Watershed Resources Management: Studies from Asia and the Pacific 9789814345880

Citation preview

Watershed Res ourc es Management Studies from Asia and the Pacific

I5EA5

INSTITUTE OF SOUTHEAST ASIAN STUDIES, Singapore

The Institute of Southeast Asian Studies was established as an autonomous organization in 1968. It is a regional research center for scholars and other specialists concerned with modern Southeast Asia, particularly the multi-faceted problems of stability and security, economic development, and political and social change. The Institute is governed by a twenty-two member Board of Trustees comprising nominees from the Singapore Government, the National University of Singapore, the various Chambers of Commerce and professional and civic organizations. A ten-man Executive Committee oversees day-to-day operations; it is chaired by the Director, the Institute's chief academic and administrative officer. The ASEAN Economic Research Unit is an integral part of the Institute, coming under the overall. supervision of the Director who is also the Chairman of its Management Committee. The Unit was formed in 1979 in response to the need to deepen understanding of economic change and political developments in ASEAN. The day-to-day operations of the Unit are the responsibility of the Co-ordinator. A Regional Advisory Committee, consisting of a senior economist from each of the ASEAN countries, guides the work of the Unit.

~~~

EAST-WEST CENTER, Honolulu

The East-West Center is a public, nonprofit educational institution established in Hawaii in 1960 by the United States Congress. The Center's mandate is "to promote better relations and understanding among the nations of Asia, the Pacific, and the United States through cooperative study, training, and research?' Some 2,000 research fellows, graduate students, and professionals in business and government each year work with the Center's international staff on major AsiaPacific issues relating to population, economic and trade policies, resources and development, the. environment, culture and communication, and international relations. Since 1960, more than 25,000 men and women from the region have participated in the Center's cooperative programs. Principal funding for the Center comes from the U.S. Congress. Support also comes from more than 20 Asian and Pacific governments, as well as private agencies and corporations. The Center has an international board of governors. The East-West Environment and Policy Institute was established in October 1977 to increase understanding of the interrelationships among policies designed to meet a broad range of societal needs over time and the natural resources on which these policies depend or which they affect. Through interdisciplinary and multinational programs of research and training, the Institute seeks to develop and apply concepts useful in identifying alternatives available to decision-makers and assessing their implications. Progress and results of Institute programs are disseminated in the East-West Center region through books, occasional papers, working papers, newsletters, and other educational and informational materials.

ISEAS Environment and Development Series

Watershed Resources Management Studies from Asia and the Pacific edited by

K. William Easter John A. Dixon Maynard M. Hufschmidt

15EA5 ASEAN Economic Research Unit Institute of Southeast Asian Studies Singapore

Environment and Policy Institute East-West Center Honolulu

Cataloguing in Publication Data Watershed resources management : an integrated framework with studies from Asia and the Pacific I edited by K. William Easter, John A. Dixon, and Maynard M. Hufschmidt. Rev. ed. (ASEAN environment series) 1. Watershed management. 2. Watershed management--Asia--Case studies. 3. Watershed management--Pacific Area--Case studies. I. Easter, K. William. II. Dixon, John A., 1946III. Hufschmidt, Maynard M. IV. Series. HD1691 W33 1991 sls90-142756 ISBN 981-3035-74-9

The responsibility for facts and opinions expressed in this publication rests exclusively with the authors and their interpretations do not necessarily reflect the views or the policy of the Institute and the Center or their supporters. First edition published 1986 in the United States of America by Westview Press, Inc.; Frederick A. Praeger, Publisher; 5500 Central Avenue, Boulder, Colorado 80301. This revised edition published 1991 by the Institute of Southeast Asian Studies, Heng Mui Keng Terrace, Pasir Panjang, Singapore 0511 in cooperation with the Environment and Policy Institute, East-West Center, Honolulu, Hawaii.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior consent of the Institute of Southeast Asian Studies. © 1986 East-West Center, Honolulu, Hawaii © 1991 Institute of Southeast Asian Studies, Singapore Composition for this book was provided by the East-West Center. Printed by Loi Printing Pte Ltd ·

Contents

List of Figures and Tables Foreword, Charles W. Howe Preface List of Contributors

ix xiii xv xvii

Part I: Framework of Analysis for Integrated Watershed Management Integrated Watershed Management: An Approach to Resource Management, John A. Dixon and K. William Easter Watersheds as a Social and Physical Unit An Integrated Watershed Management Approach to Resource Management The Approach of the Book References

3 4 6 12 15

2 A Conceptual Framework for Watershed Management, Maynard M. Hufschmidt

17

Major Elements of the Framework Application of the Analytical Framework Role of the Analytical Framework References

17 25 30 30

3 Biophysical Aspects in Watershed Management, Lawrence S. Hamilton and Andrew J. Pearce Biophysical Effects of Land Uses Biophysical Information Needed

v

33 33 48

vi

Contents

---------

Conclusion References

49 49

4 Economic Analysis at the Watershed Level, John A. Dixon and K. William Easter

53

An Integrated Economic Analysis Types of Economic Analysis Conceptual Issues within a BCA Framework Implications and Conclusions References

54 58 61 68 69

5 Economic Policies and Watershed Management,

Alfredo Sfeir-Younis

Accounting Unit Economic Policy Environment Intertemporal Questions in Economic Policy Policy Agenda Conclusion References

6 Behavioral and Social Dimensions, George W. Lovelace and A. Terry Rambo The Human Ecology Perspective Structural Parallels between Social Systems and Natural Systems Social and Behavioral Obstacles to Integrated Watershed Management Conclusion References 7 Institutional and Organizational Concerns in Upper Watershed Management, Christopher J.N. Gibbs Watersheds as Secondary Regions Institutions and Organizations Conclusions and Implications for Watershed Management References 8 Program Implementation, K. William Easter Program Implementation Evaluation of Implementation Implementation Issues Conclusion References

71

71 73 75 78 79 79 81

81 83 86 88 90 91

91 93 100 101 103 104

109 112

116 117

Contents

9 The Potential Role of Agroforestry in Watershed Management, Napoleon T. Vergara Agroforestry: A Potential for Upland Watersheds Agroforestry in Upland Watersheds: Constraints and Opportunities References

Part lL' Applications 10 Watersheds in Hawaii: An Historical Example of Integrated Management, Joseph R. Morgan Ancient Hawaiian Society and Land Use Lessons from the Past References

11 Annexation, Alienation, and Underdevelopment of the Watershed Community, Anis A. Dani The Ecology of the Hindu Kush-Himalaya The Process of Incorporation Incorporation of the Watershed Community Restoring the Balance Conclusion References

12 The Role of Extension: A Northern Thailand Watershed Case Study, Peter WC Hoare Northern Thailand Setting The Role of Extension Conclusion References

13 Watershed Management in Indonesia: The Case of Java's Densely Populated Upper Watersheds, David S. McCauley Evolution of Upland Land Use Renewed Attention for Upland Areas Present Trends in Addressing the Problems Conclusion References

vii

119 120 126 129

133 134 142 143

145 147 148 150 154 156 156

159 159 166 172 173

177 179 182 184 187 188

viii

Contents 14 Estimating Erosion Costs: A Philippine Case Study

in the Lower Agno River Watershed, Nicomedes D. Briones

191

The Lower Agno River Watershed The San Roque Multipurpose Project Analysis of the Project Conclusion References

191 194 195 203 204

15 Implications for Integrated Watershed Management, K. William Easter and John A. Dixon A Conceptual Model A Research Agenda Lessons Learned Reference

Index

205 206 208 210 213

215

Figures and Tables

FIGURES 1.1 A natural and social system schematic of a watershed 2.1 The five stages of integrated watershed management 2.2 Watershed management as a planned system 2.3 Generalized watershed management system in physical output terms 2.4 A three-dimensional analytical framework for watershed management 3.1 The hydrologic cycle 3.2 Erosion and sedimentation processes in a watershed

26 36 39

3.3 Average monthly streamflow and the increase in flow during seven years of annual recutting of deciduous forest on Watershed 17, Coweeta, USA

44

3.4 Stylized stream hydrograph response to a storm event 4.1 The private and social perspective of a watershed

45 57

4.2 Typical stream of project benefits and costs 6.1 A simplified model of human-environmental relations in the watershed context 8.1 Problems of implementation in watershed management and the role of implementation research ix

8 18 20 22

58 84 110

x

Figures and Tables

9.1 Natural terraces formed over time by contour hedges or strips in an agroforestry system 9.2 Nutrient losses from, and transfers within, an agroforestry system 9.3 Natural and man-induced nutrient inputs into an agroforestry system 10.1 The radial draining pattern of streams in Kauai and the location of Halelea District with its nine historical ahupuaa 11.1 Highland-lowland relationships in watershed management 12.1 Location of Highland Agricultural and Social Development (HASD) Project sites in the Far North of Thailand 12.2 The Royal Thai Government organizations for hilltribe development with emphasis on the Highland Agricultural and Social Development (HASD) Project, 1980 13.1 Map of Java and Madura indicating the major river systems and areas over 200 m in elevation 14.1 Map of Agno River, Upper and Lower Agno River watersheds in Luzon, Philippines

121 123 124

139 149

160

165 178 192

TABLES 1.1 Definitions of watershed-related terms

5

1.2 Summary of rationale for a watershed approach to rural development

6

2.1 The three major activities of watershed management 2.2 Examples of watershed management tasks required at the planning stage, classified by management activities and management system elements 2.3 Examples of tasks involving implementation tools, classified by stages of the management process and management activities 3.1 Examples of a few resource utilization and management practices to meet two specific watershed management goals

24

27 28

34

--------

-~

~

---

Figures and Tables

~

---

-~-

---

3.2 Some vegetative cover factors for West Africa 4.1 A comparison of financial and economic analysis 4.2 Present value of a future net return of $100 at four discount rates 4.3 Relationship between the goods and services associated with watershed management projects and location 8.1 The watershed management process and the watershed management elements for program implementation 9.1 Nitrogen yields of selected tree legumes 12.1 Records from three problem-solving meetings in Northern Thailand 13 .1 Reported land use of Java and Madura by province in 1980 14.1 Total volume of sediment deposits trapped in the proposed SRMPP Reservoir over a 50-year period, without watershed management 14.2 Total volume of sediment deposited by storage zones through year 50 for proposed SRMPP Reservoir, without watershed management 14.3 Economic loss due to sedimentation of the planned SRMPP Reservoir, without watershed management 14.4 Volume of sediment in the proposed SRMPP Reservoir, with watershed management 14.5 Volume of sediment and value of loss by active storage pools, with and without watershed management conditions 14.6 Present values and benefit-cost ratios of watershed management program under assumed sedimentation rates 14.7 Sensitivity analysis, SRMPP Watershed Management Project 14.8 Comparative economic analysis of mine tailings disposal methods, Lower Agno River Watershed

xi

37 56 63

65

104 124 168 180

196

197 198 199

200

201 202 203

Foreword The impacts of upper watershed land-use practices on resource systems lower in the watershed or river basin have long been recognized as a major problem area: siltation of the river bed, reservoirs, and irrigation systems; impacts on estuarine mangroves and coastal fisheries; increased severity of flooding and drought; and deposition of chemical residues. The watershed is thus comprised of a sequence of linked resource systems, but the linkage is one way. This complicates the integrated management of watershed resources, for not only do different agencies typically have responsibility in different parts of the watershed and for the management of different resources, but the private parties in the upper watershed are not motivated to take into account the costs they impose on the lower watershed. In addition to the traditional problems, the "Green Revolution" is now spreading to the upper watershed: the search is on for ways of increasing the productivity and population carrying capacity of the higher elevation lands. This requires consideration of nontraditional patterns of land use and even greater integration of upstream and downstream considerations. This will require institutional innovation to overcome traditional barriers to interagency cooperation and to manage watershed activities from a basinwide viewpoint. It will require open reviews of interagency rivalries and not just the passing of another law. This volume discusses all of these complexities within a unified conceptual framework, which is followed by interesting case studies. The book should be of great utility to watershed and river basin managers, consultants, and scholars doing further work on the biophysical, economic, social, and institutional aspects of watershed and river basin management. Charles W. Howe General Editor

Charles W. Howe is professor of economics at the University of Colorado, specializing in water resource development and related topics.

xiii

Preface This book is the outgrowth of independent and collaborative work on water resources and watershed management in developing countries at both the East-West Center and the University of Minnesota. Researchers at these institutions have worked extensively on irrigation problems, techniques for planning and evaluating projects, and the biophysical aspects of tropical watersheds. After studying the broad questions of water resource management, it became clear that some important aspects of watershed management research had been neglected. These gaps included the socioeconomic aspects of watershed use and methods for integration of these aspects with the biophysical elements. Three workshops held at the East-West Center helped highlight the effects of watershed mismanagement on soil erosion, slope stability, and channel and reservoir sedimentation. The first of the workshops held in January 1983 focused on how technical information and knowledge can be used to help generate alternative policies for soil and water conservation research and monitoring activities. The workshop participants concluded that one of the greatest problems facing policymakers is the translation of soil and water conservation principles into effective policies. Such translation requires a thorough assessment of the nature, extent, and impact of soil erosion and an evaluation of its effects. The necessary information is often sparse or absent; therefore, the policymaker must rely on strategies that keep options open. The second workshop held in early May 1984 was on the "Effects of Forest Land Use on Erosion and Slope Stability:' Workshop speakers emphasized forest removal and road construction as causes of the serious increase in landslides and accelerated erosion rates. Most of the research on these problems has been done in moist temperate steepland forests in western United States, Japan, and New Zealand; thus, a serious research gap exists regarding tropical forests in much of Asia. At the third workshop held at the East-West Center during the third week of May 1984, the participants presented a wide range of papers on river and reservoir sedimentation and associated watershed management questions (many of which have been published in the December 1984 and March 1985 issues of Water International). Hufschmidt pointed out that there has been no rigorous analysis of the root causes of the failures in implementing watershed management projects. Such evaluations require carefully constructed case studies of the Asian watershed management experience based on an integrated framework of analysis. XV

xvi

Preface

In an effort to stimulate research to help fill some of the gaps identified in the earlier workshops, a fourth workshop was held, with support from USAID, at the East-West Center in January 1985 on "Research for Integrated Watershed Management in Developing Countries:' As a key part of this workshop, six papers were prepared by the Environment and Policy Institute (EAPI) staff working closely together to develop an interdisciplinary approach to watershed management. These papers were revised to provide the framework of analysis presented in this book. We hope that this book captures the exciting interaction that went on as the conceptual papers were being written. Even more important, we hope that this book will provoke discussion and further research into the complicated but fascinating and important area of watershed management. Many people contributed to the creation of this book. Valuable comments on both the substance and organization of the book were received from Prof. Herbert H. Stoevener, Department of Agricultural Economics, Virginia Polytechnic Institute and State University; Prof. Charles W. Howe, Department of Economics, University of Colorado; and Prof. Samir A. El-Swaify, Department of Agronomy and Soil Science, University of Hawaii. We owe special thanks to many colleagues at the East-West Center: Helen Takeuchi, EAPI editor, and Joan Nakamura and Betty Schweithelm, typists, carefully handled the many revisions of the manuscript. Louise Fallon, research fellow, helped edit and rework several chapters. The authors of the various chapters responded quickly to the suggestions of the editors for revisions and requests for additional material. This book was originally published in 1986 by Westview Press, Boulder, Colorado, as a volume in its "Studies in Water Policy and Management;' Prof. Charles W. Howe, general editor. This edition contains all of the text of the original edition with the exception of one chapter from "Part II: Applications" section of the book. K. William Easter John A. Dixon Maynard M. Hufschmidt

Contributors Nicomedes D. Briones

East-West Center Honolulu, Hawaii

Anis A. Dani

International Center for Integrated Mountain Development (ICIMOD) Kathmandu, Nepal

John A. Dixon

East-West Center

K. William Easter

University of Minnesota St. Paul, Minnesota

Christopher J.N. Gibbs

East-West Center

Lawrence S. Hamilton

East-West Center

Peter W.C. Hoare

Chiang Mai University Chiang Mai, Thailand

Maynard M. Hufschmidt

East-West Center

George W. Lovelace

East-West Center

David S. McCauley

East-West Center

Joseph R. Morgan

University of Hawaii and East-West Center

Andrew J. Pearce

Forest Research Institute Christchurch, New Zealand

A. Terry Rambo

East-West Center xvii

xviii

Alfredo Sfeir.:Younis

World Bank Washington, D.C.

Napoleon T. Vergara

East-West Center

The Editors K. William Easter is professor of agriculture and applied economics at the University of Minnesota. John A. Dixon is research associate at the Environment and Policy Institute at the East-West Center, Hawaii, where Maynard M. Hufschmidt is a senior fellow.

