WTP- 127 WORLD BANK TECHNICAL PAPER NUMBER 127 Watershed Development in Asia Strategies and Technologies John B. Doolette and William B. Magrath, editors EEDUV IM ODITIE'w DUE rtSTUDiE %-^ BDEVELI EVE OPMEN THEOIZ wf-_ _ BN AND Tl :N_I`R-YEHO SOURI 0ACA A _MNAUNINSTITU t?J2vN* AN TENU LMNG STA RECENT WORLD BANK TECHNICAL PAPERS No. 63 Mould, Financial Information for Management of a Development Finance Institution: Some Guidelines No. 64 Hillel, The Efficient Use of Water in Irrigation: Principles and Practices for Improving Irrigation in Arid and Semiarid Regions No. 65 Hegstad and Newport, Management Contracts: Main Features and Design Issues No. 66F Godin, Preparation des projets urbains d'amenagement No. 67 Leach and Gowen, Household Energy Handbook: An Interim Guide and Reference Manual (also in French, 67F) No. 68 Armstrong-Wright and Thiriez, Bus Services: Reducing Costs, Raising Standards No. 69 Prevost, Corrosion Protection of Pipelines Conveying Water and Wastewater: Guidelines No. 70 Falloux and Mukendi, Desertification Control and Renewable Resource Management in the Sahelian and Sudanian Zones of West Africa (also in French, 70F) No. 71 Mahmood, Reservoir Sedimentation: Impact, Extent, and Mitigation No. 72 Jeffcoate and Saravanapavan, The Reduction and Control of Unaccounted-for Water: Working Guidelines (also in Spanish, 72S) No. 73 Palange and Zavala, Water Pollution Control: Guidelinesfor Project Planning and Financing (also in Spanish, 73S) No. 74 Hoban, Evaluating Traffic Capacity and Improvements to Road Geometry No. 75 Noetstaller, Small-Scale Mining: A Review of the Issues No. 76 Noetstaller, Industrial Minerals: A Technical Review (also in French, 76F) No. 77 Gunnerson, Wastewater Management for Coastal Cities: The Ocean Disposal Option No.78 Heyneman and Figerlind, University Examinations and Standardized Testing: Principles, Experience, and Policy Options No. 79 Murphy and Marchant, Monitoring and Evaluation in Extension Agencies (also in French, 79F) No. 80 Cernea, Involuntary Resettlement in Development Projects: Policy Guidelines in World Bank-Financed Projects (also in Spanish, 80S, and French, 80F) No. 81 Barrett, Urban Transport in West Africa No. 82 Vogel, Cost Recovery in the Health Care Sector: Selected Country Studies in West Africa No. 83 Ewing and Chalk, The Forest Industries Sector: An Operational Strategy for Developing Countries No. 84 Vergara and Brown, The New Face of the World Petrochemical Sector: Implications for Developing Countries No. 85 Ernst & Whinney, Proposals for Monitoring the Performance of Electric Utilities No. 86 Munasinghe, Integrated National Energy Planning and Management: Methodology and Application to Sri Lanka No. 87 Baxter, Slade, and Howell, Aid and Agricultural Extension: Evidencefrom the World Bank and Other Donors No. 88 Vuylsteke, Techniques of Privatization of State-Owned Enterprises, vol. 1: Methods and Implementation No. 89 Nankani, Techniques of Privatization of State-Owned Enterprises, vol. II: Selected Country Case Studies No. 90 Candoy-Sekse, Techniques of Privatization of State-Owned Enterprises, vol. III: Inventory of Country Experience and Reference Materials No.91 Reij, Mulder, and Begemann, Water Harvesting for Plant Production: A Comprehensive Review of the Literature No. 92 The Petroleum Finance Company, Ltd., World Petroleum Markets: A Frameworkfor Reliable Projection No. 93 Batstone, Smith, and Wilson, The Safe Disposal of Hazardous Wastes: The Special Needs and Problems of Developing Countries (List continues on the inside back cover) WORLD BANK TECHNICAL PAPER NUMBER 127 Watershed Development in Asia Strategies and Technologies John B. Doolette and William B. Magrath, editors The World Bank Washington, D.C. Copyright ©D 1990 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 July 1990 Third printing November 1994 Technical Papers are published to communicate the results of the Bank's 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 appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. Some sources cited in this paper 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 boundaries, colors, denominations, and other information shown on any map in this volume do not imply on the part of the World Bank Group any judgment on the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to the Office of the Publisher at the address shown in the copyright notice above. The World Bank encourages dissemination of its work and will normally give permission promptly and, when the reproduction is for noncommercial purposes, without asking a fee. Permission to copy portions for classroom use is granted through the Copyright Clearance Center, Inc., Suite 910, 222 Rosewood Drive, Danvers, Massachusetts 01923, U.S.A. The complete backlist of publications from the World Bank is shown in the annual Index of Publications, which contains an alphabetical title list (with full ordering information) and indexes of subjects, authors, and countries and regions. The latest edition is available free of charge from the Distribution Unit, Office of the Publisher, The World Bank, 1818 H Street, N.W., Washington, D.C. 20433, U.S.A., or from Publications, The World Bank, 66, avenue d'Iena, 75116 Paris, France. ISSN: 0253-7494 John B. Doolette is a senior agriculturalist in the Agriculture Division, Asia Technical Department. Wiliam B. Magrath is a land resource economist and was in the Environment Policy Research Division, Environment Department. He is now in the Agriculture Division, Asia Technical Department of the World Bank. Library of Congress Cataloging-in-Publication Data Watershed development in Asia. (World Bank technical paper, ISSN 0253-7494 ; no. 127) Includes bibliographical references. 1. Soil conservation-Asia. 2. Watershed management- Asia. 3. Water conservation-Asia. 4. Rural development projects-Asia. I. Doolette, John B., 1928- . II. Magrath, William. III. Series. S625.A78 1990 333.73'095 90-12722 ISBN 0-8213-1606-0 - iii - CONTENTS Page No. FORWARD ..... ................................. ....................... v OVE VIEE........ RV....................W.. vii CHAPTERS 1. STRATEGIC ISSUES IN WATERSHED DEVELOPMENT....... 1 William B. Magrath and John B. Doolette 2. SOIL AND MOISTtRE CONSERVATION TECHNOLOGIES: REVIEW OF LITERATURE ..35 John B. Doolette and James W. Smyle 3. ECONOMIC ANALYSIS OF SOIL CONSERVATION TECHNOLOGIES. 71 William B. Magrath 4. ECONOMIC ANALYSIS OF OFF-FARM SOIL CONSERVATION STRUCTURES 97 William B. Magrath 5. REVEGETATION TECHNOLOIES..... o.. 109 Ajit K. Banerjee 6. LAND TENURE ISSUES IN WATERSHED DEVELOPMENT. 131 Augusta Molnar 7. A FRAMEWORK FOR PLANNING, MONITORING AND EVALUATING WATERSHED CONSERVATION PROJECTSO. .. . 159 Glenn S. Morgan and Ronald C. Ng S. BIBLIOGRAPHIES ON SOIL AND MOISTURE CONSERVATION TECHNOLOGIESO O.. 173 Global Studies on On-Farm Impacts. .. 173 Impacts on Soil Moisture and Surface Runoff......... 191 Impacts on Erosion and Sedimentation . ...... 203 Impacts on Crop Yield and Productivity .. .. 219 v FOREWORD This review of watershed development issues arose from the realiza- tion that a number of current and piai.Led World Bank-supported projects in the Asia region deal with the linkages between upl,nd productivity and environmen- tal conditions and are, in various ways, motivated by concern with downstream impacts such as flooding and sedimentation. A collaborative effort emerged, involving the Environment Department (within the Bank's Policy, Research and External Affairs Complex) and the Technical Department of the Asia Region. From the start, it focused on deepening the Bank's collective understanding of watershed development. High priority was attached to identifying discrete operational problems that could be better understood from review of existing data and analysis. In addition, the review was to provide overall guidance to the Bank's dialogue with borrowers on strategies for resource management. Working papers on six issues of direct operational concern were ini- tiated, to be conducted by World Bank staff and consultants in the context of ongoing operations. These working papers, presented as chapters in this report, illustrate methodological approaches to project analysis, summarize the state of the art on solutions to technical problems and discuss institu- tional and social processes that bear heavily on the viability of watershed management projects. In addition to the research and operational work that has gone into the working papers, a colloquium on watershed management was held at the World Bank in October 1988. Experts from research organizations and other agencies presented results of their work on a number of topics, including the impact of erosion on crop yields, sedimentation processes, and the impact of land tenure on development investments. The first chapter draws heavily from the presen- tations at the colloquium, discussions with officials in the region and a review of literature on watershed management. Daniel G. Ritchie Director Asia Technical Department Washington, D.C. July, 1990 - vii - OVERVIEW Watersheds as hydrologic units provide appropriate units for conceptualizing and implementing development investments. They comprise combinations of arable and nonarable land and drainage lines and are utilized by permanent and transient populations with varying skills and commitment to long term resource husbandry. The range of issues relevant to watershed management is enormous and includes environmental issues, crop and livestock production, a whole range of social and cultural concerns, infrastructure planning and entire questions of governance and control. This volume presents the results of a highly selective program of research and consultation. In Chapter 1, Magrath and Doolette present a discussion of the major watershed development problems of the Asia Region. Taking a policy and investment perspective, the chapter tries to sort out what can and cannot be reasonably expected from watershed management efforts. While not minimizing the importance of linkages between upstream landuse and downstream environmental quality, the authors suggest that there are severe limits to our ability to manage these linkages in a cost effective manner. However, they observe a wide range of technological opportunities for intensifying productive activities in the uplands that, in addition to being privately profitable, will ultimately have positive impacts on downstream areas. In light of this they conclude by proposing an overall approach to watershed development that focuses on small farm development and common property management. In Chapter 2, Doolette and Smyle examine the fundamental building blocks of watershed management projects. They present a careful review of the impacts of a broad range of land management technologies and illustrate the potential and constraints facing projects that attempt to influence erosion, runoff, and agricultural productivity. They make the point that, despite the availability of a wide range of options, most development projects have relied on a limited and generally high cost set of interventions. Emphasizing emerging, low cost methods of vegetative soil and moisture conservation, the authors reinforce the conclusion that agricultural productivity in upland areas can be intensified in an environmentally sound and sustainable manner. In Chapters 3 and 4, Magrath demonstrates how benefit-cost analysis techniques can be used to assist in the selection of watershed management technologies. Chapter 3 compares alternative systems for soil and moisture conservation and shows an approach for integrating information on the physical and economic dimensions of erosion and for overcoming uncertainty about the impact of new technologies. Chapter 4 employs a similar approach to small multiple purpose conservation structures that are seldom subject to analysis. In Chapter 5 Banerjee takes on revegetation of denuded forest land. The discussion addresses both technical issues, especially the need for moisture conservation practices that enhance survival rates, and social considerations, focusing particularly on generating participation in the early stages of project planning. - viii - Molnar, in Chapter 6, explores one of the most analytically troublesome subjects in conservation policy, land tenure. The chapter discusses the range of land tenure systems found in the uplands of Asia and examines the forces that operate within these systems to encourage or discourage conservation. Project interventions have often interacted with these forces to produce unanticipated results. The analysis in this chapter argues that efforts to understand local social processes, to utilize existing local groups and to identify privately profitable technologies will have high returns. Morgan and Ng, in Chapter 7, elaborate the connections between planning, monitoring and evaluation of watershed projects. These projects, which often are spread over large and remote areas, employ new and sometimes unproven technologies, and call for unusually high levels of community participation and local involvement, present unique management problems. In addition to conventional, sound management systems, watershed development projects can benefit from special studies, with carefully thought out experimental designs. Morgan and Ng point the way toward the incorporation of efforts of this kind into standard project practice. These topics were selected and addressed on the basis of available expertise, and the operational priority of pressing issues. The list of other topics that could have, and need to be, examined is indeed long. However, to have attempted to treat them all would have precluded serious examination of any. There continues, for example, to be pressing need for policy research on livestock management systems in upland areas, for work on environmentally sound upland infrastructure and for serious examination of connections between upland development strategies and water quality. Notwithstanding, we believe that together, the chapters in this volume provide a unified treatment of what we consider to be significant and generally unappreciated topics in watershed management. - 1 - 1. STRATEGIC ISSUES IN WATERSHED DEVE[OPMFNI William B. Magrath and John B. Doolette Following a description of the significance of watersheds to the Asia region, this chapter defines the problems in watersheds as loss of agricultural productivity due to ero- sion, deforestation, population pressure and poverty, sedi- mentation of infrastructure downstream, flooding and erra- tic stream flows. The main themes emerging from analyses of watershed problems are summarized and a strategy to address them is presented. The rationale for a watershed management approach is explored in the context of physical, economic and political linkages and takes into account the interplay between upland and lowland areas. In proposing how investments may be made to solve watershed problems, a case is made that the most technically and economically efficient approach would focus on site-specific technolo- gies that are environmentally benign. Specific actions that development agencies should pursue in their operations and in discussions with governments are presented. INTRODUCTION Land that can be defined as watershed in the Asia region is a signi- ficant proportion of the total.l/ Of a gross area of 1,700 million hectares (ha), about 236 million (142) has slopes exceeding 30X (upper watershed) and another 664 million (39%) in the 8-30X slope category (lower watershed). The 900 million hectares covering more than half the region constitute what is conventionally accepted as the watershed area. However, it is difficult to ignore nonirrigated land below 8Z slope, because most strategies discussed for steeper lands are relevant and can usefully be treated on a watershed basis. About 65Z of the region's rural population of 1.6 billion live and earn their livelihood in these rainfed, watershed areas. Despite the existence of soil conservation agencies and watershed management authorities in Asian countries, the real managers of these lands are local farmers and villagers. Constrained by poverty and technology, their pursuit of arable land, food and fodder has profound effects on the land and water resources of both upland and lowland areas. Mounting pressure on scarce land and forest resources, stemming from rising human and animal populations, is leading to severe environmental degra- dation throughout the region. The extent of degradation has not been exactly measured, but it is manifest in numerous ways, including high rates of soil erosion and declining 1/ The simple definition of watershed acknowledges the hydrologic concept of a watershed being the dividing line between two catchments: this has devolved to include the area of land contained within a drainage divide above a certain specified point on a stream. The latter is intended herein, and the land area above 30? slope is regarded as upper watershed and that between 8X and 30X slope as lower watershed. yields on large areas of agricultural land, reduced livestock-carrying capa- city, sedimentation of dams, reservoirs and irrigation systems, and clearance of forests with consequent loss of biological diversity ind forest products. Together these trends threaten the ability of upland people to sustain an already precarious existence. It is misleading to speak of the watershed problem of the Asia region. There are, in fact, multiple problems, some directly amenable to solution through physical actions requiring investment, some requiring policy reform and research first, and some, principally the consequence of geology and climate, which require continued adaptation and accommodation. These watershed problems are, however, connected by the fact that they can best be understood and dealt with in the context of physical planning units defined by the flow of water. It is important to recognize significant physical differ- ences between watersheds of the large river systems of the Hindu-Kush-Himalaya region, characterized by high rates of erosion linked to ongoing processes of tectonic uplift, and the smaller, steeper watersheds of insular Southeast Asia. The latter, which result from quite different geologic processes including volcanic activity and upheavals of the ocean floor, offer consider- ably different responses to human activity. The review attempts to distin- guish between these differences where appropriate and avoid inappropriate generalizations. DEFINING PROBLEMS AND RESPONSES Major Watershed Problems Loss of Agricultural Productivity Due to Erosion. The uplands of Asia include widely diverse land forms. The Hindu-Kush-Himalaya region extends from Pakistan across northern India, Nepal, Bhutan and China and includes the world's highest mountains and poorest countries. The range is in a continuous state of formation as tectonic drift drives the Indian plate under the Eurasian Plate at a rate of 5 cm per year, lifting the Himalayas 1 cm per year in altitude. Volcanic activity, on the other hand, is responsi- ble for the original formation and continuous change of much of insular South- east Asia. These geologic processes, combined with intense tropical rainfall, are also responsible for the formation and high agricultural productivity of the alluvial plains of the region. Uplifted marine sediments provide yet another facet of the Asian uplands' fragile calcareous soils. The estimated distribution of the region's major soils is summarized in Table 1.1. The diversity of soil types is often just as pronounced at the local level as for the region as a whole. An important common characteristic of the soils of the region is their susceptibility to productivity loss due to erosion. Those soils that are particularly susceptible, principally the Luvisols, Acrisols, Nitrosols, Lithosols and Ferrasols, constitute nearly 75Z of the arable upland area. Of the remainder, volcanics (Andosols), deep loess deposits and some alluvial soils (Fluvisols), which together account for only about 102 of arable area, have deep effective rooting depth and are highly insensitive to productivity loss from erosion. 1-aLl 1.1- -ST Of ULOPD1i LAD Di TtE ASA hamN BY SOIL TYPE AD SLOPE CLASS (00 ha) Papuia bOgladeb UOwtmn uren Chin0l IWdia Indoneia Kaopucha Korm Lao FM Nalaysia Nepal Guinea Philippin.a Sri Lanka Thailand Vietnan Tetal Percent of total -- Rl ing to HillIy (a - St Sltp.) ---------- Aerleola 20I516 7.022 10.794 2.403 3.210 J14225 214 Comblool* 1379e 6 13.614 9on 4iS 29 1 Sol S0. 7' 2.4 .ra-o.am Uthet-' i0 to2 t27 o - Parraeoel 1JO i 12 1,8i - Listhanlan, 971 6.1I" En 7.748 0.4 LAkviaois 786 ~~~40 U8.m 8.669 .4833 42433 Htellirees lye70e,s lS mJiJ2 ,4 , I,lo i 1,OS2J 240,0JJ iS8 Padma l- 2,5 e " O,U 7,2 J 4t12t J t t2104 48 Arepaol,# 14.20 S 440 42 4tO iA4PORGIS el- J17 2,ff 7 * OeJ ti4 I,ltiO -917 - AMA0014 47 1.0 1.U ,t 12S, 8j45 0 4t Xero.ola o Yermla aq,u~~~ ~ ~~~~~~~~o.s 1933 1.1 Satantal mailing to hilly 5.780 aRe,t 91,819 28.27 4 .2 8 24 1.741 , .0 12° 8.3 1.0C 286 240,033 13.0 Stel ieseegt edo lbatsteaeu (Slop.. ) amS Aet leole t.66 J 84 t55, l04.,09 7.1 i, 48.67t 448 1.28 gm8 167.110 9.6 Tetm l 1.63 I4 4 9.956 4.4t0 141,90 18,794 1.206 S.978 90.2gm 1u4 41,795 8.0 Fanalaol 2..24- 61060. otellbSet FJ J 7 iO O D 7 U t S2 0 U 7 * t 4t t lt 9 0006 3 1S9 Litoeola113&.eO"6.4 1.77`5 1775 0.1 Litbamis 947 2M 1 tD10 367244 29042 61 49 42 47 400 3.108 691 106 1.866 907 40D08 2O.4 Loviaola 817 2.6915 V.427 3.0064 1.160 448 688 406 is:1in 0.9 Orowasma 9.348 7,747 0.1 Nitrmoesl lo 7.684 491 814 13 1.735 1.243 9.494 0.4 e _toi Stoepls diseectedw.. 199 .1 9184 3. to eouatainoue 1.111111 4.450 46.849 472.447 62.710 M5.789 42 34200 9- `"2.9 36921 1.4 " 13 6104 30 Total sloping 81.312 563.399 134.089 69.010 4g211 9.082 10.620 31.680 17.194 2.674 1.994 1.831 931.357 82.6 Total area 13.400 4.700 163.100 942600 297.800 161.200 17.700 9.6100 113.100 89.600 18.700 43.000 29.600 6.,00 51.200 82.300 1.727,60o Railing to hilly 6~~~~~~.7 10.0 80.7 14.5 89.0 12.7 6.6 43.6 15.9 0.7 13.9 Parcest, 18.9 94.7 619.2 60.7 21.1 80.6 0.2 14.1 0.3 6.1 96.3 89.1 12.1 25.2 8.9 4.0 40.0 Steeply dleaeete to .mteiaoua Perem sl tapies 1.9 94.7 73.0 60.6 51.6 46.8 0.2 46.2 0.3 6.2 79.0 70.3 7.7 41.1 3.9 4.6 88.9 J_see; Wl,ADM , qIrld toil 1 -. ToAl Ild I_aM. fr y Oeld ReeMUM. Ieatlt4S. V1tnj Rftaa,, i Ft_ort (16). -4- A study of the costs to the economy of soil erosion on the uplands of Java illustrates the magnitude of these damages. Based on analysis of factors .causing erosion, the impact of erosion on the productivity of different soils, and the economics of alternative cropping systems, it was estimated that ero- sion costs the economy US$315 million annually. In Nepal, overall yields of cereal fell by over 12 per year from 1970/71-1980181 in the Hills (Yadav, 1987) (Table 1.2). aErosion's role in this decline is not known but in the Terai where erosion is less significant, yields were essentially constant. The impact of runoff and erosion on productivity and yield is discussed in Chapter 2. Table 1.2: NEPAL - ANNUAL GROWTH RATE OF CEREAL CROP AREA, PRODUCTION, AND YIELD, BY REGION, 1970/71-1980/81 Hills Terai Total Nepal Crop Percent t-statistic Percent t-statistic Percent t-statistic Paddy Area 1.98 8.93 0.71 2.96 0.93 4.10 Production 0.72 1.45 0.73 0.87 0.73 1.03 Yield -1.24 -3.25 0.02 0.02 -0.20 -0.39 Maize Area 0.72 4.94 -0.86 -3.99 0.24 1.81 Production -1.14 -2.30 -0.88 -1.11 -1.06 -2.62 Yield -1.84 -4.70 -0.02 -0.02 -1.30 -3.68 Wheat Area 2.08 3.41 8.50 6.60 5.82 10.18 Production 3.86 6.05 12.79 5.17 8.64 7.54 ,Yield 1.74 3.51 3.96 2.13 2.67 2.74 Barley Area 0.24 1.00 -3.35 -1.89 -0.71 -2.35 Production -1.61 -3.48 -0.20 -0.08 -1.19 -1.83 Yield -1.85 -4.16 3.25 2.61 -0.49 -1.26 Millet Area 0.69 2.07 -0.24 -0.56 0.54 1.78 Production -0.54 -0.86 0.16 0.31 -0.44 -0.81 Yield -1.22 -3.54 0.41 2.41 -0.97 -3.32 Total cereals Area 1.27 8.77 1.50 5.90 1.42 8.13 Production 0.12 0.38 1.44 2.30 0.95 2.10 Yield -1.14 -3.87 -0.06 -0.12 -0.46 -1.39 Notes: Computed from unpublished data on area, production, and yield from the Department of Food and Agricultural Marketing Services, Nepal. 1979/80 is omitted because it was a drought year. -5- Investments in agricultural research and extension oriented toward upland crops have lagged behind those for lowland areas, as evidenced by con- siderably lower rates of penetration of high-yielding varieties in the uplands. In Nepal, for example, high-yielding varieties account for 33Z of rice, 91Z of wheat, and 302 of maize in the Terai, while in the Hills, respec- tively, only 211, 87S and 16S, and in the mountains only 61, 721 and 72 (see Table 1.3). Table 1.3 AREA COVERED BY IMPROVED VARIETIES IN NEPAL (Percent of Total Cropped Area) (1985) Crop Mountain Hill Terai Nepal Paddy 6.0 21.0 33.0 20.0 Maize 7.0 16.0 30.0 18.0 Wheat 72.0 87.0 91.0 84.0 Potato 6.0 5.0 5.0 5.0 Source: R.M. Joshi and M.K. Khatiwad, 'Agriculture Handbook Nepal," 1986. Deforestation. Approximately 192 of the region is under closed forest. It is estimated that forest cover is receding at around one percent per year and that, in addition, degradation through overcutting and grazing is reducing productivity on much of the remaining stand. In much of the region forest resources are integral to the agricultural system as sources of fodder and minor products. In India, for example, fodder available from forest and wasteland is estimated to be almost 302 of the total availability (600-620 million tons, dry matter). It is important to understand the dimension of degeneration in forest areas. In the Asia region, the steeper upper watershed areas were mostly naturally forested. Over time, characteristically they have been overex- ploited for timber, fuel and fodder, and now many areas are no longer forested and in others the forest is extremely degraded. Regarding runoff/erosion, evidence suggests that trees by themselves in such circumstances provide little soil conservation benefit except to the extent that they foster the understory.2/ This understory of shrubs, herbs and litter is what protects the soil surface, maintaining the natural higher rates of infiltration. Without it, even dense forests, as seems to be the case with some tree species, have high rates of erosion even when reasonably managed, while degraded forest with dense undergrowth of grass and shrubs with intact litter have low erosion rates. Data on volcanic soil with 102 slope in Indonesia 2/ This discussion pertains to rill and interrill (sheet) erosion. At the same time it is readily acknowledged that the root system of trees is probably the best of all kinds of vegetation in providing slope stability in landslip-prone areas. show the estimated rates of surface erosion in tons/ha/year as 5 tons or less for both degraded forest with dense undergrowth and a pristine forest with litter intact, yet greater than 75 tons for both a 40-year-old teak plantation and a forest with all litter removed (Carson, 1989). In discussing the ambivalent effects of vegetative cover for soil erosion control, Stocking (1988) points out that even at slopes of 55Z, rates of erosion from undisturbed forest are usually less than 0.5 tonslha/year, whereas planting a monoculture of Eucalyptus species as an erosion control measure stifled ground cover and accelerated sheet erosion. Population and Poverty. The population of the upper watershed areas in Asia is roughly 128 million, of which 27 million people live in the Himalayan region, 50 million in the steep uplands of China and about 30 mil- lion in the uplands of insular Southeast Asia. Table 1.4: UPLAND POPULATIONS OF THE ASIA REGION (Million) Upper Watershed Population Country Rural Population Slope >30Z Bangladesh 84.62 0.58 Bhutan 1.25 1.25 Burma 28.88 China 577.10 50.00 India 586.05 17.50 Indonesia 124.80 12.00 Kampuchea _ _ Korea 14.94 6.55 Lao PDR 3.15 Malaysia 9.98 Nepal 15.81 8.46 Papua New Guinea 2.92 1.34 Philippines 41.50 18.00 Sri Lanka 12.72 2.92 Thailand 43.13 8.42 Vietnam 50.64 Total 1,598.19 127.72 Sources Bank Staff Estimates. There are few reliable data to indicate whether poverty and landless- ness are more acute in upland areas than elsewhere in the region. Data from Indonesia indicate that landlessness is more common in lowland villages than in the uplands of Java, but conclude that many households in the uplands are among the poorest in Java (World Bank, 1989a). In the Philippines, it is estimated that recent immigrants to the uplands have an averaRe per capita - 7 - income of Peso (P) 2,168, well below the official poverty line (World Bank, 1989b). Downstream Sedimentation. The deposition of eroded material in reservoirs and irrigation systems is a major management problem throughout the region, yet a relatively small percentage of the total number of watersheds have such infrastructures. It is clear that sedimentation imposes a high cost in terms of shortened investment life, high maintenance requirements and reduced services. Sedimentation on Java alone is estimated to cost the econ- omy about US$26-91 million per year (World Bank, 1989a) (Table 1.5). Compa- risons of the design and currently estimated lives of reservoirs in India show that erosion and sedimentation are not only severe and costly, but accelerat- ing (Table 1.6). It is now obvious that the original project estimates of expected sedimentation rates were faulty, based on too few reliable data over too short a period. Table 1.5: TOTAL ESTIMATED ANNUAL COSTS OF SOIL EROSION ON JAVA (US$ million) West Java Central Java Yogyakarta East Java Java On-Site 141.5 29.1 5.7 138.6 315.0 Off-Site Irrigation System 1.7-5.7 0.8-2.7 0.1-0.5 1.2-4.0 7.9-12.9 Siltation Harbor Dredging (1984185) 0.4-0.9 0.1-0.3 - 0.9-2.2 1.4-3.4 Reservoir 9.0-41.3 3.5-16.3 - 3.8-17.3 16.3-74.9 Sedimentation Total 152.6-189.4 33.5-48.4 5.8-6.2 144.5-162.1 340.6-406.2 Source: World Bank (1989a). Flooding. Although floods are a natural feature of the lowland areas of the region, they nonetheless impose severe hardship on local population and national economies. In India between 1953 and 1976, 1,240 lives were lost, 77,000 cattle destroyed and annual property damage occurred in excess of Rupees (Rs) 2 billion annually due to floods. In 1988, the flood in Bangladesh claimed some 1,500 lives. As Rogers et al. (1989) point out, flood disasters in lower basins are caused by two much rainfall (or snow melt) in too short a time for the soil and the channels to handle, and are determined more by basin characteristics, river constrictions by bridges and roads, large amounts of compacted surface in cities, glacial outbursts, landslides, inade- quate levees and increasing flood plain occupancy, than by forest use or even forest conversion per se. - 8 - Table 1.6: SILTATION OF SELECTED INDIAN RESERVOIRS Expected Life as Percent of Reservoir Assumed Rate Observed Rate Design Life (acre-feet per annum) Z Bhakar 23,000 33,475 68 Maithon 684 5,980 11 Havurakshi 538 2,080 27 Nizam Sugar 530 8,725 6 Panchet 1,982 9,533 21 Ramgange 1,089 4,366 25 Tungabhadia 9,796 41,058 24 Ukai 7,448 21,758 34 Source: Brown and Wolf (1984). There are no reliable estimates of the economic damages caused by flooding. In addition to direct damages, floods, or more properly, the expec- tation of flooding, reduce perceived returns to investment and probably slow growth to a significant but unmeasurable degree. While it is likely that the floods of 1988 were the worst on record from an economic perspective, there is evidence that the physical severity of flooding has not worsened. An often neglected positive effect of flooding downstream is the delivery of nutrients to agricultural land. Soil moisture stored when flood waters recede also contributes to dry season yields. For example, in the dry season following the 1988 Bangladesh floods, production reached near-record levels. Dry Season Stream Flows. A direct consequence of excessive surface runoff that contributes to flash flooding is the reduced temporary storage of water in the soil profile and groundwater aquifers. Some of this stored water would normally have rejoined the surface water and contributed to stream flow in the dry season. Reduced dry season stream flow has serious consequences on downstream uses for power, irrigation and municipal supplies. Policy Responses Governments. Governments in the region and multilateral and bilat- eral donors have attempted to respond to the various watershed management problems of the region in a number of ways. Responsibility for watershed management is typically disbursed across a number of government agencies, including agricultural and forestry line agencies and in some cases free- standing watershed development agencies and soil conservation services. The common administrative approach is to focus on the implementation of physical investments on public and private land, often with a predominant single tech- nical solution, and on encouraging the adoption of conservation-oriented farm- ing practices on private land. Traditional low-technology farming systems have frequently been not only risk-minimizing, but also soil-conserving, - 9 - whereas cash crops are much riskier in both respects. Cash cropping has some- times exacerbated erosion problems (Carson, 1989). Donors. Greater interest in environmental issues by donor agencies has led to an increase in the level of support for watershed management proj- ects and programs. World Bank involvement in watershed management has, over the past decade, primarily been through forestry, agriculture and integrated rural development projects. To the extent that there has been a Bank strategy towards watershed development, it has focused on attempting to improve the productivity of smallholder agriculture, thereby leading to a reduction in environmental deterioration. Table 1.7 lists Bank-financed projects now under implementation in the Asia region. Table 1.7: WORLD BANK-ASSISTED WATERSHED MANAGEMENT RELATED PROJECTS UNDER IMPLEMENTATION China Red Soils Area Development Gansu Provincial Development Indonesia Yogyakarta Rural Development Upland Agriculture and Conservation Forestry Institutions and Conservation Thailand Northern Agricultural Development /a Philippines Central Visayas Regional Watershed Management & Erosion Control Bhutan Forestry Development India Kandi Watershed and Area Development /a Himalayan Watershed Management Haryana and Jammu Kashmir Social Forestry National Social Forestry Pilot Project for Watershed Development Nepal Second Rural Development Mahakali Hills Community Forestry Development and Training Second Forestry Marsyandi Hydroelectric /a Recently completed. The Evolution of Conservation Technology. The pattern of investment and organizational design followed by governments and donors has failed to keep pace with the rapid evolution of soil and moisture conservation technol- ogy for tropical situations that has occurred during the past 20 years. Early approaches to soil conservation were developed for large landholdings in tem- perate regions and were based on structural and engineering treatments, such as graded earth banks and broad grassed waterways. Attempts to apply these - 10 - approaches to developing country agriculture, characterized by small holdings, diverse cropping systems, extremes of topography and climate and severe limits on financial resources and skills, have proven disappointing. With rigorous attention to appropriate design criteria, engineered systems can, under some circumstances, function in tropical environments. However, experience has shown that these criteria are usually lacking. Moreover, the high unit costs of these technologies and their indivisibility imply that they are beyond the means of all but the most favored farmers and communities unless heavily sub- sidized. Even with government subsidies for initial investment costs, recur- rent costs in the form of maintenance and land taken out of production seem, judging by farmer response, to make these approaches uneconomic. 'While recognition of the technical shortcomings of traditional approaches to conservation--so far as smallholdings in the tropics are con- cerned--is growing, alternative technical and institutional approaches are emerging. For example, although it has long been known that maintaining con- tinuous vegetative cover is the most effective way of reducing sheet erosion, it has been difficult to promote heavy applications of mulch and retention of significant areas under permanent cover on small farms. Typically, the pres- sure on smallholders to cultivate all available land and to utilize all avail- able fodder militates against maintenance of permanent vegetation. Even when individuals farmers are inclined toward such practices, the widespread prac- tice of free grazing makes such a strategy impractical. The concept of conservation-oriented farming in the uplands in which farming systems and individual production treatments combine to conserve soil and water and improve total production and net benefit is now recognized. Currently, two complementary strategies for the development of conservation- oriented upland farming are evolving. The first is the adoption of a problem- solving approach aimed at identifying, on a site-specific basis, the key con- straints to and opportunities for expanding output. The second, possible because of the uniquely nonsite-specific characteristics of vetiver grass, Vetiveria zizanioides, is the widespread promotion of this grass for use as a contour hedgerow. Properly balanced, these two strategies can form the basis for a comprehensive approach to treatment of upland areas in the Asia region. A focus on small farmer development does not deny the seriousness of down- stream watershed problems in the region. It is, however, a recognition of both the difficulties of reducing downstream problems and of the complementa- rities between agricultural development and environmental improvement. The next section elaborates on the difficulties of dealing with downstream damages through land-use changes, such as reforestation and development of vegetative barriers, in upper watersheds, and the following section returns to the theme of strategies for small farmer development and common property management. INTERSECTORAL LINKAGES Discussions of watershed management are generally dominated by con- cern about physical linkages related to movement of soil and water within drainage basins. While the significance of the hydrologic cycle for water resource planning cannot be overstated, research and project experience, how- ever, show that conventional approaches to watershed management have little effect. Often neglected in analyses of watershed management are political, economic and social linkages between upstream and downstream. Understanding - 11 - of and intervention in these three areas provide an underexploited avenue to improve productivity and the quality of life of upland populations. Physical Linkages The need for watershed management arises from the interconnected nature of soil, water and land-use systems. Explicitly recognized is the fact that upstream land use generates not only direct outputs such as timber, crops and fodder, but downstream impact on sedimentation and water yields. In the absence of unified management, or some comparable arrangement, upstream users will adopt practices without regard for impacts on downstream residents. An additional consequence is that the party that undertakes watershed management will not be able to charge for all the services it provides. Charges are feasible for certain products, such as timber, but experience with cost recov- ery through charges for reductions in sedimentation and flood control are not encouraging. Types of Physical Linkages. The physical connections between upland and lowlands are manifested in two specific ways: (a) sedimentation, the delivery of eroded material into or adjacent to waterways and infrastructure; and (b) stream flows, the quantity, distribution and timing of flows from upper catchment to lower channels. The two are intimately connected by the fact that a major source of sediment movement is provided by raindrops as they strike and flow across the soil surface. In addition, the ability of soil to permit rainfall infiltration and (at least temporarily) retain moisture tends to be associated with its ability to withstand detachment and transport. Another major source of sediment when it occurs mostly in young active moun- tain systems, subject to high rainfall, is mass wasting. The initial movement of soil particles is termed detachment and ulti- mately all eroded material is deposited downhill and then downstream. However, the processes of delivery are highly dependent on the local environment, including catchment size and topography, levels and intensity of rainfall, slope, soil type, vegetation and land use. The time that may elapse between the initial detachment of the soil and its final flushing out from the system is frequently measured in decades for watersheds larger than 100 km2. Flows of water are much faster but no less complex. The movement of water from its original landfall, to minor and major channels and ultimately to the sea, is influenced by antecedent moisture conditions, the inherent infiltration and storage capacity of the soil, vegetation and land use. Typi- cally, in a mild rain and during the initial phases of more intense storms, water that reaches the ground through any vegetative canopy first enters into storage in the upper layers of the soil. A portion of this stored water will evaporate, some will be taken up by plants and transpired, and the remainder will percolate to the groundwater from which some portion will be returned as surface water. As the storage capacity of the soil diminishes, water accumu- lates on the land surface and moves downslope through various processes of inter- or overland flow. Depending on the length and intensity of the storm, topography and other factors, this inter- or overland flow may continue to a watercourse or may end in percolation into the downhill soil. If a storm of sufficient duration occurs and soils begin to be saturated or a storm of suf- ficient intensity occurs such that the soil's infiltration capacity is - 12 - exceeded, then several processes (surface and subsurface) deliver water to the drainage lines and streams. Vegetation affects these processes by storing water on leaf and stem surfaces, increasing surface roughness (thus slowing runoff and affording greater opportunities for infiltration), adding to the storage capacity of the soil by the presence of roots, reducing the initial intensity of raindrop impact, and to a minor extent, by immediately absorbing moisture through the root system. The two important aspects of stream flow are quantity and distribu- tion over time. Quantity of flow has obvious implications for the viability of downstream investments in power, irrigation and municipal water. Alterna- tive land uses can have significant impact on water yields principally through the substitution of more or less moisture-using vegetation. Hamilton and King (1983), reviewing the literature on the impact of forests on stream flow, found that forests are heavy water users and that conversion of forested watersheds to agricultural or other uses tends to increase total water yield, but increase peak flow and stream flow for any storm event. The distribution of flow throughout the year also has obvious impli- cations for downstream investments and stream bank erosion, but peak flows are more importantly related to the danger of flooding. There is some uncertainty as to the effects of environmental conditions in upper watersheds on the fre- quency and severity of flooding, because the effects diminish as distance down the watershed increases. Disentangling the impact of upland land use from other factors in determining the frequency and severity of downstream flooding has proven an elusive task. Both the quality and quantity of historical data on land-use changes and flood occurrence are so poor as to make statistical analysis impossible. The large number of variables involved similarly makes theoreti- cal analysis speculative and somewhat hazardous. A consensus among policy analysts is now starting to emerge, however, that suggests that agriculture and forestry in upper watersheds play a relatively minor role in exacerbating the effects of major catastrophic flood events (Hamilton, 1987; Ives and Messerli, 1989; Rogers et al., 1989). Data on annual runoff, sediment load and high and low flows for the Brahmaputra river system for the period 1955-79 show no definitive trend towards a deterioration in environmental quality. Moreover, the data on high and low flows are not consistent with the generally accepted expectation of land degradation causing higher peak flows and lower dry season flows (for details see Ives and Messerli, 1989, pp. 136-7). Data on the incidence and physical severity of floods in the Ganges-Brahmaputra delta do not support the hypothesis of a trend toward worsening floods. It is likely, however, that concern over flood damages is growing as a function of greater economic activ- ity in flood-prone areas.3/ 3/ See, for example, Kumra and Rao (1985) who found that the value of agricultural flood damage has been falling in Bangladesh while nonagricultural damage has been increasing. - 13 - Erosion-Stream Flow Interaction. As noted, the tendency of some soils to resist erosion is often associated with their infiltration and stor- age capacities. In addition, land management practices to reduce erosion frequently have the effect of reducing runoff (see Chapter 2). Another way in which erosion and stream flow may interact to exacerbate the danger of flood- ing is via the accretion of river bottoms. Diminished channel capacity due to sedimentation lowers the magnitude of flow required to cause flooding and may contribute to the frequency and severity of flood damage. Watershed Management Investments. In practice, most watershed man- agement projects have multiple objectives. In many circumstances, it is pos- sible to improve the environment and increase the output of goods and services at the same time, particularly when a project also averts long-term economic losses due to productivity declines. This aspect of trade-offs in watershed management is frequently misunderstood. The fact that watershed management projects can accomplish more than one objective has important policy implications. Generally, tax or subsidy schemes are recommended to resolve an externality problem in order to "inter- nalize" it. Farmers might be given subsidies to adopt soil conservation prac- tices or be taxed if they fail to adopt. Subsidies are the more frequently used approach and their budgetary cost is justified on the basis of reduced downstream damages. However, when watershed management practices make both upstream and downstream residents better off and leave no one worse off, this simple justification no longer holds. Some subsidies may, however, be justi- fied on other grounds, such as an inability on the part of poor farmers to finance purchased inputs or to wait for the maturation of tree crops (that is, capital market imperfections).4/ In order to be both multipurpose and to leave no one worse off, watershed management projects must be able to address a characteristic of unimproved watersheds, namely, technical inefficiency, or to introduce tech- nological change. Typically, projects do both. Economic Linkages While physical linkages remain the basis for watershed management interventions, a strategy that also takes advantage of social, economic and institutional linkages between upstream and downstream provides the greatest opportunity for success. Upland areas have critical connections with national economies in three significant ways. Sources of Raw Materials. Despite difficult conditions, upland areas often possess a comparative advantage in the production of certain commodi- ties. In much of the Asia region, timber and grazing represent the primary resources with potential in upland areas. In Indonesia (Roche, 1987), upland 4/ It is conceivable that situations will arise where there are real and significant conflicts between alternative uses. This situation was first analyzed by Gregory (1955) and (1957). Such conflict and the tax-cum-subsidy schemes it suggests, are more relevant to developed country, low-population density watersheds. - 14 - areas may have advantages in specialty crops such as clove, high-value horti- cultural crops and animal products. Significantly, many agricultural products for which upland areas are well suited possess a high income-elasticity of demand, so that income growth in lower areas can contribute to strengthening economic linkages between uplands and lowlands. In so doing, the strengthen- ing provides opportunities for expanding sustainable agriculture in the uplands. Sources and Sinks for Labor. Upland areas in the region were histo- rically, with some exceptions, notably Nepal's Terai, relatively sparsely populated. Recent increases in population pressure in more favored downstream environments has resulted in increased migration to the uplands. This has been offset to some extent by growth in nonfarm employment opportunities in urban areas. Seasonal employment opportunities in lowland agriculture and urban areas are increasingly important contributors to upland income. Shifts in the opportunity cost of labor brought about by changing opportunities off the farm also have important implications for farmer interest in adopting more intensive soil conservation measures. On the one hand, they may be more likely to afford the costs but, on the other, have less time. Markets for Downstream Production. Because of low incomes and high transport costs, upland areas have generally not been major markets for goods produced in lowland areas. Upland areas, however, where incomes have grown and infrastructural investments have reduced transport costs, do constitute significant markets. Political Linkages Upper watersheds, in addition to being physically remote, are often politically remote as well. The attention of national policymakers is natur- ally drawn to the concerns of urban and more affluent lowland agricultural populations. To the extent that developments in upper watersheds are a major item on the national agenda, it is because of their impact, via the physical linkages related to movement of sediment and water, on the well-being of down- stream groups. Political forces further bias policy and investment against upland areas in other ways. Watersheds are physical units that frequently do not conveniently overlap with administrative boundaries. Although the down- ward flow of the physical consequences of land use do not observe these boun- daries, limits on the ability of government agencies to transcend them are severe. Asymmetry and Rigidities of Linkages. From the perspective of investment analysis and policy-making, the most important aspect of the vari- ous linkages described in this chapter is whether or not they can be manipu- lated. In this respect, certain of these linkages are either rigid and hence not amenable to manipulation or they are asymmetrical. The most obvious asymmetry is the one-way flow of soil and moisture in watersheds. This implies a role for involvement by public agencies. Temporal Asymmetry. As noted, the movement of sediment in watersheds may involve extended periods of time. A corollary to this is that remediation may also involve long time delays after the intervention. Pearce (1986) esti- mates that the time required for sediment to be flushed from drainage systems - 15 - of large basins may exceed 40 years. In a present value sense, any reductions in sediment made possible by the adoption of watershed management practices will be of minimal economic significance. However, there are downstream bene- fits to be achieved within shorter time periods from conservation land treat- ments, like contour hedge rows, as the focus moves upstream to subwatersheds. These benefits relate to both sediment and flooding in nonmajor storm events (Hamilton, 1988). Geologic Erosion. While in principle it is possible to stop erosion, the underlying potential of soil to move is frequently so great as to preclude stopping erosion at any reasonable economic cost. This underlying rate of soil loss is termed geologic or natural erosion and provides a partial but useful guide to the design of soil conservation policies. Much of the steep uplands of the Asia region are naturally prone to erosion due to their geol- ogy. Asymmetry of Policy Linkages. Agricultural policy linkages, particu- larly price and incentive policies, between uplands and lowlands have not received the attention they deserve. Theoretical analysis of farmers' incen- tives for soil conservation, however, have been largely,unable to demonstrate how these linkages operate. Barbier (1988), for example, using an optimal control framework, modeled farmers' decisions to implement soil conservation practices and showed that price of products and other agriculturall policies could play a significant role in determining privately profitable soil manage- ment strategies. Barrett (1988), using a similar approach, found product price to be unrelated to soil management.5/ Taking a different approach, Roche (1987) analyzed the impact of these policies and growth patterns on upland land use in Indonesia. He notes that rapid lowland income growth, due in part to the successful intensifica- tion of irrigated rice production and industrialization, influenced by high income-elasticities of demand for vegetables and fruits, has created an incen- tive for upland farmers to shift to cropping patterns that are less likely to cause erosion. Observations further indicate that in upland areas with access to good markets, and particularly where the demand in these markets for meat and livestock products is strong, the incentive for establishing and maintain- ing permanent vegetation is also strong. The aggregate environmental impact of these incentives has never been assessed. Hyde (1988) explored the consequences of unemployment in lowland areas of the Philippines on the environment of the uplands through a general equilibrium model. The model, which allowed migration as an equilibrating process, showed that tax policies which in the aggregate favor capital have a significant influence on migration to upland areas. Similarly, rice subsidies were found to increase both the agricultural labor force and the upland popu- lation. Trade policies that would encourage exports were found to have a positive impact on the uplands via an expansion of lowland industrial employ- ment. 5/ Similar models have been presented by McConnell (1983), and Bhide, Pope and Heady (1982). - 16 - INVESTING TO INFLUENCE LINKAGES AND SOLVE PROBLEMS Altogether, the data reviewed on watershed linkages suggest that although physical connections shape the environment for investment planning and policy interventions, there are marked rigidities in both time and space. These rigidities limit the scope for economically viable investments aimed primarily at resolving off-site and downstream problems. Fortunately, there is ample opportunity for directly productive investment in upland areas. Most of the approaches that would fit into such a strategy are also consistent with the long-term objective of preventing or ameliorating the downstream conse- quences of watershed deterioration. Essential elements of a strategy for upland development are the same as would apply in lowland areas and include the need for a positive incentive framework and the availability of appropriate technical innovations. In con- trast to lowlands, upland areas are characterized by much greater agro-ecolo- gical diversity, are less amenable to large-scale investments (especially irrigation), and generally face runoff and soil erosion problems. Accord- ingly, strategies for upland areas require greater emphasis on generating a capacity for site-specific recommendations, and particularly a focus on improving rainfed agriculture through low-cost methods of soil and moisture conservation. Due to the greater reliance of upland farm households on non- arable land such as forest and communal grazing land, farm development strate- gies in the uplands also need to focus more on diversification than farming systems in the lowlands which can be more commodity-oriented. Technologies and Techniques for Improving Upland Agriculture Although the development of agricultural technologies for upland areas lags far behind those for the lowlands, the general principles for increasing yields are known and numerous interventions can be recommended for specific applications. The two key constraints to improving agriculture in upland areas relate to soil and moisture conservation. In practice, productivity decline due to soil erosion is related to the following soil characteristics (discussed in more detail in Chapter 2): rooting depth, water reserves available to the plant, distribution of plant nutrients in the soil profile and the chemical/physical properties of the subsoil. Of primary importance for design of an upland strategy is the con- nection between soil moisture and erosion. In tropical soils, water-use effi- ciency (kg dry matter produced/liter of water use) can be cut more than half, despite high rates of fertilizer application, if topsoil is progressively removed up to 35 cm. General Approaches to Enhancing Upland Agriculture Better agronomic techniques, improved varieties, higher-quality seeds and improved pest management and tillage practices often provide the best opportunities for increasing agricultural output. A key technique that is integral to improvements in rainfed agriculture is contour cultivation. Com- pared with the traditional up-and-down slope cultivation, contour cultivation - 17 - and ridging across the slope 6/ have produced 6-662 yield increases on 3-32% slopes, with further increases if combined with other treatments such as mulching. The evidence to support the general recommendation that all rainfed cropping activities, annual or perennial, be carried out on the contour is overwhelming. Recent and less frequently the subject of published research is the application of cropping/farming system technology to on-farm soil conserva- tion. It is motivated by the acknowledgement that social, economic and tenur- ial factors influence farmers' ability and willingness to adopt and maintain soil conservation measures. Farmers, especially poor smallholders, need direct short-term benefits from any innovation in their farming systems. Investments in soil-conservation measures apparently have not met this criter- ion and, indeed, frequently have been perceived as taking away from the farm- er's limited resource base by demanding space for banks and water disposal structures. By contrast, downstream farmers in irrigated areas respond quickly to investments in similar structures that have immediate benefit, as is the case with levees for rice paddies. Farming systems research has not commonly been applied to soil con- servation. Yet attempts to implement physical conservation works and land-use planning are usually frustrated by lack of acceptability. It becomes impor- tant, therefore, to find points within farming operations where practices that meet soil conservation objectives and increase incomes without unduly increas- ing risks can be introduced. Specific Techniques While there is clearly a need to design a package of conservation and yield-increasing interventions to be consistent with the needs of a specific site, there are several generic approaches which have widespread usefulness as well as potential for misapplication. They can be grouped as structural and vegetative/cultural. Current conservation practices in the Asia region focus on structural approaches and there is a need to assess the potential for fuller utilization of alternatives. While Chapter 2 provides a fuller discus- sion of these approaches, some aspects are relevant to a strategy for upland development. Structural Treatments. Structural treatments, earth banks, land leveling, and terracing have been applied extensively in watershed projects throughout the region. Experimental and project-level results with structural measures have been mixed but generally poor. These observations may seem inconsistent with the fact that terraces, in particular, are a widespread and integral part of the agricultural landscaping of the region (see Box 1.1). However, several features of structural approaches may account for their gen- erally poor performance. 