Vol.20, pp.6ff. Environment Nexus in Sub-Saharan Africa. Duxbury, I. 1995. "The Significance of Greenhouse Washington: World Bank. Gas Emissions from Soils of Tropical Clark, E.H., H.J. Haverkamp, and W. Chapman. Agroecosystems." In R. Lal, J. Kimble, E. 1985. Eroding Soils: The Off-Farm Impacts. Levine, and B.A. Stewart, eds, Soil Management Washington: Conservation Foundation. and Greenhouse Effect: Advances in Soil Science. Boca Raton: CRC Press. 43 44 The Global Environmental Benefits of Land Degradation Control on Agricultural Land English, J., M. Mortimore, and M. Tiffen. 1994. Lal, R., and T.J. Logan. 1995. "Agricultural "Land Resource Management in Machakos Activities and Greenhouse Gas Emissions from District, Kenya, 1930-1990." Environment Paper Soils in the Tropics." In R. Lal, J. Kimble, E. No.5. Washington: World Bank. Levine, and B.A. Stewart, eds, Soil Management Global Environment Facility (GEF). 1996a. "A and Greenhouse Effect: Advances in Soil Science. Framework of GEF Activities Concerning Land Boca Raton: CRC Press. Degradation." Washington: GEF. Lampietti, J.A. and U. Subramanian. 1995. "Taking Global Environment Facility (GEF). 1996b. Stock of National Environmental Strategies." "Incremental Costs." GEF Council Report Environment Department Paper No.10. GEF/C.7/Inf.5. Washington: GEF. Washington: World Bank. Global Environment Facility (GEF). 1997. "GEF Lutz, E., S. Pagiola, and C. Reiche. 1994. "Cost- Operational Programs." Washington: GEF. Benefit Analysis of Soil Conservation: The Goudriaan, J. 1992. "Biosphere Structure, Carbon Farmers'Viewpoint." The World Bank Research Sequestering Potential and the Atmospheric 14C Observer, Vol.9 No.2, July, pp.273-295. Carbon Record." Journal of Experimental Botany, Magrath, W.B., and P. Arens. 1989. "The Costs of Vol.43, pp.1111-1119. Soil Erosion on Java: A Natural Resource Hansen, S. 1997. "Incremental Costs as it Relates to Accounting Approach." Environment Land Degradation: Current Practices and Department Working Paper No.18. Methodological Issues." Paper presented at the Washington: World Bank. STAP Meeting on Incremental Costs and Mearns, R. 1996. "When Livestock are Good for the Global Benefits in Land Degradation, Enviromnent." IDS Working Paper No.45. Amsterdam, 17 June 1997. Brighton: Institute of Development Studies. Hassan, H., and H.E. Dregne. 1997. "Natural Nelson, R. 1990. "Dryland Management: The Habitats and Ecosystems Management in 'Desertification' Problem." Technical Paper Drylands: An Overview." Environment No.116. Washington: World Bank. Department Paper No.51. Washington: World Oldeman, L.R., R.T.A. Hakkeling, and W.G. Bank. Sombroek. 1990. World Map of the Status of Houghton, R.A. 1994. "The Worldwide Extent of Human-Induced Soil Degradation: an Explanatory Land-use Change." BioScience, Vol.44 No.5, Note. Revised Second Edition. Wageningen: May, Special Issue on the Global Impact of International Soil Reference and Information Land-Cover Change, pp.305-313. Centre. Kellenberg, J., and D. Cassells. forthcoming. Ottichilo, W.K. 1996. "Wildlife-Livestock "Incorporating Global Externalities in Forest Interactions in Kenya." Paper presented at the Conservation and Management." Global Seminar on Balancing Livestock Production Overlays Program. Washington: World Bank. and the Environment, Washington, 27-28 King, K. and K. Kumari. 