PART I

Framework of Analysis for Integrated Watershed Manag ement

CHAPTER 1

Integrated Watershed Management: An Approach to Resource Management John A. Dixon and K. William Easter

Watersheds, and the proper management of them, have become a major focus of resource managers in countries around the world. Much of this interest is the result of land-use practices that have led to increased soil erosion. Recently there has been an increasing number of reports warning of high levels of soil erosion and deterioration of major watersheds (Bowonder eta!. 1985; Brown 1981; Eckholm 1978). Sediment is building up in reservoirs and streambeds resulting in reduced irrigation and power production while increasing the incidence and severity of flooding. For example, Bowonder et a!. report a 60 percent loss in the storage capacity of the Nizamsagar Reservoir in Andhra Pradesh due to severe soil erosion in the upper watersheds. The actual rate of sedimentation was 25 times the original assumed rate. The impact of such soil erosion is felt by rural people throughout the watershed through reduced incomes as well as inadequate supplies of wood and clean water. These watershed problems are especially acute in developing countries where growing populations are exerting intense pressure on increasingly scarce land and water resources. Most people in these areas live and work on the land. So, as rural populations increase, lands formerly farmed extensively are now being farmed more intensively, while formerly fallow lands, usually more susceptible to erosion, are being cultivated. This, in combination with similarly motivated overuse of grazing lands, has dramatically increased the potential for erosion and downstream damages. Deforestation of upland areas, a result of more intensive shifting agriculture and excessive timber extraction, has also accelerated soil erosion and downstream damages. Although the immediate cause of concern-increased soil erosion and associated downstream problems-is a physical factor, these watershed problems result from a mixture of biophysical, economic, social, political, 3

4

John A. Dixon and K. William Easter

and institutional factors. Some consequences of these factors are reduction in the productivity of forests, fisheries, and agricultural and grazing lands; decreased returns from investments in hydroelectric power generation and irrigation projects; and losses of property and impairment of human health. The practices required to reduce these losses by protecting watersheds from degradation are well recognized. However, the management of watersheds has been largely unsuccessful, partly because the concentration has been almost exclusively on biophysical aspects such as slope, soil texture, and vegetative cover, without proper regard for socioeconomic aspects. Although important gaps do exist in our knowledge of biophysical watershed processes, particularly in the case of watersheds located in the tropics and subtropics, fairly accurate biophysical input-output approximations can be made in many situations. The difficult task is to interpret and apply the biophysical information to questions dealing with overall project planning and implementation. For successful project implementation, economic, social, political and institutional considerations are paramount. The socioeconomic side, although known to be important, is often ignored because of the difficulties involved in dealing with social issues (Blaikie 1985). The level of concern about watershed management is evident by the wide variety of programs and institutions actively involved in the study of and management of watersheds. In Asia, for example, the recently established International Center for Integrated Mountain Development (ICIMOD) located in Kathmandu, Nepal, has a primary focus on the management of intensively used upland watersheds. The U.S. Agency for International Development (USAID) has funded many watershed projects in Thailand, Indonesia, and the Philippines, as well as provided funding for the Association of Southeast Asian Nations (ASEAN) Watershed Project. The Food and Agriculture Organization of the United Nations (FAO) has funded watershed management activities in many Asian countries including Afghanistan, Pakistan, India, Nepal, Burma, Thailand, Indonesia, and the Philippines. Most of these rural development efforts have been concentrated in upland areas. However, other rural development activities fit equally well within the watershed framework. WATERSHEDS AS A SOCIAL AND PHYSICAL UNIT This growing level of concern for watersheds is prompted by many factors, biophysical and social as well as economic. Watersheds vary tremendously in size, ranging from 100 km' to ones that include large parts of entire countries. Table 1.1 defines many watershed-related terms; in this book the term watershed refers to a drainage subarea of a major river basin. Traditionally watershed management was viewed as a biophysical, engineering problem complicated by the presence of people. More recent

An Approach to Resource Management

5

Table 1.1 Definitions of watershed-related terms A watershed is a topographically delineated area that is drained by a stream system. The watershed is a hydrologic unit that has been described and used both as a physical-biological unit and as a socioeconomic and sociopolitical unit for planning and implementing resource management activities. When the term watershed is used in this book, it refers to a subdrainage area of a major river basin. A river basin is similarly defined but is of a larger scale (for example, the Mekong River Basin, the Amazon River Basin, and the Mississippi River Basin).

Integrated watershed management is the process of formulating and implementing a course of action involving natural and human resources in a watershed, taking into account the social, political, economic, and institutional factors operating within the watershed and the surrounding river basin and other relevant regions to achieve specific social objectives. Typically this process would include (1) establishing watershed management objectives, (2) formulating and evaluating alternative resource management actions involving various implementation tools and institutional arrangements, (3) choosing and implementing a preferred course of action, and (4) thorough monitoring of activities and outcomes, evaluating performance in terms of degrees of achievement of the specified objectives. The watershed approach is the application of integrated watershed management in the planning and implementation of resource management and rural development projects or as part of planning for specific resource sectors such as agricultural, forestry, or mining. Imbedded in this approach is the linkage between uplands and lowlands in both biophysical and socioeconomic contexts. Source:

Easter and Hufschmidt 1985.

thought considers watershed management as an integrated process whereby a natural resource is managed in conjunction with human use to produce a series of goods and services. Watershed management must explicitly recognize the range of physical, social, economic, and political factors that result in the observed patterns of use. Even when people are explicitly included in the analysis, the differing perspectives of lowland and upland residents in a watershed must be considered. As discussed at some length in Chapters 7 and 11, upland areas and upland residents were frequently regarded as those that lowland people "do things to?' This dichotomy has been described by various theories in which the two areas have been given labels such as core-periphery, primary-secondary, or center-hinterland. The idea is essentially the same. The proposed solutions to watershed management problems frequently have been designed by lowland residents and then imposed on upland resource users. This approach to management has largely been unsuccessful.

6

John A. Dixon and K. William Easter

AN INTEGRATED WATERSHED MANAGEMENT APPROACH TO RESOURCE MANAGEMENT This book proposes that an integrated watershed management approach be used to understand the range of factors affecting resource use and development in many watersheds. That watershed boundaries are appropriate boundaries within which resource management and development can be conducted is not a new idea. Much of the river-basin planning and small watershed development efforts in the United States were based on a similar concept. As mentioned earlier, the approach also is being applied in developing countries. The rationale for a watershed approach to rural development projectswhether it is directed toward agriculture, forestry, rangelands, or water resources-is that the approach has a strong biophysical and economic logic (Table 1.2). There is a certain natural imperative about gravity. Water flows downhill and, in a watershed, this creates a strong undirectional dimension to the cause-and-effect relationships. Soil, nutrients, and agricultural chemicals are all transported in the water medium. People may move in either direction depending on various influences, such as employment and trade opportunities, but their movement is likely to be influenced, at least partially, by the topography of the watershed.

Table 1.2 Summary of rationale for a watershed approach to rural development I.

The watershed is a functional region established by physical relationships.

2.

The watershed approach is logical for evaluating the biophysical linkages of upland and downstream activities because within the watershed they are linked through the hydrologic cycle (see Figure 1.1).

3.

The watershed approach is holistic, which enables planners and managers to consider many facets of resource development.

4.

Land-use activities and upland disturbances often result in a chain of environmental impacts that can be readily examined within the watershed context.

5.

The watershed approach has a strong economic logi-c. Many of the externalities involved with alternative land-management practices on an individual farm are internalized when the watershed is managed as a unit.

6.

The watershed provides a framework for analyzing the effects of human interactions with the environment. The environmental impacts within the watershed operate as a feedback loop for changes in the social system.

7.

The watershed appoach can be integrated with or be part of programs including forestry, soil conservation, rural and community development, and farming systems.

Source:

Easter and Hufschmidt 1985.

An Approach to Resource Management

- - - - - - - - - - --------

7

The uplands and lowlands within the watershed are linked through the hydrologic cycle. Many of the impacts of land-use activities on ecosystems (both upstream and downstream) are related to changes in hydrologic processes caused by the land-use activities. Because of these physical and hydrological reasons, the watershed is a naturally occurring geographic area that is reasonably well defined and comprises a "neat" unit for planning and analysis. When a resource management action is applied, most benefits and costs will accrue to people and communities within the watershed and adjacent coastal areas. Many of the impacts, which would have been external to traditional project analyses, are accounted for when the watershed approach is used. This interrelationship of physical and social forces in a watershed is seen in Figure 1.1. Since the watershed is a subunit nested within the larger river basin, however, some impacts may still be external when the watershed framework is used. In most cases, however, the major externalities can be accounted for within the watershed framework without the problems related to the unwieldy scale of the river basin. Thus, by broadening the approach to the point where the focus is on the watershed, one can generally improve program planning and implementation of development efforts. This is particularly true where actions motivated by individual interests conflict with the biophysical imperatives for sustainable use of the watershed. When the physical and economic aspects of the problem override the administrative and institutional impediments that may exist, then the watershed is a good management unit.

Boundary Questions The traditional Hawaiian land management system, discussed in Chapter 10, is one of the few examples where the two sets of boundaries coincide. The fact that the administrative and political boundaries are almost always different from watershed boundaries creates ·several problems. First, watershed programs may cut across several administrative units or even countries, and the success of the program depends on coordination among these different units or countries. Second, in other cases only part of an administrative unit is covered by a special watershed program. Residents who are part of the unit or district but not in the program will complain that they should also have program benefits. Third, economic and demographic information and data are collected on the basis of administrative or political units. The first two problems occur during program implementation and should not affect planning. Watershed programs can be planned on a watershed basis but may be administered using different boundaries. However, to do this will require some overall responsibility for coordination as discussed in Chapter 8. In many instances the watershed management framework can be used if the administrative structure and institutions can be adjust-

8

John A. Dixon and K. William Easter NATURAL SYSTEMS

SOCIAL SYSTEMS

Upland Forest (rate ot clearing l

Reservoir Marg1ns (soil erosion • ! )

Lowland Alluv1um (product1v1ty ! o)

Estuary (product1v1ty •

F1shermen (population dens1ty j i) Short term

OCEAN lncreasmg

Long term

I

About the same

Figure 1.1 A natural and social system schematic of a watershed (Source: Hamilton and King 1984).

ed to fit more closely the watershed management needs. For example, in the Mae Chaem watershed project in Northern Thailand, overall administration and coordination of the project was shifted from Bangkok to the governor of Chiang Mai to provide better representation for local interests. In other cases, watershed districts have been established to coordinate the various land- and water-related agencies and activities within the watershed.

An Approach to Resource Management

9

In cases where the necessary adjustment or coordination cannot be accomplished, watershed boundaries will be inappropriate for some development projects but are still useful in the planning stage. The third problem makes planning difficult since economic data are generally collected for a county or district that does not fit watershed boundaries. Thus data must be adjusted before they can be used for watershed planning and analysis-sometimes a difficult task. Yet if data are collected on a disaggregated basis such as suggested in Chapter 11, then it can be aggregated to the watershed level. Despite these problems, the basic premise of this book is that watersheds, analyzed and managed in an integrated manner, are a useful tool for planning and implementing rural development efforts. The emphasis of this book is on management of both a biophysical resource and the interaction of that resource with human needs. The question, therefore, is how the resources within a watershed should be used to obtain a socially optimal level of production over time.

Socioeconomic Issues Within the integrated watershed management approach, there are a set of issues concerned with program management in which economic, social, political, and institutional considerations are paramount. The emphasis on these socioeconomic questions, particularly as found in upland watershed areas, is a common thread running throughout the book. Economics. The primary focus of economic analysis of watersheds has been on project and practice appraisal. These appraisals have been used to help guide managers and planners in their selection of management actions and practices to stabilize upland areas. Research shows that private incentives are inadequate to obtain farmer adoption of many practices that have a significant part of their benefits in the distant future or where the benefits are mostly downstream. This means that some tree planting and terracing practices will require subsidies or public installation if they are to be widely used in upper watersheds. The basic economic problem is the difference between the small-scale farmer's need for short-term soil exploitation and society's desire to maintain long-term soil productivity and reduce downstream damages. Some of the practices that offer the best physical protection may not pass the "social economic" test. Even with low discount rates and inclusion of downstream benefits, the costs still exceed the benefits. In these cases alternative practices will need to be developed, which at least offer a reasonable social rate of return. These practices will probably involve less soil protection but offer higher short-run benefits to the farmer with opportunities to sustain or increase food production. Complete soil protection is rarely economically rational because of the high costs involved.

10

John A. Dixon and K. William Easter

To determine whether a practice or project will have a high social rate of return is not an easy task. The most difficult part of the analysis is to value the downstream damages prevented by the alternative management actions. This is one of the serious gaps in information concerning conservation practices for upper watersheds. At the national level, watershed management will be subject to the influences of economic policies concerning trade and commodity prices. These policies can have either a positive or a negative effect on attempts to reduce soil erosion and improve watershed management. For example, a policy to increase log exports may well aggravate watershed problems by stimulating timber harvesting, unless it is coupled with a reforestation program. Import taxes on energy commodities will increase the demand for a country's fuelwood and charcoal, again increasing tree harvesting (Nelson and Cruz I985). Sociopolitical concerns. From a technical standpoint, many alternative physical measures can be used to reduce soil erosion in upland tropical and subtropical watersheds. However, since the sociocultural aspects of many of these options have not been studied, their potential for adoption is highly uncertain. In addition many practices or programs may have unexpected consequences because of the sociocultural impacts. For example, will a road cause migration rather than open up markets for agricultural products? Are certain practices beyond the villagers' experience and therefore require technical assistance and training to be implemented? If so, are trained personnel available to provide the technical support and education of the kind recommended in Chapters 11 and 12? These and other questions will have to be answered as technical alternatives are considered for dealing with different watershed problems. Finally, as discussed in Chapter 8, one of the most difficult factors to assess is the political experience and commitment of the administration to improve watershed management. The potential for competition and noncooperation among agencies and ministries is high in watershed management. For example, in Indonesia no less than seven ministries are involved in watershed management. Thus, it takes both political experience and commitment to the watershed approach to overcome this potential for competition and conflict. An additional political problem involves the conflicting uses of the resource in an upland watershed by different groups including forestry firms, those practicing extensive agriculture, and those practicing intensive agriculture. The forestry firms want to extract timber as cheaply as possible with little concern for soil erosion. The upland community wants to continue to practice slash-and-burn agriculture and has limited resources to spend on conservation practices. The intensive agricultural producers are usually the ones most damaged by the soil erosion so they have an in-

An Approach to Resource Management

11

centive to reduce soil erosion. However, they may not be in a position to do much about it since much of the erosion may occur upstream. A closely related issue is the role of farmers in watershed management. One of the conclusions from socioeconomic research is the need for farmer or user participation in watershed management (Cernea 1985), meaning farmers have a role to play in both program planning and implementation. "Top-down" approaches to watershed management, which ignore local communities and farmer incentives, are typically failures. Central governments in many developing countries lack capacity for sustained local action and can only adapt in limited ways to local conditions. As illustrated in Chapter 11, there are serious problems in obtaining local participation. Upland areas in much of Asia are administered by forestry departments, which consider the occupants as squatters who are interfering with their management activities. Foresters have tended to resolve disputes by exercising police power. Thus, in many areas, introducing a more participatory approach to upper watershed management will be a difficult task. Successful soil and water management programs and policies must focus on the many land users (farmers, graziers, miners, and foresters) whose actions directly affect soil loss rates. Policies must be translated through the various levels of government to influence these many individuals who have different objectives and priorities-a challenging task for any developing country. Institutions. Property rights are unclear in many upland areas. Farmers do not have secure claims over the long-term benefits of their actions. The incentives to invest in practices that conserve land and water resources do not exist. Most uplands and forests are usually defined as public lands, and the occupants have limited rights to the lands. And, as pointed out above, many occupants of public lands are considered by the government to be farming illegally. Yet this does not mean that private property rights are the only alternative. "Privatization" or central government ownership of upper watersheds do not offer simple solutions to watershed management problems. This is particularly true when large externalities (off-site damages) are involved, and the government has neither the personnel nor funds to implement any intensive management program. Many communities have used common property rights effectively to manage upland resources. In developing countries, common property resource management may be the rational response to spreading risk and reducing uncertainty for the individual farmer (Runge 1983). The literature on common property resource management in upland watersheds is limited but shows that collective management is capable of providing stable and productive resource use over a long time period (McKean 1984). These studies are primarily of upland grazing and for-

12

John A. Dixon and K. William Easter

estry management and show that no one property arrangement appears to be best. The whole range of property rights needs to be considered to determine which ones best fit the conditions of a particular watershed. Yet in Asian watersheds, where a large number of small landholders are involved, the individual decisions may have to be supplemented by some community group as the focus of watershed management decisions. One of the key components in developing institutions for watershed management is to devise institutions that minimize the transactions costs of collective action. For example, there will be a variety of transactions costs involved in coming to an agreement concerning who should pay for reducing the sedimentation rate in a reservoir, including information, administration, and enforcement costs. Devising the appropriate set of property rights can help minimize these costs. THE APPROACH OF THE BOOK This book presents an introduction to watershed management using an integrated, multidisciplinary approach. The approach is based on several key assumptions: • Watersheds need to be managed to meet a variety of goals; these goals are social and economic, as well as biophysical. • People are an integral part of the problem of watershed management and therefore are a key component of any successful solution. • Watersheds as a physical unit provide a useful analytical framework or boundary of analysis. • Prevailing institutions and organizations significantly influence the effective implementation of watershed management strategies. • Economic analysis can be a powerful tool in understanding the complex interactions that occur in a watershed. While not providing "the" answer, economic analysis is a necessary part of any watershed management plan. Given these assumptions, this book stresses the economic, social, and management dimensions of watershed management. Although biophysical forces are of utmost concern, they are well covered in the extensive scientific literature on soil conservation, soil erosion, and land use. Chapter 3 highlights some of the key biophysical factors involved.

The Intended Audience In-depth coverage of all factors needed to perform an integrated watershed analysis is not possible. Rather, this book presents a rationale for an integrated analysis and an operational approach, based partly on economic analysis. The chapters are written to introduce the basic concepts and concerns to a variety of individuals including university students, government

An Approach to Resource Management

13

planners, and resource managers or analysts. An economist will benefit from the discussion of sociocultural and biophysical factors. Similarly, the professional forester or soil expert will be introduced to the basic concepts of economic analysis. The objective of the book is to help each individual involved in watershed management projects to better understand the range of important interrelated factors and thereby to work more effectively as part of a management development team. The book is not designed to make the economist into a soil scientist or turn the forester into a sociologist.