6/ Contour cultivation refers to cultural treatments that follow surveyed guidelines linking points of the same elevation marked at 2-3 m vertical intervals, whereas across-the-sloPe refers to treatments at right angles to the general slope direction and, hence, deviate at times from the true contour. - 18 - Box 1.1: ROLE OF BENCH TERRACES IN ASIA Bench terraces are part of the landscape of the Asia region, especially Southeast Asian countries, China and the Philippines. Reverse- sloped bench terraces and outward-sloped bench terraces are used in steep uplands of humid and semi-arid regions, respectively, to change the slope of the land in order to increase the area that can be cultivated "safely' and "control" runoff. Level bench terraces (irrigation-type) are used for rice paddy and conservation bench terraces are used in arid regions to harvest rainfall on part of the slope and direct it to a level bench. While it may be questionable whether bench terraces represent the best treatment for the respective locations in terms of land-use capabil- ity or meeting the needs of the population, they are in place over vast areas and future land development programs should start from this reality. A review of research on the effectiveness of terraces of the first two types in uplands regarding sediment yield, runoff and productiv- ity (Chapter 2) highlights widely divergent results. Although terrace technology is well understood and engineering design readily available, it is frequently poorly applied. There are three common shortcomings: fail- ure to relate soil type-rainfall characteristics-cropping pattern to design; not viewing the water disposal component of terrace systems as integral--farmers are reluctant to lose the 3-5Z land required--and com- promise in design increases runoff and damage; shoddy operation and main- tenance by farmers compared with level benches used for paddy, which implies a questionable economic situation. Developing a program for correcting these shortcomings should start by defining the treatment options. Carson (1989) discussed the limitations of terraces in several agro-ecological zones in Indonesia and suggests some appropriate soil conservation strategies. The key issue is identifying a farm production system attractive enough to induce the majority of occupants to become involved and finding a way for them to convert to it. After this, the treatment options can be laid out, with priority going to vegetative-cultural measures which should be cheaper to implement and maintain. For example, if the horizontal grade of the existing terrace exceeds 1Z, judicious use of a vegetative barrier such as vetiver grass would induce natural and rapid leveling and at the same time control the effluent point, which in turn would change the dimension of waterway rehabilitation. Only then would structural treatments be consid- ered for problems still without solution. The upper limit of slope is normally 60% for any sort of terrace. Beyond this slope, riser height and width are too great, bench width too narrow, and the net arable area down to about 50X. - 19 - (a) High Unit Costs. Costs for terracing in Indonxesia are estimated to range from US$400-1,000/ha. Construction of earth bunds in India is estimated to cost between US$23 and US$150ha depending on soil type and slope. Aside from high initial costs and the financing burden they impose, structural techniques inevitably require high levels of maintenance. Failure to maintain structures properly can lead to their total failure and can actually accelerate soil loss. (b) Inappropriate Design. With structural measures there has been wide- spread failure to adjust designs and standards to accommodate the engineering properties of particular soils and local rainfall pat- terns. The inherent instability of some soils can result in massive failure of structures. For example, saturation of the topsoil over relatively impervious subsoil results in soil slumping. (c) Inadequate Drainage. Operating on the principle of slowing water, with structural measures water is usually directed along field boun- daries toward natural drainage ways. Drainage ways need to be large enough to accommodate peak flows, otherwise they will be overtopped, damaging the adjacent field or the drains themselves will fail. Moreover, the natural drains may receive more runoff than they are capable of safely handling, resulting in a danger of gully erosion. The planners' incentive to design drains to accommodate peak flows runs counter to the farmers' desire to minimize land taken out of production. This conflict usually results in no drains or undersized drains prone to failure. (d) Exposing Subsoil. Construction of soil conservation structures usu- ally entails earth movement that exposes infertile subsoil. This reduces yields in early years of the structure and amounts to an additional construction cost. Research has added very little to traditional farmers' understanding of the potential use of structural measures. Detailed analysis of terrace designs and maintenance in Nepal, for example, has shown that use of outward- sloping terraces is an effective means of allowing surplus water to move off the terrace while causing minimal surface erosion. Attempts to reduce 'off by introducing backsloping terraces resulted in collapse of the improved terrace due to concentration of water. Similarly, the apparently low levels of maintenance and poor-quality construction of terraces supplied by projects in Indonesia may reflect the interaction of farmer perception of the dubibus value of terracing and the attractiveness of the assorted subsidies andtincen- tives provided. In the absence of compelling evidence that a significantly new and attractive on-farm structural technology can be suggested to farmers, there seems limited justification for the central role such struciures now play in watershed development projects. Vegetative/Cultural. Vegetative/cultural measures to improve upland agriculture include contour cultivation, techniques to reduce tillage, addi- tion of new crops and changes in timing or cropping pattern (intercropping, etc.) or stand architecture to provide for more continuous and effective soil cover. In some cases the use of vegetative treatments is intimately mixed with cultural practices, such as contour cultivation with grass strips, while - 20 - in other cases vegetative measures stand alone as in the establishment of permanent cover. Vegetative measures have been shown to be highly effective in minimizing erosion by reducing the impact of raindrops as they strike the soil. Mulches, certain agroforestry options and permanent cover crops can perform this function. Plants can also be used to form a physical barrier to slow runoff and arrest already moving soil. For a long time, suitable species have been sought and several have been proposed for use in this manner, including napier grass, vetiver grass, and the tree species Leucaena. The utility of different species in this capacity will vary, depending on circumstances. The particu- lar features of vetiver grass, discussed in detail in Box 1.2, make it partic- ularly well suited for this application. Box 1.2: VETIVER GRASS - CONTOUR SYSTEM FOR SOIL AND MOISTURE CONSERVATION The notion of carrying out all farming operations especially cultivation and plant- Ing on the contour In any rainfed situation, on any slope, for any crop is overwhelming. Customarily, barriers are constructed on the contour at certain vertical intervals according to slope to break the length of the slope so as to check the velocity of runoff water and trap slit. These also serve as guidelines for contour cultivation. Graded earth banks (bunds) usually with a horizontal gradient of up to 1X to feed excess water into a prepared waterway have been employed extensively. These structures have severe limitation in the tropics (Chap- ter 2) and do not fit small holdings due to loss of arable area for the bank itself and the waterway. Vegetative barriers on the contour have distinct advantages over earth banks, namely, vegotative barrier requires about one tenth of the space and no water disposal system is necessary, vegetative barrier slows down surface runoff and causes it to deposit the silt load while the water seeps through spread out, with increased opportunity to infiltrate. This also avoids the problem of waterlogging which is common behind the bank. A plant suitable for a vegetative barrier requires particular morphological charac- terlstics. Its root system should be aggressive and deep without rhizomes or stolons so as not to spread out of line; the crown should be below the surface for protection against fire and overgrazing; the culms tough and unattractive to animals and pests; and the flowers, If any, essentially sterile so as not to permit spreading by seed. The plant should be a peren- niSl and persistent, tiller freely and intermingle with its neighbors (so" clump grasses do not). To date, the only plant know to meet these criteria is votiver grass, (Votiveria zizanloides). It has an extremely wide range of climatic conditions over which it is adapted and urther exhibits adequato growth over a wide range of soil types, including those with highly unfavorable properties for many plants. Votiver grass has been used for this purpose and as permanent field boundaries for a long time, and hence It is known to persist, once establIshed, without maintenance indefinitely. It is propagated by root *lips which the former may plant himself on a roughly surveyed contour lines. Given moderately favorable conditions, the hedge would be complOte after throe growing seasons, fewer with high fertil- ity, high rainfall and close planting. Apart from physical advantages, establishing and maintaining the system is low-cost and can be carried out entirely by the former. Compare this with the engineered system which requiros oarth-moving equipment, complete cooperation among neighbors especially for water disposal components that may impose an Intolerable burdon on those furthor down the slope, and regular rebuilding every throe to five years. Votivor grass has other applications due to Its unique morphology. Among these ore protecting paddy banks, dam catchments and drainage lines from siltation, roadsides and stream banks from erosion, and performing the soil- and moisture-conservation function when planted In V-ditches with fruit and forest trees. The same technology can be used for the stabilization of degraded nonarable lands. Vetiver grass can be used, but in these nonarable situations, shrubs that can be coppicod for fuel or fodder could also be used as barriers on the contour. The search for suitable shrubs continues. - 21 - Vegetative systems, of whatever species, have a number of advantages over structural systems: (a) Cost. Vegetative measures for soil conservation generally can be promoted at low cost. Costs for establishing vetiver grass hedgerows in India are estimated to be US$18/ha. In many cases the major cost item for promoting these measures is for extension advice. (b) Adaptability. Unlike structural measures which require detailed engineering and site planning, vegetative approaches are relatively insensitive to issues such as proper alignment on contours, irregu- larities in field boundaries and minor errors in placement. Hence, surveyed contour guidelines can be replaced by planting across the slope. (c) Farmer-Controlled. Because vegetative methods are relatively inex- pensive and do not require use of machinery or sophisticated survey- ing, individual farmers can take the initiative in adopting conserva- tion measures. An indigenous system of contour alley cropping using bands of Leucaena has been used widely in the steep lands of Cebu in the Philippines. A particular advantage is that the cropping area sacrificed to the conservation measure is considerably less than with the typical structural approach, and is especially true in the case of grass contour hedgerows. Farmers' willingness to devote arable land to essentially permanent cover is often largely dependent on the degree to which livestock are integrated in the farming system. Investing in Nonarable Areas A large proportion of a typical watershed anywhere in the region is nonarable in the sense of not being suitable for agriculture due to soil or slope characteristics. Yet the consequences of runoff/erosion on nonarable lands are quite significant. Productivity is lost, and more so than on arable lands, sedimentation and local flash flooding are increased, and dry-season stream flows reduced. Land use and ownership are generally less complicated in arable areas, where crop-based agriculture and mostly private ownership prevail, than in the nonarable areas divided among forest, grazing and community lands and variously owned by government (mainly forest department), communally and pri- vately. The condition of nonarable land is further complicated by de jure and de facto rights of access by both landed and landless rural families. This diversity of use\and ownership has exacerbated the effect of degradation and makes remediation more difficult on nonarable land. In order to redevelop nonarable areas, something has to be done first to restore soil moisture status. Contour vegetative hedge treatments improve infiltration. Lowering livestock populations would reduce soil compaction, as would less use of heavy equipment in forest harvesting. Controlling fire, which induces water repellancy in some soils, can also assist. Redefininig land use becomes the important next step. While it may be the most advantageous use of land and the best soil conservation strategy to - 22 - try to return bare land or degraded forest areas to the original mix of spe- cies, other options could include closed mixed-species forest, single-species plantations, fuelwood plantations, silvipastoral plantations or pasture. It is not within the scope of this study to develop guidelines for determining which option might be employed under any given situation. At the same time, the importance of doing so ought not to be underestimated since it takes account of the needs of the population and may ultimately determine the suc- cess of the investment. Treatment of Forest Areas Stabilization. Denuded slopes in nonarable areas need to be stabil- ized by using vegetative barriers on the contour at approximately three to four meter vertical intervals first, before any of the options are applied. This treatment cuts down runoff, increases infiltration and traps erosion products. This stabilization technology can be applied on village common land, grazing land and wasteland, as well as forest land. The vegetative barrier should ideally comprise indigenous, locally adapted shrubs that are unpalatable, deep-rooting, easily propagated and capable of forming a dense hedge, planted into a V-ditch or trench. In the absence of suitable shrubs, vetiver grass would form a suitable hedge in most circumstances in the region. As noted above, the cost of of establishing this treatment is likely to be on the order of US$18 per hectare. It would be applied regardless of what the interhedge spaces might later be used for. Revegetation. Artificial forestation is the option most commonly applied to nonarable land, probably because most areas are under the jurisdic- tion of forest departments whose mandate is to plant trees and manage planta- tions. The success rate for forestation is low and costs per hectare high, which call the technology into question. Questioning the technology is valid, but only when the issues of stabilization/soil moisture status and land use are resolved. Chapter 5 reviews the methods of revegetation presently prac- ticed in Asia. There are key shortcomings regarding selection of species and quality of planting materials, land preparation, methods of planting and planting geometry, protection and management. A serious nontechnical short- coming has been that forestation has been carried out without the support or agreement of local people who may customarily harvest some resources from these areas. The technical shortcomings are well understood and little or no additional research is required to be able to grow most tree species success- fully. Addressing all these issues so as to do everything well, however, results in costs per hectare in the range of US$500-1,OOO. This cost is gener- ally too high to be replicable over wide areas. The Problem-Solving Approach Beyond doubt, removing the vegetative cover causes accelerated ero- sion so, if over time an undisturbed vegetative cover can be recreated, the problem is solved. However, (a) time may not be an option, in which case intervention is required to accelerate growth of vegetative cover; - 23 - (b) natural regeneration may not be an option, if the soil moisture and nutrient status have been changed due to degradation; and (c) an undisturbed vegetative cover may not be an option, if land is to be used for arable agriculture, grazing, fuelvood or fodder produc- tion, in which case a vegetative/cultural farming system, or struc- tural intervention or combination of interventions is required. The issue is how to decide, within the choices, which to take and how to avoid the common mistake of opting for a single solution to a complex prob- lem. Information specific to the site and a clear understanding of impact are required. There are two main questions. The first is one of scale: what should be the size and definition of the planning unit? The larger the area of the planning unit the greater the heterogeneity of land use, land capabil- ity, microclimate, soils, vegetation and people. The larger the size the more likely that one or two widely applied solutions will fail. The second ques- tion relates to what must be achieved. If, for example, the objective is solely to reduce downstream sedimentation, then it might be achieved by a sim- ple technique such as a checkdam, but the dam would have no effect on erosion- induced productivity decline in arable areas. The objectives are rarely simple and hence invariably require a set of solutions. Folly of a Single Solution. Terracing as a treatment is a common choice in the tropics and done properly can be effective. Yet terraces by themselves do not necessarily conserve soil or moisture, improve productivity or decrease sedimentation. In fact, poorly farmed terraces may result in greater degradation, and inappropriate types of terraces or terracing inappro- priate soil types may reduce productivity and accelerate soil loss. Even when properly used, terrace technology is only part of a system. Whether or not terracing is a sound proposal depends on the type of terrace in relation to rainfall, soil depth, drainage and structural strength, land ownership pat- terns, crops to be grown on the terrace, farming systems and quality of the extension services. This is true of all soil conservation interventions. There is no one technology that applied in isolation will achieve a soil con- servation benefit on anything but a very small scale. Importance of Planning at a Micro Level. For selecting appropriate on-site soil con-servation practices, it is possible to construct guidelines that can be easily followed at the local level. At the regional planning level it is more difflcult-due to increased heterogeneity and the limited quality of information available. Information on where and under what circum- stance various soil conservation technologies are applicable is available; whether this information can be used depends on the quality of the information from within the planning area. Given the importance of scale and linkages ascribed to economic functions, ascertaining the nature and optimum dimension of the planning unit is critical. There are advantages and disadvantages in using a hydrological (physical) unit, an administrative (political) unit or a set of villages (social unit) and a case can be made for each. Since villages tend to be located close to drainage lines in order for occupants to exploit lower arable and higher nonarable lands, the boundaries of a village's area of influence frequently coincide roughly with watershed boundaries. As a general rule, the hydrological unit is preferred. A typical watershed (100,000- 200,000 ha) comprises a series of subwatersheds (5,000-15,000 ha) which in - 24 - turn are made up of five to ten microwatersheds (500-2,500 ha). Experience seems to suggest that the subwatershed, as it is defined here, is a convenient planning unit, provided that plans respond to an aggregate of information from constituent microwatersheds. Having avoided the single solution trap, the preferred approach in developing a conservation strategy is to define the problems that are evident in the subwatershed and to define the objectives of possible solutions. Since there is no immediate linkage in very large watersheds (above 200,000 ha) between erosion in the upper catchment and downstream sedimentation, the most supportable programs are those whose objectives are to raise farm incomes and increase on-site sustainability of both arable and nonarable land in the upper watershed. Techniques such as rapid rural appraisal (see Box 1.3) can be used in problem definition through interactive planning with the population. In a short time, it would be possible to identify the population's dependencies, needs and aspirations, perceptions of the dimension and causes of degradation, to describe the microwatershed itself in terms of land classes (arable, non- arable, private, village, public) and identify respective needs for treatment, and by this method to come up with objectives, a strategy and action plans. Menu of Solutions. Within any subwatershed there will be a number of treatments appropriate to addressing the problems defined in the interactive planning process. A list of eligible treatments can be assembled easily, given what is known about the efficacy of each in particular agro-ecological situations, their synergism and cost. It remains then to match solutions to the problems, an exercise that has several dimensions. For private arable land, farmers will undoubtedly choose treatments that are income-enhancing in the short term. Although support may be required, little coercion would be needed. Treatments such as stabilizing or revegetating nonarable areas or treating drainage lines would require the population's consent and coopera- tion, and would mostly be implemented by an agency or organization rather than individual farmers. Incentives for Participation in Watershed Development Three well recognized factors influence farmers' willingness to par- ticipate in watershed development programs, and more specifically to implement soil conservation treatments: (a) land tenure; (b) profitability of the farming system and scope to improve profitabil- ity; and (c) economic status, whether a cash or subsistence farming system, and portion of income derived from the farm. - 25 - Box 1.3: PROBLEM DEFINITION IN A MICROWATERSHED Interactive Planning Numerous approaches have been developed around the world for carrying out interactive village planning in microwatersheds. These plan- ning approaches have as their objective to put planners, agency staff, and villagers on a common ground for identifying key problems, analyzing their causes, and devising realistic action plans that reflect local needs and the availability of government and local resources. Successful approaches are those which include techniques for collecting and discussing informa- tion in an open-ended way, which draw strongly upon indigenous technical knowledge as well as professional expertise, and which are conducted in stages to allow villagers to participate in devising action plans, rather than simply reacting to plans drawn up by government extension agents or officials. The Technique Rapid rural appraisal is a technique that is often employed in interactive planning. It is not a methodology, but a set of investigative tools adapted to short-term analysis of particular sets of problems of natural resource management. It is often used to gain an initial under- standing of problems on the basis of the analysis of secondary data com- bined with a structured field investigation. In combination with other formal surveys, it can be used in monitoring program performance and eval- uating program efficiency or program impact. Unlike traditional research, rapid rural appraisal teams include planners as well as researchers, and their investigative tools are designed to encourage as much interaction with villagers as possible. These tools includet (a) group and individ- ual interviewing; (b) cross-checking information (triangulation); (c) direct observation; (d) use of sketch maps, diagrams, village tran- sects; (e) sampling tailored to a shortened time frame; and (f) redesign of plan as hypotheses change and new options emerge. A Sample Application to Watershed Development A watershed development program which includes soil and moisture conservation, forestry, on-farm tree planting, and pasture improvement is being implemented by several government agencies in a subwatershed. A team of one or two persons trained in the technique and government exten- sion agents would visit villages to analyze people's needs and conduct individual interviews with different types of households. The team reviews environmentalleconomic problems with the villagers, adding to villager statements with their own observations, and plotting the informa- tion with villagers on sketch maps showing village areas of influence. Conflicts over use of the resources within the village or between villages and over government regulations or uses are important topics, as are the institutional mechanisms for resolving these problems. Villagers discuss options that they feel will help to resolve their problems and with the team draw up an action plan, based on their own time and resources and the available government inputs, programs, and resources of the extension departments represented. - 26 - Land Tenure. Tenurial arrangements on arable land are complex. Insofar as they affect implementation of soil conservation measures, the issues go beyond guaranteed long-term access to land to include access to credit, ability to make decisions on land development, the proportion of returns that accrue to the user, and the ability to transfer rights. About eight categories of tenure, ranging from very secure for privately owned land with title to very insecure for some forms of sharecropping and for private cultivators on public lands, pertain. The attributes, categories and conse- quences for participation in development programs on arable land are discussed in Chapter 6. An important conclusion of that review is that tenurial arrangements do have an important influence on the land user's decision to participate or not to participate. On private land, low adoption rates have frequently been attributed to tenurial constraints whereas poor technology options are the real constraint. Two significant changes in recent Bank- assisted projects are: (a) greater use of vegetative, cultural and farming systems-related con- servation treatments that are, overall, more effective and more amen- able to a wider range of tenure categories, and (b) presentation of a menu of technical options from which the farmer may choose according to personal conditions rather than a single package that may be intimidating and therefore rejected. Profit, Sustainability and Risk. An on-farm production system that is more lucrative than the current one and involves minimal increase in risk provides incentives to farmers to participate. Evidence suggests that in all but the most marginal situations (shallow stony soils, steep slopes), total production can be improved by managing soil moisture and restoring soil fer- tility, which techniques generally mitigate runoff and soil erosion. Yield increases, for example, from contour cultivation alone which involves very little cost can be 50Z or more: the resulting improvement in soil moisture then allows modest levels of applied fertilizer to be effective. Improved plant density due to seeding rate and adjustments in row spacing, an improved fertilizer strategy or intercropping can also improve yield and soil protec- tion. Support for Participants. However attractive the incentives may be, most smallholders need support in several ways to implement new initiatives: (a) to lay out, establish and maintain hedges on-the contour (for exam- ple, with vetiver grass); (b) to reshape inappropriately designed terraces, plant surface risers with suitable fodder plants; (c) with good-quality planting materials of the most suitable cultures; (d) with demonstrations of improved farming systems and cultural treat- ments; and (e) with credit and extension. - 27 - KECONKENDED APPROACHES TO WATEERSED DEVELOPEENT The data reviewed make clear that there is no single watershed man- agement problem in the Asia region. Rather there is a complex of issues rela- ted to increasing soil and moisture loss, land degradation, sedimentation and irregular stream flows, and poverty that can best be understood in the frame- work of watersheds as physical planning units. The analysis suggests that the donor community should continue, and accelerate, its efforts on the develop- ment of environmentally sound and higher-productivity, upland farming systems. Greater attention should be given to the diversity of upland agriculture and the need to develop local capacity for diagnosis of constraints to productiv- ity growth and design of site-specific solutions. Forest lands and livestock grazing systems similarly need to be addressed, because few soil and water conservation measures are in place on them. Projects promoting single- solution, structural approaches to soil conservation problems should be reduced and greater effort given to the use of vegetative techniques. Tech- nology development projects, with provision for careful experimental design and rigorous testing of proposed techniques, may be required before large- scale projects can be expected to be viable. Flooding and Sediment The potential for reducing flood and sediment damage downstream in large basins by means of land-use changes, rehabilitation, and reforestation in the upper watersheds of the Asian region appears limited. Catastrophic floods seem to be the result of heavy rains largely falling on already satu- rated soils and nonabsorbant surfaces in lower reaches; and by nature, deltic regions, formed by flood-borne deposition of eroded material, are subject to flooding. No statistical evidence indicates a secular increase in the occur- rence of flood events, yet increases in flood damage are largely explained by increased population and higher-value land use in floodplains. Data on sedimentation support the argument that in large river basins whose headwaters lie within the geologically young, unstable mountain ranges of Asia (for example, Himalayas), the largest portion of sediment load is the result of natural processes, and the damages therefrom are to the same extent ascribable to geological processes. Human-induced sedimentation, while having severe impact within smaller watersheds, has much less impact within the con- text of larger river basins. Reducing geologic erosion is not practicable and reducing human-induced erosion in areas with already widespread disturbance by altering land use or reforesting, for example, will not necessarily have the impact required. This relationship is particularly true, given the prolonged residency of sediment within a catchment whereby sediments currently stored within channels and floodplains will continue to move through the system1for extended periods. Nonetheless, at some point, watershed rehabilitation must be undertaken to begin the process of reducing the transport and deposition of sediment, though for large basins the time lag for noticeable reduction down- stream may be decades or centuries. While the problems caused by sedimenta- tion are real and significant, they should be dealt with by such means as operating practices, dredging and appropriate infrastructure design. In par- ticular, decisionmakers should give greater attention to assessing the valid- ity of assumptions about sedimentation rates and predictions of investment life. In some cases, it may be necessary to either accept or reject water _ 28 - resource developments, depending on whether the expected lifetime in the face of sedimentation is sufficiently long. However, the likelihood 'of eventual sedimentation must be recognized. Similarly, recognition of the probability of flooding should be fac- tored into policy-making with respect to downstream areas. The design of infrastructure and buildings and the decisions to site developments in flood- plains and to promote settlement should recognize the certainty of an eventual flood. Cost Sharing and Cost Recovery A corollary of the limited impact of upstream land-use changes on downstream damages is that there is limited justification for schemes to com- pensate upland farmers and communities for adopting conservation practices. There is considerable scope for identifying techniques that will reduce or at least not increase erosion and runoff and that are profitable for upland farm- ers. Various subsidies and compensation schemes may be required to bridge the gap between adoption of a conservation measure and the realization of a sus- tainable net return. If so, such compensation should be seen as transitional and not as part of a policy of ongoing subsidy. h-Experience with subsidy schemes for adopting conservation measures has not been encouraging. Unless carefully designed, subsidies can lead to an overemphasis on construction of structural measures and neglect of maintenance requirements and serve as a disincentive to less-expensive measures that would otherwise be adopted by farmers on their own. An example of a promising -trategy has been used in the Central Visayas Regional Project in the Philippines. To promote adoption of contour hedgerows, the project lends breeding cattle to farmers, conditional on establishment and maintenance of a hedgerow sufficient to support stall feeding of the offspring. This provides a powerful incentive for both establishment and maintenance of the hedgerow and has been extremely effective. Rural Infrastructure Road and trail construction in upland areas can contribute to either environmental improvement or deterioration and is one of the few ways of sig- nificantly affecting downstream sedimentation. The extension of road networks can lead to deterioration of watersheds by facilitating access to fragile remote areas. On the other hand, improved'access to markets can improve incomes which can generally be expected to lead to adoption of more conservation-oriented farming. Road and trail design and construction methods need to incorporate adequate safeguards to minimize erosion and sedimentation. As attention is shifted from use of structural conservation measures to agro- nomic approaches, the engineering expertise of soil conservation agencies can be reallocated to road construction and rehabilitation. Analytic Methods There are no special characteristics of watershed development proj- ects vis-a-vis other development projects that require a fundamentally differ- ent approach for their economic analysis. Standard approaches to the analysis - 29 - of agricultural projects, based on a with-without comparison, will produce a satisfactory estimate of project worth. Special attention may be required, however, to understanding incentives as perceived by farmers and communities, given the importance of common property resources and the often precarious nature of land tenure systems. While this review concludes that the downstream impacts of land-use changes within large river basins will generally be of marginal importance over any time period of economic interest, there are no conceptual constraints on taking any such benefits into account. Meaningful judgments on the physi- cal impacts of a specific intervention can be developed through approaches such as sediment budgeting, as well as on the basis of long-term measurement and modeling studies. Well-established approaches can then be used to trans- late physical impacts into economic terms. More relevant to most watershed development projects is the need to integrate technical judgments on the impact of erosion and of conservation practices on crop yields with economic and financial analysis of cropping systems. Although data are seldom available for a particular project site, there is usually sufficient evidence from other sites to guide economic analy- sis. Input from experienced agricultural specialists is needed to ensure the validity of assumptions made in these calculations. Input from social scien- tists is also important in order to assess constraints to adopti'on, especially on common property and public land. Guidelines There appears to be no need to prepare technical guidelines for watershed development projects. Annex 1.1 lists guidelines issued by several government and international agencies on various aspects of watershed develop- ment projects. Funding Procedures New approaches to disbursing project funds are required, given that: (a) overall project costs can be estimated by extrapolating from detailed plans developed at the microwatershed level; (b) a number of different govern- ment and quasi-governmental agencies and nongovernmental organizations can legitimately be involved in implementing watershed actions; and (c) not all activities need to be synchronized. For example, it is reasonable to envisage a program-type watershed development fund from which approved agencies may be reimbursed for completing treatments eligible for reimbursement.''This requires definition of the eligible treatments, definition of the organiza-' tions who may implement them and verification procedures relating to implemen- tation. The Bank and other development agencies would be advised to explore more effective funding procedures. Need for Commitment Watershed management projects are complex interventions that require effective multidisciplinary collaboration, commitment by governments and local communities, and sustained efforts. For development agencies to be effective - 30 - partners in this process, it is necessary to recognize that watershed proj- ects, while not necessarily large or expensive, require heavy inputs of staff, particularly in preparation and supervision. Agencies also need to recognize and act on the need for government commitment in resolving watershed problems. Without serious commitment by governments and their field staff, investments are unlikely to succeed. - 31 - ANNEX 1.1 Guidelines for Watershed Management Dani, Anis, and J. Gabriel Campbell. 1986. Sustaining Upland Resources: People's Participation in Watershed Management, ICIMOD Occasional Paper No. 3, International Centre for Integrated Mountain Development and FAO: Kathmandu. FAO. 1986. Strategies, Approaches and Systems in Integrated Watershed Man- agement, FAO Conservation Guide 14, FAO: Rome. Gregersen, H.M., K.N. Brooks, J.A. Dixon, L.S. Hamilton. 1987. Guidelines for Economic Appraisal of Watershed Management Projects, FAO Conservation Guide 16 (East-West Center, SIDA, FAO) Rome. Hamilton, L.S. and P.N. King. 1983. Tropical Forested Watersheds: Hydrolo- gic and Soils Responses to Major Uses or Conversions, Westview: Boulder. Hufschmidt, M.M., D.E. James, A.D. Meister, B.T. Bower and J.A. Dixon. 1983. Environment. Natural Systems and Development - An Economic Valuation Guide, The Johns Hopkins University Press: Baltimore. Kunkle, S., et al. 1987, Monitoring Stream Water for Land Use Impacts: A Training Manual for Natural Resource Management Specialists, Water Resources Division, U.S. National Park Service: Fort Collins, Colorado. Pearce, Andrew J. and Lawrence S. Hamilton. 1986. Water and Soil Conserva- tion Guidelines for Land Use Planning Report of a Seminar Workshop, East- West Center: Honolulu. Spears, J.S. and R.D.H. Rowe. 1981. "Preliminary Guidelines for Designing Watershed Rehabilitation Projects for Bank Financing" in World Bank, Proceedings of the Second Agricultural Sector Symposium, pp. 408-423. World Bank: Washington, D.C. United Nations Environment Programme. 1982. Environmental Guidelines for Watershed Development, UNEP Environmental Management Guidelines No. 3, UNEP: Nairobi. - 33 - REFERENCES Barbier, Edward. 1988. "The Economics of Farm-Level Adoption of Soil Conser- vation Measures in the Uplands of Java," World Bank Environment Depart- ment, Working Paper No. 11. Barrett, Scott. 1988. "Optimal Soil Conservation and the Inadequacy of Agri- cultural Pricing Reforms,n unpublished, Cambridge University, Cambridge, England. Bhide, Shashanka, C. Arden Pope and Earl 0. Heady. 1982. "A Dynamic Analysis of Economics of Soil Conservation: An Application of Optimal Control Theory," CARD Report 110, SWCP Series III, The Center for Agriculture and Rural Development, Iowa State University, Ames, Iowa. Brown, Lester and Eric Wolfe. 1984. "Soil Erosion - Quiet Crisis in the World Economy," Worldwatch Institute Paper 60, (Washington, D.C.), Sep- tember. Carson, Brian. 1989. "Soil Conservation Strategies for Upland Areas of Indonesia," Occasional Paper No. 9. Environment and Policy Institutes, East-West Center, Honolulu. FAO/UNESCO. 1982. Soil Map of the World, (Rome/Paris) Gilmour. Gregory, G. Robinson. 1957. "Economic Research Needs in Watershed Manage- ment," Proceedings, Society of American Forestry, Syracuse, New York. . 1955. "An Economic Approach to Multiple Use," Forest Science, Vol. 1, No. 1, March. Hamilton, Lawrence S. 1988. *The Recent Bangladesh Flood Disaster Was Not Caused by Deforestation Alone," Environmental Conservation, pp. 369-370. _ and P.N. King. 1983. Tropical Forested Watersheds: Hydrologic and Soils Response to Major Uses or Conversions (Westview: Boulder). Hyde, William. 1988. "General Public Policy Impacts on Upland Resources and the Environment in The Philippines." Draft, World Bank. Ives, J.D. and B. Messerli. 1989. The Himalayan Dilemma. Rutledge: New York. Joshi, R.M. and Khatiwada, M.K. 1986. Agricultural Handbook Nepal. Agri. Publication Series: Kathmandu. McConnell, Kenneth. 1983. American Journal of Agricultural Economics, Febru- ary, pp. 83-89. Pearce, Andrew J. 1986. "Erosion and Sedimentation," paper prepared for the Workshop on Ecological Principles for Watershed Management, April 9-11, East-West Center, Honolulu. - 34 - Roche, Frederick. 1987. "Sustainable Farm Development in Java's Critical Lands: Is a Green Revolution Really Necessary?" (unpublished manuscript), Cornell University. Rogers, Peter, Peter Lydon and David Seckler. 1989. "Eastern Waters Studys Strategies to Manage Flood and Drought in the Ganges - Brahmaputran Basin," report prepared for USAID, April. Stocking, M.A. 1988. "Assessing Vegetative Cover and Management Effects" in R. Lal, ed., Soil Erosion Research Methods, Soil and Water Conversion Society, Ankeny, Iowa. UNEP. 1986. Environmental Guidelines for Rural Roads, UNEP Environmental Management Guidelines No. 13, Nairobi. World Bank. 1989a. "Indonesia: Forest, Land and Water: Issues in Sustain- able Development," Report No. 7822-IND. . 1989b. "Philippines: Forestry, Fisheries and Agricultural Resource Management," (FFARM Study). Report No. 7388-PH. Yadav, Ram P. 1987. Agricultural Research in Nepal: Resources Allocation, Structure and Incentives, Research Report 62, International Food Policy Research Institute, Washington, D.C. - 35 - 2. SOIL AND MOISTURE CONSERVATION TECHNOLOGIES: REVTEW OF LITERATURE John B. Doolette and James W. Smyle This chapter reviews research literature concerning on-farm impacts of soil and moisture conservation technologies on surface runoff, erosion/sedimentation and productivity and yield. It is preceded by a brief discussion of the range of treatments applied in relevant World Bank projects. The review includes data from more than 200 studies globally that appear to be based on valid experimental methods. The limits to extrapolating from the data to other sites and projects are noted. The literature covered is presented in bibliographies in Chapter 8. INTRODUCTION .The process of watershed improvement involves several important aspects, not the least of which is the selection and application of technical methods for bringing about stabilization of degraded land surfaces, that is, the reversal or arrestment of degradation, or protection against it in newly exposed watersheds, and redressing the loss in agricultural productivity due to diminished soil and nutrient status. Erosion Effects and Control Measures World Bank Proiect Interventions. The World Bank has financed many projects with substantial watershed development components. Staff appraisal reports on 35 relevant projects (20 in the Asia region), spanning 1976-87, show estimated total project costs (in current dollars) in excess of US$500 million. In most cases, unit costs for individual treatments were estimated at appraisal, but project benefits were less frequently estimated. Table 2.1 shows the technologies to be implemented over the 35 projects, their unit cost and the impacts anticipated. Categorized by on-farm and off-farm conservation interventions, respectively, the unit costs and assumed benefits are looked at more closely in Tables 2.2 and 2.3. This substantial investment, combined with the sizable commitment implicit in projects under preparation or in the pipeline, prompted regional staff to question the technologies and treatments from the standpoint of effi- cacy, synergism between them, the benefits accruing from them, and relative cost and replicability. This questioning called for a systematic review of the performance of on- and off-site 1/ soil and moisture conservation treatments in terms of physical effectiveness and benefits. The literature I/ On-site refers to the area under treatment in a micro- or subwatershed, whether it be on-farm or off-farm, whereas off-site is further downstream below the area under treatment and outside of the micro- or subwatershed. - 36 - review herein especially focuses on determining benefits and, to the extent possible, on establishing links between physical treatments and socioeconomic analysis. Perceptions of the Effects of Soil Erosion. A basic premise is that soil erosion is undesirable for reasons that are generally easy to support in quantitative and economic terms. It is important to understand that erosion effects and runoff effects are closely related and, indeed, in terms of cause and effect, runoff comes first. Defining the effects of runoff/erosion is important in determining the relevance and benefits of technical measures. In this cqntext, surface runoff and the concomitant soil erosion result in: (a) loss of agricultural productivity through: (i) reduced rainfall infiltration and moisture-holding capacity in the soil; (ii) reduced depth of topsoil; (iii) impaired soil surface characteristics and seedbed quality; and (iv) loss of soil nutrients. (b) increased sedimentation which impacts on: (i) downstream agriculture through siltation of irrigation canals and deposition of silt on farm lands; and (ii) reduced reservoir life, damage to fisheries, reduced water qual- ity, pollution from agricultural chemicals which are adsorbed to soil particles, increased maintenance costs in waterways and harbors, and stream aggradation. (c) flooding; and (d) decreases in dry season stream flow. Types of Runoff and Erosion Control Measures. The total range of treatments is better seen when grouped in terms of land-use categories, namely, on-site--arable land, nonarable land including forest and drainage lines--and off-site--downstream drainage lines and compacted areas. Within each group there may be the option of structural or vegetative/cultural mea- sures. The following measures have been applied and to varying extents have been studied: On-Site (a) arable land - contour farming; - contouring with vegetative (vetiver grass) barri- ers; - 37 - - contouring with earth banks and waterways; - earth banks on field boundaries; - furrowing, ridging, ridge tying; - tillage practices, subsoiling; - vegetative ground cover, mulching, manuring; - grass cover, grass strips, grass barriers; - improved farming (cropping) systems; - agroforestry; - terracing; and - land leveling, smoothing. (b) nonarable land - vegetative barriers on the contour; - earthen or rock barriers; - afforestation, reforestation, revegetation; - area closure; - reduced grazing pressure, stall feeding; - pasture improvement; - silvipastoral plantations; - buffer zones; and - trail, rural road and forest road treatments. (c) drainage lines - gully control structures; - checkdams, silt traps; - diversion drains; and - vegetative stabilization of natural drainages. Off-Site (a) drainage lines - grassing of artificial waterways; - stream bank protection; and - channelization. - 38 - (b) compacted areas - roads; overall design, retaining walls for cut batters; and - settlements; diversion drains. LITERATURE ON THE IMPACT OF CONTROL MEASURES Methods The body of work on soil erosion/degradation processes is extensive, yet predominantly empirical in nature, making results site-specific. A review of the literature pertaining to the on-farm impacts of soil conservation tech- nologies on surface runoff, erosion/sedimentation, and site productivity was conducted. The sources for the review were obtained through comprehensive searches of the USDA's Agricultural Library (Beltsville, Md., USA), and the World Bank's Sector Library (Washington, D.C., USA). In both libraries the computerized on-line bibliographic search facilities 2/ were used. The review considered only that literature which appeared to result from valid experimen- tal design and methodologies. Even so, the research conclusions available from the selected literature are included without making judgment as to their validity. Such judgments would require extensive analysis beyond the purview of this exercise. In addition, a special review of the Indonesian experience with soil conservation technologies was carried out within the country. A lack of standardization in field and laboratory research methodologies makes comparison of results difficult. Problems of scale compound the difficulties of applying research results, for example, when modelers try to apply pan or small-plot data to answer questions that arise on a watershed scale. Consid- erable work is still required for a more complete understanding of the physi- cal/chemical/biological processes that drive the erosion/degradation process. That work must be accomplished before an accurate "universal soil loss' process-based model can be formulated. Quantitative studies on the wide array of approaches to soil conservation technologies are scanty and only available in a few widely scattered research locations. Gaps exist in work that is applicable to tropical areas, especially high rainfall, mountainous areas. Lack of standardized methodologies and problems of scale also create problems with the applicability of these studies. Despite the problems with modeling, quantifying erosion/soil degradation and predicting quantitatively the impacts of proposed soil conservation technologies, there is sufficient understanding to allow the application of soil conservation treatments in a straightforward, systematic manner. The review groups the literature on the impact of control measures into three subjects, namely: (a) on soil moisture and surface runoff; (b) on erosion rates and sedimentation yields; and (c) on productivity and yield. 2/ USDA library facility = 'AGRIS' and 'SILVER PLATTER'. - 39 - Erosion is a term easily misused. This review is concerned with what is often called "upland erosion' in the literature; it is what the universal soil loss equation estimates: rill and interrill erosion. Rill erosion takes place when water is concentrated into tiny rivulets and proceeds predominantly by the erosive force of flowing water. Interrill erosion (often referred to as sheet erosion) proceeds predominantly by the force of raindrop impact. A further source of confusion may relate to whether soil erosion or soil loss is meant. Erosion takes place if a soil particle is at all disturbed; soil loss takes place if the particle is at all moved. The difference then is one of measurement. Since it is almost impossible to measure average soil distur- bance in any fashion that is meaningful, due to spatial and temporal variabil- ity, researchers set up collectors and capture sediment. The sediment cap- tured represents soil loss and the data used to infer erosion rates. Soil loss is then defined by where the collector is placed and, therefore, so is the erosion rate. The problems of scale are beyond the scope of this report. The only confusion that can be resolved easily is semantic; the terms to be used will be werosion rate' and "sediment yield.' Evidence suggests that tropical soils erode more quickly when dis- turbed and that the impact of erosion is greater than in temperate counter- parts (27, 50, 51, 93, 95, 97, 158, 160, 215).3/ El Swaify (51) also reports that the downstream impacts of sediment on water quality may be greater from oxidic tropical soils than from temperate soils.4/ On-site erosion/sedimentation control practices that have been studied may be broken down into two broad categories, namely vegetative/ cultural practices and structural practices. Since these two categories are not mutually exclusive and are often mutually supportive, the question of which approach to take must be decided on the basis of site-specific factors and goals. Often a combination of the two approaches is necessary, such as when vegetatively stabilizing a stream bank, it is necessary to control struc- turally the forces that cause further degradation so that the vegetation may be established (22). IMPACT ON SOIL MOISTURE AND SURFACE RUNOFF 5/ ClearinR Land for ARriculture Evidence from several studies on small watersheds shows that land clearing increases surface runoff, especially when carried out with heavy equipment (6, 62, 100, 118, 166, 201). The one exception is where heavy ash inputs, following manual slash-and-burn clearing, increased soil permeability 3/ Numbers refer to list of references in Table 8.1, which are then cited in Tables 8.2, 8.3, and 8.4. 4/ Where possible, soil types are presented using the FAO-UNESCO (1974) soil taxonomy. The translation from USDA to FAO-UNESCO was done based on: Breimer, R.F., van Kekem, A.J., and H. van Reuler, 1986. Guidelines for Soil Survey and Land Evaluation in Ecological Research. MAB Technical Notes No. 17, UNESCO, pp. 29-30. 5/ Literature cited in Table 8.2, page 191. - 40 - so that surface runoff decreased (181). Conversion of land from forest to grass has been found to increase surface runoff and total water yield (20, 62, 77, 134, 160, 176) except when grass production is high (77, 176). Conversion to agricultural lands also has significant impact on surface runoff from small watersheds (63). The impact is, however, use-dependent. Poor land management will increase surface runoff; significant reductions in surface runoff are usually associated with intensively managed areas (155). Little effect of soil/moisture conservation practices has been noted on water yield from areas larger than 120 hectares (155). Vegetative and Cultural Measures Grass Cover and Strips. The effect of a grass cover on surface run- off varies--runoff reductions from 56? to 902 have been observed on slopes from 0.5-Z to 46? when compared to surface runoff from other agricultural types (50, 68, 102, 103, 173, 182)) and one study (161) showed no difference in surface runoff between a grass cover of Imperata cylindrica or Saccharum spp. compared with secondary or plantation forest on 36-70? slopes. Yet, Pennisetum spp., Cynadon sUP., Urochloa spp., Panicum spp., and Desmodium spp. showed no effect on surface runoff when compared to conventional cultivation in two cases with, respectively, 5? and 46Z slopes (102, 202). Effects of grass barriers or strips have not been well investigated, though it appears that they do have an impact on surface runoff. On 12-22? slopes, grass strips reduced surface runoff 9-14? (3, 183), and planting grass strips on bench terraces on 28X slopes reduced surface runoff 69Z, compared to unplanted ter- races (102). The spacing of strips and types of grass affect the impacts on surface runoff--2.5 meter spacing of grass barriers gave a 592 greater reduc- tion in surface runoff than grass barriers spaced 5 meters apart, and Eragrostis spp., guinea grass, and South African pigeon grass barriers reduced surface runoff 30? and 25?, respectively, on 0? and 23? slopes (102). Mulching. Mulching protects the soil from raindrop impact, reduces evaporation from the soil surface, and slows down runoff, giving more time for water to infiltrate (96, 160). Measurements of soil moisture storage on mulched plots showed an increase of 38? when mulch was applied at 1.1 t/ha; by increasing the amount of mulch to 4.4, 8.8 and 12 t/ha, moisture storage was increased 61?, 93Z and 104%, respectively (180, 197). Mulch applied at 9 t/ha conserved the equivalent of 23 days of a soybean crop's moisture requirements (189) and 0.66 t/ha of mulch increased soil moisture storage 3Z (38). Surface runoff, an indicator from which increases or decreases in soil moisture stor- age could be inferred, has been shown to be reduced by 16Z to 91? on slopes ranging from 1Z to 46? as the result of mulch applications (54, 97, 102, 122, 188, 189, 192, 199, 213). Rainfall simulator studies have shown that mulch has no effect on surface runoff from low-intensity rainfall (171), yet reduces surface runoff by 11-96? on 52 slopes (with mulching rates of 0.63-4.94 t/ha) for higher-intensity events (113). One study (21) on 2? and 8? slopes showed a combination of mulch and manure at different rates to reduce surface runoff from 64? to 93? compared to untreated plots. Stubble mulching with crop resi- dues has also been shown to increase soil moisture storage (180) and to decrease surface runoff (84, 116). Cultivation practices may also be effective in reducing runoff. As a general principle, tillage and cultivation reduce the size of soil aggregates - 41 - and promote pore clogging which can result in decreased infiltration (66, 133). Increases in surface runoff of 150-1,3172 have been reported as a result of taking land out of permanent grass and putting it under cultivation (158). However, infiltration increases ranging from 72Z to 860? have been observed as a result of different tillage and cultivation practices, compared with untilled land, as a result of increased surface roughness slowing down or detaining surface runoff (28). Poor tillage practices, such as up-and-down slope cultivation, can increase runoff, even when combined with beneficial practices such as cover cropping or mulching (54, 68). No-till/minimum till methods on 1-18? slopes have decreased surface runoff from 7? to 86Z in compa- rison to traditional and conventional cultivation methods (90, 116, 122, 157). No-till systems combined with mulching have reduced surface runoff from 43? to 92X on slopes from 3.5Z to 14?, compared to unmulched, traditional cultivation (116, 190). On slopes as high as 212, no-till carried out on the contour reduced runoff 43? compared to clean tillage (73). Increased surface runoff has been observed in a no-till/stubble mulch/chemical weed control system compared with conventional tillage/stubble mulch. The study reported a 16? increase in surface runoff attributed to chemical weed control (212). Disk plowing/stubble mulching were shown in one study (116) to reduce surface run- off about the same amount as a no-till/mulch system. Contour Cultivation. Reports from India on contour cultivation and planting, based on 30 years of work at an agricultural experiment station, suggest that this practice alone will reduce surface runoff 25? when compared to up-and-down slope cultivation (46). Other studies have found a decrease of 70? in surface runoff on a 25? slope (68) and an 82? decrease on a 20Z slope when contouring was combined with mulching (102). These two studies were also in comparison to up-and-down slope cultivation. A 29Z and 48? decrease in surface runoff was observed on 2.2? and 6? slopes, respectively, when the only difference was contour cultivation (16, 60). Direct measurement of soil mois- ture showed an 8? increase in storage on medium- and fine-textured soils on contour-farmed lands and no soil moisture benefits in coarse-textured soils that are contour-farmed (24). Compared to a control continuously under shrub and grass, contour-cultivated land showed a decrease in surface runoff of 82 on a 25? slope and an increase of 14? on a 302 slope (54). Ridge and Furrow. Ridged and furrowed plots have demonstrated deeper percolation of water (164, 218), with increases of 121? in soil moisture stor- age (47) and decreases in surface runoff on the order of 31-86% (122, 131); combined with broadbedding or field banks, surface runoff has been reduced on the order of 38-682 (130, 142, 180). There have been, however, concerns that runoff may increase in situations where water is allowed to concentrate in the furrows (178). In a structurally unstable soil, broadbed and furrowing caused a 38? increase in surface runoff and a 67? increase in peak discharge on a slope of less than 1X (142). It has also been noted that there are few or no moisture conservation benefits from this practice in sandy soils, in clay soils which become compacted from in-field operations, on rough or broken land, on steep slopes, or where large amounts of sediment will accumulate (78, 120, 200). - 42 - Structural Measures Earth Banks. Earth banks require frequent repair and maintenance (26, 68) and in some structurally poor soils (142) or poorly drained soils are impractical--as one author, working in vertic soils, commented that tradi- tional farming consisted of "...breached banks, local varieties and no ferti- lizer" (203). Also, ponding of water behind banks may occur on poorly drained soils (68). Grading of banks and putting them on the contour can reduce run- off 23-37Z when compared to ungraded/uncontoured banks; however, a 50? decrease in depth of percolated water (168) and a 702 increase in surface runoff (169), as well as 182 and 37Z decreases in runoff (142, 187), have been observed with contour-banked compared with unbanked plots. Land Leveling. Little work was encountered on soil moisture/runoff impacts of land leveling. One study showed that a leveled area accumulated in seven months the soil moisture that 19-21 months of fallow would accumulate (126). But this same study also concluded that benefits were dependent on the timing and distribution of rainfall; this may explain the lack of any effect of leveling on soil moisture found in the other study (112). Terraces. Studies conducted on terracing generated a wide range of findings. No changes in surface runoff were found between terraced land and contour-planted corn on 2-18? slopes (177); between terraced and unterraced land (10, 74); and on coarse-textured soils and other soils with low water- holding capacity (75). Direct measurement of soil moisture found either no changes from terracing (76), that any changes were temporal and may only be observed during the rainy season (58), or that there was a 50? decrease in the depth to which water percolated (168). Terraces, when compared with graded furrows on gentle slopes, were found to increase surface runoff on the order of 25Z, with the greatest increases coming in small storms (84, 155) or to perform no differently (68). Compared to unterraced plots, terraced plots have been found to increase surface runoff if antecedent moisture conditions are high or decrease it if they are low (10). As the result of terracing, increases in surface runoff of 20? (10), 116? (169), 140Z (182) and 31-201l (183) have been observed. Significant increases in peak discharges and time to peak discharge, decreases in soil moisture, and low flows have been observed as the result of completely terracing small watersheds (108, 110). Conversely, a 28-year study reported that terracing decreases peak flows with the magnitude of decrease inversely proportional to the size of the watershed (10), and in an 11,500-ha watershed there was a reported 11Z decrease in sur- face runoff after 42? of the watershed was terraced and a 44? decrease in surface runoff after 75? of the watershed was terraced (89). When comparing terracing to other practices, such as clean cultivation on slopes of 9-47?, terracing reduced surface runoff 31-86? (72, 102, 103, 163, 191). The mois- ture retention function provided by terraces in some situations (75, 124) has also created difficulties in carrying out tillage as a result of excess soil moisture (17) in others. It is apparent, however, that when terraces are used in conjunction with other improved practices they do enhance moisture conser- vation. In general, the experience in the United States has been that contour-farmed terraces reduce surface runoff 9-37Z compared to unterraced agricultural land (74, 219). Inclusion of such practices as conservation cropping, permanent covers, mulches, rotations, and deep tillage have been shown to decrease runoff 20-90? on terraced lands (10, 11, 103, 167). - 43 - IMPACT ON EROSION RATES AND SEDIMENT YIELDS 6/ The effect of treatments on erosion rates and sediment yields is arranged by climatic zones, which is not to suggest that research from one climatic zone is not applicable to another. Physics is not changed by rain- fall and temperature regimes; what does change, however, is the relative importance of individual parameters in the erosion/sedimentation process. For example, antecedent moisture conditions are much more important in a monsoonal climate than a semi-arid climate. Since soil types are also loosely correla- ted with current climatic regimes, climatic zone seems a useful basis for compartmentalizing research results. Climatic Zones Equatorial Monsoon - receives both monsoons, no regular dry season [precipitation (Pt) = 2,000+ mm/yr], and, Continuously Wet Tropics - humid for more than nine months (Pt = 1,400+ mm/yr) The review found limited quantitative research on soil conservation in these two climatic zones. It is reasonable to assume that given their large volume of annual precipitation, soils are often near or at saturated conditions. Extensive overland flow in well-vegetated areas could be expected as a common phenomenon (176) and erosion/sedimentation hazards high in dis- turbed catchments. Research available shows that for slopes of 3.5-22Z tech- niques such as minimum tillage with mulching could reduce sediment yields 56-992, compared with unmulched traditional farming systems and that grass strips reduce sediment yields 932 compared to bare soils (3, 190). Commercial fertilizer applications and manuring were also found to reduce sediment yields 992, compared to nontreated plots, due to increased yield (21). These results suggest the linkage between a productive cover crop, infiltration, and reduc- tion of runoff velocities. Some evidence from these zones supports the idea that traditional hill-slope agriculturalists practice farming systems that minimize erosion (160). No research on structural soil conservation treat- ments in these two climatic zones was encountered. Dry/Wet Monsoon - One monsoon predominates, receiving rela- tively little precipitation from the other (Pt = 1,800+ mm/yr). A wide variety of research exists for this climatic zone. Work has been done on permanent grass covers in comparison to clean cultivated agricul- tural lands (7, 30, 41, 68, 102, 152, 161, 167, 202); research findings taken from slopes ranging 5-70Z show decreases in sediment yields of 50-94Z. Grass strips alone have proven effective on slopes up to 30Z (no available research at slopes beyond this figure), reducing sediment yields from 71X to 99.72, compared to clean cultivation (33, 102, 103, 205). The effectiveness of grass strips in reducing short-term erosion rates, however, may not be as great. Compared to clean cultivation, erosion rates ranged from not significantly 6/ Literature cited in Table 8.3, page 203. - 44 - different to 242 less on areas with grass strips (4,183). The addition of mulch in conjunction with grass strips was very effective on slopes of 23-28Z, reducing sediment yields 97-992 compared to clean cultivation (102). Mulch alone reduced erosion rates from 162 to 99Z (1, 188, 189) and sediment yields from 642 to 1522 (41, 116) on slopes up to 352. Research on cultural treatments for soil conservation in this zone found that no-till/stubble mulch systems can decrease sediment yields 64-932 (116), and that contour cultivation by itself (based on 30 years of experiment station research in India) decreases sediment yield 30X in comparison to up- and-down slope cultivation (46). Research in other areas found reduction in sediment yield from contour cultivation to range from 302 to as high as 80S when used in combination with mulching (68, 102) in comparison with uncon- toured plots. One study reported that at higher slopes, 252 in this case, contour cultivation alone was ineffective and yielded 14 times more sediment than a bench-terraced plot (143). Other cultural practices such as mixed cropping reduced erosion rates by 74-992 on 20 slopes when compared to strip cropping or monocultures (106, 213), and the addition of manure (16 t/ha) reduced erosion rates 422 on a 162 slope (2). Finally, when compared to seed- bed preparation by plowing on a 102 slope, burning and dibbling reduced ero- sion rates 382 in a corn field (30). The structural treatment of contour banks (syn. bund) appears to have a limited life span of 2-5 years in this zone and at slopes above 112 to fill rapidly with silt and have trap efficiencies on the order of 30-502 the first year and 02 the second (79, 116). Contour banks also have been found to be useful only where soils have good drainage (116, 159), otherwise they are susceptible to breaching and failure (68, 79, 116, 159, 203). They have, however, been shown to decrease sediment yields 32-462 on well-structured soils, in comparison to unbunded plots (169,187). Many types of terraces have been tested in this zone, such as inward-sloping, outward-sloping, level- drainage and level-retention bench terraces. As a generic type of structural practice, over a range of slopes from 102 to 302, terraces have been found to decrease sediment yields 50-902 (33, 68, 102, 103, 104, 143, 169, 205); they have also been shown to reduce erosion rates 20-95Z on slopes ranging from 92 to 402 (71, 152, 163, 184, 191). Compared among themselves, it is evident that evaluation of slope, soil, and rainfall/runoff regime should decide the type of terrace used. Sediment yield increased 17 times with outward-sloping, compared with inward-sloping bench terraces (104) and no difference in sedi- ment yields was found in two studies between the less costly approaches of graded terracing and hillside ditching, compared with bench terraces (68, 169). Erosion rates may also be effected by terrace type; flat-bench terraces and sloping-bench terraces increased erosion rates 1252 and 800Z, respec- tively, when compared to ridge terraces in an upland situation (140). Over the range of slopes researched in the dry/wet monsoon climatic zone, 5-302, it appears that vegetative/cultural approaches such as grass cover, grass strips, and mulch are almost as or are as effective as structural approaches. - 45 - Wet and Dry Tropics - Clearly defined wet and dry seasons. Dry season persists for 3 to 6 months (Pt - 1,000+ mm/yr). A variety of soil conservation work has been carried out in this zone. No research was encountered that investigated the effects of grass strips. Permanent grass cover has been found to reduce sediment yields 84-1OOZ, compared with clean cultivated plots (54, 62, 182, 215). Mulch, tested on slopes of 3-8X, reduced sediment yields 38-98Z in comparison to traditionally cultivated plots (122); on 302 slopes, the mulching of plots that were cultivated up and down slope reduced sediment yields by 28Z (54). Experiments with a 50Z cover of asphalt mulch reduced erosion rates 681 (149). Poor cultural practices such as cultivation up and down slope have been shown to accelerate erosion rates, even when combined with good practices such as cover cropping or mulching (54). When compared to conventional till- age and bare soil, no-till systems have proven to decrease sediment yields by 61-98% on slopes ranging from 1Z to 52X (27, 67, 122, 157); when combined with mulching, a no-till system reduced sediment yields 99Z, compared to conven- tional cultivation (139). Land clearing and cultivation, especially on sandy soils, may increase erosion rates drastically (up to 115 t/ha/yr) (97, 98, 100). Burning to clear a site for agriculture, then practicing no-till culti- vation decreased sediment yields 85Z, compared to conventional tillage (90); bulldozing, leveling and conventional tillage increased sediment yields 105Z versus burning and conventional tillage (90). Hand cultivation versus mecha- nized cultivation reduced sediment yields 382, compared to plowing and harrow- ing or mouldboard plowing and 70Z, compared to plowing and bare fallowing (139). Other practices found to be effective when compared to up-and-down slope cultivation are cross-slope and contour cultivation and ridge and furrow systems which have reduced erosion rates 43-96Z on slight slopes (16, 90). One study (122) showed that on one of two sites no-till, minimum till, and contour cultivation actually increased sediment yields 33-205Z when substitu- ted on 32 slopes for a traditional mixed cropping scheme. Tillage practices appear to be site-specific in impact. Deep tillage when compared to shallow tillage on a erodible sandy soil subject to compaction decreased the erosion rate 63% (36); on a highly weathered Ferrasol, deep tillage increased the erosion rate 1002 in comparison to shallow tillage (114). Shifting cultiva- tion by traditional practitioners on steep slopes appeared to result in little accelerated erosion (150, 209), but practiced by newcomers to the area resulted in disaster (209). Terracing, compared to unterraced lands and the same cultural prac- tices, showed 26-89Z decreases in sediment yield on slopes of 5-45Z (56, 94, 129, 182). When comparisons were made among terrace types, a study on 5Z slopes (56) demonstrated that sediment yields from different types of terraces could range from a 70Z decrease to a 692 increase, depending on whether the terrace was backsloped or broad-based in comparison with level or out-sloping. One contour banking study showed that earthen banks were superior to stick or stone banks on 30Z slopes. The earthen banks were 34-39Z more efficient at reducing sediment yields versus the stick or stone type; compared to bench terracing, however, the earthen banks were 58-70% less efficient on these steep slopes (129). - 46 - Temperate Zone - Rainfall every month or well marked rainy season. May be snow in winter (Pt = 600+ mmiyr) The majority of work done on soil conservation has been done within the temperate zone. The applicability of this work to the tropics has always been in dispute, though Sanchez (160) states that for soils that are Entisols, Vertisols, Yermosols, Solonetz, and others with little or no iron or aluminum oxides, the classic concepts of soil conservation developed in the temperate zones are entirely applicable. Research on grass strips has found a 40-702 reduction in sediment yields in farm fields on relatively flat lands, compared to no grass strips (31, 74, 217). Vegetative riparian buffers and very wide (> 50 ft) grass strips were found to have a trap efficiency for sediment of 84-99Z for runoff moving at low velocities (39, 217). Mulches alone, at rates of as little as 0.33 t/ha, have been shown to reduce sediment yields by 42% (171). Doubling that rate of application to 0.66 tIha of mulch reduced sediment about 66% and mulch at about 2.5 t/ha reduced sediment yields 80-97% (99, 113, 123 171). Cultural operations such as contouring were considered to be most effective on slopes of 3-9%, reducing sediment yields by 50-86% compared with uncontoured fields (73, 219). Most research on cultural operations involved a treatment package of different tillage types with mulch/chemical weed con- trol/crop residue management or stubble mulching. Ranges of sediment yield reductions were 26-99.9Z on nearly level to 21Z slopes (12, 31, 73, 74, 137, 178, 195, 219). There is also evidence that some tillage practices, for exam- ple, ridge tillage without crop residue management, are inappropriate and may increase erosion/sedimentation when not used in combination with other prac- tices (178). Structural approaches to soil conservation were, again, found to be site-specific in their application. On 9% slopes in shallow erodible sand, contour banks increased sediment yield 200Z (121), whereas on a 32Z slope, contour banks reduced sediment yields 252, 772, and 99.9Z, depending on whether the banks were placed every 10 rows, every 5 rows, or every row (148). Research has characterized terraces as effective, but not necessarily more effective than grass or graded furrows, at reducing sediment yield (74, 177, 219) (155, 177); expensive ("more expensive per ton of soil erosion reduction than any other alternative for soil erosion control") (198); and prone to failure if used in structurally unsound soils (23, 87). Semi-arid, Tropics - One rainy season, marked by 4-8 month dry season (Pt = 400+ mm/yr), and Semi-arid, Temperate - One rainy season or patchy rainfall dis- tribution (Pt - < 600 mm/yr). In these two climatic zones, natural upland erosion can reach its peak. Density and distribution of vegetation are patchy, long-term wetting and drying cycles create hydrophobic conditions in soils; initial infiltration rates, especially in sandy soils, can be quite low; and finally, what precipi- tation does occur often comes in the form of high-intensity rainfall. The - 47 - little soil conservation research encountered for these climatic zones was mostly of the cultural practice-type, where methods to increase infiltration and contain runoff for future crop use predominate. Permanent grass cover as an alternative to wheat was found to decrease sediment yields 98Z on 8% slopes (50), but as an alternative to natu- ral cover on 0.5-1Z slopes to increase sediment yields 6-12z. In comparison to other agronomic treatment, though, the grass-covered plots produced from 74Z to 92Z less sediment (173). One study on grazing effects (211) showed a 2002 increase in sediment yields when short-duration grazing (a few days of high-density grazing) was practiced in place of moderate continuous grazing. Cultivated fallows are a common practice in semi-arid zones as a means of storing soil moisture for later crops. This practice has been called into question due to erosion hazard and, on very gentle slopes, sediment yields from these fallows have been shown to be 850-1,239% higher than natural cover and to be 22-214% higher than when in crops (86, 173). Manuring and mulching have proven effective in reducing sediment yields on slopes of 2-25Z: decreases of 73-98%, depending on whether used alone or in combination, have been reported in comparison to bare fallow and no sediment yield increase has been reported in comparison to a vegetated fallow (21, 167). Contour cultivation has been found effective on slopes ranging from 3% to 82; at less than 3Z, cross-slope cultivation approximates the contour and at greater than 8% increased washouts negated short-term benefits (54, 58). Broadbed and furrow systems have been found to increase sediment yield on poorly drained soils compared to better drained soils (130, 142) for tradi- tional cropping systems. One study (142) warned of structural instability for furrowing or bunding in a fine-textured Alfisol. A broad-based terrace study, the only terrace study encountered that was concerned with erosion/sedimenta- tion, demonstrated a 92Z decrease in sediment yield in comparison to an unter- raced field (144). IMPACT ON PRODUCTIVITY AND YIELD 7/ Inevitably, erosion decreases the productivity of a site. If the subsoil characteristics of eroded sites are favorable, then erosion necessi- tates higher production costs without affecting yields (93, 95). Generally, soils with favorable subsoil characteristics are Andosols and Cambisals, while soils with unfavorable subsoil characteristics are often Ferrasols, Acrisols and Nitosols. On shallow, infertile tropical soils, productivity may decline more rapidly than in similar temperate soils (93, 95). Some studies on the effect of natural erosion on yield have found a yield decrease of an average 0.14 tons/ha per mm of soil loss (95),8/ or that yield declines 3-7.52 after 1 mm of soil loss and declines 10-25% after 8 mm of soil loss (115). Other studies have shown that an erosion rate of 5-14 mm/yr (77-216 t/ha/yr) resulted in yield declines of 50-70% (57). In comparing crop yields from plots that had eroded down to subsoil and plots where topsoil was intact, a decrease of 15-28% was found for the various crops planted in the exposed 7/ Literature cited in Table 8.4, page 219. 