1997. "Incremental Costs September 1996. of Focal Area Activities that also Address Land Pagiola, S. 1993. "Soil Conservation and the Degradation." Paper presented at the STAP Sustainability of Agricultural Production." PhD Meeting on Incremental Costs and Global Dissertation, Stanford University. Benefits in Land Degradation, Amsterdam, 17 Pagiola, S. 1994. "Soil Conservation in a Semi- June 1997. Arid Region of Kenya: Rates of Return and Kiss, A. 1990. "Living with Wildlife: Wildlife Adoption by Farmers." In T.L. Napier, S.M. Resource Management with Local Participation Camboni, and S.A. El-Swaify (eds), Adopting in Africa." Technical Paper No.130. Conservation on the Farm. Ankeny: Soil and Washington: World Bank. Water Conservation Society. Lal, R.C. 1987. "Effects of Erosion on Soil Pagiola, S. 1995. "The Effect of Subsistence Productivity." CRC Critical Reviews in Plant Requirements on Sustainable Land Use Sciences, Vol.5 No.4, pp.303-367. Practices." Paper presented at the Annual References 45 Meetings of the American Agricultural Soule, and P.L. Woomer. 1997. "Soil Fertility Economics Association. Indianapolis, 6-9 Replenishment in Africa: An Investment in August 1995. Natural Resource Capital." Paper presented at Pagiola, S., and M. Bendaoud. 1994. "Long-run the Workshop on Soil Fertility Replenishment Economic Effects of Erosion on Wheat in Africa, Nairobi, 3-9 June 1997. Production in a Semi-Arid Region of Morocco: Schiff, M. and A. Valdes. 1992. The Political A Simulation Analysis." Agricultural Economy of Agricultural Pricing Policy Baltimore: Economics Staff Paper No.AE 95-12. Johns Hopkins University Press. Washington State University. Schroeder, P. 1993. "Agroforestry Systems: Pagiola, S., and J.A. Dixon. 1997. "Land Integrated Land Use to Store and Conserve Degradation Problems in El Salvador." In El Carbon." Climate Research, Vol.3, pp.53-60. Salvador: Rural Development Study. Report Scoones, I. 1995. "New Directions in Pastoral No.16253-ES. Washington: World Bank. Development in Africa." In I. Scoones (ed.), Pagiola, S., J. Kellenberg, L. Vidaeus, and J. Living with Uncertainty. London: IT Srivastava. 1997. Mainstreaming Biodiversity in Publications. Agricultural Development: Toward Good Practice. Sombroek, W.G., F.O. Nachtergaele, and A. Hebel. Environment Paper Number 15. Washington: 1993. "Amounts, Dynamics and Sequestering of World Bank. Carbon in Tropical and Subtropical Soils." Parton, W.J., G. McKeown, V. Kirchner, and D. Ambio, Vol.22 No.7, pp.417-426. Ojima. 1992. CENTURY Users' Manual. Fort Srivastava, J., N.J.H. Smith, and D. Forno. 1996. Collins: Colorado State University. "Biodiversity and Agriculture: Implications for Pieri, C. 1992. Fertility of Soils: A Futurefor Farming Conservation and Development." Technical in the West African Savannah. New York: Paper No.321. Washington: World Bank. Springer-Verlag. Steinfeld, H., C. de Haan, and H. Blackburn. 1997. Pieri, C., J. Dumanski, A. Hamblin, and A. Young. Livestock-Environment Interactions: Issues and 1995. "Land Quality Indicators." Discussion Options. Brussels: European Commission paper No.315. Washington: World Bank. Directorate-General for Development. Pratt, D.J., F. Le Gall, and C. de Haan. 1997. Stoorvogel, J.J., and E.M.A. Smaling, 1990. "Investing in Pastoralism: Sustainable Natural "Assessment of Soil Nutrient Depletion in Sub- Resource Use in Arid Africa and the Middle Saharan Africa: 1983-2000." Report No.28. East." Technical Paper No.365. Washington: Wageningen: Winand Staring Centre. World Bank. Thiollay, J.