The Organization of the Book This book is organized into two sections. Part I contains nine chapters that introduce the proposed approach and some of the key factors required for an integrated analysis. Following this introduction, Chapter 2 presents a conceptual framework for analysis of watershed management programs. This framework is based on a management model that includes three decision-making dimensions-watershed management as a process, as a planned system of management, and as a set of linked activities and tasks. While the conceptual framework sets the broad management context, the next five chapters examine various dimensions of integrated watershed management. Chapter 3 discusses key biophysical relationships involved with land and water resources in a watershed. Chapters 4 and 5 cover economic aspects of watershed analysis. Chapter 4 focuses on project analysis and micro considerations, while Chapter 5 examines the effects of macroeconomic policies on watersheds and vice versa. Behavioral and social dimensions of different patterns of resource use within watersheds are presented in Chapter 6; this chapter discusses the crucial impacts that social factors can have on resource management. Chapter 7 discusses the institutional and organizational factors that influence how people relate to their environment and to new programs or projects. The combination of all factors presented in Chapters 3 to 7 affect the implementation of an integrated watershed management plan. Plan implementation is discussed in Chapter 8 and is seen as a "process" that is even more important than the planning "process:' Chapter 9, the last chapter in Part I, discusses agroforestry, one biophysical technique frequently used as part of watershed management plans. The discussion highlights the biological, economic, and social factors that come into play when implementing agroforestry components of watershed management plans. The chapter illustrates how any disciplinary approach, or biophysical technique, is likely to fail if examined in isolation from the range of factors influencing watershed management. Part II of the book presents a series of case studies that illustrate many of the points discussed earlier. The studies are drawn from five different

John A. Dixon and K. William Easter countries or regions and include historical accounts as well as reports of current activities. Chapter 10 presents the historical example of integrated watershed management practices in Hawaii. Based on physical landforms, the Hawaiian management plans recognized the interdependence of human needs and biological forces in a watershed. Although organized on an autocratic basis, the ancient Hawaiian watershed management system allowed sustainable resource use that met basic human needs. The next two chapters consider different aspects of involving residents of upper watersheds in management of the resource base. Chapter 11 discusses the process of ''incorporation'' of upper watershed society into national societies. Incorporation has frequently occurred in a destructive fashion, which the author describes as the processes of annexation, alienation, and underdevelopment. These processes break the close socialphysical links that previously existed as watershed communities managed their resources. Chapter 12 examines an upland watershed in Northern Thailand, and the important role that extension can play in involving the watershed community in the development process. The case study illustrates how extension workers and upland farmers can work together to develop practical watershed management plans. The carrying capacity of land (i.e., the density of population it can support) is a function of availability of resources. Lowland rice areas may have three times as many people per cultivated hectare as an upper watershed, and yet the upland area may be the area under more severe population pressure. Chapter 13 presents a case study of watershed management problems in the upland area of Java. Although Indonesia's densely populated island of Java has access to upper watershed areas, the sheer press of people on the land and the lack of realistic alternatives make watershed management especially challenging. The last case study, Chapter 14, presents an economic analysis of management options for the upper watershed of a large multipurpose dam in the Philippines. The study examines the benefits and costs of alternative management options and quantifies some of the economic benefits of reduced soil erosion. The concluding chapter derives implications for integrated watershed management from the espoused analytical framework and the case studies. The clear message that comes through is that although it is essential to do correct biophysical, economic, and social analysis, success in actual program implementation is frequently dependent on institutional and organizational factors. 14

An Approach to Resource Management

15

REFERENCES Blaikie, P. 1985. The Political Economy of Soil Erosion in Developing Countries. New York, NY: Longman Group. Bowonder, B., K.V. Ramana, and T. Hanumantha Rao. 1985. Management of watersheds and water resources planning. Water International 10:121-131. Brown, L. 1981. Eroding the basis of civilization. 1 Soil and Water Conservation 36(5):255-260. Cernea, M.M., ed. 1985. Putting People First (Sociological Variables in Rural Development). Washington, D.C.: The World Bank. Easter, K.W., and M.M. Hufschmidt. 1985. Integrated Watershed Management Research for Developing Countries. East-West Center Workshop Report, Honolulu, Hawaii. Eckholm, E.P. 1978. Losing Ground: Environmental Stress and World Food Problems. London: Pergamon Press. Hamilton, L.S., and P.N. King. 1984. Watersheds and rural development planning. In Traditional Life-Styles, Conservation and Rural Development, ed. J. Hanks, 80-86. IUCN Commission on Ecology Papers 7. Gland: International Union for the Conservation of Nature and Natural Resources. McKean, M.A. 1984. Management of traditional common land (iriaichi) in Japan. Paper for Board on Science and Technology for International Development, National Research Council. Nelson, G.C., and W. Cruz. 1985. Macro policies and forestry. Paper presented at the A/D/C-JCIE Forestry Seminar, Sapporo, Japan. Runge, C.F. 1983. Common property and collective action in economic development. Paper for Board on Science and Technology for International Development, National Research Council.

CHAPTER 2

A Conceptual Framework for Watershed Management Maynard M. Hufschmidt

The framework presented here is an adaptation to the specific case of watershed management of a broader conceptual framework developed for water resources management (Bower and Hufschmidt 1984). Although the framework is not the only valid way to conceptualize watershed management, it provides a useful analytical approach that encompasses the wide range of issues confronting an interdisciplinary team engaged in watershed management. This chapter presents an outline of the conceptual framework. Various aspects of the framework are elaborated on in subsequent chapters in Part I of this book. Not all aspects of the conceptual model will be important in every watershed management program. However, ignoring the problems associated with project implementation and institutional arrangements is the common weakness of many of the existing watershed management efforts. MAJOR ELEMENTS OF THE FRAMEWORK The framework includes three dimensions, each representing a different analytical approach to watershed management; yet, each one of them is also related to the two other approaches. Individually none is adequate but, when combined, they provide a comprehensive picture of watershed management. They are: I. Watershed management as a process involving separate but closely linked stages of planning and implementation. 2. Watershed management as a planned system of management measures and implementation tools applied to a watershed through a set of institutional and organizational arrangements. 3. Watershed management as a set of linked activities for which specific management tasks are required. 17

18

Maynard M. Hujschmidt

-----------, PLAN FORMULATION

J DESIGN

I

I

I

I

I

I I I

Monitoring

_____ _] 1·---

+

1 I

I I

I I

I I I I I

INSTALLATION

I

Monitoring

I

_____ _j 1·--I

1

I I

~

I

I

I

1----. I I

I

I I I

PLANNING

OPERATION Monitoring

_____ j 1·--1

I I I

I

+

8

IMPLEMENTATI ON

I

I

I

I

I I

I

I I

1·---

MAINTENANCE Monitoring

____ _j

I

-----------.INF ORMATION FLOWS

Figure 2.1 The five stages of integrated watershed management.

The Process of Watershed Management The conventional delineation of watershed management as sequential steps of planning, design, installation, operation, and maintenance-w ith monitoring and feedback of information to earlier steps of the processis adopted for the initial dimension of the framework (Figure 2.1). This

A Conceptual Framework

19

delineation conforms to the realities of many water and land resource management activities in developing countries where the project is the major decision unit. In the classical case in a developing country, watershed management begins with the perception of a problem such as deterioration of land and water resources, or a management task for a specific resource such as forests, which leads to a decision to prepare a project or program plan. Plan formulation leads to the preparation of a project feasibility report that provides the basis for a decision to implement the project. The planning stage is completed with the design of the project. Typically, implementation begins with installation of watershed resource utilization and management practices, often involving substantial capital expenditures. The first stage of implementation, which may involve detailed designs, may take several years and is followed by an ongoing program where the land resources and facilities are utilized and maintained. This may continue indefinitely or at least until some major natural or social perturbation requires a change. There is no conceptual reason why planning should be a separate step that is far removed in time from implementation. Planning and implementation can proceed more or less in tandem, with information gained in implementation fed back promptly into planning. Yet this rarely seems to be the case for watershed management projects in developing countries, partly because plans must be approved before a project can be authorized. For example, implementation of the Mae Chaem Watershed Development Project on 4,200 km' of upland, used primarily for agriculture in Northern Thailand, was preceded by seven years of planning (Roth eta!. 1983). Observing watershed management as a process helps us to understand how planning and implementation must be meshed if management is to be effective. For example, lessons learned from monitoring performance during the implementation stage would be used to make appropriate revisions in management programs.

Watershed Management as a Planned System In the second dimension of the framework, watershed management is seen as a planned system of (1) resource management actions including land uses, on-site resource utilization and management practices, and off-site management practices; (2) implementation tools for putting management measures into effect through public and private actors; and (3) a set of institutional and organizational arrangements within which implementation proceeds (Figure 2.2). Two important points can be made by such an observation of watershed management. The first is the clear distinction drawn between management actions---.! 'things to be done' '-and implementation tools---.! 'ways of getting things done?' Watershed management planners often concentrate on formulating alternative packages of resource management actions without

N 0

RESOURCE MANAGEMENT ACTIONS

IMPLEMENTATION TOOLS

• Major land-use ass1gnments • On-site agricultural. forestry, grazing, mining, and other resource utilization management practices • Off-s1te management pract1ces

For each category of management measure: • Regulations • Licenses. permits, and fines • Pr1ces, taxes, and subsidies • Loans and grants • Technical help • Education and information • Direct installation by public agencies

+

INSTITUTIONAL AND ORGANIZATIONAL ARRANGEMENTS

+

For each category of management measure: Nonorganizational: • • • •

Tenure systems Legal codes Economic policies Informal arrangements

Organizational (public and private agencies): • Planning and management • Extension services • Credit agencies

Figure 2.2 Watershed management as a planned system.

A Conceptual Framework

21

formulating alternative packages of implementation tools to an equivalent level of detail, including determination of who is to take resource management actions and who is to apply implementation tools. This often results in neglect of the problems of implementation at the planning stage of watershed management. The second point is that institutional and organizational arrangements are included within the scope of the planned system. Again, this serves to highlight the key role of institutions and organizations in the success or failure of implementing watershed management plans. Since both institutions and organizations are political entities, politics plays a major role. A typical situation encountered in many Asian countries is the fragmentation of watershed management activities among public sectoral agencies and private enterprises. Thus, upstream watershed management responsibilities may be lodged in the Ministry of Forestry, but usually may be shared with the Ministry of Agriculture. The ministries of Irrigation, Energy, or Public Works also may have major responsibilities in specific cases involving watersheds above reservoirs. Other governmental entities that may be involved include those concerned with industry, land reform, rural development, and local government. Added to this are the many private, individual, and nongovernment al cooperative groups that are dependent on watershed resources for their sustenance. Clearly, effective watershed management requires that problems and issues of implementation , institutions, and organizations be addressed at the planning stage along with the physically oriented management measures. This second dimension of watershed management highlights this requirement, which is considered further in Chapters 7 and 8. In a somewhat broader context, in physical output terms only, watershed management can be seen as a planned system that uses management inputs along with natural inputs to produce outputs of useful goods and services, with consequent on-site and off-site effects on natural systems (Figure 2.3). From an economic viewpoint, the watershed management system is a particular kind of production process in which economic costs are incurred by use of management and natural inputs, and economic benefits are gained from the resultant outputs. Depending on the nature of the resulting on-site and off-site natural systems effects, additional economic costs or benefits are generated. Thus, Figure 2.3 highlights the principal elements of a benefit-cost analysis. The figure also distinguishes between the on-site and off-site natural systems effects. To the extent that watershed management reduces adverse onsite effects on soil and vegetation such as soil and nutrient losses, it will have a positive effect on maintaining or increasing outputs. Most of the economic benefits of improving on-site conditions will be in the form of maintaining or increasing levels of on-site outputs.

N N

MANAGEMENT INPUTS Labor, materials, energy, equipment, and management skills for planning, design, installation. operation, and maintenance

OUTPUTS NATURAL INPUTS Land and water resources. cl1mate

THE WATERSHED MANAGEMENT SYSTEM • Resource management actions • Implementation tools • Institutional and organizational arrangements

• Agricultural crops • Forest products • Livestock products • Minerals

• Fish • Recreation • Water

NATURAL SYSTEMS EFFECTS

r------------------,

I On-Stte: 1 : Changes m the state of the system 1 as a result of mass wastmg, soil 1 1 and nutrient losses, and changes m I water yieldmg capacity I L - - - - - - - - - - - - - - - ____ J Off-Site: Changes in time pattern of streamflows and ground-water flows; channel and reservoir sedimentation. channel degradation; water quality

Figure 2.3 Generalized watershed management system in physical output terms. Note: This schematic can be used to depict a system in the planning, design, installation, or operation stage.

A Conceptual Framework

23

Measurement of off-site (downstream) effects on natural systems may require an extended physical and economic analysis, including impacts beyond the physical boundaries of the management area (e.g., coastal fisheries). The extended analysis would also include inter-farm effects involving the impact of erosion and sedimentation on neighboring farms, roads, houses, and other rural infrastructure. Thus, the objective of watershed management is to maximize the net socioeconomic benefit of land-use activities in the watershed. Specific targets are difficult to formulate without reference to a particular watershed. However, the general idea is to improve upstream conditions to maintain or increase existing levels of on-site outputs, while minimizing the adverse downstream consequences of these land-use activities. In Chapter I the watershed was defined as a drainage subarea of a river basin. As a result, determination of what are on-site and off-site effects depends on how the boundaries of the watershed system and its subsystems are delineated. To the extent possible, boundaries should be established so that all major effects are internalized within the system. Establishment of the appropriate boundaries for a watershed management program is a crucial, early step in the management process.

Activities and Tasks of Watershed Management Watershed management can be subdivided for analytical purposes into many specific steps that watershed management agencies or other actors must perform to produce the desired outputs and effects on the natural systems. These steps can be identified by analyzing watershed management as a set of linked activities for which specific management tasks are required. For example, as a first activity, the entire watershed area must be subdivided into various types of existing or prospective major land uses: agriculture, grazing, agroforest, commercial forest, protection forest, mining, transportation, urban and aquatic (Table 2.1, Panel I). For each land type, there may be multiple uses. To illustrate, a commercial forest may be used for grazing, wildlife habitat, and recreation as well as for production of wood products. The second activity is, for each operating unit in a given land use, to develop the on-site resource utilization and management practices (Table 2.1, Panel 2). For agriculture the practices would include types and rotation of crops, quantity and time patterns of inputs, methods of tillage and input application, and construction and maintenance of on-site conservation practices. For commercial forests these practices would include selection of tree species, rotations, quantity and timing of inputs, methods of tree planting, methods of harvesting including siting of roads, and water handling practices on roads, skid trails, and landings. Finally, to further reduce adverse downstream effects of the on-site landuse activities, a set of downstream management practices would be installed

N .,..

Table 2.1 The three major activities of watershed management Panel I.

Panel 2.

Divide watershed into major land uses • Agroforestry • Agriculture Irrigated • Forestry Commercial Rain-fed Mixed use • Grazing Preservation • Horticulture

Mining Transportation Urban Lakes, reservoirs, stream channels, and wetlands

Develop set of resource utilization and management practices for each operating unit within each major land use Irrigated Agriculture Commercial Forestry Agroforestry

• Types of crops • Rotation of crops • Quantity and timing of water, fertilizer, pesticides, labor, animal power, and machinery inputs • Methods of tilling (e.g., contour plowing) Panel 3.

• • • •

• Methods of application of water, fertilizer, and pesticides • Installation and maintenance of buffer strips, grassed waterways, terraces, on-farm check dams

Develop set of downstream management • • Stream bank protection by • reserve buffer strips, revegetation, and riprapping •

• Types of tree species • Rotation and spatial distribution of tree crops • Quantity and timing of inputs • Methods of tree planting, thinning, and fertilizing • Harvesting methods, erosian control practices, road siting, construction and mair1tenance practices Debris removal Channel dredging Harbor, estuary dredging

• Types, spatial distri- • Installation and bution, and rotation maintenance of of tree and row erosion control crops measures and road • Quantity and timing siting, construeof resource inputs tion and main• Methods of tilling tenance and tree cropping • Methods of application of water, fertilizers, and pesticides • Treatment of intake water • Wastewater treatment • Check dams

A Conceptual Framework

25

instream and along stream borders (Table 2.1, Panel 3). These practices include structural measures such as sluices in reservoir diversion gates, channel improvements, and stream bank protection by riparian buffer strip preservation, revegetation, or riprapping. APPLICATION OF THE ANALYTICAL FRAMEWORK Each of the three dimensions of the framework represents a different analytical "cut" or dimension of the watershed management problem. Taken together, they form a three-dimensional analytical framework that can be depicted as a cube (Figure 2.4). This cube has 45 individual cells, each of which could provide a basis for analysis. If we assume that the unit of analysis in each cell is the task, the following types of questions can be asked: What specific tasks are required to plan for major land-use assignments? What tasks are required to provide implementation tools to encourage the adoption of resource utilization practices? In theory, an investigation of the content of each of the 45 cells would be required for a complete management analysis. In fact, however, the content of many of the cells may have little management significance; only three cells are involved in the planning question, while five cells are important to the implementation tools question. The analytical task, therefore, can be simplified by constructing twodimensional tables of the type shown in Table 2.2 and selectively analyzing certain cells of the table considered to be of special interest. For example, if the focus of concern is on resource management actions ("things to be done'') at the planning stage (the set of cross-hatched blocks in Figure 2.4), one could start by identifying the tasks required to plan for major land-use assignments. This list might include: • Land capability and suitability analyses, • Formulation and benefit-cost analysis of alternative land-use plans (including assumptions as to associated management practices). If desired, the analysis of management actions at the planning stage could be extended to the second block of the cross-hatched set in Figure 2.4 to identify the tasks needed in planning for on-site resource utilization and land management practices for one or more specific land uses. For agroforestry uses there would be forestry, agronomic, and economic analyses of vegetation types, spatial distribution, and rotation of tree and row crops, and planning of tilling and cropping methods and erosion control practices. The analysis could be extended further to identify tasks needed to plan for off-site management practices for agroforestry and other land uses. Further extension to the entire "planning" slice (the top nine boxes) of Figure 2.4 would allow one to identify all of the tasks required at the planning stage, to formulate a set of management actions, implementation tools,

MANAGEMENT SYSTEM ELEMENTS

MANAGEMENT ACTIVITIES

Resource Management Implementation Inst1tut1onal Actions Tools Arrangements

On-Site Resource Utilization and Management Practices Off-Site Management Practices

Planning rJ) rJ)

w u 0

Des1gn

a:

a_ f-

z

w w

2

Installation

(C)

t----

"'>

~

50

-

-

1-

r--

r--

r---

r--

25

May

Jun

July

Aug

Sept

Oct

Nov

Dec

Jan

Feb

Mar

Apnl

Figure 3.3 Average monthly streamflow and the increase in flow during seven years of annual recutting of deciduous forest on Watershed 17, Coweeta, USA (After Douglass and Swank 1975).