8/ For purposes of conversion of soil depths or soil volumes to soil mass, a conversion factor of 1.4 grams per cubic centimeter was used. - 48 - subsoil (31). Certain volcanic soils (Andosols) and loessial soils have fairly uniform physical, chemical, and structural characteristics present to great depths in the soil profile. In such soils the loss of several centime- ters of topsoil will have little or no effect on crop yields (93); convincing the cultivators of such soils to employ soil conservation measures may be difficult. Causes of Productivity Decline Causes of erosion-induced yield decline differ among soil types, ecological environments, climatic conditions and crops, with no single factor or combination explaining yield variability. There are obvious causes such as decrease in net arable area or burial by erosion products. There are indirect causes such as when timeliness of farm operations is disrupted by difficulties in seedbed preparation or delayed planting. The significant causes of yield or productivity decline are primarily due to changes in three soil variables, namely, plant-available soil water, plant-available soil nutrients, and orga- nic matter. Soil Moisture Loss. The conclusion of the United States National Soil Erosion-Soil Productivity Research Committee was that the primary effects of erosion on productivity are due to loss of plant-available water (93). The reduction of soil water reserves available to the plant is due to several factors, namely: (a) Increased Surface Runoff. The increase is dependent on slope, vege- tation (type and distribution), soil characteristics, climatic regime, level of disturbance, and subsequent management practices. Estimates of loss due to increased surface runoff vary widely. (b) Decreased Soil Depth. The decrease directly reduces water-storage capacity and effective rooting depth, especially in soils with shal- low/rooting lithic contacts, unfavorable subsoil characteristics or restrictive layers. (c) Decreased water-storage capacity due to physical changes in soil structure (e.g., loss of noncapillary porosity, crusting) and the preferential depletion of soil organic matter and certain clay frac- tions by the erosion process; organic matter and clays contribute disproportionately more to water-holding capacity relative to other coarser soil fractions. The literature does not disaggregate the separate effects of these factors, nor does it distinguish moisture from the other key soil factors. It mostly concerns water-use efficiency. Drought effects are magnified on eroded soils (95) and water-use efficiency of crops is decreased (186). A 4-5? decrease in plant-available water in an eroded soil caused a 12-36% decrease in yield (61). Yost et al. (186) report reduction in water-use efficiency due to soil loss. Kilograms of dry matter per liter of water used ranged from 0.66 kg to 0.7 kg in uneroded control, 0.39-0.64 kg when 10 cm depth was removed and 0.07-0.45 kg for a loss of 35 cm. Water-use efficiency for a soil that lost 35 cm was only 50% that of an uneroded control. - 49 - Soil Nutrient Loss. The impacts of erosion have been investigated by removing soil from plots and attributing subsequent yield differences to the depth of soil removed. These soil removal studies are useful only as indica- tors of the effects of erosion. This is true, in part, because the process of erosion under natural conditions preferentially removes soil components, namely, organic matter and certain clay particles that are important in main- tenance of cation exchange capacity (CEC) and moisture-holding capacity. Soil removal, on the other hand, impartially removes all soil fractions. The expe- rience with soil removal studies has been, averaging across different soil and crop types, that the removal of 5 cm of soil reduces yields on the order of 60Z, of 10 cm reduces yield on the order of 65Z, and of 20 cm reduces yield on the order of 802 (95, 96, 118, 221). It should be noted that these percent- ages are useful only as indicators of the magnitude of the response, in that it is the removal of the top few centimeters that has the greatest impact. Lal (93) states that the available data on yield reduction per unit loss of topsoil is more drastic for the tropics than for the temperates. Other stu- dies have combined soil removal and fertilizer inputs to look at the unrecov- erable productivity decline associated with soil loss (117, 221). One study (96) analyzed sediment and surface runoff from different slope classes to determine the amount of nutrient export taking place. Combining both parts of the study, nutrient flux off site for slopes of 1-15% ranged from 9 kg to 235 kg/ha/yr nitrogen, 0.7-9 kg/ha/yr available phosphorus, 4-6 kg/ha/yr potassium, and 50-3,070 kg/ha/yr organic carbon. Looking at nutrient loss as a function of soil loss, it was observed that on slopes of 3-6.5Z, losses of organic carbon and nitrogen were soil-dependent (52). On well-drained sands, losses were 0.97 kg of nitrogen and 10.7 kg of organic carbon per ton of soil loss per year. On other soils, losses were 2.1 kg and 15.4 kg of nitrogen and organic carbon, respectively, per ton of soil loss per year. Losses of phos- phorus were the same irrespective of soil type, with 0.16 kg of phosphorus per ton of soil loss per year. On 18Z slopes, nutrient losses per ton of soil loss per year were 30 kg of organic matter, 1.5 kg of nitrogen, 1.0 kg of phosphorus, and 2.0 kg of potassium (30). Artificial stabilization of a soil surface decreased the loss rate of major nutrients by more than 90% (44). Maintaining and/or replacing nutrients is the basis for sustainable productivity. The most common approaches to nutrient management have been to use organic or commercial fertilizers. Organic matter is essential in unfer- tilized systems--it supplies nitrogen and sulfur, blocks phosphorus fixation, maintains CEC, improves structure of the soil, and forms complexes with micro- nutrients (160). Organic matter is important in soils with a low CEC or in poorly aggregated sands. Commercial fertilizers, by providing for increased on-site productivity, increase soil organic matter through the decomposition of roots. In choosing between organic or commercial fertilizers, the deci- sion, if soils are adequate, should be based on economics, transport, accessi- bility, and social criteria (160). Organic Matter. Organic matter effects soil moisture by enhancing soil permeability, infiltration capacity and moisture retention. In terms of soil nutrients it supplies most of the nitrogen and sulfur and half of the phosphorus taken up by unfertilized crops (the slow release pattern from orga- nic matter is an advantage over chemical fertilizers), it limits phosphorus fixation and can form complexes with micro-nutrients to restrict their leach- ing. Its depletion in eroding soils is serious. Typically in tropical soils, - 50 - the zone of enrichment is narrow and removal of the top layer exposes subsoil layers with very little organic matter, poor structure, porosity and nutrient content. Under forest conditions, tropical soils are high in organic matter due to the rapid rate of replacement, but once cleared, even without erosion, the high rates of decomposition, mineralization and leaching lower organic matter content rapidly. Losses of organic matter are high with erosion prod- ucts and rates of replenishment low because of the competition for crop resi- dues and other sources for fuel and fodder. Organic matter in whatever form is essential to preserve and protect top soil, for optimizing soil water and for the efficient use of chemical fertilizers. It is the only soil amendment that can be produced on or near the farm. Treatment Effects. Soil treatments variously affect productivity and yield and the literature reviewed isolated the effects of several treatments as follows: (a) Cover. Vegetative soil covers have been shown to conserve or increase soil nutrients. On a grassed plot, soil nutrient decline was only about 192 of what it was on a bare tilled plot (182), and on plots cover cropped with legumes there was about a 522 increase in major nutrients compared to clean cultivated plots (85). Mulches have been shown to increase yields from 7Z to 188Z in comparison to yields from unmulched plots (38, 91, 105, 149, 153, 174, 185, 193). (b) Clearing of land with heavy equipment for agricultural use, compared with clearing by slash-and-burn, has had yield impacts ranging from none (40) to 16-74Z declines in various crops (136, 166); subsequent fertilization did not return yields from the mechanically cleared plot to yield levels of the slash-and-burn plots (136). The observed yield differences were attributed to the benefits of ash on the slash-and-burn plots versus soil compaction and topsoil disturbance by the heavy equipment. (c) Tillage, by itself, significantly reduces nutrient loss in agricul- tural systems by incorporating fertilizers into the soil (44). A comparison of shallow hand cultivation (to 5 cm), deep mechanical cultivation (15-20 cm) plots, and uncultivated plots showed a 22-103% yield increase and a 8-73X yield increase for hand and mechanical cultivation, respectively (36). Deep tillage was shown to increase yield an average of 282 in 10 crops, over four years, in seven regions of India (153). One author (158) warns that physical degra- dation and productivity decline caused by mechanized agriculture is more rapid in the tropics than the temperate regions; therefore, the beneficial effects of tillage will not last without liming or inclu- sion of rotations with deep-rooted grasses. No-till/minimum till systems have shown a range of impacts from a 21Z yield decrease (in peanut yields, attributed to the low growing habit of peanuts, exa- cerbating the effects of weed competition) to a 51Z increase in yield compared to traditional cultivation (97, 122). In combination with mulches, no-till has shown a range of impacts, from a 7% decrease (in peanut yields) to a 139Z yield increase (mung beans) compared to traditional cultivation (190, 212). - 51 - (d) Contour cultivation and ridginR across the slope have shown yield increases of 6-662 on 1.5-32% slopes, when compared to traditional or up-and-down slope cultivation (16, 34, 46, 104. 105, 122, 131). Based on 30 years of research station experience in India, contour farming increases yields up to 352 (46). The inclusion of mulch with contour farming increased yields on a 28% slope by 1052, compared with up-and-down slope cultivation (123). (e) Vegetative Barriers on Key Contour Lines. The vegetative barrier- contour cultivation system using vetiver grass, Vetiveria zizanioides (Chapter 1, Box 1.2), being developed and promoted in World Bank projects in India, has not yet been referenced in the research liter- ature and data to quantify its benefits are just being collected. A recent set of data from Karnataka, India compared two factors, namely, soil moisture in the top 15 cm and yield, on a contour- cultivated field with vetiver hedges and a field cultivated up-and- down slope. It showed that, on the contour-cultivated field, the soil's wilting point was delayed 14 days, the yield of finger millet without fertilizer increased by 252 from 550 kg/ha and with fertili- zer by 572 from 790 kg/ha: in short, a classic pattern of increased growing season, yield and response to fertilizer. Data from the first year of experiments in Maharashtra showed the same yield pat- tern and reductions in soil loss, ranging from 382 to 73X; the maxi- mum observed difference was a decrease in soil loss from 33 tons/ha to 8 tons/ha when cross-slope cultivation was changed to contour cultivation with vetiver grass contour hedges. Surface runoff decreased 20-602 with treatments of vetiver grass, explaining the observations of increased crop-available soil moisture. (f) Ripped furrows in semi-arid zones have shown 2502 and 3002 yield increases over nonfurrowed areas (25, 47). Furrow dams in a semi- arid study area increased yield 15-20% compared to no-furrow dam areas (180). (g) Construction of banks as soil or moisture conservation structures has failed to show any significant or stable increase in yield in con- trolled experiments at ICRISAT (86). In poorly drained Vertisols, contour-banked plots showed yield decreases in each of eight years, due to ponded water interfering with tillage and damaging crops (68). A 1962 experiment (209) found 352 yield increases with banking and 98Z yield increases with banking and land leveling combined, and on a structurally stable, high-quality soil, contour banks were as effec- tive at increasing yields as contour banks in conjunction with ter- racing or mulching on 9-122 slopes (5). (h) Land levelins and level pans have shown yield increases ranging from 02 to 1802 in comparison to unleveled plots in semi-arid areas (88, 112, 126). (i) Terracing is a technology whose experimental effects on yield vary widely. On the negative side, terraces have been found to have any- thing from no effect on yield to a 322 decline in yield (17, 18, 35, 56, 68, 76, 104, 105, 182). Reported are impacts such as significant - 52 - increases in loss of major nutrients (excepting N) (182), yield reductions persisting for six years due to topsoil disturbance (17), topsoil disturbance resulting in the necessity of high levels of commercial fertilizer inputs (18), and poor drainage resulting in degradation of soil tilth from tillage operations and crop damage from excessive moisture (35, 68). As an agronomic practice, terrac- ing alone-may have no effect on yield (74, 76). On the positive side, terraces have been shown to increase yields from 16? to 100? (34, 56, 68, 75, 105, 111, 124). The important question is, what caused these differences? Soil disturbance is one causal factor--a study of five different terrace types, four structural and one vege- tative, found that yield was 18-45Z higher on the vegetative terrace due to lack of soil disturbance (42, 68); another study found that replacement of two inches of topsoil on bench terraces (to cover subsoil exposed by terracing) increased yields as much or more than the highest fertilizer applications (18). Terraces, and bench ter- races especially, require deep fertile soils on moderate slopes to permit the topsoil removal that is necessary for leveling (17). Shallow droughty soils do not have the level of productivity required to justify the annual maintenance cost, let alone the initial cost (30). Construction of the wrong type of terrace is another problem. Level-absorption terraces and conservation-bench terraces are for use in climates where retention of moisture is important and will not be appropriate in humid climates or in areas with poorly drained soils. Outward-sloping terraces are useful in areas with well-distributed, low-intensity rainfall, and well-structured permeable soils. Inward sloping and level drainage terraces are useful in humid climates. All terraces require soils that have good engineering properties-- silts and fine sands are unsuitable soils in which to terrace. Cer- tain landscapes in Indonesia such as the upland marls are physically unsuited to classic bench terraces and can be severely destabilized by their construction (30). Depth of soil over parent material and the parent material itself are important. Shallow soils over slick parent materials (e.g., granites and shales) (23) that are not well- weathered can result in massive terrace failures. The proper terrace type may also be crop-specific; in Indonesia (135) rice yields were 18? higher on bench terraces compared to ridge terraces, but the more expensive bench terraces produced no greater yields for peanuts or cassava than did the ridge-terraced lands. Whether or not to choose terracing, and then, which type(s) to choose are site-specific ques- tions. NEED FOR STANDARDIZED RESEARCH METHODS There is an urgent need for standardizing methodologies to increase the reliability and accuracy of data on soil erosion, soil moisture, produc- tivity and yield. This is particularly important in determining magnitude on a global or regional basis. Despite the extensive literature, reliable quan- titative data are limited, having been derived by survey and visual assessment and experiments lacking standardized methodologies. At the local level, sound results from well-designed and properly equipped experiments are required to enable scientists and policymakers to develop watershed management strategies. Further, as the review discovered, the types and range of experiments must be - 53 - expanded to cover the extreme situations experienced by small farmers, partic- ularly those farming the steep slopes of upper watersheds. The International Society of Soil Science through a subcommittee on soil conservation and environment has published 'Soil Erosion Research Methods*9/ which attempts to standardize methodologies. The book addresses the issues of evaluating erosion problems with nonstandard methods, data pre- cision and reliability, and then deals with the methodologies involved in laboratory, field runoff plots, and large river basins; the design and use of rainfall simulators; modeling soil erosion processes; methods of monitoring erodibility and erosivity and of canopy cover; and assessing the impact of erosion on productivity. For more effective physical monitoring of Bank-supported projects, a number of these techniques may be considered. Generally they are labor- intensive but relatively inexpensive and their application in projects would add greatly to the understanding of the effectiveness and benefits of treat- ments. 9/ Soil Erosion Research Methods, 1988, R. Lal, ed., Soil and Water Conservation Society, Ankeny, Iowa. StYaRY OF WORLD SANK PRWJECTS REPORT O CO"RTS! PROJECT TECHNOOGY AREA ' uANTITlY tilT COST ASSWEO IWACTS Of TECHNOOGIES ERR REMAS -- --- - -- -- - - -- - - - - - - - -- ............. . ... .... .. . ............... -- - --- - - - ... ... <......... ................. -- - - - - - - - - - - - - - - - - - - 4317-111 INDIA HIALAYANM WAERSHED 1963 M -T. UTTAR PRADESH -GRAS -- 23S REPORTS REDUCTION OF SOIL L055ES FUEL00D * TIUER FUELUOD PLANTATION u.5-3 ton/bh FRM 30.40 tn/h/yr TO 1-3 tV~yl PLANTATIONS - NIXED 59,000 ha USS 274/ho(Govt) R yWr 1.6,11-18 111 IN INDIA FOLLOWIING INLEMENTlINE SPECIES USS 379/ha(CIv) FODOER PLANTATION * 2-4 tnVha SOIL flNSERVATION MEASURES OSINILAr a ye 1-20 TO THOSE PROPOSED. NO CITATION FODDER TREE 27,80 ha USS 1S8/ha(Civ) -FUELWOOD -- GIVEN FOR THIS FICUl. P9UIM tonS WS Ufihatriv) MEVXO PKOMI MIO % n-LI tos; a yr - II 20 - domnot eqmt 6OW h of FUELWOWD & FODDER 81,000 ha Il/A FODOER PLANTATION - 18 ton/ha terraces, rether 6000 ha of PLANTATIONS a yr 10 Ł 20 tend Will have terra-e FARM FORESTRY a 5 tan/ha a yr cantructed an it. BRUSH &STONE 1.650 uss Ile.. 101a20 CHECK DAMS -FODDER -- FUELWOOD PLMNTATION-.l- .S ton/ha CRATE WIRE DOAS 1,300 US 316 ca. a ye- 3-20 FODDER PLAITATION * 1-S taVha DROP STRUCTURES 700 ust 1,053 ea. a yrs 3-20 -SOIL CONSERVAT ION/SEDINENTAT ION TERRACES 6.000 h * USS 526/ha KENEFITS CATTLE EXCHANGE PROGRAH 10,500 ha US$ 368/ha ..*.** ...-***.** .... ... *,**,..*** ..**** **e.,.ee.e.**e*.,*,,,, ....a.a....,...... ,*.*... *. ..e..... *..e..o.....a..* *...aaaaoa......e.*.*.... *.--t -e.e.....oee.ee*.e 4.*. *561-IN INDIA PILOT PROJECT FOR 1983 WATERSHED DEVELOPMENT -AT YEAR 13 46S IN RAINFED AREAS FUELWtD - 194.000 d3 DRAINS AND 300 km US 15-21/ha FODDER - 130,000 tot REPORTS THAT IWRK PERFORMED a WATERWAYS ICRISAT (6 YR. STUDY) PARTITIONaED -SOIL CONSERVATION/MOISTURE RAINFALL: ON A FALLOW -- GRADED BIUNDS/IERRACES 140,000 ha USW 41-51/ha BENEFITS 25 - UIOFF,252 - EVAPOPATION (SOIL 9X - DEEP PERCATION LAND SHAPING 10,500 ha USS 41-51/ha 41S - MELD S SOIL hOISTURE ' ON CROPPED FIELD - - 1S% - 51110F. LAND SWOTHING 60,000 ha US$ 15-26/ha SOIL ORDER * VERTISOL (NIGO SHRINK/ SllELL h01TllILLIHITIC CLAY) REHA. EXISTING SOIL 30,000 ha N/A CORSER. STRUCTURES BRUSHUKX Ł STONE 10,000 H/A IMS a FIELD 81ERDIES LITTLE CHECK OAKS EFFECT MONENT OR RETENTION OF WATER. CROPS MAY SUFFER FROM WATER WIRE CRATE DAMS 400 H/A STAGNATION.UID GM REACHES- FLASH FLOODING. SOLUTION a DRAINS TO DROP STRUCTURES 100 H/A WATERWAYS & GRADED fMrS (DEPEflD. ON SOIL TEXT. MWD PERMEABILITY) SOWING FODDER LEGUIES 45,000 ha USS 16/ha DISTRIBUTE SEEDLINGS 20,000,000 No/A TO INDIVIDUALS Table 2.1 SMUARY OF EORLD BANK PROJECTS ItEPORT C COUTRY PROJECT TECHNOLOGY AREA \ QUANTITY tNITl COST ASS N ED ISPACTS OF TECHNOLOGIES ERR REMRKS -- - - - ---------- ------ ------ ------ -- -- -- - -- -- -- - ....... .... ..... ...... ................. - - - - - - - - - - - - - - - - ............... ................... ........ .......... .. FARM FORESTRY 30X OF FALLOUED USS 115/ha EPOIRT IS PRELINMARY REPORT AM LANDS CONTAINS K0 USEFLIL QANTITATIVE INFO REFORESTATION 30 X OF EXISTINCG US 344/ha ON IMPACTS Of PROPOSED TECtLOGIES FOREST RESERVE .. .* ..a.., *-**......a .. ....... ....... .. *. .a... ... ..... ..... ..a,..... ........ . ........ *.a,a..*........*. * .*.*..***. **.*..*... *O.. ** *a******.*** *............ *...* ..* *.*******. 643-cNA CHINA GANSU PROVINCIAL DEVELOPMENT PROJ. AGRI. CttPOENT BENCH TERRACES 13,500 ha USL 594/ha EPORT ONTAINS ND QUATITATIVE INFO TERRACES 21.000 ha USS 216/ha LAND S5100THING (mini.) 40,000 ha USS 27/ha LAND StOOTHING (mech.) 10.000 ha LISS 270/ha PLANT GRASS/LEGUtES 40.000 h LUSS 54/ha GULLY PLUGS 2.400 km USS 810/km ****,,*O.......................**. **O***O***********a,* *...*.....................................................................*...*.....................*..*.........*...*..........*..*..*..*..*. . ...*_ ..... 3316.-TUt TUNISIA UORlTNEST RtRAL 1961 DEVELOPMENT - WPRISA. EASED ON PILOT PROJECItl PHASE I (SEDJENAI RNE MtOtNT.) AND SOIL b' REFORESTATION/WOODLOTS 22,080 ha USS 1,657/ha -FORESTRY PROlUCTtION IKCL. RURAL 16X CONS./FORESTRY PROJECT (Etj^ PJICJ ENERCY SUPPLY - US. 5.9N/YR ONLY INFO GIVEN IS tUAW PRlJcltS gtm PASTURE(IMPROVE./EST.) 72.64 ha USS 59,S00,000 C FULL DEVELOPEENT SUCCESSFUL" CONTOUR BlltS 3.300 ha (TOTAL FOR ALL -SOIL CNSERVATION TO REDUCE OR CONTOUR GUIDELINES 21,580 h PASTURE & SOIL ARREST EROSION Ł DECLINE IN GULLY STRUCTURES 1,740 ha CONSERVATION) AGRICULTURAL PRDDUCTION -REDUCE SEDIMENTATION IN INTRO. NEW AG. TECH. 41,550 ha N/A RESERVOIRS R- IEPOT DOES NOT CONSIDERRD PENAB.UDER SOIL CERVATION. ROAD IMPROVENENT +. 1.940 km USS 7,216/km SNOL KE CONSIDERED AS EWEFIT IF **-**-- --***a* ----..*-.-----,. *e*..**............- **...,,......... --.....a......a... --...a.a,..... **O.*O-**.*.a,e*.*.aa a.. 4831-BEu BNUTAN FORESTRY DEVELOPMENT 1964 REPORT STATES TEAT PR JECT AREA CONSERVATION IBRKS WHERE NECESSARY EST. EXPENDITURE COPRISES AtEA OF: "EASILY ERWIELE CHECK DANS,GULLY PLUGS USS 105,000 SOILS. GENERALLY STEEP SLOPES". HEAVY PRECIPITATION" (3.950 =03 ft PROTECTION PLANTATIONS - 4.870 wA16 ft) per year). SEVERE ON SLOPES > 60X; 700 ha N/A ERIOSIN AM SEDIMENTATION NAZARD AS ROTATION = 60 YEARS RESULT OF LAN CLEARING. LOGGING, AN PLANTATION ESTABLISHMENT AND TIGT. PRACTICES NOT CONSIDERED. REPORT CONTAINS NO QUNTITATIVE INFO 2336-ltD INDONESIA YOGYAKARTA RURAL Table 2.1 SUUARY OF NaRLD BANK PROJECTS REPORT 4 COUNTRY PROJECT TECNIIOGOY AREA N LBANTITY RWIIT COST ASSURED IMPACTS OF TECHNOOGIES ERR REMS ............ . ----------- .......-------------....-- . ---...........----------..--------..-----.-.. -----....--------- ------ ----- ------ ----- ----- .. .... ------------------------------------- 1979 DEVELOPMENT -REDUCE SILTATION RATES IN 14S + . - COSTS ASED Ol MEAN SlEOf OF GULLY PLUGGING,DRAIS, OORIMSTREAM IRRIGATION WORKS 13X 25S FOR DRTIIND AND 15S FOR LULAD TERRACE IMPROVEMENT USS 48/ha -ARREST SOIL EROSION. DECLINE TERRACES. COST FIOURE ICLES (NNECGARDENS) IN SOIL FERTILITY, ASD CHNAGES CONSIRSCTeION AMI/OR RENAIILITATION IN STREANFLOWS OF TERRACE DRAINS AND WiATE ATS DRYLAND TERRACES + -90S PER CAPITA INCOME INCREASE (8ENCN-TYPE.RAINFED) FOR PARTICIPAMTS IN SILVIPASTURE C- OSTS IICWUDE PLANTATION NEW TERRRACES 390 ha USS 355/ha -20S PER CAPITA INCOMlE INCREASE MAINTENIINCE TILL YEAR 5. PLANTATION RENAB. EXISTING 740 ha USS 192/h. FOR PARTICIPANTS 1N TERRACING ESTABLISIET COST * Rp. 74,OOC/ha -INCREASES IN AGRICULTURAL LaWLAND TERRACES PRODUCTION OUE TO TERRACING COSTS INCLUDE MAINTENANE TILI NEW TERRACES 100 ha USS 82/ha CASSAVA - S YEAR 5. ESTAULISIESUT COST - SNEET POTATO - 400X * Rp. 145.000/ha REFORESTATION 350 ha USS 149/ha * MAIZE - 2002 DRYLAND RICE - 300S PECT CONTAIIED RESEARCN AND SILVI-PASTURE SYSTEMS 350 ha USS 262/ha GROUDIIUTS - 150S DEMONSTRATION PLOT CGWOMOEhT. *.......* ***...*-**. ...*.*.**-***....*.o. ......**.*.a.****a.4..* ..4....4........ *....aa....a**-. *-*-.....-**-**.** .1*-** .*aa..,* a.... * -**.f ......**-* **. 4773-PAR PAKISTAN INTEGCRATED HILL 194 FARMING DEVELOPMENT -DOUBLE AGRICtIETtAL PRaOUCTIII 22S AFFORESTATION 5,680 ha *. - SEE REARKS -FUELUD0D -- 50,000 d3 S yr 10 4 - COSTS MT ITEMIZED. TOTAL 340,000 d1 a yr 15 COSTS FOR ALL PLANTATIONS AND SOlt FuELUDOC PLANTATION 3,800 ha C lSERVATIO WM = USS 2,600.000 DISTRIBUTE SEEOLINGS 12,000,000 REPORT GIVES HD QANTITAIIVE IRV TO FARMERS CNECK DANS NOT SPECIFIED P-2139- PAKISTAN MILL FARMING PAK ECHIICAL DEVELOPMENT -"DEVELOPMENT OF OIH AN POECT ACTIVITIES INCLUDED STaRT ItP 1977 NO PARIICULaR INPROVED TECHNIOGY IN FORESTRY OF 4 EXPERIMENTAL/DEIONSTRATION TECHNOLOGIES PROPOSED AND A SYSTEN OF USER FARSS. RICIENBARLE TO CNECK FOR FOR SOIL CONSERVATION PARTICIPATION IN ESTAtLISNING RESUILTS SUSTAINED FtELEOD SUPLIES,WILL PROVIDE A BASIS FUR FUTURE REPORT CONTAINS uO WANTITATIVE INFO LARGE-SCALE REFORESTATION PROG. I .O PREVENT FURTHER DEGRaD. OF STATE'S FORESTS.SOILEWATER.. 2174.-IN INDIA KANDI WATERSHED AND 1980 AREA DEVELOPMENT -INCREASED PRODUCTION OF FODOER 12S - COST IS DIRECT COST ONLY VEGETATIVE S STONE -INCREASED INFILTRATION THUS CHECK DAMS/CHANNEL INCREASED GROUNDUATER,DECREASED REPORT liTAINS IO UATITAtIVE INFO GRADE STABILIZERS/ 24,000 ha USS 42/ha 4. FLOWOING,DECREASED SEDIMENT DROP STRUCTURES/CRATE -TIMBER -- 320,000 . 8 FULL WIRE DAMS/DEBRIS BASIN DEVELOPMENT -FODDER -- 11,000 tons a FULL SILVI-PASTURE 18,250 ha USS 188/h. DEVELOPMENT Table 2.1 SUI1RY OF WORI RA PROECTS OMY 9 BRUTE? PROJCT TEBWmoOGY AREA . GMITITY MIT1 COST ASSUIJED IMPACTS OF TECNNOLOGIES ERR RENARKS LAW LEWELLIUG/ TERRACES/IMTER 3.700 ha US5$ 214/l.a LMARSTIlUG STRULTIMS 1E1CN TERRACES S00 ha INS 536/ha LAND LEVILLING 3.000 ba USI$ 262db. GALT RECGAMT? 3 SASINS/CRATE WINE 1,100 km us5$ 357/ha OANS/DROP STRUCTURES/ TRtEE II SIll P1.1T11G 2269-TI TMIAEL EUTRER AGICIOLTIMA 1979 ~~~OwLaPPIEMr -R&UCE 1*6EA US6 FOR SMIFfINS 131 RtEPORT corAlsINS W UMTIrTATVE INFO VILLAGE WENDLOTS 7,150 ha 115$ 252/ha CtILTIVATION BY 26,000 ha -PROTECT 138,000 ha OF WATERtSNED KOOM TERRACES 2.600 ha NOT SPECIFIED Flag MNIMMG & DESTRUCTIWE 031T11MG. RW IPWOVEPIENT 500 km 1155 1Z350/ba -LOW1 DEGRADATION OF UOIL.URTER. AND FOREST RESS*CES FORST PLANTATIOR 7.600 ha 115$ 193/ha -SOIL CONSERVATION -FUELWIOOD--II2,OOO .3/yr a yr 10 -POLE$--?7.Om.3/yr 3 yr 10 V -SAIR.OGS--6,100 al/yr 3 yr 10 - 609-ET ETUIOPIA FORSESTY 1906 FOREST PLANTATION 11,000 ha 115$ 1,909/ha -INCREMENTAL PR0OtUCTION OF 161 4 INCLWDES COSTS Of COWSTRITIMG W1*1000.3a lima15 1* PEW. 210 KM OF ACCESS AND FEEDER 380S INmOADE EXISTING 13.000 ha US5$ 1.107/ha a. TR 12, MEAN A1111 MMW. Of FOREST PLANTATION 577,000.03 UIASSS IN YEARtS -INCLISES COSTS OF COISTUMTIMG 6011.0111G. AtELVMW CRWOMT Of 90 KM OF ACCESS AND FEEDER MM0 CONIUITT FORESTRY 3301N35 MPAIS? a 16-241 OF PL.ANTATIONS 1.600 ha FUELMOOD CEIS OF PRJECT AREA, REE1OT CONTAINS NO BMMTITATIVE INFO ALLEVIATINMG SOl PRtESUME FRO UPGRAE EXISTING RENINIMS FOREST LADISS. cm.m1I rT FORtESTRT 1.000 ha -SOIL CrmSERVATION,S1WPE PIAMTATIONS STAMILIZATION.NICRO-CLIMAT[C. RESTORATIOR 09O SOIL FERTILITY, SELF-SIP FORESTRT 8,500 ha -PROTECT 01FlUE AZING 9 PLANTATIONS TRAMPL.ING WILL RESUILT IN UI SSI 400/ha INCREASED INFILTRATION russ UPGRADE EXISTINMG 1PIIFRNINS WATER SUPPLY? & SELF-NELP fORtESTRY 2.700 ha AGRICULTURAL PRWUCTION. PLANTATIONS FAI/VUESTEAD PLANINIGS 210 ha Table 2.1 SaO*av OF WELD BM PROJECTS REPORIT CMEtTRY PROJECT TECNOLOGY AREA \ TITl UNll COSt AStUMD INPACIS OF TECINIOGIES ERR REMAKS .......... ........... ---- --- --- --- --- -------------------- ....... .......... ---, ......... ... . ,,..................... ................... .. .... ...... . .......... ... ... .... ..... .... ...... ... . INSTITUTIONAL ha -. PLANTINGS ....*.o* .... **.** .*.*...*.**. *....a.****. ***.****,*.*.*.*...* *******.****.** .** ..*....*...*.e.......*.. *.*e*ee*.*-.*.*e **-****** *****t**O*O***-*O*O*,**h********** ** ************* 6012-CNA CHINA RED SOILS AREA 1986 DEVELOPMENT - E0 VIH-TI A ER CO S EFORtt CONTAINS GUTITTIV INFO TERRACING -I E ENlthEWtA DIt S BROAD & BENCH -REDUCE SURFACE RtLOPF -FUELU0DI SSUPY CONTOLUR PLANTING -DEVELOP SOIL COXSERVATION NAPPOCES' AFFORESTAT ION 27,000/ha USS 884/ha DROP STRUCTURES CHECK DANS DRAINS P.42%- LESOTHO LESOTHO HIGHLANPS LSO UATER ENGINEERING N/A N/A REOT CONCERNED ONLY I TN DAN 1986 SEE REMARKS C--STRUCTION WORKS; DOES NOV CONTAIN SOIL CONSERVATION C caeO DESPITE OBVIOAS CAUSEIEIPECI HAIUN I OF EROSION/SEDIMENTATION A &ISTBW)1N%A LIFE, CHANNEL AGARDATION I& UtAVV o CONSTRUCTION 1I FLODD PLAIN. ..*.*.. ***.****** *eee**** * ... ..... ....*Oe* ***.*OOO,***,***O.**.* .*.*..*...,... *.*****.* ..... ****.** .... ***..****..e...*.. *.*.. ,.........-.*--****- ............... ---.... .*-. P-3106-CO COLOMBIA UPPER MAG6ALEMA N/A 1981 PILOT WATERSHED MGMT. -PROJECT WILL SELECT AND REfINE tEPORT CMCERNS WIHN SETTIM UW SEE REMRtKS TECHNICAL MENS AND MA EhENT RtESEARCN AS DEVELOPNENT PM TO SKILLS FOR PUlSEE MATERIE'D IDENSifY WPIt l I&E EPfORESIAIION PROTECTION PROGRAMS SECIES, SOIL CONSERVATION AND SUSTAINABE AGRICULtURE TECNIIES FOR UPPER MAGDALENA RIVER BASIN. PtOJECT TO E CURRIED OUT BY INDERENA OVER 4 YEARS. PROECT SNOlWt NAVE cOIETED, CNECK FOR RESULTS. *-*.. *. .***. ........ . *.... ......... *** ***.*.*O *.O..........~**. * ***e**** **** * .*.e. ..* ......*... *. .*e.* ** *.* ***e .*e** ........ * *********O**** ...... ... **CC *C*.*O****O*O*C**** ..........*S*@**** *.**** 4501A:6/ INDONESIA UPLAND AGRICULTUIRE 13/84 AND CONSERVATION -INCREASED PRDUtTION Of AG. (IRR) CILUTUII IATERSNED IIEST JAVA - 5291-IND REMOVAL OF SILT FOREST AND GRASS DUE TO 12X SEDIIENT LOSS INCREASED FR-lMtyr 1964 FROM SEDIMENT TRAPS UsS 1.40L SILT SOIL/MATER CO4ERVATION IMPACTS IN 1911 > 2_ 1935 > 1m 1960.s ON 1. PREVENTING LOSS Of SOIL (E: Sb FlY et al in Wt . Svs. for IUJTRIEITS AND DEGRADATION Of Develop.. R. Carpenter,.d.. 1963) SOIL, 2. PREVENTING LOSS Of SOIL MOISTURE 110DING CAPACITY, EST. OF SOIL LOSS OR 16 SITES (see 3. BETTER MOT. PRACTICES table 4, p.16 of rep4rt) RAWGE FROM Table 2.1 naima Of uiNw Mam uoARcvS *EPI B mum1 * CT TECtIOTOGY AMA %WAf1tT 133? 005 Ass119 wAS Of tECNosIES M mAiS (e.g. terraces). 0.24 - 10.6 .w. SoIL FllII -REUCED EXTERM coTS OF RATE EST. a 2.4 _yr. (erte: this OOTIEAN SSDIKuTAITIO a i i eceedingly big. rate). FLOWING. -PRESERVATION OF OTI1 FO A SEU TAKE 1.2.3 p.4-S5.E I FM UICEMAIX FUt_E. SAE FACTOS. SEf a 13,V.I., SOL OMSEiVATIO P0 05TAIV AET, for the SOIL BESEJOCN IST. OF IN10., 1950,1981,1962 OMT. COCEU VITI EI n UP CT OFFICE FOR EMTS. 43SI-mEp *PAL 0ECO FRESPRY 1963 ClUSIVY REFOREST. 2,000 he US 27/ha 40 v" - aE POT CONTAIN 0 sWTITATIVE I10 FM 00 - 6.4 N ml, POLES - PItVATE REFOREST. 12,500 ha IM 104/ha I, N Fttins - 1.4 N as, mom S 1.7 is at. MIT STRIP PLATUtAIS - .3 N mt. (roodcuts,rlwr/canai 2,SSO ha -SAVW 6;900 d3/yr FUE VIIITX liii" tOlt 2,2CO hr |-SOIL COM AVAtI0PbTECTIO SEIUSSSIAIION uinots 2.200 ha USUtAL lUTES amyWPlTIoo- AUO-FUESTEY PLOITS 5.900 ha cS 9S4/he CaINllC,Sl FETliTY IEEFIS FUEL IFFICiEwT 5w0c 1IIIG STOVES 20,000 US$ 12/stw. 3628-r oCC NIO0L1 ATLAS/CETIRAL 195i hal. amte SLVI PAI.r-am 05 TR l-* 30.M la 3M 24% ROM CSIAUS as UM3IIATIVE IWO SYLVI-PA IURE (r_no OUK 900;e,=bZ dt F lt WI I lUl 1t tftorags plantinP, 45,100 ha Mi 69/ha -vER 36 d3 MWAVE 7 0.M fortl i zatlon,tree CUM4, 15,000 .3 fIILvOO. tbimfngsl -TEARS 30-34 --PbsS Mm.0 PIE T33155. -INCEAS FORMK T51L5 FOREST/ AMSE T6 OECIEASSI LIVWtOCK DARM 5676-rn t co FORESiRi i962 RI5AS. EXISTING -U20 190 NC0 yr FOP PuLPWOOD 25 _T CUSIIS as EIStAE .TATIW IIFO FOREST PLATATlONS 20.000 be 9fJcUos. 470,000. assu US$ 410/ha2 FOR SMTRY - VALUE v U36555 I REFQESTATION 10,000 ha JPP9EV IX (720,0 0) oft A PASTURE Il_ EIITS 2,600 ha UI 3/h FUEe KEEN oF anT. Table 2.1 _M Of OF sam U NsAcS no maui Imct ucmoy ma taAmilwT MIT cmt AURE wctn OF ua tl0WEus m f - PDSTE P ESIlAET mli!30hCOW a1 _ 1 1s -- Lna * LE lS OF DIER Lfu. CEORNE- "'Am immnacis 10 kS>r / 30r rotatfml. - 3f13 nelEU *EIES *IS8ON 11 he -PInu coaii-o MlLOATDU,0LU1105 CdC O CIIECi 1I 85,D/ IO |IR O -_ hlEslt IMDES Of SOILuEmA IOU FIIUCEai 1 ,151 a * 1 um LISE. u t CAanm MO ItNasi;vs 1 EFORESTA I1 6OtD4JSIFOUU TREEIS/ 7.r50 km FIIU1 TIEES 1952 IDESIE - .gpy PLsO nS if o45 1 N T UTAIn N MlIATIVE INFO VILLAE 6or0s s. he0 use 44 - 471/h P58 (1 3/pmraty) NMRILLT. PASUCE 416,000 013/Yr sEUSII.EFORESTiTIN 1tvOmO 11 1S 494 - 529S/he -PROp50E 465.00.1D UWLL Tirna a 2.0 a LUDI POE3/y. F O FNESTRY 19t.= be us 71 - 106/h -SOIL ONSVATION.1LUL/SA) hKM STABILIZATIO11,11CR-CLINAIE ETLAW PfMTISNS 2,600 km 81 4711fes AND SOIL FERTILITY SIEFITS STRIP PUIINGS 1,mk hUS 5al m VILLAE w0tOTs 12.0 km SS 4109 - 527/ha AFFOESTAT ICI: ALKALI LAWS So km LESS 66/ao Sam 31S 1S,60 km M 3761k FPA FOaEnnlm 30,000 km -.5$ INS 106b/. STRIP PLANTINS 9,50 hk US 612 - 624/I. Slb- IN I3IA NTIONAL SOCIAL *9ES AMESTIY I-EUCE SOIL ERSION 21X *D OST BES SO IE E FAIM F(STFT 46.000 km 85 16/k 4.b -INCREASE TI83CR SU!PP1. EVLOPENT. AVERA COS/IECtAcE V TNOUT 335Emw OEMLONT CR01' TREE TEIIRE PLANTING 4.000 km USS 3.121/h US5S lW/a; WITN mASERi COST UESS 253/ha. COmaItY FOREST 95.0O hk u5$ 15h - SS 2TOAh S ASTELAI PLAITATIONS 77,940 ko USS 650/ha Table 2.1 OF Of 10LD UlK PROJECTS NET 8 CMRT POJECT TECW40LO6t AREA UMITITY UNIT ST ASSUED INACTS Of TECOMOGIES EN RENAS REPORT CONAINOS G UIITITATIVE INFO fUEiOED EFFICIENT STOVES/CRENATONIA N/A US 553.700 .--*.*-**---..... ..m...*.........e..,..... *........e......... .. --*.* ..**.*.... -***a..*4---- ... *-* . ****_.....*.....*....*,..*.* ..*..*.t.*.*. -.*..*......*.....,* 6424 EPAL REVIEW OF Al. 1984 DEVELOPIENT SEE HENIEtS N/A V/A N/A N/A E1T IS A FINAL PROJECT P'RtOUACE REPOR. 3 PERTINET PIECES OF INFO IN KEPORT 1. EIOSION CNTS n rl TOO IN FaE OF 1wU of t 2. POOOv ChTWCTEN PROJCT 0DS MTED IlAOSl USES 3. LACK UIESTAWIIUS OF FA_U'S DECISION-116 PRESS RESULIO It on FP8ISIIIWA CIS fRO ERISION CTM. & FORESTRY OEVLOPN CtlElt OF fuO.CI ~ **. -*..*.**.***..** ****-****-**O**** *O****. -*-**.*t**-*...,a....t,a .ea,a*oe*..-.,e . 4642-TAR 1EN CtNTRAL NI1SANDS MIA U/A N/A N/A 116 AWICIULTURAL DEVELOP. REPORT US MO SOIL RVATION SEE RENARKS WENT. E OMTAT POECT WISHES TO tNUllE IIRDITIONAL AGICIOtUALt NE XO.0GY W/0 ILUDISNG SOIt I CONSEATION EXTENSION CINPOREI. PSOECT IS AN UPLAND PSOJECI 0. --.S*S t*S.S i*-**-**SS~ ****-***S.****Sh*-***. *-*-*-*-*SS*-*.* ****S.**** ***S**.St**-e*O* **S***z*-*- *- *----*--*--* - - - - -.- t 2920-Pf PILIPPIS WAIEU NUT. AMD 116 EROION CONTRX -Ea)tRl OF ERC SION tax EROSION CONTROM STIlY IINC. LOANS REFORESTATION (aure- -LIllTING rm IUCT .LSED ReING 1222-PI. 1976. w ClECK FOR RESUtS forestry/tim ercrop) 32.100 ha t1SS M/ha -NININIZE fLASN FLOODS -EIINACE PNUCTIVITY REPORT COITAIMs D MO AITITATIVE INFO FOREST POTECTION/ -REDUCTION OF UDINENATION fiRE PEVIENTION -LEAFISA *- I tmu/ha/pi- OM rela .,soil 155,000 ha USS 2/ha FROM TEA S ON orles,an cntrwl) - tFUELO -- LEMAUENA - 301 mi/lh YEAR 10 ON FOREST ROAD NINT. 410 ha US 3,170/kb CASURINA - 14 h/h YEAR 10 OR CN -CA 1001 INCREASE -PULP/TilIER YEAINE - 67o.3/h TOTAL MM TiAR 7-15I, REUSr PINE - 1t- 32 ml/he TOTAL FROM YEARS 13-25 HARM & NAIaw - 787 r 3or he TOTAL YEAS 20-40 4369b-NA HAITI SECON !Ul *EVELOP.- - * a e . .. ***5- ********t***S** ** *S**SSOS..S- Ff63 . rPRCF (I mN u0t- -ENEFITS OCEIVED Of INI -TEUS 23 TECIIOLOIES OUER A fEXTENSIONSOIL Table 2. 1 amSUU OF URtRD ANK PJECTS REuPON) I C rT PRWJCT tfECNoT ARA k tWA ITY UNIT COST ASUIED INDACTS OF TECNOOGIES EIn MAKS .......... - ----- ......_. ............ .......,. ........ .... .......... ..... --------- ..... _........ .. .........-... - ~ ... ... .............. ...... .. ---- M. xTENSION/SOIL OF INCREASED FAtILY ICIXE COSERVA IONU ARE T*IOSE NIICR CONSERVArlON (cantow VITN INCREASE ll COWO YIELDS PIEvEYafS EXPERIEIICE FN 10 SŁ strips,hodges,strip 9,000 he Us5 166/ha OF 332 10 152 WOSJ PALAIABLE 10 SILL FARIEUS. SOll cr1,1 entour bis, CONS1ERYVATI STRUCTURES (e.g. muIching terrsces) WERE THE LEAST ACCEPtALf. lROD AlITilEANCE 201 kw N/A ltoo CONTAINS MN) NItAytvE I[fa .*.aa. *****e** *A* .A.**14........ *W4***.*.*.**.* .........*. ..****O .... ****.***~4 *.** ........ ** *.**~***4* ** **e* . **........ *e* *** . *,**.** .*h.h,.***. ..** ....** . * ****.*.* **.. 35B-PA P tARAM CAAPA ARE DEVELOP. 19$Z SOIL CONSERVATION -UENFITS CO tCEIVED OF H TERMS 20S REPOT CONTAINS NO QUANtITATIVE INFO (conmtw tittsge.orwas OF INCEASED FAMI"ItTAX atrips/uatertns, 6,750 ha US$ W/he INCOME tO STATE AIn A 20X protection planting) (tabSEco) TO 602 (sugrcuu) INCEASE IN CROP YIELDS FOPEST NAEMENT 130.000 he USS I5/ha ROAOSIMAINTENANCE 270 kw t5 62,592/ha .e.a ...e.*,e*e*.e..*.* **-**e***..*,ea..*.. *****.a...*.*e.e* .a*.a.*a*ee **.e*e.*eoea.**.*,a.**. a.e*. *@*********,*,*h**W*,,***.hg*,*********.****.*.***.****.-.* * 3776.-NA 4Atti FORESTRY 1912 SPECIES TRIALS/PILOT -WOW -- 600,000 03 a CHARCOAL, Nl/A REPORT CONTAINS NO LAUTITATIVE ItFO FUELWOW PLANltATIONS 200 ha US$ 2,M/h. POLES,LUISE2 OWING I4 tEAt WlLEMENTAItION PU#S, FOREST AXAGMENT 32,000 he us$ 30/ha SUSTAINABLE TINSES YIELD FROM YEARt 10 WALtED 1 US$ 18.5 DEVELOP FtEL EFFICIENT N/A US 117,000 ILLION/TEAR. cOm StOvES FUELWOW VAUED & U15 250 OF KEROSENE/NA. -"SOIL CONSERVAtIIN *1AD ERtOSIN CONTROL MILL INIOVR 11TH THlE PIOJECT AND DETERIORATEt IGlNIFICANTLY VITNCU IT* * _.ee **.e**.. **a*e.**.**********.* * *he**.,a.*.ee..eaa e***ee*.e*a.a..e..C.. ,.**3***@***.***... 2473-SO SBOIVIA OMASUYOS-LOS ANIDES 199 RURAL DEVELOPMENT -*REFORESTATION OF PtXLIC AIlM 361 REPORt CONTAINS NO aUllIIATIVE INFO FORESTY 600 he UISS 497/ha PtIVATE LAWDS (WILL) COiAtRIITE TO NtTER SOIL COSERVATION AIM ROADS REDUCE TIE PIACTICE OF EURING IIwOVmENEN 50* km U$ 3,628 km 0161 TERIEBY FACILITATING MAINTENACE SO km USS 3,628 kw NATURAL NITROGEN FERTILIZATION" 2e*a ..ePe. IIP **W**e*.g.r..* *e.aa..ea2 ,. *e.**...*.. *edei***Ca.ae**.'*ee..-*e*....... ****' *'***i*****C****'--'*C*-***t*---*-** 2"301EP NEPAL COIUSRil l FORESTRY 1980 DEVELOPEN/TRAINING -F.ELO0 *.t POIDE 16 ItEPORT CONTAINS NO QGUAtITATIVE INFO VILLAGE FORESTY REUIREMENTS ('1 t3perSSt5Vre3ul) C(DSRJITY FORESTS If, 730 ha U15$ 69/ha OF 190.000 PEOPLE (1/3 of CoNREITY PROTECTION target *r#e. FORESTS 39. 100 ha U5$ 10/h. -FOODER -- FORt 132,000 CATft.E PRIVATE PIANTINGS 900,000 trotsa US$ 0.01/gres (330.000 Iinws/yser). INCREASE Table 2.1 aUsuwU OF VJL RtD POECTS ftNT S CITRY PROJECT TECNLOGY AREA k ou.DUTrTV LUIT COST ASSUED IMPACTS Of TECNNOCIES EN REMARKS ,,........ ............. ............._. ... .... ----------------------,,,,,, ,,,, ............... .. ... .,, , _. ........... ............ .,,,........... __........ --- .................. ------------------- fOW P#UCUTION BY SUSSTITUTING IUIISE1/FORESTRY FLMLkD FOR OUIG/CROP RESIDUES OEVEIOPNENT USS 3 28W0.000 (qusivlent 156.000 tam wIze at peak Year). SAVE 25,000 IIiMWNED STOVES 15.000 iSt 16/stove - SAVE 25,000 t2amlyear FUECVOW WITH STOVES -SOIL AIID IOIStURE CONSERVATION BENEF17S. *-*...*.. *********** *********.****,.,*.*** ***.*******,*********O* **.***@e********A ****-*****e****** ****.**.*C******.****.*.O******** *.*.* ***-*******,*,*,********************* 2695-PI1 NILIPPIVES RAINFEI AGRICltULUAL 1960 DEVELOPNGNT (IOIOlt) N/A K/A .+ - SD0l CONSERVATION PORtION Of SOIt CONSERVATION z M PROJECT IS A PILOT PROGRM REFORESTATION COVER CROPPING REPORT CONTAINS NO aLJATITATIVE INFO COCTCUR HEDGES 400/ha USS 135/ha DIVERSIDO DITCHES rIREtINES ********* *****O**** **.**@*********** ....****..4.***.**O*.C***.** **.**..*O***.* .. *********~**,** e....e...e.ee...*ee......*e* ***e *****e***,***S******* .................. **. .***4** ***~****..* **-- 6197 KfEA REVIEW Of THE 1966 111H0 WATERSHED AREA N/A N/A * W - WORKS TO BE PRIMARILY BENCN DEVELOIfENT/VONG CA TERRACES, WHAT OtIlER TECHNO1OGIfS GAUG IRRIGATION WILL BE USED Is 1501 SPECIfIED PROJECt STAGE II ItVAIN RECtAMATION :*4 SENCh TERRlCES 1,200/he USS 2,417/he * 1976 DOLLARS, REPCRI IS A 'OR PROJECT C(JMPLEtION REPORt. iUtNS ALLOCATED 1976. UPLAND RECLAMATION COIEHNNI WAS NOOT CARRIED OUT. REPORT CONTAINS tlO 11ANTITATIVE IIFO 3925 KOREA REVIEI Of RMU 1962 UINFASESIICTURE FUELEOD 127.000 hI USt 714/he.. -MEAN ANIML EMLWD YIELD OF 163 ^ . - COST ICtLUDES ESTALISHNENT. 5 tans/he/yr a 635.000 tons/yr IPSLAD RECLAt AION 4,4eo he USS 3.233/ha FOR 10 -5 Ti EARS. 19 D A MNAINTENANCE/1ARVESTING UNTIL -RESIEOINIG SOIL FERTILITY TEAR 21 -REDUCED EROSION -REDUCED FLASN FLOODS U*5 6X a FUELWO0D 193 = UPLAND REC. GREATER AG. tUC(NTION -AMENITY VALUE OF FORESTLAND REPORT CONTAINS NO tlANI fIIfE 11- ***4***** ***** *****^**************- *************.**e***** *e.ae*,.a*.*e*.e *****4***ae*eb** ****e*...e**e.*e*a*********.**** -**** ***.-****.** **** ****i **-4C --**--**--- 95&-lEP NEPAL RIAL DEVELOPMENT 1976 180013 TREES 1 050 he :REDUCE EROSION/SEDIMENTATION 22X FOR EROSION CONT1RO *REDUCE PRtESUJR ON EXISTING REGEIIEIRAE VGETATION 1.000 he FOREST REPT CONTAINS ND QNTITATIVE INFO H | -REDUCE AGRICOtTtAL USAG Of Table 2.1 SINAUT OF UORLD BANK PRWECTS REPORIT calmRv PROJECT TECHNLOGY AREA N 1NTITY tIT COST ASS UIED IMPACTS Of TECHNOOGIES ERR RENAKS FOREST PROTECTION 5000 he | | # 49/ha RGINAL LANDS FUELID 1,050 hp SLOPE STABILIZA11 IO TREE PLANTINGS 450 ha EROSION CONTROL UNSECIFIED tUSS 846 000 TRACK/TRAIL DEVELOP 9 PROTECT 214 km USS 5,140/km *-* *--.-.*.. *-**---*-******.*.* ....................... ..... *..*....*.. *.***-**.*.*a ...a............ *.* *~***... *..... **.**-**-*aa*-*.a*.*a****** **.*.@**O.* ****O***-*O*-***. ******* 702INID INDONESIA fCRESTRY IsItUlTUTS 1907 AND CONSERVATION -INCREASE CROPPING INTENSITY 271 22X W A - TERSHED CONSEEVATION SOIL CONSERVATION -INCREASE YIELDS OF UPLAID RICE. RENABILITATION OF NAME.& CASSAVA 70 - O0; EXISTING TfRRACES 22.000 ha NAIZE/GROUNDRWT 90 - 130X -SILT EXTRAFMENT BY CHECK DMS GULLY HEAD WORKS 250 WILL PROVIDE CtLTIVATABLE TERRACE IN 3-7 YRS., GULLY PLUGS (EARTHEN) 1.300 NPV * U#S 1.3 " -REDUJCED SILTATION OF RESERVOIR CHECK DAMS 160 1. PRODUCTION OF ELECTRICITY FOR LONGER PERIODRNPV2USS .2 N SLOPING/GRASSING 2. D(ISTEAN IRRIGATION FOR GULLY SIDES 5 ha uss 9,896,400 LONGER PERIOD,uNpVuss 5.7 N -LIMITED HRVESTING OF FORESTRY 0' STREAM BlANK PROTECTION 10 km PLANTINGS, NPV SS .79 N (GAllONS) ROASIOE PROTECTION (RESHAPE EXISTING DRAINAOE COANNELS, 50 km GRASSING ANKS,DROP STRUCTURES) REFORESTAION I COtlITY FOREST 5,500 ha | USS 2,675,500 CONSERVATION FOREST 3,500 ha MNAAGEMENT OF EXISTING U#S 6,388,700 CONSERVATION AREAS on-farm soil conservation technologies, costs and expected benefits from selected World Bank Staff Appraisal Reports TECHNOLOGY PROJECT COST PER HA EXPECTED BENEFITS STRUCTURAL TECHNOLOGIES Terraces China - rural development (1987) USS 216 None stated Bench terraces China - rural development (1987) USS 594 None stated Terrace rehabilitation Indonesia - soil conservation (1987) not specified Conserve soil, reduce sediment, increase crop productivity 100l Bench and broad-base China - area development (1986) not specified Conserve soil, reduce surface runoff terraces Terraces India - watershed management (1983) USS 526 Conserve soit, reduce sediment Terraces/graded bunds India - watershed development (1983) USS 41 - 51 Conserve soil and moisture Contour bunds Haiti - rural development (1983) not specified Increase crop productivity, increase farm family income Contour burids Tunisia - rural development (1981) not specified Conserve soil, reduce sediment, arrest decline in productivity Bench terraces and India - watershed development (1980) USS 536 (bench) Conserve soil and moisture, reduce lowland terraces USS 214 (lowland) sediment Stone terraces Morocco - rural development (1980) not specified Increase crop production, increase farm family income Bench terraces and Indonesia - rural development (1979) USS 355 (bench) Conserve soil, reduce sediment, lowland terraces USS 82 (lowland) increase crop productivity by > 200X increase income 20X co Bench terraces Thailand - agri. development (1979) not specified Conserve soil, slow degradation of t soil and water resource Bench terraces Korea - watershed development (1976) USS 2,417 Upland reclamation Land smoothing China - rural development (1987) US$ 27 (manual) None stated US$ 270 (mechanical) Land shaping and India - watershed development (1983) USS 46 (shaping) Conserve soil and moisture smoothing USS 21 (smoothing) I.and leveling India - watershed development (1980) US# 262 Conserve soil, reduce sediment, increase infiltration, increase groundwater, increase productivity Drains China - area develoment (1986) not specified Conserve soil Drsins and waterways India - watershed development (1983) USS 15 - 21 Conserve soil waterways Paraguay - area development (1982) not specified None stated Diversion ditches Phillipines - agri. development (1980) not specified None stated Drains Indonesia - rural development (1979) not specified Conserve soil FS (Continued) On-farm soil conservation technologies, costs and expected benefits from selected World Bank Staff Appraisal Reports Table 2.2 TECHNOLOGY PROJECT COST PER HA EXPECTED BENEFITS VEGETATIVE TECHNOLOGIES Contour ptanting China - area development (1986) not specified Conserve soil, reduce surface runoff Contour strips and Haiti - rural development (1983) not specified Increase crop productivity, increase hedges, strip crops farm family income Contour tillage Paraguay - area development (1982) not specified Increase crop productivity, increase family income, increase tax income Contour guidelines and Tunisia - rural development (1981) not specified Conserve soil, reduce sediment, contour farming arrest decline in productivity Contour farming Morocco - rural development (1980) not specified Increase crop productivity, increase farm family income Contour hedges Phillipines - agri. development (1980) not specified None stated Plant grass and Legumes China - rural development (1987) not specified None stated a' Mulching Haiti - rural development (1983) not specified Increase crop productivity, increase a' farm family income Grass strips, protection Paraguay - area development (1982) not specified Conserve soil plantings Vegetative and crop Tunisia - rural development (1981) not specified Conserve soil, reduce sediment, cover _mnagemmnt arrest decline in productivity Cover cropping and Phillipines - agri. development (1980) not specified None stated sodding Off-farm soil conservation technologies, costs and expected benefits from selected World Bank Staff Appraisal Reports TECHNOLOGY PROJECT COST PER HA EXPECTED BENEFITS STRUCTURAL TECHNOLOGIES Gully head works, gully Indonesia - soil conservation (1987) not specified Reduce sediment, increase reservoir plugs, check dams, life, increase life of irrigation gully shaping/planting, works, Increase life of electricity gabions, drop strucs. producing dams, silt entrapment providing arabte land Gully plugs China - rural development (1987) USS 810/km of gullies None stated Check dams/drop strucs. China - area development (1986) not specified Conserve soil Check dams, gully plugs Bhutan - forestry development (1984) not specified None stated Check dams Pakistan - hill farming develop. (1984) not specified None stated Brush/stone check dams, India - watershed fhnagement (1983) USS 11/check dam Conserve soil, reduce sediment wire crate dams, drop USS 316/wire dam structures US$ 1,053/drop struc. Brush/stone check dams, India - watershed development (1983) not specified Conserve soil wire crate dam, drop structures Gully structures Tunisia - rural development (1981) not specified Conserve soil, reduce sediment Vegetative/stone check India - watershed development (1980) USS 357 Reduce sediment dm,, wire crate d_, drop structures, debris basins, check dams Check c | Norocco - rural development (1980) USS 330 None stated Gulty plugs Indonesia - rural development (1979) not specified Conserve soil, reduce sediment FORESTRY Community forest and Indonesia - forestry institution USS 297 Reduce silt, provide wood products conservation forest and conservation (1987) Forest plantation, Ethiopia - forestry (1986) USS 1,909 Conserve soil, restore fertility, Upgrade existing USS 1,707 liprove water supply, alleviate plantations, pressure on remaining forest lands Comunity and farm USS 400 provide fuelwood and sawlogs forestry plantations Farm forestry, India - social forestry (1985) USS 16 Conserve soil, increase timber Tree tenure planting, US, 3,121 supply Ceunmity forest, USS 515 Wasteland plantations USS 650 Afforestation, Pakistan - hilt farming dwemlopent (1984) not specified Provide fuestwood > Fuelwood plantation not specified Distribute seedlings to not specified farmers (Contimued) Off-farm soil conservation technologies, costs ard expected benefits from selected Worid Bank Staff Appraisal Reports TECNXOLOGY PROJECT COST PER NA EXPECTED 0ENEFITS FORESTRY Protection plantations Bhutan - forestry developiunt (1984) not specified none stated an slopes > 6OX FuelwoodVtiiber India - watershed munagement (1983) USS 274 (Govt.) Conserve soil, reduce sediment, plantation, USS 379 (Civil) provide fuetwood,timber,fodder Fodder tree plantation USS 158 (Civil) - - USS 12 (Private) -Fern forestry, India - watershed management (1983) USS 115 Conserve soil, provide fuelwood Reforestation USS 344 nd fodder Coaiity reforestation, Nepal - forestry (1983) USS 277 Conserve soil, protect rural water Private reforestation, USS 104 supply, improve soil fertility, Agro-forestry plots, USS 954 benefit microclimate, provide tinber, Strip plantings US, 250 sawlogs, and fuelwood, stabilize roodcuts, river and canal banks Rehabilitate existing Morocco - forestry (1982) US$ 410 Provide pulpwood for industry and forest plantations, 1X of annual fuetwood needs Reforestation not specified Village woodlots, India - social forestry (1982) USS 447 - 471 Conserve soil, stabilize hillsides, Reforestation/plantation US, 494 - 529 and sand dunes, improve soil rehabilitation, fertility, benefit microclimate Fern forestry, S71 - 108 Village woodlots, US, 489 - 527 Strip plantings, US# 58 - 618 oD Sand dune plantings US, 376 Forest management Paraguay - area development (1982) US, 15 Not stated Species trials/pilot Haiti - forestry (1982) US, 2,775 Conserve soil, provide sustainable fuelwood plantations, yield of wood products, offset cost Forest mnagement US, 30 ioported kerosene for fuel Fuelwood plantations Korea --rural infrastructure (1982) USS 714 Conserve soil, improve soil fertility, -reduce flash flooding, amenity value of forest Silvi-pasture Morocco - agricultural development (1981) US$ 69 Increase forage yields, decrease livestock daewae, provide timber ard fuelwod Reforestation/woodlots Tunisia - rural development (1981) -US, 331 Conserve soil, reduce sediment, --provide fuslwood and timber Reforestationlwoodlots/ Morocco - rural development (1980) not specified Provide wood prouicts, forage, fruit forage trees/fruit-trees Reforestation, Phillipines - watershed management (1980) US, 729 Conserve soil, limit forest fires, Forest protection/ US, 28 minimize flesh floods, reduce fire prevention sediment, provide wood products Copnpity forests, Nepal - commanity forestry (1980) USS 69 Conserve soil and moisture, increase Comunity protection -US 10 soil fertility by substituting - -I forests, fuelwood for duig, provide wood 9 (Continued) w off-farm soit conservation technologies, costs and expected benefits from setected Vorld Bank Staff Appraisal Reports TECHNOLOGY PROJECT COST PER NA EXPECTED BENEFITS FORESTRY Private plantings, Nepal - comn nity forestry (1980) USS 0.01/Tree products and fodder Nursery/forestry (contimsed) USS 3,280,000 Total development Reforestation, Phillipines - agri. develop ent C1980) not specified Not stated Forest fire prevention not specified Forest protection/ fire Phillipines - watershed mEnagement (1980) USS 28 Conserve soil, reduce sediment, prevention minimize flash flooding Reforestation, Irdonesia - rural development (1979) US$ 149 Conserve soil, reduce sediment, Silvi-pasture USS 262 regulate stremftlow Village woodlots, Thailand agricultural development (1979) USS 252 Conserve soil, watershed protection, Forest plantation t1SS 193 slow degrodation of soil,water, and forest resource, provide poles,timber,fuelwood Forestry Bolivia - rural develompent (1979) US5 497 Conserve soil, reduce burning of forests Fodder/fuelwood trees, Nepal - rural developent (1976) USS 49 Conserve soil, reduce sediment, forest protection, slope reduce pressure on existing forest, stabilization tree reduce agri. use of mrgitial lands plantings, regenerate '0 vegetation PASTIRE Pasture improvements Norocco - forestry (1982) 1SS 339 Not-stated Pasture establishment Tunisia - rural development (1981) 11S 90 Conserve soil, reduce sediment Pasture establishent Norocco - rural development (1980) 1S5 330 Increase farmer income e.g So - 71 - 3. ECONOMTC ANALYSTS OF SOIL CONSERVATION TECHNOLOGIES William B. Magrath Comparing recent vegetative techniques for soil and water conservation using vetiver grass (Vetiveria zizanioides) and the standard practice of employing earthen bunds, this chapter explores the economic aspects of these alternative techniques. In addition to indicating the relative eco- nomic advantages of a conservation system based on vetiver grass, the exercise also helps set the agenda for the col- lection of additional physical and economic data. The paper consists of four parts: discussion of conceptual issues in the economics of soil conservation investments and the model used to implement this analysis; description of data used; discussion of the results of the basic analy- sis and, due to its speculative nature, modeling of a range of plausible combinations of parameters; and conclusions and recommendations for further research and development. CONCEPTUAL ISSUES It is widely accepted that erosion lowers agricultural productivity (see Chapter 2) and that soil conservation raises and preserves it. However, there is little agreement on exactly how productivity is related to erosion or on the quantitative impact of erosion on yields. In part, the uncertainty arises from the difficulty of defining fertility, as well as of conducting controlled experiments to identify and measure erosion-related yield changes. Erosion involves changes in soil structure that influence root growth and water availability and in the availability and relative concentration of plant nutrients. Soil conservation practices minimize the occurrence of these changes and often induce other reactions that directly improve conditions for crop growth, such as improved response to fertilizer or a delayed wilting point. Nor is there widespread agreement on how erosion influences the eco- nomics of agricultural production. A decline in the underlying productivity of the resource base does, presumably, lower the profitability of farming but not necessarily in a simple or direct way. Erosion-induced losses involve declines in both current and future incomes, but their effects can be masked and at least partially overcome by the use of different or additional inputs. And, like all aspects of agricultural production, they interact with forces of nature largely or totally beyond farmers' control. Similarly, conservation measures often have hidden costs and may generate benefits only over long periods of time. MODELING THE IMPACT OF EROSION AND SOIL CONSERVATION Four Types of Data In view of these difficulties, the most practical approach to devel- oping an understanding of the potential role of soil conservation measures in - 72 - a farming system is to employ an engineering economics approachs the impacts of erosion and conservation are applied to an economic model of crop produc- tion and the value of conservation is calculated on the basis of a with- without comparison. The basic data required for the analysis are crop produc- tion budgets, an understanding of local cropping patterns, evidence on the effects of erosion on yields, and evidence of the impact of specific soil conservation measures on crop yield. The model developed for this analysis consists of two series of linked crop budgets that represent the consequences of erosion, on the one hand, and conservation, on the other, over a 30-year horizon. Cost and reve- nue items included in the budgets are shown in Table 3.1 and include all pur- chased and farmer-supplied inputs which can be valued at market or economic prices. Provision is made for outputs of crops and by-products as well as output (if any) produced from soil conservation practices. Assumptions The case being developed assumes that erosion affects farm income by reducing yields and costs, directly in the case of those which are harvest- related and less directly by making it unattractive for the farmer to add inputs such as the seed of improved varieties or to apply optimal amounts of fertilizer. Conservation treatment is assumed to change costs--by reducing cultivation costs and permitting cultivation and planting to be timely and by increasing costs related to harvesting a higher yield. Direct conservation costs are entered as separate items where applicable. Table 3.1 summarizes the structure of the crop budgets and the assumed impacts of erosion and con- servation. These are restrictive assumptions and do not fully account for the range of impacts and cost-averting opportunities available to farmers. They do, however, probably account for the most direct impacts of erosion, at least in the early years of the planning horizon. The model makes provisions for two crops per year or the rotation of two crops. Output The basic outputs of the model are projections of the flow of net farm income (returns to land and management) without and with the project. The incremental flow can be cast in present value at any interest rate or summarized in an internal rate of return. A variety of extensions are possi- ble, among them; evaluating the impact of cost and benefit sharing, alterna- tive planning horizons and, in the case of vetiver grass, assessing the incen- tive to abandon soil conservation for oil-root harvesting. There is considerable question as to the magnitude of the physical dimensions of the problem, and important parameters will vary from case to case. Therefore, results can be presented to show the impact of changes in key assumptions or the range or combination of parameters that favor one con- clusion or another. - 73 - Table 3.1: IMPACT OF EROSION AND CONSERVATION ON THE MODEL CROP BUDGET Item Impact of Erosion Impact of Conservation Costs Seed 0 ) } } } Fertilizer 0 1 } Manure 0 } } These inputs decrease in pro- Bullock Rental 0 ) portion to land taken out of } production Pesticides 0 } Labor } Land Preparation 0 } Fertilizing 0 1 Cultivating 0 1 Harvesting Decrease in Increase in proportion to pro- proportion to ductivity increase productivity loss Revenues Harvest (Product and By-Product) Decrease in Increase in proportion to pro- proportion to ductivity; decrease in propor- productivity loss tion to area taken out of production DATA Crop BudRets and Two Conservation Techniques The model is used to compare the relative economics of the proposed vetiver grass-based technology with the more conventional approach of con- structing earthen bunds. To the extent possible, data were assembled to reflect conditions on Alfisols in the semi-arid zones of India. Cost, yield and input data for the initial budgets are based on estimates provided by World Bank field staff, which in turn are based on research by state agricul- tural universities. The budget represents a rotation of sorghum intercropped with red gram (Caianus caian) and castor (Ricinus communis). Initial crop budgets are given in Annex 3.1. - 74 - The principles behind the two conservation techniques considered here are similar.l/ By interrupting the length of the field, both techniques are intended to slow movement of water down the slope, which reduces the movement of soil particles and allows for greater absorption of moisture into the soil and hence increased yields. Vetiver grass hedges are said to be more effec- tive in slowing water and eventually form terraces as soil accumulates along their upslope side.2/ Earthen bunds promote some additional absorption of water but are also designed to channel surplus water into drains and water- ways. Loss of water from the root zone via waterways probably accounts for the smaller increases in yield obtained under bund technology. Inadequate provision for disposal of excess water can cause failure of bunds during intense storms and damage to crops in the downslope vicinity. Even so, farm- ers are reluctant to allocate scarce land for water disposal. Table 3.2 sum- marizes data on the impact of selected soil and moisture conservation technol- ogies on soil loss and runoff. Table 3.2: IMPACT OF SOIL AND MOISTURE CONSERVATION TECHNOLOGIES ON EROSION AND RUNOFF Erosion or Technology Sedimentation Runoff Location Reference --- (Z Reduction) ---- Contour Cultivation 10-50 USA Wischmeier and Smith (1978) Contour Cultivation 30-60 10-70 India Gupta, et al. (1971) 25 India Dhruva Narayana (1986) Grass Strips 93 Indonesia Abujamin, et al. (1985) Grass Strips 40-60 USA Carter (1983) Contour Bunds 43 -70/a Thailand Sheng et al. (1981) Contour Bunds 62 Sierra Leone Millington (1984) /a Runoff increased 70Z. 1/ Data reported in this section on the effects of conservation technology are largely based on literature reviewed in Chapter 2 and Tejwani (1989). 2/ This effect has been noted with other grass used in soil conservation work in India. See Sud et al., 1975, as discussed by Tejwani (1989). - 75 - Erosion and Productivity Decline Despite scientific uncertainty over the impact of erosion on produc- tivity and crop yields, as noted, data from a variety of experiments provide some indication of the magnitude of productivity declines. Experiments on Alfisols in Africa showed that mechanical removal of the top 10 cm of soil resulted in yield declines of 73Z for maize. Similar experiments at Dehra Dun, India showed more modest impacts (Table 3.3). Actual levels of erosion on cropland, as reported by El-Swaify, et al. (1984), which appear conserva- tive, are only a fraction of the amounts experimentally removed. Results reported by Lal (1987, and personal communication) argue strongly that the damage from naturally occurring erosion is much more severe than that produced in artificial experiments, largely because natural erosion tends to remove preferentially the most productive constituents of the soil. In this analy- sis, data from Dehra Dun were used to represent the without-conservation case. These are highly conservative and almost certainly serve to understate both actual damage from erosion and benefits from conservation. Table 3.3: EFFECT OF SOIL REMOVAL ON MAIZE YIELDS ON ALFISOLS Depth of Soil Removal Nigeria India (cm) (Z Decrease in Yield) 5.0 30.5 12.5 10.0 73.6 25.1 15.0 33.3 22.5 93.5 39.1 30.0 45.0 Source: Lal (1987) for Nigeria; Hegde (1988) for India. El-Swaify, et al. (1984) estimate that Alfisols on relatively mild topographies at ICRISAT (Hyderabad, India) have a mean annual erosion hazard exceeding 40 tons/ha or approximately 5 mm. Applied to the data from Dehra Dun, India, this implies yield declines of 1.252/year for five years, decreas- ing to 0.95Z/year thereafter (see Figure 3.1). In all likelihood, yield declines would become more severe in the absence of soil conservation as sheet erosion gives way to rill erosion leading to larger soil losses. On actual farmers' fields, it could be expected that some effort would be devoted to soil conservation but, at this point, there are no data to indicate what adjustments might be made. - 76 - Figure 3.1 EFFECT OF EROSION ON PRODUCTIVITY 0.9 0.8 - o 0.7- 0 0 0. 0.6 -J z 0.2 2 0.4 1 2 3 4 5 6 7 8 9 10 11 12 13 14 151617 1819 2021 2223 2425 2627 2829 30 YEAR - 77 - Soil Conservation Treatment and Crop Yields Vetiver Grass. The effects of alternative soil conservation technol- ogies on crop yields have been studied in a number of experiments in India and elsewhere. Despite the apparent simplicity of the questions being asked, there are as yet no definitive answers. Experimental designs are weak and researchers have often focused on questions that are peripheral to impact on yields. Table 3.4 summarizes available data on the impact of alternative soil conservation technologies. In the case of soil and moisture conservation using vetiver grass hedges, the quality of the crop cutting experiments that have been conducted is questionable. There are, however, other data from trials with other grass barriers that provide some indication of likely bene- fits. In addition, it is likely that a significant proportion of the yield increase attributed to the vetiver grass system results from the accompanying practice of contour cultivation.3/ Using Greenfield (1987) as a base case, a 50Z yield increase from vetiver grass technology has been assumed. Table 3.4: IMPACT OF SELECTED SOIL AND MOISTURE CONSERVATION TECHNIQUES Technique Impact on Employed Crop Yield Crop Location Reference (% increase) Contour Cultivation 36 Sorghum India Dhruvm Narayans (1988) 12 Sugarcane Taiwan Liao (1972) q60 Cotton USA Unger (1984) * 6 Potatoes India 9 * 48 Maize India U 3 25 Sorghum Kanpur, India Bhatia and Chaudharye a 15 Barley Kanpur, India Sloping Agricultural Land Technology 107 Maize Philippines Watson and Lnquihon (n.d.) Earthen Bunds 10 Maize Chandigarh, India Sud, et al.* * 36 Maize Uttar Pradesh, India Khan* * 18 Setaria Tamil Nadu, India Kanitkar* 11 Cotton Tamil Nadu, India 17 Sorghum Tamil Nadu, India 24.5 Sorghum Maharastra, India Tamhane* 26.2 Pearl Millet Maharastra, India U 3866 Sorghum Tamil Nadu, India * 2.64 Pearl Millet Tamil Nadu, India U 21.4 Wheat Punjab, India 3 16 Grain Punjab, India * 19.7 Maize Punjab, India U 13.9 Pearl Millet Punjab, India 0 * In Tejwani (1989). Constructing Bunds. Similarly, experiments with bunding treatments provide only limited evidence of their efficacy. It is even possible that the standard practice of field bunding may have no positive impact on yields. However, in the calculations discussed, as a base case using Tejwani (1989), it has been assumed that bunds will increase yields 30% over the without- project case. For both bunding and vetiver grass treatments, the percentage 3/ Because the vetiver grass system essentially forces the adoption of contour cultivation, no effort has been made to separate these effects. - 78 - of yield increase has also been treated as a parameter and results have been presented for yield increases ranging from 0 to 1402.4/ Comparative Cost of Alternative Treatments Investment costs for vetiver grass and bund treatments have been taken from estimates made by World Bank staff in preparing the Integrated Watershed Development Project--Plains (1989) in India. Costs for vetiver grass include labor in man-days (Md), bullock power (Bprs), fertilizer, and contingencies. Planting material is valued at full cost, including transpor- tation, 25Z contingencies, plus a 50? markup. Costs are detailed in Tables 3.5 and 3.6. These cost estimates are now considered generous. If vetiver grass technology is widely adapted, the cost of planting material will eventu- ally fall to be only the labor cost associated with harvesting, and planting slips from existing hedges. For the purposes of costing, a hectare of cropland is assumed to require 250 linear meters of contour hedge. At a width of 0.5 m, the hedge would occupy 125 sq m.5/ In addition to initial planting costs, allowance has been made for hedge maintenance in the second and third years, after which it is assumed that hedges would be fully established. Costs for bunds are intended to reflect the same parameters. Costs are based on earth work on light soils and assume a bund cross-sectional area of 0.5 sq m, which appears to be the current standard. Land estimated to be taken out of cultivation by bunds consists of the width of the bund (1.7 m) and berm (0.3 m) plus one half of the borrow pit (1.7 m) and a provision for drains and waterways (0.3 m). These costs are detailed in Table 3.7, and total Rs 863/ha. The cost of grassing and maintaining bunds has been excluded from the analysis. Poor maintenance is one of the main causes of bund failure and necessitates their frequent replacement; the base case assumption is that bunds require replacement every five years. The use of vetiver grass strips as a source of fodder has been observed in southern India, but there are currently no data on its value as fodder or on sustainable yield. Fodder yields have not been incorporated into the benefit flows but this could easily be done as additional data become available.6/ 4/ There is considerable doubt as to whether these yield increases can actually be reached. Percentage of yield increases, of course, depends on the base, which in the case of the semi-arid zone, can be highly variable. It is clear that yield increase due to moisture conservation can, in percentage terms, be very high in drought years, but in good years that the same absolute increase would be small in percentage terms. 