-M. 1995. "The Role of Traditional Repetto, R., and W. Cruz. 1991. Accounts Overdue: Agroforests in the Conservation of rain Forest Natural Resource Depreciation in Costa Rica. Bird Diversity in Sumatra." Conservation Washington: World Resources Institute. Biology, Vol.9 No.2, pp.335-353. Sanchez, P.A., R.J. Buresh, and R.R.B. Leakey. Tiffen, M., M. Mortimore, and F. Gichuki. 1994. 1996a. "Trees, Soil and Food Security." Paper More People, Less Erosion: Environmental presented at the Discussion Meeting on Land Recovery in Kenya. Chichester: John Wiley. Resources: on the Edge of the Malthusian Tinker, P.B., J.S.I. Ingram, and S. Struwe. 1996. Precipice?, London, 5 December, 1996. "Effects of Slash-and-Burn Agriculture and Sanchez, P.A., A.-M. Izac, I. Valencia, and C. Pieri. Deforestation on Climate Change." Agriculture, 1996b. "Soil Fertility Replenishment in Africa: Ecosystems and Environment, Vol.58 No.1, pp.13- A Concept Note." Paper presented at the 22. Workshop on Achieving Greater Impact from Tucker, C.J., H.E. Dregne, and W.W. Newcomb. Research Investments in Africa, Addis Ababa, 1991. "Expansion and Contraction of the Sahara 26-30 September 1996. Desert." Science, No.253, pp.299ff. Sanchez, P.A., R.J. Buresh, F.R. Mwesiga, A. Uzo Mokwunye, C.G. Ndiritu, K.D. Shepherd, M.J. 46 The Global Environmental Benefits of Land Degradation Control on Agricultural Land Umali, D.L. 1993. "Irrigation-Induced Salinity: A Woomer, P. and M.J. Swift (eds). 1994. The Growing Problem for Development and the Biological Management of Tropical Soil Fertility. Environment." Technical Paper No.215. New York: John Wiley & Sons. Washington: World Bank. World Bank. 1992. World Development Report 1992: Unruh, J.D., R.A. Houghton, and P.A. Lefebvre, Development and the Environment. New York: 1993. "Carbon Storage in Agroforestry: An Oxford University Press for the World Bank. Estimate for Sub-Saharan Africa." Climate World Bank. 1996. Desertification: Implementing the Research, Vol.3, pp.39-52. Convention. Second Edition. Washington: Van Orsdol, K.G., I.G. Girsback, and J.K World Bank. Armstrong. 1993. "Greenhouse Gas Abatement World Bank. 1997. Guidelinesfor Climate Change Thought Non-Forest Biomass Production: Global Overlays. Environment Department Allocating Costs to Global and Domestic Paper No.47. Washington: World Bank. Objectives." Environment Working Paper No.59. Washington: World Bank. World Bank. 1998. New Opprtunitiesfor Walling, D.E. 1988. "Measuring Sediment Yield Development: The Desertification Convention. from River Basins." In R. Lal (ed). Soil Erosion Washington: World Bank. Research Methods. Ankeny: Soil and Water World Resources Institute. 1992. World Resources Conservation Society. 1992-93: Toward Sustainable Development. New Williams, J.R., K.G. Renard, and P.T. Dyke. 1983. York: Oxford University Press. "EPIC: A New Method for Assessing Erosion's Young, A. 1989. Agroforestryfor Soil Conservation. Effect on Soil Productivity." Journal of Soil and Wallingford: CAB International. Water Conservation, Vol.38 No.5, pp.381-386. Wisniewski, J., R.K. Dixon, J.D. Kinsman, R.N. Sampson, and A.E. Lugo. 1993. "Carbon Dioxide Sequestration in Terrestrial Ecosystems." Climate Research, Vol.3, pp.1-5. The World Bank 1818 H Street, NW Washington, D.C. 20433, USA o Telephone: (202) 477-1234 Facsimile: (202) 477-6391 Telex: MCI 64145 WORLDBANK MCI 248423 WORLDBANK 8 World Wide Web: http://www.worldbank.org/ E e-mail: Books@worldbank.org 0 CD ISBN 0-8213-4421-80( ISN08134421816