tions of severe erosion or soil compaction following land clearing, the ability of the soil to absorb water is so greatly reduced that baseflow during dry periods is not maintained. After-the-fact reports have suggested that this has occurred in some degraded lands, but few, if any, controlled experiments have been done (Hamilton 1983). Practices such as reforestation or afforestation in a large part of the watershed will normally reduce streamflows during the dry season. This is exactly what has happened in Fiji following large-scale afforestation with pines in former grasslands, and serious policy and planning questions have been raised following streamflow decreases (Kammer and Raj 1979). One answer is to have continuous cutting in the form of thinnings or final harvests in mosaic patterns over the reforested watershed. Meanwhile, reforesta-

Biophysical Aspects

45

Baseflow

Time

Figure 3.4 Stylized stream hydrograph response to a storm event.

tion has reduced erosion and will continue to do so if the harvesting is done carefully. Second, there is the stream response to a single rainfall event or storm. A generalized streamflow response curve (storm hydrograph) has both a baseflow and a stormflow (Figure 3.4). Basejlow is that portion of the overall streamflow that is maintained by groundwater and that would occur had the storm not taken place. The stormflow is the increase in stream discharge due to the rainfall event. Some watersheds characteristically have high stormflow peaks that occur soon after the rain starts, followed by a quick return to baseflow (a "flashy" stream). Other watersheds have hydrographs that are more sluggish in response time and with lower peaks (Hewlett 1982). The rate of rise and fall of hydrographs is controlled more by rainfall characteristics, soil and subsoil properties, and, in larger catchments, the storage geometry of the channel than they are by the type of vegetation present (Pearce and McKerchar 1979). Stormflow peaks and stormflow volumes for a given watershed are usually somewhat higher if there is a large percentage of the watershed in grazing and cropland rather than in forest, and the peaks frequently occur more quickly. Watersheds that are forested or with a substantial percentage of forest may have slightly fewer "flashy" flood events on their streams. Other things being equal, the greater the effectiveness of soil and water conservation practices in the nonforested lands of the watershed, the lower the storm flow peaks and the fewer the floodflows from frequent rainfall events. In major, infrequent storm events, however, overbank flows (flooding)

46

Lawrence S. Hamilton and Andrew 1 Pearce

will develop even though all of the watershed is in pristine forest. When considering the disastrous floods on the downstream reaches of large river systems such as the Brahmaputra, the Yangtze, and the Chao-Phraya, it is a mistake to attribute these to upland watershed land-use practices. These catastrophes result mainly from too much precipitation (or snow melt) occurring too quickly or with too great an intensity for soil storage, interception losses, or transpiration to have an effect. Such factors as rainfall pattern, storm direction, basin size and shape, soil mantle thickness, river channel constrictions through human works, flood plain occupancy, and the amount of pavement and roof area in urban catchments are all more influential in determining the seriousness and nature of major floods than is the manipulation of vegetative cover in the far upper watersheds (Hamilton 1985). Poor watershed land-use management may exacerbate the effects of the floods through siltation, but it is not the cause of major floods in downstream areas. Practices such as forest protection, forest planting, fire protection, and many agricultural practices that either reduce soil compaction or delay water movement off-site may reduce peakflows in local streams though they may not reduce stormflow volumes appreciably. For the frequent storms, overbank floodflows may not occur as often. This effect, however, diminishes rapidly as one moves farther down the basin and as many tributary watersheds contribute to the total flood volume.

Changes in Water Table Groundwater is an important source of well water for domestic and agricultural use, and it supplies springs and feeds streams where the water table emerges at the ground surface. Confined aquifers below an impermeable or confining soil or rock formation are largely beyond the influence of land-use effects (except contamination by chemicals and drawdown through pumping for irrigation). When the water table is near the surface in unconfined aquifers, the kind of vegetation and land use can influence the groundwater. The water table is recharged by infiltration of precipitation and its percolation through the soil pore spaces to a zone of saturation that constitutes the water table. To assure maximum recharge of this important water source, those land uses and vegetative situations that maintain high infiltration capacities are "good" watershed management practices. Forest soils in humid areas have high infiltration capacities in the uppermost layers; capacities that are nearly always greater than normal rainfall intensities. If the forest soil is altered by burning, or is compacted by logging, road building, grazing, or shifting cultivation, then there is some reduction in infiltration. If infiltration capacity is reduced below normal rainfall intensities, then there is a reduction in groundwater recharge. Grazing almost always results in soil compaction. Land managers have the option

Biophysical Aspects

47

of adjusting the stocking levels of animals to reduce the problem. In farm cropping systems, good "tilth" can maintain favorable infiltration capacity. In such systems the use of heavy machinery is a principal cause of compaction and therefore no-till or low-till systems represent good watershed management practices. Evaporation of rainfall intercepted by forest canopies reduces the net rainfall reaching the ground under forests. Because of this factor alone, groundwater recharge under forests may be less than under other vegetation. Soil water transpiration by vegetation also influences groundwater by reducing recharge. By reducing both interception loss and transpiration, forest removal usually results in the water table moving closer to the ground surface (Wicht 1949; Sharma et al. 1982). Subsequent abusive land use leading to soil compaction, erosion down to subsoil, and gullying, however, can reduce infiltration and increase rapid overland water flow so that groundwater recharge is reduced. As a result, the water table will be lower and groundwater discharge that produces baseflow in streams may decrease. If, however, proper soil and water conservation practices are employed after forest clearing and conversion, the usual result is higher water tables, greater reliability of springs, and higher baseflow discharge into streams. If salt accumulates in the subsoil, forest removal and a resultant higher water table can bring the salts closer to the surface and result in soil surface salinization. Lateral movement of salts in groundwater seepage may enter streams, thereby reducing their usefulness for irrigation or domestic water (Peck 1978). Deforestation in several areas of Australia has produced serious problems of this sort. Conversely, afforestation or reforestation usually reduces groundwater levels (Holmes and Wronski 1982). Waterlogged surface soil in parts of China is being improved by planting groves of trees, thus lowering the water table and permitting the interplanting of an annual crop in an agroforestry system. However, it is possible to encounter severely degraded, eroded sites where eventually the planting of forests would provide greater recharge by building up a porous, more permeable, organic-rich topsoil. This topsoil would have higher infiltration and greater soil water storage capacity that might more than compensate for the increased evapotranspiration losses. Trees or shrubs with functioning roots in the saturated soil water zone are called phreatophytes. These plants generally grow along margins of streams and rivers. In water-scarce areas of semi-arid zones, streamflow has been increased by removal of phreatophytes using groundwater in the riparian zone. However, the replacement vegetation also uses water, and the productive new vegetation (especially if it is irrigated to maintain productivity) can use more than the savings from phreatophyte removal, thereby reducing streamflow (Culler et al. 1982). In addition, any water

48

Lawrence S. Hamilton and Andrew J. Pearce

yield increases must be balanced against the possible loss of valuable riparian wildlife habitat, the probable increases in streambank erosion and hence in sediment, the possible effects on aquatic food chains, and the increased water temperatures that may harm fish. BIOPHYSICAL INFORMATION NEEDED Detailed information in the form of airphotos, maps, inventory data, and ecological processes in operation is often not available to planners, managers, or watershed analysts, especially in developing countries. The basic information needed to assess how a watershed, or an area within it, will respond erosionally and hydrologically is suggested below: 1. Geology and Terrain. From geologic maps (if available), preplanning surveys, local knowledge, and professional judgment, some assessment of slope stability and mass erosion hazard should be made. Helpful information in this determination is:

• deep or shallow weathering profile; • consolidation or "looseness" of parent rock; • strength of soils (low or high); • rock strata parallel to slope direction; • fault zones and other zones of tectonic disturbance; • stepped slopes, even if subtle, on soft rocks; • groundwater close to surface or seeps at surface; • slope angles and slope shape; • evidence of high natural erosion. 2. Soils. Detailed soil surveys with maps are desirable; but in their absence, some assessment or rapid survey should be made of the following: • fertility and problems of maintaining it; • soil depth; • infiltration characteristics; • susceptibility to compaction; • susceptibility to surface erosion; • susceptibility of lower surface horizons to dispersion; • presence or absence of an impervious or slowly permeable pan at some depth in the soil; • potential for development of salinity problems with changes in water regime. 3. Hydrologic Behavior. Key items of information include: • precipitation data as localized as possible on amount, seasonal distribution, and intensity and duration of storm events;

Biophysical Aspects

49

• drainage network density and pattern; • permanence or ephemerality of streams; • extent and location of perennially or seasonally wet areas; • nature and position of groundwater table; • flood history information (frequency, extent, and timing). More detailed discussion of the information needs in watershed planning may be found in Terrain Evaluation (Mitchell 1973) and in Assessment and Evaluation for Soil Conservation Policy (Perrens and Trustrum 1984). CONCLUSION An understanding of biophysical relationships that cause changes in erosional and hydrological processes is important in determining such things as: • why certain undesirable situations are occurring (e.g., estimated reservoir lives being greatly reduced-even halved-by sedimentation), and what to do about them; • the soil and water effects of existing land-use trends (e.g., continued expansion of shifting agriculture into forest lands); • the physical consequences of proposed projects (e.g., establishing dendrothermal energy plantations on Imperata grasslands); • what land-use shifts or management practices might be instituted to capture a new soil or water benefit (e.g., increased low flow from a watershed). The effects of rural land uses on soil and water, especially uses involving forests, are widely misunderstood or misinterpreted. They are the subject of much propaganda based more on myth than on fact. The processes and relationships, however, are now fairly well known in temperate regions and can be extended albeit with some caution to tropical watersheds where long-term, high-quality research is still insufficient. Two major tasks are apparent. One is to gain broad acceptance of the principles and trends revealed by existing knowledge about the soil and water effects of land-use change. The second major task is better quantification of specific changes in specific circumstances, so that costs and benefits can be expressed in terms useful in watershed planning. Such data are gradually becoming available, along with detailed case studies in social and economic evaluation of watershed projects (Gregersen et a!., forthcoming 1986). REFERENCES Bosch, J.M., and J.D. Hewlett. 1980. Sediment control in South African forests and mountain catchments. South African Forestry 115:50-55.

50

Lawrence S. Hamilton and Andrew J. Pearce

Bosch, J.M., and J.D. Hewlett. 1982. A review of catchment experiments to determine the effect of vegetation changes on water yield and evapotranspiration. Hydrology 55:3-23. Boughton, W.C. 1970. Effects of land management on quantity and quality of available water: A review. Australian Water Resources Council Research Project 68/2, Report 120, Univ. of New South Wales, Manly Vale. Culler, R.C., R.L. Hanson, R.M. Myrick, R.M. Turner, and F.P. Kipple. 1982. Evapotranspiration before and after clearing phreatophytes, Gila River Flood Plain, Graham County, Arizona. U.S. Geological Survey Professional Paper 655-P. Washington, D.C.: U.S. Government Printing Office. Douglass, J.E., and W.T. Swank. 1975. Effects of management practices on water quality and quantity: Coweeta Hydrologic Laboratory, North Carolina. In Proc. Municipal Watershed Management Symposium, 1-13. USDA Forest Service Northern Forest Expt. Sta. Gen. Tech. Rept. NE-13, Upper Darby, Pennsylvania. El Swaify, S.A., EW. Dangler, and C.L. Armstrong. 1982. Soil Erosion by Water in the Tropics. University of Hawaii College of Tropical Agriculture and Human Resources Research Extension Series 024, Honolulu, Hawaii. Gilmour, D.A. 1977. Logging and the environment with particular reference to soil and stream protection in tropical rainforest situations. In Guidelines for Watershed Management, 223-235. FAO Conservation Guide 1, Rome. Gregersen, H.M., K.N. Brooks, J.A. Dixon, and L.S. Hamilton. Forthcoming 1986. Guidelines for Economic Appraisal of Watershed Management Projects. FAO Guidelines Series, FAO, Rome. Hamilton, L.S. (with P.N. King). 1983. Tropical Forested Watersheds: Hydrologic and Soils Response to Major Uses or Conversions. Boulder, CO: Westview Press. Hamilton, L.S. 1985. Overcoming myths about soil and water impacts of tropical forest land uses. In Soil Erosion and Conservation, eds. S.A. El-Swaify, W.C. Moldenhauer, and A. Lo, 680-690. Ankeny, lA: Soil Conservation Society of America. Hewlett, J.D. 1982. Principles of Forest Hydrology. Athens, GA: University of Georgia Press. Holmes, JW., and E.B. Wronski. 1982. On the water harvest from afforested catchments. In Proc. First National Symposium on Forest Hydrology, eds. E. O'Loughlin and L. Bren, 1-6. Barton, Australia: Institution of Engineers.

Biophysical Aspects

51

Kammer, R., and Raj. 1979. Preliminary estimates of minimum flows in Varaciva Creek and the effect of afforestation on water resources. Fiji Public Works Department Technical Note 79/1, Suva. Lembaga Ekologi. 1980. Report on study of vegetation and erosion in the Jatiluhur catchment. Institute of Ecology, Bandung. Mathur, H.N., Rambabu, P. Joshie, and B. Singh. 1976. Effect of clearfelling and reforestation on runoff and peak rates in small watersheds. Indian Forester 102:219-226. Megahan, W.F. 1981. Nonpoint source pollution from forestry activities in the western United States: Results of recent research and research needs. In U.S. Forestry and Water Quality: What Course in the 80's? An Analysis of Environmental and Economic Issues, Proceedings, 19-20 June 1980, Richmond, VA. Water Pollution Control Federation, Washington, D.C. Megahan, W.F., and J. Schweithelm. 1983. Guidelines for reducing negative impacts of logging. In Tropical Forested Watersheds: Hydrologic and Soils Response to Major Uses or Conversions by L.S. Hamilton (with P.N. King), 143-154. Boulder, CO: Westview Press. Mitchell, C. 1973. Terrain Evaluation. London: Longman Group. Mosley, M.P. 1982. The effect of a New Zealand beech forest canopy on the kinetic energy of water drops and on surface erosion. Earth Surface Processes and Landforms 7(2):103-107. O'Loughlin, C.L., and R.R. Ziemer. 1982. The importance of root strength and deterioration rates upon edaphic stability in steepland forests. In

Carbon Uptake and Allocation: A Key to Management of Subalpine Ecosystems, ed. R.H. Waring, 70-78. Forest Research Lab, Oregon State University, Corvallis. Pearce, A.J., and A.I. McKerchar. 1979. Upstream generation of storm runoff. In Physical Hydrology: New Zealand Experience, eds. D.L. Murray and P.J. Ackroyd, 165-192. Wellington: N.Z. Hydrological Society. Peck, A.J. 1978. Salinization of non-irrigated soils and associated streams: A review. Australian J. Soil Research 16:157-168. Perrens, S.J., and N.A. Trustrum. 1984. Assessment and Evaluation for Soil Conservation Policy. East-West Center Workshop Report, Honolulu, Hawaii. Roose, E.J. 1977. Application of the universal soil loss equation of Wischmeier and Smith in West Africa. In Soil Conservation and Management in the Humid Tropics, eds. D.J. Greenland and R. La!, 177-187. Chichester: John Wiley and Sons. Sharma, M.L., C.D. Johnston, and R.JW. Barron. 1982. Soil water and groundwater responses to forest clearing in a paired catchment study

52

Lawrence S. Hamilton and Andrew J Pearce in south-western Australia. In Proc. First National Symposium on Forest Hydrology, eds. E. O'Loughlin and L. Bren, 118-123. Barton, Australia: Institution of Engineers.

Van Lill, W.S., F.J. Kruger, and D.B. VanWyk. 1980. The effect of afforestation with Eucalyptus grandis (Hill ex Maiden) and Pinus patula (Schlect. et Cham.) on streamflow from experimental catchments at Mokobulaan, Transvaal. Hydrology 48:107-118. Wicht, C.C. 1949. Forestry and water supplies in South Africa. Department of Agriculture, South Africa Bulletin 58, Cape Town. Wischmeier, W.H., and D.O. Smith. 1978. Predicting rainfall erosion losses: A guide for conservation planning. USDA Agriculture Handbook 537. SEA/USDA, Washington. Yost, R.S., S.A. El-Swaify, E.W. Dangler, and A. Lo. 1985. The influence of simulated soil erosion and restorative fertilization on maize production in an oxisol. In Soil Erosion and Conservation, eds. S.A. El-Swaify, W.C. Moldenhauer, and A. Lo, 248-261. Ankeny, lA: Soil Conservation Society of America. Zadroga, F. 1981. The hydrological importance of a montane cloud forest area of Costa Rica. In Tropical Agricultural Hydrology, eds. R. La! and F.W. Russell, 59-73. New York, NY: John Wiley and Sons.