5/ A square, one-hectare plot with a slope of 2.5?, would require approximately this much material. 6/ Potentially significant, especially regarding adoption, given the frequent importance of livestock to small farmers. - 79 - Table 3.5: COSTS OF PRODUCING VETIVER SLIPS IN A NURSERY Per Ha Cost/ No. of Total Cost items Units Units Units Costs (Rupees) (Rupees) Labor and Machinery Plowing Bprs 45 10 450 Breaking Clods Md 12 50 600 Spreading Manure Md 12 10 120 Forming Ridges & Furrows Bprs 45 5 225 Transport Planting Material Md 12 10 120 Treatment Dressings Md 12 15 180 Pruning and Sorting Md 12 20 240 Planting of Slips Md 12 75 900 Weeding Bprs 45 15 675 Weeding & Topping Md 12 150 1,800 Uprooting Clumps Md 12 25 300 Subtotal 5,610 Inputs Planting Material '000 10 62.5 625 Manure Ton 50 25 1,250 Diammonium Phosphate (DAP) Kg 3.5 250 875 Urea Kg 2.6 375 975 Atrazine (ai) Kg 167 1.5 250 BHC (10X) Kg 2 25 50 Irrigation Total 250 Subtotal 4,275 Base Costs 9.885 Contingencies, Losses, etc. Z 25 2,471 Total Costs 12,356 Outputs slips* '000 1,875 Average Cost per Slip* Paisa 0.66 Sales Price** Paisa 1.00 * Basis for costing purposes is 30 slips/clump. ** Assumes 50Z markup. Bprs = bullock pair-days. Md = labor in man-days. - 80 - Table 8.6: COST OF ESTABIUSHINO VETIVER GRASS HEDGES (1989 COSTS) Unit No. of Units Yr of Establishment Total Unite Cost Yr I Yr 2 Vr T Yr 1 Yr 2 Yr 3 Cost (Re) (Rs) LABOR/INPUTS Labor Opening Furrows La Spra 46 0.6 22.5 0.0 0.0 22.5 Forming Bunds Md 12 6 60.0 0.0 0.0 60.0 Pruning, separating, loading & unloading Md 12 2 0.4 24.0 4.8 0.0 28.8 Planting A dressing Md 12 4 0.8 48.0 9.6 0.0 57.6 Weeding Md 12 2 24.0 0.0 0.0 24.0 Subtotal 178.5 14.4 0.0 192.9 Inputs Purchae Cost of Slips L '000 10 40 8 400.0 30.0 0.0 480.0 Transport of Slips Le x 10 40.0 0.0 0.0 40.0 DAP Kg 8.5 20 70.0 0.0 0.0 70.0 Urea (8 split dressings) Kg 2.5 60 150.0 0.0 0.0 150.0 BHC (10X) Kg 2 40 4 80.0 8.0 0.0 88.0 Contingenciso x 10 10 74.0 8.8 0.0 82.8 Subtotal 814.0 96.8 0.0 910.8 TOTAL COST m.5 111.2 0.0 1 108 7 Rond dCo-t 990 ilT1TU0 TREATMENT COST PER HECTARE /d Labor 44.6 8.6 0.0 48.2 Inputs 20386 24.2 0.0 227.7 TOTAL COST 248.1 27.8 0.0 275.9 R ounded Cost 2 2 PROJECT COST PER HECTARE Labor X of above 100 100 44.6 8.6 0.0 48.2 Inputs X of above 100 100 208.5 24.2 0.0 227.7 TOTAL COST 248.1 27.8 0.0 276.9 Round-d Cost 276 / Costs entered as bullock pair days. Soe nursery costs, Table 8.5. From nursery to field site. Basod on 40 a horizontal interval, equlvalent to 250 . per hectare (1 m vertical Interval). - 81 - Table 3.7: COST OF CONSTRUCTING EARTHEN BUNDS (1989 COSTS) Slope (Z) Unit 1 2.5 4 Construction Costs Average bund length per ha 100 250 400 Average earth works sq m/ha 50 125 200 Field bunding costs Rs 300 750 1,200 Associated Costs* Rs 45 113 180 Cost per gross hectare Rs 345 863 1,380 Loss of Arable Land Affected width sq m of bund 4.00 4.00 4.00 Adjusted width sq m of bund 3.00 3.00 3.00 Area affected sq m 400 1,000 1,600 Net loss sq m 300 750 1,200 Proportion affected X 4.0 10.0 16.0 Net Loss Z 3.0 7.5 12.0 Cost per net hectare Rs 356 932 1,368 * For associated diversion channels and vaterways--15Z of direct costs. Assumptions: Bunds establ$shed at one meter vertical interval, bund cross- section equals 0.5 sq m and labor rate is equal to Rs 6/sq m for earth work. RESULTS Comparative Viability The results of calculations are sumarized in Table 3.8, and illus- trated in Figures 3.2 and 3.3 for vetiver grass and earthen bunds, respec- tively. Using the base case assumptions, both systems appear economically viable. However, vetiver grass with a net present value (NPV) of Rs 8,543/ha (IRR-95Z) is clearly superior to bunding (NPV-Rs 3,436/ha, IRR-28Z). The dominance of the vetiver grass technology, of course, is essen- tially complete for any plausible combination of parameters, mainly due to the cost advantage of vetiver grass. Figure 3.4 illustrates the impact of alter- native productivity assumptions. Even if it is assumed that the impact of vetiver is only to prevent erosion, a yield increase from bunds of nearly 40Z (higher than the optimistic base-case assumption) would be required before bunding would become the more desirable option. - 82 - Figure 3.21DISCOUNTED IMPACT OF VETIVER GRASS 50 YIELD IMPACT 1.30 - 1.20 - w 1.10 c< 1.00 I Z 0.90 I 0 o -0.80 / -0.70 ra 0 (Ac 0.40 I-1 0 0.30 - l Wi 0.20- + 0 o 0.10 WIT VETIVER++ ++ + 4ffU VETVE INRMNA. ,. 0.00 - w z -0.10 -0.20- -0.30 - 0 1 2 3 4 5 6 7 8 9 10 11 12 1314 1516 171819 20 212223 24 252627 2829 30 YEAR, WITH VETIVER + WrITOUT VETIVER - INCREMENTAL - 83 - Figure 3.3 DISCOUNTED IMPACT OF FIELD BUNDS 30% YIELD IMPACT 0.90 - 0.80 - 0.70 C),6-0°60 070 . z 0.30 (. 0303 V) 0.20- 5~~~~~~~~~~~~ 0.10 -0.9 -0.10 <2 -0.20- ".~'-0.30- -0.40 -0.50 0 o -0.60 Z -0.70 Id -0.80 z -0.90 -1.010 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 192021 222324252627282930 YEAR O WITH BUNDS + WITHOUT 'BUNDS - INCREMENTAL - 84 - Figure 3.4 IMPACT OF ALTERNATIVE PRODUCTIVITY ASSUMPTIONS ON RATES OF RETURN 1.6 - 1.5. 1.4- 1.3 1.2 1.1 h- 0.809 Id 0.8 0 wi 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0.2 0.4 0.6 0.8 1 1.2 1:-4 ASSUMED PRODUCTIVITY IMPACT 0 FIELD BUNDS + VETIVER - 85 _ Table 3.8s RESULTS OF ECONOHIC ANALYSIS OF ALTERNATIVE SOIL CONSERVATION TECHNOLOGIES Yield Erosion Increase Loss Pre- NPV la En Technology (2) vented (2) (Re/ha) (2) Vetiver Grass 5O0 0.95-1.25 8,543 952 Earthen Bunds 352 0.95-1.25 3,436 282 Vetiver Grass 502 0 6,765 872 Earthen Bunds 35Z 0 1,659 22Z Earthen Bunds 352 0.95-1.25 4,719 342 (no replacement required) /a At 102 discount rate. If damage from erosion is ignored, the impact of a 502 yield increase from the vetiver grass treatment still shows a return of 872. On the other hand, under the same assumption, a 352 yield benefit from bunding returns only 222. Since, in fact, neither conservation technology will completely stop erosion, the actual rate of return would lie somewhere between these two esti- mates. The assumption that field bunds need to be replaced every five years has relatively minor impact on the profitability of the technology. From the base-case rate of return of 282 with replacement, the rate of return rises only 62 to 342 if bunds are assumed to last the 30 years. The present value of future costs of replacement are so small as to have little impact at that high an implicit rate of discount. At a more modest discount rate of 102, the impact and the present value rise from Rs 3,436/ha to Rs 4,719/ha. Nonethe- less, even if bunds are maintenance-free, vetiver grass technology is more cost-effective. Land Tenure Issues Regarding land tenure issues in watershed development, two issues are relevant heres the importance of farmers' time horizon and the role of cost and benefit distribution as incentives for adoption of soil conservation (Chapter 6). Presumably, cultivators with less secure tenure will be more reluctant to invest in conservation techniques because of the longer time required to reap benefits from their investment. Conceptually, of course, this implies a high rate of time preference and can be expressed as a high discount rate. Alternatively, and perhaps more intuitively, this insecurity - 86 - can be modeled by imposing a shorter time horizon. Figure 3.5 illustrates the rate of return to the two technologies as a function of planning horizon. For any planning horizon long enough to motivate adoption of either technology (>3 years), vetiver grass technology will yield higher returns. Benefit and Cost Distribution In addition, the impact of benefit and cost distribution on the incentive to adopt is relevant to government cost-sharing policies. From the point of view of land tenure. assuming that a small share of benefits accrues to the cultivator is comparable to assuming that profits from conservation are fully capitalized into land rent. Assuming that costs are shared between cultivator and landlord is conceptually the same as some form of government cost-sharing. To illustrate these issues, the basic model was extended to separate the costs and benefits of conservation in order to analyze alterna- tive combinations of cost/benefit sharing. The internal rate of return for various combinations was calculated and liso-returns curves (the locus of points representing equal rates of return) are plotted in Figure 3.6. Figure 3.6 can be used to distinguish the combinations that would lead a profit- oriented cultivator to adopt from those combinations which would be considered most attractive. The lower curve represents the vetiver grass technology and a 401 rate of discount. It essentially implies that cultivators operating under a discount rate as high as 40% would be willing to adopt vetiver grass technology if they expect to receive only 20% of the benefit, even if they bore the entire cost. The higher curve attempts to represent the decision calculus of a more patient cultivator (discount rate = 20Z) considering earthen bunds. The higher curve indicates more stringent requirements and the greater slope more sensitivity to cost/benefit sharing. In order to adopt bunds, a farmer would require 30% of the benefits before shouldering 50% of the costs and would require nearly 70% of benefits before investing the entire cost of bunding.7/ Concern about Vetiver Root Harvesting One aspect of vetiver grass technology that raises concern is the occasional occurrence of a lucrative cash market for vetiver grass root (see Annex 3.2). The harvesting of vetiver grass lines that are intended for soil conservation purposes can have quite the opposite effect. Uprooting vetiver grass can create deep furrows that are especially subject to erosion. This has been observed in areas of commercial vetiver grass production in Indonesia, as well as in a soil conservation project in Haiti. A farmer's decision to pull out the soil conservation lines in this fashion would presum- ably be related to his failure to understand their purpose, special needs for cash that the family might have, or a host of other considerations. More generally, however, even with full knowledge about consequences, combinations of prices and discount rates do exist that would lead a profit-maximizing farmer to adopt such an environmentally harmful practice. 7/ The choice of different discount rates appears awkward but is due to the fact that at a 20Z discount rate, vetiver grass technology is accepted at essentially any combination, while at the 40Z discount rate earthern bunds are refused at all but the most generous combinations. The curves shown are relatively favorable to bund and unfavorable to vetiver grass technologies. - 87 - Figure 3.5 TIME HORIZON AND CONSERVATION 1- 0.9- B o o 0.8 - 0.7- 0.6 - 0.5 - 0.4 - 0.3 X z 0.2- 1 0.1 0' IL o -0.1 iLi -0.2- i -0.3 -0.4- -0.5- -0.6- -0.7- -0.8- -0.9 -1 -1.1 1 1 13 PLANNING HORIZON (YRS) a VETIVER + BUNDS - 88 - Figure 3.6 EFFECT OF COST AND BENEFIT DISTRIBUTION ON INCENTIVE TO ADOPT SOIL CONSERVATION 0.9 - 0 l--:1. 0.8 - o 0.7 - zD 0.4 - m 0 w 0.6 - U w 0. 0.4- 0.3- 0 0.2 - <- 0.1 A 0.5 0.7 0.9 SHARE OF COSTS BORNE BY CULTIVATOR 0 VETIVER (40X) + BUNDS (20%) - 89 - This problem can be modeled by defining the keeping of vetiver grass lines in the ground as the without-project situation and harvesting for oil production as the project. There are three key parameters that need to be estimated for this calculation--the cost of harvest, the value of the root and the impact of harvest on erosion and yield. Harvest cost and root price have been treated as parameters and solely for the purpose of calculation, it has been assumed that root harvest will lead to an erosion rate of approximately 2 cm/year, leading to productivity declines of 5Z per year. The range for harvesting cost is suggested by the costs given in Table 3.4 for nursery oper- ation but are expected to be higher for oil production because the roots will be larger and deeper. The range for root price is based on root prices reported in Indonesia (see Annex 3.2). Figure 3.7 illustrates the results of this analysis showing the com- binations of harvest cost and root price that will lead to either root harvest or conservation. The curves shown represent a discount rate of 10, 20 and 30Z, respectively, and illustrate that results are quite insensitive to choice of discount rate. These results, together with experience in Haiti and Indonesia, suggest that the combination of parameters leading to abuse of vetiver grass is quite plausible and is an issue requiring attention during project preparation. It is important to realize that the model is severely limited in this respect and that harvesting vetiver grass roots will drastic- ally alter the structure of the farm management problem in ways not envisioned here. Moreover, the world (let alone local) market for vetiver oil is presum- ably not perfectly elastic, and it is reasonable to think that appropriate measures could be designed to depress returns to oil production. Nevertheless this is one issue that seems to require additional consideration. CONCLUSIONS Value of Structured Economic Analysis The foregoing discussion illustrates that despite obvious gaps in knowledge, a structured analysis of soil conservation investments can generate useful insights, and perhaps most usefully, can highlight specific issues on which additional research is necessary. Most notably, these are not questions of economic methodology. Rather they are largely technical questions about the impact of erosion and conservation on yields. In addition, there is the obvious need for more reliable cost data on crop production and a better understanding, both technical and economic, of farmers' responses to erosion. Economic analysis is not merely a device for project justification. Rather, it should be seen as a tool for decision-making and for understanding the resource allocation problems facing farmers and project planners. In this context, an important role of economic analysis is to identify and articulate areas of uncertainty and potential trade-offs. Consequently, an important aspect of this analysis is the manner in which uncertain results are presen- ted. - 90 - Figure 3.7TRADE-OFF OF CONSERVATION AND OIL SALES LEVEL SETS OF BREAKEVEN DISCOUNT RATES 10.5 - 10 M 9.5 - 9 O 0.~~~~~~~~~~~~~~~~~ w 8.5 O c 8 SELL ROOT a. 8 - 0 o C) o + °r 7.5- irC w 0 I;i CONSERVE 6.5- 0 + 0 6 + + 5.5 r 0 0.2 0.4 0.6 0.8 1 (Thousands) VETIVER HARVEST COSTS 0 . 1 + .2 0 .3 _ 91 - Better Data Needed to Verify Assumptions As a final point, the primary purpose of this analysis is normatives to assess whether a particular technology is profitable and should be promoted to farmers. The more positive issue, whether farmers will adopt a technology, depends on whether the assumptions on prices (including discount rates) and quantities are essentially correct and on the extent to which the assumption of present-value maximization holds true. The ex ante benefit/cost literature on soil conservation generally appears to have been overly optimistic about benefits and costs, has often failed to adequately address the subsidy compo- nent of conservation promotion schemes, seldom accounts for maintenance requirements and most importantly seems to be at odds with evidence on adop- tion rates. More careful attention to the dynamics of the adoption of soil conservation techniques is clearly needed. - 93 - ANNEX 3.1 DEVELOPMENT OF HECTARE CROP BUDGETS FOR TECHNOLOGY ANALYSIS Red Gram Castor Unit Total Unit Total Units Qty. cost cost Qty. cost cost (Rs) (Rs) (Rs) (Rs) Crop Expenses Seeds Main crop kg 10 1.5 15 25 3 75 Intercrop kg 1 6 6 Fertilizer Urea kg 20 3 60 0 3 0 Phosphorus kg 20 4 80 0 4 0 Potash kg 0 0 Manure ton 4 80 320 4 80 320 Bullock rental Ipra a so 400 8 50 400 Pesticides kg 0 0 Subtotal inputs 1,225 1,200 Labor Land preparation Hd 7 15 105 7 15 105 Fertilizing Md 2 15 30 2 15 30 Cultivation Md 5 15 75 5 15 75 Harvesting Hd 13 15 195 18 15 270 Threshing Md 0 15 0 0 15 0 Subtotal labor Md 27 15 405 32 15 480 Total Crop Expenses 1.660 1,680 Conservation Expenses Libor 0 Inputs 0 0 Total Conservation Expenses 0 0 Revenues Crop harvest Main product kg 500 2 1,000 600 4 2.400 By-product kg 2,000 0.16 320 300 0.16 48 Crop harvest Main product kg 150 6 900 0 By-product kg 600 0.08 50 0 Other harvest Main product kg 0 0 By-product kg 0 0 Total Revenue 2,270 2,448 Returns Net revenue return to land and management 610 768 - 94 - ANNEX 3.2 Cost and Returns for Vetiver Oil Production in Indonesia Background. Until recently, the economic use of vetiver grass was for the extraction of aromatic oil, for use in perfume manufacture, from the plant's roots. Other, relatively minor, uses are the manufacture of hand fans, screens for evaporative coolers and aromatic sachets from the plant's root. Harvesting the root can be environmentally destructive. For example, in Indonesia harvesting is carried out up and down the slope to take advantage of erosion that naturally loosens the root from the soil. As harvesting pro- ceeds, bare soil is exposed to accelerated erosion. Although it is conceiva- ble that a sustainable cultivation system could be developed for commercial vetiver production, to date none is available. Cost and Returns. In the vicinity of Garut, West Java, a large area of vetiver grass is cultivated for oil extraction. Harvested roots are macer- ated by hand and distilled 12 hours with water in two-ton batches, producing 6 kg of oil. Each of several stills at a site is used for 7-10 batches per week. Costs and revenue estimates for vetiver distillation are given in Table Al. Table Al: ESTIMATED COSTS AND RETURNS FOR VETIVER OIL DISTILLATION IN WEST JAVA Item Unit Price Quantity Total (Rp) (Rp) Root /a kg 150 2,000 300,000 Kerosene liter 200 300 60,000 Labor Md 1,500 6 10,000 Returns Oil kg 79,000 6-10 474,000-790,000 Return on variable costs per run 104,000-420,000 Return on variable costs per still per year /b 41.6 - 168.0 million /a Cost to distillery based on yield of 20 tons/ha/yr and labor input of 60 man-days/ha. Vetiver is planted 80,000-100,000 slips/ha (spacing of 40-50 cm between rows. lb Eight batches per week, 50 weeks. Markets and Prices. Not much is known about the world market for vetiver oil. According to distillers, export markets have been growing rapidly. Prices rose in 1989 from Rp 46,000/kg to Rp 79,000/kg. - 95 - REFERENCES Abujamin, S., A. Abdurachman, and Suwardjo. 1985. "Contour Grass Strips as a Low Cost Conservation Practice," ASPAC Extension Bulletin No. 221. Carter, D.L. 1983. "Erosion and Sediment Losser In Furrow-Irrigated Land" in El-Swaify et al., ed., pp. 355-364. Dhruva Narayana, V.V. 1986. "Soil and Water Conservation Research in India," Indian Journal of Soil Conservation, 14(3), pp. 22-31. El-Swaify, S.A., T.S. Walker and S.M. Virmani. 1984. Dryland Management Alternatives and Research Needs for Alfisols in the Semi-Arid Tropics. International Corps Research Institute for the Semi-Arid Tropics: Patancheru, Andra Pradesh, India. El-Swaify, S.A., W.C. Moldenhauer, and A. Lo, eds. 1983. Soil Erosion and Conservation, Soil Conservation Society of America: Ankeny, Iowa. Greenfield, J.C. 1987. Vetiver Grass (Vetiveria Zizanioides) A Method of Soil and Moisture Conservation. World Bank: New Delhi. Gupta, S.K, D.C. Das, K.G. Tejwani, S. Chittaranjan and Srinivas. 1971. "Mechanical Measures of Erosion Control" in Tejwani et al., eds. Hegde, B.R. 1988. "Concepts and Components of Watershed Development" in reading material proposed for the Workshop on Watershed Management for Rainfed Farming Systems in Hill States, December 2-4, National Centre for Management of Agricultural Extension, Rajendranger, Hyderabad. Lal, Rattan. 1987. "Effects of Soil Erosion on Crop Productivity" CRC Criti- cal Reviews in Plant Sciences, 5:4, pp. 303-367. Liao, M.C. 1972. "Study on Soil and Water Conservation in Sugarcane Land" TSES Report, vol. 55, pp. 75-90. Millington, A.C. 1984. "Indigenous Soil Conservation Studies in Sierra Leone" in Challenges in African Hydrology and Water Resources, IAHS Pub. No. 144. pp. 529-537. Rao, Y.S., M.W. Hoskins, R.T. Vergara, C.P. Castro. n.d. Community Forestry: Lessons from Case Studies in the Asia and the Pacific Region. Regional Office for Asia and the Pacific of the FAO and Environment and Policy Institute, East-West Center: Bangkok and Honolulu. Sheng, T.C., J.K. Jackson, J. Kraayenhagen, J. Nakasthin, P. Watnaprateep. 1981. "The Effects of Different Structures on Erosion and Runoff" in South-East Asia Regional Symposium on Problems of Soil Erosion and Sedi- mentation, pp. 301-310. - 96 - TeJwani, K.G. 1989. 'Consultant's Report on Benefits of Soil and Water Con- servation Technologies," unpublished report prepared for the World Bank, New Dehli. Tejwani, K.G., S.K. Gupta and H.N. Mathur. 1971. Soil and Water Conservation Research 1956-1971, Indian Council of Agricultural Research: New Dehli. Unger, P.W. 1984. Tillage Systems for Soil and Water Conservation, FAO Soils Bulletin 54. FAO: Rome. Watson, H.R. and W.A. Laquihon. n.d. "Sloping Agricultural Land Technology (SALT): A Social Forestry Model in the Philippines" in Rao, Y.S. et al. (eds.) pp. 21-44. Wischmeier, W.H. and D.D. Smith. 1978. "Predicting Rainfall Erosion Losses - A Guide to Conservation and Planning," USDA Agricultural Handbook. - 97 - 4. ECONOMIC ANALYSIS OF OFF-FARM SOIL CONSERVATION STRUCTURES William B. Magrath Although off-farm structures such as checkdams and water- harvesting structures are often a substantial portion of costs in watershed projects, they are seldom subjected to benefit-cost analysis, due to their typically small size and dual objective of providing environmental and produc- tive benefit. The benefit-cost analysis in this chapter, based on data collected in Indonesia, demonstrates that a rudimentary analysis can help distinguish viable invest- ments from unprofitable ones, even when values for environ- mental benefits are uncertain. Furthermore, the effort required for benefit-cost analysis is minimal, compared to the engineering and construction input. Also noted are some strategic considerations involved in evaluating off- farm structures for watershed projects. CONCEPTUAL ISSUES Investment Objective In most respects, checkdams and water-harvesting structures can be approached as conventional minor irrigation reservoirs. They are distinct in that one of their significant functions is to capture sediment, and conse- quently their life span is often quite short. While siltation is frequently considered cause for alarm, it is important to note that these structures are often built solely or primarily for the purpose of capturing silt and that if they were not eventually silted, the investment could be considered a failure. In addition to determining the life span of a structure, siltation also influences the stream of irrigation benefits during the structure's life. In large reservoirs the ratio of capacity to sediment deposition is so large that, for practical purposes, reductions in service due to siltation can be ignored. In small structures, often associated with highly degraded catch- ments, benefit streams are more likely to be critically affected by sedimenta- tion. Dynamics of a Reservoir This can be reflected in an economic analysis by explicitly relating benefit flow to the current status of the reservoir which needs in turn to be related to sediment delivery. The dynamics of the reservoir can be represen- ted by the following difference equation: CT+1 = CT - ST (eq. 1) where CT+l = Reservoir capacity at time T + 1 CT = Reservoir capacity at time T ST = Sedimentation during period T _ 98- Initial storage capacity (CO) is known from engineering and site studies. Current sedimentation can be estimated for small catchments through a variety of techniques. These include use of empirical soil loss equations (e.g., the universal soil loss equation (USLE), construction of sediment budgets, or monitoring data on stream-suspended sediment and bedloads. Sediment Rates. Sediment rates are highly variable and dependent on both climatic variables and land use. In most cases, it will be necessary to use estimates of long-term average sedimentation rates. These, however, must be used with caution because there is seldom a sufficiently long time-series of data on which to calculate highly reliable estimates. If land-use changes or improvements are expected in the catchment, sedimentation can be treated as a variable dependent on other investments.l/ Projections of reductions in sedimentation need to be viewed skeptically. Significant improvements can probably only be reasonably expected in relatively small catchments and after some lag time. Reservoir Capacity. The relationship between current reservoir capa- city and benefit flows is dependent on a number of variables, including type of structure and operating rules. This can often be clarified by analysis of engineering studies. More typically, economic analysis of explicitly examined reservoir sedimentation has treated benefits as proportional to capacity. For example, an XZ reduction in reservoir capacity can be assumed to produce an XZ reduction in irrigated area. Provided that irrigation water is efficiently used in the first place and that there are limited opportunities for substi- tuting less water-demanding crops, this is probably not an unreasonable first approximation.21 Valuing Sediment Capture. Explicitly valuing sediment capture is another issue raised in the analysis of off-farm structures. This is often not strictly necessary when irrigation benefits can be shown to justify con- struction, but is relevant where irrigation benefits are insufficient or where the purpose of the analysis is to aid in project design. The value of sediment capture is site-specific and dependent on the nature and value of the receiving downstream area. Deposition of silt in reservoirs and irrigation systems can, for example, raise operation and main- tenance costs and/or reduce operating efficiencies. Deposition of gravelly and sandy material on agricultural land can reduce the value of this land or even take it entirely out of production. Yet, sediment deposit can be benefi- cial where sediment has desirable properties as a building material or for other uses.3/ 1/ In which case the cost of land-use changes, including both direct costs and opportunity costs, need to be included in the analysis. 2/ See, for example, Southgate (1986). 3/ In the Kali Progo Irrigation Project, Indonesia, removal of 32,000 m3 of sediment from silt traps costs Rp 23 million (Rp 718/m3) per year. However, this material has a ready market and can be sold for Rp 2,000/m3, yielding a net return of Rp 44 million/year. - 99 - Reasonable approaches to shadow prices for sediment capture can be based on costs of dredging (or other ameliorative measures), where practiced, or on the value of service reductions. It is useful to note that the costs of service reductions should not exceed the costs of dredging; otherwise, it would be in the interests of the downstream authority to institute such a program. An alternative approach to shadow pricing, especially in the case of reservoirs, is to note that what is actually being valued is water storage. Reservoir capacity, once constructed, is a depletable resource. It is well established in the literature on depletable resources that their shadow price will rise at the rate of discount until reaching a backstop price.4/ The backstop price, in the case of reservoirs, could either be dredging (seldom practiced in developing countries) or the construction of additional storage by raising dam height or new dam construction. A CASE STUDY: INDONESIAN REGREENING PROGRAM Appraising Economic Efficiency of Checkdams Checkdam construction is a major element of the Indonesian Regreening Program, accounting for as much as 602 of total annual expenditure (World Bank, 1989). The economic questions posed by this investment program are typical of those discussed in this chapter. The checkdams are intended to serve multiple functions, including provision of irrigation water, silt cap- ture and land preservation and reclamation. To illustrate an approach to appraising the economic efficiency of these investments, data were collected on 23 checkdams constructed under the program in Central Java between 1983 and 1988. Data. Costs and initial storage capacity data were available for all 23 dams.5/ Information on catchment size, erosion rate and command area were available for only seven dams planned for FY88/89. These data are summarized in Table 4.1. 4/ The classic treatment is Hotelling (1931); the ratio of a backstop price was popularized, in the context of energy resources, by Nordhaus (1969). For discussions of the application of this approach to reservoirs and watershed management, see Magrath and Grosh (1985) and Southgate (1986). 51 All costs were converted to 1988 Rupiah using the IMF wholesale price deflator. - 100 - Table 4.1:, DESCRIPTIVE DATA ON 23 INDONESIAN CHECXDAMS Units Average Minimum Maximum Catchment area ha 140.5 70.5 285 Ponding area ha 1.2 0.75 2.725 Storage m3 26,970 7,860 125,200 Erosion rate mm/year 4.6 2.43 5.53 Expected,life year 3.6 1 7 Calculated life year 4.2 Dam length m 74 Dam height m 8 Dam base m 4 Command area ha 3 Cost Rp million 33 19.7 49.2 Cost/r3 Rp 1,668 392.7 .3.539.0 Assumptions. Results presented here are based on an analysis of a representative checkdam, using average,values and construction cost derived from a regression equation relating costs with initial storage capacity. The equation was of the form: ( B , (eq.2) cost/r3 - (Storage Capacity m-7 where B is a parameter to be estimated. A relationship between storage capa- city and command area was estimated based on the sample of seven dams. This relationship was,assumed to be of the,form: Command Area = A x [Storage] (eq. 3) where A is another parameter to be estimated. The form of equation 3 was chosen to reflect the economics of scale in dam construction. Equation 2 reflects the assumption that command area (irrigation benefit) is linearly related to capacity. Both equations are restricted to have a zero intercept on the basis that a reservoir of zero capacity has neither cost nor command area. The results of ordinary least-squares estimation are shown in Table 4.2. In both cases, the estimates are statistically significant and of good fit. - 101 - Table 4.2i REGRESSION ESTIMATES A B Coefficient 0.000227 28484004 t-statistic 8.93 19.45 R2 0.594 0.690 It is further assumed that after complete siltation the surface of the reservoir is salvaged as arable land. Physical Benefits Based on these assumptions it is possible to trace, using the logic of equation 1, the evolution of physical benefits from the representative checkdam, using the laws of motion given in Figure 4.1. Figure 4.1s CHECMDAh SYSTEM EVOLUTION Command area - 0.000227 * Current storage (ha) (m3) Sediment capture - Erosion rate * catchment area * 10 (m0) (mm) (ha) Current storage - Lagged storage - (Erosion rate * catchment area * 10) (m3) (m3) (mm) (ha) Together these assumptions generate projections of physical benefit flows from the representative damrn A base-case scenario is given in Table 4.3. Table 4.s: PHYSICAL BENEFIT STREAMS FOR REPRESENTATIVE CHECKDAM Year - O 1 2 a 4 5 6 7 B Cowmnd area (ha) 6.1 4.7 8.2 1.7 0.8 0.0 0.0 0.0 Sediment trapped (68) 0,463.0 6,408.0 0,468.0 6,463.0 1,118.2 0.0 0.0 0.0 Land reclalmed (ha) 0.0 0.0 0.0 0.0 1.2 1.2 1.2 1.2 Current storag. (p8) 26,970.173 20,507.2 14,044.2 7,561.2 1,110.2 0.0 0.0 0.0 0.0 - 102 - Irrigation is valued at Rp 1,200 per hectare, based on calculations presented in the appraisal of the Forestry Institutions and Conservation Proj- ect (World Bank, 1988). Sediment capture is initially valued at Rp .60/m3, based on the cost of constructing water storage in large reservoirs on Central Java.6/ Based on the assumption that reservoir storage is a nonrenewable resource, it is further assumed that this value will rise over time,at the rate of interest. This is a low value of storage compared to the average cost of storage in checkdams (Rp 1,056). However, it reflects the fact that the cost of storage is subject to large economies of scale. Prevention of silta- tion in other applications, especially the protection of irrigation systems, can have considerably higher value. Mechanical silt removal from silt traps in Central Java costs roughly Rp 600/m3. Accordingly, although the primary purpose of the sampled checkdams is to protect reservoirs, higher values vere also tried to examine the viability of using dams to protect high-value installations. Lastly, land reclaimed is valued at an annual rental rate of Rp 2,225,000 per hectare, based on estimated returns to land and management. Economic Benefits Applying these values to the physical benefit flows in Table 4.3 yields the following profile of value flows:7/ Tobl. 4.4: ECONOMIC COST AND BENEFIT FLOWS FOR REPRESENTATIVE CHECKDAM (Rp'OOO) Year O 1 2 3 4 5 8 7 Costs (Rp'OOO) -7-oistruction 28,484 B.n.fits oi-mand area (he) 7 6 4 2 0 0 0 0 Sediment trapped (m3) 428 474 523 678 110 0 0 0 Land reclaimed (ha) 0 0 0 0 2,226 2,225 2,225 22,248 Total -28.484 435 479 527 580 2.33S 2.225 2,225 22.248 Under these conditions the investment has a net present value 8/ of Rp 12,672,000 and an internal rate of return of 1%. These results stem pri- marily from the high costs of storage and the low value of outputs. To explore the sensitivity of these results to alternative assumptions, several sensitivity analyses were conducted. In Table 4.5, the implications of higher sediment capture values and various values for irrigation services are shown. 61 The Kedung Ombo Dam, Indonesia, was constructed at an average cost of Rp 55.8/m of storage. 7/ Land returns accruing after the eighth year are capitalized at a 1OZ discount rate. 8/ All present values are based on a lO2 discount rate. - 103 - Table 4.6: NET PRESENT VALUE OF REPRESENTATIVE CHECKDAM UNDER XLTERNATIVE IRRIGATION AND SILT CAPTURE PRICES Value of Irrigation (Rp'OOO/ha) 0.0 1.0 2.0 3.0 4.0 5.0 0.00 -14,328 -14,313 -14,300 -14,287 -14,274 -14,260 0.06 -12,961 -12,946 -12,936 -12,922 -12,909 -12,696 0.10 -11,596 -11,683 -11,570 -11,557 -11,548 -11,58o 0.15 -10,231 -10,218 -10,205 -10,191 -10,178 -10,16 0.20 -8,86e -8,653 -8,840 -8,826 -8,813 -6,800 Value of git capturo 0.26 -7,501 -7,488 -7,474 -7,481 -7,448 -7,436 (Rp'OOO/m4) 0.30 -6,136 -6,122 -6,109 -6,096 -8,083 -4,070 0.40 -3,405 -3,392 -3,379 -3,386 -3,858 -3,840 0.50 -876 -662 -649 -836 -623 -609 0.80 2,056 2,068 2,081 2,094 2,108 2,121 0.70 4,786 4,798 4,811 4,826 4,836 4,861 1.00 12,978 12,989 13,002 13,016 13,028 18,041 These show that considerably higher values for sediment capture, close to those associated with dredging of irrigation systems, are required to make the representative checkdam economically viable. This result is highly insensitive to irrigation values. Table 4.6 gives net present values for alternative combinations of construction cost and silt capture. The sensitivity of NPV to construction cost illustrates its role in producing the negative estimate of profitability. Table 4.6: NET PRESENT VALUE OF CHECKDAM UNDER ALTERNATIVE CONSTRUCTION COSTS AND SILT CAPTURE PRICES Cost of Construction (Rp'000) 0.00 16,000 20,000 26,000 30,000 35,000 40,000 0.00 -826 -5,828 -10,828 -15,826 -20,826 -25,826 0.05 639 -4,461 -9,461 -14,481 -19,461 -24,461 0.10 1,904 -3,098 -8,098 -13,098 -18,096 -23,096 0.15 3,269 -1,731 -6,731 -11,731 -16,731 -21,731 0.20 4,634 -366 -6,386 -10,3e8 -15,368 -20,366 Value of gilt capture 0.25 6,999 999 -4,001 -9,001 -14,001 -19,001 (Rp'000/m3) 0.30 7,364 2,364 -2,636 -7,638 -12,638 -17,638 0.40 10,094 5,094 94 -4,906 -9,906 -14,906 0.50 12,825 7,826 2,826 -2,175 -7,175 -12,176 0.60 16,656 10,656 6,656 656 -4,445 -9,44s 0.70 18,285 13,285 8,286 3,28s -1,715 -8,715 1.00 26,478 21,478 18,476 11,476 6,476 1,478 An interesting aspect of the checkdam investment is its built-in obsolescence. Tables 4.7-4.9 show the results of different catchment erosion rates (and implicitly different reservoir life span), sediment capture and irrigation values. For all combinations, higher erosion rates (or shorter life spans) yield higher returns. This is due to the earlier onset of land reclamation benefits. This result is particularly interesting in that concern - 104 - is often raised about the short life of these structures. Moreover, a stan- dard recommendation is to require catchment stabilization prior to construc- tion--a practice which, if followed, reduces the value of the representative checkdam. Table 4.7: EROSION RATE (m/yr) 8 4 5 6 7 0.00 -16,948 -14,809 -12,792 -12,798 -11,128 0.05 -16,571 -12,942 -11,428 -11,481 -9,762 0.10 -14,199 -11,576 -10,064 -10,069 -8,402 0.16 -12,827 -10,208 -6,700 -8,707 -7,041 0.20 -11,454 -6,840 -7,888 -7,846 -6,681 Value of lit captur, 0.26 -10,082 -7,478 -5,972 -6,983 -4,820 (Rp'000/al) 0.80 -8,710 -.,106 -4,608 -4,621 -2,969 0.40 -5,965 -8,872 -1,879 -1 ,897 -288 0.50 -8,221 -886 849 827 2,488 0.60 -476 2,096 8,577 8,651 6,204 0.70 2,266 4,831 6,805 6,275 7,926 1.00 10,502 18,088 14,489 14,447 16,089 (Reservoir Life) (year) 6.4 4.8 8.8 8.2 2.7 Table 4.8: EROSION RATE (ma/yrl Silt capturo Valued at Rp 60t a 4 5 6 7 0.0 -16,816.7 -12,686.6 -11,189.8 -11,172.2 -9,502.88 1.0 -16,800.0 -12,671.2 -11,167.8 -11,161.1 -9.492.32 Value of lrrIgation 2.0 -15,283.4 -12,656.7 -11,144.9 -11,160.1 -9,482.29 (Rp'000/ha) 8.0 -16,266.7 -12,642.2 -11,132.6 -11,139.0 -9,462.23 4.0 -15,260.1 -12,627.8 -11,120.0 -11,128.0 -9,482.23 6.0 -16,233.4 -12,613.8 -11,107.6 -11,117.0 -9,462.20 Table 4.9: EROSION RATE (mm/yr8 SilT Capture Valued at Rp 800/m 8 4 5 a 7 0.0 -498.108 2,079.077 8,661.702 8,637.684 5,192.401 1.0 -479.449 2,093.641 3,674.138 3,548.700 6,202.433 Value of Irrigation 2.0 -482.795 2,108.006 8,686.689 8,669.738 6,212.486 (Rp'000/ha) 8.0 -448.141 2,122.489 8,699.003 3,670.778 5,222.498 4.0 -429.487 2,136.934 8,611.437 3,581.809 6,232.630 5.0 -412.833 2,151.398 3,623.870 3,692.846 5,242.582 (Reservoir Life) (year) 6.4 4.8 8.8 8.2 2.7 - 105 - STRATEGIC CONSIDERATIONS The inclusion of off-farm structural works in watershed projects raises issues that are not solely economic and which need to be considered in the context of an overall approach. It may lead to either inclusion of works that are strictly speaking not economically viable or exclusion of viable ones. This chapter discusses approaches to valuing environmental benefits that are usually not considered. In many countries, off-farm structures have become ingrained in watershed investment and they offer costs and benefits of quite a different sort. For example, the ability of soil conservation agen- cies to provide upland areas with irrigation benefits, even on a costly and probably inequitable basis, may be of value in gaining the cooperation of local communities for the introduction of other conservation techniques (see, for example, Society for Promotion of Wastelands Development, 1990). On the other hand, structural works often provide an opportunity for rapid expendi- ture of project funds with little meaningful local participation. Further, the inclusion of off-farm works, even if economically desirable, may also serve to bias agency efforts away from on-farm and forestry measures. Thus, while it is feasible to conduct benefit-cost analyses of check- dams and other small structures, there may be good reasons for overriding the results. At a minimum, however, it is reasonable for donors and planning agencies to require implementing agencies to include simple economic analysis in the planning of such structural works and to document the justification for proceeding with structures that do not appear viable. This requirement could be imposed on structures greater than a certain size or that require more than a minimum level of site surveying and engineering. In general, it is clear that checkdams and other small structures are an expensive adjunct to downstream water/sediment storage capacity. Because of the considerable economies of scale present in reservoir construction, small structures need to offer significant directly productive benefits or be sited to provide protection to highly valued infrastructure which would other- wise require expensive maintenance. Other off-farm conservation works, for example, gully plugs which are primarily intended to stabilize channels and prevent loss of adjacent land, can be analyzed in similar fashion. It would be imprudent here to consider detailed analysis of all the small, scattered structures typically included in a watershed management project. The practice of benefit-cost analysis is itself an exercise in benefit-cost analysis and judgments as to the value of incremental information to the decision-making process is an essential ele- ment. - 107 - REFERENCES Dixon, John A., Daniel E. James and Paul B. Sherman. 1990. Dryland Manage- ment: Economic Case Studies. Earthscan Publications: London. Hotelling, Harold. 1931. "The Economics of Exhaustible Resources," Journal of Political Economy, Vol. 39, pp. 137-175. Magrath, W.B. and M. Grosh. 1985. 'A Dynamic Approach to Watershed Manage- ment with Emphasis on Soil Erosion,' paper presented at the Ninth World Forestry Congress, Mexico City, Mexico. July. Nordhaus, W.D. 1973. "The Allocation of Energy Resources," Brookings Papers 3, pp. 529-70. Sfeir-Younis, A. 1985. 'Soil Conservation in Developing Countries: A Back- ground Report", draft manuscript. The World Bank: Washington, D.C. Society for Promotion of Wasteland Development. 1990. 'Economic and Social Changes in a Small Rural Community in the Degraded Lower Shivalik Hill Range in North India' in Dixon, James and Sherman. Southgate, Douglas. 1986. 'The Off-Farm Benefits of Soil Conservation in a Hydroelectric Watershed", paper presented at the Annual Meetings of the American Agricultural Economics Association, Reno, Nevada, July. World Bank. 1988. Indonesia Forestry Institutions and Conservation Project, Staff Appraisal Report, Report No. 7002, Washington, D.C. . 1989. Indonesia: Forest, Land and Water: Issues in Sustainable Development, Report No. 7822, Washington, D.C. - 109 - 5. REVEGETATTON TECHNOLOGIES Ajit K. Banerjee Revegetation is a significant technology employed in the rehabilitation of upper watersheds in the Asia region. It traditionally encompasses both enrichment planting and forestation of bare areas. This chapter recognizes that much plantation work has been less successful than hoped and discusses the reasons, both technical and nontechnical, for the lack of success. Technical components covered are: local need-based planning, choice of species, nurseries, ground preparation, planting methods, planting designs, protection and management. For each topic, the issues and options are presented. Poor success rates are attributed, variously, to these components, but also to social factors, such as the critical need for active participation by local inhabitants in revegetation programs, which is now often lacking. INTRODUCTION The relationship between the upper watershed and the forest is impor- tant to appreciate. Prior to settlement, most watersheds, both lower and upper, were covered with mixed species forest. In a pristine state, the upland forest comprised multilayered vegetation with a fully covered forest floor that held erosion to a minimum. Now mostly cleared to accommodate agri- culture, forests have become restricted to the steeper, higher-elevation shal- low soils or droughtier regions less suited to agricultural production, and even so, are being degraded by fuel-gathering and grazing. The degradation has reduced the vegetative cover and litter to various extents, often down to bare surface, inducing runoff and erosion and changing soil moisture status. With some exceptions, forests today provide marginal economic returns to gov- ernment; yet, they are of considerable importance to villagers who exploit them for timber, firewood, minor forest products, food, and sometimes game and feed for their animals. Revegetation, a significant technology in the rehabilitation of upper watersheds in the Asia region, encompasses enrichment planting (supplementing standing forest crops) and forestation (planting on bare land). Given the effect of degradation on soil and moisture loss, on-site soil and moisture conservation is an integral part of revegetation. Engineering structures for soil and moisture conservation are expensive to construct and maintain and are short-lived; therefore, they are not considered here, although it is acknow- ledged that under uncertain circumstances, especially in gully rehabilitation, such structures may be required to allow vegetation to establish. Rather, the focus is on vegetative conservation measures. - 110 - TECHNICAL COMPONENTS OF REVEGETATION Local Need-Based Planning While technical factors (such as species, spacing, soil moisture) are vital, the success of revegetation programs is also dependent on benefits accruing in the short and long run to local users who have traditionally depended on vegetative resources on adjacent common or public property for fuel, fodder or income. Population pressure and the consequent rising demand for fuel and fodder from the same land resources cause upland degradation. For various reasons, including poverty and lack of fuel alternatives, rural families cannot forsake biomass fuel (Leach, 1987) and this dependence will persist for some time to come. Similarly, stall-feeding of animals to relieve pressure on forest vegetation, although feasible, is unlikely to be widely adopted in the near future because of the need to grow fodder and the invest- ments needed. Also, in India for example, poorer villagers often subsist by illicitly collecting and selling fuelwood in towns (Fernandes, 1987). Genera- tion of employment through revegetation programs could possibly break this ongoing poverty-land degradation-poverty cycle. Revegetation planning that does not take into account local needs is inadvisable, because new plantations cannot survive the pressure of fuel collection for consumption and commerce, nor can they survive unrestricted grazing. The correct approach to revegeta- tive technology, therefore, would consider both the site and needs of the local population. All common land is tied in some way to users, who may be individuals, groups, communities or villages, and although usage may not be recorded, it is traditionally accepted by all concerned. For practical reasons, the identity and size of the planning unit for revegetation are important. Furthermore, needs of subunits may differ and identifying those that can be addressed requires care: marginal farmers may own few cattle and need little fodder, while the intermediate-size farmers may need more; the landless may require employment from the program, while farmers would not. Discussion with repre- sentatives from various economic groups within the planning unit is therefore necessary to provide a full-bodied and accurate depiction of needs, and in this discussion, women as a social unit should be involved. Once needs are assessed and land and financial resources of the unit analyzed, technology suitable for the site and aimed at meeting these needs can be developed. Continuous consultation within the client community and adaptation from one site to another regarding planning, execution and mecha- nisms for distributing benefits should, therefore, be major elements of revegetation programs (Banerjee, 1987). Forestation projects by nongovernmental organizations (NGOs) have succeeded in many countries (though on a small scale) due to project planning, execution and protection by local inhabitants (see Chipko in Garhwal Himalayas and Appiko in the Western Ghat mountains in Hegde, 1987). In contrast, many government projects, without local involvement and enthusiasm, have failed in spite of suitable technology. Issues of Participatory Planning. Given the fact that local partici- pation is essential for successful revegetation programs, some important ques- tions emerge: - 111 - (a) What should be the geographical unit whose inhabitants relate to the project site: village, a cluster of villages, district or some other unit such as a users' group? (b) Should an institution represent the people? If an institution is necessary, should it tap into an existing one or be newly formed or should a spontaneously created group be promoted? (c) What should be the mechanism of participation? Recent experience shows that the smaller the unit in relation to the revegetation site the more likely people will take a cooperative interest in the project and the more likely it is to succeed. A village, often encompas- sing a users' group, is usually a reasonable unit for participatory planning. Existing institutions, if genuinely representative, are sound choices. The mechanism for consultation may involve complexities best sorted out by spe- cific circumstances. The poitit to keep in mind is that the views of a cross- section of a village, usually made up of heterogeneous elements, should be incorporated in the technical plan, designed to satisfy plural objectives. With any choice, coordination of local interests is necessary--whether it involves a community movement spawned by local initiative in ecological resto- ration; a community association, such as the village forestry associations in the Indo-German Dhaulader project in Himachal Pradesh, India, that assist in planning, protection and management of revegetation of degraded hills; or existing institutions such as the panchayat in India. Choice of Species Monoculture and Exotics. Over the past three decades of forestation programs, monoculture has become an regular practice. Only a few species have been used and often they are exotics: Eucalyptus tereticornis, Casuarina equisetifolia, ProsoDis iuliflora, Cryptomeria iaponica, and Pinus patula in India; Acacia manRium and Paraserianthes falcataria (prev. Albizzia falcataria) in Bangladesh; Pinus caribaea in Malaysia; Leucaena leucocephala in the Philippines and Thailand; Eucalyptus camaldulensis and Eucalyptus citriodora in many of the South Asian and insular Southeast Asian countries. Two important reasons for selecting exotics and monoculture are fast growth and easy management, which characteristics have obvious advantages in areas suffering from biomass shortage and a lack of technical manpower. How- ever, the adverse effects of monoculture could outweigh its advantages. To ensure adequate stocking, close espacements are used, resulting in early canopy closure which discourages undergrowth, for example, in plantations of Cryptomeria iaponica, Eucalyptus tereticornis, Pinus roxburghii and Tectona grandis. Some species, particularly Eucalyptus, are said to tap too much groundwater and nutrients. Fast-growing species must use these essential elements in absolute quantities, but often utilize each molecule of water and nutrient to produce more biomass than slower-growing species can. Some spe- cies may allocate more energy towards wood production at the expense of crown development, for example, Casuarina equisetifolia. -This benefits fuel produc- tivity but limits the supply of twigs and leaves for local people and reduces recycling of nutrients. Exotics may provide a new food source for pathogens - 112 - and could be vulnerable to pests and diseases due to growth stresses related to poor site conditions. Corticium salmonicolor is a serious pest of Eucalyptus grandis in Kerala, India and Agrilus opulensis, a wood borer, severely attacks Eucalyptus deglupta in the Philippines (Evans, 1982). Impact on Erosion. Lack of undergrowth in hill conditions is a seri- ous deficiency which results in soil erosion. Care is needed to select spe- cies with relatively light crowns in order to encourage ground cover. Since fast-growing species require more water, special efforts are needed during planting to ensure maximum percolation of rainfall into the soil. When exces- sive nutrients are likely to be lost due to disproportionate wood/leaf produc- tion, green manure plants should be introduced. Rao (1967) reports a decline in yield of C. equisetifolia from 185 tons at the end of the first rotation to 140 tons at the end of the third, in Nellore, Andhra Pradesh, India. Since the C. equisetifolia had to be replanted, the seed source may have been the cause of yield decline, but possibly different crops should be used in alter- nate rotations. Quite often exotic species are mistakenly introduced in enrichment planting also. Planting of Acacia auriculiformis in degraded forests of Shorea robusta in Bihar, Eucalyptus tereticornis in moist deciduous forests of West Bengal and Kydia calycina in tropical rain forests of Arunachal Pradesh, India, have been unrewarding. Response to Site Conditions. Several issues need to be considered in selecting species for Asian uplands which are fragile ecosystems that deterio- rate quickly. The process of deterioration is progressive: reduced protec- tion of the soil surface by removal of cover permits increased runoff with a concomitant reduction in percolation and loss of soil, leading to a sequential reduction in available soil moisture and nutrients. Therefore, choice of species has to take into account, not only site and local needs, but also the degree of deterioration. Three categories of degeneration can be recognized. If the site is bare, degraded to the point that it cannot support trees, then grasses, legumes and local shrubs may be the only alternatives. On a par- tially degraded site with some scrub vegetation left, enrichment planting of aggressive indigenous shrubs and management by protection may be appropriate. But, if the site is just beginning to deteriorate so far as canopy cover is concerned, a mixture of trees and shrubs to establish a two- or three-tier forest should be introduced. Bare sites are characterized by thin soil cover, low moisture-holding capacity, low fertility, and ongoing erosion. These areas may also be subjec- ted to overgrazing and periodic burning of whatever surviving weeds or grasses remain. Attempting to revegetate them with tree species, even if indigenous, is bound to fail, based on experience in the western Himalayas of India, where Pinus roxburghii and Quercus incana, both indigenous, have been planted with- out adequate success. Both these species, which were respectively the pioneer and climax tree species, now stand as "relicts'.l/ Due to biotic disturbance, the cites have retrograded ecologically and the appropriate vegetation required to reverse the situation has to be more akin to that found in newly 1/ Persistent remnant of otherwise extinct flora. - 113 - exposed sites. Grasses, herbaceous weeds, hardy shrubs, and legumes are the early harbingers in natural succession and should be planted to begin revege- tating such sites. On degraded sites with scrub vegetation, the situation is quite dif- ferent. Often they are occupied by unwanted shrubs such as Lantana spp. and Eupatorium spp. in Nepal and India, by single-stemmed bamboo Melocanna bambusioides in Burma, and Imperata cylindrica grass in many of the South Asian countries. While these species appear to stabilize the site and prevent erosion, they seldom allow other planted species to grow and are mostly unwanted by local people. However, their removal in order to replant with desirable species can induce erosion. In such cases, enrichment planting of desired species that will gradually suppress the weeds is the appropriate measure. The choice of species is difficult, because they have to be aggres- sive with the capability of competing with the shallow, matty root system of weeds which use soil moisture rapidly and, at the same time, withstand fire, cutting and grazing. Not many trees, except some belonging to the Leguminoseae family such as Acacia mangium and Robinia pseudoacacia, can satisfy these demands. On the other hand, many shrubs have the desired properties and are at the same time useful to local people. On sites beginning to deteriorate--with poor tree stocking but rea- sonable soil depth--selected species should include all forms of vegetation: trees, shrubs, herbaceous legumes and grasses. They will be able to establish a multitier plantation crop. The tree species can be either local or proven exotics, while the shrubs preferably should be local ones. Some examples of indigenous tree species used in forestation programs are Pinus roxburghii in Indonesia, Eucalyptus deglupta in the Philippines and Papua New Guinea, Araucaria huntsteinii and A. cunninghamii, Tectona grandis, Dalbergia sissoo, P. roxburghii and Cedrus deodara in India. Exotic species used include Pinus caribaea in Malaysia, P. patula and Pseudotsuga menziesii in New Zealand, and Eucalyptus spp. in most countries. There have been very few attempts to com- bine tree and shrub species in plantations, which has led to a number of draw- backs. Unless the tree is a multipurpose species, the plantation cannot sup- ply more than one category of need to local people. Secondly, tree species usually take a few years to produce a useful product unlike the shrubs which mature faster. Thirdly, the tree canopy generally fails to prevent erosion to the extent a multitiered forest can. The correct choice of species to estab- lish multistoried plantations, however, can meet most of these deficiencies. Issues Concerning Choice of Species. Species selection is related to the type of site being revegetated and preferences of the local population. On degraded sites with shallow soils, useful grasses, legumes and indigenous shrubs are likely to be better adapted than tree species. They will develop a root mat and surface cover to minimize runoff and soil moisture losses. When the site already has a cover of aggressive, nonusable species, locally availa- ble hardy shrub and tree species, particularly tree legumes like Acacia mangium and Robinia pseudoacacia, should be introduced to develop a multilevel canopy. On poorly stocked sites with adequate soil depth, exotic trees with indigenous shrubs are suggested, aimed again at a multistoried canopy. The propagation technology of indigenous shrub species is poorly understood and will require field research. - 114 - Nurseries Seed Quality. Currently, the weakest aspect of nursery operations is the use of poor-quality seeds. Seedlings raised from them produce trees of poor form and growth rate. Such trees are found in large numbers in planta- tions of many afforestation projects in Asia. Since most countries are increasing their planting activities with inadequate attention to seed selec- tion, unsatisfactory results are likely; in other words, the biomass potential of the ecotype will remain unrealized. A few short-term measures can substantially improve seed quality. Most countries plant a large number of tree species, but only a few in large numbers. Initially, the most commonly used species in forestation programs should be selected for quality improvement. The first step would be to select a number of phenotype (candidate plus trees) of these species; the selected trees should be mature and have characteristics considered by their planters to be desirable when ready for harvesting. The number of trees to be selected would depend on the quantity of seeds produced by a tree, seed germination percentage and the total quantity of seed required for the program. The second step would be to mark, protect and maintain the selected trees to obtain the maximum yield of seeds. The third step would be to engage trained personnel for seed collection and to eliminate seeds from any other source. These selected seeds then should be supplied to all nurseries assigned to the project. Records of seed origin, at least by provenance, should be main- tained. Long-term measures also need to be taken. These measures would include provenance trials, selection of the best provenances, identification of their outstanding trees, etc. In addition, establishment of a seed and testing certification department that would act as the core department to control quality of seeds distributed would be necessary. Root-Shoot Ratio. Another important aspect of nursery operations is the root-shoot ratio of the seedlings raised for planting. A commonly held notion among foresters is that larger seedlings have a better chance of survi- val and establish themselves faster than smaller ones. This is erroneous and needs to be strongly countered. Smaller seedlings with a bushy root system and a woody erect shoot should be the preferred product for planting. The ratio of root to shoot varies across species, but in all cases the root system should be sufficient to provide the seedling with required water and nutrients in the initial period of its life. Modern nurseries raise containerized plants in root trainers which prevent root coiling and ensure that growth commences quickly after transplanting. Ground Preparation Ground preparation consists of one or more of the following opera- tions: removal of ground vegetation, windrowing, burning and soil cultiva- tion. Depending on the objectives of the revegetation program, nature of the site and availability of resources, these operations are carried out in vary- ing degrees. Each measure has the potential for increased erosion and must be carried out carefully if required. - 115 - Removal of Ground Vegetation. Ground vegetation on degrading uplands prevents soil erosion and conserves moisture but also competes for moisture, nutrients and light with planted species. It is a controversial matter as to whether the existing vegetation should be removed or left undisturbed. All sloping areas are vulnerable to erosion under certain circumstances. There are few species which compete with undisturbed ground vegeta- tion. Most Eucalyptus spp. prefer a completely cultivated and weed-free site for rapid early growth. In Papua New Guinea, a trial of eucalyptus failed completely because of competition from Imperata cylindrica. Similarly, growth of Araucaria cunninghamii and A. hunsteinii in Papua New Guinea is poor when there is lack of weed control (Evans, 1982). In contrast, many species of acacia such as Acacia auriculiformis, A. mangium and Faidherbia albida (prev. A. albida) grow well, albeit at a reduced growth rate, in competition with ground vegetation. When deciding on the degree of ground clearance to be undertaken, the objectives need to be clear. If the objectives are mainly soil conservation and rehabilitation of the site, existing vegetation should not be disturbed and loss of growth of planted seedlings should be considered as a trade-off for the achievement of the objectives at hand. On the other hand, if increas- ing the cash value of the crop with marketable species is the intention, removal of existing vegetation becomes-imperative, while soil and moisture conservation measures have to be undertaken as a part of the plantation opera- tion. A compromise of the above two extremes is removal of alternate strips of vegetation at specified intervals and stacking them up against the undis- turbed vegetation (Cassels, 1983). Clearance of vegetation can be accomplished mechanically or manually. Removal of ground vegetation is a capital-intensive task if heavy tractors, bulldozers and other attachments are used. Manual operation is preferred on steep hills, as it disturbs the soil less, but is time-consuming and expen- sive. Sixty-five man-days were necessary to clear a light savannah area in Nigeria with 9-sq m basal area of wood, whereas two 180-hp tractors did a similar job in about 18 minutes. Bulldozers cannot operate beyond a certain slope, and in these situations, manual operation remains the only available option. Windrowing and Burning. The cleared vegetation is heaped and should be windrowed across slopes by machines. However, as harvesting is usually done up and down slope, windrowing across slope is inconvenient. On most forestation sites, there is hardly any valuable harvestable product and wind- rowing on the contour is then possible, which reduces soil erosion. The heaped or windrowed vegetation is burned when dry and the ash provides a good nutrient-rich seedbed for seedlings. Several studies have shown that seed- lings grow well in these ashbeds. Soil Preparation. Soil preparation is one of the most important operations in the plantation operation as it affects soil erosion, moisture conservation, plant growth and plantation cost. Soil manipulation can range from no-till to intensive bulldozing, plowing and harrowing operations. The most common soil preparation methods now in practice in the uplands are: - 116 - (a) no-till; (b) pitting; (c) patch or strip soil working; (d) herringbone or fish-scale plantations; (e) tie ridging; (f) contour stone walls; (g) gradoni or banquettes; (h) broken contour line ditches; and (i) V-ditch and contour banks; No-till. There is now a considerable body of opinion that soil should not be manipulated on any slope beyond 202 if the soils are loose. Vulnerability of such sites to erosion is most visible in Kandi Siwaliks, India. The rocks in this upland chain consist of fragile sandstone and'siltstone which disintegrate to'sand or silt on minimum disturbance. They then roll down the hills as slides or wash down with rain water to silt-up downstream agricultural land. There- fore, no-till is generally appropriate in such areas. Pitting. Pits are usually cubes, each side being 30-45 cm, dug at 2-4 m intervals. The soil is heaped on the side of the pit for 1-3 months before being put back in the pit for planting. Pitting is a common practice all over India. In Fiji, small seedlings of Pinus caribaea are inserted in holes dug only with a crowbar (Evans, 1982). A popular misconception exists that a pit planting design on the con- tour is beneficial as an anti-erosion measure. In fact, except for initial assistance to the seedling by providing loose soil, pits whether on contour or not have little impact in reducing erosion, nor do they conserve any significant moisture. Patch or Strip Soil Working. Sometimes small square patches, each side being 1-2 m, at a certain spacing are loosened by hand to pre- pare for sowing and planting. The main reason for such a simple soil operation is to reduce cost. This method does not significantly assist seedling growth. Some species, however, need shade initially to grow and patch sowing under the shade of other trees can be a suc- cessful practice. Azadirachta indica in semi-arid Vertisol soils in submontane and low hills of central India are sown successfully in this manner. Eucalyptus pilularis and E. grandis are grown by patch sowing in Australia. Strip working is more intensive; strips are plowed at intervals. In the hills, the method is not recommended on bare slopes as it accelerates erosion significantly. On slopes with some vegetation, however, the uprooted shrubs and tree stumps are often stacked along the lower edge (Chapman and Allan, 1978). Any - 117 _ dislodged soil is to a large extent trapped both by stacked material and the undisturbed strip. Tie Ridging. This method covers the entire surface with basin-like furrows scooped out along the contour with special plows, and the soil is thus ridged. The contour ridges are then interconnected to create a number of basins for water accumulation. The method is very expensive and not suitable for areas with moderate to steep slopes. Contour Stone Walls. Chapman and Allan (1978) describe contour stone walls with bases of 30-40 cm and height of 20 cm and cross ties every 5 m along the contour wall. Whatever soil is available is dragged back to seal gaps in the wall to make a reverse slope. Seedlings are planted on either side of the cross tie, thus using the low point of water concentration. This method is inexpensive if sufficient stone is available at the site and useful in arid and semi-arid uplands with shallow soil. Herringbone or Fish-Scale Method. These methods direct water from a small surrounding area to the planted seedling. With the herringbone pattern (Dalwaulle, 1977), a crescent-shaped low ridge is raised on the downslope side of the tree and small channels are dug into the upslope side leading to the tree. In the fish-scale method, a low dike is raised on the downslope by soil collected from about 2-4 sq m from the upslope side to provide a depression for water collection against the dike where the seedling is planted. The method is an inexpensive moisture conservation method, but in the course of a few years, the depression is filled up by soil moving in from outside and the advantage of pooling is lost. Gradoni or Banquettes. These are narrow terraces built along the contour on the hillside with the outside rim higher than the inner edge. The terraces are generally discontinuous and are staggered between rows. The runoff water not only loses velocity at the ter- race, but also is collected in the terrace because of the negative slope. Erosion is thus reduced and seedlings planted in the gradoni receive additional water. In Morocco, Eucalyptus gomphocephala and Pinus halepensis are successfully raised on gradoni terraces (Chapman and Allan, 1978). In Gujarat, India, gradoni soil working is the standard practice to forest degraded stony areas of low hills. Broken Contour Line Ditches. This method is practiced in arid and semi-arid hill slopes of parts of India. The ditches dug on the contour are of different sizes (usually 30 cm x 30 cm x 60 cm), done either by hand or by heavy tractor-bulldozers, spaced 3 m apart center to center in the row and 3-6 m between rows, in a staggered fashion. The soil dug out from the ditches is heaped on the down- slope side of the ditch. The trenches collect the water which bene- fits the seedlings planted either in the ditch or on the mound. Unless the ditches are at close intervals, which is expensive, this method is inadequate as the ditch fills quickly with soil washing down from the upslope side. The seedling bed in the trench is also - 118 - not a suitable microsite for root development, because on degraded hill slopes the subsoil is rocky. Broken contour line ditches, however, are better for soil and moisture conservation than simple patch soil working or pitting, the two most commonly practiced methods. V-Ditch and Contour Banks. This is a totally mechanized operation executed by heavy crawler tractors provided with rippers and angle dozer-blades. Contour subsoiling is possible on up to 30Z slope but is most satisfactory on slopes below 22-252 (Shepherd, 1986). The latest available machines with new ripper designs shatter a total width of 3 m to a depth of 50 cm. In order for the subsoiling to be effective, it is carried out when the soil is dry. The rip lines are usually 5-6 m apart. After subsoiling is complete, the tractor's angle blade makes a V-ditch in the ripped line and simultaneously forms a ridge on the downhill side of the ditch. Seedlings are planted in the ditch 1-2 m apart in holes dug with a spade. This pattern of earthwork is most suitable where the soil is shallow and is underlain by hard pan. The ripper, by shattering the pan, improves water percolation and enhances root development. The V-ditch on the contour combined with subsoiling is an effective way to conserve moisture in-situ and to reduce soil erosion. In the Pilot Project for Watershed Development in Andhra Pradesh, India, this method has been used to obtain uniform and significantly better growth of Eucalyptus tereticornis, Dalbergia sissoo and several other tree species planted in degraded Alfisols. Issues Concerning Ground Preparation. Several issues relate to pre- paration methods: (a) ground vegetation removal may contribute to erosion, but nonremoval leads to competition with the planted species for light, moisture and nutrients, so careful site reconnaissance is needed before vegetation removal; (b) burning releases nutrients locked up in waste biomass. This may have a beneficial "ashbed" effect, but also may kill harmful pathogens and beneficial symbiotic bacteria, insects, etc.; may seal the soil pores on the surface leading to faster runoff; and boost weed growth, thus affecting planted seedlings. On the other hand, controlled early burning, having a cooler burn and being easier to supervise, may be beneficial. The pros and cons of burning during site preparation need to be evaluated; (c) manual soil manipulation is labor-intensive, while machines are capital-intensive, have adverse effects on soil by compaction and are difficult to operate in steep rocky sites. However, tractors equipped with rippers can shatter rocks, makes trenches, terraces, and ditches economically which assists moisture infiltration and in- situ moisture retention. In some instances, soil manipulation increases land productivity, while in others it may contribute to accelerated erosion; hence, the situation on each site needs review. - 119 _ Decisions regarding the above questions require an understanding of context: (a) these operations are generally in hills and vulnerable to ero- sion; (b) highly degraded lands are frequently in areas with acute shortages of fuelwood and fodder; and (c) where there is a high level of unemployment and poverty. Technologies which promote soil and moisture conservation, increase productivity of biomass, provide conditions for reducing poverty and generate employment are obviously more appropriate. However, there cannot be one model to cater for the plurality of requirements. Each site, with its individual micro- and macro-environment comprising geographic, historical, social, political and ecological aspects, needs a customized model. Planting Methods Alternative planting methods involve direct sowing of seed, planting of bare root or containerized seedlings. Other methods such as stumping, grafting, and root suckering are rarely used in large-scale plantation opera- tions and are not discussed. Direct Sowing. 'Sowing of seeds has not been used to its maximum potential in enrichment planting and forestation. The high mortality of seed- lings emerging from direct sown seeds compared to those raised by planting seedlings have made it unpopular. The mortality is caused principally by a combination of post-emergence drought and weed competition. Many of the com- monly planted exotics have tiny seeds (for example, Eucalyptus spp.) and emergent seedlings are small, needing special care and attention which are seldom provided. The unpopularity of direct sowing is unfortunate, because it is a low-cost method of regenerating indigenous species, seeds of which are seasonally plentiful and locally available. Extensive areas can be vegetated by direct sowing with less manpower and infrastructure than formalized trans- planting if some special precautions are taken. Firstly, selection of species is important. Seeds should have low dormancy and high germination percentage, and the emerging seedlings should have the capability to survive weed competi- tion. Secondly, more attention than hitherto should be given to breaking dormancy and reducing germinating time by proper seed treatment. Thirdly, where insect or fungal,damage, erratic rain, or low soil fertility are expec- ted, seeds should be pelletted with a poly-layered coating which may contain pesticides, anti-drying chemicals and fertilizers. In addition, seed should be coated with mycorrhizae or bacteria, as may be necessary if the species belong to the families Pinaceae, Casuarinaceae or Leguminoseae. Broadcast sowing, patch and line sowing or aerial seeding may be used. Planting Eucalyptus spp. by direct seed sowing has been successful in Australia. Acacia auriculiformis is successfully grown by line sowing in plantations in West Bengal, India, as is Dipterocarpus turbinatus in Bangladesh. Aesculus indica and Juglans spp. are' grown by direct sowing in the Kashmir Himalayas, India. Shorea robusta and Dipterocarpus macrocarpus also grow well from direct sowing. Prosopis juliflora has been grown very successfully by direct sowing in some arid states of India. All these exam- ples show the potential of sowing as a method of plantation establishment. Aerial sowing of pelletted seed is an interesting option, which has been suc- cessfully utilized in New Zealand, Canada and Indonesia. The technique may be valuable in arid and semi-arid wastes and is a possibility that should be explored. - 120 - Bare-Root Planting. Bare-root planting is less expensive than con- tainer seedling planting, due to lower nursery and transporting costs, and should be adapted wherever possible. While some species are very sensitive and cannot stand bare root transplanting (e.g. Acacia catechu, A. nilotica), others are amenable, provided the seedlings are carefully handled. Eucalyptus tereticornis, E. camaldulensis, Casuarina equisetifolia, Pinus roxburghii are some of the species which have been successfully raised with bare-root stock. Container Seedlings. The use of container seedlings is the most common planting system used today in revegetation programs. Two types of containers are used: semipervious and impervious containers, with the latter used more extensively. Semipervious (and pervious) containers were exten- sively used before polyethylene bags were introduced. These were made up of leaves (Shorea robusta, Butea frondosa, Bauhinia villosa in India), waste veneers (by Picop in the Philippines), or bamboo pots (by Nalco in the Philippines), all locally available and low-cost. One of the disadvantages of semipervious containers is that roots penetrate the container and form a mat with the roots of adjacent containers, making it difficult to separate them during transplanting. This type of planting stock has less mortality than bare-root stock. The semipervious containers are not used widely anymore, but renewed attention to this method and evaluation of its potential would be useful. Two types of impervious container, the flexible polyethylene bag or sleeve and the more rigid root-trainers, have been developed over the past decade. In tropical countries, flexible polybags or tubes are most commonly used: they are frequently 4-mm thick black or transparent polyethylene and of varying diameter and length. Disadvantages are that they are not biodegrada- ble, add to seedling production costs and may constrict root development. However, advantages are many: seedlings have a better chance of survival, planting stock size can be controlled and there is little damage during trans- portation. To prevent root-constriction, root-trainers have been developed. They are normally hung slightly above the ground, causing air pruning of roots and have the additional benefit that seedlings do not significantly deterio- rate if held back in the nursery due to erratic rains or other delays in planting schedules. Issues Concerning Planting Methods. The following issues are impor- tant regarding planting methods: (a) direct sowing as a method of revegetation is unpopular. As the method is low-cost and has a lot of potential, provided the seeds are pretested and pelletted, a fresh look at this method for extensive use is recommended; (b) bare-root seedling planting is successful for many species, provided seedlings are planted during good climatic conditions. Containeriz- ing them to ensure success is not necessary under such circumstances. It would be worthwhile to list species that do well by this method of propagation and standardize the procedure of growing bare-root seed- lings in nurseries; - 121 - (c) semipervious containers, particularly of leaves, once popular have been practically discarded. Yet the method has potential. Not only are they low-cost but also biodegradable. Species which can with- stand some root disturbances during transfer can be successfully grown in these containers; and (d) containerized seedlings often have abnormal shoot/root ratios and coiled root systems. Controlling size of shoot and shoot/root ratio through the use of the best container size needs to be introduced as a standard nursery practice. For species requiring containerization, root trainers should be used where possible. Planting DesiRn Planting design involves the pattern and spacing between trees. The pattern commonly used is the square or rectangular. In the square pattern, the distances between plants in the row and between rows are the same. In the rectangular pattern, distances between rows exceed those in the row. Examples are: Eucalyptus deglupta is spaced at 4 m x 4 m, Paraserianthes falcataria at 4 m x 2 m in the Philippines and Araucaria cunninghamii 3 m x 2.50 m in Papua New Guinea. Planting pattern and spacing are determined by the silvicultural requirements of the species, the objectives for which the trees are being raised, plantation management methods and cost considerations. Species with spherical crown shape (for example, Mangifera indica) are appropriate for the square pattern, but those with a cylindrical form are better suited for rec- tangular planting. Some species are raised very close in a square pattern for self-pruning in the early period of life to obtain knot-free timber. Larger rectangular spacing, on the other hand, can provide open space between rows for agroforestry. If management aims at getting intermediate yield, closer planting initially allows mechanical thinning at a young stage. This allows for recovering some of the higher investment cost due to the greater number planted. These planting patterns and spacing are suitable for industrial plan- tations but inappropriate for revegetation where soil and moisture conserva- tion and production of fuel and fodder are more important. In this latter case, planting on the contour in V-ditches or the equivalent, with contour barriers of suitable shrubs or vetiver grass is the preferred option. The pattern and spacing of trees would be approximately 1 m apart in the rows and about 3.5 m between rows, depending on the slope. The steeper the slope, the lesser the distance with contour hedges of grass or shrub grown between tree rows. Issues Concerning Planting Methods. Planting design for watershed revegetation should avoid the usual square or rectangular pattern and spacing followed in plantation forestry. The dual objective of providing benefits to people and of soil and moisture conservation is best served if trees are planted close in rows in contour V-ditches, or in pits, and between rows a contour hedge of useful shrubs or grass. The hedge will assist in soil and moisture conservation, provide twigs and leaves for low-quality fuel, and the trees can supply most of the other products needed by people, including fod- der, poles, small timber, etc. - 122 - Protection Protection of revegetated sites during the establishment period may be done by fencing, watch and ward, social consent or by a combination of these methods. Fencing types used include cattle-proof ditches, stone walls, barbed wire fences, electric fences and hedges. Watch and ward employs per- sons to guard the plantation. Social consent infers that the people volunta- rily desist from using the plantations for a certain period. Protection by social consent is the most appropriate and inexpensive method but also the most difficult to execute. A minority group or even one person in a village may damage the plantation. An unconcerned man may accidentally burn the young crop. Stray cattle, by browsing among the seedlings, may put back the growth of trees by several months. Fencing becomes ineffective if local people do not accept it. Protection by watch and ward is effective if people cooperate with the guard. Hence, low-cost fencing, such as a vegetative hedge, and watch and ward with community cooperation perhaps would be the ideal combina- tion. Fencing. Several options in terms of efficacy, cost, material, and level of technology and maintenance pertain to the use of barriers to protect plantations in upland Asia. Cattle-proof ditch is a common method of fencing in many parts of India. They are dug around the plantation, having vertical side walls, variable sizes (generally 60-90 cm wide, 100 cm deep) with the soil heaped as a ridge on the plantation side. The trenches are interrupted by half-cut transverse walls at intervals to ensure that the trench does not form a running gully. The trenches are not desirable for a number of reasons. They are expensive, costing about 252 of the total plantation cost, hard to dig where there are rock outcrops on the surface or at shallow depths, promote severe erosion where ditches run along the slope, and are ineffective if any part is breached or filled with soil washing down the ridge or collapsed side walls. Stone walls are convenient if sufficient stone is available on site. The walls should be at least 1.5 m high and 0.3 m wide. Often in the hills, irregular shape and hence lengthy boundaries make walls very expensive. Barbed wire fencing is very effective against men and animals and can be recommended, even if expensive. Wire is often stolen as it has resale value and continuous repair makes it even more costly. Electric fences have been introduced recently in parts of India and Southeast Asia, particularly to protect crops from cattle and wild animals. Their use can be totally negated by human interference as damage to one portion makes a large part inoperative. In addition, forest workers in many developing countries are not trained to keep them under repair. - 123 - Live hedae is the least-cost and most appropriate form of fencing. Selection of species for the hedge can be made from a large number of shrubs in each site. In India, some common he,dge plants are Agave spp., Iomea spp., EuDhorbia app., Duranta app., Cactus spp., Vitex spp., Barberis spp. Such hedges not only protect the plantation, but can provide fuelwood and other products if properly managed. The difficulty with the hedge as a fence is that it has to be raised at least two years in advance of plantation planting, unlike other fences which can be built in the year of plantation. Watch and ward is the most common method of protection in Bangladesh, Burma, India, Nepal, Pakistan, and Sri Lanka. Guards selected from the local village are effective if villagers cooperate. A guard can look after an area of 10 hectares in the plains and about 6 hectares in the hills. Employment is a recurrent cost and can be prohibitively expensive if a guard has to be employed beyond the first 1-2 years. Social Consent. People will normally agree to protect the plantation voluntarily if they are involved in planning, execution, management and are recipients of the benefits that may accrue from it. But rarely in the past have villagers been consulted in revegetation projects in the uplands. Aguilar (1982) reports indifferent success of four forestry projects in upland Philippines due to noninvolvement of people. Many community woodlands raised under social forestry projects in India have been destroyed by the people themselves. On the other hand, there are examples of forest protection by social consent. Villagers of Chamoli district in the Western Himalayas of India, who initiated the Chipko movement to ban Government tree felling in 1972/73, have now focused their attention on raising plantations and protecting forests (Agarwal, 1986). In West Bengal, villagers on their own initiative protected about 120,000 ha of Shorea robusta forest by 1973 (Banerjee, 1978) and 150,000 ha by 1988. Eckholm (1979) reports successful planting of over 643,000 ha of village woodlot by village forestry associations in South Korea. Issues Concerning Protection. Protection by social consent is the most inexpensive and desirable method of protection. But a minority of one uncooperative person may damage the revegetation effort. Hence, social con- sent combined with some form of fencing or watch and ward is necessary for protective fencing. Among fencing options, the live hedge seems to be the ideal method. Not only does it protect the area from stray cattle, but the hedge can be managed to provide fuelwood and other useful products for distri- bution to people. The difficulty with the hedge method is that it should be established at least two years in advance to be effective by the year of plan- tation. Other forms of fencing using materials such as barbed wire or stones, though commonly used, are generally ineffective, expensive, and should be discontinued wherever feasible. Cattle-proof trenches, as used in India, are a source of erosion and should be discontinued. Management For this discussion, the term management refers, narrowly, to only silvicultural practices that exploit a revegetated area for sustained yield. As mentioned earlier, soil and moisture conservation and meeting local needs - 124 - for fuel, fodder and food are the primary objectives of revegetation. Manage- ment practices, therefore, have to be oriented-towards these objectives, yet few examples of managing upland revegetation can be cited that have attempted to attain these objectives simultaneously. Current management practices can be broadly divided into two catego- ries: protection and fixed-rotation harvesting. The first involves a hands- off approach that uses vegetation as a protective cover and induces indirect' yields through water and soil conservation. Such areas are fenced and looked after by intensive watch and ward for a few years to keep local people and cattle out. Hardly any silvicultural treatments are carried out, except for weeding, fire protection and climber cutting in the initial 2-3 years. The second involves exploitation in short rotations'. Usually such areas are cop- piced (5-15 year rotation) if the species are suitable, or they are clear- felled and the area replanted. Eucalyptus spp., Leucaena leucocephala,' Prosopis Juliflora are examples of species having coppicing ability. Those clearfelled include Acacia nilotica, A. auriculiformis, Cassia siamea, Faidherbia albida and Pinus roxburghii. Before harvest by coppicing or clear- felling, there is no other operation except grass cutting or-weeding. Both are, to a large extent, adversely affected by local disturbance through illi- cit grazing, hacking and intentional burning. The objectives of soil and moisture conservation, on one hand, and fixed-rotation harvesting for supply of products to upland villagers, on the other, can be in conflict unless management methods are substantially changed. Disturbance of the understory and forest floor (litter, humus layer, root mat) during the removal of trees in upland areas exposes the soil and accelerates erosion. At the same time, harvesting trees is essential to provide round wood for people. The supply of forestry, products only at the rotation age (say at 5th, 10th, 15th, 20th years if the rotation age is 5 years) is unsat- isfactory, as the community needs small amounts of products such as fuelwood and fodder regularly. Silvicultural operations such as annual brashing, low and high prun- ing, pollarding, coppicing, selective thinning every year are some of the management methods most applicable for arranging regular supply. For example, on a plantation of Prosopis iuliflora and Acacia nilotica in Karnataka, India,' the rotation of A. nilotica is fixed at 15 years, while P. iuliflora is regu- larly lopped and the harvest supplied to people after it has reached the age of three years. P. iuliflora continues to develop leafy shoots which act as a soil -cover in spite of annual lopping. Another interesting;example of dual- purpose management is that of Leucaena leucocephala regenerating profusely under its own shade in Haryana, India. The overhead canopy is retained with a long rotation, while cattle are allowed to graze the vegetation daily. Annual silvicultural operations such as brashing, pruning,'thinning and coppicing are expensive and ought not to be done by regular employees. Under the technical guidance of the responsible line department, beneficiaries should take over the management and benefit distribution of such plantations. Social forestry projects in India have community land and roadside plantation components that aim at local management. Dani and Campbell (1986)-report that - 125 - one third of the villages in Nepal have at least one local forest under local management and the number is increasing. Issues Concerning Management. Very few examples are available in upland revegetation programs of successful management to attain soil and mois- ture conservation and regular production of goods for local people. Either they are managed as protection forests, or else they are cut at fixed-rotation intervals to produce fuel and other products. The dual-purpose technology for soil and moisture conservation and production of goods in the same area has to be designed. Raising of shrubs and grasses, such as vetiver grass, on con- tours as hedges under the overhead tree canopy is an appropriate design for this dual purpose. The hedge can act as the soil and moisture trap and sup- plier of twigs and branches for fuel, and the tree the major supplier of other products. Regular supply can be arranged only if annual silvicultural opera- tions such as brashing, low and high pruning, topping of the hedge and thin- ning the top canopy when possible are introduced. The hedge should be a spe- cies that can withstand annual cutting and topping and yet develop enough shoots to protect against soil loss and continue as a hedge. All these opera- tions are expensive and cannot be repeated unless the management is carried out by local beneficiaries who voluntarily execute the job for their own consumption. CONCLUSION Upland revegetation efforts in Asia have seldom been successful. Although failure is generally attributed to techniques, in many instances biotic factors such as cutting, unrestricted grazing, and intentional burning by the local population have dominated. The introduction of technology with- out the active participation of local inhabitants in the revegetative program is ineffective. Their participation can be assured only if they are involved in different stages of the revegetation program including benefit-sharing. The technology should aim not only at producing fuelwood, fodder, and employ- ment opportunities and at promoting soil and moisture conservation, but also at involving them in its adaptation and use. A review of available technology reveals a number of shortcomings. Choice of species has overemphasized trees, exotics and monoculture. It is argued that the introduction of shrubs, indi- genous legumes and grasses, on the contour as a hedge, should find a place in land stabilization and the regeneration of vegetation. Ground preparation includes removal of ground vegetation, windrowing, burning, and soil manipulation. Removal of ground vegetation, an established practice, may have serious erosion implications, although planted seedlings cannot compete otherwise with existing vegetation. Burning supposedly releases nutrients from biomass waste, but may also increase erosion. In soil manipulation, pitting is the common practice in Asia, often done manually and rarely by machines. In planting methods, reintroduction of direct sowing and more use of bare-root planting and biodegradable containers to reduce cost are proposed. In planting design, it is suggested that planting with a shrub hedge on the contour in between the tree rows should replace the present method of square planting. Regarding protection, social consent with watch and ward or with a live hedge raised at least two years in advance is prefer- able to other forms of fencing. Management of revegetated areas presently - 126 - comprises two methods: complete protection without any harvesting and regular harvesting at the rotation age. Both methods fail to consider the needs of local villagers for a continuous supply of products. This can be satisfied by introducing silvicultural practices such as annual brashing, low and high pruning and topping of the hedge and thinning of the top canopy of trees. - 127 - Table 5.1: TREE SPECIES CITED Genus Species Common Name Acacia auriculiformis Northern black wattle, ear podwattle Acacia catechu Khair (Nepal) Acacia mangium Mangium, brownalwood, hickory wattle Acacia nilotica Egyptian thorn; red-heat, kudupod, wsweet smell," babul (India); kiker, bhbar (Pakistan); sunt, ruikperul (Arabic) Aesculus indica Lekh-pangar, Nark (Nepal) Araucaria cunninghamii Hoop pine Araucaria huntsteinii Klinki pine Azadirachta indica Neem, nim Bauhinia villosa Orchid tree Butes frondosa Flame-of-the-Forest, dhak (India), Palen* (Nepal) Cassia siamea Yellow cassia, minjri, moung, angkanh, kasof- tree, Bombay blackwood, cassia (French) Castanse app. Chestnut Casuarina equisetifolia Casuarina, Australian pine, agoho (Philippines), ru, atru (Malayasi), nokonoko (Fiji) Cedrus d-odara Deodar; Dior, Devadaru (Nepal) Cryptomeria japonics Sugi (Japan); Dhupi saIla (Nepal) Dalbergia sissoo Sisso, sarsou, shisham, nlkaer, karri, tanach, tali, yette Dipterocarpus macrocarpus Hollong (India) Dipterocarpus turbinatus Carjan Eucalyptus camaldulensis Red gum, river gum Eucalyptus citriodora Spotted gum, lemon-scented gum Eucalyptus deglupta Kamarere (Papua New Guinea), bagras (Philippines), leda (Indonesia) Eucalyptus gomphocephala Tuart Eucalyptus grandis Flooded gum, rose gum Eucalyptus pilularis Blackbutt Eucalyptus tereticornis Blue gum, mountain gum, red gum; Mysore gum (India) Faldherbia aIbida (Acacia albida) Apple ring acacia, Haraz Juglans app. Walnut Kydia calycina Kubindeh (Nepal) L,ucaena l-ucocephala Leucaena, ipil-ipil (Philippines), subabul (India) Mangifera indica Mango, amp (Nepal) Paraserianthes falcateria (Albizzia falcataria) Beta!, Molucca albizzia, sengon, falcata, vaival, puah, white albizzia, tamalini, marfa, placata, djeungjing Pinus caribaes Caribbean pine, putch pine, pino de Is costs, pino clorado Pinus halepensis Aleppo pine, pino carrasco, sansoubar halabi Pinus patula Patula, Mexican weeping pine Pinus roxburghii Chir pine Prosopsis juliflora Mesquite, algarroba Pseudotsuga menziesii Douglas fir Quercus incana Bluejack oak, sano banjh (Nepal) Robinia pseudoscacis Black locust, robinia Shorea robusta Sal Tectona grandis Teak, sagwan, sengon, - 129 - REFERENCES Agarwal, Bina. 1986. Cold Hearths and Barren Slopes. Riverdale Co. Publish- ers: Maryland, USA. Aguilar, Filomeno V., Jr. 1982. 'Social Forestry for Upland Developmentt Lessons from Four Case Studies.' Institute of Philippine Cultures Quezon City, the Philippines. Banerjee, A.K. 1987. Microplanning--A Tool for Social Forestry Implementa- tion. National Wasteland Development Board: New Delhi. _ 1978. 'Local Needs of Small Timber and Fuelwood in India,' in K.R. Shepherd and H.V. Richter (eds.), Forestry in National DeveloDment. The Australian National University Development Studies Center, No. 7. Cassells, D.S. 1983. 'Reducing the Impact of Plantation Establishment on Soil and Water Resources--Some Research Requirements." in K.R. Shepherd and R.O. Squire, Establishment of Coniferous Plantations. Australia. Chapman, G.W. and T.A. Allan. 1978. 'Establishment Techniques for Forest Plantations,' FAO Forestry Paper t8. Rome. Dalwaulle, J.C. 1977. 'Species, Techniques and Problems of Semi-arid Zones (the Sahel)' in Savannah Afforestation in Africa. FAO: Rome. Dani, Anis and J. Gabriel Campbell. 1986. 'Sustaining Upland Resourcess People's Participation in Watershed Mangement,' ICIMOD Occasional Paper No. 3. International Centre for Integrated Mountain Development and FAOQ Kathmandu. Eckholm, Erik P. 1979. 'Planting for the Futures Forestry for Human Needs,' World Watch Paper #26. World Watch Institute: USA. Evans, J. 1982. Plantation Forestry in the Tropics, Clarenden Press: Oxford. Fernandes, Walters. 1987. 'Voluntary Organization in Afforestation and Energy Savings Project,' in India Papers. International Tree Project Clearinghouse: NGLSlNew York. Hegde Pondurang. 1987. "Appiko (Chipko) Movements Lessons of Grassroot Action,' in India Papers. NGLS/New York. Leach, Gerald. 1987. 'Household Energy in South Asia,' Biomass 12. Rao, E.V. 1967. 'Economics of Plantation particularly of Casuarina and Euca- lyptus with special reference to the sandy soil of Nellore, south divi- sion of Andhra Pradesh," Proc. Eleventh Silvi. Conf. FRI, Dehra Dun. Rockwood, W.G. and R. Lal. 1974. 'Mulch Tillage--A Technique for Soil and Water Conservation in the Tropics,' Span 17. Shepherd. Kenneth R. 1986. Plantation Silviculture. Martinus Nijhoff Pub- lishers: Boston. -131- 6. LAND TENURE TSSUES IN WATERSHFD DEVELPMENT Augusta Molnar Intended to document why tenure is relevant to watershed projects and to provide a conceptual framework for making decisions related to tenurial issues during project design, this chapter is grounded in a review of the literature on land tenure and conservation and a review of relevant proj- ects in China, India, Indonesia, Nepal, the Philippines, and Thailand. It begins with an overview of the importance of land tenure in watershed projects, describes some gen- eral principles of land tenure and categories of tenure systems found in the Asian uplands, discusses why the rela- tionship between conservation and tenure has not been clearly understood, presents a conceptual framework for analyzing that relationship based on case studies, identi- fies measures for addressing tenurial issues in watershed projects, and, finally, suggests studies that should be undertaken to further the World Bank's operational under- standing of this issue. INTRODUCTION Basic Obiectives of Watershed Proiects Watershed development projects aim at improving the overall produc- tivity, sustainability and equity of land use in fragile, arable and nonarable lands. Land tenure can be an important factor in achieving these goals. Productivity is linked to tenure because tenurial arrangements can affect both the profitability of present and improved farming systems and the ability of the cultivator to make the transition to an improved technology. Sustainabil- ity is linked to tenure because tenure affects both the initial decision to adopt practices to slow land degradation and the ability to maintain land improvements over the longer term. Finally, equity concerns are linked to tenure because project investments can reduce people's access to resources, including land; and new technologies may induce changes in land tenure, dis- placing people from their land and their employment base without providing viable alternatives. It is generally difficult to identify strategies that optimize all three of the above objectives. Choosing a strategy for addressing tenure issues will inevitably require some trade-offs in project goals. For example, in the Amazon region, titling programs to induce squatters to adopt a long- term view of their holdings also resulted in increased land speculation and the opening up of new areas for agriculture, ultimately leading to further land degradation (Cullins and Painter, 1986). In the Southeast Asian uplands, private agriculture on steep slopes is not the ideal use of the land, but because of population pressure, improving local agricultural practices is more sustainable and equitable than trying to eliminate such land use (World Bank "ARM Study, 1989). Thirdly, closing lands to open grazing in South Asia and -132- inducing graziers to adopt stall-feeding may lead to more sustainable or pro- ductive land use, but land-poor herders without pasture who cannot afford to stall-feed may be forced out of animal husbandry. Each case involves diffi- cult choices. Tenurial Issues in Watershed Projects Land tenure includes the formal (state-recognized) and informal (cus- tomary) rights of access to different kinds of land, the rights to control products of that land, obligations to maintain the land, the rights of trans- fer, and rights to determine changes in use of that land. Formally and infor- mally held rights in land may be in harmony or there may be conflicting rules within the system.. Land tenure is not unchanging, but responds to new eco- nomic, social, political, and cultural factors on a continuous basis. Often, changes occur in the system when different interpretations of rights--by other farmers, the village council, the state, the forestry bureaucracy, or the courts--gain precedence in the resolution of a tenure conflict. There are several reasons why it is important to be concerned with land tenure issues in watershed projects, all of which can be illustrated by examples from recent, World Bank-assisted upland projects. (a) the type of tenure may affect the profitability, adoption rates, and the impact of proposed measures; (b) efforts to change land tenure patterns to provide land-users with better incentives for sound management may have unintended, negative consequences; (c) introduction of specific soil and water conservation technologies may affect land tenure patterns and have an unanticipated impact on local smallholders; and (d) the failure to understand existing tenure patterns may lead planners to overlook promising opportunities to develop lands under particular tenure arrangements, especially informal systems or systems of group- management. Land Tenure Systems in the Asian Uplands There are many systems of land tenure in the Asian uplands adapted to specific ecological/economic conditions, cultural/historical traditions, and population densities. A summary of the range of tenure types is presented in Table 6.1. Some of the most familiar systems in Asia are: (a) watersheds with relatively stable land tenure systems, such as the island of Java or Taiwan. These have few communal lands, more or less clearly demarcated state-owned lands, and complex owner-cultiva- tor and tenant-cultivated arrangements, many of which are not clearly reflected in the official land registers; -133- (b) watersheds with relatively ancient, state-recognized land tenure systems, such as Burma, China, India and Nepal, where collectively- managed resources, usually called common property resources (CPR), are an important part of the formal and customary (informal and de facto) tenure rights; and (c) watersheds in "frontier' areas, such as the Philippines, north Thailand, Malaysia, or the Outer Islands of Indonesia, where a limi- ted proportion of cultivators has state-recognized tenure rights, and a large portion of forested and unforested land is usually designated as state forest. Cultivators in such areas include indigenous ethnic minorities (including tribal groups) who have a long-standing land- use traditions and informal tenure based on ancestral rights; set- tlers, often belonging to majority ethnic groups whose actual tenur- ial arrangements tend to emulate those elsewhere in the country; and recent migrants on newly-cleared lands. These tenurial differences are Important because each system may lend Itself to different forms of technical and social intervention. Tabj I a 1 CHARACTEISTICS OF DIFFEPB4t TYPES OF TENRE ARRAN4BMTS IN THE ASIA RiBION RELEVANT TO THE ADOPTION OF SOIL AND MOISTURE CONSERVATION Cu ranteed Abi I ity long-term of user Ability acc-s- Access to to make to protect Perentae of to land cr;dit/ dev*loement land froe returns to Ability to transfer Tenure category (security) Capital decisions others user of land rights of ownership Privately-owned with High High High High High High title Prlvately-owned without Medium Medium High High High Medium Le and from relative Low Medium Medium High Medium Loe Fixed renter from unre- LOe High if owner Lea Medium/high Medium LOW I*ted owner contributes Sharecropper from resi- Loe Lew (medium LOe High Low to medium Low dent owner if owner contributes) Sharecropper from Low Low In gen- Low High LOW Low absentee owner oral Private cultivator on Low Low High Depends on High if not LoW pub If lends management paying rent system to peudo- landowner Perum Perhutani model Medium High Medium/high Low/medium High Low at present of Ionger taungye I agse (Java) Forest association. Medium High Low Medium High Low at pre"en (FOSAs) (Philippines) Local grezier on vil- High Low Low Depends on Depends on Depends on lages common land management management management system system Local grazier on state Low Low Low Low Depends on Depends on revenue lands management management system Tres tenure/leases on Medium Low High Low High Low state revenue land (Ind ia) Cost-sharing of forests High High Low High High (could High within village giving uee~r protec- be higher if tion dutios (West products Bengal, Indi.) expanded) Recognizing customtry High Low Depends on High Can be high High within village rights over forests Foreatry Department/ village relations -134- Pooulation Pressure and Land Tenure One characteristic that makes land tenure systems so distinctive in most of the Asian uplands is the high population pressure on scarce resources. While population densities in the lowlands are widely recognized as affecting socioeconomic conditions in Asian countries, their effect has not been as apparent for the uplands, because the high proportion of nonarable lands tends to make densities seem low. Statistics comparing lowland and upland popula- tion densities are misleading, because they do not usually factor in the limi- ted amount of cultivable land in the upland areas. In Thailand, Cohen (1983) calculated the number of persons per km2 of cultivated land and found that pressure exceeds 500 persons/km2 in 58 out of 64 districts of seven northern provinces. Comparable figures for other upland areas are 560-1,000 persons/ km2 for upland Java, 530 persons/km2 for upland Nepal and 215 persons/km2 in the Philippines' uplands. If the lower productivity of upland soils is factored in, this is a striking picture. Given the importance of population density in determining land use in much of Asia, this review has focused on case studies of tenure systems and projects in relatively populated watersheds; examples are also included from the relatively underpopulated Outer Islands of Indonesia, parts of Malaysia, Bhutan, and Papua New Guinea, where the tenure issues and trade-offs are less constrained by competing demands on scarce resources. Factors Obscuring the Impact of Tenure in Watershed Proiects The data are ambiguous regarding the effect of tenure on adoption rates in ongoing or completed watershed projects and the analysis has been confused by a number of factors, such as quality of the technology packages, subsidies, and lack of knowledge about the land tenure situation in project areas. (a) Many of the technologies promoted (gully plugs, bench terracing, vegetative contouring or afforestation) have not been as economically attractive as anticipated. Either they have been too expensive (in terms of labor and capital) for the cultivator to adopt, given the long time before benefits start to flow, or they have not led to the projected increases in yields and cash returns, with the result that neither owner-cultivators nor tenants had an incentive to adopt. Projects have also ignored variations in the availability of house- hold labor and assumed profitability would be the same in differing households. In such cases, land tenure has often been cited as a problem, whereas the issue is related to poor choice of technology. (b) Because of the high, initial cost of adoption, many technologies have been heavily subsidized. In such cases, it is impossible to separate the effects of land tenure on adoption rates from the incentives provided by the subsidy. In some projects, in sites with considera- ble differences between recorded tenure rights and actual tenurial arrangements, project staff have no idea whether subsidies were paid to owners or cultivators/ land managers, so separating their effects becomes doubly difficult. -135- (c) There has been little monitoring of the effect of tenure on adoption rates. Where data have been collected, project staff have often failed to understand the tenure system and either lumped a number of distinct tenurial types into one analytical category or failed to recognize the existence of a range of customary land tenure arrange- ments. In India or Nepal, common property resources (CPRs) have been falsely perceived either as open access or private lands, when in fact there were effective, informal systems for their management. Alternatively, truly open access lands have been lumped together with collectively-managed lands for purposes of analysis, leading to an overly pessimistic picture of the scope for CPR management. In both South and Southeast Asia, corporate-owned private lands have been mixed with individually-owned lands, with negative results when development programs attempt to interact only with individual land- owners. Improving Analysis of Land Tenure Factors A significant problem with attempts to evaluate the effects of land tenure is that studies have been too simplistic. The commonly studied, prede- termined sets of tenure categories--owner-operator versus tenant-operator on private lands and private land tenure versus public land tenure for the system as a whole--are largely meaningless for analyzing the relationship between tenure and adoption rates in most Asian settings. A sound study would first identify the full range of tenure types in the project area, the characteris- tics of tenure that affect the adoption of proposed technologies, and then select the tenure types that should be evaluated. In a dynamic tenure system, there are multiple rights and obligations, not just the rights of long-term access to the land commonly associated with the word *title.' These other rights will affect adoption rates. Another important point is that distinct tenure types represent points on a continuum in any tenure system and may not fall squarely in a preset category such as "privately-owned,2 "corporately-owned,w "communally- owned," or 'publicly-owned." This is not to say that the rights and obliga- tions pertaining to a tenure type are ambiguous, but that these groupings obscure important tenure dynamics. For example, village grazing lands in northwest India include shamlat i deh, a tenure type which allocates a legally specified set of landowning households primary rights to that grazing land in proportion to the size of their private holdings in the village. Resident landless families have only secondary rights to grazing that land at the behest of village landowners. In addition, any trees planted in this "common" or "corporate" land belong to the person who planted them. This "tree right" is not found in village lands in other regions of the country. Shamlat i deh thus functions as a place to plant privately-owned trees, a corporately-owned grazing land, and through the concessional rights to landless, as communal land (Stewart, 1989). In addition, since anyone is entitled to right of way for this land, it functions as a public land. In the conceptual framework presented in Table 6.1, there has been no line drawn between public and pri- vate land tenure types. Some of the tenure types on the public end of the continuum have as many characteristics which support adoption of improved land practices as tenure systems on the private end of the continuum. -136- CHANGES IN LAND TENURE AND THEIR IMPACTS Impact of Tenure on Adoption of New Technologies In the current World Bank-assisted watershed development projects in the Asia region, a number of soil and moisture conservation technologies are being promoted, including: (a) structural works, such as bench terraces, contour banks, gully plugs and water diversion channels and drainage channels, to manage surface runoff and sedimentation; (b) vegetative barriers on the contour, such as vetiver grass, fodder grasses, or mixed conservation hedges of shrubs, grass, and trees to keep moisture and silt on the site; (c) cultural treatments and improved cropping systems to improve farm production in a manner that protects soil and moisture status; (d) agroforestry technologies that provide a long-rotation crop for steep, fragile soils; and (e) silvipastoral or pastoral technologies that provide ground cover, fodder, and forest products. A pertinent question in the Asian watersheds concerns the extent to which tenurial arrangements affect the adoption rates of these technologies. A logical assumption has been that cultivators will be less likely or unlikely to adopt technologies which have a long time-lag before they begin to generate benefits, or stated in a another way, which require a high level of investment without a commensurately high, short-term return. Following from this assump- tion, where tenure status is weak, the most appropriate technologies to pro- mote should be those with quick returns and minimum levels of investment. These include the vegetative barriers and mixed silvipastoral models mentioned above. Tenure has a significant impact on the profitability, adoption rates, equity, and long-term sustainability of interventions. The farmer's sense of long-term tenure security will affect his willingness to undertake land improvements. Private cultivators with secure title rights, tenants who are relatives of the owners, or corporate groups with secure land-use rights are more likely to take a long-range view of land productivity. Sharecroppers with short-term or indeterminate contracts, nonresident landowners who are more concerned with land as an investment than in its annual productivity, and village graziers with only concessionary rights to forest or grasslands are all unlikely to take a long-range view of the resource. Tenure may also affect a farmer's access to credit for land improvements where written title is a prerequisite of creditworthiness. Effect of Tenancy on Improved Practices Tenancy arrangements affect the profitability of an intervention differently, depending on the rent paid out, the terms of produce-sharing and -137- the length of the contract. There is a wide range of tenancy arrangements in the Asian uplands, and when targeting technologies, it is useful to understand the particular range found in the project area. The case-study data from India and Java document many informal produce-sharing and rental arrangements that do not conform to the legal code, which makes it difficult to collect accurate data on their extensiveness (Mackie, 1989; Khasnabis and Chakravarty, 1982; Cohen, 1983). Arrangements differ depending upon the productivity of the land, the reason the owner is leasing out the land, the market orienta- tion, and the relationship between owner and tenant. Studies indicate that as many as 502 of the cultivators who rent or sharecrop land may be friends or relatives of the owner, and that these tenants receive more favorable terms and more latitude in decision-making than other tenants. In Thailand, rela- tives avoid 'key money' (right-to-lease payments made to the owner distinct from the regular terms of the actual contract) and in Java, produce shares are greater (Mackie, 1989; Cohen, 1983). Different categories of landlords also respond differently to such technologies. Some poorer farmers temporarily rent out parcels of land to get short-term capital or to deal with scarce labor problems. They will not con- tribute to inputs or encourage tenants to develop the land, lest they lose claim upon it. Another consideration in adopting new practices is whether the tenant or owner is expected to provide the inputs. A disadvantage for tenants who have secure tenancy contracts is their lack of access to credit for inputs. Case studies from South India indicate situations in which owners provide sharecroppers with inputs, either when tenants are destitute and can- not even afford seed or when the owner's investments can clearly increase profitability to both owner and tenant. In this latter situation, there may be considerable scope to encourage landowners to help tenants invest in high- return, soil and moisture conservation measures. Absentee landowners are, however, a category of owners who are unlikely to encourage land improvement. As nonresidents of the village, they tend not to concern themselves as much with long-term land productivity and tend to discourage tenants from undertak- ing improvements lest they weaken the owner's claim to the land. One excep- tion to this rule is in regions where land reform legislation is enforced, as in parts of South India, where tenants are often the main decisionmakers on the estate and absentee owners are afraid to set terms. Impact of Land Titling Recognizing the linkage between tenure and the incentives to under- take land improvements, a number of upland projects have included components to strengthen tenurial status. By far, the most common of these is land titl- ing. Watershed projects have included titling components to provide incen- tives for farmer participation in programs, to enhance the farmer's tenure security, and to provide the farmer with access to formal sources of credit. A mixed picture emerges on the relationship between titling and land improve- ments. The only in-depth study for Asia was undertaken in northeast Thailand (Feder et al., 1988). It compared the productivity of farmers with titles and those with certificates of occupancy and concluded that titles had a signifi- cant impact, mainly because of their value for obtaining loans with land as collateral. Since farmers had been resident in the area for some time, secur- ity did not seem to affect the decision to invest in the land. The Thailand study argues strongly for titling components in projects that promote soil and -138- water conservation and other land improvements in order to give farmers access to capital and enable them to transfer title and improvements to their heirs. Unfortunately, there are few comparative data to generalize to other regions or countries. The ongoing titling project in Thailand has been very successful, but no new data have been collected from it on the relationship of titling to agricultural productivity. In India, where rural credit is pro- vided at subsidy, it has usually been monopolized by larger farmers and few marginal and small farmers have gotten formal credit, regardless of whether they had titles to the land they cultivated. Some analysts of the Thai case argue that once titling programs increase the number of farmers seeking credit, a similar problem of credit rationing will occur. Where technologies require formal credit, however, as in the case of fruit tree schemes in Java, providing certificate of title may be an important inducement for participa- tion in the scheme. Lack of tenure security seems to be more important where customary rights are not strong and farmers without title are worried about eviction. Unfortunately, the lack of data makes it difficult to evaluate perceptions of security. Titling has been carried out as part of the Yogyakarta Rural Devel- opment Project in Java where customary rights are recognized, but land records still pertain to surveys carried out when the province was a princely state. No formal study has been made of its effect on productivity, though a consul- tant's report indicates that some farmers with newly titled home gardens have used their new access to credit to invest in still-untitled drylands (Mackie, 1989). Land Certificates or Land Leasing In the "frontier" areas of Southeast Asia (Malaysia, Thailand, the Philippines, Outer Islands of Indonesia, Papua New Guinea), large tracts of steep land designated as state forest land have been converted (sometimes prior to their designation as state land) to shifting or permanent cultiva- tion. As governments become more interested in preserving natural resources, there is an increasing controversy over whether to move such people out of designated forest areas or legitimize their land rights. In most areas, it has proven counterproductive to evict already-settled inhabitants of lands designated as forest. Programs to increase tenure security in such areas include formal titling of holdings, granting of limited-use certificates in state forest lands, and/or formalizing customary land rights. In Thailand, limited-use certificates have not been very effective (Feder et al., 1988), because farmers already have customary security of tenure and would only change their agricultural strategy if they have more access to formal credit channels through titling. In contrast, in the Philippines or parts of the Outer Islands of Indonesia, where residents may not have the clout to protect their holdings from outsiders, certificates may serve an important function as an intermediate tenure step to prevent transfer of land (World Bank, FFARM, 1989). It is, therefore, important that limif'6d.Y' used certificates be carefully evaluated as to the rights they are intended to provide. -139- In some areas, limited-use certificates have been used to ensure that the farmers plant only specified crops, such as trees, on steeper slopes. However, this has not been very effective. The certificates of use issued in the Philippines for state forest lands specify that farmers must put 1O of their land under trees for conservation reasons. This rule is not complied vith unless the farmer finds an economic as vell as conservation reason for adopting such a practice. Programs encouraging tribal farmers in Thailand to change their swidden cultivation practices to more intensive agriculture have also had limited success due to the poor returns from recommended technologies (Hoare and Larchrojna, 1986). Unless an intervention is perceived as eco- nomic, no farmer will adopt it, regardless of tenure. Land Consolidation and Land Redistribution Other tenure changes have been promoted through land redistribution or land consolidation programs. In both instances, there is little evaluative information and what exists does not indicate a strong relationship between implementing these programs and increasing levels of productivity and sustain- ability. Two positive examples exist of land consolidation measures. One is a World Bank-assisted Moroccan project (Meknes Rural Development Project) in which land consolidation was carried out to create more viable holding sizes for agricultural development and land stabilization (World Bank, Meknes PCR, 1987; Adams and Seddon, 1983). The success of this initiative was linked to the identification of enough different soil types so that farmers could main- tain a diverse farming strategy. The second example comes from Gujarat state in India, where the soil and water conservation service has reallocated some field boundaries along the contour in consultation with farmers to make adop- tion of conservation measures less cumbersome. In general, land redistribution and land consolidation programs are tricky to implement and require political will and an efficient bureaucracy. The evidence collected so far (Herring, 1983) is that such programs are best implemented for their equity objectives, rather than to increase productivity. There are numerous instances in which reforms were not accompanied by adequate technological packages or support services, and many of the cultivators who were allocated land did not stay on that land over the long term or rented out the land due to economic need. Special Issue: Tribal Land-Use Rights in Forest Areas To date, inadequate attention has been paid to the scope for intro- ducing improved agricultural practices that are based upon traditional systems in forest areas. Bank guidelines on indigenous peoples and development expressly state that 'the Bank will not assist development projects that know- ingly involve encroachment on traditional territories being used or occupied by tribal people unless adequate safeguards are being provided.' Government and Bank policy on this point has been confused by the general lack of under- standing of the dynamics of tenure and the resultant attempt to fit a variety of indigenous systems into tight categories of 'private farms' and 'public forest.' Despite some progress, the issues of what land-use rights should be recognized is largely unresolved in traditional swidden cultivation areas and in those areas where indigenous people may have moved in response to pressure from more powerful migrants. A case in point is the situation in the Outer -140- Islands of Indonesia where tree crop schemes have provided outsiders with title over lands that indigenous groups perceive as theirs. Where governments have been willing to recognize the rights of local people to forest lands, such as in Thailand under the Highland Agriculture Development Project, a mistake has been to try to adjust customary land rights to fit preexisting legal categories, rather than to adjust legal categories to traditionally viable systems. In the Thai case, applying the statutory land ceiling in the uplands resulted in uneconomic holding sizes, given the produc- tivity of the land and potential household economic strategies. Even with the intensification of agriculture proposed in the project, farmers did not think they could support themselves and most farmers refused to participate in the land titling program. With a more flexible body of land legislation, ceil- ings could have been adjusted in the tribal area and made to conform more closely to customary systems. The same problem has been recognized in the Philippines. There, a proposal has been made to the government to formalize ancestral, communal rights to lands in areas with tribal residents, rather than titling these lands to individuals, and preserve the more sustainable, traditional system (World Bank FFARM Study, 1989; Lynch and Talbot, 1988). THE SCOPE FOR COMMON PROPERTY RESOURCE (CPR) MANAGEMENT EmerRin2 Principles In those Asian countries with long-standing tenure traditions, including lands designated as common property resources (Bangladesh, Burma, China, India, Nepal), watershed development has emphasized individual, owner- cultivated lands to the neglect of noncultivable lands held by corporate groups or public lands managed as a common property resource (CPR). This is an important issue for watershed stabilization, since nonagricultural lands make up a large proportion of land in the watershed and can be important sources of products within a farming system. Under high population pressure, it is almost impossible to manage state lands by excluding local people. Projects are therefore exploring sys- tems for management of public lands by corporate groups or surrounding commu- nities. However, early evaluations of the World Bank's experience in South Asia in promoting common property resource management on village lands, state revenue lands, and forest lands were generally pessimistic (Naronha, 1985; World Bank, India National Social Forestry Project, 1985, Uttar Pradesh and Gujarat Social Forestry Projects, 1988; Blaikie, et al., 1985; Ljungman, McGuire, and Molnar, 1987). The conclusion has been that such strategies fail due to the heterogeneity of most communities and their control by a small group of elites, the breakdown of indigenous systems in the face of population pressure and market orientations, the privatization of many CPRs by industries and encroaching farmers, and the lack of equity in distribution systems estab- lished for products. Recently, however, more positive experiences with CPR management are being documented on both communal and state-owned lands, and projects are becoming more sophisticated in their understanding and approach to the dynam- ics of CPRs (Stewart, 1989; Dani and Campbell, 1985; Arnold and Campbell, 1986; Guhathakurta, 1989; Brara, 1987). What is striking about the recent -141- literature, however, is that the principles for effective common property resource management that are being discussed directly contradict the strategies promoted in earlier forestry and watershed development projects. Addressing Problems in CPR Management Under the World Bank-assisted social forestry project, which started in 1982, farmers in West Bengal were encouraged to plant trees on marginal lands that had been allocated to them on 99-year, unlimited-use leases as part of a land redistribution scheme. These lands had been considered too degraded to support crops, but tree planting proved to be a profitable, alter- native land-use strategy. This program, locally dubbed "group farm forestry" because of the joint responsibility taken by blocks of farmers for protecting the trees, was successful (World Bank, 1985; Molnar, 1986; Shah, 1987). How- ever, this system did not work in states which did not have active land reform programs, but tried to emulate the scheme in a more restrictive manner by giving landless cultivators plots of government land to be used solely for tree-planting. Unlike the West Bengal leases, which allowed individuals to do anything they wished with the land, in other states the lease was limited to tree planting and the lease period confined to the life of the plantation. These schemes failed to reach their targets, due to recipients' lack of capi- tal and labor, lack of confidence in their long-term use rights, the power of local elites to co-opt some lands, and the fact that communities believed the government to be alienating lands for which they had informal rights. The schemes also created a legal precedent for further undermining common property rights. More recent interventions in India and Nepal have found models for generating sustainable land management of common property resources; each shares the following characteristics: (a) management resting with a locally constituted users' group, building on both customary and formal institutional arrangements (regulating long-term access, use and protection of the resource); (b) publicly-acknowledged rights and duties of the users regarding the resources in question, and the ability of users to make decisions about development or utilization; (c) a regular flow of outputs valued by the users, i.e., at least an annual flow of produce, not merely a one-time harvest; (d) a distribution system that reaches diverse elements in the population without excluding the interests of power brokers in the community; (e) a protection system that has clear, easily enforced rules of compli- ance; and (f) a public authority (such as the forest department or district coun- cil) at a higher level that supports the local-level authority over the resource against outsiders. -142- These characteristics were missing from earlier social forestry models. Shortcomings were that management was given to an artificial, political body with no customary management responsibilities; rights and duties were not known to the majority of users (that is, formal agreements with the forest department were only shared among a few key village leaders); distribution was skewed away from local power brokers; and the protection system was alien to local customs and difficult to enforce by local people. While the problems of sustained management of state lands in upper catchments are not solved, the basic principles and how to apply them are beginning to be clarified. Lessons learned from South Asia are, first, that traditional systems of CPR management rarely provide equal returns to all users, that is, distribution of benefits tends to be skewed toward the elite and vested power brokers. Effective systems are those that provide substan- tial benefits to both resource-rich and resource-poor individuals, not neces- sarily equal benefits to both. Second, approaches must be tailored to the sociopolitical context of the project location, rather than fit an ideal type of CPR management. Legal Issues in CPR Management In devising an effective intervention, three aspects of CPR manage- ment are essential: consideration of the resource's potential productivity including the range and kinds of products that can be generated for local or national needs; review of the formal and informal institutions that govern use, access, management, and development of that resource; and knowledge of the formal, legal status of rights in a particular resource and how the legal rights relate to customary rights or to the actual use of the resource. A common mistake in project planning for nonagricultural lands has been not to investigate the legal as well as customary tenure status of lands designated as common or public. Few countries have a consistent set of rights and obligations for different users of common-property resources. There can be conflicting rights over a CPR that may each be valid from a different legal standpoint. Where customary rights pertain to a piece of land, these may be equal to or have precedence over the formal rights prescribed in written law. When designing measures to encourage CPR management, it is important to analyze correctly the interrelationship of conflicting tenurial claims to a piece of land. Where the desired intervention to strengthen management by a community of users is a legal one, the correct point of legal intervention must be well understood. Tenurial rights governing the use of a piece of land can be legally present at any of five distinct levels (C. Singh, 1989). These include: (a) customary or traditional rights (e.g., village grazing rights, tribal forest rights); (b) administrative orders regarding use of lands (e.g., forest department rules concerning collection of headload fees or forest closure orders); (c) court rulings regarding existing legislation; (d) state and national legislative statutes regarding the rights over lands (e.g., the Indian Forest Conservation Act); and (e) constitutional law regarding citizens. rights in land. Conflicts over use rights occur because there is a discrep- ancy in the rights at two or more different levels. (See Table 6.2.) To -143- identify the proper measure to strengthen CPR management, it is often neces- sary to identify conflicting claims and take legal or policy steps to allevi- ate the conflict. At each level outlined above, the solution will be differ- ent. The solution to weak CPR management may rest upon the formal recognition of existing customary rights (local, level 1), or may require changing forest department regulations (administrative, level 2), or there may be need for a formal change in the state forest legislation (state law, level 4). Table 6.2: A LEGAL TYPOLOGY OF TENURIAL RIGHTS Level ---------------------------------------------------- 1. People Customary Law ---------------------------------------------- 2. Administration Govt. Orders 3. Courts Legal Cases 4. Legislative Bodies Statutes 5. Legislative Body - National Constitution For example, a study of village grazing patterns in central Rajasthan, India revealed conflict resulting from the claims of (a) tradi- tional village users of the grazing lands, (b) panchayat (village government) officials entrusted with responsibility for these lands as the smallest unit of state administration, and (c) private individuals who had encroached upon these lands over time. When these disputes led to court cases, their resolu- tion varied. While the common ruling was to give decision-making power to the panchayat, more recent rulings have upheld the traditional rights of villagers to retain control of these lands for grazing. Thus, villagers today may be gradually losing control of grazing lands to panchayats and private individ- uals, due to their lack of awareness that the courts might rule in their favor if they sued the panchayat. The dilemma of community woodlots in the Indian social forestry pro- grams stem in part from the differing perspectives of the government imple- menting the program and the local population regarding tenurial rights over the woodlot and the responsibilities stemming from those. When a panchayat gave permission to the forest department to establish a woodlot on village grazing land, with an agreement that the forest department would recover its costs at the time of harvesting, everyone saw the agreement differently. To the panchayat, it was much like renting the land out to the forest department on a 50/50 share basis, since cost recovery usually led to this division of profits. To the local village, it meant a loss of grazing land to the panchayat and forest department, with no assured returns to the village. To the villager admonished by the forester for grazing his cattle inside the enclosure, it was evidence of the forest department's assumption of tenure over the land, even if on behalf of the environmental needs of the villager. To the forester, it was panchayat and village land, and the people were -144- responsible for protection. Thus, no local sense of responsibility for pro- tection or plantation maintenance developed in the intended direction of sus- tained CPR management. Rather than reinforcing local village conceptions of