CHAPTER 4

Economic Analysis at the Watershed Level John A. Dixon and K. William Easter

Just as there is a physical, natural logic to a watershed and its use as an organizing unit for analysis, there is also a strong economic logic for the use of a watershed as an analytical unit. This follows from the flow of physical factors and the fact that actions in one part of the watershed (e.g., upland deforestation) can have effects on another, sometimes distant and usually downstream, part of the watershed (e.g., increased sedimentation or changes in flood peak flows). A basic concept in economics, which encompasses these interrelationships, is that of externalities-usually defined as the situation when "some of the benefits or costs of an action are external to the decision maker; that is, some of the benefits accrue to, or some of the costs are imposed upon, individuals who play no part in the decision" (Randall 1981, 157). For example, consider the case of adjoining fields or upstream and downstream areas managed by different decision makers. Externalities exist because improper upland cultivation pract'ices can result in increased erosion and sedimentation on downstream fields or areas. These are negative externalities or costs. The reverse also holds-an active reforestation program can lead to decreased erosion and consequent downstream benefits. One role of the watershed approach is to internalize these externalities. Economic analysis of watersheds includes a search for ways in which externalities can be first identified and then explicitly incorporated into decisions. Since many watershed management actions are implemented in one location to cause a change in another location, there is not the direct cause-and-effect relationship that economic theory predicts as leading to efficient decision making. For example, an upland slope stabilization program may be designed to prevent sedimentation in a downstream reservoir and may not yield any measurable benefits in the uplands (or at least not enough benefits to cover project costs). In this case the benefits are largely 53

54

John A. Dixon and K. William Easter

external to the project site but would be included in a broader watershed analysis. Because of its broad scope, the watershed is a logical organizing unit for an economic welfare analysis because most of the externalities are internalized within the boundary of the watershed. As such the benefits or costs of any proposed project are included in the economic welfare analysis because few are external to the watershed unit. AN INTEGRATED ECONOMIC ANALYSIS Economic analysis of a watershed project or plan will usually be done in a benefit-cost or project evaluation framework. These approaches assist decision makers in deciding among alternative projects or plans. Given constraints on money, managerial skills, and time, building of all useful projects for improving a watershed is not possible. Once the goals have been set (upland stabilization, increase in productivity, sustainable resource use, reduction of adverse off-site effects), the planner-analyst will formulate various alternative management measures and implementation tools to meet the desired goals. These alternatives will then be evaluated and compared to determine the most desirable course of action. Figure 2.3 portrayed the range of inputs, outputs, and other effects that should be included in the evaluation. Two important points must be made regarding the proper role of economic analysis in the planning of watershed management projects. First, the economist must work with other scientists. This is particularly true in the formulation of alternatives. Given the biophysical nature of watersheds, many complex physical relationships must be understood and incorporated into the economic analysis. Such factors as sediment delivery ratios, expected impacts of various vegetative regimes, soil management programs, and hydrologic characteristics may all enter into an economic analysis of a watershed management program. Second, an economic analysis by itself is necessary but insufficient for decision making. When properly done, an economic analysis provides information to the decision maker concerning how resources can be allocated to achieve higher project returns. This information becomes part of the decision-making process. Additional information about social, cultural, equity, and political factors is also considered by decision makers, along with economic analysis. How much weight will be given the economic analysis depends on the quality of the analysis, the decision makers involved, and the importance of other factors. Most watershed programs involve multiple objective decision-making analysis that could include sustainability, stability, economic efficiency, and equity as objectives. Economic analysis can fit into such a framework by providing information concerning the opportunity cost of meeting other objectives at the expense of economic efficiency. In addition, it can deter-

Economic Analysis at the Watershed Level

55

mine the project and resource allocation that attains the highest net present value, given specified levels of the other objectives that are to be achieved. For example, the analysis could show how the net present value of new farming systems changes as the variance in yield is reduced. One would expect to find that high yields are associated with an increased yield variance. Thus, the decision maker would have to determine how much in net present value (due to higher yield) must be given up to increase yield stability.

Accounting Stance In economics the problem of incidence of benefits and costs can be addressed in several ways. One approach is to talk about the accounting stance, or point of view, used in the analysis. Accounting stance merely means the set of "givens;' which are assumed in making a decision. Economists usually use two alternative accounting stances-that of an individual and that of society. The individual accounting stance considers the project from an individual or private perspective. This is usually called ajinancial analysis and considers only those benefits and costs that directly affect the individual or firm. Financial analysis is what individuals and private companies or firms do as separate entities. This is both correct and natural from their individual private perspective. The individuals or the firms are concerned about the benefits they receive and the costs they pay. Externalities, by definition, are not included in their analysis. When a societal accounting stance is used, benefits and costs affecting society's welfare are counted; therefore, what would be considered externalities by individuals are included. This type of analysis is referred to by several names-social welfare, economic welfare, or, more simply, economic analysis. There are many differences between the ways various factors are counted in a financial and economic analysis, as illustrated in Table 4.1. The economic analysis of watershed projects incorporates a wider range of impacts and externalities than does a private/financial analysis. This dichotomy leads to an interesting problem. A wider, social analysis may show that a certain watershed management plan is attractive economically (the benefits exceed the costs). However, the financial analysis of the farmers, or those who must implement the program, may show that the plan is uneconomic for the farmers (costs exceed benefits). Therefore, acceptance of the plan by farmers is low. As shown in Figure 4.1, individual farmers base decisions on their view of the watershed-the small piece of land that the farmer controls and the benefits and costs that are perceived. The watershed management plan, however, includes many individual farmers, along with streams, rivers, and public-owned lands. When the results of the analyses from the two perspectives are not reinforcing (i.e., a project is socially profitable or economic from the watershed perspective but privately unprofitable for the farmer), implementation tools such as subsi-

Table 4.1 A comparison of financial and economic analysis Financial

Ul

0">

Economic

Focus

Net returns to equity capital or to the private group or individual

Net returns to society

Purpose

Indication of incentive to adopt or implement

Determine if government investment is justified on economic efficency basis

Prices

Prices received or paid either from the market or administered

May require "shadow prices" (e.g., monopoly in markets, external effects, unemployed or underemployed factors, overvalued currency)

Taxes

Cost of production

Transfer payments and not an economic cost

Subsidies

Source of revenue

Transfer payments and not an economic cost

Interest and loan repayment

A financial cost; decreases capital resources available

A transfer payment and not an economic cost

Discount rate

Marginal cost of money; market borrowing rate; opportunity cost of funds to individual or firms

Opportunity cost of capital; social time preference rate

Income distribution

Can be measured regarding net returns to individual factors of production such as land, labor, and capital but not included in financial analysis

Is not considered in economic efficiency analysis; can be done as separate analysis or as weighted efficiency analysis with multiple objectives

Adapted from Hitzhusen 1982.

57

~-----------------------------------~

II

I

I

,-----------------1

I

I I

I

\'------Project boundary tor a ' broader community-level analysis

I

! 1i

1 11-!

J_ ,4- 1-.1~£ i.Lt ;{..

£~

~

-

I

L-----~---------------------------~ Figure 4.1 The private and social perspective of a watershed.

58

John A. Dixon and K. William Easter

1

~------Bt ~------NBt

2 ;;::

Q)

c

Q)

en Time ({)

(j)

---------------------------- ct

0

u

j Figure 4.2 Typical stream of project benefits and costs (Source: Hufschmidt et a!. 1983).

dies become necessary. These implementation tools seek to transfer resources to make the project also profitable from the private perspective. TYPES OF ECONOMIC ANALYSIS There are different types of economic analysis but almost all compare expected costs and benefits of a project in some way. The assumption here is that the watershed is managed on a project basis for a public body. Therefore, the social or economic accounting stance should be used for the analysis.

Benefit-Cost Analysis The most common type of analysis is some form of benefit-cost analysis (BCA) whereby the streams of benefits and costs of a project are analyzed over a fixed period of time. This type of analysis requires decisions about the appropriate time horizon (how many years to include in the analysis), the discount rate, and which benefits and costs to include. In general, project costs are large in the first few years of a project and then benefits begin and grow over time. Figure 4.2 presents a stylized picture of a typical set of benefit (B 1) and cost (C 1) streams. An important

Economic Analysis at the Watershed Level

59

question is the size and timing of the net benefit stream (NB 1); that is, do the discounted benefits exceed discounted costs, and if so, by how much?

Cost-Effectiveness When it is difficult to estimate the volume or value of benefits, or when the goal of a project is to attain some predetermined level of environmental quality (e.g., sediment loss in tons per hectare), cost-effectiveness can be used instead of benefit-cost analysis. The cost-effectiveness approach evaluates the various alternative ways to reach a certain goal (e.g., reduction in chemicals in a water supply) and seeks the least cost way to reach that goal. This common-sense approach has been used for some time by engineers, planners, and economists in comparing alternatives. With the cost-effectiveness approach, two major factors should be remembered before it is applied. First, cost-effectiveness does not consider whether the benefits are sufficently large to warrant the expense. Since resources are usually scarce, there may be other projects or programs more beneficial to society. The second factor is that alternatives frequently do not produce the same levels of control; therefore, the choice is not a simple one. For example, assume three projects are being evaluated to reach a target of 100 parts per million (ppm) for some water pollutant. Project A costs $20 million and attains a level of 95 ppm (or a higher level of water quality than the target). Project B costs $35 million and also attains a level of 95 ppm, while Project C costs $5 million and attains a level of 105 ppm. Which project is better? Project A is definitely better than Project B on a cost- effectiveness basis. They both reach the same water quality level, but A is much less expensive than Project B. But what about Projects A and C? Project C is the cheapest and just misses the target by 5 ppm. Is it worth the extra $15 million to reach the Project A level? There is no easy answer. In this case the decision maker will compare the alternatives, the potential damage from a water pollution level of 105 rather than 100 ppm (or 95 ppm), and the alternative uses for the funds. If the water quality level is established by law, there is no alternative but to choose Project A or to change the law so that project C can be legally adopted.

With-and-Without Analysis In analyzing projects involving natural systems such as watersheds, withand-without analysis should be conducted. This can be done with a regular benefit-cost analysis or a cost-effectiveness study. A with-and-without analysis compares the situation with the project to that without the project. This is not the same as a before-and-after comparison because over time changes will occur due to natural causes. The appropriate comparison is between what the situation will be if the project is implemented as compared to if there is no project.

60

John A. Dixon and K. William Easter

For example, poor land use in a catchment area leads to sedimentation in a reservoir. Natural erosion processes of streambanks and streambeds also contribute to sediment load. The proper evaluation of a catchment management program is to determine the sedimentation rates with and without the project. A before-and-after analysis may give a misleading result, especially if naturally occurring erosion was increasing. In this case, even the best catchment management program may not eliminate all sedimentation.

Ex-Ante and Ex-Post Analysis A final and different point concerns the timing of the analysis. Should one do ex-ante analysis, which refers to an analysis done before the project is implemented and many assumptions concerning variables have to be made, or ex-post analysis, which is done after the project is completed and actual results are measured! Ex-ante analysis is essential in helping to decide between alternative management actions; ex-post analysis is also essential, both to learn from experience and to plan better future projects. Unfortunately, ex-ante analysis is far more common than ex-post analysis. In fact, programs and projects are seldom examined to determine what worked or why they failed to meet expectations. Ex-post analysis can reveal the accuracy of models and prediction techniques. The feedback from ex-post analysis can improve future ex-ante analysis by helping eliminate the use of unrealistic prices or inappropriate benefits. For example, a recent ex-post analysis of a water supply and flood control project found that although the ex-post benefit-cost ratio was quite close to the ex-ante ratio, the mix of benefits had changed dramatically. The ex-ante analysis estimated that 92 percent of the benefits would be from domestic water supply and 8 percent from flood control. In contrast, the ex-post analysis found that flood control accounted for 72 percent of the benefits; recreation, I 9 percent; and commercial fishing, 9 percent. The domestic water supply benefits were zero, while the flood control benefits were 37 times higher than expected. In addition, costs were about 100 percent higher, primarily due to an underestimation of land values (Palanisami and Easter 1985). Some form of ex-post BCA can also be used to monitor and guide the implementation of projects or programs during the operations stage. This involves ongoing evaluation and monitoring of performance so that adjustments can be made when results do not match expectations. Unexpected events and impacts make it difficult to predict the exact outcome of a project or program. Therefore, some procedure for evaluation and adjustment must

I. This is not to be confused with the discussion of an evaluation with and without a project. Both ex-ante and ex-post analyses would compare the situation with and without the project.

Economic Analysis at the Watershed Level

61

be built into the project or program if it is to have a reasonable chance of fulfilling objectives.

Regional Impacts The benefit-cost analysis could be expanded to include analysis of regional or local economic impacts. This would be particularly important for low-income or economically backward areas such as the upper watersheds to which the government assigns high priority for economic development. Input-output or economic base analysis can be used to estimate the project or program impacts on regional income or employment. Multipliers can be calculated for income and employment in major sectors of the region. In many of the upper watersheds, this will only involve the forestry, crops, and livestock sectors. In regions expected to generate their own resources to support local infrastructure development, the analysis should consider the local tax base. How will the project affect local finances needed to support local government? This support will be important in areas where local governments are expected to play an active role in project implementation. CONCEPTUAL ISSUES WITHIN A BCA FRAMEWORK This chapter excludes the details of how to do an economic analysis of a watershed management project, which would require much more time and space than are available. Rather, the goal is to introduce the reader to the concepts and an approach to economic analysis of a natural system such as a watershed. This section will cover many key concepts for designing and conducting an economic analysis of a watershed management project. Most of these concepts focus on designing the analysis (i.e., what to include and how). (For additional details concerning economic analysis of watershed projects and soil conservation practices, see Chapter 14, Gregersen and Brooks 1980, Hitzhusen et al. 1984, Nelson 1984, Sfeir-Younis 1985, Tolley and Riggs 1961, and Gregersen et a!. 1986.)

Boundary Previously it was argued that the watershed was a physical unit that also had an economic logic. Its use allowed the inclusion of most effects considered externalities. Therefore, setting the physical boundary is crucial in determining what will be included in the analytical sequence of identification, quantification, monetization, and then economic evaluation. One question that arises is "How far should the physical boundary on the downstream end be extended?" Some analyses end at the juncture of the river or stream with a major river or water body, but coastal areas should also be included if fisheries might be affected by upstream actions. The general rule for economic analysis is to include all downstream ef-

62

John A. Dixon and K. William Easter

fects as far as is feasible and realistic, implying recognition of near-shore marine effects if identifiable and measurable. However, even if the watershed drainage area is appropriate from an economic and a physical viewpoint, concerns for user participation may constrain the boundary selected for management. If users of the watershed are to participate fully in the planning, design, and implementation of projects to improve the watershed, then the management unit cannot be a whole river basin. It must be some smaller part of the river basin that can be managed as a watershed unit. The size of this watershed unit will vary from country to country depending on transportation, communication, institutional arrangements, political jurisdictions, and other factors (Nicholson 1981). Another type of boundary is a temporal one. The appropriate time horizon for the analysis has to be determined. This is particularly important in watershed management decisions because of the inherent instability of hydrologic phenomena during a short time period (less than 20 years). In theory, an economic analysis should extend long enough to include all benefits and costs of a project. In practice, two factors are important in selecting an appropriate time horizon: (1) the expected useful life of the project in terms of yielding the outputs for which it was designed, and (2) the level of the discount rate used in the economic analysis of the project. Concerning the first factor, when usable project outputs diminish or cease altogether, the effective project life can be considered as terminated. As for discount rates, the higher the rate, the shorter the economic time horizon. This is because discount rates act progressively to reduce the present value of benefits and costs obtained in future years. For any given discount rate and value of benefits (or costs), the more distant the year in the future, the smaller the present value of the output and benefits for that future year. Accordingly, for a project with a long, useful life in terms of outputs (assume 100 years) but with a high discount rate (assume 10 percent), the effective time horizon used would be much shorter than 100 years. For example, $10,000 received 100 years hence is only worth $1 today (see Table 4.2). This fact leads to the general rule that the appropriate time horizon for a project is the shorter of (1) the useful physical life of the project, or (2) the economic life of the project measured by the year when discounted net benefits no longer add significantly to the project's net present value. This rule presents a quandary for watershed management projects. If the management project is successful in reaching a sustainable yield equilibrium, the appropriate time horizon will be infinite. The net benefits stream has no natural cutoff point. In practice, however, discounting and a desire to simplify calculations frequently result in shorter time horizons being selected. Discounting resolves the quandary since any benefits or costs beyond 40 years will be so small that they will have little impact on

Economic Analysis at the Watershed Level

63

the net present value of a project (Table 4.2). For the evaluation of watershed management projects, therefore, a time horizon of 30 to 40 years should be sufficient to capture most benefits and costs. This leads to the question of discounting.

Discounting Discounting is the mechanism whereby benefits and costs occurring at different points in time can be weighted and compared. In order to use discounting, two preconditions are required. The first precondition is that all of the variables to be discounted (e.g., resource costs and benefits from project outputs) have to be converted into common units. For ease of analysis, a monetary unit-dollars, yen, rupiah, or marks-is usually used. The second precondition is the acceptance of the assumption that individuals value a unit of present cost or benefit more highly than a unit of cost or benefit in the future. That is, if individuals have a choice of receiving $1 (or one apple) today or $1 (or one apple) next year, they would place a greater value on the dollar or apple received today. This is referred to as time preference of consumption. The discount rate is the factor reflecting the difference in value that people attach to having a good now rather than in the future.