common grazing land management, which had been undermined by population pres- sure, the woodlot model introduced a new arrangement for which no one had clear responsibility. LAND USE MANAGEMENT IN STATE-OWNED FOREST LANDS Forest lands subject to claims of state ownership and traditional common property resource rights fall into several different, state-created categories: (a) undemarcated forest lands to which people often have traditional use rights for subsistence products; (b) production forests, which are lands allocated for timber production, and generally more restricted to local people; (c) reserve forests, including parks and reserves, which are closed to local people (although indigenous peoples may have customary and conflicting claims to these areas); (d) forest lands of all categories which have been put under shifting or permanent agriculture; and (e) forest lands allocated on concession or lease to industries, coopera- tives, associations, etc. (pulp industries in India, forestry associ- ations granted stewardship contracts in the Philippines, village resource societies with grass leases in Haryana, India). For the ex-colonial countries, it is important to note that forest land demarcation tends to follow patterns set up during the colonial adminis- tration (Java, the Philippines, and India), with the result that many custom- ary rights have been revoked in law and others remain ambiguous. The situa- tion is complicated in India by the fact that some areas of India remained princely states up to independence and in these areas, customary rights were never legally overridden at independence. In general, recommended measures to encourage sustained resource management on public lands include: (a) formally recognizing local people's rights of access to public lands, sometimes with written agreements (woodlot agreements in the Indian National Social Forestry Project, forestry management plans in the Hills Community Forestry Project in Nepal); (b) transferring control over resources to local groups of users or political authorities; (c) instituting systems of joint management and cost-sharing of final product between local users and government agencies; and (d) extending leases to coop- eratives or associations for forest or pasture development (Forestry Steward- shin Associations in the Philippines). These are also critical ingredients in the management of forested lands. -145- Transferring Control to Local Communities In Nepal, where few forests have been demarcated or reserved, the policy for improved forest management in the hills, and more recently in the Terai, is to transfer control for local protection and management to communi- ties over two categories of government forest lands: those rehabilitated through plantation and those already-forested lands near villages. In many areas of Nepal, this policy is simply legalizing an already existing, tradi- tional system of common property management, and thereby providing more encouragement for its sustained continuance. The present policy reverses an earlier one (1957), which had transferred control over all forest areas to the state in an attempt to arrest deforestation resulting from rapid population growth. The first step in the change transferred management to the local administrative council, the panchayat, with a forest committee organized to handle the management of the forest area. It is now proposed to transfer management to a more natural unit, namely the village or villages with tradi- tional rights of access to the forest for their subsistence needs. With this change the role of the forest department would shift from manager to advisor, assisting the local community to develop a sustainable management plan and providing technical guidance on silvicultural matters. Cost-Sharing Model of Community/Forest Department Management Transfer of resource control has been successfully implemented on a broad scale in West Bengal, India. The forest department on its own initia- tive has been experimenting with a cost-sharing model for producing timber in natural but degraded sal (Shorea robusta) forests. Villagers agree contrac- tually that in exchange for protecting growing timber, they will be employed in operations to cut all but the main shoot from degraded sal stands, receiv- ing concessional prices for the discarded stems. They are assured a fixed share (25Z) of the final harvest, distributed in cash individually to each villager, and have access to the forest for fallen wood, grass, and sal leaves (which can be a major source of income when sold in local markets). This model has been so successful that about 150,000 hectares of regenerating sal forest is being managed in this fashion. Several other state forest depart- ments, including Haryana, are planning to implement a similar model. Increased Rights Over Produce from Forest Lands Cost-sharing is matched by a general trend to increase substantially local rights over produce from forests established under social forestry. In all Indian states with these programs, government orders regarding the conces- sionary rights of local communities have been revised to provide increasing amounts of produce to surrounding areas--either through the panchayat on auc- tion or by allowing more collection of intermediate produce. In addition, there is experimentation with new technologies that are directed toward improving the environment rather than timber production, and which also pro- vide a range of forest products needed by the poor and marginal villagers. Emphasis is on silvipastoral plantation models that generate grass for stall- feeding of animals as well as trees. More use is being made of shrubs and hedges that stabilize the soil as well as provide a regular source of medium- quality, but continually accessible supply of household fuel (World Bank, 1988b). -146- Increased Tenure Rights in Taungya Models In response to a very low success rate with afforestation of produc- tion forest lands in Java (under the jurisdiction of the forest corporation), the government in cooperation with the Ford Foundation has been experimenting with increasing tenure rights of laborers working in afforestation schemes under a taungya system. Individuals are allocated plots for a longer time period than the original three years and are allowed to plant intercrops of grasses and fruit-bearing trees between rows of timber and pulp species for their own profit, in addition to seasonal crops planted before closure of canopy (Peluso, 1988; Ford Foundation, 1988). From the cultivator's perspec- tive, the model is somewhere between a stewardship contract and the West Bengal cost-sharing model. This revised scheme is emulating principles emerging for successful CPR management in South Asia. There is a sustained flow of benefits from the initial planting of seasonal crops, rows of fodder grass, and eventual har- vesting of fruit. On a pilot basis, the plots have been targeted to marginal villagers, but on a broad scale, considerable opposition from better-off vil- lagers is likely if they are, de facto, excluded from access to plots. Rights and responsibilities are clearly defined for the plotholder and foresters. In some areas, foresters have stopped enforcing restrictions on the gathering of minor forest produce in villages where the new scheme is in effect (Sri Palupi, 1988); this appears to be a conscious effort to change the relation- ship between village and forest department and has strengthened villager faith in the program. The scheme raises the issue of what the villagers' rights of access should be to other forest lands. Should the government consider the income-generating potential of other forest lands, except for parks and reserves, or continue its present policy of exclusion, except under the limi- ted taungya plots? Wells (1989) argues for rights of access in forest areas adjacent to parks and reserves, so that these areas can realistically serve as buffer zones. This is also a contention of the Dutch-assisted Kalikonto watershed project staff in East Java, which is tackling similar forest land- use issues (Jacques Beerns, personal communication). Leases to Cooperatives and Associations There are two models for extending leases to cooperatives and associ- ations in the case studie.s examined. One is in the Central Visayas Regional Project in the Philippines, in which groups of upland residents who had been illegally exploiting forest lands for timber in areas of extensive in-migra- tion have been given legal stewardship over these lands for forestation pur- poses. The residents of the area are organized into associations (FOSAs) under the project by the forestry extension staff and project staff. They establish nurseries for forestation and plant forest lands with timber and fuel species, over which they have exclusive rights of harvest within the terms of a 25-year, renewable lease. Initially, the associations are allowed to harvest the remaining timber on these lands in a controlled manner to gen- erate income in the early stages. Since most FOSA members have some land under agriculture as well, the project also provides tree crop seedlings to members to encourage more sustainable agricultural practices, along with slopy area land (SALT) technologiea promoted in less slopy agricultural sites. -147- Since FOSAs are quite new, there is not enough information to judge their via- bility. This model merits careful study, particularly of inputs and returns, as an alternative to private or village management of resources. The second model comes from the state of Haryana in India. A Ford Foundation-assisted program for the degraded watershed catchment near the capital city, Chandighar, developed small-scale water-harvesting structures for irrigation and created village resource management societies to protect the upper catchment and the flow of water into the harvesting structure. The lands in the upper catchment are state forest lands that were formerly given on lease to contractors for grass harvesting. In some villages in the new model, these societies have purchased lease rights from the forest department for the grasses. These are used for rope-making and as fodder for local dairy cattle. Given the proximity of these watersheds to the capital city, there is a good market for milk and dairy products, and villagers have a strong incentive to protect the catchment for grass production. Analysis has shown that in many sites, the water-harvesting benefits are quite small, because of limited potential, and that the benefits from grass are more promising. Irri- gation structures provide an initial, strong incentive for village participa- tion and mobilization, while it is the grass management that offers the most sustained returns to the village population as a whole (Stewart, 1989; Arnold and Stewart, 1989). Interestingly, even though the irrigation facilities created are limited and do not reach the majority of villagers, they serve as a strong incentive for protection of the catchment by all villagers, including those without irrigation facilities. This is the case even where the returns from grasses and resulting fodder production are actually greater than from the water. It would seem that small irrigation structures in hilly areas may provide a good reward for catchment protection in state watershed programs. Potential for Small-Scale Forest-Based Enterprises Recently, a growing interest has developed in the potential to increase the income-generating potential and thereby the sustainability of forest lands through support for small-scale, forest-based enterprises. In areas of natural forest, which traditionally supply a range of nontimber forest products, more attention to the income-generating potential of these forests for local people is recommended. Potential encompasses income from collection and sale of products and from better access to markets and greater participation in value-adding processing activities. In the villages partici- pating in the cost-sharing model in West Bengal for sal forest rehabilitation, women earn substantial income from such nontimber products. Not yet recogniz- ing this benefit, foresters have neither tried to increase nontimber produc- tivity nor assisted villagers with processing for higher returns. Attention to nontimber products is also being recommended for forest areas on Java in several projects. For the areas under taunaja forestation schemes, Kalikonto project staff are promoting higher-value products such as oilseeds that can be locally processed by cultivators and are more difflcult to steal than the fruits now grown on often-distant plots. Several social forestry projects in India have placed increasing attention on growing nontlm- ber products, such as medicinal plants, but again have not explored the best channels for marketing or processing them in order to generate mazimum income -148- for local people. An innovative NGO in Delhi, India has started an experimen- tal industrial estate which will support a small village from enterprises drawing on raw materials produced in a nearby plantation on government land. CHOOSING EFFECTIVE PROJECT MEASURES Measures that provide checks and balances in project design against the unintended consequences of introducing technological or institutional change include: (a) provision of appropriate support services, especially to those categories of farmers with inadequate resources for adoption; (b) choice of technological options that generate additional income to beneficiaries; (c) mechanisms for local institution-building; and (d) strategies for media- tion at the local level. Support Services. Farmers' groups are an excellent conduit for sup-. port services. The Central Visayas Regional Project in the Philippines and the Highland Agriculture Development Project in Thailand both attribute a large measure of their success to the strengthening of existing farmers' groups and providing extension services through them. In Thailand, such groups were a focus for offering alternatives to formal credit. Formation of farmers' groups for improved extension services has also proved important in relieving labor constraints to the adoption of soil and moisture conservation on arable lands in the Yogyakarta Rural Development Project in Indonesia. Where extension services are effective, a much wider range of farmers will be interested in the technologies offered. Income-generating opportunities from projects also address tenure- related constraints. In Java, farmers traditionally invest much less in rain- fed plots than in irrigated rice or home-and-mixed gardens, particularly when the distant rainfed plots are difficult to protect and time-consuming to reach. The Dutch-financed Kalikonto watershed project in eastern Java has increasingly put emphasis on identifying income-generating components. Research efforts have focused on cropping systems with high returns for home- and-mixed gardens on the premise that enough additional income from them would reorient marginal farmers towards heavier agricultural investment in their rainfed plots and make them less dependent on off-farm employment. Mediation as a Form of Extension Support. In "frontier" areas, gen- erating adequate information about tenure rights, mediating between cultiva- tors and other land claimants, and informing landowners about conservation measures are all crucial to adoption of new practices (Feder, 1988; World Bank, FFARM Study, 1989; Hoare, 1986). Programs carried out by the Bureau of Forest Development in the Philippines relied upon mediation between individ- uals claiming rights to the same lands (Seymour, 1985). Mediation has led to informal guarantees by de facto owners that cultivators will benefit from land improvements. This mediation is less threatening to the existing power brok- ers, yet seems to be simultaneously increasing benefits to cultivators, who can now adopt more sustainable practices. NGOs as Mediators. Some local NGOs have established group planta- tions on wastelands allocated to them. They are successful largely because of the social pressure that NGOs can place on local elites to protect the groups from power brokers competing for access to these newly productive lands. Yet -149- NGOs have not been able to help groups gain formal lease-certificates for these wastelands, since the legal procedures are formidable and not well understood by government personnel at different levels of the administration. There is potential to expand NGO programs under Bank-assisted projects through support to training of NGO personnel and encouraging policies that streamline government procedures and recognize customary rights. CONCLUSIONS Land tenure issues in watershed development are complex. Certain kinds of tenure changes can have a positive impact on adoption, and in some cases, titling or land consolidation on private land in areas with socially recognized tenure rights is effective. The most scope can be found in increasing tenure security in the "frontier" areas and in recognizing and strengthening systems of corporate and common property resource management, where they exist. Some positive measures that can be included in projects to support land tenure changes or to broaden the range of adoption within existing tenure systems include: providing increased extension support, often through tar- geted groups; providing sources of credit; focusing on technologies with quicker and high returns; strengthening local institutions; and providing mediation or legal aid to participants. It becomes clear from this review that the World Bank needs to con- tinue to study the relationship between land tenure and the adoption of soil and water conservation technologies, so that clearer directives can be given to task managers as to what strategies are most productive, sustainable, and equitable in the different Asian settings,. More testing is needed of these measures through controlled study of ongoing project experience. Monitoring and evaluation systems are already overburdened by information requirements and are in any event unlikely to yield the culturally sensitive data needed. With Bank guidance, studies could'be carried out by host country institutions, providing value both to the borrower and the Bank task manager. - 151 - ANNEX 6.1 Project Cases Examined in This Review 1. Yogyakarta Rural Development Project, Indonesia, (WB/Government) which included a titling component for private, dryland plots to encourage farmers to adopt improved technologies. 2. Upland Agriculture and Conservation Project, Indonesia (USAID/WB/Govern- ment), which has selected demonstration plots on farms cultivated by the owners for the dissemination of bench terrace technology and agrofores- try. 3. Social Forestry Project of the State Forest Corporation, Indonesia (Ford FoundationfGovernnent), which has provided extended leases for taungya/ tumpang sari cultivation and allowed farmers to plant perennial crops in between timber trees. 4. National Social Forestry Project, India (USAIDIWB/Government), which is the most recent of the Bank's social forestry programs funded for indi- vidual states in India. 5. West Bengal Social Forestry Program, India (WBIState Government), which includes a Bank-assisted project with a component termed ogroup farm forestry" encouraging farmers with marginal holdings to undertake block tree planting with shared labor and shared protection systems. 6. West Bengal forest department experiment in regeneration of natural forest through allocation of forest land to local villages (WB/forest department program), which has provided villages with intermediate yields and a substantial share of the final timber harvest in rehabili- tated sal forests. 7. Village Resource Management Societies, Haryana, India (originally Ford Foundation-assisted), which are a local-level society created for water- shed management and maintenance of small, water-harvesting structures for irrigation. 8. Forest Panchayats, Himachal Pradesh and Uttar Pradesh, India (pre- independence institutions), established in the Himalayas and Siwaliks for community management and protection of forests on behalf of the forest department. 9. Central Visayas Regional Project (CVRP), Philippines (WB/Government), which is a watershed and coastal rehabilitation project extending the sloping area land technologies (SALT) for land improvement to upland farmers and providing stewardship leases to farmers cultivating state forest lands. 9b. Forestry Associations (FOSAs), CVRP, Philippines (WB/Government), which is a component of the above project for organizing upland migrants to exploit lands over 18Z slope into associations (FOSAs) for the estab- lishment and extraction of forest plantations on degraded forest lands and providing them with stewardship leases to these lands. - 152 - ANNEX 6.1 10. Highland Agriculture Development Project, Thailand (Australia/WB/Gov- ernment), which is a project in north Thailand for rehabilitation of lands cultivated by tribals under an extensive (slash-and-burn) agricul- tural system. 11. Gansu Provincial Project, China, (WB/Government), which has allocated pasture lands to individuals for development on long-term leases (the period of the lease has steadily increased since the inception of this program). 12. Red Soils Provincial Project, China (WB/Government), which has allocated lands to individuals for development on long-term leases, and which includes a parastatal institution providing capital and marketing sup- port for farmers developing these lands. 13. Kalikonto Project, Java, Indonesia (Dutch-assisted), which has experi- mented with arable and nonarable land watershed development, concentrat- ing recently on increasing economic returns from home gardens to reduce farmer dependence on off-farm employment strategies. - 153 - REFERENCES Adams, Martin and David Seddon, and Overseas Development Institute. 1983. Land Tenure in Irrigation Planning: Two Examples, Irrigation Manage- ment Network Paper 76. Overseas Development Institute: London. Anderson, Jock R. and Dodo Thampapillai. 1988. "Soil Conservation in Developing Countries: Project and Policy Intervention," draft paper for Agricultural and Rural Development Department, World Bank. August. Arnold, J.E.M. and William Stewart. 1989. "Common Property Resource Man- agement in India: A Desk Review,n Report for the Asia Environment Division and the India Agriculture Division, World Bank. _ and J.G. Campbell. 1986. "Institutional Dynamics: The Evolu- tion and Dissolution of Common Property Resource Management," A Panel on Common Property Resource Management, Board on Science and Technology for International Development, National Research Council, Proceedings of the Conference on Common Property Resource Management, pp. 425-454. National Academy Press: Washington, D.C. Bandhopadhyay, J. and Vandana Shira. 1985. "Eucalyptus in Rainfed Farm Forestry: A Prescription for Disaster," Economic and Political Weekly 20(40). Batie, Sandra and Alyson G. Sappington. 1986. "Cross-Compliance as a Soil Conservation Strategy," American Journal of Agricultural Economics, 68(4):880-885. November. Blaikie, Piers. 1987. The Political Economy of Soil Erosion in Developing Countries. Longman Press: London. Blaikie, Piers, J.C. Harriss and A.N. Pain. 1985. "Public Policy and The Utilization of Common Property Resources in Tamil Nadu, India," Report presented to the Overseas Development Administration. April. Brara, Rita. 1987. "Shifting Sands: A Study of Customary Rights in Graz- ing." Mimeo. Institute of Development Studies: Jaipur, India. Campbell, J.G., Rajendra P. Shrestha and Fred Euphrat. 1987. "Socio- Economic Factors in Traditional Forest Use and Management." Prelimi- nary Results from a Study of Community Forest Management in Nepal. Banko Jankari (Special Issue, Community Forestry Management Workshop), 1(4). Cernea, Michael. 1985. Putting People First: Sociological Variables in Rural Development. Oxford University Press (for World Bank): Oxford, United Kingdom. Cohen, Paul T. 1983. "Problems of Tenancy and Landlessness in Northern Thailand." Developing Economics, 21:244-266. - 154 - Cullins, Jane and Michael Painter. 1986. 'Settlement and Definition in Central America: A Discussion of Development Issues,' Working Paper in IDA Series; 31. Worchester, Massachusetts: Clark University Interna- tional Development Program. Damodaran, A. 1987. 'Structural Elements of the Fodder Crisis: A Village Study in Karnataka." Indian Economic and Political Weekly Review of Agriculture, 13(22):A16-A22. Dani, Anis and J. Gabriel Campbell. 1986. 'Sustaining Upland Resources: People's Participation in Watershed Management," ICIMOD Occasional Paper No. 3. International Centre for Integrated Mountain Development and FAO: Kathmandu. Edgerton, J. 1987. 'Review of Land Registration/Titling and Farmer Parti- cipation,' Draft, Yogyakarta Rural Development Project Consultants' Report to Indonesia Agriculture Department (World Bank Internal Docu- ment). Feder, Gershon, Tongroj Onchan, Yongyuth Clamawong, and Chira Hongladaron. 1988. Land Policies and Farm Productivity in Thailand (World Bank Research Publication). Johns Hopkins University Press: Baltimore. Gibbs, Christopher J.N. 1983. 'Institutional and Organizational Concerns in Upper Watershed Management," in Watershed Resources Management: An Integrated Framework with Studies from Asia and the Pacific. pp. 91-102. Waterview Press: Boulder. Guhathakurta, Prabir. 1989. 'The Arabari Experience in West Bengal,' paper presented to the National Wasteland Development Board: Workshop on Involving Women In Social Forestry. New Delhi, India. February. Hart, Gillian. 1986. Power, Labor and Livelihood: Processes of Change in Rural Java. University of California Press: Berkeley. Herring, Ronald J. 1983. Land to the Tiller: The Political Economy of Agrarian Reform in South Asia. Yale University Press: New Haven. Hoare, Peter and Somphob Larchrojna. 1986. "Change in Traditional Manage- ment of Forests: A Study of Thailand's Karen Hill People,' Community Forestry: Lessons from Case Studies in Asia and the Pacific Region; Rao, Y.S., M. Hoskins, N.T. Vergara, and C. Castro, eds. Regional Office for Asia and the Pacific of the Food and Agriculture Organiza- tion of the United Nations, Thailand and the Environmental and Policy Institute, East-West Center: Honolulu, Hawaii. Jodha, N.S. 1989. 'Management of Common Property Resources in Selected Areas of India," in Dani, Anis, and J.G. Campbell, eds., Local Institu- tions and Resource Management. International Center for Investigation on Mountain Development (ICIMOD): Kathmandu, Nepal. - 155 - Kalikonto Project, Konto River Project, Phase III. 1988. Indonesia and Kingdom of *he Netherlands Proiect Working Paver Series. Directorate General of Reforestation and Land Rehabilitation (DGRRL), Ministry of Forestry. KEPAS [Kelompok Penelitian Agro-Ekosistem). 1985. The Critical Uplands of Eastern Java: An Agroecosystems Analysis. Brawijaya University Research Center, East Java Agroecosystem Working Group. January. Rhasnabis, Ratan and Jyotiprakash Chakravarty. 1982. 'Tenancy, Credit and Agrarian Backwardness: Results of a Field Survey. Indian Economic and Political Weekly, pp. A21-A32. March. Kulkarni, Sharad. 1988. 'Encroachment on Forests: Government versus People.' Indian Economic and Political Weekly, pp. 55-59. Kunstadter, Peter and E.C. Chapman, Sanga Sabhasri and Mahawitthayalia Chiang Mai, eds. 1978. Farmers in the Forest: Economic Development and Marginal Agriculture in North Thailand. East-West Center, Univer- sity of Hawaii: Honolulu. Ladejinsky, Wolf. 1988. Land Reform as Unfinished Business, Selected papers of W. Ladejinsky edited by Louis Walinsky. Oxford University Press (for World Bank): New York. Librero, Aida R. 1984. 'Socio-Economic Considerations in a Soil Erosion Management Program: Case Studies of Two Provinces in the Philippines.' Soil Erosion Management, CGIAR Proceedings Series, 6. Australian Centre for International Agricultural Research: Canberra, Australia. Ljungman, Lennart, Doug McGuire, and Augusta Molnar. 1987. Social Fores- try in India: World Bank Experience. South Asian Agriculture Depart- ment Draft Review, World Bank: Washington, D.C. Lynch, Owen and Kirk Talbot. 1988. "Legal Responses to the Philippines Deforestation Crisis." Journal of International Law and Politics, Spring. MacAndrews, Colin and Lincoln Institute of Land Policy. 1986. Land Policy in Modern Indonesia: A Study of Land Issues in the New Order Period. Delgerchlager, Gunn and Hani: Boston, Massachusetts. Mackie, Cynthia. 1986. 'Land Tenure and Conservation Practices in the Upper Watersheds of Java." Report submitted to Asia Environment Divi- sion as part of Regional Sector Review on Watershed Management and Development, World Bank: Washington, D.C. June. Magrath, William. 1989. The Challenge of the Commons: The Allocation of Nonexclusive Resources. Environmental Department Working Paper No. 14, World Bank: Washington, D.C. and P.L. Arens. 1989. "The Costs of Soil Erosion on Java--A Natural Resource Accounting Approach,' Environment Department Working Paper No. 18, World Bank: Washington, D.C. - 156 - McCauley, David. 1988. 'Overcoming Institutional Organizational Con- straints to Watershed Management of the Densely Populated Island of Java.' Paper presented to 5th International Soil Conservation Confer- ence: Land Conservation for Future Generations: Bangkok, Thailand. January. _ 1989. Project Completion Report Annex for Citanduy Project: Annex 5, Policy Analysis. USAID Document. McKinnon, John. 1977. 'Who's Afraid of the Big Bad Wolf." Faculty of Agriculture, CMU and NADC's 77th Seminar in Agriculture in Northern Thailand Series. April. McLain, Rebecca J., and Douglas M. Steenberger. 1988. Land Tenure and Land Use in Southern Haiti: Case Studies of the Les Anglais and Grande Ravine du Sud Watersheds. Land Tenure Research Paper 95: Madison, Wisconsin. April. Molnar, Augusta. 1986. 'A Review of the Social Forestry Experience in India and Nepal Regarding Community Woodlots, Improved Cookstoves and Women.' Report for the South Asia Agriculture Division, World Bank: Washington, D.C. National Academy of Sciences. 1986. Common Property Resource Management. National Academy Press: Washington, D.C. Naronha, Raymond. 1985. 'Sociological Aspects of Forestry Project Design.' Draft report to The World Bank: Washington, D.C. Pant, Chandrasekhar and ICRISAT. 1981. 'Tenancy in Semi-Arid Tropical and Input Use.' Economic Program, Progress Report 20, International Crops Research Institute for the Semi-Arid Tropicss Hyderabad, India. Peluso, Nancy. 1988a. 'Rich Forests, Poor People, and Development: For- est Access Control and Resistance in Java.' PhD dissertation, Cornell University. 1988b. 'Troubles with Tenure: Farmer Agroforestry on State Lands in Indonesia." Paper prepared for 1989 annual meetings of the American Association for the Advancement of Science: San Francisco. January. -__ _ .1987. 'Social Forestry in Java: Can it Work?' Consultants' Report to the Ford Foundation: Jakarta and New York. Rahman, Mushtaque, and M. Aminui Islam and Dirpica Baglhi. 1981. Agrarian Egalitarianism: Land Tenures and Land Reform in South Asia. Iowa State University Research Foundation, Kentall/Hunt: Dubuque, Iowa. Ramsey, Ansil. 1985. 'Population Pressure, Mechanization, and Landless- ness in Central Thailand." Journal of Developing Areas, 19:351-368. April. - 157 - Rosenberg, Jean G. and David Rosenberg. 1980. Landless Peasants and Rural Poverty in Indonesia and the Philippines. Cornell University Rural Development Committee (Special Series on Landlessness and Near Land- lessness, No. 3): Ithaca, New York. Royal Thai Government, Public Welfare Department. 1988. "Highland Agri- cultural and Social Development Project Completion Report." World Bank, Australian International Development Assistance Bureau, Royal Thai Government: Thailand. Sajise, Percy and Terry Rambo. 1985. Agroecosystem Research in Rural Resources Management and Development. Selected papers presented at the Second SUAN-EAI Regional Symposium of Agrosystem Research: Philippines. Seddon, David. 1983. "Results of a Failure to Make an Early Investigation into Land Tenure in Morocco." Land Tenure in Irrigation Planning: Two Examples. Irrigation Management Network, Overseas Development Insti- tute. Seymour, Frances. 1985. "Ten Lessons Learned from Agroforestry Projects in the Philippines." Paper for Ford Foundation Office: Manila, Philippines. Shah, Tushaar. 1987. "Gains from Social Forestry: Lessons from West Bengal." Social Forestry Network Paper 5E. Agriculture Administration Unit, Overseas Development Institute. Singh, Chatrapatti. 1989. Presentation at a Methodology Workshop on Com- mon Property Resource Management, sponsored by USAID, World Bank and Development Alternatives (India). New Delhi. February. Singh, Inderjit. 1988. "Tenancy in South Asia." World Bank Discussion Paper No. 32, Washington, D.C. Soentoro, William Collier and Sri Hartoyo. 1981. Land Tenure and Labor Markets in East Java, Indonesia. Agro-Economic Survey: Bogor, Indonesia, August. Sri Palupi. 1988. "Pengaruh Program Perhutanan Sosial [Social Forestry] Perhadap Perranan Wanitadalam Ekonomi Rumahtanggu: Studie Kasus di Desa Pitu, Kecamatan Pitu, Kapupaten Naroi-Jawa Timur." Bogor Agricul- tural University, Socioeconomic Department, Agricultural Faculty: Bogor, Indonesia. Stewart, William. 1989. "Common Property Resource Management: Status and Role in India." Draft paper, World Bank. April. Stoler, Ann. 1978. 'Garden Use and Household Economy in Rural Java." Bulletin of Indonesian Economic Studies, 14(2):85-101. Australian National University: Canberra, Australia. - 158 - Tampobolan, S.M.H. 1988. "An Economic Analysis of Soil Conservation Prac- tices in the Citanduy River Basin, Java, Indonesia.* PhD dissertation (draft), University of the Philippines at Los Banos. November. Tingfu, Gus and Liu Shen. 1988. "Reliance on the Broad Masses of Farmers to Control Soil Erosion in the Upper Reaches of the River Valley of China." Paper presented at 9th Annual Symposium on FSR/E: Arkansas. October. USAID and the World Bank. 1988. 'National Social Forestry Project Mid- Term Review." New Delhi, India. . 1984. 'Upland Agriculture and Conservation Project Paper., Republic of Indonesia. Wade, Robert. 1988. Village Republics: Economic Conditions for Collec- tion Action in South India. Cambridge University Press: Cambridge. Wells, Michael. 1989. 'Can Indonesia's Biological Diversity be Protected by Linking Economic Development with National Park Management? Three Case Studies from the Outer Islands.' Report for The World Bank, Envi- ronment Division. March. Wiradi, Gunawan, Chris Manning and Sri Hartoyo. 1989. 'Employment, Rural Labor Market and Land Tenure: A Preliminary Report of a Recensus in Nine Villages in Java.' Draft paper for Agroeconomic Survey Founda- tion: Bogor, Indonesia. April. World Bank. 1989. Indonesia: Forest; Land and Waters Issues in Sustain- able Development, Report No. 7822-IND. June. _ 1989. Philippines Forestry Fisheries, and Agricultural Resource Management Study (FFARM). Internal Country Department II. _ 1988. Indonesia: Yogyakarta Rural Development PCR, Phase 1. _ 1988a. India: Review of Rainfed Agriculture and Watershed Development, vols. I and II, Report No. 7138-IN. February. . 1988b. India: Uttar Pradesh and Gujarat Social Forestry Project, PCR, 1988. . 1987. Meknes Agricultural Sector, PCR. Operations Evaluation Department. . 1985. India: National Social Forestry Project, SAR, 1985. Young, Michael. 1987. 'Land Tenures Playthings of Governments or An Effective Instrument,' in Anthony Chisholm and Robert Dumsday, eds., Land Degradation Problems and Policies. pp. 175-186. Cambridge Uni- versity Press: Cambridge. - 159 - 7. A FRAMEWORK FOR PLANNTNG. MONITORTNG. AND EVALUATTNG WATERSHED CONSERVATION PROJECTS Glenn S. Morgan and Ronald C. Ng This chapter reviews the World Bank's operational experi- ence with planning, monitoring and evaluation (PME) of watershed development projects; develops a rationale for promoting and developing local competency in PME as part of watershed development projects; and describes the essential components of a PME program that should be established prior to, or as part of, project implementation. To be successful, PME components should be developed within a clearly articulated framework for organizing key functional tasks. The approach described herein recommends that these tasks be organizationally structured under three closely related administrative cells. DEFINITION OF TERMS The terms planning, monitoring and evaluation are used to refer to three mutually supportive, though discrete, project activities. Planning, used traditionally, refers to a systematic process of establishing watershed development objectives (spatial and nonspatial), formulating alternative development actions, and selecting technically sound and socially acceptable courses of action. Monitoring denotes continuous assessment of project activ- ities in the context of implementation schedules, the use and allocation of project inputs by targeted beneficiaries, and the measurement of progress against stated goals. Evaluation refers to periodic assessment of the rele- vance, performance, efficiency and impact of the project in the context of its stated objectives. WORLD BANK EXPERIENCE WITH WATERSHED PROJECTS Careful planning, monitoring and evaluation of project activities have been cited as essential components of virtually every World Bank-assisted watershed development project. As reflected in project appraisal reports, the components' overall goals within watershed projects appear consistent with recommendations available in the technical literature. Financial allocations for such components have typically ranged from 12 to 62 of total project costs, with the average expenditure being about 2Z. Available project documentation reveals that the most often cited goals for planning, monitoring and evaluation procedures are to: (a) establish baseline information; (b) track and document physical progress and achievements; (C) account for financial expenditures; - 160 - (d) develop an information system adequate for comparative analysis of project components; (e) provide a mechanism for evaluating the need for project adjustments; and (f) execute special-purpose studies to evaluate project performance. Still, World Bank experience with implementation and execution of such tasks within watershed development projects seems to have varied consid- erably. Benefits of PME, though thought to be large in relation to expendi- tures, have proven difficult to quantify. The effort required to design and implement PME activities successfully has probably been underestimated in most projects, given that knowledge from many fields such as accounting, sample survey design, and land-use surveying must be coordinated. Supervision reports indicate that few PME components have yielded satisfactory results; most report delays in implementation and relatively poor output. The constraints faced by designers of PME components are certainly not unique to watershed development projects. At least ten issues have been consistently identified as major problems in implementation encountered by project supervision teams: (a) shortages of trained and competent staff; (b) a weak, in some cases nonexistent, information base; (c) inadequate interdepartmental cooperation and coordination requiring information exchange; (d) overall lack of institutional commitment to PME tasks; (e) poor understanding of external reporting requirements; (f) poorly communicated objectives regarding the users and uses of moni- toring and evaluation data; (g) poorly supervised and enforced standards of report preparation; (h) redundancies in reporting requirements (e.g. quarterly reviews and semiannual, making the same points over and over); (i) lack of clarity about responsibilities and obligations of different units; (j) inability of field supervision teams independently to verify reported progress. Organizationally, at the project level, the responsibility for devel- oping the details of PME components has usually been left to the individual project coordination or project implementation units. As a consequence, there appears to be little consistency across watershed projects about: how much and what type of baseline information is collected; the expenditure incurred to develop and maintain an ongoing management information system; the level - 161 - and detail of supporting information required; the frequency of surveys and other data collection exercises; or the methodology employed or the technolo- gies utilized to collect and manage information. This lack of consistency in approach makes the task of comparing project experiences somewhat more diffi- cult. As with functional responsibilities, there are also no clearly defined organizational and institutional blueprints relevant to all watershed development programs. Some watershed development projects have recommended the establishment of new project-level institutions, varying by level of decision-making authority and management responsibility. For example, the Indonesia Forest Institutions and Conservation Project, for which implementa- tion commenced in 1988, has recommended establishing a project planning and implementation unit (PPIU) for each project component (forestry, soil conser- vation, pasture and grazing, etc.). Under this model, the individual unit has responsibility for monitoring its disciplinary specialty under the general guidance of a project liaison office. Another approach has been to establish formal, centralized PME units with more autonomy from the project implementing staff. Often, the tasks of PME are contracted out to independent research institutes, universities, or private consulting operations and are supervised by small monitoring units. Yet another institutional option, tested in the Pilot Project for Watershed Development in Rainfed Areas of India (implementa- tion commenced in 1984), is to establish monitoring and evaluation responsi- bilities with a state-level watershed development cell. Each approach has enjoyed some success and endured some disappointments in implementation. What appears to be consistent in the review of previous projects is the recommenda- tion that PME tasks and their execution remain functionally separate from implementing agencies in order to ensure objective evaluation of project per- formance. Despite some setbacks, many program planners remain committed to the idea that a major objective of watershed development projects should be to establish, within appropriate lead agencies, capacities to carry out planning, monitoring and evaluation tasks. In past projects, such tasks have included, inter alia, the execution of diagnostic surveys, preparation of land-use and land-capability maps, development and maintenance of a watershed information system describing the baseline conditions of the watershed to be treated, the preparation of treatment plans, the execution of special research and investi- gative studies, training of staff in line agencies in the application of cost- effective planning tools and procedures, and training of field staff in appro- priate techniques for eliciting public input in the watershed development process. RATIONALE FOR PLANNING, MONITORING AND EVALUATION CELLS The approach that will be described refers to a planning, monitoring and evaluation (PME) unit having three separate but closely related cells, one each for planning, monitoring and evaluation. While the conventional term, monitoring, is retained for the second of these cells in the general model, it is more functionally a management cell. Watershed development projects are predicated on the principle that watersheds are viable and meaningful physical units for development planning and program implementation. Most importantly, watersheds are viewed as convenient and logical units for evaluating the bio- logical and physical linkages between upstream and downstream activities. - 162 - These activities are linked naturally through the hydrology of the watershed, because physical changes in upper catchments can result in a chain reaction of physical impacts downstream. This can only be done through a careful land-use planning process. Watershed land-use planning is fundamentally concerned with delineat- ing where, by whom, and in what sequence conservation and rehabilitation actions will be undertaken. The specific spatial arrangement and timing of activities are of critical importance to the success of the investment package and therefore should be approached systematically. In addition, the approach' must be executed in a way which simultaneously considers the place of an indi- vidual watershed in its regional environmental and economic context (strategic planning) and the important role of public participation in the preparation of site-specific treatment plans (micro-planning). It is generally believed that, in the absence of a systematic approach to watershed planning, conservation and development authorities will be constrained in their ability to describe accurately the current land-use situation, monitor the dynamics of land-use change, evaluate the state of land degradation, or predict the likely impact of proposed development projects. The difficulties created by this situation are cross-sectoral with implica- tions for forestry, agriculture, water management and pasture development. In the context of watershed development projects, a planning cell must be capable of addressing several interrelated tasks. Typically these tasks include: (a) establishment of watershed development priorities on a regional basis; (b) coordination of implementation responsibilities among line agencies; (c) elicitation of people's participation in the establishment of invest- ment priorities; (d) development of technically sound, sustainable land management alter- natives through micro-level village planning; and (e) effective and timely communication of project activities to the local inhabitants affected by the project. The realization of these planning goals depends on several prerequi- sites, such as institutional ability to evaluate development proposals in a regional environmental and economic context; existence of a process which allows communications among line departments on a regular basis; existence of trained and committed staff to work with people at the grass-roots level; existence of local community organizations to articulate community needs and desires; and technical ability to articulate and implement sound and sustain- able technical solutions to land management problems. In the generic model, a monitoring cell would have responsibility for translating information and regional priorities defined by the planning cell into location-specific action plans, that is, a management function. Clearly, there is explicit linkage and perhaps a slight overlap in the responsibilities of the two units. Monitoring of watershed development projects will involve both socioeconomic household surveys (longitudinal, cross-sectional, informal - 163 - interviews, etc.), as well as physical monitoring (erosion rates, revegetation rates, etc.). Ideally, one would wish to be able to compare a series of sur- vey results with a reliable baseline of information. Unfortunately, this is not possible if baseline data are not available. While the establishment of baseline data is certainly a welcome long-term goal, it is not necessarily a prerequisite of monitoring programs. Several key issues need to be addressed during the design of monitor- ing components: namely, the choice and articulation of key indicators to be measured and collected; the procedures and methodologies to be employed in collecting data; the temporal frequency of collecting information; the size of the sample; and how data will be utilized in future planning exercises. Each project will vary according to staff skills, available budget, institutional capacity; nevertheless, the work of a monitoring cell should parallel the strategic planning efforts of the planning cell. Thus, it is critical that an adequate watershed management information system ('WHIS) be established. It should be designed with long-term objectives in mind and at a minimum should address the functional tasks described below in the following section. Evaluation tasks typically involve independent assessment of a wide range of considerations, including assessment of the replicability and sus- tainability of project objectives; the quality of project preparation and appraisal; the performance and efficiency of watershed investments in the context of original goals; the sequencing of project components; and the extent to which investments in different components were mutually reinforcing. In practice, evaluation has focused primarily on measuring physical progress and less on measuring behavioral changes of participating communities. Within watershed projects, evaluation is generally regarded, somewhat like monitoring, as an ongoing management mechanism for identifying design and implementation problems and making appropriate adjustments to the project. In the longer term, evaluation reports are useful in the design of follow-up projects. Ideally, evaluations should be undertaken by independent organiza- tions and budgetary allocations should be separated from the planning and monitoring budgets. As with other elements of a PME unit, blueprints for success are rarely available. FUNCTIONAL TASKS OF WATERSHED PLANNING, MONITORING AND EVALUATION CELLS Regardless of the organizational framework selected for PME compo- nents, the staff responsible for implementing planning, monitoring and evalua- tion components of watershed projects should address the following functional areass Technical Records Management An important function of the PME Unit is to establish a systematic approach to recording and retrieving data on field trials and experiments, technical studies, species performance under different agroclimatic condi- tions, planting methods, costs, returns and effectiveness. These data have been particularly scarce in watershed projects and to paraphrase one project supervision report, "when it comes to hard facts regarding the technical or - 164 - economic merits of recommended practices in relation to controlling erosion and sedimentation, to their impacts on agricultural productivity and farmers' incomes, to the rate of adoption under different conditions and to several related factors, no systematic, quantitative assessments are generally availa- ble." In practical terms, establishing a database of technical records involves the use of desktop computer workstations and laptops, together with appropriate database management software. In many jurisdictions, the data are available in numerous reports and documents, but need to be collited and recorded in a format more accessible to planners. In practice, the technical records database would be compiled both from existing data, as well as new field data, as it is collected and reported from experimental stations, demon- stration plots, and operational field trials. Geography (GIS) and Cartography A second objective of a planning cell should be to improve the man- agement of spatially referenced information such as maps, air photos, and satellite imagery. This function could be effectively supported through the development of a geographical information system (GIS). A GIS, in the context of watershed rehabilitation and stabilization projects, is a computer-based workstation which permits the integration of baseline data (physical and social) deemed relevant to watershed development planning and management. Typical GIS systems link computer mapping software and tabular database man- agement software into one comprehensive system. A GIS database is organized on a modular basis, each module corres- ponding to important planning parameters. The strength of the GIS approach is that data handling and analysis can be carried out on the spatial and nonspa- tial elements of the database simultaneously. These capabilities can be used either by themselves or in conjunction with other simulation or statistical modeling techniques. The main modules of a GIS database usually consist of some variation of the following: (a) base module (administrative boundaries, transportation, communica- tions facilities, settlements); (b) terrain/soil module (land unit descriptions, soil types, geomorphol- ogy, erosion characteristics); (c) land cover/land-use module (vegetative cover, major land-use types); (d) hydrology module (rainfall, climate, river discharge data); (e) socioeconomic module (farming systems, demographics). If organized and implemented in a rationale manner, a GIS can be a valuable tool which facilitates the storage, retrieval and production of maps, statistical reports and, in some cases, the production and processing of sat- ellite imagery. It is also thought to be an appropriate mechanism for stor- ing, analyzing and disseminating field data from monitoring and evaluation field teams. GIS analytical techniques can be used on information derived from intensive ground studies or can be based on data derived from the use of remotely sensed imagery such as aerial photographs or space-craft pictures. - 165 - Village-Level Information Management In many situations, reasonably reliable social and economic data exist for individual villages within a watershed, but often require systematic storage and retrieval to make the information more readily available to con- cerned agencies. To formulate site-development plans, additional information on the social and economic conditions prevailing in the village needs to be collected. Village-level data are usually collected by revenue officers and updated on a cyclical basis. The standard approach is to collect and record village-level and other relevant socioeconomic information (fuel, fodder demand and supply, population, etc.) using previously designed data collection forms. The infor- mation on the forms could be efficiently managed on a small personal computer with appropriate spreadsheet or database management software. Watershed Strateaic Planning The extent to which dryland degradation in upland watersheds can be reversed will depend on a much improved understanding of many factors: among them, the status and rate of vegetative denudation, the severity of local soil erosion processes, the intensity of human and livestock pressures, the ability to introduce sustainable measures to control degradation, and the ability to communicate unambiguous benefits to practitioners of these.. As a basis for priority selection of large-scale investments in the stabilization, conservation, and reclamation of degraded lands, planners require knowledge about the status of lands across an entire region. In most situations, evidence of widespread degradation is anecdotal or based on small- scale, special-purpose, local surveys. For most jurisdictions, almost no maps of regional degradation exist and rarely has systematic assessment of degrada- tion been undertaken. An important element in the planning, monitoring and evaluation therefore is to assist in the establishment of regional baseline databases. Strategic planning is used to refer to the process by which priori- ties for watershed development are articulated and recorded. This usually requires close interaction and coordination among line agencies and the resource users. The process of setting regional priorities for watershed programs should be based on careful consideration of: (a) physical evidence of degradation (extent of gully and sheet erosion, erosion rates, deforestation, water quality); (b) social and economic indicators of degradation (fodder and fuel defi- ciencies, decreasing farm yields); and (c) likelihood of introducing sustainable technical packages (slope and soil characteristics, current land-use practices, local willingness). On the surface, land-use planning appears to be a logical and neces- sary, if not essential, part of any watershed development activity. Numerous planning models have been proposed and tested in the field with varying levels - 166 - of success. Still, many practitioners have questioned the overall utility of land-use planning exercises. Put simply, the development of land-use plans has frequently been described as burdensome, time-consuming and expensive. In situations where the planning goal has been to develop rigid village or regional master plans, the results often do not live up to expectations. In practice, many examples could be cited where detailed land-use plans have been drawn up based on elaborate soil surveys and land capability studies, only to be subsequently shelved and rarely consulted. In most cases, the disappointment with land-use planning arises out of a fundamental confusion between the land-use plan as an end product and land-use planning as a mechanism for articulating local needs, resolving con- flict, ensuring rights, and enforcing obligations of land users. The chal- lenge to watershed development planners is to avoid the syndrome of focusing on elaborate programs aimed at developing the master plans and shift attention towards an iterative process of matching local needs to local constraints. Rather than being abandoned, land-use planning for watershed development should be reoriented towards village-level consultation, program flexibility, selectivity and institutional coordination of predetermined responsibilities. Formulation of Site-Development Plans and Annual Work Program Site development plans should flow naturally from the strategic plan- ning performed in the planning cell. The tasks of such a unit include the management of public participation in the planning process and the development of site-specific action plans. The annual work program involves the schedul- ing of tasks to be performed, delegating responsibility to appropriate agen- cies, budgeting, and coordination of field activities. The steps involved in preparing site-development plans seem well developed and understood. In most projects, site plans are formulated based on some variation of the following process: (a) conduct diagnostic survey and local needs assessment; (b) evaluate local resource potential and constraints; (c) articulate local technical, financial and logistical prescriptions (the plan); (d) secure local endorsement, commitment and support for proposed action plans; (e) implement program; (f) conduct periodic reviews to assess progress and revise prescriptions as necessary. Many different types of reporting formats have been developed which may be adopted for each step of the interactive planning process. Whatever the format chosen, the village-level diagnostic survey should provide detailed descriptions of the specific location of areas requiring remedial treatment, the current status of these lands with respect to land-use practice, land - 167 - capability, land ownership/tenure, the land users, and a description of treat- ment options to be considered. The identification of treatments and options is largely a technical exercise based on diagnostic surveys, but the final decisions on the type, scale and timing of measures to be installed must be determined by the needs and preferences of local land-users. The operational challenge for watershed development practitioners is twofold: Firstly, how to efficiently obtain information and opinions from the rural residents. Secondly, how to convert and integrate that information into programs which are locally relevant, technically sound and which also serve broader, regional resource management goals. Unfortunately, little effort has been made toward incorporating or soliciting the active participation of the watershed land-users in the planning of site-development actions. The need for public participation is especially important in water- shed development schemes when the project treatments involve: (a) private or communal lands; (b) temporary land closures; (c) changes in land management practices (e.g., grazing patterns, cut and carry, rotational management, etc.); (d) investment of time and/or financial resources by local inhabitants (e.g., planting vetiver grass hedges or private construction of water-harvesting structures). Several approaches to soliciting popular participation as well as monitoring beneficiary impact of watershed investments have been commonly used in the field. These approaches include: (a) one-on-one interaction with local people (personal interviews, ques- tionnaires, household surveys, individual observations); (b) communication and interaction with community leaders (interviews with village leaders, traditional leaders, local entrepreneurs, progres- sive farmers, informants); (c) dissemination of information and participation through community meetings (e.g., public forums, hearings, presentations); and (d) interaction through community organizations (nongovernmental organi- zations (NGOs), religious organizations, local councils, regional organizations]. Monitoring Progress Monitoring progress involves regularly scheduled site inspections to assess works completed and should include statements of expenditure incurred and problems and delays encountered. Monitoring progress is an essential feedback mechanism as these findings define how annual work plans should be - 168 _ modified or corrected. Monitoring activities use several different tech- niques. The most common approach is to have site inspectors record required information on pro-forma recording documents. These documents would be fed back into the WHIS as part of the permanent database record. In practice, independent field verification of progress reports has proven to be quite difficult in watershed projects, particularly when the watershed areas are remote or access is difficult. It has proven extremely difficult, for example, for World Bank supervision teams to verify the numbers of checkdams and gully plugs constructed or to make reliable estimates of total reforested areas treated by the project. A complementary approach is to utilize remote- sensing technology, such as high-resolution satellite pictures to assess phys- ical progress. For example, site inspectors could clearly see the progress made, especially regarding forestry plantations in upland watershed areas. While these images could possibly be used quite effectively, for example, to evaluate the total impact of larger projects, they are probably not useful for monitoring small catchments of less than 3,000 hectares. Another approach, though not used widely as yet, would be to perform field monitoring using low- cost, hand-held video cameras to record construction or plantation activities as they are executed. MonitorinR Results Monitoring results also involves field inspections and supervision, but rather than simply indicating physical progress, attempts to assess the effectiveness of activities in terms of survival rates, erosion control, fuel- wood produced. Closely related is the concept of monitoring benefits. Sample surveys should be done on a regular basis (e.g., annually or quarterly) to assess the beneficiaries' perception of the project's effect on pasture reha- bilitation, improved farm and fuelwood yields, and reductions in erosion rates. Village-level meetings should be held on a regular basis to elicit beneficiary response to the progress of works completed. Again, monitoring results could use a combination of personal interviews, video recordings, public meetings, and formal questionnaires to record information. Monitoring results could also include more detailed long-term studies related to project progress, including information on disbursements, the type and location of erosion control structures financed, and dissemination of erosion control technology through extension services. Other programs might focus on physical monitoring for recording the effect of project investments on the rates of erosion. Monitoring Benefits Monitoring benefits attempts to describe the annual increments to production attributable, either directly or indirectly, to project invest- ments. This monitoring function focuses on the interaction between project activities and reactions of the target populations. In watershed projects, it includes, but is not limited to, assessment of grazing, farming, and other benefits. Such monitoring is usually based on periodic household surveys and should be supported for a reasonable time period following completion of the project. Typically, monitoring beneficiaries would attempt to account for changes in household incomes and relative abundance of land-based resources such as fodder, fuel, access to grazing, or water supply that could be related to project investments. - 169 - Interim Project Evaluations This function involves the careful review of experience during imple- mentation, including institutional problems, technical achievements and con- straints, and socioeconomic benefits derived. Interim evaluation would be based on field information collected in the monitoring stages of the project. Interim reporting must go beyond an accounting of project expenditures and physical progress and should focus on the relevance, performance, efficiency and impact of the project in the context of stated goals. Ideally, both interim and overall project evaluation should be performed by an independent external agency such as a university, research institute, or qualified consul- tants. Evaluation of Impact of Physical Treatments Impact evaluation involves detailed and objective evaluation of the overall effectiveness of watershed treatments such as soil conservation, pas- ture rehabilitation or land management practices. Ideally, the evaluation should also be carried out by a university, independent research institute, or qualified consultants. With respect to watershed projects, impact evaluation can be a very difficult task as many factors external to the project can influence the viability of investments. Report Generation (Internal/External) Report generation is an important element of all PME units, though typically it is overlooked. This function addresses the requirements for periodic reporting to the state on physical progress. The process is essen- tially manual, although word-processing facilities could be introduced to facilitate report writing. The reports would be generated directly from the data collected and systematically recorded in the watershed management infor- mation system. Evaluation teams should also have the capacity to generate special- purpose reports for external institutions such as development agencies, NGOs, research groups, or public information campaigns. This capacity should be flexible enough to address the needs of formal evaluations such as mid-term reviews, project completion studies, and ex-post reviews, as well as informal management reviews. Like many other elements of the PME unit, report genera- tion must rely on the quality and timeliness of information contained within the management information system. The reports and accounts generated by an evaluation team could be presented in a number of different formats such as formal written reports, diagnostic analyses or verbal and visual presenta- tions. SUMMARY Watershed planning, monitoring and evaluation tasks are frequently cited as among the most important components of successful projects. Never- theless, the potential of PME activities will be achieved only if it is clearly understood who is to use the results and for what purposes and if the practical limitations of available methodologies and techniques are accepted. In general, there is agreement that monitoring and measurement of physical processes (erosion, deforestation) or physical delivery of project inputs is - 170 - easier than measuring the less tangible, people-centered goals of watershed rehabilitation programs such as increasing beneficiary participation. In designing PME components of watershed projects, the mutually rein- forcing nature of these tasks must be clearly understood. Casley and Kumar (1987) note that "monitoring and evaluation are separated by their objectives, reference periods, requirements for comparative analysis, and primary users. But having emphasized the differences, we need to enter a qualification: in spite of these distinct functions, there are common features which highlight the relation between them. In many cases, the same data are used for both and the indicators for monitoring may be included in the range of information required for evaluation, but they will be reviewed over longer time spans, with the use of comparative analytical techniques, and a larger group of users will be addressed." One conclusion should be made clear; PME methodologies and procedures need to be flexible and well adapted to local circumstances. Due to the wide variation in objectives, scale and scope of investments in watershed projects, there can be no blueprint which is universally applicable or acceptable in all situations. Each situation will vary regarding the technical capabilities of existing staff, institutional commitment to PME tasks, availability of histo- rical, time-series data and so on. In all situations, PME must be seen as an evolutionary process, building slowly on existing capabilities over relatively long time periods. This is particularly important when monitoring the effects of project investment on the productivity of natural systems such as forests, soils, or water. - 171 - REFERENCES Casley, Dennis J. and Krishna Kumar. 1987. Proiect Monitorina and Evaluation in Agriculture. Published for the World Bank, The Johns Hopkins University Press: Baltimore. _ 173 - 8. BIBLIOGRAPHIES ON SOIL AND MOISTURE CONSERVATION TECHNOLOGIES The more than 200 studies cited in Chapter 2 concerning on-farm impacts of soil and moisture conservation technologies on surface runoff, erosion/sedimentation and productivity and yield are arranged in four tables. This separate presentation make them more readily accessible to the reader who wishes to locate specific references. Table 8.1 is the complete list of research studies in alphabetical order. Tables 8.2, 8.3 and 8.4 pertain respectively to the three topics: surface runoff, erosion/sedimentation and productivity and yield, giving pertinent information from the references which are listed approximately in the order of appearance and grouped according to the main treatment being studied. TABLE 8.1: Global Studies on On-Farm Impacts 1. Abdurachman, A., Barus, A. and U. Sudirman. 