Table 4.2 Present value of a future net return of $100 at four discount rates Time of Net Return (year) 0

Discount Rate (o/o) 2 $100.00

5

8

10

$100.00

$100.00

$100.00

10

82.03

61.39

46.32

38.55

20

67.30

37.69

21.45

14.86

25

60.95

40

45.29

60

30.48

5.35

0.99

0.33

100

!3.80 ................

0.76

0.05

0.0!

··············· 29.53 14.60 9.23 ·················· 14.20 4.60 2.2! ...................

Source: Dixon and Hufschmidt 1986. Note: Different combinations of discount rates and time will yield the same present value of some net return received in the future. For example, a present value of $14-15 is yielded by a $100 net return received 100 years in the future if the discount rate is 2°lo; at a 5"7o discount rate the present value of $100 received in year 40 declines to $14; for an 8% discount rate the decline to $14 occurs in year 25; and with a 10"/o discount rate it occurs in year 20 (see the dotted line in the table).

64

John A. Dixon and K. William Easter

A higher discount rate rapidly decreases the present value of a benefit or cost that occurs in the future (see Table 4.2). For example, $100 in benefits received 40 years in the future has a present value of $45 if discounted at 2 percent, $14 at 5 percent, and only $2 at 10 percent. What then is the correct discount rate to use in an economic analysis? This is not an easily answered question. To answer this question, several conditions will be taken as given: • Only one discount rate will be used in any single economic analysis although the analysis may be repeated several times using different discount rates (sensitivity analysis); separate discount rates will not be used for the cost and the benefit streams in the analysis or for different categories of benefits (environmental or developmental). • The discount rate used does not reflect inflation and is therefore not the market interest rate; all prices used in the analysis are real or in constant dollars. • Discounting will be done on an annual basis. This assumes that the costs and benefits occur uniformly throughout the year or occur at the end of the year. Whereas in financial analysis an interest rate, which reflects market rates for investment and working capital, is generally used, the discount rate used in economic analysis is usually not readily observable in the economy. Economists have developed many approaches for determining and justifying a discount rate for economic analysis. These include the opportunity cost of capital, the cost of government borrowing, and the social rate of time preference. The actual rate to be used in economic analysis will be country-specific and should be established as a matter of government policy. Important factors governing the choice of rate will be the opportunity cost of capital, donor or lending agency requirements, cost of money to the government, and the government's views of the private-sector consumption-investment mix in relation to its concerns for future generations and the level of public spending. (For a fuller explanation, see Hufschmidt et a!. 1983, Dixon and Hufschmidt 1986, Gittinger 1982, and Baumol 1968.) Our position is that project analysts should seek guidance from responsible policymaking agencies of government on the discount rate to be used. In the absence of such guidance, analysts should conduct project economic analyses using a range of rates reflecting those recently or currently in use in the country for public and private investment projects. However, the same range of rates should be used for all projects.

Valuation After deciding the physical and temporal boundaries for the analysis, the next task is to identify the various effects for each alternative so that they

Economic Analysis at the Watershed Level

65

can be quantified and monetized. A time sequence of inputs and outputs for each plan must be quantified and values calculated for each category. This means quantifying environmental quality changes and other externalities as well as the more conventional outputs. The primary outputs from agricultural development projects will be changes in crop and livestock production. For irrigation projects changes in yields and/or increases in net farm income from new cropping systems will be used to calculate benefits. When the project is flood control, benefits will be measured in terms of damages averted. For navigation and hydroelectric projects, benefits should be measured as cost savings for transportation and electric power production, respectively. Here one must measure the real cost savings to users and not just changes in rates charged by government. An agricultural development project in a watershed will produce a variety of goods, services, and effects-some of which occur on-site and others off-site. The simple matrix shown in Table 4.3 illustrates how the variety of goods, services, and effects of a project can be divided into whether

Table 4.3 Relationship between the goods and services associated with watershed management projects and location Location of Goods and Services On-site Types of goods and services

Marketable Nonmarketable

Off-site II

Ill

IV

Adapted from Dixon 1984, and Hamilton and Snedaker 1984. Quadrant I

II

Food crops, forage for livestock, animal products, fuelwood, pulpwood, lumber, and other wood products, minerals, water, fish Fuelwood, animal products, food crops, forage for livestock, water for drinking, fish, irrigation water, hydroelectric power generation, municipal and industrial supplies

III

Aesthetic values, wildlife habitat protection, health benefits of high quality water supplies, protection of aquatic ecosystems, landslide-mudslide control (minimization), preservation of gene pools (natural vegetation and fauna)

IV

Protection of downstream riparian and aquatic ecosystems, high quality water for recreation-aesthetic uses, navigation, flood control benefits, sediment control for avoiding losses of reservoir benefits, etc.

66

John A. Dixon and K. William Easter

they are marketable (and have an observed market price) and by where they occur-on-site or off-site. For example, there are usually markets for crop, livestock, wood, and fish products. Once the quantities produced are known, the valuation is straightforward unless the new production affects the market price. If increased production is likely to lead to lower market prices, then a demand analysis may be necessary to obtain estimates of new equilibrium prices. Even though irrigation water, hydroelectric power, and drinking water are listed in Table 4.3 as marketable goods, in most countries there are no markets for them. Instead, government agencies sell water and power to users at a fixed rate ranging from zero up to their full cost or more. If price distortions exist, other techniques can be used to determine values. To value irrigation water, for example, the demand for water can be derived from the value of the final products produced such as milk, wheat, rice, or vegetables. Similarly, drinking water may be valued by examining health benefits or the value of time saved by having a new, more convenient source of drinking water. In cases where markets do not exist, the valuation problem is more difficult. The analyst must look elsewhere to obtain values for many of the environmental quality changes and some of the other externalities. Although use of the watershed as the boundary of analysis internalizes many of the off-site impacts within the analysis, it does not eliminate the measurement problems. Off-site effects such as changes in sediment loads and water quality will be difficult to value. However, considerable work has been done recently to develop procedures for valuing environmental services and effects that traditionally have not been valued. Surrogate market approaches, including travel-cost and property value procedures, and survey-based valuation techniques are now being widely used to value environmental effects (see Hufschmidt et al. 1983). In a simple case study based on the Nam Pong Reservoir in Thailand, Ruandej Srivardhana examined the potential benefits from better management of the watershed above the reservoir (Dixon and Hufschmidt 1986). The case study estimated the value of the four major types of benefits derived from the project-hydroelectric energy, irrigation, flood damage reduction, and the reservoir fishery. Information from the watershed was used to estimate rates of erosion and silt deposition in the reservoir under different resource management actions. The connection among reservoir sedimentation rates, the reduced reservoir capacity, and the consequent decline in benefits generated over a 50-year period was used in the analysis of the costs and the benefits (reservoir losses avoided) of various resource management actions, including the "without project;' no-management option. The comparison between alternatives-no resource management actions and several different levels of resource management actions-was done in

Economic Analysis at the Watershed Level

67

terms of net present value. The results indicated that, under the estimated conditions, a medium-scale resource management program would be economically justified above the reservoir while a full-scale resource management program would produce fewer net benefits than the no-management option.'

Uncertainty and Sensitivity Analysis There may be considerable uncertainty about many of the variables used in a benefit-cost analysis. This is particularly true for a project with an expected life of more than 10 years. For example, an analysis of diesel fuelpowered tube wells in 1972 would have underestimated the price of diesel for the 1974-84 period and made the wells appear much more profitable than they actually were. There are two general ways of accounting for uncertainty. First is to use sensitivity analysis for key variables such as the prices and yields of the major crops to be grown in an irrigation project. The benefit-cost analysis would be rerun for a range of prices and yields to determine the impact on the benefit-cost ratio. If the yield assumption is critical in determining whether the project is profitable, then additional information should be collected to reduce the uncertainty about the yield estimates. The new information would then be used to redo the benefit-cost analysis. A second approach is to introduce probabilities into the analysis. This requires that there is some knowledge about the range and mean of possible values for the key variables. If enough information is available, benefitcost ratios can be calculated with different probabilities assigned to various outcomes. For example, one could estimate that 90 percent of the time the B/C ratio would be greater than 1 and that 80 percent of the time it would be greater than 1.5. This provides the decision maker with specific information about the certainty of the estimates. Each benefit-cost estimate has an implicit probability attached to it. The sensitivity analysis just makes the probabilities explicit but to do so requires additional time and data. Information for analysis is not free, and the cost of information must be compared with the benefits from improved decisions to determine how detailed an analysis to conduct.

Benefit Distribution In watershed projects there is always the trade-off between upstream or on-site benefits and downstream or off-site damages. Even though society may reap net benefits from a specific watershed management program, the program may not be feasible because of what users must sacrifice in

2. It should be emphasized that the hypothesized erosion and sedimentation rates are not estimates of possible future conditions; rather, the rates were selected merely as illustrations for the case study.

68

John A. Dixon and K. William Easter

the upstream areas. A shift from cassava production to trees may mean disaster for the net income of individual farmers, especially in the short run. The problem is that costs and benefits accrue to different people. Thus, the project analysis needs to show who benefits and who pays the costs of a program or project. This information can then be used to design ways of influencing that distribution. Institutions or administrative procedures might be established to assist upstream people who take actions that benefit downstream people. This could involve implementation tools such as subsidies for upstream users who plant the desired crops. The government or the downstream communities could also buy the rights to grow crops from the upstream users, who would still own the land but could only grow trees or grass. In general, a reduction in downstream damages will mean lower incomes for upstream farmers, at least in the short run. Unless institutions and implementation tools are developed that establish a new set of incentives for upstream users to adopt the desired practices, downstream damages will continue. A second and related point is the transactions costs involved in establishing appropriate resource management actions. 3 Were it not for transactions costs, the downstream users would get together and pay the upstream users to reduce the downstream damages, i.e., buy the rights to grow crops in the upstream areas. Thus, one of the functions of new institutional or administrative arrangements would be to reduce transactions costs. When these transactions costs are higher than the benefits that can be gained from an agreement about resource management actions, no action will be taken.

IMPLICATIONS AND CONCLUSIONS An economic analysis is only one part of the watershed management process. But just as it is important to get the facts straight on the physical aspects of a project, one should also have the project economics clearly defined. When developing values for the various inputs and outputs of a project, one must look to economists who have a broad understanding of project and environmental analysis. In evaluating projects or programs, three points should be emphasized. First is the difference in perspective between private individuals and society. The individual will measure benefits and costs differently from society. Therefore, analysis should be done from both perspectives. If the outcome is the same, then corrective action is not needed. Yet, if the analysis from the public perspective calls for implementing the resource manage-

3. Transactions costs · · commg · to an agreem . .are the var1· e t Y o f costs t h at would be mvolved ment for a reductwn m downstream d amages. c osts may be mcurred · · collectmg · mforma· . m tlon on downstream damages • in orga mzmg · · farmers, an d m · admm1stenng · · · and pohcmg · · the agreement.

Economic Analysis at the Watershed Level

69

ment action while from the private perspective the answer is no, then public action may be needed. The second point involves measurement of environmental impacts. In the past, benefit-cost analysis has been criticized because some of the important environmental impacts were omitted; however, this situation is rapidly improving. Environmental and natural resource economists have developed ways of valuing many environmental effects. New procedures for valuing project costs and benefits will make the analysis more comprehensive (Hufschmidt et a!. 1983). Finally, in the analysis of watersheds one of the critical gaps has been knowledge of downstream impacts. Although the types of downstream effects are known, little information concerning the quantity or value of the impacts is available. These downstream impacts probably are even more important than those from upstream or on-site. Thus, a concerted effort is needed to quantify downstream effects so they can be better represented in the decisions concerning watershed management. REFERENCES Baumol, W.J. 1968. On the social rate of discount. American Economic Review 58:788-802. Dixon, J.A. 1984. Economic valuation: Its role and use in assessment and planning. Paper for the International Training Course on Environmental Impact Assessment and Land-Use Planning, January, Hong Kong. Dixon, J.A., and M.M. Hufschmidt, eds. 1986. Economic Valuation Techniques for the Environment: A Case Study Workbook. Baltimore: Johns Hopkins University Press. Gittinger, J.P. 1982. Economic Analysis of Agricultural Projects. Baltimore: Johns Hopkins University Press. Gregersen, H.M., and K.N. Brooks. 1980. Economic analysis of watershed projects: Special problems and examples. In Economic Analysis of Forestry Projects, 137-176. FAO Forestry Paper 17, Supplement 2. Rome: FAO. Gregersen, H.M., K.N. Brooks, J.A. Dixon, and L.S. Hamilton. Forthcoming 1986. Guidelines for Economic Appraisal of Watershed Management Projects. FAO Guidelines Series, FAO, Rome. Hamilton, L.S., and S.C. Snedaker. 1984. Handbook for Mangrove Area Management. Published by EWEAPI, IUCN, UNESCO, and UNEP. Hitzhusen, F. 1982. The "economics" of biomass for energy: Towards clarification for non-economists. Ohio State University. Hitzhusen, F., R. MacGregor, and D. Southgate. 1984. Private and social cost-benefit perspectives and a case application on reservoir sedimentation management. Water International 9(4):181-184.

70

John A. Dixon and K. William Easter

Hufschmidt, M.M., D.E. James, A.D. Meister, B.T. Bower, and J.A. Dixon. 1983. Environment and Natural Systems and Development: An Economic Valuation Guide. Baltimore: Johns Hopkins University Press. Nelson, M. 1984. Economic/institutional issues in sedimentation management in developing countries. Water International 9(4):165-168. Nicholson, N.K. 1981. Applications of public choice theory to rural development: A statement of the problem. In Public Choice and Rural Development, eds. C.S. Russell and N.K. Nicholson, 28-29. Research Paper R-21. Washington, D.C.: Resources for the Future. Palanisami, K., and KW. Easter. 1985. Ex-post evaluation of flood control investments: A case study in North Dakota. Water Resources Research 20(12):1785-1790. Randall, A. 1981. Resource Economics. Columbus: Grid Publishing. Sfeir-Younis, A. 1985. Soil conservation in developing countries. The World Bank, Washington, D.C. Tolley, G.S., and F.E. Riggs. 1961. Economics of Watershed Planning. Ames, lA: Iowa State University Press.

CHAPTER 5

Economic Policies and Watershed Management Alfredo Sfeir-Younis 1

The previous chapter considered economic analysis at the watershed management level. The focus was on the correct economic analysis of watershed management activities or projects, both from the individual's and from society's perspective. While such an analysis is necessary, broader macroeconomic policies and external factors can have important effects on a watershed. The understanding of these economic policy-watershed interactions is the topic of this chapter. Sustained economic development rests on proper management of natural resources. Existing economic policies have often ignored this fact and assumed that somehow proper adjustments in the allocation and use of such natural resources as soil and water will eventually take place. Environmental changes have been defined as residuals to most macroeconomic interventions. This approach to economic policy has resulted in serious degradation of human and natural environments. For example, soil degradation represents a major worldwide environmental threat, particularly in developing countries. Soil erosion, one of the most important causes of land degradation today, is a primary concern in formulating watershed management plans. Economic policies on prices, income, and other incentives-coupled with institutional problems and social factors such as population growthhave often resulted in soil erosion leading to irreversible environmental damages in the upper and lower watershed. ACCOUNTING UNIT By design or by default, the general assessment of the welfare impacts of alternative economic policies has chiefly focused on such traditional ag-

I. The views expressed in this chapter are solely the author's and should not be attributed to the World Bank or any of its affiliates.

71

72

Alfredo Sfeir-Younis

gregates as income, employment, and foreign exchange. National accounting systems, and the accompanying indicators of economic performance, disregard aggregates showing policymakers the potential or actual changes taking place in the natural environment (whether positive or negative). The same applies to the way in which development planning is conducted. For example, environmental planning in general or land-use planning in particular are seldom integrated with development planning. A major reason for this unfortunate disjunction is the analytical unit of account used today in the formulation of economic policy. As discussed in earlier chapters, the use of a watershed unit for planning will help eliminate this disjunction.

Agricultural Development Based on the Farm Unit It is customary to see reports dealing with agricultural development using the "farm" as the basic unit of account. Because the agricultural sector is conceived as an aggregation of these farm units, policy recommendations from such reports deal mainly with incentives to improve efficiency at the farm level. In Asia this often means many small, interdependent units within which one is trying to maximize returns. Concomitant to this, investment decisions, which would determine the level of financial commitments to a sector, are also based on analytical approaches based on the farm (i.e., the basic production unit). Accordingly, success is measured most often by quantifying the total net value of production per unit of land. Within this context, measuring yields and financial returns on a per hectare basis becomes the central theme. Even when decisions are made from the "economy's point of view;' aggregate analysis is similar to farm analysis except for the type of pricing system used. For example, the macro-analysis will use economic prices, or "shadow prices;• instead of financial prices. The basic fact remains: The analytical unit of account is the farm, and regional investments (or projects) represent an aggregate of those farms. The "project;' as an aggregation of farms, artificially imposes a boundary to the economic analysis. This narrow boundary becomes limiting in the context of natural systems such as watersheds, particularly to the analysis of the allocation of soil and water resources. This approach to the economic analysis of investment decisions has often been referred to as ''traditional'' benefit-cost analysis. The traditional approach has prevailed in the past for several reasons. First, there was no reason to believe that major external effects across farms or across sectors had a significantly large economic value (i.e., external effects are assumed to be minimal or absent). Second, there was little understanding of the interrelated nature of resources. Finally, there was little attention paid to the intertemporal relations between costs and benefits of different activities, except those which concern the individual project.

Economic Policies and Watershed Management

73

In sum, both spatial and intertemporal externalities were assumed to be too small to make it worthwhile to incorporate them into the analysis of investment decisions. Therefore, the traditional approach to development decisions in some cases has resulted in investments that may be "optimal" from the farmer's point of view, but that are "suboptimal" from society's point of view. In addition, the traditional concept of opportunity cost has provided a misrepresentation of costs to the economy of many investments and policy decisions. Three components of opportunity cost need special attention in natural resource decisions: (1) off-site or downstream effects, (2) intergenerational trade-offs, and (3) irreversible changes.