1985. Peranan Pola Tanam dalam Usaha Pencegahan Erosi pada Lahan Pertanian Tanaman Semusin. Pemberitaan Penelitian Tanah dan Pupuk 4:41-46. 2. Abennar. W. 1986. Pengaruh Pupuk Kandang dan Tanaman Terhadap Erosi dan ar. W. 1986. Pengaruh Pupuk Kandang dan Tanaman Terhadap Erosi dan Limpasan Permukaan Pada Regosol Abu Vulkan. Tesis. Fakultas Pertanian Universitas Brawijaya. Malang. 3. Abujamin, S., Abdurachman, A., and Suwardjo. 1985. Contour grass strips as a low cost conservation practice. ASPAC Extension Bull. No.221.pp.1-7. 4. Abujamin, S., Abdurachman, A. and U. Kurnia. 1983. Strip Rumput Permanen Sebagai Salah Satu Cara Konservasi Tanah. Pemberitaan Penelitian Tanah dan Pupuk 1:16-20. 5. Ahmad, A.R. 1983. Pengaruh Pengelolaan Tanah dan Tanaman Terhadap, Erosi dan Limpasan Permukaan. Tesis Fakultas Pertanian. Universitas Brawijaya, Malang. 6. Alegre, J.C., Cassel, D.K., and D.E. Bandy. 1986. Effects of land clearing and subsequent management on soil physical properties. Soil Sci. Soc. Am. J. 50(6):1379-1384. 7. Ambasht, R.S., and K.N. Misra. 1980. Conservation studies of a hilly grassland. In : Proceedings of the Fifth International Symposium on Tropical Ecology. Ed. J.I. Furtado. Kuala Lampur, Malayasia. pp. 133-139. - 174 - 8. Arsjik, A.H. 1976. Pengaruh Kemiringan Lereng dan Intensitas Hujan Terhadap Jumlah Aliran Permukaan dan Kehilangen Tanah Pada Tanah Podzolik Jonggol dan Latosol Citayam. Tesis Fakultas. Kehutanan Institut Pertanian Bogor. Bogor. 9. Artaban. 1985 Peranan Pengelolaan Tanah dan Tanaman dalam Menekan laju Erosi di Daerah Aliran Sungai Waduk Bening, Madiun. Tesis Fakultas Pertanian, Universitas Pembangunan Nasional Veteran, Surabaya. 10. Baird, R.W., Richardson, C.W., W.G. Knisel. 1970. Effects of conservation practices on storm runoff in the Texas Blackland Prairie. ARS/USDA Tech. Bull. no. 1406. 31 pp. 11. Baird, R.W. and C.W. Richardson. 1969. Effects of conservation treatment on water yield. In : Effects of Watershed Changes on Streamflow. Eds. W.L. Moore, and C.W. Morgan. Univ. of Texas Press, Austin, Texas. pp. 69-78. 12. Beasley, R.S. 1979. Intensive site preparation and sediment losses on steep watersheds in the Gulf Coastal Plain. Soil Sci. Soc. Am. J. 43(2):412-417. 13. Bennet, H.H. 1939. Soil Conservation. McGraw Hill. New York, New York. 14. Bertoni, S., Lombardi, F. and R. Benatti. 1975. Equacao de perdas de solo. Boletin Tecnico no. 21, Secao Conservacao do Solo. Instituto Agronomico de Estado de Sao Paulo, Brazil. 15. Bertoni, S., Pastana, I., Lombardi, F. and R. Benatti. 1972. page 2 of 16 Conclusoes gerais das pesquisas sobre Conseracao de solo no Instituto Agronomico, Circular 1 20 de la Divisao de Solos, Instituto Agronomico do Estado de Sao Paulo, Brazil. 16. Bhatia, K.S. and H.P. Choudhary. 1977. Runoff and erosion losses and crop yields from slopy and eroded alluvial soils of Uttar Pradesh in relation to contour farming and fertilization. Soil Conservation Digest, Vol. 5, No. 2, October 1977. pp.16-22. 17. Black, A.L. 1968. Conservation bench terraces in Kansas. Trans. of ASAK. 11:(3)387-388. 18. Black, A.L. 1968. Conservation bench terraces in Montana. Trans. of ASAE. 11:(3)393-395. 19. Blong, R.J. 1983. Gully sidewall development in New South Wales, Australia. In : Soil Erosion and Conservation. Eds, S.A. El-Swaify, W.C. Moldenhauer and A. Lo. Soil Conservation Society of America. Ankeny, Iowa. pp.574-584. 20. Bo-Myeong Woo. 1984. Status of watershed forest influence research in Korea. In: Country Papers On Status of Watershed Influence Research in Asia and the Pacific. Ede. L. Hamilton, M. Bonnel and D.E. Mercer. East-West Center Working Paper. Honolulu, Hawaii. pp.153-173. _ 175 - 21. Bonsu, M. 1983. Organic residues for less erosion and more grain in Ghana. In : Soil Erosion and Conservation. Eds. S.A. El-Swaify, W.C. Moldenhauer and A. Lo. Soil Conservation Society of America. Ankeny, Iowa. pp.615-621 22. Bowie, A.J. 1982. Investigations of vegetation for stabilizing eroding stream banks. Trans. of ASAE. 25:(6)1601-1606,1611. 23. Bradley, G.R. and H.Y. Kim. 1964. Site selection and soil characteristics. In: Technical Guide for Upland Development Using the Bench Terrace Method. Coursebook from training and demonstration held 9 February-18 April 1964 at Sosa, Kyonggi Do, Korea. pp.3-4. 24. Branson, F.A., Miller, R.F., and I.S. McQueen. 1966. Contour furrowing, pitting and ripping on rangelands of the western United States. Jour. Range Mgmt. 19:182-190. 25. Brown, A.L., and A.C. Everson. 1952. Longevity of ripped fourrows in southern Arizona desert grassland. Jour. Range Mgmt. 5:415-419. 26. Brown, D., Hallman, R.G., Lee, C.R., Skogerboe, J.G., Eskew, K., Price, R.A., Page, N.R., Clar, M., Kort, R. and H. Hopkins. 1986. Reclamation and Vegetative Restoration of Problem Soils and Disturbed Lands. Noyes Data Corporation, Park Ridge, New Jersey. 560 pp. 27. Buol, S.W., Hole, F.D., and R.J. McKraken. 1973. Soil Genesis and Classification. Ames, IA. Iowa State Univ. Press. 28. Burwell, R.E., Allmaras, R.R. and L.L. Sloneker. 1966. alteration of soil surfaces by tillage and rainfall. Consv. 21:61-63 Structural J. Soil Water 29. Campbell, I.A. 1983. The partial area concept applied to sediment source area. In : Soil Erosion and Conservation. Eds. S.A. El-Swaify, W.C. Moldenhauer and A. Lo. Soil Conservation Society of America. Ankeny, Iowa. pp.128-138. 30. Carson, B. 1989. Soil Conservation Strategies for Upland Areas in Indonesia. Occasional Papers of the East-West Environment and Policy Institute. Paper No. 9. 120pp. 31. Carter, D.L. 1983. Erosion and sediment losses in furrow-irrigated land. In : Soil Erosion and Conservation. Eds. S.A. El-Swaify, W.C. Moldenhauer and A. Lo. Soil Conservation Society of America. Ankeny, Iowa. pp.355-364. page 3 of 16. 32. Castellanos, V. and J.L. Thomas. 1980. Application of multiple-use research on a watershed in Honduras. Proc. from IUFRO/MAB Conference: Research on Multiple-use of Forest Resources. 33. Chan, C., 1981. Evaluation of soil loss factors on cultivated slopelands of Taiwan. ASPAC Technical Bull. No. 55. 28 p. - 176 - 34. Chang, S.M., Huang, C.T., and U.C. Wang. 1968. Soil and water conservation experiment on a pineapple orchard. Fengshan Tropical Horticultural Experiment Station, No. 55. 35 Chakravarty, R.K., Vasudevaiah, R.D., Guha, D.P. and S.P.S. Teotia. 1966. Soil moisture studies in the terraced land to explore its utility for 'rabi' cultivation in the upper catchment of Damodar Valley. J. Soil Water Consv. India. 14(3&4):27-32. 36. Charreau, C. 1972. Problems in agricultural usage of tropical soils for annual cropping. Agron. Tropicale 27:905-929. 37. Chiun-Ming, L. 1983. Impact of check dams on steep mountain channels in northeastern Taiwan. In : Soil Erosion and Conservation. Eds. S.A. El-Swaify, W.C. Moldenhauer and A. Lo. Soil Conservation Society of America. Ankeny, Iowa. pp.540-548. 38. Choudary, P.C. and B.N. Chatterjee. 1967. Moisture conservation under mulches in the eroded terraced soils of Ranchi. J. Soil Water Consv. India. 15(1&2):61-71. 39. Cooper, J.R., Gilliam, J.W., Daniels, R.B., and W.P. Robarge. 1987. Riparian areas as filters for agricultural sediment. Soil Sci. Soc. Am. J. 51(2):416-420. 40. Cordero, A. 1964. The effect of land clearing on soil fertility in the tropical region of Santa Cruz, Bolivia. M.S. Thesis, Univ. of Fla., Gainseville. 102 pp. 41. Costales, E.F. and A.B. Costales 1985. Stabilization of streambanks and riparian zones by riprap combined with selected vegetative engineering structures. Sylvatrop Philipp. For. Res. 10:(1)17-33. 42. CSWCRTI. 1982. 25 years of Research on Soil and Water Conservation in Southern, Hilly, High Rainfall Regions. Monograph No.4. Central Soil and Water Conservation Research and Training Institute, Dehradun, India. 212 pgs + annexes. As reported in : K.G. Tejwani. 1989. White Cover Report - Draft Material - Chapter . World Bank Internal Document. February 5, 1989. page 4 of 16 TABLE 4 43. Dallaire, G. 1976. Filter fabrics bright future in road and highway construction. Civil Engineering 46(5):61-65 44. De Boodt, M., Van Den Berghe, C. and D. Gabriels. 1979. Fertilizer losses associated with soil erosion. In: Soil Physical Properties and Crop Production in the Tropics. Eds. R. Lal and D.J. Greenland. John Wiley and Sons. Chichester, England. pp.455-464. 45. Dendy, F.E. 1982. Distribution of sediment deposits in small reservoirs. Trans. of ASAE. 25:(l)100-104 46. Dhruva Narayana, V.V. 1986. Soil and water conservation research in India. Indian Journal of Soil Conservation. 14(3):22-31. - 177 - 47. Dickenson, R.E.,Langley, B.C., and C.E. Fisher. 1940. Water and soil conservation experiments at Spur, Texas. Texas Agri. Exper. Station Bull. 587. 67 pp. 48. Doty, R.D., Wood, H.B., and R.A. Merriam. 1981. Suspended sediment production from forested watersheds on Oahu, Hawaii. Water Resources Bull. 17(3)298-307. 49. Dunne, T. 1979. Sediment yield and land use in tropical catchments. Journal of Hydrology (Amsterdam). 42(3/4):281-300. 50. Edwards, K. 1983. Analysis of Runoff and soil loss from long-term plots. In : Soil Erosion and Conservation. Eds. S.A. El-Swaify, W.C. Moldenhauer and A. Lo. Soil Conservation Society of America. Ankeny, Iowa. pp.472-479. 51. El-Swaify, S.A., Dangler, E.W., and C.L. Armstrong. 1982. Soil Erosion by Water in the Tropics. College of Tropical Agriculture and Human Resources. Univ. of Hawaii, Manoa. 173 pp. 52. Elwell, H.A. and M.A. Stocking. 1988. Loss of nutrients by sheet erosion is a major hidden farming cost. The Zimbabwe Science News. Vol.22, Nos.7/8 July/August 1988. pp.79-82. 53. FAO. 1976. Soil conservation for developing countries. FAO Soils Bull. No. 30. pg.81 54. Felipe-Morales, F., Meyer, R., Alegre, C. and C. Vittorelli. 1979. Losses of water and soil under different cultivation systems in twp Peruvian Locations, Santa Ana (Central Highlands) and San Ramon (Central High Jungle), 1975-1976. In: Soil Physical Properties and Crop Production in the Tropics. Eds. R. Lal and D.J. Greenland. John Wiley and Sons. Chichester, England. pp.489-499. 55. Ffoliot, P.F. 1980. Watershed Resource Management and Environmental Monitoring in Humid Tropical Ecosystems. UNESCO/MAB/AID. 56. Figueroa, J.F.L. and M.A. Garduno. 1981. Evaluacion de cinco tipos de terrazas en suelos de ladera de la cuenca del rio Tezcoco. 57. Filipovic. 1968. Erosion of arable land in the upper Jasenica watershed and its influence on winter wheat yield. Arh. Polypr. Nauke. 21(74) 16-17. 58. Finkel, H.J. 1986. Semiarid Soil and Water Conservation. CRC Press, Boca Raton, Florida. 126 p. 59. Florido, L.V. 1985. Check dams for the control of gully erosion in the pine forest watersheds. Sylvatrop Philipp. 10:(1)9-16. 60. Forest Service USDA, Region 5. 1934. Guideline for Watershed Improvement Measures. 69 pp. - 178 - 61. Frye, W.W., Ebelhar, S.A., Murdock, L.W. and R.L. Blevins. 1982. Soil erosion effects on properties and productivity of two Kentucky soils. J. Soil Sci. Soc. Am. 46:1,051-1,055. 62. Gayoso, J.A. and A.A. Iroume. 1984. Soil disturbance from logging in Southern Chile. In : Symposium on Effects of Forest Land Use On Erosion and Slope Stability. edited by C.L. O'Loughlin and A.J. Pearce. 1984. East-West Center. Honolulu, Hawaii. pp.203-209. 63. Gintings, A.N. 1982. Aliran Permukaan dan Erosi dari Tanah yang Tertutup Tanaman Kopi dan Hutan Alam di Sumberjaya-Lampung Utara. Laporan No.399, Balai Penelitian Hutan. Bogor. 64. Graham, O.P. 1984. Gully erosion. J. Soil Cons. N.S.W. 40:(1)31-37. 65. Greenland, D.J. and R. Lal. 1979. Towards optimizing soil physical properties for sustained production from soils in the tropics. In: Soil Physical Properties and Crop Production in the Tropics. Eds. R. Lal and D.J. Greenland. John Wiley and Sons. Chichester, England. pp.529-530 66. Grohman, F. 1960. Distribucao de tamanho de poros en tres de solos do Estado de Sao Paulo. Bragantia 19:319-328. 68. Gupta, S.K., Das, D.C., Tejwani, K.G., Chittaranjan, 5., and Srinivas. 1971. Mechanical measures of erosion control. In : Soil and Water Conservation Research, 1956-1971. Eds. K.G. Tejwani, S.K. Gupta and H.N. Mathur. Indian Council of Agricultural Research. New Delhi. pp.146-182. 69. Haas, H.J. and W.O. Willis. 1968. Conservation bench terraces in North Dakota. Trans. of ASAE 11(3):397-398,402. 70. Hamilton, L.S. 1985. Overcoming myths about soil and water impacts of tropical forest land uses. 71. Hardjono, D. 1987. Demonstarsi UPSA dan Permasalahannya dalam Lokakarya Pelaksanaan Rehabilitasi Lahan dan Konservasi Tanah Secara Terpadu di Sub-DAS Konto. Malang/Batu, 11-12 Maret 1987. Proyek Cali Konto Dirjen. Reboisasi dan Rehabilitasi Lahan (RRL) Departemen Kehutanan RI. Bekerjesama Dengan Dirjen Kerjasama Internasional (DGIS) Departamen Luar Negeri Kerajaan Belanda. Hal. 2.1-2 s/d 2.1-7. 72. Harjono. 1979. Soil erosion study in Upper Solo river basin. Balai Teknologi Pengelolaan DAS Solo. Surakarta. 73. Harrold, L.L. and W.M. Edwards. 1972. A severe rainstorm test of no-till corn. J. Soil Water Conserv. 27:30. TABLE 4 74. Harrold, L.L. and F.J. Dragoun. 1969. Effect of erosion control land treatment on flow from agricultural watersheds. Trans. of ASAE 12:(6)857-861. - 179 - 75. Hauser, V.L. 1968. Conservation bench terraces in Texas. Trans. of ASAZ. 11:(3)385-386,392. 76. Hauser, V.L. and M.B. Cox. 1962. Evaluation of the Zingg conservation bench terrace. Agri. Eng. 43: (8)462-467. 77. Hibbert, A.R. 1969. Water yeild changes after converting a forest catchment to grass. Water Resources Research 5:634-640. 78. Houston, W.R. 1965. Soil moisture response to range improvement in the Northern Great Plains. Jour. Range Mgmt. 18:25-30. 79. Hurni, H. 1982. Soil erosion in Huai Thung Choa - Northern Thailand: Concerns and constraints. Mountain Research and Development 2(2): 141-151. 80. Ives, R.M. and C.F. Shaykewich. 1987. Effect of simultated soil erosion on wheat yeilds on the humid Canadian Prairie. J. Soil Water Consv. 42(3):205-208. 81. Iwamoto, M. and K. Murikami. 1983. Effect of check dams on a torrent. J. Jap. For. Soc. 65:(12)458-464. 82. Jaiswal, S.P. 1966. Influence of grasses under natural conditions on the improvement of soil fertility. J. Indian Soc. Soil Sci. 14(3):205-209. 83. Jepson, H.G. 1939. Prevention and control of gullies. U.S.D.A. Farmers Bulletin no. 1813. 59 pp. 84. Jones, O.R. and R.N. Clark. 1987. Effects of furrow dikes on water conservation and dryland crop yields. Soil Sci. Soc. Am. J. 51(5)s 1307-1314. 85. Kalbande, A.R. and E. Velappan. 1977. Effect of cover cropping on soil fertility under arecanut plantation. J. Soil Water Consv. India. 27(1&2):72-75. 86. Kampen, J. 1979. Watershed management and technology transfer in the semi-arid tropics. In : Development and Transfer of Technology for Rainfed Agriculture and the SAT Farmer, Proc. of the Inaugural Symp. at ICRISAT. ICRISAT, Andra Pradesh, India. pp.111-119. 87. Keep, T. 1987. Investigation of terrace breaching in Missouri. In% Optimum Erosion Control at Least Cost - Proceedings of the National Symposium on Conservation Systems. Am. Soc. Ag. Eng., St. Joseph, Michigan. pp.384-390. 88. Khan, A.D. 1962. Measurement of increase in productivity by adopting soil and water conservation practices. J. Soil Water Consv. India. 10(1&2):25-33. 89. Koelliker, J.K. 1981. Impact of terrace systems on water yield. ASAE Paper no. 81-2051. 6 pp. - 180 - 90. Koichi, 5., Sakurai, Y. and K. Takase. 1984. Soil loss and runoff characteristics of shifting cultivation land in Iriomote Island. In Symposium on Effects of Forest Land Use On Erosion and Slope Stability. edited by C.L. O'Loughlin and A.J. Pearce. 1984. East-West Center. Honolulu, Hawaii. pp.147-153. 91. Kurnia, U. and R.L. Watung. 1985. Pengaruh Pengolahan Tanah'dan Mulsa Sisa Tanaman Terhadap Produktivitas Tanah Latosol Kadipaten. Makalah Lokakarya Hasil Penelitian Pengelolaan DAS dan Dampak Pemanfaatan Sumber Alam Terhadap Lingkungan. Cisarua, Bogor 9-10 Maret 1985. 92. Laflen, J.M., Highfill, R.E., Amemiya, M. and C.K.' Mutchler. 1985. Structures and methods for controlling water erosion. In: Soil Erosion and Crop Productivity. edited by R.F. Follett and B.A. Stewart. SSSA Publishers. Madison, Wisc. pp.431-442. 93. Lal, R. 1987. Effects of soil erosion on crop productivity. In : CRC Critical Reviews in Plant Science. Vol. 5,'Issue 4. Pg.'303-367. 94. Lal, R. 1984. Mulch requirements for erosion control with the no-till system in the tropics: a review. In: Challenges in'Africian Hydrology and Water Resources (Proc. Harare Symposium). IAHS Publ. no. 144.. pp.475-484 95. Lal, R. 1983. Soil erosion and productivity in tropical soils. In : Soil Erosion and Conservation. Eds. S.A. El-Swaify, W.C. Moldenhauer and A. Lo. Soil Conservation Society of America. Ankeny, Iowa. pp.237-247. 96. Lal, R. 1975. Role of mulching techniques in tropical soil and water management. Int. Inst. Trop. Agric. Tech. Bull. No. 1. 97. Lal, R., Kang, B.T., Moorman, F.R., Juo, J.S.R., and J.C. Moomaw. 1975. Soil Management problems and possible solutions in western Nigeria. In: Soil Management in Tropical America. Eds. E. Bornemisza and A. Alvarado. North Carolina State Univ., Raleigh. pp. 372-408. 98. Lal, R. 1974. Soil erosion and shifting agriculture'. In: Shifting Cultivation and Soil Conservation in Africa. FAO Soils Bull. 24, FAO, Rome. pp.48-71. 99. Lattanzi, A.R., Meyer, L.D., and M.F. Baumgardner. 1974. Influences of mulch rate and slope steepness on interrill erosion. Soil Sci.'Soc. Am. J. 38(6):946-950. 100. Le Buanec, B. 1972. Dix ans culture motorisee sur un bassin versant du centre Cote'd'Ivoire. Evolution de la fertite' et de la production. Agron. Tropicale 27:1191-1211. 101. Leopold, L.B., Wolman, M.G. and J.P. Miller. 1964. Fluvial Processes i Geomorphology. Freeman, San Francisco, Calif. 522 pp. - 181 - 102. Liao, H.C. 1981. Soil Conservation measures for steep orchards in Taiwan. In: South-East Asian Regional Symposium on Problems of Soil Erosion and Sedimentation. pp.301-310. 103. Liao, M.C., Chang, H.H., and R.S. Lin. 1975. A comparison of runoff from bench terraces and hillside ditches. JCSWC 1:24-32. 104. Liao, N. C . 1972 . Study on soil and water conservation in sugarcane land. TSES Report Vol . 55, pp . 75-90. 105. Lin, C P F. 1969 . Experimental study of erosion and runoff of Yellowish-Brown Gravelly Red soil in Pinchin Tea District. Experimental Reports of Soil and Water Conservation in Taiwan, 1:139-152. 106. Litaniwan, B.G. 1986. Pengaruh Vegetasi Penutup Tanah Terhadap Erosi dan Aliran Permukaan pada Tanah Tegalan di DAS Tritis Kecamatan Wdaslingtan Kabupaten Wonosobo. Skripsi Fakultas Kehutanan, Universitas Gadjah Mada. Yogyakarta. 107. Lombardi Neto, P., and J. Bertoni. 1975. Tolerencia de perdas para solos do Estado de Sao Paulo . Bol . Tec . Inst. Agron. 28 :1-12. 108. Luft, G. and G. Morgenschweis-. 1984. View on problems deriving from large-scale terracing in the wine-growing area of the Kaiserstuhl Mountains J. Rural Eng. and Develop . pp .138-148. 109. Luft, G., Morgenschweis, G. and A. Vogelbacher. 1982. The effects of large scale terracing on hydrological processes. Proc. Symp. Hydrolog. Research Basins. Sonderh. Landeshydrologie, Bern, 1982. pp 543-553. 110. Lugo-Lopez, M.A., J . Juarez, and J .A. Bonnet. 1968 . Relative infiltration rates of Puerto Rican soils. J. Agr. Univ. Puerto Rico 52: 233 - 240. 111. Lukito, H . 5 . 1984 . Pengaruh Jenis Tanaman Terhadap Erosi dan Limpasan Permukaan di Desa Gubug Klakah, Kecamatan Poncokusumo, Malang. Tesis Fakultas Pertanian, Universitas Brawijaya, Malang. 112. Lyles, L. 1967. Effect of dryland leveling on soil moisture storage and grain sorghum yield. Trans . of ASAE 8(4): 523-525. 113. Hannering, J.V. and L.D. Meyer. 1963. The effects of various rates ofsurface mulch on infiltration and erosion. Soil Sci. Soc. Am. J. 27 (1) :84-86. 114. Marques, J .Q.A. and J . Bartoni . 1961. Sistemas de preparo do solo em relacao a producao e a erosao. BraRantia 20: 403 -459. 115. Marsh B. 1971. Immediate and long- term effects of soil loss. Proc Australian Soil Conservation Conf.- 1971. - 182 - 116. Marston, D., Anecksamphant and R. Chirasathavorn. 1983. Soil conservation and land development in Thailand. In i Soil Erosion and Conservation. Ede. S.A. El-Swaify, W.C. Moldenhauer and A. Lo. Soil Conservation Society of America. Ankeny, Iowa. pp . 634 - 641 . 117. Massee, T.W., and H.0 Waggoner. 1985. Productivity losses from soil erosion on dry cropland in the intermountain area. J. Soil Water Consv. 40(5):447-450. 118. Mbagwu, J.S.C., Lal, R., and T.W. Scott. 1984. Effects of desurfacing of Alfisols and Ultisols in southern Nigerias I. Crop Performance. Soil Sci. Soc. Am. J. 48(4) :828-833. 119. Mbagwu, J.S.C., Lal, R., and T.W. Scott. 1984. Effects of desurfacing of Alfisols and Ultisols in southern Nigeria: II. Changes in soil physical properties. Soil Sci. Soc. Am. J. 48(4):834-838. 120. McCorkle, J.S., and T. Dale. 1941. Conservation practices for the rangelands of the southern Great Plains. S.C.S./U.S.D.A. 18 pp. 121. McPhee, P.J., Hartman, M.O., and N.F. Kieck. 1983. Soil erodibility and crop management factors of soils under pineapple production. ASAS Paper No. 83-2073. 122. Mensah-Bonsu and H.B. Obeng. 1979. Effects of cultural practices on soil erosion and maize production in the semi-deciduous rainforest and forest-savanna transitional zones of Ghana. In: Soil Physical Properties and Crop Production in the Tropics. Eds. R. Lal and D.J. Greenland. John Wiley and Sons. Chichester, England. pp.509-519. 123. Meyer, L.D., Wischmeier, W.H., and G.R. Foster. 1970. Mulch rates required for erosion control on steep slopes. Soil Sci. Soc. Am. J. 34(6):928-930. 124. Mickelson, R.H. 1968. Conservation bench terraces in Eastern Colorado. Trans. of ASAE. ll:(3)389-392 125. Mickelson, R.H. 1966. Level pan system for spreading and storing watershed runoff. Soil Sci. Soc. Am. J. 30(3)t388-392. 126. Mickelson, R.H. 1966. spreading runoff. Level pan construction for diverting and Trans. of ASAE 9:(4)568-570. 127. Mielke, L.N. 1985. Performance of water and sediment control basins in northeastern Nebraska. J. Soil Water Consv. 40(6):524-528 128. Miller, C.R., Worlbrown, R. and R.R. Turner. 1962. Upland gully sediment production. IASH Pub. 159. 129. Millington, A.C. 1984. Indigeneous soil conservation studies in Sierra Leone. Int Challenges in African Hydrology and Water Resources (Proc. Harare Symposium). IAHS Publ. no. 144. pp.529-537. - 183 - 130. Miranda, S.M., Pathak, P., and K.L. Srivastava. 1982. Runoff management on small agricultural watersheds: The ICRISAT experience. National Seminar on A Decade of Dryland Agricultural Research in india and Thrust in the Eighties. Hyderbad, India. 131. Mittal, S.P., Singh, P., Agnihotri, R.C., Dyal, S.K.N. and A.D. Sud. 1986. Effects of conservation farming practices on runoff soil loss and yield of maize in Shiwalik foothills. Indian J. Soil Conservation. Vol. 14, No. 1. April 1986. pp. 1-6. 132. Morgan, R.P.C. 1986. Soil Erosion and Conservation. John Wiley & Sons, New York, New York. 297 pp. 133. Moura, F.W. and S.W. Buol. 1972. Studies of a Latosol Roso in Brazil. Experientiae 13:201-234. 134. Mueller-Dombois, D. 1973. A non-adapted vegetation interferes with water removal in a tropical rainforest area in Hawaii. Tropical Ecology. 14(l):1-18. TABLE 4 135. Murad, A. 1985. Penilaian Tindakan Pengawetan Tanah Dengan Tehnik Sipil Dalam Usahatani lahan Kerinjg di Jampang Tengah, Sukabumi. Tesis Fakultas Kehutanan. Institut Pertanian Bogor. 136. North Carolina State University. 1973,1974. Research on Tropical Soils. Annual Report. Soil Science Dept., North Carolina State University, Raleigh. 137. Norton, L.D., Cogo, N.P. and W.C. Moldenhauer. 1983. Effectiveness of mulch in controlling soil erosion. In : Soil Erosion and Conservation. Eds. S.A. El-Swaify, W.C. Moldenhauer and A. Lo. Soil Conservation Society of America. Ankeny, Iowa. pp.598-606 138. O'Loughlin, C.L., and A.J. Pearce. 1976. Influence of cenozoic geology on mass movement and sediment yield response to forest removal, North Westland, New Zealand. Bull. Int. Assoc. Engineering Geology 14:41-46 139. Osuji, G.E., Babalola, 0. and F.O. Aboaba. 1980. Rainfall erosivity and tillage practices affecting soil and water loss on a tropical soil in Nigeria. J. Environ. Management 10:207-217. 140. P3HTA. 1987. Penelitian Terapan Pertanian Lahan Kering dan Konservasi. Progress Report 1986-1987. P3HTA. Salatiga. 141. P3HTA. 1988. Penelitian Terapan Pertanian Lahan Kering dan Konservasi. Laporan Tahunan 1986J1987. P3HTA. UACP-FAS. Salatiga. 141. P3HTA. 1988. Penelitian Terapan Pertanian Lahan Kering dan Konservasi. Laporan Tahunan 1986-1987. P3HTA. UACP-FAS. 142. Pathak, P., Miranda, S.M. and S.A. El-Swaify. 1983. Improved rainfed farming for semiarid tropics. In : Soil Erosion and Conservation. Eds. S.A. El-Swaify, W.C. Moldenhauer and A. Lo. Soil Conservation Society of America. Ankeny, Iowa. pp.338-354. - 184 - 143. Patnaik, N. 1975. Soil erosiont A menace to the nation. Indian Farming 24(11):7-10. 144. Patnaik, U.S., Satry, G. Sharda, V.N., Juyal, G.P., and V.S. Katiyar. 1986. Soil conservation engineering practices for agricultural lands. Indian Journal of Soil Conservation. 14(3):85-89. 145. Pearce, A.J. and L.S. Hamilton. 1986. Water and Soil Conservation Guidelines for Land-use Planning. Environmental and Policy Institute, East-West Center. Honolulu, Hawaii. 43 pp. 146. Peterson, Branson. 1962. 147. Pickup, G., Higgins, R.J., and R.F. Warner. 1981. Erosion and sediment yield in the Fly River drainage basins, Papua New Guinea. IASH 132:438-457 148. Pinczes, Z. 1980. 'A muvelesi agak es modok hatasa a talajeroziora'. Foldr. Kozl.. 28:357-359 149. Pla, I., Florentino, A. and D. Lobo. 1983. Soil and water conservation problems in the Central Plains of Venezuela. In : Soil Erosion and Conservation. Eds. S.A. El-Swaify, W.C. Moldenhauer and A. Lo. Soil Conservation Society of America. Ankeny, Iowa. pp. 66-77. TABLE 4 page 11 of 16 150. Popenoe, H.L. 1957. The influence of the shifting cultivation cycle on soil properties in Central America. Proc. Ninth Pacific Sci. Congr. (Bangkok) &:72-77. 151. Powell, G.M. and L.P. Herndon. 1987. Maintenance of field conservation structures. In: Optimum Erosion Control at Least Cost - Proceedings of the National Symposium on Conservation Systems. Am. Soc. Ag. Eng., St. Joseph, Michigan. pp.374-383. 152. Proyek P3DAS Solo. 1982. Pengaruh Teras Bangku dan Penghijaun (Silvopasture) Terhadap Laju Erosi. Proyek Pusat Pengembangen Pengelolaan Daerah Aliran Sugai (P3DAS). Surakarta. 153. Randhawa, N.S. and J. Venkateswarlu. 1979. Indian experiences in the semi-arid tropics: prospect and retrospect. In : Development and Transfer of Technology for Rainfed Agriculture and the SAT Farmer, Proc. of the Inaugural Symp. at ICRISAT. ICRISAT, Andra Pradesh, India. pp.207-219. 154. Reed, L.A. 1978. Effectiveness of sediment control techniques used during highway construction in Pennsylvania. USGS Water Supply Paper no. 2054. 155. Richadson, C.W. 1973. Runoff, erosion, and tillage efficiency on graded furrow and terraced watersheds. J. Soil Water Consv. 28(4):162-164. - 185 - 156. Richardson, C.W. 1972. Changes in water yield of small watersheds by agricultural practices. Trans. of ASAE 15s(3)591-593. 157. Rockwood, W.G. and R. Lal. 1974. Mulch tillage: A technique for soil and water conservation in the tropics. Span 17s77-79. 158. Roose, E.J. and F.X. Hasson. 1983. Consequences of heavy mechanization and new rotation on runoff and on loessial soil degradation in northern France. In : Soil Erosion and Conservation. Eds. S.A. El-Swaify, W.C. Moldenhauer and A. Lo. Soil Conservation Society of America. Ankeny, Iowa. pp. 24 - 33. 159. Sahasrabuddhe, X.R. 1964. Preliminary trials of bunding in deep black soils of the Deccan region of Maharashtra. J. Soil Water Consv. India. 12(3&4):67-79. 160. Sanchez, Pedro A.. 1976. Properties and Management of Soils in the Tropics. John Wiley and Sons, New York. 618 pp. 161. Saplaco, 5., Lapitan, R., and M. Aquino. 1980. Hydro-meteorological characterization of upland ecosystems at Mt. Makiling. In s Proceedings of the Fifth International Symposium on Tropical Ecology. Ed. J.I. Furtado. Kuala Lampur, Malayasia. pp. 411-425. 162. Scott, G.A. 1969. Vegetation and avalanching in rainfall areas of Oahu. MA Thesis, University of Hawaii. 98 p. 163. Sembiring, H. and A.M. Fagi. 1989. Paket Usahatani dalam Mengatasi Erosi di DAS Brantas, Jawa Timur. P3HTA Malang. 164. Semple. 1937. 165. Seno, K., Ohyagi, 5 . and Y. Fukushima. 1981. Erosion control on a granite mountain in Japan. In: South-East Asian Regional Symposium on Problems of Soil Erosion and Sedimentation. pp . 95-105. 166. Seubert, C . E . 1975 . Effects of land clearing on crop performance and changes in soil properties in an Ultisol of the Amazon jungle of Peru. M.S. thesis, North Carolina State University, Raleigh. 152 pp- 167. Sfeir-Younis, A. 1985. Soil Conservation in Developing Countries: A Background Report. Unpublished report to the World Bank. 582 p. 168. Sharma, R. C. and S. S. Khanna. 1978. Impact of various soil conservation practices on soil fertility and moisture conservation. J. Soil Water Consv. India. 28(1&2) :50-60. 169. Sheng, T.C., Jackson, J.K., Kraayenhagen, J., Nakasthin, N. and P. Watnaprateep. 1981. The effects of different structures on erosion and runoff. In: South-East Asian Regional Symposium on Problems of Soil Erosion and Sedimentation. pp. 301- 310. - 186 - 170. Sheng, T. C., and H. R. Stennet. 1975. Forestry development and watershed management in the upland regions - Jamaica. FAO Working Document SF/JAM 505. Rome: FAO-UNDP. 1981. Effect of mulch rate. 171. Singer, M.J., Matsuda, Y., and J. Blackard. on soil loss by raindrop splash. Soil Sci. Soc. Am. J. 45(1) :107-110. 172. Singh, A. 1974. Selection of structures for soil and water conservation. Soil Conservation Digest 2(2) :49-53. 173. Singh, G., Rajbans, D., and S.N. Bhola. 1967. Soil and water loss (run-off) studies under different vegetative covers on 0 5 and 1.OZ slopes at Kota. 1967. J. Soil Water Consv. 15(3&4) :17-23. 174. Singh, G., Singhal, A.K., and K.K.M. Nambiar. 1967. Effect of mulching on moisture retention and yield of wheat under rainfed conditions. J. Soil Water Consv. India. 15(1&2) :102-107. 175. Sinha, A.K. and C.N. Rai. 1980. Economics of common terracing methods in the uplands of Chotanagpur. J. Soil Water Consv. India. 30(1&2) s 42-46. 176. Smyle, J.W., 1987. Sediment production from small catchments in the mid-elevation wet zone of Sri Lanka. M.S. Thesis, University of Georgia. Athens, Georgia. 49 p. 177. Spomer, R.G., Saxton, K.E., and Heinemann, H.G. 1973. Water yield and erosion response to land management. J. Soil Water Consv. 28(4) 168-171. 178. Stein, O.R., Neibling, W.H., Logan, T.J., and W.C. Moldenhauer. 1986. Runoff and soil loss as influenced by tillage and residue cover. Soil Sci. Soc. Am. J . 50(6) :1527-1531. 179. Stern, P. 1981. Gabions for hydraulic structures. Appropriate Technology. 7:(4)6-8. - 187 - 180. Stewart, B.A., Unger, P.W. and O.R. Jones. 1983. Soil and water conservation in semiarid regions. In : Soil Erosion and Conservation. Eds. S.A. El-Swaify, W.C. Holdenhauer and A. Lo. Soil Conservation Society of America. Ankeny, Iowa. pp.328-337. 181. Suarez de Castro, F. 1957. Las quemas como practica agricola y sus efectos. Fed. Nac. Cafetaleros de Colombia Bol. Tec. 2. 182. Suarez de Castro, F. and A. Rodriguez. 1955. Perdidas por erosion de elementos nutritivos bajo diferentes cubiertas vegetales y con varias practicas de conservacion de suelos. Federacion Nacional de Cafeteros de Colombia Bol. Tec. 14. 183. Subagyono, K. 1988. Pengaruh Penanaman Berbagai tanaman Penutup Tanah Secara Strip Terhadap Erosi dan Limpasan Permukaan pada Kediteran Desa Jatikerta Sumberpucung Kalang. Tesis. Fakultas Pertanian Universitas Brawijaya. Malang. 184. Sub Proyek Bantuan Teknik UNDP Solo. 1979. Evaluasi Penelitian Erosi Pada Plot Percobaan Proyek TA IN5/72/006. Kerjasama antara Sub Proyek Bantuan Teknik UKDP Solo/TA IN5/72/006. Surakarta dengan Fakultas Kehutanan UGK Yogyakarta. Surakarta. 185. Sudirman and A. Abdurachman. 1981. Pengaruh Kadar Ciri Tanah, Mulsa dan Pupuk Organik Terhadap Pertumbuhan Jagung dan Pemakaian Air. dalam Kumpulan Makalah Pertemuan Teknis Proyek Penelitian Tanah di Cipayung - Bogor. 10 s/d 13 November 1981. Proyek Penelitian Tanah. Puslit Tanah - Bogor. Hal. 445-458. 186. Sudirman, Simkkaban, N., Suwardjo, H. and 5. Arsyad. 1986. Pengaruh Tingkat Erosi dan Pengapuran Terhadap Produktivitas Tanah. dalam Pemberitaan Penelitian Tanah dan Pupuk No. 6, 1986. Pusat Penelitian Tanah, Bogor. Hal. 9-14. 187. Sunardi, H. 1989. Penggunaan tanaman Strip untuk Pengendalian Erosi dan Limpasan Permukaan pada Tanaman Tumpangsari Jagung dan Ubikayu tanpa teras. Tesis. Fakultas Pertanian Universitas Brawijaya. Malang. 188. Sutrisno. 1987. Pengaruh Pemberian Jerami dan Sifat-sifat Terhadap Erosi dan Aliran Permukaan. Tesis Jerusan Tanah Fakultas Pertanian. Institut Pertanian Bogor. Bogor. 189. Suwardjo, Suhardjo, H. and S.H. Talauhu. 1988. Pengaruh Panjang Lereng dan Cara Pengelolaan Lahan Terhadap Erosi dan Pertumbuhan Tanaman Kedelai dalam Prosiding Petremuan Teknis penelitian Tanah. Cipayung 18-20 Karet 1986. Pusat Penelitian Tanah. Badan Litbang Pertanian. Bogor. 190. Suwardjo and Sofijah Abujamin. 1983. Crop residue mulch for conserving soil in uplands of Indonesia. In : Soil Erosion and Conservation. Eds. S.A. El-Swaify, W.C. Holdenhauer and A. Lo. Soil Conservation Society of America. Ankeny, Iowa. pp.607-614. - 188 - 191. Suwito, H. 1984. Pengaruh Pengelolaan Tanah dan Tanaman Terhadap Erosi dan Limpasan Permukaan di Daerah Aliran Waduk Karangkates. Tesis. Fakultas Pertanian Universitas Brawijaya. Malang. TABLE 4 page 14 of 16. 192. Team Konservasi Tanah dan Air. 1981a. Pengaruh Hulsa Sisa, Tanaman dan Pengelolahan Tanan Terhadap Erosi, Sifat-sifat Tanah, Pertumbuhan dan Hasil Tanaman Padi Pada Tanah Latosol Citayam. dalam Progres Penelitian Konservasi Tanah dan Air 1980/1981 Proyek Penelitian Tanah, Puslit Tanah. Bogor. Hal. III.10-1 s/d III.10-7. 193. Team Konservasi Tanah dan Air. 1980d. Pengaruh Mulsa Sisa Tanaman dan Pengelolaan Tanan Terhadap Erosi. Sifat-sifat Tanah dan Petembuhan Tanaman pada Tanah Latosol di Citayam dan pada Tanah Podsolik. Di Pekalongan dan laporan Progres. Penelitian Konservasi Tanah dan Air 1979/1980 Proyek Penelitian Tanah. Lembaga tanah. Bogor. Hal. IX.1 - IX.13.v 194. Tejwani, K.G. and B. Verma. Undated. Reclamation and Management of Ravine lands in India. Presented at: International Conference on Watershed Mgmt. and Land Development in the Tropics, IITA, Ibadan, Nigeria. 195. Tregubov, P.S. 1981. Effective erosion control in the USSR. In : Soil Conservation: Problems and Prospects. Ed. R.P.C. Morgan. Wiley & Sons, New York, New York. pp.451-459. 196. Trimble, S.W. 1981. Changes in sediment storage in Coon Creek Basin, Driftless Area, Wisconsin, 1853-1975. Science 214:181-183 ) 197. Unger, Paul W. 1984. Tillage systems for soil and water conservation. FAO Soils Bulletin 54. FAO, Rome. 278 pp. 198. USDA. 1981. Soil, water and related resources in the United States: Analysis of resource trends. 1980 RCA Appraisal, Part II. USDA. Wash., DC. 199. Utumo, T.W. 1986. Pengaruh Penutupan Mulsa Jerami Padi Terhadap Sifat Fisik Tanah dan Erosi Pada Latosol Darmaga yang Ditanami Padi Gogo Selama Satu Musim Tanam. Tesis Jurusan Tanah Fakultas Pertanian. Institut Pertanian Bogor. Bogor. 200. -Valentine, K.A. 1947. Effects of water-retaining and water-spreading structures in revegetating semi-desert rangeland. New Mexico Agri. Exper. Station Bull. 341. 22p. 201. Van der Weert, R. 1974. The influence of mechanical forest clearing on soil conditions and resulting effects on root growth. Trop. Agr. (Trinidad) 51:325-331. 202. Vasudevaiah, R.D., Teotia, S.P.S. and D.P. Guha. 1965. Runoff -- loss determination at Deochanda Experiment Station: II Effect of annually cultivated grain crops and perenniak grasses on sZ slopes. J. Soil Water Consv. India. 13(1&2):36-46. - 189 - 203. Verma, G.P. 1981. Erosion control on vertisols under comparatively high rainfall conditions. In: South-East Asian Regional Symposium on Problems of Soil Erosion and Sedimentation. pp.301-310. 204. Walker, S.H. and K.R. Rushton. 1986. Water losses through the bunds or irrigated rice fields interpreted through an analogue model. Agric. Water Manage. 1l:(1)59-73. 205. Wang, 1969 page 15 of 16. 206. Ward, T.J. 1983. Sediment yield modeling of roadways. In : Soil Erosion and Conservation. Eds. S.A. El-Swaify, W.C. Moldenhauer and A. Lo. Soil Conservation Society of America. Ankeny, Iowa. pp.188-199. 207. Warner, R.C., and M.C. Hirschi. 1983. Modeling check dam trap efficiency. ASAE Paper no. 83-2082. 10 pp. 208. Watnaprateep, P. 1980. First year study on soil loss and runoff from cropping under a forest plantation. Paper presented at ASEAN seminar on management of tropical forests, ASEAN forests for social welfare. December 1980, Chiang Mai, Thailand. 10 p. 209. Watters, R.F. 1971. Shifting cultivation in Latin America. FAO For. Dev. Paper 17, pp.291-299. 210. Wein, R.W. 1970. Microenvironmental effects of erosion control treatments. Utah State University, Ph.D. Dissertation. 181 pp. 211. Weltz, M, and M.K. Wood. Short-duration grazing in central New Mexico: Effects on sediment production. 41(4):262-266. 212. Whitaker, F.D., Heinemann, H.G., and W.H. Wischmeier. 1973. Chemicaliweed controls affect runoff, erosion, and corn yields. J. Soil Water Consv. 28(4):174-176. 213. Widajati. 1989. Kajian Pengaruh Sistem Penanaman dan Pemberian Mulsa Terhadap Aliran Permukaan dan Erosi Pada Lahan Tegalnan Daerah Hulu. Thesis Fakulatas Kehutanan. Universitas Gajah Mada. Yogyakarta. 214. Wieczorek, G.F., Jishan, W., and L. Tianchi. 1987. Deforestation and landslides in Yunnan, China. In: Proceedings of Conference XVII, International Erosion Control Association. IECA Pinole CA. pp.181-196. 215. Wiersum, K.F. 1983. Effects of various vegetation layers in a Acacia auriculiformis forest plantation on surface erosion in Java, Indonesia. In : Soil Erosion and Conservation. Eds. S.A. El-Swaify, W.C. Moldenhauer and A. Lo. Soil Conservation Society of America. Ankeny, Iowa. pp.79-89. 216. Wilder, C.A. and M.T. Rains. 1976. Site Improvement Techniques On The Y-LTt A Status Report. Forest Service USDA, Southwestern State and Private Forestry. 12pp. - 190 - 217. Wilson, L.G. 1967. Sediment removal from flood water by grass filtration. Trans. of ASAE 10:(1)35-37. 218. Winkworth, R.E. 1963. Some effects of furrow spacing and depth on soil moisture in central Australia. Jour. of Range Mgmt. 16:138-142. 219. Wischmeier, W.H., and D.D. Smith. 1978. Predicting rainfall erosion losses a guide to conservation and planning. USDA AH 1 537.58 pp. 220. Wischmeier, W.H., and D.D. Smith. 1965. Predicting rainfall-erosion losses from cropland east of the Rocky Mountains. Agric. Handbook no. 282. USDA, Wash.,D.C. 221. Yost, R.S., El-Swaify, S.A., Dangler, E.W. and A.K.F. Lo. 1983. The influence of simulated soil erosion and restorative fertilization on maize production on an Oxisol. In : Soil Erosion and Conservation. Eds. S.A. El-Swaify, W.C. Moldenhauer and A. Lo. Soil Conservation Society of America. Ankeny, Iowa. pp.248-261. IMPACTS OF SOIL CONSERVATION TECHNOLOGIES ON SOIL MOISTURE AND SURFACE RUNOFF CITE TECHNOLOGY\TREATMENT LOCATION CLIMATE SLOPE LAND USE SCALE IMPACT REMARKS uins$sminueawasufs.nuus=sus=wsnr=s=n33csss=g=fs==nswa==s:.B==si================-====================-z=====-== ===== 155 Conservation practices USA 7 N/A Agricultural Watershed Agricultural practices have significant In 30 years of studies little effect on on-site runoff from small effect of conservation practices watersheds. Effects management dependant. on areas larger than 120 ha has Poor management will increase water yietds. been noted -- or studied. Significant runoff reduction associated intensely managed areas. 110 Ferralsols (Oxisols) Puerto Rico U/C N/A Undisturbed Watershed Max/min infiltration rate = 15.4 / 8.4 cm/hr Range of observed infiltration Acrisols/Xitosols (Ultisols) Max/min infiltration rate = 23.6 / 7.4 cm/hr rates in undisturbed Puerto Rican Chernozems/Phaeozems Max/min infittration rate = 19.5 / 8.2 cm/hr soils. (lol ltisols) Luvisols/Podzoluvisols Max/min infittration rate = 11.5 / 2.7 cm/hr (Alfisols) CambisoIs (Inceptisols) Max/min infittration rate = 13.2 / 2.7 cm/hr Ftuvisots/Gteysots/Regosots Max/min infittration rate = 27.5 / 2.3 cm/hr (Entisols) Vertisols Max/min infittration rate - 9.5 / 0.1 cm/hr 119 Removal of 5 cm of topsoil Nigeria W/D NG NG Field 61X decrease in water hotding capacity in Measurements taken at 7 months Ultisols, 12 - 23X decrease in water holding after topsoit removal capacity for Alfisols 181 Land clearing Colombia WD/ Steep Agriculture Field Decrease in surface runoff. Attributed to increase in permeability from burning; soil AndosoL (Andept) 100 Land clearing Ivory Coast U/D NG Agriculture Field Large increase in surface runoff Soil a sandy Atfisot 166 Land clearing - bulldozer Peru CU NG Agriculture Plot 95f decrease in infiltration capacity Comparison to clearing by slash and burn method 201 Land clearing - KG blade Trinidad U/D NO Agriculture Field 291 increase in soil bulk density in top Comparison to pre-treatment 16 cm -- decreased infiltration capacity. bulk density 6 Land clearing - slash/burn Amazon CW 2-41 Agriculture Field 37X decrease in Infiltration rate Land clearing - bulldozer 96X decrease in infittration rate with straight blade Land clearing - bulldozer 90X decrease in infiltration rate with shear blade 160 forest clearing Ghana Cu NO Forestry Field 131 decrease in non-capillary porosity Soit l sandy toam 77 Conversion of forest to USA T Steep Grassland Uatershed Greater water yield, higher groundwater Comparison to pre-conversion grassland levels in deep soils. no significant parameters difference in stormflow, peak stormflow, and stonrflow duration when grass dense and vigorous ar 77 Conversion of forest to USA T NG Grassland Uatershed No significant difference in total discharge Comparison to pre-conversion f grassland when grass production high. Decrease in parameters production increased discharge 12 cm/yr 0 over predicted forest discharge Table 8.2 IMPACTS Of SOIL CONSERVATION TECHNOLOGIES 011 SOIL MOISTURE AND SURFACE RUMOFF CITE TECNMOLOGYMTREATMENT LOCATION CLIMATE SLOPE LAND USE SCALE IMPACT REMARKS MM-aft.sX.Musawasaaawas .... au.s .auz ......a x.%XDazz;Xaaa ...==S===== ..===Mxu ... =t,S=.= ..===M==MMsZaut .. ...===;s=S,,= === = ====;;Sts====X = ======= 134 Conversion of forest to Hawaii CU I MG Grassland Uatershed Increase in runtoff rates Attributed to decreased rate grassland of evapotranspiration by grass relative to natural rainforest cover 20 Bare soil Korea T 272 Forestry Watershed 132X increase in surface runoff Comparison to grassed; Hardeood plantation 36X decrease in surface runoff soil = sandy ctay loam; Coniferous plantation 422 decrease in surface runoff Pt = 1,370 uu/yr 176 Fern cover Sri Lanka EM 50 Forestry Watershed 110X increase in surface runoff Conparison to closed canopy Degraded grass cover 50X 552 increase in surface runoff Pinus caribea ptantation; Grass cover SO No significant difference in surface runoff Soils - sandy ctay toam Nitosol (Typic Tropudult), Pt a 5,640 me/yr 176 Fern cover Sri Lanka EN 502 Forestry Watershed 2052 increase in average peak discharge Comparison to ctosed canopy Degraded grass cover 502 2202 increase in average peak discharge Pinus caribea plantation Grass cover 502 582 increase in average peak discharge Soils a sandy cLay toam Nitosot (Typic tropudult), Pt = 5,640 mn/yr 62 Clearcut Chite I/D 302 Forestry Plot 732 increase in surface runoff Comparison to 30 yr. old Pinus Pasture 302 Pasture 522 increase in surface runoff plantation; soil = ctay toam; 6 yr. old Pinus plantation 302 Forestry 142 increase in surface runoff Pt = 2,000 met/yr 182 Grass cover Colombia lID 222 Pasture Plot 702 decrease in surface runoff Coarison to monthly tilled bare soil. 33 Grass cover NSW, Austratia SATr 82 Grassland Plot 812 decrease in surface runoff Coirparison to wheat; so i = Grass cover 8.52 Grassland Plot 752 decrease in surface runoff calcic and chromic luvisots, D respectively; Pt * 644 and " 561 on/yr, respectively 202 Grass cover (Peimisetun, India D/UM 52 Agriculture Field No significant difference in surface runoff Comparison to conventional Cynodon,Urochloa,Panicum) cuttivation of gorapaddy,urid corn.peanut; soil z sandy clay, 161 Grass - Imperata,Saccharum PhiLippines D/IM 36-70X Grassland Field No significant difference in surface runoff Comparison to secondary forest; Plantation forest - Forestry No significant difference in surface runoff Soit = cLay loam, Pt = 2,600on/yr Gtiricidia,Leucaena Kaingin Agriculture 1652 increase in surface runoff 160 mutch/Cover crop Pan-tropical N/A N/A Agriculture N/A Protection fron compaction and Luvisols, Nitosols, Acrisols decreased infiltration (Ultisols and Alfisols) with sandy topsoils susceptible to compaction and resuitingly lower soit moisture leveLs following cultivation. 97 Mulch Nigeria WID 12 Corn Plot 694 decrease in surface runoff Comparison to unmuLched forest cover 12 Forest 732 decrease in surface runoff Hutch sX Corn 81X decrease In surface runoff Forest cover 52 Forest 97X decrease in surface runoff Table 8. 2 IMPACTS OF SOIL CONSERVATION TECHNOLOGIES ON SOIL MOISTURE AND SURFACE RUNOFF CITE TECHNOLOGY\TREATKENT LOCATION CLIMATE SLOPE LAND USE SCALE IMPACT RENARKS Mulch 10X Corn 87X decrease in surface runoff Forest cover 101 Forest 96X decrease in surface runoff Hulch 15X Corn 89X decrease in surface runoff Forest cover 15X Forest 89X decrease in surface runoff 102 Desmodiui app. cover Taiwan D6/14 46X Litchi Field No significant difference in surface runoff Comparison to clean cultivation; Bahia grass 46X 98X decrease in surface runoff Pt = 3,494 mm/yr Eragrostis app, mutch 46X 691 decrease in surface runoff 102 Eragrostis barrier/mulch Taiwan D/WM 231 Banana Field 301 decrease in surface runoff Comparison to clean cultivation; Guinea grass barrier/mutch 23X 251 decrease in surface runoff Pt =2,348 m/yr South African pigeon grass 23X Mo significant difference in surface runoff barrier/sutch 3 Grass strips (O.Sm wide - Java EM 15-221 Agriculture Field 9X decrease in surface runoff Comparison to bare soil Brachiaria, 1m wide - Paspal up) 183 Vegetative strip (Mucuna Indonesia D/W4 121 Corn & cassava Plot No significant difference in runoff Comparison to corn & cassave w/o utilis) vegetative strips Veetative strip (Pucraica 14X decrease in surface runoff phaseoloide) Vegetative strip (mimosa 101 decrease in surface runoff invisa) Veetative strip (Peamisetum 321 Increase in surface runoff purpureum) 160 Hulch Pan-tropical N/A N/A Agriculture N/A Decrease soil moisture losses from Ferrasots (Oxisols), Andosols evaporation (Andepts). and oxidic soits havel a low range of available moisture (gravity draired 2 0.1 bar) 96 Mulch Nigeria U/D N/A Agriculture N/A Increased soil water, decreased runoff, Best management practice for decreased evaporation rates Luvisols, Podzoluvisots (Alfisols) 197 Hulch <1.1 tCha) USA r MG Sorghum fallow Field 38X increase in soil moisture storage Comparison to umuilched Mulch (2.2 t/ha) HG Field 39X increase in soil moisture storage Mulch (4.4 t/ha) MG Field 61X increase in soil moisture storage Mutch (8.8 t/ha) MG Field 931 increase in soil moisture storage Hulch (13.2 t/ha) MG Field 104X increase in soil moisture storage 180 Mulch (12 t/ha) USA SATap NG Agriculture Field 1041 increase in soil moisture storage Comparison to no mutch 171 Mulch (0.33 t/ha, 22X cover) USA T NG Agriculture Plot No significant difference in surface runoff Comparison to no mutch under Mtulch (0.44 t/ha, 371 cover) No significant difference in surface runoff rainfall simulator Hulch (0.66 t/ha, 461 cover) No significant difference in surface runoff Mulch (0.88 t/ha, 58X cover) No significant difference in surface runoff Mulch (1.21 t/ha. 811 cover) Ho significant difference in surface runoff Mulch (1.87 t/ha, 87X cover) go significant difference in surface runoff Hulch (2.47 t/ha, 92X cover) No significant difference in surface runoff 113 mulch (0.63 t/ha) USA T 5X Agriculture Plot 11X decrease in surface runoff Comparison to no mutch; soil = Table 8.2 IMPACTS OF SOIL CONSERVATION TECIINOLOGIES ON SOIL MOISTURE AND SURFACE- RUNOFF CITE TECIINOLOGY\TREATMENT LOCATIONI CLIMATE SLOPE LAND USE SCALE IMPACT REMARKS ..... a .....=.as,=,5===,==X= ... ... ..... ... =====S,g,=====S-====-=g==============lt=========W=== ==== .... -.= Mulch (1.23 t/ha) 5S 44X decrease in surface runoff silt loam; rainfatt simulator Mulch (2.47 t/hs) 52 89. decrease in surface runoff study Mutch (4.94 t/ha) 5S 96X decrease in surface runoff Mulch (9.88 t/ha) 5 100X decrease in surface runoff 38 Mutch (5 cm rice straw) India D/ a 12 Wheat/barley/ Fietd 32 increase in soil moisture storage Cotparison to uromutched gram/, inseed 192 Mutch Indonesia D/ l NG Corn Plot 862 - 912 decrease in surface runoff Comparison to urmulched corn 189 Nutch (9 t/ha) Indonesia I/D MG Soybean Plot Conserved equivalent of 23 days of plant's Comparison to unmutched water requirements throughout growing season 199 Mutch (0.6 t/ha) Indonesia D/ 1 72 Uptand rice Plot 502 decrease in surface runoff Comparison to urnnutched (rice Mutch (1.0 t/ha) 69° decrease In surface runoff straw); soil = Latosol hutch (1.6 t/ha) 832 decrease in surface runoff Mutch (0.6 t/ha) 92 242 decrease in surface runoff lutch (1.0 t/ha) 692 decrease in surface runoff hutch (1.6 t/ha) 862 decrease in surface runoff Mutch (0.6 t/ha) 142 262 decrease in surface runoff Mutch (1.0 t/ha) 482 decrease in surface runoff Mutch (1.6 t/ha) 832 decrease In surface runoff 188 Mulch (0.6 t/ha) Indonesia D/14N 7-142 Uptand rice Plot 332 decrease in surface runoff Comparison to umutched (rice Mutch (1.0 t/ha) 592 decrease in surface runoff straw); soit = latosol Mutch 01.6 t/ha) 842 decrease In surface runoff 213 Mixed cropping A nutch Indonesia D/11I 202 Corn & peanuts Plot 432 decrease in surface runoff Comparison to corn & peanuts - (4.4 t/ha) strip cropped 4 Mixed cropping No significant difference in surface runoff 106 Multicropping Indonesia D/ 1 202 Ctove/banana/ Ptot 992 decrease In surface runoff Comparison to corn monculture cassava/ungor taut/coconut 9 Corn L teak Intercrop Indonesia SATr NG Agricutture Plot 2X - 17X decrease in surface runoff Comparison to corn monoculture; Pt - 715 - 848 snayr 173 Natural cover India SATr 0.52 Naturat cover Field 99.92 decrease In surface runoff Comparison to cuttivated fallow; Naturat cover 1X Natural cover 99.9X decrease in surface runoff soil = silty clay loam, Grass cover (Cynodon spp.) 0.52 Grass 72X decrease in surface runoff Pt = 800 mn/yr Grass cover (Cynodon spp.) 12 Grass 562 decrease in surface runoff Peanut 0.52 Peanut 69x decrease in surface runoff Peanut 1X Peanut 522 decrease In surface runoff Gram 0.52 Gram 552 decrease in surface runoff Gram 12 Gram 252 decrease in surface runoff Jowar 0.5x Jowar 402 decrease In surface runoff Jowar 12 Jowar 182 decrease In surface runoff 208 Agroforestry (intercropping) Thailand D/WN MG Gmetina arborea Field 8402 increase In surface runoff Comparison to GOetina arborea and rice cropped alone 63 Coffee - 1 yr old Indonesia D/IN 46-o66 Agricutture Plot 1352 increase in surface runoff Coqparison to undisturbed natural Table 8.2 IMPACTS OF SOIL CONSERVATION TECHNOLOGIES ON SOIL IIOISTURE AIID SIAFACE RUNOFF CITE TECHNOOGfTREATNENIT LOCATION CLIMATE SLOPE LAND USE SCALE IMPACT REMIRKS Coffee - 3 yr old 3485 Increase in surface runoff forest. Coffee 16 yr old 726X Increase in surface runoff 8 Tobacco Indonesia D/Wi 30X Agriculture Plot 301X increase in surface runoff Comparison to ureeded cassava; Potato 8891 increase in surface runoff soil amndosol Corn + beas 369X increase in surface runoff 2 anure (16 t/ha) Indonesia D/Wi 16X Agriculture Plot 451 decrease in surface runoff Comparison to ursanured; soit - votcanic, dusty regasoli 21 Cow dung (5 /ha) iGhana Cm 7.51 Corn Plot 901 decrease in surface runoff Comparison to bare fallow; Wood shavings (5 K/ha) 7.51 Plot 921 decrease in surface runoff Pt a 1,340 usyr Poultry wmnre (5 t/ha) 7.51 Plot 931 decrease in surface runoff Cow/poultry manure (5 t/he) 7.51 Plot 911 decrease In surface runoff Comercial fertilizer 7.51 Plot 591 decrease in surface runoff 21 Cow dung (2 t/ha) Ghana SATr 21 Sorghum Plot 641 decrease in surface runoff Cowparison to bare fatlow; Cow "ua (4 t/ha) 2X Plot 771 decrease in surface runoff Pt a 320 ur/yr Cow dung (4 t/ha) * straw 2X Plot 891 decrease in surface runoff mulch (4 t/ha) Cow dung (5 t/ha) 21 Plot 821 decrease in surface runoff Cow dung (5 M/he) & straw 21 Plot 911 decrease in surface runoff eulch (4 t/ha) 190 Mulch and mini . tillage Indonesia EN 3.5X Cassava/rice/ Plot 75X decrease in surface runoff Comparison to urnulched. Bare soil 3.51 corn/pearats/ Plot 701 increase in surface runoff traditional cultivation; hulch and minim tillage 9X beans Plot 641 decrease in surface runoff in order of treatment pairs v. Bare soil 9X Plot 431 increase in surface runoff soils a mlitosol (Tropudult). Iulch and minim tillage 101 Plot Ho significant difference in surface runoff Nitosol (Tropudalf)I Bare soil 101 Plot No significant difference in surface runoff Nitosol (Tropudatf). Mulch and minim. tillage 141 Plot 631 decrease in surface runoff Ferralsol (Haplorthox). Bare solt 141 Plot No significant difference in surface runoff Nitosol (Tropudalf). Mulch and minia* titlage 14X Plot 92X decrease in surface runoff Bare soil 141 Plot No significant difference in surface runoff 180 Stubble mulch tillage USA SATcp MG Uheat Field 14am increase in soil moisture storage Comparison to clean cultivation, Clean cultivation/fallow Field 35mm increase in soil moisture storage continuous vheat 116 No-till Thailand D/Wi MG Upland rice Plot 671 decrease in surface runoff Comparison to clean cultivation Studble mulch MG 421 decrease in surface runoff Bare fallow NG 2081 increase in surface runoff 116 Disk plow/stubble mulch Thailand D/Wi MG Upland ries Plot 841 decrease in surface runoff Comparison to disk plowed and lo-tiIl/mulch HG Plot 8X decrease in surface runoff no mulch 116 Disk plow/stubdble mech Thailand D/Wi HG Peanuts Plot 381 decrease in surface runoff Comparison to disk plowed and No-till/mulch NG Plot 50X decrease in surface runoff no mulch 212 No-till with stubble mulch USA T 31 Corn Field 161 increase in surface runoff Comparison to conventional (chemical wed control) tillage with stubble mulch 84 Furrow dike with stubble USA T HG Wheat-sorghum- Field 25 - 30 m/yr decrease in surface runoff Soil - clay loos Xerosol tulch fallow (Torrertic Paleustolt). Table 8.2 IMPACTS OF SOIL CONSERVATIOSI TECHNOLOGIES ON SOIL MOISTURE AND SURFACE RUNOFF CITE TECHNNOOG1YTREATMENT LOCATION CLIMATE SLOPE LAND USE SCALE IMPACT REMARKS Pt = 1,270 s/yr 54 Rows up 6 down slope Peru SATr 25X Potato Field 71X increase in surface runoff Comparison to contirnous fallow Contoured 25X 8X decrease in surface runoff Cover cropped (Lupinus) and 25X 44X increase in surface runoff rows up & down slope Mulched 25X 16t decrease in surface runoff 54 Rows up 6 down slope/burned Peru W7/ 30X Potato Field 19K increase in surface runoff Coqnprison to continuous fatlow Cover cropped and rows up 30X 6X increase in surface runoff a down Contoured 30X 14X increase in surface runoff Mulch and rows up & down 30X 15K increase in surface runoff stope 122 Bare fallow Ghana UW/ 7.5X Corn Field 214X Increase in surface runoff Conparison to traditional mixed Zero-tillage 7.5X 51X decrease in surface runoff cropping; soil = sandy toam to Nulching 7.5X 85K decrease in surface runoff sandy clay. Pt = 1,500 ms/yr Ridging across slope 7.5X 47K decrease in surface runoff Minimu tilitage 7.5X 37X decrease in surface runoff 122 Bare fallow Ghana U/D 3K Corn Field 722X increase in surface runoff Comparison to traditional mixed Zero-tillage 3X 18K decrease in surface runoff cropping; soil = sandy clay toam Mutching 3X 82K decrease in surface runoff to sandy clay, Pt = 1,400 am/yr Ridging across slope 3X 31X decrease in surface runoff Minima tillege 3X 7n decrease in surface runoff 66 Cultivation Brazil cU NG Agriculture Field 50K decrease in percentage of large soil Soil * Ferrasol, Acrisols aggregates, pore clogging Nitosols (Oxisols & Ultisols) o 133 Cuttivation Brazil cU NG Agriculture Field 85K decrease in infiltration capacity Resutt of 15 years cropping; attributed to coaetion by mchinery and illuviation of clay soil a Ferrasol (Oxisol) 157 No-till Nigeria WU/ 1K Agriculture Field 86K decrease in surface runoff Coqparison to conventional Bare fallow 1X Field 127X increase in surface runoff plowing No-till 5X Field 80K decrease in surface runoff Bare fallow SX Field 130K increase in surface runoff No-till 10X Field 77K decrease in surface runoff Bare fattow 10X Field 90X increase in surface runoff No-till 15K Field 76K decrease in surface runoff Bare fallow 15X Field 134X increase In surface runoff 73 Plow,clean till,contour USA T 6K Agriculture Field 48K decrease in surface runoff Comparison to plowuclean till, No-till,contour 21K Field 43K decrease In surface runoff and sloping rows 90 Burn to clear/no-till Japan UW/ 12-18K Agriculture Field 31K decrease in surface runoff Coeparlson to burn to clear Bulldoze to 4K slope & till 3-5S 47n decrease in suface runoff I tilled; soil * sandy clay loam; Pt * 2200 asyr 16 Contour cultivation India U/D 2.2K Barley/jowr Plot 29K decrease in suwface runoff Cualrison to up & down slope cultivation Table 8.2 IMPACTS OF SOIL CONSERVATION TECHNOLOGIES ON SOIL MOISTURE AtlD SURFACE RUNOFF CITE TECHNOLOGY\TREATMENT LOCATION CLIMATE SLOPE LAND USE SCALE IMPACT REMARKS 46 Contour rows India D/A NG Agriculture Field 25S decrease in surface runoff Coamparison to up & down slope cultivation; based on review of 30 years of experiment station projects 28 Plough USA T NG Agriculture Plot 858X increase In total water infiltrated Coaparison to urtitted; Plough-disk-harrow 250X increase in total water infiltrated attributed to swface roughness, Cultivate 279X Increase In total water infiltrated not to increased permeability Rotovate 722 increase in total water infiltrated 158 Animal traction France T 5X Winter barley Plot 99.92 decrease in surface ruroff Cofparison to perwmnent grassland Broken grassland 52 Grassland Plot 350X increase in surface runoff Rainfall simulator study - 1 hr. tilt,stubble plow,subsoil 5 Agriculture Plot 150X increase in surface ruroff 33m rainfall itll,stubble plow 52 Agriculture Plot 6002 increase in swface runoff Heavy equipnent. post harvest SX Potato/endive Plot 1,317X increase in surface runoff 178 Ridge-tillage USA N HG Agriculture Field May increase surface runoff Increase in urface runoff due to concentration of flow in the furrowa wnd ridges inreasing sIpe. 24 Contour furrowing USA SAImp HG Rangeland Field 8X increase in soil moisture storage in top Generatly, soils with mediL. to 75 Co of soil fine texture show mot consistent beneficial response to furrowing 78 Furrowing USA SATrp NG Rangeland Field Little benefit as soil moisture conservation tool on clayey soils if compaction of soil surface occurs %0 200 Furrowing USA SATmp HG Rangeland Field Ineffective as soil moisture conservation tool in sandy soils 47 Furrowing USA SATmp NG Rangeland Field 1211 Increase in soil moistuwe storage Cariason to non-furrowed; soil * clay lom 164 Furrowing USA 1 IIC Crassland Field Rainfall pereolated 6-16 inches deeper Comparison to non-furrowed 218 Furrowing Australia SATr HG Grassland Field Water storage upto 1 meter deeper Coaparison to non-furrowed 142 8roadbed & furrow India SAlr 0.42 Pearl millet/ Field 61X decrease in surface runoff. Coparison to traditionat flat sorghum 56X decrease in peak discharge cropping system with amnsoon 8roadbed & furrow 0.62 intercrop with Field 48X decrease in surface runoff, fattow; soilt a very fin et ay Cajanus cajan 31X decrease in peak discharge Vertisol (Typic Petlustert; Broaubed & furrow 0.62 Field 66X decrease In surface ruioff Pt a 760/yr field bumds S6X decrease in peak discharge Flat on grade 06X Field 365 decrease In surface runoff, no significant difference In peak discharg 142 Broaded & furrow India SATr 0.62 Pearl sitlet/ Field 382 increase In urface rnoff. Caarlean to traditional 672 increas in pea* dischrge crpping system jath field Flat an grade with graded 0.6a sorghu Field 23X decrease In surface runoff. bunds; soil - fine clay bunds intercrop with no significant differnce in peak dischge CUdic RhodouBtalf). Table 8. 2 IMPACTS OF SOIL CONSERVATION TECHNOLOGIES ON SOIL MOISTURE AND SURFACE RUNOFF CITE TECHNOLOGY\TREATMENT LOCATION CLIMATE SLOPE LAND USE SCALE IMPACI REMUS *uuxu=-== ..... noun .....== .. X=====ss== === =======sa=====aS .... == g====nou .... ......=U$sssssss=Stu==s==,,ssssssass _- Traditionat with contour NG Cajanus cajan Field 37X decrease In surface runoff, Alfisols unstable in bods L bunds 33X decrease in peak discharge furrows. Pt a 760yr 130 Broacbed and furrow India SATr NG Agriculture Field 50 decrease in surface runoff CoWarison to traditional flat land uith unds 11 Cross-slope planting India D/U 1.5X Grains Plot 21X decrease in surface runoff Cooparison to up & dounslope Ridge & furrowv60 cm spacing) 86X decrease in surface runoff cultivation; Soil * well-drained. sandy loam (inceptisol) 180 Furrow dams USA SATmp 0-5 sorghum Field 35-88mm of surface runoff conserved Scm storage capacity before overtop; soil * clay loam 72 Bench terrace Indonesia D/A 10-47K Agriculture Plot 50K decrease in surface runoff Comparison to unimproved local Afforestation 70X decrease in surface runoff practices; Pt * 2.000-3,350 mayr 167 Terracing & reforestation Indonesia EN HG Agriculture Uatershed 50K decrease in runoff coefficients Comparison to pre-treatment levels 177 Level terrace USA T 2-18K Corn Field No significant difference in surface runoff CaoarIson to contour ptanted Pasture 2-18K Pasture 37K decrease In surface runoff corn; soil deep ioess Pt - 10X decrease in peak runoff on both 815 am/yr. 102 Bench terrace Taiwan D/W1 28K Citrus Field 74K decrease in surface runoff Coqarison to clean cultivation;,_. Mulch/Sahia grass strips 28X 92K decrease in surface runoff Pt a 1,634 Vyr t0 Hulch/Bahia grass cover 28X NG Go 102 Level terrace Taiwan D/W1 24K Banana Field 77K decrease in surface runoff Coeparison to clean cultivation, Grass barrier (2.5m spacing) 24K 67K decrease In surface runoff Pt - 2,274 miyr Grass barrier (Sm spacing) 24X 29K decrease in surface runoff Mulch/Bahia grass cover 24K 87X decrease in surface runoff 103 Bench terraces Taiwan D/WM 28K Citrus orchard Field 75K decrease in surface runoff CaWparlson to clean cuttivation; Bench terraces with grass 28K 90K decrease in surface runoff soil * clay loa cover (Bahia,Love) or mutch 103 Bench terraces with Bahia Taiwan D/Ul 28K Citrus orchard Field 48X decrease in surface runoff Comparison to bench terraces and Hillside ditches with Bahia 28K 75K decrease in surface runoff hillside ditches without Bahia 102 Reverse slope bench terrace Taiwan D/lfl 20K Pineapple Field 86K decrease in surface runoff Cooparison to planting up I Mulch & close planted on 20X 82X decrease in surface runoff dowe slope; Pt a 1i373 Jyr the contour 68 Contour rows India D/01 25K Potato Field 70K decrease in surface runoff Comparison to up & dam slope LUp & down stope planting/ 10X decrease In surface runoff cultivatIon contour furrows Permanent grass 70K decrease in urface runoff Bench terrace 50- decrease in surface runoff 10 Terraces with increase in USA T 2-3X Agriculture Watershed 24K decrease in surface runoff 28 yr study; Soil * calcarem permanent grass/3 yr. crop Clay. Pt a 839 myr rotation/deep tiltage Table 8.2 IMPACTS OF SOIL CONSERVATION TECHNOLOGIES ON SOIL MOISTURE AND SURFACE RUNOFf CITE TECNNOLOGYITREATMENT LOCATION CLIIIATE SLOPE LAND USE SCALE IMPACT RENARKS .........=== ==== == .a.. ==== =*=====a====*==....==zS =======- ... .....== ==--= ...-==- Zanum =-S=-g.a===-S*== =--=.S= ......ACg=-==g4-g-- Terraces with shallow tillage 2-3X 20X increase in surface runoff and grazing of crop residues Terraces with no inproved 2-3X No difference in surface runoff practices 11 Terraces with conservation USA I NG Agriculture Watershed 20 - 33X decrease in surface runoff 200 ha watershed; Soils = black cropping & permanent cover Solis 74 Contour-strip on terraces USA T MG Agriculture Watershed 30X decrease in anmual surface runoff Cofiparison to pre-treatment Terraces GC Watershed No significant difference in annuat surface runoff 219 Contour farmed terraces USA T NG Agriculture Watershed 9 - 37X decrease in surface runoff General statement for USA 84 Graded furrows vs contour USA T NG Wheat-sorghum- Field 25X increase in surface runoff from graded Greatest increases In siall storm tilled graded terraces fallow terraces relative to graded furrows events; Soil catsy toma Xerso5t (Torrertic Paetastott); Pt - 1,270 &Wyr 155 Graded furrows vs terraces USA T 2-3X Cottonxsorghun/ Watershed 21X decrease in surface runoff from graded Graded furrow system htd up oats furrows relative to terraces. to storm that caused extensive terrace faiture; Soit a black clay, Pt - 853 tyr 112 Land leveling USA I NG Sorghum Fiteld No significant difference in soil moisture CoWerisen to unteveted; storage Pt * 673m/yr 187 Contour bunds Indonesia D/WI 12X Agriculture Plot 18X decrease in surface runoff Carlson to unuded 204 Field bunds Indonesia EM NG Irrigated rice Field Water efficiency of rice 25-30K, attributed Results tally with other studies! to farers maintaining water heed too high range of efficiencies recorded - thus increasing lateral flou into bunds 18K (sads) - 45K (cay) In 10 awd water lost to vertical percoiation. countries. 126 Level pans USA T NC Agriculture Field 12.7cm - 22.9cm increase in soil moisture Increase dependent on timing nd storage. distribution of rainfatt 126 Level pans USA SATnp NG Sorghum Field 7 month moisture storage quivalent to Pt * 422 ma/yr 19 - 21 month fallow 182 Terrace Colombia W/D 45K Coffee Plot 116X increase in surface runoff Coqerison to unterraced coffee. 10 Terraces USA T 2-3K Agriculture Watershed Reduction In peak runoff - mgnitude of 28 yr study; Soil c lcareous reduction Inversely proportional to size of clay. Pt a 889 mmVyr watershed 10 Terraces USA T 2-3K Agriculture watershed Wet soils - Increase in surface runoff, 28 yr study; Soil * calcareous dry soils - decrease in surface runoff clay, Pt * 889 _/yr 191 Bench terrace Indonesia 0/WI 9K Corn & cassava Plot 44X - 65K decrease in surface runoff Comparison to unterraced Bench terrace Sweet potato 50K decrease in surface runoff Bench terrace Corn 31X decrease In surface runoff Bench terrace Cassava 52K decrease in surface runoff Table 8.2 IMPACTS OF SOIL CONSERVATION TECHNOLOGIES ON SOIL MOISTURE AND SURFACE RUNOFF CITE TECHNOLOGY\TREATMENT LOCATION CLIMATE SLOPE LAND USE SCALE IMPACT REMARKS 140 Ftat bench terrace Indonesia D/ M NG Agriculture Plot 171 increase in surface ruinoff Coaparison to ridge terrace Sloping bench terrace 293X increase in surface runoff 163 Bench terrace Indonesia D/I N IG Agriculture Plot 31X increase-15X decrease In surface runoff Courparison to uniaproved locat Ridge terrace 80X - 201X increase in surface runoff farming systm. 76 Zingg conservation bench USA T NG Agriculture Field No significant difference In soil moisture Coaparison to unterraced. terrace storage 17 Zingg conservation bench USA T 1.51 Sorghuw/wheat/ Field Excess soil moisture made tillage operations Soil a silty clay terrace fallow difficult. 124 Zingg conservation bench USA SATrp 1X Sorghum field 7 month soil moisture storage in terrace Soil a silt lom to clay lo"; terrace equivalent to 19 month storage in fellow Pt a 424Wyr 75 Zingg conservation bench USA SATep 1.51 Sorghuu/uheat field Soil moisture storage the same continiously Soil a silty clay loam, terrace cropped as level terrace fallowed 11 onths 75 Zingg conservation bench USA SATmp 1.51 Sorghum/uheat field No soil moisture conservation benfits in terrace coarse or low water-holding capacity soils 69 Zingg conservation bench USA SATrp 1-5X WHeat/corn/ Field No increase in overwinter soil moisture Soil = silt toam to silty clay terrace bromegrass A storage, increase occurred in spring. oam. Pt a 445 mayr alfalfa 169 Contour bunds Thailand D/WI NG Agriculture field 701 Increase in surface runoff Comparison to unterraced control' Bench terrace NG 1401 increase in surface runoff soil a Acrisol (Typic Peleudult); Hillside ditches NG 701 increase in surface runoff Pt a 1.612 m/yr. Increased Intermittent terrace NG 601 increase in surface runoff runoff undesirable. 68 Graded vs Bench terraces India D/WI 25X Agriculture Field No significant difference in surface runoff between these two terrace types 68 Contour bunds/ridge-type India SATr & NG Agriculture Field Ponding due to insufficient drainage Soil = Vertisots and other terraces/tevel terraces/ AT poorly drained soil types absorptive terraces 168 Contour bunzding/bench India SATr & NG Agriculture Field 50X decrease in depth to which water Comparison to adjacent untreated terracing/land levelling CW/N percolated lands 103 Bench terraces vs hillside Taiwan D/WI 281 Citrus orchard Field Runoff less from bench terraces relative If bench terrace topsoil not ditches to hillside ditches replaced, percolation rates are low and runoff rates are high; 4 years were necessary to consolidate soil on bench terrace so that less than from hillside ditches 89 Terraces - 421 watershed area USA T NG Agriculture Uatershed 11X decrease in surface runoff. 361 decrease 11,500 ha watershed in total discharge from 10 yr. storm Terraces - 751 watershed area NGC 44 decrease in surface runoff, 601 decrease in total discharge from 10 yr stors Table 8.2 IMPACTS OF SOIL CONSERVATION TECHIOLOGIES ON SOIL MOISTURE AND SURFACE RUNOFF CITE TECHNOLOGY\TREATNENT LOCATION CLIUATE SLOPE LAUD USE SCALE IMPACT REKARKS 108 Large-scale terracing Switzerland T NG Vineyards Jatershed Significant increase in peak discharges, Effects attributed to: increase 110 Large-scale terracing Switzerland T NG Vineyards Watershed time to peak discharge, steeper slope of of asphalted road anS drainage dry weather hydrograph recession curve, systen (gulliea & pipes) and reduction in base and low ftows greater destruction of soil structure by variability in runoff. Decrease in soil's mechanicat action of heavy hydraulic conductivity, decrease in soil equipnt In terrace building. moisture. Soit * deep loss. 58 Check dams to harvest water Israel SATP NkG NG Watershed lprwoved output from dowstra shallow Effective where installed in aquifers series alon sam river, hydrologic/hVdrogeologic inputs are knoim. puWs md energy available. and technical staff to operate nd mintain 206 Dirt roads USA T 0.2-30X Dirt roads Plot 24X - 96Z of rainfall occurs as runoff Range of runoff values fram rainfalt simulator studV. N IMPCTS OF SOIL COuSERVATIOU NECNOCIES ON EISION AM SEDIUNIATION CIIE TECIUOGY/TItREATMENT LOCATION CLIMATE SLOPE LAW USE SCALE IMPACT CGWMISON ISMA 145 Forestry Pan-tropical Various 45.1 Forestry Watershed Mini-ite soil erosion N/A Most intenive nnUIUft Agroforestry 35-601 Agroforestry Watershed Miniitze soll erosion cptl at th givn c Pasture 10-45X Pasture Watershed Ninimize soil erosion of a ats t Inidge silt Bench terraces & intensive 10-3SX Agriculture/ Watershed Minimfie soil erosion of Io e to i zs crop mngement Horticultur *ero on,the shloW the Bench terraces 10-25X Agriculture/ Watershed Minimize solt erosion nd of the rage. Hort icul ture 215 RemOve trees Java EM 9X 5 year old. Plot 200 increas in edimnt yield I/A Campd to undisturbad; Remove uiergrowth 9X Acaei app. Plot 4001 increae in sedimnt yield soit * clay. Gletsot Iteve trees/undergrowth 9X plantation Plot 1001 increase in sedimnt yield (CTpic Tropmqapt) Remo uidergrouthllitter 9n Plot 26 0o00 Increase In sedimnt yield Rem" alI 91 Plot 4,1001 increas in sedimnt yield 176 Fern cover Sri Lanka EN 501 Forestry Watershed 5M66 Incrs In sedimnt yeltd OS.6 tlhWyr Cqrwson to ctesed campy Degraded grass cover 1 171 increse in sedimnt yield Plan cribm plentaticn Cress cover SOX No significant difference in sdimnt yield 100 Land clearing Ivory Coast W/o MG Agriculture Field Large increase in erosion rate. N/A Soil e sd* Al isot 93 Lnd clearing Nigeria W/b NG Agriculture Field Large incree In erosion rate I/A Soil * sandy Aifisot 97 Plowing Nigeria W/o NC Agriculture Field Erosion rate upto t15 t/ha/yr I/A Soll * sandy Alfisol 58 Contour cultivation Israel SAIr 3-81 Agriculture Field Sltes ltes than 3X ne difference fre I/AI cross-slop cultivation, slpes greater than SX ha washouts nd incresed orosien 86 Cuttivated faltow India SAIr MG Agriculture Feled uestionabbe practice due to erosion hazard. N/A Sol * heay Soll *.g. Vertisol Bunds NG Erosion between bunds way be stebatantiai, and Inadqute mintennce often results In breche 160 Traditional cultivation Amzon Cm Steep Agriculture Field No Indications of Incred erosion N/A Solt * Ultisol 209 Shifting cultivation Venezuela W/o Steep Agriculture field Little erion igst traditinal N/A cultivators svere erosion mPnst rwoers to shifting cultivation 70 Conversion of forest to World-wide All N/A Grassland Watershed No significant difference In areroVi / aems wsigpoeto grassland borld-t tde at HtA Cresst nd h sediment yield nc gras establihd with conv lesin to grmlen 134 Conversion of forest to Hawaii Cl MG Grassland Watershed Increased erosion/sedimentation N/A Attributed to incrsd rmff grassland dw5 to reduced evapotranspirat ion br Aioon on rainfest solla 162 Natural erosion Hawaii CU Steep Natural cover Watershed Landtids served onty tu r fogest wd IN/A NW" ValI ey fern cover. not grass CrA S Table 8.3 IMPACTS OF SOIL CDYSERVATICY TECHNOLOGIES ON EROSION AND SEDIMENTATION CIFE TEC O0011RtErAFENT LOCATION CLIMATE SLfPE LAND USE SCALE IMPACT CPARJISN BEiS 147 Natural erosion New Guinea EN Il/A laturst cover Watershed Dominant sediment aowcos are ladslides /A Atl rareoff a 5,3O0-7.2001yr and slopewsoh uwder forest cover 30 Forest. litter renoved/burned nAdonesia D/ l 10X Various Plot 1,4001 increase In erosion rate 5 t/ha/yr Comriaon to zmdlldturbsd forest; 40 yr. old teak plantation 1,400Z increase In erosion rate soil a deep volcanic soil Ptowed. reisnfed mize 11900X increse In erosion rote Fired. dibbled rainfed mize 1.300S Increae in erosion rate Bnch terraced, rainted 3001 inrcrase in erosion rate Mize/cassava Coffee plantation Ho significent difference in erion rote Dense Irperatu grasstand No significant differenc, In creasin rate Degraded forest, shrub, dense No significant difference In erosion rate uidergrouth 176 Fern cover Sri Lanka EN SO Forestry Watershed 901 of sedimnt yield In 301 rainfalt events M/A Soila * sad clay tal MItosol Degraded grass cover s50 Forestry 901 of sedbient yietd in 301 rainfalt events (Typic IropudLit). Grass cover s50 Forestry 9OS of sadimlnt yield in 10D raInfatl events Pt * 5.460 onyr Closed cnpy Pinus s50 Foretry 90o of sedismnt yield in 105 rainfall eveta plantation 4U Forest momi CY MG FYorest Watershed 9o of sediment yield In 2 rainfoll evnts B/A 29 Mile River basin Egypt SAlr H/A Mixed Watershed 1001 of sediment from 10X of basin H/A Red Deer River basin Canand T H/A Nixed Watershed 901 of sediment from 101 of basin Amazon River basin Brazil Cl H/A Mix ed Watershed W0S of sedimnt at the mouth from the Andes Yellow River basin China D/WN H/A Mixed Watershed 901 of edmlent from 401 of basin united States USA T I/A Nixed Watershed 1.1 t1/myr edient frm river batns Spring Creek Canada T N/A Nixed Watershed 36o increse in sdiment yiltd following disturbwice of 0.451 of the watershed Smtl watershed USA T I/A IG Watershed 51S of sedimnt from 1S of watesthed 32 lRods and traits Honduras w/I usG Mixed Watershed 451 of sediment produced, 21 of ra h/A Niltside cuttivation Agriculture 201 of sediment prodicd, IS3 of are Grazing lad Pasture 201 of sedimnt proAed. 20 of oea Burned foreat land Forestry 41 of sediment pred1 of ea orush lands Mixed 1S of sediment pred 20 of area forests Forestry 1S of edimnt produced. 3S5 of ae 49 Roat ard trails Kenya V/0 N/A Mixed Watershed Large frwction of sedient leaving B/A Bad on ediment yield analysis *asricultural catchent contributed by fr te 1 Łazinentew roads e d trailts 206 Dirt road USA 1 H/A Mixed Plot 3 - 60 t/ha/yr sediment yield I/A Sig of so Il tlo value reported In literature. Soil loss us d_psbnt. 14 Soit lose toternmes Brazil CV I/A AgriCutture Fleld tlthotal * 2-7 t/ha/yr IVA Soil tos tolerance is that Podeot * 5-13 ta/yr qmtity of Solt that may bt Dark red latotoo * 12-16 t/hyr erodd without a sites ability to auttsn protuctivity being affected. Table 8.3 IMPACTS OF SOIL CONSERVATION TECHNOOGIES ON EROSION AND SEDIMENTATION CITE TECHNOLOGY/TREATNENT LOCATION CLIMATE SLOPE LAND USE SCALE IWACT CoDpARlsoN REMARKS .......... *a..... * .... ft- ........ am- ........... ................... .............................. X .S ..........a.................=*-....--Xl==-a --=a=== == ...... =..... ....... 107 Soit toss tolerances Brazil CU N/A Agricutture Field Acrisols ard Nitosots (Ultisols uith N/A argillic horizon), Lithosols, Regosols - 4.6 - 16.6 t/ha/yr 220 Soll toss toterances USA I N/A Agricutture Field 2.2 12.1 t/ha/yr I/A 13 Soit loss toterances USA T N/A Agriculture Field Topsoil formtion rate (urider tillage) * N/A Wrox. 30 yr/cn or 12.4 t/ha/yr 27 Soil loss toterances WIorldwide All I/A t/A I/A Range of soil formtion rate, in N/A literature * 1.3 - 750 yr/cm or 286 - 0.5 t/ha/yr -- dependent on cloiate, parent mterial. nd criteria (profile development vs total soil depth) 51 Soil loss toterances Pan-tropical EN A N/A Agriculture Field an basis of erosion impacts, highly N/A D/W Ł ueathered tropftal soils should have lower Cll soit losstoterances than their temperate 1/0 counterparts. 50 Soil loss toterances NSW, Australia SATr N/A H/A N/A Soit formation rate: 350 yr/ca In eiluvium. N/A longr in bedrock 2 Ibnure (16 t/he) Indeia D/Wi 16X Agriculture Plot 42S decreae in erosion rate 341 t/ha Coeritwn to unmnured t 21 Cow dung (5 t/he) Ghana CM 7.5S Corn Plot 99X decrease in sedimnt yield 64 t/ha/yr Coeristn to bare fatlow; I Wood shavings (S t7/h) 7.5X Plot 9M1 decrease in sediment yield Pt * 1,340 oyr Poultry wsiure (5 t/ha) 7.5 Plot 99M decrease in sediment yield Cow/pouttry m_nure (5 t/ha) 7.5 Plot 99M decrease in sedimant yield Coercial fertitizer 7.5X Plot 941 decrease in sediment yield 21 Cow dun (2 M/ha) Ghan SATr 2X Sorghum Plot 73S decrease in sediment yield 5.2 t/he/yr Coaerisn to bare fallow; Cow dung (4 t/ha) 2S Plot 81S decrease In sediment yeltd Pt * 320 ma/yr Cow dung (4 t/ha) Ł strew 2S Plot 981 decrease in sediment yield witch (4 t/ha) Cow dung (S t/he) 2X Plot 835 decrease in sediment yield Cow dun (5 t/ha) & stra 2S Plot 9JS decrese In sedimnt yield Mtch (4 t/ha) 96 lutch Nigeria W1/D N/A Agriculture Field Protection fri coepation aId NA Best wom sgemnt prartice for increased erosion Luvisole, Podzoluvisols (Af1iso1s) 149 502 asphalt mitch Venezueta w/D 4S Sorghum Field 680 decrease In erosion rate 57 t/ha/yr Crerison to no-mutch; 100X asphalt wilch 42 Field 99.9v decrease in erosion rate soil * sandy to_n (Typic Heplustalf) 123 llutch (0.62 - 1.23 t/he) USA T 15X Agriculture Field 662 decrease in sedimnt yietd N/A Coarison to no mulch *.alch (2.5 t/ha) 152 822 decrease in sedimient yield Table 8.3 IMPACTS OF SOIL CWOSERVATION TECHNOtOGIES ON EROSION AND SEDIEIINIATION CITE TECI10LOGY/1REATHENT LOCATION CLIMATE SLOPE LAND USE SCALE IIPACT C*WPARISON REARlKS MIteh (5 - 10 t/ha) lS 95f decrease in sediment yietd 99 Mutch (0.55 tlba) USA I 2X Agricutture Fleld 401 decrease In interrill erosion rate N/A CoWrlson to no mulch; sOil I Mulch (2.2 t/ha) 21 80X decrease In Interrill erosion rate silt tom (Typic NeplustlIf) Mutch (6.8 tCha) 2X Interriil erosion rate negligible InIerril erosion * erosion by ratidrop splash 171 Mutch (0.33 t/ha. 2 cover) USA 1 NG Agriculture Plot 421 decrease in sedimnt yiltd N/A Coarison to no mulch urder Mutch C0.4 t/ha, 376 cover) 431 decreas In sediment yield rainfall simulator Mluch (066 tsha. 46x cover) 621 decrease in sediment yield M3lcuh (0.88 tEhaa, t cover) 62X decrease in sediment yield Mutch (1.21 88ha, 811 cover) 81X decrease in sedimnt yied Nutch (187 W/ha. 87a cover) t81X decrease In sedtimnt yield lulch (2.47 C/a 921 cover) 921 decrease in sodiment yRied 113 Nulch (0.63 t/ha USA I 5 Agriculture Plot 69U dcerease In sediment yield 31 t/ha Coerison to no malch; soi I Hulch (1.23 t/ha) 51 89X decrease in sedimo t yfeld silt lam; rainfall simurator Mulch (2.94 I/ha) SI 971 decrease ib sediment yield sup 6Ibtdh t1 04 t/ha) SsX99.91 dereas in sediment yield mulch (9.88 t/Va) St 99.9X decrease in sedimant yield 199 Mutch (0.6 tIba) Indonesla DIai 7-1 Upland rice Plot 78X decrease in erosion rate 18.6 C/ba Coarison to uwmlched; soil Mulch (1.6 t/) 921 decrease in erosion rate Reddish-bromn Latosot; rice Mulch (0.6 t/ba) 968 decrease in erosion rate straw much Mtuch (0.6 W/ba) 91 241 decrease In erosion rate 22.3 t/ha Multc (1.0 Meh) 441 decrease In erosion rate Mulch (0.6 Me/b) 5 decrease In erosion rate Mulch (1.0 C/ha) 141 163 decrease in erosion rate 27.9 E/ba Mulch (1.6 C/ha) 591 decreae" in erosion rate 0 59X ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~r decee Inottoving rat 188 Mulch (0.6 t/ba) Indoneia Is 910/11 7-142 Ufpland rice Plot 351 decrease in erosion rate Comprison to uwuichbed;si Nutch~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~n (1.i ttlon. $oi Mulch (11.0 t/he) 611 decreaes. in erosion rate Reddish-brown Ltatsol; rice Mulch (1.6 E/ba) ~~~~~~~~~~~~~~~~~~~~681 decrease In erosion rate straw mulch 213 Mixed croging IL mAch Indonesia 0/ill 201 Corn IL peanuts plot 741 decrease in erosion rate 250 lbs. Comarison to corn A pemmts Nixed croping No significant differene in etroion rot tr teropped 160 Mulch/Cover Crop Pan-tropical NIA N/A Agriculture Field Protection f rm coation and VIA Luvistoso Nitosotls Acrisola Increased erosion (Ultisols wod Alfigols) with san* topsoils susceptible to compation and rincreased erosion following eaposure and cultivation. 190 Milch and minimug tjIlage Indeneia En 3.51 Cassea/rice/ Plot 9M decrease In sediment yield 16 t/ha Compiaisn to wesalched. SW* Sol t 3.5% corro/peouts/ Plot 6202 incron" in udlmnt yield ~136 C/ha traditional cultivation; Nuich wW minisum tillop 91 beans Plot 921 decrease In sediment Yield 332 W/ho In order of treatment pairs Dare, soilI 91 Plot 2451 increase in sediment yield 195 E/ba seils * Niosed (Cropuialt), Mulch and minirnem Ciiage 101 Plot 563 decrease In sedtiment yield is C/ba Nitosol (Utropndalf). Table 8. 3 IMPACTS OF SOIL CONSERVATIOII TECHHOLOGIES ON EROSION AND SEDIMENTATION CITE TECIINOLOGY/TREATMENIT LOCATION CLIMATE SLOPE LAND USE SCALE INPACT COMPARISON REMRKS Bare soil 1DX Plot 25K increase in sedi_mnt yield litosol (Tropudatf), Mutch nd minimu tillat 14K Ptot 05X decrease in sediment yield Ferralsol (Haplorthox), lare solt 14X Plot 149X increse In sediment yield Nitosol (Tropudtlf), Nutch and minimu tiltag 14S Plot 96K decrease in sediment yield Bare soil 14K Plot 431K tncresc in sedimnt yietd 116 Disk plow/stubble mulch Thailand D/h I NG Lpland rice Plot 64S decrease in sediment yield 5.5 t/ha Carison to disk plowed and No-tit II/mch N Plot 75K decrease in sedimnt yield no mtch 116 Disk plow/stidbbe mutch Thailand D/A I N Peanuts Plot m decrease in sediment yield 3 t/ha Coaerisen to disk ptowed and No-till/mitch II Plot 83K decrease in sedimnt yield no Mich 41 Fascine Philipines DAN 20-351 Grazing/kaingin Field 154K increase in 5 yr sediment yield 48.O t/ha Cowprison to grass sod of Love Mattling 142 increase in 5 yr sedie nt yield end ermad grasses/Alnus,Pinus, Mutching 145K increase in S yr sediment yield Paidium,and Eucatyptus seedtings; Bare soit 1.110X increase In S yr sedimnt yield soil * clay oam, Pt-2679 _/yr 41 Fascine Philipines DA/ 20-35K Grazing/kaingin Field 110X increase h sediment yield 1.05 t/ha Cearison to sediment yield at Wattling 9ff increise in sediment yield years 4 4 5 of grass sod of Love Mulching 152X Increase in sedi_mnt yield wad Bermud grass/Ainus.Pinus. Bare soit 14.500K increase in sediment yield Psidia.Eucalyptus seedlings, soil a clay toam Pt-2.679 /yr 102 Desodium spp. cover Taiwan D/N 461 Litchi Field 95K decrease in sediment yield 55 t/ha Caperison to clean cultivation; laia gras 99.61 decrease in sediment yield Pt * 3,494 _tyr Eregrostis app. mIdc 96K decrease In sedimnt yield 162 Grass cover Coledia 1/D 22K Pasture Faied 97X decreeas in erosion rate 0.3 Itha Coserisan to mnthly tilled " bote soft. 0 50 Gre" coer NSW, Autralia SAtr aX Grassland Plet 98K decreas in sedimnt yield 47 t/ha Ceqarison to heat; soilt Gras cover 8.K S rstwd Plot M9X decrease In sedi_nt yield 140 t/ha calcic and chroic luvisot;. respectively; Pt * 644 and 561 #yr. respectively 215 Shaded grss Nigeia W/o NG Agroforestry Field 4K increase in eroion rate 0.1 t/ha Coarison to tree plantation Unahadad grass 91K deereas" In erosion rate 202 Grea cover (Paniseta, India D/WI 5K Agriculture Field 70K decrease in In sedimnt yield N/A Coerison to conventional Cynodoi,U1rochloe,Panicum) cultivation of gorapeddy.urid. corn.pewsut; soil * sxndy clay. Pt * 1,302 rnyr 7 Grass - ugrazed 15 years Indis 0/Wi Gentle Pasture Field 63X dcrese in sediment yield 8 t/ha/yr C l rison to ffnemd wnd lightly Grass - heavily grazed 163K increase in sediment yfied grazed 211 Short dLwation grazing USA SATap 0-3X Pasture Ftild 200X increse In sediment yield N/A Cperison to moderate, (few day high d renity continuous grazing; soil * fine grazing, oway day rested) lom to lom Trermeot (Aridisol) 161 Grass - loperata,Saccharut Philippines D/Wi 36-70K Grassland Field no significant difference in pediment yield 0.4 t/ha/yr Coarison to secondary forest Ta.ble 8.3 IMPACtS OF SOl COMIVATION TlECINOOGIES ON EROSION AI SEDilUlIATU CITE TECHNOXOGY/TREATMENT LOCATION CLIMATE SLOPE LAID USE SCALE IMPACT CORPARISON REMARKS Plantation forest - Forestry No sIonIficant difference in sediment yietd ClIricidie.Leuceena Kaingin Agricutture 2,7602 increase in sediment yield 217-Grass strips USA 1 M Wast6 treatmnt field 99S decrease In sediment concentration 5,215 pps COmrison to concentration nd 67S decrease In SOD levels and WD of water prior to movement through grass strip 217 10 ft. uide grass strip USA I MG Agriculture Plot 50S decrease in sediment yield N/A Coarison to sediment Load 25 ft. wide grass strip 70S decrease In sediment yield of wter prior to movement through grass strip 102 Eregrostis barrler/mutch Taiwan D/lWl 23X eanan Field 97.5S decrease In sediment yield 53,t/ha/yr Coseriason to ctean cultivation; Guinea grass barrier/.atch 23S 972 decrease in sediment yield Pt * 2.3U _/yr South African pigeon grass 23S 972 decrease In sed8tmt yield barrier/mutch 39 Vegetative buffers USA T 0-202 Agriculture Watershed 54-902 trap efticiency by riparian buffers N/A Trap efficiency a X of eroded sedients deposited in riparian one 3 Grass strips (0.5m wide - Java EN 15-222 Agriculture field 932 decrease In sediment yield N/A Comprison to bare soil Brachiaria, im wide - Papa"it.) 31 sediment basins USA T NG Beans/pess/ Watershed 802 decrease in sediment yield N/A Casrison to calibration Grass strips NG eugarbeets/ Watershed 40 602 decrease in sediment yield period; soil * silt toam Mulch in furrows NG alfalfe Plot 902 decrease in erosion rate 4 Grass strip (P. natatt;, Indonesia 0/mI NG Agriculture Plot 212 decreae in erosion rate MG CoMarison to no grass strip I a wide) Grass strip (B. decuebena 242 decrease in erosion rate 0.5 * wide) 183 VegetatIve strip (muram Indonesia 0/WI 122 Corn a cessava Plot 132 decrease in erosion rote 89 t/ha Coaparison to corn Ł cassava w/o (utilis) vegetative strips Vegetative strip (Pucralc so significant differenc in erosion rate phaseoloide) Vegetative strip ( mimosa 8X decrease in erosion rate inviss) Vegetative strip (Pemnisetun no significant difference in erosion rate purpureus) 74 Reforestation USA T NG Forestry Watershed 99.9n decrease In sedimnt yield N/A Comparison to pre-treatment Seeding native grass MG Grassland Waterehed 99.92 decrease in sediment yield Contour strips MG Agriculture Watershed 50X decrease in sediment yield Crop rotation * fallou MG Agriculture Watershed 502 decrease in sediment yield 22 Vegetative stabilization USA N/A Streaanks Watershed Mt eliminate forces that cause bed N/A Chinmsl structures will become of streaanks degradation before possible to Ineffective or deteriorate due Table 8.3 IMPACTS Of SOIL CONSEtVAltON TECIUIOLOCtES ON EROSION AlN SEDIMENTATION CITE TECNIOLOGY/TREATMEIT LOCATION CLIMATE SLOPE LAND UOSE SCALE INPACT COMPARISON REMARKS stsbitise banks vegetatively, to hvdrology changes; must have mintenance. 137 mulch and no-till USA I 5.52 Corn/soy/oats Plot 942 reduction in sediment yield #/A Carison to sa tillage Muich nd eutiboard plow 5.5 Plot 415 reductlon in sediment yield type with less than 15S crop mutch and seep tillege 5.52 Plot m reduction In sediment yield residue cover as mulch; Mulch nW chislt S.5 Plot 452 reduction in sediment yield soli - fine tom Luvisot Nulch wr chtset A disk .57. Plot 26ft redcttoen in sediment yield tlypic Ilpiudalf) 116 No-till Thailantd l /l MG Rice Plot 93 decrease In sedimnt yield 40 t/ha Camrison to clean Sttbe Mlch NG Plot 902 decrese in sediment yield cultivation iar teallow MC Plot 3752 increase in sediment yield 212 No-till with stete mulch LUSA T 3S Corn Field 602 Increase in sedi_nt yield N/i Cowparlson to conventional (chamicat weed controt) titlag with stutbte mulch 173 Cultivated fataow India SAT 0.52 Fellow Field 1,239S increase in sediment yield N/A Coarison to natural cover; Cultivated fallow 1S Fallow I502 increas In sediment yield solt * sitty clay lom. Cress cover (Cynodon spl.) 0.52 Grass 122 increas in edimnt yield Pt a *00 rn/yr Crass cover (Cynodon spp.) 12 Crass 62 decrease in sediment yield Peant 0.52 Peunut 3272 increase In sdiment yield Pauut 12 PC~ut 3952 increae In sediment yield Grm 0.52 Grm 691X increase in sdiment yield Crm 12 Grm 4302 increase in sediment yield ° Jowar 0.52 Joar 9972 increae In sdimnt yield Jioar 12 Jouer 5272 increae In sediment yield 63 Coffee - I yr old lndnsi 0/DM 46-66X Agriculture Plot 526X Increase in erosion rate 0.31 t/ha Caristn to urdisturbed natural Coffee - 3 yr old 4063 increase In erosion rete forest Coffee - 16 yr old 3102 increase in erosion rate 8 Tobacco (cultivated) Indlnesia D/N 30X Agriculture Plot 138 increase In erosion rate 3.2 tVhaeZo Comrison to non-cuttivated Potato (cultivated) 557S increase In erosion rate cassava Corn * be (cultivated) 522 increase in erosion rate 54 toe. optL dam slope Peru SAl 252 Potato Field 1762 inrese In sediment yield 4.7 t/ha Comparison to continuous fallou Contoured 122 Increase in sediment yield Cover cropped (Lupiius) nd 22 increaeu In * dim_t yield rastp & does slope mucided No increase In sediment yield 54 Rows up & dom slope/burned Peru W/o 302 Potato Field 68E0 increase in sediment yietd 16.4 t/ha C rerlson to continuous fallow Cover croped rows up 1572 Increase in sediment yield & daem Contoured 5762 increase in sediment yield sulco and rows up 4 dous 4904 increase in sediment yield 122 Bare fellow Ghara ID/ 7.52 Corn f Ield 9647. Increase in sediwAmt tield 13.1 t/he/yr Camarison to treditional mixed Zero-tli bae i 7.52 802 decrese in sediment yield croppinr; soil * sandy tom to Mulching 7.5X 98 decrease in sediment yield sandy clay. Pt * 1.500 t wyr Table 8.3 IMPACtS OF SOIL CONSERVATION IECNOLOGIES ON EROSION AND SEDIMENTATION CItE IECONIKOLGY/TREAtIENT LOC/ATIN CLIMATE SLOPE LAND USE SCALE INPACT COMPARISON REMRKS Ridging aroes tip. ?75X 81S decreae in sedimnt yield Minim tillag 7.51 83X decrese in sedimnt yield 122 Bare falltew Ghiu WV/ 3 Corn Field 75 Increase In sediment yield 2.1 t/ha/yr Coprison to traditional mixed Z re-tiltoge 3X 2M1E incraesc in sedimnt yield croping; soil * sandy eey lom Iikchin 3S 38S decrease In sedlmnt yield to sandy clay. Pt * 1,400 _/yr Ridging aeros slope 3X 62S Increase in sediment yield liniom tiltiw 31 33S increase ein sedimnt yield 165 EeforeststionJmluchiplate Jopen T 701 Forestry Watershed 99.71 decrease in sediment yfeld 69 t/ha/yr Ctarifon to bare soit terraces 2 AIrofarestry (intercrepping) Theilnd D/IM NG Omelina rubwre Field 101 increase in sediment yield N/A, Ccarison to Gmetinr orboree and rice crapped aetne 9 Agrefrestry Indonia SATr NG corn Ł teax Plot 121 - 311 decrease in erosion rate 6-287 tiha Corison to corn nonocutture 72 Acacia auriculiformis Indenecl / DAN N Forestry Filtd 241 increase fit rainfall erosivity betlow N/A Ct rison to rainfall erosivity plantation - S yr old plantetion canopy outside of plantation 12 Clearcut/brush chopping USA F 301 ForestrV iaterahed 16.3 t/ha sediment yield N/A All 3 site prep Nethods increase (371 bore soil) sediment similarly; soil Clearcut(uith ahear)/windrou 302 16.5 t/ha sedimnt yield sandy lom (Ultisoia/Alfisols). (531 bore soil) Contour bedding difficult in Clearcut/contour bedding 301 21.7 t/ha sediment yield steep, stump covered areas. (681 bere soil) leproepr bedding cuses gullies. 62 Ctlrcut Chile u/D 30X Forestry Plot 2.140X increase in sedinent yield 0.2 t/ha/yr Comrlson to 30 yr. otd Pinus Posture 301 Posture 1001 increase In sediment yield plantation; soil a clay loam; t'3 6 yr. old Pinu plantation 30S forestry no significant difference In sedimnt yield Pt - 2,000 us/yr 215 Coffee A shad Costs RICa W/o NO Agroforestry Field 781 decrease in erosion rate 0.4 tIhe Cerison to coffee/no shade 215 Tea, 651 cover Inqis 0/ll NGC Agroforestry Field 871 decrease In erosion rate 4.6 t/ha/yr Coperlson to tea with 151 Tea, 951 cover IG 951 decrease in erosion rate Cover 215 Multi-story tree garden Psn-tropical Various Various Agroforestry Various 981 decrease In erosion rate 2.8 t/ha/yr Median observed values from Natuwal forest Various 895 decrese in erosion rate literature - Cosarison to Shifttig cultivation, Various 951 decrme In *erosion rate shifting cultivation during fallow period cropping period Forest plantation, Ver low 791 decrease in erosion rate twdisturbed forest plantation, burned, Various 1t800% increase in erosion rate no litter layer Tree crops with cover crop Various 731 decrease in erosion rate and mulch 1susuy Various 881 Increase in erosion rate Tree crops clean weeded VerIous 1,6101 Increase In erosion rate 215 Multi-story tree garde Pen-tropical Various Various Agroforestry Various 99.81 decrease in erosion rate 70 i/ha/yr Maxim. observed values from Natural forest various 911 decrease In erosion rate literature - Comparison to Table 8.3 INPACIS Of SOIL COIISEEVAIION TECIUOGIES CO EROS10 AND SEDINEINAT IO CITE TIECaOMOST/ATEuNT LOCATION CLINATE SLOPE LAiD USE SCALE INPACT CmPARIsOu REHARKS Shifting cultivetion, Vartous 895 decrease In erosion rate shifting cultivation during follow period cropping period Foret ptmntation. Varilr 91X dareaoe in erosion rate .mldisturbead Forest plantation. burned. Vrou 0S ncre In erosion rate no titter tayr Treo errop with cover crop Various 92S decre in erosion rate nd mtch l1url Variou 293S increase In erosion rate Tr crops clean d Varius 1612 inerease in erosion rate 216 1 yr old reforestation USA T us Various Watershed 1X decrease in sediment yields N/A Cosarison to before treatment. I yr old reforestatin a NG 65S decrers. In sediment yields brush dea 2 yr old reforestation NG 4S decreose In sedimnt yitlds 2 yr old rtefre tation a II 70S decrease In sedimnt yields brush dw 5 yr old reforestation NG 10S decrease if sedimnt yields 5 yr old refoestation & Im 751 decrease In sdimnt yields brush cb e yr etd reforestation NG 3i decrease in edimnt yields e yr old reforo tation & ilG 95 decre In sedint yields brush dar 12 yr old reforestation NG 9ff decreas In sediment yiotds 12 yr old reforestation A Ns 9Si decre in sedimnt yields brush dew a0 Ilare sofl Korea T 271 Forestry Watershed 1.01101 rincrease in sediint yield I.? tlhe/yr CoMrison to 9rassed- Msrdeood plantation 113T increa" in sedient yield soil * sandy clay tlom; Coniferous plantation 100S increase in sedint yield Pt * 1.370 rnyr 20 sa-eso il dom T l Se forestry Watershd 38 9001 increase in sedimnt yield 0.3 t/ha/yr Comperison to grssed; Watl-foreted 4311 increase in sediment yield soil sarndy Touns plantstion 3000X increa In sedimnt yield 138 Cler eutting e" Zeoland I Sttep Forestry Witersied 3.900 increase In erofian rate 14 t/hayr Ceprlson to pro-cleareut 141 Crop rotation I ines D/A as Cor a cassava Plot 3261 increas in erosion rate SS t/hl/yr Coarison to corn nocutture Crop rotatimn l~upwrie/ 4401 rIncreas In erosion rate caroeasseva & isnuats/sres/ bone 1t lixod crqing Indonesia CAN 20S Ctlbewu/ Plot 99a decreae In rosion rate 8 tow Carison to corn noculture csee/sator [aon/Coco"si 141 Crp rotation Indonsis DAN I Pe nut/corn Plot SS decrease In erosion rate 15.2 t/ha/yrCperison to corn/Ceoib 4rotforestry with crop cO Wcssw7a-pemtnt pentandro intercrop Agreforestry with crop l~~~~~~lpnd rice/ 351 - 73 decrease in erosion rate rotation corlVcasAava/ Table 8.3 IMPACtS OF SOIL CONSERVATION TECmbOLGIES ON EROSIO AND SEDIMENTAT ION CIIE TECIOLOGY/TREATMEIIT LOCATION CLIMATE SLOPE LAND UISE SCALE 1IPACT CWOPARISON REMARKS *~~~~~~~~~~~~~...... * - .--.----.. a... on.* ............ . ...... 0 ............ ...........= ....................................................................................................................---.. a........s==*. aSt == = = beaiu/banii/ grass/Ceiba pentwndre 150 Shifting cultivation Guatemala se /lb Steep Corn/beons Fietd No significant difference in erosion N/A Low xilk density volcanic rates folleoing clearing. toii-- Absot (Aneptl 215 2 yr old corn swidden Philippines DOM MG Agriculture Fietd 275X increase in erosion rate 1.6 t/haeyr Coarison to I yr. old swidden 215 12 yr ald rice suidden Phitippine, D/WN NG Agriculture Field 7.3503 Increase in erosion rate 1.6 t/he/yr Cearison to 1 yr old swidden 73 Plow clean tilltcontour USA T 6o Agriculture Field 86s decrease in sediment yietd 51 t/he Coqrison to ptow.clean till. lo-tilt,contour 21X Field 99.9X decrease in sediaent yield and sloping rows 139 lo-till, 4 t/h mulch Nigeria U/o 2X Corn Field m decrease in sediment yield 6.5 t/hs Comarison to conventional Hnd cultivation 2X Field 38X decrease in sediment yield titlge (ptow 4 harrow) Naoutckord ptow 23 Field No significant difference in sediment yield Plow, bare ftllow 23 Field 1043 increase rn sediment yield is? Ho-tlt llNigeria W/D) 13 Agrlculture Field 983 decrease in sediwent yield 0.4 t/he CoWrison to conventionst Bare fattow 13 Field 4003 increase In sediment yield plowing No-till s5 Field 99.93 decrease in sedimnt yield 2.2 t/he Iare fatlow 53 Field 643 Increase in sedimnt yield No-t il 15X Field 99.9S decrease in sedieent yield 3.9 t/he Bafa f ilie 15X field 3103 Inerease in sediment yield 6? Mk-till Trinidad W/D S11 Corn Field 70s decrease In sediment yield I/A Comparison to bare soil; soit *' no-till fiX CoWeo 94S decrease In sedi_nt yietd s ciay Acrisot (Uttisol) Uo tilt 223 Corn 74X decree In sediment yield 223 Covpeo 87 decrese in sediment yield Ho-titt 52S Corn 61X decrease in sediment yield 52S Coupte 7M3 decrease in sediment yield 61 Ho-tIl 1rinidad 11/D 22M Corn Field 903 decreas, in sediment yield II/A C'arison to bare soit; soil Till 223 Corn 893 decreae in sediment yield sandy clay Acrisot (Ultisot) Mo-tilt 223 Coupee 853 decreae in sedimnt yield Till 223 Cowea 943 decreae in sedimnt yiold 36 Detp tillae Senegat W/D HG Agriculture Field 633 decrese in erosion rate 10 t/ha/yr CMarison to shatlow tiltage in sendy soils. 114 bep tiltlge Brazil U/D HG Cotton Field 100t increase In erosion rate 8 t/ha/yr Comarison to shattow titlage in Ferresol (Oxisot). 4 Contour rows India D/lu HG Agriculture Field 303 decrease in sediment yield I/A Cowrison to up Ł down stope cultivation; based on review of 30 years of experiment station projects 219 Contour rows USA T 1-2X Agricutture Fietd 403 decrease in erosion rate H/A Comparison to un-contoured rows Contour rows 3.-53 50 decrease in erosion rate fult benefits only in gully and Table 8.3 INPACTS OF SOIL CaNSEIVAtIO TEaCNOCIES ON EROSION AND SEOIIEiA ION CITE TECIIOMOGY/ITEATNENT LOCATION CLIMATE SLOPE LAD USE SCALE ISPACT CWASISOh REIIS *.-0 ~~~..... * ,,- ........a.....a..... . --...--. ---0.....M.88. 0__00 *O Contour row 6-9X 50X decrese In eroin rate dipreesln free fieds. If Contour raw 9-12X 402 decrease In eroin rate brtalovers of centour row Contour ras 13-16S 30X decrese in erosion rate eeer thn eron will be Contour raw 17-20X 202 decree in roeion rate grter than in uwntoured. Contour ram 21-25S 10S decre In erosiln rate 132 Contour strip-cropping Atl All Verious Agriculture Field 35-75S reduction in sedimnt yield, slqe i/A Co_risn to up A dan sl*p. dependent ti1ttg 195 Contour strip-cropping USSR T NG Agriculture feld 50S decreae in erosion rate */A CeWrisen to non-etrip-cropped, Wm stripe So-lOan uld. 142 roadbed & furrow India SATr 0.42 Corn/sorohus Field 87X decreae In sediment yield 6.6 t/ha Coerison to treditiorI $roadbed & fwrrow 0.6X intercrogped Field 822 decreae in sediment yield cropping syato; silt very roadbed A furro 4 0.42 with Cajais Field 91X decrese in sdimnt yield fine clay Vertisol field binds caln (Typic Peilustert); FIat en grade 0.6S Field 80X decrese in tdimnt yield Pt S 760 _/yr 142 Uroadbed t furrow Irdia SATr 0.61 Pearl millet/ Field 49X increase In sediment yield 2.S2 t/a C rison to treditisml Ftat on grade with graded 0.62 sorghum Fietd 722 decrese In sediment yietd crepinG syt_ with field burKe intercropped bumb; soil - fins Uidic Traditional with contow with Cajanus Field 87X decrease in sedimnt yield Ubodo intalf. Alfisole burdc IIC caj;n unstable In bnds L turrows. Pt. 740 myr 130 Bro~eXd end furrow Irdia SAT NG Agriculture Field 832 decrease in sedimnt yield i/A COprism to traditional faetfJ tamot with bunds 131 Cross-stope planting India W/o i.SX Grain Plot 432 decrease In erosion rate 25.5 tn/h C rlesn to sap A dtehaiop tidge S furros 962 decrese In erosion rate plAntino; soil - .mlt-dreaind Sandy tooe (linceptisol) 178 Rldge-titlage USA I NG Agriculture Field Nay increase sediment yield V/A Incre in sediment yield by Rige-tildue e with crop 50-862 decrease in sedimnt yield rldes-tillg turkcition of runof residies in fur-rows concentrated in furrows md rides incraing step. 90 Burn to clear/no-till Jan / 12-18 Agriculture Field 8S2 decreae in sediment yietd 20 V/he CamPeism to bun to clear lulldoze to 42 slope & till 3-521 1502 increase in sedisnt yield Ł tilled;- soil a sandy Clay Forested (tree ht. * 10 a) 12-18X 99.92 decrease in sediment yield lome; Pt * 220l0 yr 102 Benh terrace Taiwan DS/1 282 Citrus Field 972 decre"e in sedimnt yield 156 tIhyr Corieon to clean cultivation Nulch/Bdhis grass cover 982 decrease in sedimt yield Pt * 1i634 _tyr malch/Bdaia grass cover ~~~~~~~992 decreas In sedimen yietd 102 Leel terrace Taiwan t/1ig 242 Baana Field 99.57 dacrease in sediment yield 39 t/ha/yr Cq.rlion to clean cuttyetiont Grass barrier (2.5Spa cing) 972 decrease In sediment yield Pt 2,274 mly Grass barrier (5m spacing) 90X decrae in sediment yield ilulch/Bahie grass cover 99.72 decrease in sediment yield 102 Benrh terrace vs Bahia grass Taiwan O/Wm 20-28X Peremvist Field Soi conservation effect of att types of ViA Table 8.3 IMPACTS OF SOIL COUEVATION TECaCUOGIES ON E11C18 AMD UDIOEITAT?IOI IIE TECiNVOLOGY/TUATNEY LOtClaION CLJMTE SLOPE IND USE SCALE INPACT alkit bech trr m poore than that of Oak% Grassi (capt level rent ton terr ewr) 66 Contor raw India DCAM 25 Potato fiod 401 derees In tdl_nt yield I/A C _risa to Sp A don stlap Up a darn alope plentin. 301 ducrae in sedimnt yietd cuitivation contour furrow Prr_nent gr 991 decre In eadimnt yield enac terra" 9 decrene in sediment yeltd 143 Cwnto.r cultivation India DNi 251 Potato Field 1.4001 ircnrea In ediCAnt yield 1.1 tia Coqwism to bench ttrrwd; ft * 1.2ff _pr 16 Contuw cutivation India 11D 2.21 hrtsyltjoer Plot 621 dcear in esion rat. IAke CU heriw M to up I da Glep 102 ever. soIpr ben' terrace Taiwan DAM 201 Pin pple Field 99.91 decwre in sediment yield 34 t1halyr Comparison to pntw elnp t Nutch & close planted an 841 decrane lb sdient yield damn slqe; Pt a t 373 I W the contour 92 cr,s waterwa US T All Agricultur Watered V/A UIA If eosin rates w atetm mw hi than difficult to asabdirId. eapemive to mintain. I2 Vegptative gully controt India Att I/A Atl U/A Gutly stdbititetion A Itf *lpe of gAty chamt r 191 Nd drairnee mrea omit saiq Ieelfn1 ad reatatn s approprirte aoch 55 Otulide sully charel USA SA/A so Various Watardbed 25-601 decrense in sedient yield II/A Compwison to beforet treetian mesures to: decrease overland flow. spred ter. increse Infiltration Outside gutly chare NG 60-15% decrea In sedleunt yield Wmasures (sa as above) & gully structures Inside gully chanel V/A N/A Nay increne sedi_nt yield as gr tree plantings beyond sapling stag and divert flou flow Into gully wlls 121 Contour banh South Africa T 9X Pinerpteas Field 2001 Incrase in saditmt yield 33 t/he/yr COarlsn to no cent.or bind; soil a shallow, erodibte send 14t Contour urd every tO rows -ary r 32X Vlneyrd field 25X decras In sedimnt yield 4.4 t^Wbre Ccpr5son to no tsotour ba Contour btnd every 5 rows 7M decras in sedlnt yield Contour bund every row 99.9X decreae in edimtnt yietd 79 Contour bds - yr I Thailwnd DiM 11X Agriscuture Feld 100D trap efficiency 'A Trap efficiency a X of ed Contour bud - yr 2 i11 1001 trap efficiency adime rter 45Sm ad parent mterial weatbered (em sick pWuite or seals) 94 Terraces (200 inter-terrace Nigeria UW/ 1S Agriculture Plot No signifieant difference in erosin rate 6.2 tia Cmrpisen to erosion rate frm width) mn strer event on 10. inter- Terraces (200 inter-terrace 5X 81Z decreae in erosion rate 16. tljh terr ee widths width) Terraces (20e inter-terrace 10 26X decrese in ermion rate 24 tE b width) rerraces (20. inter-terrace lSX se significain differ.m in erosion rate 60 to width) 129 Bench terrece Sierra Leans U/D 30X Rice Field a4x decree In sediment yield 48 tIho/lr Comprison to rice. no Stone btding 30 30X de e in saiint yield emmnietion Stick bunding 30X 43X decreas in aedi_t yietd Centaur btxding 30X 62X decreae In asdnt yield 129 Bemh terrace Sierra Leone W/f 30X Cassva Field a5x decrease in sediment yield 33 tiEyr Ctarisen to casuav no Stick bunding 30X 181 decrea In adi_ ield emnsevatlion Centaur bunding 30X 49X decr In edint yield 143 Bench terrace India D/i 25k Potato Fietd 93Z decre in ediment yield 17 tEb C-rime to cantmur rews; ft * 1t2@95 _pr 144 Broad base terrace Indias SATr GC Agricutture Field 92X decrease in sediment yield /A C erison to unterraced; sel Vertleol 68 Graded vs Bench terraces India D/l 252 Agriculture field So slgnificant differece In sediment yied V/A betueen e two terrace twes 104 Ineard sloping bench terrace Taiwn D/wi NC Sugarcane Field 94k decrease in aedi_nt yield WA Corisen to eut_rd sltping bench tere Table 8.3 IMPACTS OF SOIL CONSERVATION TECHIOLOGIES ON EROSION AND SEDIMENTATION CITE TECINOLOGY/TREATIENT LOCATION CLIMATE SLOP LAND USE SCALE IMPACT COWARISN RsM3AKS 103 Bench terraces laiwan DAI 283 Citrus orchard Field 97X decrease In sediment yietd N/A Coqiison to clean cuttivation; soil * clay tlo 12? Sediment basins with pipe USA I MG Crn, Clen Field > 97S trap effieiency N/A Trap *fficiency * % of sedilit drainage frm terraces cultivated load of 1 wff drqpited In bain; soil * (EntiolotSillisol) iontrapd sedimnt * 8SU clay 30 Cully plugs 4 drop structures Indonesia DM 4-l8 Agriculture Watershid All structures failed within 2 years N/A 19 Gully hebadall structures ISV, Australia SATr N/A Gutty Iatershed Greeter then 501 of sediments derive NI/A from gully sidwalls, headeatl control will not give hort-term sdiment reutction 30 Check d A bamboo Indonis D/W 4-83 Agriculture Watershed Increased in-gulty erosion N/A Gully was stabilized until construction of ceckdm 3? Check d_m laiwn DM4 54 Mixed hardwood/ Watershed No significent difterence in reservoir N/A Csrison to sittetion rates conifer forests siltation rates following construction prior to contruction; soil of 70 eheck dams. rocky-gretly sa; Pt * 3020 _/yr 60 Check dm USA T N/A iI/A Watershed Gutty reclamion II/A Nm erous low. smi-pervious. ,. t orr_y check do e are preferble & ore economical 00 26 Check dms USA I N/A Various Watershed increas ftooding If used In pernmial or NIA Useful for drai_wges c 10 acres$ ftood susceptible stream. 81 Check d jan T N/A Varius Watershed Reduced bed gradient requtated stream N/A Concluslons based n erisl width. controtted sedimnt transprt ihotOsrwas 59 Hogpire check dm Phitippires D/I 36-119X Various Watershed 74X decreae in trap efficitecy N/A Crlwison to stone check d_; Brush check da 4S decreas in trap efficiency 5 yewrs for treated gulties to Log check dm 421 decrease in trap efficiency stabilize nd revegetate; only stem check d_o still intact at year S 207 Strw/faric check da All All NG All Watershed 50-95f trap efficiey of sediments N/A Strew/fabric d onty for flow Rock check dam Trap only fine snds nd tcoarsr sediments vetocittes of < 0.06 ft/sec 154 Rock check dms USA T NG Construction Watershed 5-10X trap efficiency N/A CWrison to edimnt tled of site runoff bove cdwck d_ 43 Fabric check da USA T MG Costruction Watershed 99.t trap efficiency N/A Ca_rslwn to sedlnt tled of runeff above check d_ 1INPACTS Of SOIL CONSERVATIONi TECHNOL.OGIES ON CROP YIELD AND PRODUCTIVITY CITE TECHNOLOGYTUREATNENT LOCATION CLIMATE SLOPE LAND USE IWPACT RMAmaW 95 Erosion Worldwide ALI N/A Agriculture Inevitable reduction In productivity. in Andosols (Anidepts) wnd C"isol5 favorable subsoilIs erosion increases (Inceptisots) have good ashsoil production costs with tittle or no losses characteristics, old and highly in yield weathered Ferrasols (Oxisols). Luvisols (Alfisola). Acrisots (UIEfSoolS and Nlteeefe (Al fisoIs and Ultiso(s) usually do not 95 Erosion Tropies All N/A AgricuLture on shallow, infertile tropical soils Drought effects are asto mlfiled productivity may decline more rapidly than in tesperate couniterparts. 93 Erosion Tropics All N/A Agriculture From available data appears that yield Attributable to lower lnAswent reduictions per aunit soil loss of topsoil fertility. concentrations of plant- is more drastic for tropical vs temerate available muitrients mW aomgaic soils. matter in top few cms. to root- restrictive edadphological ly unfavorable sA*soil uwvroiumints. 93 Erosion Worldwide All N/A Agriculture Primary affects of erosion on prodLuctivity Conclusion of the U.S. Natioanal Soil is through loss of plant-available water. Erosion-soil Prodc tivity Research Cemittee 93 I.Shatiom soils, concentration MorLdwide All N/A Agriculture 1.Severe erosion-ind&aced productivity Type I typically highly leached, of plant- available water and decline; will not respond to management In tropical Frerasolcaxisol). Acrisol ' nutrients in top few cm savery eroedphases. (Ultisol); Type 2 typically Anxosols 2. eep soils, good strteture, 2.Nay show no significant yield decline (Andepts); Type 3 would be raem. favorable distribution A higih despite significant erosion; fertility reserves f or plant-availabl, restorable with adeltions of N or organic water and nutrients matter. 3.1oils where topsoil horizon 3.May show rIeId increaesn as the result of bwried bv iess fewrojble so(( sewere frosyon' material 95 Soil loss, natural Nigeria W1/o 1 Corn .26 t/ha decrease in yield/ - of soilt loss Higher rate of yield decline at It 52 Corn .10 t/ha decreaise in yield/ - of soil loss slWop attributed to sevwer 102 Corn .08 t/ha decrease in yletd/ me of soil loss crusting ud "seaing from raindrop 15 corn .10 t/ha decrease In yield/ me of soil loss lqaect 31 Soil loss, natural USA T NO Beans 272 decrease in yield Coaprison to crop yields from NG Peas 232 decrease In yield pltst where topsoi depth m NG Sugarbeets I5% decrease In yield about the own as when the land NO Alfalfa 282 decrease In yield was first cultivated to plots where subsoll was exposed ST soil loss (0.5 - 1.4. cU/yr), Poland T NO Winter ,Aeat 502 - 70 decrease in yield naturet 115 Soil loss (0.1 cm), naturaL Australia SATr NO Agriculture 32 - 7.52 decrease in yield Soil loss (0.8 cm), natural NO 10 - 252 decrease In yield Table 8.4 IMPACTS OF SOIL COISERVATIOI TECH1LOGIES ON CROP YIELD AMID PRI0UCTIVITV CITE TECNNOLOGY\TREATNENT LOCATION CLIMATE SLOPE LAND USE INPACT REIIAICS ssus=s=zS====tA==z=======sa===fl=====lm=t==sf=stf== =n3.=====5==X=-===== - 96 Soil removal (2.54 c) Nigeria U/D us Corn 50 decrese in yield Soi1 - sandy Atfisol SoIt rmoval (7.5 cm) 902 decrease In yield 95 Soil removat (10 cm) Nigeria W/D 11G Agricultwe 652 and 382 dcreae In yield for rain md Soil Paltntlf stover. respectively Soil rmoval (20 cm) 902 md 552 decrease In yield for grin md stower respectively 118 Soil removat (5 cm) Nigeria W/D NG Corn 95 decreae in yield Cqprlwan to no soil romva; COea 632 decrease in yield Solt * Ultiboe Pt a 2U480 r/yr Soil removat (10 cm) Corn 951 dwreae in yield CoWea 712 decrease In yield Soil removal (20 cm) Corn 1002 decreae in yield Cowmea 6MM decrease in yield 118 Soil removal (S cm) Nigeria W/D NG Corn 312 decrease in yield Cerisn to no soil rel; CoW" 22 dereae In yield Soil a Atfieot. Pt a 1,307 _ryr Soil removal (10 cm) Corn 74 decrease In yield CO"ea 59X decreae in yield Soil removal (20 cm) Corn 922 decrease in yietd Coepea 651 dcrease in yield 118 Soil removal (5 cm) Nigeria w/D NO Corn 732 dcrease in yietd Cmpwiwsn to no soil rt-t; SoilI removal (10 cm) Co"a 432 decreae in yietd Soit UAfisol. Pt 1.230 {yr Corn 83X decrese in yield Coupea 332 decreae fit yield ° Soil removat (20 cm) Corn 100X decrease in yield CoW"a 812 decreas in yield 186 Soil removal (10 cm) Indonesia w/D NG Sobmi 482 decreae in seed ueight Liming of seits Increased average Soit removat (20 cm) 65X decrese In seed might ptot yietd by 152 bLt did not Soil removal (40 cm) 79X decreae in seed weight increae seed uelight Soil removal (60 cm) 86 decrese in seed amight 221 soil removaL (0 cm, Hava i i/D o NG Corn stover. 222 increase In yietd; no sigmificant Cwrisen to no soil removal 502 recomeinded fertilizer) 1st crop difference In uter use efficiency fertilizer; Soil * clay Ferrasol soi l removat (0 cm, NG Corn stover, 46X increase in yield; no significant (Trapeptic Eutrustox). 1002 reco_ended fertilizer) let crop difference in Mter wse efficiency water use efficiency = Soit removal (10 cm, no NG Corn stover. 392 decrease in yield; 442 decrease in Yietd (kg)Alater use (kg) fertilizer) 1st crop water use efficiency Soil removal (10 cm, NG Corn stover, No sionificant difference in yietd; 502 recanended fertilizer) 1st crop 312 decrease in Mter use efficiency SoiL removal (10 cm, KG Corn staver, 44 increas in yietd; no significant 1002 recofeuided fertilizer) 1st crop difference in Meter use efficiency Soil removaL (35 cm, NG Corn stover, 912 decrease in yield; 902 decrease in no fertitizer) Ist crop uater uLe efficiency Soil removal (35 cm. NG Corn staver. 35X decrease in yield; 672 decrease in 50 recomended fertilizer) 1st crop water use efficiency Soil removal (35 cm, NG Corn stover, 212 increase in yield; 36S decrease in 1002 reconwnded fertilizer) 1st crop veter use efficiacy Table 8.4 IMPACTS OF SOIL CONSERVATION TECHNOLOGIES ON CROP YIELD AND PRODUCTIVITY CITE TECINOLOGY%TREATIENT LOCATION CLIMATE SLOPE LAND USE IMPACT REMARKS fltnf3SS -esa================.n Bz=as=z===n5S=== =a.=n … sans … aaaaaan==aann-- aanS s=n8=--==snz== ===a====ns= z== e==== = = == == = 221 Soil removal (O co, Hawaii u/D NG Corn, 165X and 34X increase in grain and stover Comparison to no soil removal, no 50W recomended fertilizer) 2nd crop yield, respectively fertilizer; Soil = clay Ferrasol, Soil remval (O cm, G Corn, 317X and 64X increase in grain and stover (Tropeptic Eutrustox) 100X recaeded fertitizer) 2nd crop yield, respectively Soit removal (10 cm, NG Corn, 46X and 41X decrease in grain and stover no fertilizer) 2nd crop yield, respectively Soil renmval (10 cm, MG Corn, 83X increase in grain yield and no 50D recomended fersilizer) 2nd crop significant difference in stover yield Soil removat (10 cm. NG Corn, 302X and 51X increase in grain and stover lOOX recomended fertilizer 2nd crop yield, respectively Soil removat (35 cm, NO Corn, 100K and 91K decrease in grain and stover no fertilizer) 2nd crop yield, respectively Soil removal (35 co, MG Corn, 39K and 41K decrease in grain and stover SOX recomended fertilizer) 2nd crop yield, respectively Soit removat (35 cm MG Corn, 112X and 16K increase in grain and stover 100K recommened fertilizer) 2nd crop yield, respectively 117 Soit added (15 cm), 0 kg/ha N USA SAThp MG Sorghum 69K increase in yield Comparison to non-treated; Soil = Soil reoved (15 cm), 0 kg/ha N 46K decrease in yield loessial, calcareous, silty Soil removed (30 cm), 0 kg/ha N 61K decrease in yield Mollisol Soil added (15 cm), 34 kg/ha N 45K Increase in yield Soil reoved (15 cm), 34 kg/ha N 22K decrease in yield Soil removed (30 cm) 34 kg/ha N 27K decrease in yield Soil added (15 cm) 68 kg/ha N 34X increase in yield N Soil removed (15 cm), 68 kg/ha N 13K decrease in yietd Soit remved (34 co), 68 kg/ha N 21K decrease in yietd 96 Nutrient losses on eroded soil Nigeria w/D 1K Agriculture 50 organic C, 6 N, 0.2 available P (kg/ha/yr) 5 870 organic C, 100 N, 1.8 available P 10M 1850 organic C, 190 N 2.2 available P 15 3070 organic C, 230 N, 8.1 available P 30 llutrient losses on eroded soil Indonesia D/AI 18X Agriculture 30 organic matter, 1.5 N, 1.0 P, 2.0 K (kg nutrients/ton of soit loss/ year) 52 llutrient tosses on eroded soil Zimbabue "/D 3-6.5K Agriculture Type I: 0.97 N, 0.16 P, 10.7 organic C Type I = wetl-drained sands; (kg nutrients/metric ton of Type 11: 2.1 N, 0.16 P. 15.4 organic C Type 11 a other soils sedi ent/yr) 96 Nutrient losses in runoff water Nigeria U/D 1K Agriculture 2.9 N, 0.5 P. 4.7 K, 11.2 Ca, 2.4 Mg (kg/ha/yr) 5K 5.5 N, O.S P. 6.2 K, 17 Ca, 2.5 Mg 10 5.7 N, 0.8 P, 5.6 K, 14.9 Ca, 3.1 Mg 15K 4.5 1, 0.7 P, 4.1 K, 12.5 Ca, 3.0 Mg 61 4K to 5X decrease in plant USA T NG Agriculture 12X - 36K decrease in yield Dependant on crop and degree of available water soil loss 182 Grass cover Colombia U/D 22 Pasture Decreases in nutrient losses: N - 72X, Comparison to monthly tilled P - 85K, K - 75X, Ca - 89X, Mg - 83K bare soil. Table 8.4 IMPACTS OF SOIL CONSERVATION TECHNOLOGIES ON CROP YIELD AND PRODUCTIVITY CITE TECIIIIOLOGY\TREATSENT LOCATION CLIMATE SLOPE LAND USE IIIPACT REMARKS 82 Grasland India D/Il MG Grassland 201 Increase in N comparison to forested laytds; Solt a sanidy loam to clay loam, Pt a 1,050 mJyr; attributed to higher phosphate levels (aids in N fixation) in grasslands 85 Cover cropping with N-fixing India 0/I mimosa imvisa 0.4X increase In organic matter,30X Increase Comparison to clean cultivation ptant (Himsa invisa) - 2 years in available N, 98X inerease in avaitable P, 281 increase in K 160 Green Muwre Morlduide All N/A Agriculture only effects next crop Usually limited to mechanical agriculture as the power to incorporate green muwre into soil often beyond maal labor 160 Organic matter Uorlduide All N/A Agriculture In unfertilized soils supplies N. S. blocks EssentiaL in non-fertilized P fixation, maintains CEC, Isprove structure systems, important in tow CEC of poorly aggregated soils, form complexes soils and poorty aggregated WItht micro-nutrients sands. 160 Co_ercial fertilizer Worltwide All N/A Agriculture Increase soil organic matter due to If soils are adequate, sound increased root decomposition fertilization practices decrease need for organic matter 160 organic vs Commercial fertilizer Wortduide All N/A Agriculture Choice between predominately N based on ecornics, transport . accessability,and social criteria 21 Cow dung (5 t/ha) Ghana Cll 7.51 Corn 1052 increase In yield Comparison to comuercial Wood shavings (5 t/ha) 7.5X 531 increase in yield fertilizer (200 kg/ha Urea, 100 Poultry manure (5 t/ha) 7.5X 321 increase in yield kg/ha triple superphosphate, 120 CoW/pouttV wawre (5 t/ha) 7.51 tIZ Increase In yield kg/ha uriace of potash; Pt * 1.340 ain/yr 21 Cow dung (4 t/ha) Ghana SATr 2X Sorghum 751 increase in yield Coperison to 2 t/ha cow dung, Cow dun (4 t/ha) A straw 2X 75X increase In yield Pt a 320 _Jyr _Ach (4 t/ha) Cow dug (5 t/ha) 2X 88X increase in yield Cow dung (5 t/ha) A straw 21 100X increase In yield mulch (4 t/ha) 197 Mulch (1.1 t/ha) USA T MG Sorghum 351 Increase in crop yield Comparison to no mutch; Pt Mulch (2.2 t/ha) MG 46X Increase in crop yield 808 r/yr Nulch (4.4 t/ha) MG 671 Increase in crop yield Kutch (8.8 t/ha) MG 1071 increase in crop yield Mulch (13.2 t/ha) MG 1241 increase in crop yield 38 Nulch India D/Ill 11 Uuheat/barley/ 301 increase in yield Comparison to uwmsutched; Soil a gram/linseed sandy clay toam 153 Muich India Satr NG Agriculture 401 increase in yield Comparison to wumutched in 4 crops Table 8.4 IMPACTS OF SOIL CONSERVATION TECHNOLOGIES ON CROP YIELD AND PRODUCTIVITY CITE TECHNOLOGY\TREATlENT LOCATION CLIMIATE SLOPE LAND USE IMPACT REMARKS in 6 regions for 3 years 185 Mulch (5 t/ha) Indonesia N/A oX Corn 63XX increase in biosass Greerhouse study 91 Hutch Indones ia W/D N/G Rice 69X increase in yield Plots on 20 yr old terraces; soil Mutch Corn 10X - 188X increase in yietd latosot 193 Mutch (6 t/ha) Indonesia w/D 14X Rice 30X Increase in yield Soil = red latosol 149 50X asphalt mutch Venezuela W/D 4X Sorghum 39X increase in yietd Comparison to no-mulch; 1OO0 asphalt mutch 4X 110X increase in yield soil * sardy toam. Typic Naplustalf 174 Mulch - polyethylene sheets India SATr HG Wheat Yield increase did not justify expense Comparison to urmulched; Soil Mulch - dry grass (10 t/ha) NG 24X increase in grain yield, 30X increase in silty clay toam straw yield 153 Vertical muching (8m interval) India SATr NG Sorghuw 35X increase in yield Comparison to no vertical mulch; Vertical mlching (4m interval) NG 34X increase in yield Soil - heavy black soil 190 Mulch and minimu tillage Indonesia EN NG" Peanuts no significant difference In yield Comparison to urmudched, Mulch and minil. tillage NGI Soybeans no significant difference In yield traditional cultivation; Soil - Mulch and minim. tillage NGN Corn 58X increase in yield Ferrasol (Haplorthox) Mulch and minim. tillage NCG Cassava 33K increase in yield Mulch and minim tillage NG" Mung beans 139X Increase In yield N; Mulch and minimu tillage HGH Upland rice 24K Increase in yield W 190 Mulch and minim. tillage Irdonesia EN NC Peanuts no significant difference in yield Comparison to unmuiched, 1Mulch and minimu tillage NC Corn 15X increase in yieLd traditional cuttivation; Soil = Mulch and minim. tiltage NC Hung beans 48K increase in yield Iitosot (Tropudult) Mulch nd minim tiltage NG Uptand rice 31K increase in yield 212 No-till with stutble *utch USA T 3S Corn 7X decrease in yietd Comparison to conventional (chmdcaL wsed control) titLage with stubbie mutch 40 Land clteirng (bultdoze. plow, Bolivia W/o NG Sugarcane No significant difference In yield Comrparison to slash-and-burn level) 136 Land clearing (0-6 with Brazil Cm NG Rice (2nd crop) UX decrease in yield conventional blade) NG Rice (3rd crop) 32X decrease In yield ttG Rice (4th crop) 74K decrease In yield NG Cassava 55X decrease in yeltd NG Soybean 83K decrease in yield NIG uinea grass 16X decrease in yield 136 Land clearing (0-6 with Brazil Cl NG Rice (2nd crop) 21K decrease In yield C marison to stash-end-burn; conventional blade, addition NC Rice (3rd crop) 10K decrease in yietd Soils a Acrisols (Ultisols) of lime to pH 6.2, 50 kg/ha N, NG Rice (4th crop) 29M decrease in yield 172 kg/ha P, 40 kg/ha K. Same NG Cassava 6X decrease In yield 1 and K additions after each NG Soybean 44K decrease in yield crop or grass cutting) NG Guinea grass 24X decrease In yield TEble 8.4 IMPACTS OF SOIL CONSERVATION TECHNOLOGIES ON CROP YIELD AND PRODUCTIVITY CITE TECNOllOGYO2REATIEMN LOCATION CLIMATE SLOPE LAND USE IMPACT REMARKS 166 Land clearing (slash-wnd-burn) Peru Ci NG Agriculture Increased crop yields Cosparison to butldozed sited; attributed to benefits of ash, soil coapaction and topsoil disturbance by bulldozer 153 Deep tillge India SATr MG Agriculture 28X Increase In yield Corparison to shallow tillage in 10 crops in 7 regions over 4 years 44 Tillage Indonesia EM NG Agriculture Drastic reduction of fertilizer loss Soil x Ferrasol (Oxisol); rainftall Artificial stabilization of HG 95f decrease in N A P. and 90X in K losses simulator study performd In pans soil surface (Polyacrylamide) 97 tinlm tillege Nigeria W/o NG Corn 50X increase In crop yield Comparison to conventional Cowpeas 24X increase in crop yield titlage Soybeans 212 decrease In crop yield Sweet potatoes 24X Increase in crop yield Pigeon peas No significant difference in yield 1Z2 Ninlm. tillage Ghana I/D 7.5X Corn 16X increase in yietd Comparison to traditional mixed Zero-titlsge 7.52 82 decrease In yield cropping; soil = sandy low to lluching 7.5X 57X Increase In yield sandy clay, Pt = 1,500 suZyr Ridgi#W cross elope 7.52 16X Increase in yield 122 Minimu tillage Ghana U/D 32 Corn 1SX decrease In yield Comparison to traditionat mixed a Zero-tillte 32 51X decrease in yield cropping; soil = sandy cisy loa INichirg 3X 72 increase in yield to sandy clay, Pt = 1,400 mi/yr Ridging cross slope 32 21X increase in yield 158 Mechanized agriculture Loessial soils All N/A Agriculture Rapid physical degradation and productivity Beneficial effects of tillage do decline in tropics; process slower in not last unless lime or use temerates rotation with deep rooted grass 36 MNd cultivation C. 5cm) Senegal W/o NG Millet 242 yield Increase Comparison to non-cultivated Nehanical cultivation(IS-ZOc.) Mtilet 222 yield increase Nwd cultivation (4cm) Sorghum 24X yield increase Mecnical cultivation(IS-20c.) Sorghum 25X yield Increase MNwd cultivation (< 5cm) Corn 352 yield increase Nedcanical cultivation(1S-2kcm) Corn 73n yield increase ti d cultivation (< Sce) Rice 103% yield increase Medcnical cultivatiomi15-20cm) Rice 732 yield increase Mnd cultivation (c Sew) Cotton 25X yield Increase Nehanical cultivation(15-20cu) Cotton 382 yiteld increase iand cultivation C' 5em) Peanuts 222 yield increase Mechanical cultivation(1S-2ocm) Peanuts 8X yield Increase 16 Contour cultivation India U/D 2.2% Jower 282 increase in yield Comparison to up & do6lope Contour cultivation Barley 232 increase in yield ptanting; soil = eroded, alluviai loam 46 Contour rows India D/m Up to Sorghum Lp to 352 increase In yield Comparison to non-contour Table 8.4 IMPACTS OF SOIL CONSERVATION TECHNOLOGIES ON CROP YIELD AND PRODUCTIVITY CITE TECHNOLOGYTREATNENT LOCATION CLIMATE SLOPE LAND USE IMPACT REMARKS 3K cultivated; based on review of 30 years of experiment station projects 131 Cross slope planting India II/o 1.5X Grain 23X increase in yield Coirparison to up & downalope Cross slope planting Straw 10X increase in yield planting; soil a well-drained, sandy Ridge & furrow (60 cm spacing) Grain 59X increase In yield loam (inceptisol) Ricde 4 furrow (60 cm spacing) Straw 360 increase In yield 25 Ripped furrows USA SATpp NG Grassland 250X increase in yield Corparison to non-ripped (Boutelou spp) 47 FurroeL USA SATmp NG Grassland 300K increase in yield, 50K increase in Comparison to non-furrowed (Buchtoe spp) cover 104 Bench terrace Taiwan D/WI 12K Sugarcane 4S decrease In yield Caoparison to up & domn slope Contour cultivation 12X 10X Increase in yield cultivation 34 Contour/much/close planting Taiwan D/WI1 2aX Pineapple 10SX increase in yield Coxparison to up & down slope/wide lanch terrae 72X increase In yield planting. Contour/close planting 660 increase in yield 105 Contouring Taiwan D/i 17X Tea 6K increase in yield Coxparison to clean cuttivated; IHutchins 17X 6K decrease in yield Soil * gravelly Lam Bench terrace 17X 32X decrease in yield ContmLwing 32X 27X increase in yield mllethIr 32X 71K increase In yield ' Bench terrace 32X 16K incresm In yield 5 Contour bund & terraced Indonesia iW//D 9K Potato No significant difference In yield Coarison to contour bunding only; Contour bund & muleh No significant difference in yield solt - Andosol Contmr bund & terraced 12X Cabbage 8X decrease in yield Contour buds mulch No significant difference in yield 180 Furrow d USA SATap NG Sorghum 20K increase in yield Cotton 15K increase in yield 86 Buds India SATr NG Agriculture on significant or stable yield Increase Construction of birds as a soil or moisture conservation practice have not shown, in controlled ICRISAT experiments, to achieve either significant or stable yield increase due to moisture conservation f8 Contmw bnding/level terrace/ India SATr L NG Agriculture Pondad water damaged crops and interfered Comparison to non-treated; Soil ridge-type terree/absorptive- AT with tilltge operations resulting in lowered Vertisol tse terrace yieldo on treated sites 8 years out of 8 years on poorly drained soils a8 BudB India D/il NG Iheat/brteey/ 35K increase in yields Comparison to unbunded. unleveled. LevelIlng grcorn/juar 63U increase in yields Table 8.4 IMPACTS OF SOIL CONSERVATION TECHNOLOGIES ON CROP YIELD AM PSOUCIVITY CITE TECUOLOhGY%TREATMENT LOCATION CLIMATE SLOPE LAW USE INPACT REENKS Sunds & l eveling 9es increas In yields. 112 Lend leeing USA T N Sorgbhu No significant differeme In yield Cear ion to unleteled; Soil sandy clay lo_m fins sand loom. Pt a 614 e yr 2M Lenel pMn USA SIATp grain eorgbM 13OX Increase in y bid, Co _ rEen to plote euteldr For ee UOXsm Iner increI yietd. o pn 125 Len PM USA pSATp a Sorghum 101 Increase in yield Coeprism to untlveled reas Pt * 422 rn/yr - 46 radd ve Canch terrae India D/Wi 251 Potato 1S greter yield an bench terre TI2 Terrace Coltal VIP 4S5 Coffee Der eae In nutrient ltom : N SOS C_rlawison to unterraced Increae in nwtrient loses t P 2501. coffee. K-I* OO. toa- 331. It!- M9 74 Terae USA "TOp U Agrieulture No yield boweits from terracing slam. 42 I emd sloping bwch terrae Indb DM 25X Potato 101 Incre In yield C_prlsen to In_rd sltping beneh dB outard sloping bIc t rr e 9 incrae In yield terrace; lighet yields attributed Pueto tico structural t5rrwe 231 Increase in yield to minimized soil disturbaie on Purto lico vegtative terrae 451 increas In yield Puerto Rico vegtative terrace. * -Pt terraces * structurat or vegtative berriers built an the contour and attloed to fill with 0 iall. usu. 2-4 years to build up terrace; yield differences in 5 year total yield. 76 Cmervatien b terrae USA SATp 1.5X Sorghbtmimwt/ Little or no yield bwwfits from bench Cparison to non-terraced fatllou lveling lone. 75 Cons tin bnh terrace USA SATp 1.51 Sorgh mtl SO1 Inerme In yield an fine or medium No Increases on coarse txtured fellow textured soils with good water holding soils; Solls * silty clay lo_m & caity sBnd 17Conservation bnh terre USA T no Sorghum/wheat/ 27X Increase In yleld over the long term. anch terraces reqpire deep fertile fatllow Yield reductions persisted for 6 years soil an moderate slopes to permit following leveling, topsoil removal necessary for leveling. 124 Cons tion bench terrace USA SATop 1X Sorgha 401 increae In yield Cxarison to unterraced; Soil - silt loam to clay losem Pt a 424 _Iyr 10 Conservation bench terrace USA SATp 2-1 Whseat Topsoil loss due to leveling resulted in Subsoil properties poor.replacem_nt need for high application rates of of 2 inches of topsoil increased fertilizer to sustain yield. yields as wuch or more than highest rates of fertilizer pticatian Table 8.4 IMPACTS OF SOIL CONSERVATION TECHNOLOGIES ON ClOP YIELD AND PRtDUCTIVITY ClITE TECINIOLOGTRItEATEN LOClION CLIMATE SLOPE LAND USE IMPACT REMS over tuo year period. Soll t silt lom. Pt a 325 m yr 69 Cnraevtion bunch terrace USA SATep 1-5X wHeat 19X increem in yield Cowpsrison to unterreced; Solt 1-5X Corn no significant difference in yield silt low_ to silt clay lose 1-S5 aromialfalfa 1001 Increae In yield Pt * 111 §Wyr of study 35 Closed-Fid terrace/graded Inidia D/U 21 Barley/gru Significant yield dacreas i In all years (3) Ceia rison to nan-terraced; Solt terrac linseedlniger and In ell crops sandy clay loew. Pt . 1.170 * yr; attributed to poor dr aim resultfng in higher sol moisture causing poor tilth as a result of tillg oprations In mist clay 135 Sench terrae Indane le WV/ N/h nice `c1 Increase In yield Coeprlisen to crop an ride Sench terrece Peut No significant diffeence in yield terraced ltnd Bench terrae Ceeseva So significant difference In yield 111 Bench terrae I snei W/o N/ Agriculture 3.X - 311 Incrse In yield Crops: bmow prag corn ben 56 Lewe bmeh terrace Ecuaor W/o 4- Corn No significant difference In yield Coqreisn to untorraced Sub terre9 1.eR eutelspa 4-.4 331 increae in yield control; soils * Iose to l_o C= tin hnch t err e 4-43 12X decree in yield sard Steqp beckslep terrce 4-13 11d Incrse n yield sted bae terrae 4-63 Mo significant diffenc In yield Distributors of World Bank Publications ASULNTDA PD111.41N MSX=5 SPAIN CO&ON &S SL A^_IurnplongW Sao= 1drNd m Nm IA. OdsuleC PD.O. aft in A paaiM.u Ca_ios7 Rid. II5, 4b SFoit40 O 1ww"C6 eD.P. 21 DM 1333 Sa,lmAha Il~daM 10 MOROCCO Wrd trodMAX AUSIRIAjUAPAPAANEW GU1NE1W PIRANCIE Sod& j3atludjudigf Mw CAwdideCc3M1 PlJ,SOLOMON SLANDl WO wU. kPb6S&Ew IZ2rMo&wt1M.'Anb mmawced VANUAWU AND WESTRN SAMOA W. .avrndi. COSMIC D.A.SadhalJouris 75114PA& SRI LANRA AND THIE MAIDIVIS USWbdiho ma"d NEiERLAS Ill bdp Mitdbo 3133 GSMANX PWEItLG RA UUC Or boor.Hwfiabw. P.O. 244 ViSih UNO-Vad) P.O. Ilo 4 100 SkrOUEapdAnA. Gadin PappdadvAga 5 n4l EA LR MAcelb AUSIA D.33WnI Cdoub*2 Cad and CM NEW ZEALAND GCbon3i GRSBCI H*: thmy mE Wlb hf v I Sb a swwDE A.0St %ho KEMB PA**E Fyfl 24. IVpd W" Sb Pld). PMM NwMukg P h P._cbdumbq Aihems11635 AadmdA Lr,inSMI129.'16355 Am 'damW. C.UATVMAIA P.O. Omc 31C L11FIm P1dceSak lnS.w. EMbed Fwmsbutd.ud Ib1317TcucSlP S. CluaLl-55 The dweasnafsjoride Waw.p%vWISmA5 ZrnAl Fvdems as 8Da3.D4 NARGIAtwmi Goai..aLty Ibi m5042S Al0Sdly(MDAS) HONG KON4 iACAO NOWAY SWITZRLAND If SIced 16 AMa VW Ld. NNV4 _Wumbact FcrAukiq. 1cmdI RA/uA M _Rlk Rf0 I b*Dwm Ubr* aq Imbat 1KE B.' Sweet N.37 P.0. Om 612SItmA_d 4 nwCu. ?alw Kow Ncim N S OA6 Cmpabas36S Hnxd*s H=6X" CH IGG0ww 15i Nw AhmedSk OMAN 04we",im HVNGARY NMDREB hdamQ SurIo F l ' Kolw. PA.OmIGelSbAkpoft Ismard) P.1 74 ILDA Avrnu P.O. Uam 159 MKMI SwindmASamomw. K1%& 138BudqU cane3312 SEiCIUM INDL Mbime 94l Aan.y PubE.'Imd.Nd)crLUc Ailed Pubehetms liot US. f f,41r TANZANA Av. du Rd DZ 751 Mmt lid P.O. Du N D Odmed UnbPrm 106 0Eo M4adm-000C2 L~E3 p.a am EMS SaAm Da~~~~~~~~~~~~~~IrmMdm P4.Em_T4a C.S34beIG,d.'AM 15 JJN. HawMa MgBdiva dhS,k TJIAItAND So Paeot :amw M Doby40DW Um 3 Sam ama Wa.~ ~ ~ ~ ~~~1341 Balla IgS 010IMS.OPAmdmp IS/14AIdAIVid Banko CANADA Nw D 11lDC NLa Uea TRD & TOSAOO ANlClAU LaflEw.w SA~~~Nonam-1= mkd ok kT UMAD&OSAbATIU LAD_UR 70t ladA_Ww 2ABUDA, UAROD04 C*..UI35nwA=ptm 17?hdtarjnA,m, P.Q am "1 DOMIC.G CMNADA.CGUYANA. Srndmci0 Omfile: Cdlntla.7Wan M*rmlila JAMAICA. MONTISRRAT. ST. 1555B WITS & sVs4. st. UCta jssm Sqodw )fi_tBi ST. VaNC SNT . G WERADINES jayadevaowpumngrsot w c abA 51 Idalc Icefummsd 2mdMa5a OUPAN Sudawl . Hoeme 0-Ut Waucma Cwxqi. b 0D PSRDIe 3.-S1151 KdWwda Cnm Rod TdaM4 Weea lodiA 14mg Hyd-dd.S5lC2 PORTUGAL ibmeA Pabugal TURKE 31OMSUN 1wthm F1010, 21ld Flw ftboI CO74 AIII N_ Th N*ZBAOu U 1M LA" 1 C WA Nl. 449 ApartufoAwme5 Ahucdbebd.NO3P09 Bi. Di. SAUDI ARAUZA. QATAReuedA A COil DiVORW I6SAAdvd aP.O.PDM 3196 UGANA Cmmd'SdMrn dde DOm.'m ELumbo- 22101 R"&a 11401 Usaw"000101hoip AMiedam(CSDA) pjO Bm 714 04 B3. 541 XNONIA MiLubemdm 5w K|au A%14n%In lma Pt lIndira LimLid aa4o P. SA Rada 37 AAN5belt UNmDK ARlA EMIRATES CWRUS Pn. D Eisi A)Caa e I _ MWRRSG YCoa MUE bbISS Ii.mSaw) Jakwa P"a PIMN,mA PJD0o SrW P,O.MIbA7 P.n EM 71S6 WaA Cm AiAS.nrn SPA UN"`SD KINDOOM DENMARK VvdUS.1ScI*O.Iw _10 U O AM b hdW Mk* Ltd. }'~~~~~~~~~CM 5St2AlA- SU Sb^5f PIQ sox D1.mwAJM11 Elli Blam. P.O. Du 300 AUaM d 1.U34 2P0 migm Frodediateg c JAPAN DM DOMINICAN 751C FIuam Book Svce A3 Mwa d Hm Awd s9 URUGUAY 2dMsTdsC.pwA A.4 HMO &CnbDAakYo40123 PfL Uan we hastaWIimdtiI Ubt RSootadb ebeId . CaWl 3d7 TDe ' 5 1 1 6 AIdap.P Fe 52100 Malvr*dm saa,Dm,bi uWrA S AOS.TAIWANMYANMAS. AffiaftokSa.e.(UJLid ~ $1RUNS1I V1EN,1ZU1LA SL SALVADOR P Un 434 wmagm Pdcmm. 4!3uM SL, am hued. Nalcb poli Ud.A1&d* 3m Sd,ad PmKme ArmjkC M 7.4 N 1Ilt Adt% Rind YUGOSLm VIA Sai SUCOP S cOudKU61 XL _ 1owlMM i SIAdiw P.O. PA Om $sZl_66i_~~~~~~~~XMI ,;w&,M YUWYdw Utop SOim.aH obui MlAYPIO P.YouCn11o41A PAX Uso 1410 ACAVTOA J_SImb RECENT WORLD BANK TECHNICAL PAPERS (continued) No. 94 Le Moigne, Barghouti, and Plusquellec, Technological and Institutional Innovation in Irrigation No. 95 Swanson and Wolde-Semait, Africa's Public Enterprise Sector and Evidence of Reforms No. 96 Razavi, The New Era of Petroleum Trading: Spot Oil, Spot-Related Contracts, and Futures Markets No. 97 Asia Technical Department and Europe, Middle East, and North Africa Technical Department, Improving the Supply of Fertilizers to Developing Countries: A Summary of the World Bank's Experience No. 98 Moreno and Fallen Bailey, Alternative Transport Fuels from Natural Gas No. 99 International Commission on Irrigation and Drainage, Planning the Management, Operation, and Maintenance of Irrigation and Drainage Systems: A Guide for the Preparation of Strategies and Manuals (also in French, 99F) No. 100 Veldkamp, Recommended Practices for Testing Water-Pumping Windmills No. 101 van Meel and Smulders, Wind Pumping: A Handbook No. 102 Berg and Brems, A Case for Promoting Breastfeeding in Projects to Limit Fertility No. 103 Banerjee, Shrubs in Tropical Forest Ecosystems: Examples from India No. 104 Schware, The World Software Industry and Software Engineering: Opportunities and Constraints for Newly Industrialized Economies No. 105 Pasha and McGarry, Rural Water Supply and Sanitation in Pakistan: Lessons from Experience No. 106 Pinto and Besant-Jones, Demand and Netback Values for Gas in Electricity No. 107 Electric Power Research Institute and EMENA, The Current State of Atmospheric Fluidized-Bed Combustion Technology No. 108 Falloux, Land Information and Remote Sensing for Renewable Resource Management in Sub-Saharan Africa: A Demand-Driven Approach (also in French, 1 08F) No. 109 Carr, Technology for Small-Scale Farmers in Sub-Saharan Africa: Experience with Food Crop Production in Five Major Ecological Zones No. 110 Dixon, Talbot, and Le Moigne, Dams and the Environment: Considerations in World Bank Projects No. 111 Jeffcoate and Pond, Large Water Meters: Guidelines for Selection, Testing, and Maintenance No. 112 Cook and Grut, Agroforestry in Sub-Saharan Africa: A Farmer's Perspective No. 113 Vergara and Babelon, The Petrochemical Industry in Developing Asia: A Review of the Current Situation and Prospects for Development in the 1990s No. 114 McGuire and Popkins, Helping Women Improve Nutrition in the Developing World: Beating the Zero Sum Game No. 115 Le Moigne, Plusquellec, and Barghouti, Dam Safety and the Environment No. 116 Nelson, Dryland Management: The "Desertification" Problem No. 117 Barghouti, Timmer, and Siegel, Rural Diversification: Lessons from East Asia No. 118 Pritchard, Lending by the World Bank for Agricultural Research: A Review of the Years 1981 through 1987 No. 119 Asia Region Technical Department, Flood Control in Bangladesh: A Plan for Action No. 120 Plusquellec, The Gezira Irrigation Scheme in Sudan: Objectives, Design, and Performance No. 121 Listorti, Environmental Health Componentsfor Water Supply, Sanitation, and Urban Projects No. 122 Dessing, Support for Microenterprises: Lessons for Sub-Saharan Africa No. 123 Barghouti and Le Moigne, irrigation in Sub-Saharan Africa: T7he Development of Public and Private Systems No. 124 Zymelman, Science, Education, and Development in Sub-Saharan Africa No. 125 van de Walle and Foster, Fertility Decline in Africa: Assessment and Prospects No. 126 Davis, MacKnight, IMO Staff, and Others, Environmental Considerations for Port and Harbor Developments The World Bank Headquarters European Office Tokyo Office 1818 H Street, N.W. 66, avenue d'Iena Kokusai Building Washington, D.C. 20433, U.S.A. 75116 Paris, France 1-1 Marunouchi 3-chome L Chiyoda-ku, Tokyo 100, Japan Telephone: (202) 477-1234 Telephone: (1) 40.69.30.00 Facsimile: (202) 477-6391 Facsimile: (1) 47.20.19.66 Telephone: (3) 214-5001 Telex: wui 64145 WORLDBANK Telex: 842-620628 Facsimile: (3) 214-3657 RCA 248423 WORLDBK Telex: 781-26838 Cable Address: INTBAFRAD WASHINGTON DC 11606 WAS 100 0-8213--1606-0 WATERSHED DEVELOPMENT 400000002354 $13 .9 5 Cover design by Walton Rosenquist ISBN 0-8213-1606-0