The Watershed as a Unit of Account As a unit of account, the watershed enables one to improve the analysis of external costs; this concept was discussed in the preceding chapter. The boundaries of a watershed are larger than an individual project or farm; therefore, many of those affected by these external effects will be included in the economic analysis of watershed projects and policies. In other words, this new unit of account will enable policymakers to assess in a more comprehensive way the spatial effects of benefits and costs of different interventions. Neither production effects from upstream areas nor environmental quality assessments downstream (e.g., sedimentation, water quality) will be analyzed in isolation. The "watershed" as a unit of account does little to directly improve our analysis of intertemporal or intergenerational effects. The incorporation of these effects requires use of a different framework (see Sfeir-Younis 1985). Moreover, the inherent instability of hydrologic phenomena in a time span of less than 20 years means that the evaluation of watershed programs should include at least two generations. ECONOMIC POLICY ENVIRONMENT There is a need to focus not only on investments (or projects) but on the policy environment that surrounds those investments. Well-conceived investments, from both a technical and an economic perspective, often fail to generate the expected benefits to the economy because of the existing policy environment, that is, the incentives or disincentives faced by individuals. Incentives can be divided into two broad types: market incentives (e.g., prices, taxes) and nonmarket incentives (e.g., land tenure, property rights). Institutions, or the lack of them, create incentives or disincentives as discussed in Chapter 7. Economic units located in different areas of a watershed will respond to these incentives in different ways. For example, in a watershed characterized by insecurity in land tenure, one would ex-

74

Alfredo Sjeir-Younis

pect investments in soil conservation practices to be lacking. Many of the benefits from these practices will occur in the distant future; therefore, tenant farmers will often be reluctant to adopt such practices.

Acceptable and Unacceptable Policy Decisions Economic performance of alternative policy packages has seldom been measured by their potential impacts on the allocation and use of natural resources. This gap in analysis has also been true for the formulation of watershed management schemes and has resulted in major environmental deterioration in many parts of the world. Certain policy decisions, which were designed to improve one or more aspects of resource conservation and management within a watershed, may end up defeating the conservation efforts because the policy is formulated on too narrow a basis. Input subsidies have been advocated by many decision makers as a vehicle to increase agricultural production and farm income. However, these subsidies will decrease the private opportunity cost (in the traditional form) of the soil and its nutrients, thus encouraging the adoption of practices that only mask the effects of erosion and increase the levels of chemical pollution in the soil and water. For example, subsidizing fertilizers will encourage farmers to move into more fragile or marginal lands that, in principle, are unsuitable for cultivation. Subsidizing irrigation water will have a similar effect by making it both economically viable to cultivate lands that otherwise are unsuitable for intensive cultivation and to encourage the use of excessive quantities of water. Some of the most negative effects of excessive irrigation include salinization and water logging. Finally, subsidizing land clearing and drainage and reclamation activities (e.g., by tax concessions for land development or cost-free machinery and equipment) may result in serious environmental degradation, including destruction of vegetation, downstream salinity problems, ancl increased erosion. Because of these various effects, a policy of subsidized inputs or credit can lead to mixed results: both increased private returns to farmers and increased social costs to society because of soil degradation and other negative environmental effects. The broader, social analytical perspective described in Chapter 4 helps identify some of these trade-offs. Price supports are another popular policy instrument. Many governments of developing countries subsidize agricultural outputs. This policy makes the protected crop more profitable; consequently, farmers will allocate economic resources to the production of the "favored" crop. Intensification and area expansion are two immediate effects of such price support systems. Both effects may increase the rate of environmental deterioration. In particular, crop intensification may require agricultural practices that may exhaust soil nutrients and contaminate water through pesticide use. It can deplete soil micronutrients and cause a deterioration in soil texture

Economic Policies and Watershed Management

75

and structure due to excessive plowing. Area expansion in practical terms means the use of marginal lands. These lands are often quite fragile from an ecological viewpoint, and soil degradation is likely to occur. Undoubtedly, in several cases price support policies have increased farmers' incomes and, by expanding the supply of different products, may have benefited consumers. Thus, such policies may have been judged beneficial to the country from income, employment, and foreign-exchange earnings. This will be the most probable answer under the traditional approach; however, from an environmental viewpoint, this policy should be qualified as less successful. Import policies may not only affect soil deterioration in a country itself, but may also have a serious impact on other countries. For example, the European Economic Community's (EEC) import policy involving feed grains has had a significant impact on Thailand and the United States. Both countries have higher rates of soil erosion due to changed cropping patterns because of these policies. In Thailand the export of cassava for livestock feed to western Europe has caused a major expansion of cassava production in Northeast Thailand, resulting in cultivation of marginal lands with high rates of soil erosion. For the United States, the European import policies have encouraged soybean production relative to corn production. Unfortunately soybeans do not provide as good a soil cover as corn. Thus, contrary to what a distinguished U.S. economist has claimed, the decline in U.S. corn acreage because of replacement largely by soybeans since the 1930s has actually resulted in more soil erosion (Schultz 1982). The EEC countries are interested in how their import policies affect their farmers and are not particularly worried about soil erosion in other countries. This leaves countries such as Thailand and the United States with limited alternatives to correct this imbalance. They either allow soil erosion to increase or introduce new policies and programs to increase investments in erosion control. Unfortunately, given the existence of alternative suppliers and high stocks of soybeans and other feed grains elsewhere, it is difficult to pass on to the EEC in the form of higher prices the social costs of producing these commodities.

INTERTEMPORAL QUESTIONS IN ECONOMIC POLICY Although spatial decisions apparently are easier to express within the watershed planning context, intertemporal decisions are more complex. The spatial problems expressed previously may be solved by changing the analytical unit of account to the watershed and "internalizing" most externalities. However, a unit of account capable of dealing with intertemporal effect is more difficult to develop.

76

Alfredo Sjeir-Younis

How are decision makers engaged in watershed management to decide on intertemporal policies? What would be an acceptable package of investments and economic incentives (market and nonmarket incentives) that would intertemporally optimize the production and consumption of goods and services of a given watershed? The answers to these questions depend on the development objectives of a society or a country. These objectives will provide the basis for defining acceptable decision criteria. The allocation and use of hydrologic resources (e.g., surface and groundwater) are made by several groups in a watershed. Water may be used by the agricultural sector for upstream and downstream irrigation, by the energy sector in hydroelectric power production, by industries located along the river or in the city, and by rural and urban households for drinking water. The short-term objective of the country may be the expansion of irrigation from surface water sources to increase agricultural production. In the long-term, development of other sectors may be desired. This intertemporal decision will result in an initial mobilization of surface water and other resources for the agricultural sector. Over time, competing uses for the same source of water will develop and groundwater sources can then be developed. As expressed earlier, because of the nature of goods and services in a watershed the traditional benefit-cost decision criteria suffer from some major limitations. These limitations are particularly evident as one moves from directly consumable to environmental goods or services. As most of these decision criteria are based on discounted cash-flow techniques, these criteria will often underestimate the consumption and production merits of many environmental goods and services. This problem is often expressed by saying that "It is not economically profitable to invest in soil conservation projects because they have an internal rate of return lower than the opportunity cost of capital or have a net present value of less than zero:' How can policymakers ensure that today's economic incentives will sustain development of a watershed in the future? This question has received little attention in the watershed management literature with the exception of the idea of a safe minimum standard (Bishop 1978). A few reasonspolitical instability, low per capita income, disparities in policy effectiveness (implementation), and immediacy of social problems-may explain this lack of attention particularly within the context of developing countries. Political instability tends to focus the policymaker's mind on immediate results, and the medium- and long-term negative effects on the environment are ignored. Thus, it is not at all uncommon to see the formulation of policy packages maximizing the short-term, directly consumable outputs of watersheds such as food crops and timber production at the expense of irreversible losses of valuable top soil. Low per capita income in most developing countries tends to favor current consumption over investment for the future. Low-income farmers located in a watershed will

Economic Policies and Watershed Management

77

allow heavy soil losses so that consumable goods can be produced to satisfy their present basic needs. Decisions between consumption and investments will favor consumption today. In this particular context, policymakers should view the conservation of the watershed's resources as an "income maintenance" problem. Policy implementation failure is a major contributing factor to environmental deterioration in many watersheds. Since some policies are more difficult to implement than others, policymakers tend to favor those where immediate changes are possible and visible. Therefore, it may be more politically palatable to allocate scarce financial resources to new investments such as irrigation dams and canals rather than to the rehabilitation of old structures or to the maintenance of existing structures. This is an important factor contributing to the underinvestment in project operation and maintenance (Howe and Dixon 1984). It may also be thought better to deal with changes in prices rather than with changes in land tenure. Finally, many developing countries have been faced with severe socioeconomic problems, and these problems have directed the attention of policymakers away from basic environmental issues.

Natural Resources as Objectives or Constraints An extension of the traditional approach of maximizing income subject to input constraints has evolved in the past few decades. As many resources are being degraded or irreversibly lost, resource-related characteristics have been incorporated as constraints. In this approach increases in national income still prevails as the chief objective. It could be improved if there is a clear understanding that both the quantity (stock and flow) and the quality of natural resources are income-earning assets: poor land makes poor people. But even this modified approach has been questioned. Newly developed approaches postulate that the main objective of a given society is not to increase the value of national income; rather, the chief objective is the enhancement of the human and natural environment. It is within this context that such issues as "sustainable development" and "genetic diversity" come into play. For example, one may need to establish safe minimum standards for certain species and determine the opportunity cost of such standards. Based on these costs, it is now possible to see how much "genetic diversity" a society or a group within society is willing to buy. The question of what is the objective and what is the constraint in economic policy becomes even more complex when intergenerational elements are taken into account. Should policymakers formulate policies that increase the welfare of this generation, or should they also be concerned about the welfare of future generations? These concerns have become critical in the establishment of policy frameworks for watershed management. Certainly no simple and straightforward answers are readily available.

78

Alfredo Sjeir-Younis

POLICY AGENDA

Establishing Priorities and Standards A central role of economic policy is to establish priorities and standards for achieving goals and targets during program implementation. The watershed as a unit of accounting for economic policy will enable planners to disclose trade-offs in setting standards and in assessing intra- and intersectoral programs and the potential equity implications of different policies. Determining standards such as how much soil erosion upstream is acceptable and how much variation in water quality is allowed within the watershed system are just two examples of the central issues faced in development planning. Other standards are needed for resource management such as the protection of wildlife and wildland, industrial location, and allowable wastewater discharges into bodies of water. Clearly, the establishment of standards based on purely technical parameters such as acceptable erosion rate in tons per ha per year is not always meaningful because the value of a ton of soil lost almost always varies by location. Alternative quantity and quality standards need to be developed with an assessment of the likely costs and benefits. In this regard, economic efficiency, equity, and public savings will be central in defining standards. In some instances the watershed as a unit of accounting goes beyond an individual country. Establishing policies and standards in these international watershed will require the cooperation of several countries. A particular case in point is the management of sediments in international river basins.

Income and Distributional Policies Watershed management policies should not be established independently of other economic policies, particularly income policies. The lower the per capita income of people located in a watershed, the less their willingness to postpone consumption. To achieve a desirable level of consumption, low-income farmers will continue to put pressure on the natural resources available to them. Since most poor people live on poor lands, income policies are important. The watershed, as a new unit of accounting in economic policy, will provide a framework that will enable decision makers to assess important distributional impacts of economic policy. Five important types of distributional impacts are (1) among farmers in the upper portion of the watershed, (2) between upstream farmers and other activities located in the watershed, (3) between producers and consumers of resources and other products, (4) between land owners and land operators, and (5) between generations. These distributional effects are important in weighing decisions with regard

Economic Policies and Watershed Management

79

to which group in society should benefit and for establishing adequate incentive systems. The development of appropriate technology to meet these goals may be necessary. CONCLUSION Several important points can be drawn from this discussion of macroeconomic policies and watershed management. The first point is the need to change traditional approaches to economic policies in watershed management. At present the literature has emphasized the investment aspects of watershed development. There is little emphasis on economic incentives. A second important point is the need to change the analytical unit of accounting in watershed policy. The watershed unit of accounting will enable analysts to improve economic analysis by incorporating external costs. As outlined above, however, this new unit has some limitations that are important to remember. The third point relates to intertemporal decisions, which are more complex because of the introduction of new elements that are difficult to assess. Economic policies will have intertemporal costs and benefits, and these costs and benefits should be taken into account in public decision making. Finally, from a practical viewpoint, several conclusions may be outlined. First, macroeconomic policies, even in cases where they may appear not to be connected, have important effects on the environment. Second, assessment of macro- and even microeconomic interventions is needed to avoid unintentional impacts that defeat the same objectives these policies are supposed to accomplish. Third, economic incentives such as subsidies might aggravate the watershed degradation problem, even if the incentives improve farmers' income in the short term. Finally, the objective of any incentive scheme should be such that spatial and intertemporal effects resulting from soil and water degradation are fully accounted for. REFERENCES Bishop, R.C. 1978. Endangered species and uncertainty: The economics of a safe minimum standard. American 1 Agricultural Economics 60:1, 10-18. Howe, CW., and J.A. Dixon. 1984. Distortions in the development process: An economic perspective. East-West Center, Honolulu, Hawaii. Schultz, TW. 1982. The dynamics of soil erosion in the United States: A critical view. Agricultural Economics Paper 82:8. Paper for Conference on Soil Conservation, Agricultural Council of America, 17 March, Washington, D.C. Sfeir-Younis, A. 1985. Soil conservation in developing countries. The World Bank, Washington, D.C.

CHAPTER 6

Behavioral and Social Dimensions George W. Lovelace and A. Terry Rambo 1

Success in watershed management requires more than an examination and understanding of environmental resources and natural processes. It also requires the consideration and analysis of human behavior and a variety of social factors that affect land use as well as project planning and implementation. The inherent logic of the watershed as a natural, a functional, and an analytic unit and the watershed's suitability and utility as a planning and management unit are supported only in part by the pattern of human activity across the landscape. This is because many human activities connected with resource exploitation and rural land use are influenced by essentially "social" factors (e.g., politics, culture, history, religion, and ethnicity), which are only indirectly and partially related to the natural environment as represented by the watershed. Consequently, many of the problems that watershed planners and managers must address in their work lie at the interface of these distinct natural and sociocultural realms. They must understand both of these to work effectively. THE HUMAN ECOLOGY PERSPECTIVE Considerations of the behavioral and social aspects of environmental utilization and management normally focus on relationships between human society and the natural environment. Human ecology (the scientific study of human-environment interactions) and the "systems model of human ecology'' in particular provide a useful perspective for initially conceptualizing and examining these relationships. I. This chapter is a revised version of a longer manuscript. The authors would like to thank K.W. Easter, J.A. Dixon, M.M. Hufschmidt, L.S. Hamilton, and N.L. Jamieson, all of the East-West Environment and Policy Institute, Honolulu, for their useful comments on earlier drafts of this chapter.

81

82

George W. Lovelace and A. Terry Rambo

The Systems Model of Human Ecology The "systems model of human ecology" examines relationships between humans and their environments by focusing on the systemic interaction of the natural ecosystem on the one hand and the human social system on the other hand (Rambo 1983). The natural ecosystem is composed of a complex of biophysical factors, such as soil, water, climate, flora, and fauna, all of which are interrelated in such ways that changes in one component can lead to changes in other components. For example, a climatic change involving increased aridity over time can lead to a change in the availability of ground and surface water, which in turn affects the distribution of flora and fauna. Like the natural ecosystem, the human social system also consists of a number of interrelated elements. Examples of social system elements include demography, social organization, economics, ideology, political institutions, and language. It is the interactions of these components that largely determine the character of the social system and have considerable influence on its stability, resilience, and development over time. Natural system-social system interactions. The natural ecosystem and the human social system are interrelated and connected through exchanges of energy, materials, and information. The social system depends upon the natural system for the basic materials and energy it requires. Change in the availability of these necessary natural system inputs can thus have a tremendous effect on the social system's operation and viability. For example, siltation of irrigation channels, coupled with salinization of crop lands, is believed to have been a major factor in the decline of ancient Mesopotamian civilization (Jacobson and Adams 1958). In a similar fashion, modification of the social system can greatly affect and change the natural system. This has often been the case following major innovations in technology and/or increases in population. In Iron-Age Britain, for example, a combination of a new technology based on iron tools, new agricultural techniques, and an increased population brought about dramatic and irreversible changes in the upland ecosystem through upland deforestation and degradation and increasing soil acidity, podzolization, and erosion (Drew 1983). Through time and feedback and through selection and adaptation, each of the two subsystems shapes and is shaped by the other. The social system and the natural system undergo a process of interrelated coevolution, such as that which Norgaard (1981) has described for the Amazon Basin. P_artia/ autonomy and external inputs. In conceptualizing humanenvu?nment~l relationships in a watershed as a close, highly interdepend~~t mterac~10n of the social system and ecosystem, however, several additiOnal pomts should be made. The first point is that each of the subs~st~m~ has a degree of autonomy, a dynamic of its own that within certam hm1ts allows for both independent development and flexibility. This

Behavioral and Social Dimensions

83

autonomy allows each subsystem to resist certain kinds and amounts of change introduced by the other subsystem or from the outside. Adaptive technologies, for example, may allow the social system to preserve much of its integrity in the face of considerable environmental change. Although the ecosystem's capacity to resist change is less pronounced, the presence of limiting factors (e.g., limited water supplies or extremely difficult topography) in the environment may restrict or inhibit the development and magnitude of changes occurring in the social system. A second point is that the larger human ecological system, composed of both the natural ecosystem and the social system, is not closed. It can be affected by forces from the outside, such as external economic forces and political forces. External factors, such as a war-induced immigration into a watershed or pressure from outside commercial interests to exploit resources, can easily upset existing patterns of equilibrium between the human residents and natural resources of the watershed. Some of the external economic forces were discussed in Chapter 5. Multiple sets of human-environmental interactions. Humanenvironmental interactions in one portion of the watershed can also affect ecosystems, social systems, and human-environmental interactions in other areas of the watershed (Figure 6.1). For example, denudation of an upper watershed can significantly affect the nature of interactions between lowland river valleys and intensive agriculturalists. Similarly, an increasing intensity of human-environmental interactions in the lowlands resulting from population growth may ultimately affect less densely populated portions of the watershed through migration. In fact, this has happened in many parts of Asia and has led to serious problems of upper watershed degradation with consequent adverse downstream impacts as highlighted in the second section of this book. STRUCTURAL PARALLELS BETWEEN SOCIAL SYSTEMS AND NATURAL SYSTEMS As a geophysical entity, the watershed possesses a basic structure and a topographic pattern that, together with the law of gravity and the flow of water, shape its biotic and abiotic characteristics and processes with considerable regularity. Given this regularity, it is perhaps not surprising that human societies in different parts of the world have frequently adapted to the watershed landscape in similar manners. Indeed, there are often structural parallels between the ways in which human groups are organized and spatially distributed and the natural physical pattern of the watershed (Hamilton and King 1984). Just as the watershed can be broken down into separate physical units (e.g., upper catchments, river valleys, and coastal estuaries), rural populations in many regions of Asia can be divided into different kinds of groups on the basis of adaptations to different portions of the overall watershed

84

Figure 6.1 in the watershed an- env iro nm ent al rela tion s A simplified mo del of hum context.

Behavioral and Social Dimensions

85

landscape. It is quite common, for example, to find small-scale hunting and gathering societies and groups of forest-based shifting cultivators occupying upper watersheds juxtaposed with intensive agricultural communities that are concentrated in adjacent river valleys. These different subsistence and settlement strategies frequently represent highly specialized adaptations to particular microenvironments within the overall landscape, adaptations that have evolved over centuries of combined social and environmental adjustment. These adaptations are not easily changed and frequently they should not be changed. They should, however, be taken into account in preparing management plans for these watersheds.

Symbiotic and Trans-Environmental Adaptations Specialization of the sort mentioned previously has sometimes led to the development of symbiotic relationships between different groups within the overall watershed. For example, over the course of many centuries, hill groups and lowland communities in southernmost Yunnan Province in southwest China have not only evolved relatively separate adaptations but also mutually beneficial trade relations. Upland groups such as the Hani and the Yao exchange crude tea and other upland products for grain and cloth with the lowland dwelling Han Chinese and Dai peoples (Pei 1984). These types of upland-lowland trade arrangements are quite common and have often existed since ancient times in various parts of Asia (Dunn 1975). While these relationships can be described as "symbiotic" in that each partner derives needed resources from the other, the terms of trade are frequently unfavorable to the upland groups. As a result, upland residents are often resentful of lowlanders in general, including government agencies seen, often correctly, as representing lowland interests. This antagonism, described in detail in the Himalayan areas in Chapter 11, can impede acceptance of even extremely well-motivated programs for upper watershed management. In contrast to the groups specifically adapted to particular resources and environmental conditions and involved in trade networks within the watershed, there are also groups whose settlement and subsistence patterns crosscut different portions of the landscape. In North Pakistan, for example, certain ethnic groups display a pattern of transhumance, that is, seasonally shifting between valley bottom agricultural lands and summer pasturages on neighboring hillsides and mountain slopes (Barth 1956).

Directional Emphasis of Social Processes Although both people and social and cultural influences are equally able to spread both upstream and downstream, the dominant trend, especially during the modern era, seems to be in an upstream direction. Rapidly increasing lowland/urban populations are now expanding into forested upland areas in search of new agricultural lands. Lowland-based governments

86

George W. Lovelace and A. Terry Rambo

and societies extend their transportation and communication networks into these upland areas to increase and enhance resource exploitation and to allow for greater control (Romm 1981). In some instances, particularly under the increasing influence of mass communications and educational programs, uplanders are also attracted to the lowlands and urban centers seeking new opportunities in employment and education and a "better" way of life. Regardless of direction of movement, however, the strong and increasing impacts of the lowlands and "lowland culture" upon the uplands are unmistakable in both developed and developing countries. SOCIAL AND BEHAVIORAL OBSTACLES TO INTEGRATED WATERSHED MANAGEMENT Behavioral and social phenomena operate on a number of different levels (e.g., individual, community, intergroup, national and international) with considerable variability within each level. Social factors also show up in the context of institutions and organizational arrangements (an in-depth discussion of nature and roles of institutions in watershed management is presented in Chapter 7). Many of the variables contained in the conceptual framework for watershed management offered in Chapter 2 have social and behavioral dimensions. The outcome of a rural development project or environmental management strategy can be affected by social and behaviorial elements associated with any of these levels and variables.

Natural Versus Sociopolitical Units and Boundaries Cultural and social factors and historical events have frequently resulted in political units and boundaries that bear little resemblance to ecological reality. This, as pointed out in Chapter 1, often makes project management extremely difficult. To the extent that different portions of the same watershed or river system are controlled by different nations, as in the case of the Mekong River, integrated management may be faced with virtually insurmountable barriers. A related problem is faced by environmental planners and managers in Bangladesh. Rural development and environmental programs in that lowland country must frequently confront environmental problems aggravated, if not caused, by improper land use occurring upstream in India and Nepal. This problem is not confined to situations involving national boundaries. It can occur within countries, not only in provincial or similar units, but also in governmental agency boundaries and mandates. It is common to find different portions of the same watershed under the administration of different agencies. Upper watersheds, for example, are typically administered by forestry-related agencies, while hydrological processes are under the control of hydroelectric power- or irrigation-related authorities.

Behavioral and Social Dimensions

87

Class and Ethnic Relations Watershed land use and management may also be affected by class and ethnic relations. Conflicts between different groups in the same watershed, or between groups that reside in the watershed and lowland-based governments, are common. The upper watersheds of many parts of the world are inhabited by poor farmers and by ethnic minorities who are only partially assimilated into national cultures. These groups often perceive themselves as being, and frequently are, discriminated against. Recent migration of lowlanders into upland areas has also contributed to increasing social tension throughout the developing world. Further, the upper watershed with its difficult, yet easily defended, terrain often offers a place of refuge for social and ethnic groups that are out of favor with central governments. Consider, for example, the difficulties the Burmese government would face in attempting to plan and implement integrated watershed management in upland northeast Burma, an area inhabited by independence-minded and often armed Shan cultural groups. Even where actual conflicts do not exist, it may still be difficult to obtain cooperation from and between societies that have traditionally operated with considerable independence and with little positive regard for each other. The case study from Thailand in Chapter 12 illustrates some of this concern. Attempts to convince upland groups, whose life-sustaining environmental activities are having a negative effect downstream, that they should change for the benefit of lowlanders are likely to fail unless special incentives are provided. Competition between different social classes and ethnic groups for the same resources lies at the heart of many of these problems. Upland groups with little or no political power are usually suspicious, often with good reason, of development programs initiated from lowland/urban-based political and administrative structures. (See Chapter 11 for further discussion of this issue.)

Planners and Planees Another area requiring brief comment involves what might be referred to as "planner-planee relationships:' In many, if not most, countries, there is considerable distance (social, economic, cultural, technological, educational, political, and spatial) between those individuals in charge of resource planning and project implementation, and those individuals and groups whose land-use practices are the subject of management. Typically, decision makers, administrators, and many field workers are from urban-elite rather than rural-poor backgrounds, are lowland oriented rather than upland oriented, and are from ethnic groups other than those that inhabit the watershed. The involvement of international development agency representatives and foreign consultants frequently makes the pattern of class and ethnic interaction even more complicated.

88

George W. Lovelace and A. Terry Rambo

Given the gulf that often exists between planner and planee, it is not surprising that neither fully understands nor appreciates the other and that each tends to act toward the other in terms of vague, often incorrect, and usually negative stereotypes (Jamieson 1984). Therefore, one should not be amazed that upland watersheds are often managed and developed largely in terms of what they have to offer to lowland and urban areas. This lowland/urban bias and the lack of understanding or appreciation of the needs and interests of other groups often hinder cooperation from upland societies and prevent the successful implementation of integrated approaches to the watershed.

CONCLUSION As should be clear, given the diversity of social groups and institutional arrangements found in the world's watersheds, it is not possible to prescribe an exact format for behavioral or social analysis of watersheds. The purpose of the foregoing discussion is to point out the diverse and sometimes subtle ways in which social factors and forces affect watershed land use, planning, and project implementation. Experiences in rural development have repeatedly shown that the failure to take account of local social and behavioral conditions can hinder project success. In order to design successful watershed management projects, information about the local social and behavioral context needs to be collected and integrated into project design and implementation. The range and diversity of potentially relevant research topics point out the need for many different kinds of research-especially the need for cooperative, multidisciplinary research efforts. Understanding of specific human-environmental interactions and their consequences, for example, will frequently require intensive field studies and comparative analyses by researchers trained in fields such as human ecology, rural sociology, anthropology, agronomy, and resource economics. The gathering and analysis of data related to larger issues such as social and ethnic relations and political processes, on the other hand, may require broader-based social surveys, studies of ethnicity, political and socioeconomic analyses and historical studies. Research on these latter topics often deals with issues about which government authorities and villagers may be extremely sensitive. In developing a watershed management plan, the following social or behavioral considerations need to be addressed:

• Baseline data. Because general approaches to watershed management must be adapted, refined, or modified to enable their successful use in different settings, there is a crucial need for baseline data on existing social and behavioral conditions and human-environmental relations.

Behavioral and Social Dimensions

89

This initial data base should be viewed as a prerequisite for successful management in the same way that prior knowledge of existing biophysical conditions and processes is crucial for natural resources planning. An attempt must also be made to assess the interrelated social and environmental consequences of alternative resource management actions and institutional changes.

• Identify different sociopolitical groups. Just as a watershed can be examined in a biophysical sense to analyze differences in soils, topography, and land use, a watershed should also be examined in a sociopolitical sense to identify different social or ethnic groups living in the watershed. This is crucial to designing projects with objectives that involve economic, biophysical, and social concerns. Watershed management projects frequently fail if any of these concerns are not addressed.

• Recognize the cultural basis of different land-use patterns. Any given watershed may have a variety of different land-use patterns. These patterns are frequently as much a result of social, cultural, or rural economic factors as physical factors. Programs may have to be designed to meet different human needs in different parts of the watershed, even if the physical conditions are the same.

• Pay special attention to ethnic minorities and the rural poor. The upland areas of many watersheds are inhabited by ethnic minorities and by groups that represent the poorest and most disenfranchised segments of society. These groups usually depend on the watershed to meet basic requirements of life. The special perspectives and needs of these groups have to be taken into consideration. This is especially true if the lowlandupland, top-down planning frictions and tensions are to be avoided. Direct and indirect impacts of proposed changes in watershed utilization upon these groups should also be determined. Where these impacts are negative, alternative courses of action or programs for mitigation should be considered.

• Learning from other areas. The information and institutional requirements necessary to address and solve social and behavioral problems are tremendous. Some of the needed information is now being collected in the context of specific development projects, but much more data need to be gathered and interpreted. To a certain degree, it may be possible to transfer knowledge that has been gained and developed in other parts of the world. But just as there are problems in applying temperate zone environmental models to tropical environments, there are also obstacles to transferring models of behavior, social relations, and social change that have evolved in one ethnic, social, and cultural context to the often different contexts of the developing world.

90

George W. Lovelace and A. Terry Rambo

REFERENCES Barth, F. 1956. Ecologic relationships of ethnic groups in Swat, North Pakistan. American Anthropologist 58:1079-1089. Drew, D. 1983. Man-Environment Processes. London: George Unwin and Allen. Dunn, F.L. 1975. Rainforest Collectors and Traders: A Study of Resource Utilization in Modern and Ancient Malaya. Kuala Lumpur: Malaysian Branch of the Royal Asiatic Society. Hamilton, L.S., and P.N. King. 1984. Watersheds and rural development planning. In Traditional Life-Styles, Conservation and Rural Development, ed. J. Hanks, 80-86. IUCN Commission on Ecology Papers 7. Gland: International Union for the Conservation of Nature and Natural Resources. Jacobson, T., and R.M. Adams. 1958. Salt and silt in ancient Mesopotamian agriculture. Science 128:1251-1258. Jamieson, N.L. 1984. Multiple perceptions of environmental issues in rural Southeast Asia: The implications of cultural categories, beliefs, and values for environmental communication. East-West Environment and Policy Institute Working Paper, Honolulu, Hawaii. Norgaard, R.B. 1981. Sociosystem and ecosystem coevolution in the Amazon. J. Environmental Economics and Management 8:238-254. Pei, Sheng-ji. 1984. Plant products and ethnicity in the markets of Xishuangbanna, Yunnan Province, China. Paper for Conference on Ethnic Diversity and the Control of Natural Resources in Southeast Asia, University of Michigan Center for South and Southeast Asian Studies, Ann Arbor. Rambo, A.T. 1983. Conceptual Approaches to Human Ecology. Research Report 14. Honolulu, HI: East-West Environment and Policy Institute. Romm, J. 1981. Environment and development in Southeast Asia: Trends, themes and issues. In A Colloquium on Southeast Asian Studies, eds. Tunku Shamsul Bahrin, C. Jeshrun, and A.T. Rambo, 195-218. Singapore: Institute of Southeast Asian Studies.

CHAPTER 7

Institutional and Organizational Concerns in Upper Watershed Management Christopher J.N. Gibbs

The conceptual model described in Chapter 2 identifies three major dimensions of watershed management, one of which involves a planned system of resource management actions, implementation tools, and institutional and organizational arrangements. This chapter explores the meaning and significance of institutional arrangements for watershed management with special emphasis on upper watersheds. Since very little institutional analysis of watershed management has been completed, conclusions must be tentative. However, a persuasive argument can be made that our understanding of institutions for watershed management is weak and should be strengthened. The word institution is used here to include two distinct but complementary concepts: (1) institutional arrangements, which define property rights in watershed resources and the rights and obligations of individuals and groups, and (2) organizational arrangements, which include the ordered groups of people who use watershed resources purposefully. The distinction made here is somewhat artificial in that the legitimacy of organizations is closely linked to the prevailing institutional rules. However, maintaining the distinction facilitates the exploration of the separable effects of rights of access to watershed resources and collective action for management. WATERSHEDS AS SECONDARY REGIONS Watersheds are defined by hydrological processes that make functional sense for foresters, hydrologists, and soil conservationists. However, while understanding of the technical aspects of watershed management is progressing, knowledge of the influence of institutional and organizational factors in the special context of upper watersheds as secondary regions remains relatively undeveloped. 91

92

Christopher J.N. Gibbs

Geographically, economically, politically, and organizationally, regions can be classed as either primary or secondary. Within this classification system, secondary regions are those that are marginal in all these categories (Koppel 1981). Geographically, secondary regions are often remote and inaccessible-at the periphery rather than the center. Economically, secondary regions are at the lower margin of production, where, even under optimum conditions, land barely yields enough to cover costs of production. People in secondary regions are politically weak and are not well assimilated into the larger political system. Organizationally, they often appear to be beyond the effective reach of national agencies of government administration. Chapter 6 highlighted some of these aspects in its discussion of the "planner-planee" relationships. The primary regions of Asia are the highly productive core areas in the lower watersheds that include the valleys of the continent's major rivers. These core areas are the densely populated and highly productive alluvial lands that have supported the rise of the great civilizations of Asia (Rambo and Sajise 1985) and produce the bulk of the major staple cereals today. Core areas are closely and efficiently linked to markets, sources of information, and centers of decision making. It is in the core areas of Asia that the "Green Revolution" has had undoubted success at developing intensive culture techniques for the relatively homogeneous and ecologically stable lands where institutional and organizational arrangements are well defined. In contrast, the secondary regions of Asia are characterized by their ecological and social diversity. Secondary regions are composed of many small, local ecological and social systems that are distinct and rarely well understood. This distinction between primary and secondary regions is of fundamental importance to watershed management, since approaches to development that have succeeded in the core areas are likely to be inappropriate in the hinterlands. Approaches that apply packages of technology to broad areas in primary regions will not work in the diverse and less accessible secondary regions. Watershed management as public policy is advocated most frequently on the grounds that sustainable uses of economic resources of national importance are being diminished. For instance, the degradation of public forest lands is of concern because the decreased productivity of these areas results in a weakened capacity to contribute to foreign exchange. Human actions by an upstream minority are perceived as producing heavy economic costs for the downstream majority. Frequently upland land uses are evaluated in terms of the potential threat posed to downstream investment or interest, such as reduction of the economic lives of investments made in infrastructure for irrigation, power generation, or flood control. Consequently, the traditional focus of watershed management in upland areas has been on biophysical factors that relate land uses to land cover, slope,

Institutional and Organizational Concerns

93

soil, rainfall, and the hydrological cycle. Land management measures that protect infrastructure and downstream production systems by reducing runoff and soil loss are stressed. Although deforestation, soil erosion and sedimentation are recognized as significant problems, the welfare of people in upper watersheds is rarely given adequate priority. In fact, evidence shows that considerable damage to upper watersheds is the result of deliberate national policies of resource utilization that seldom benefit watershed residents. Three principal causes of natural resource degradation in the tropics have been identified (Bromley 1985): (1) the deliberate exploitation of natural resources for export to earn foreign exchange; (2) national conservation policies that fail to take account of local resource management systems; and (3) the unintended side effects of development policies that stimulate movement of population into watershed areas. These are all institutional matters. INSTITUTIONS AND ORGANIZATIONS Institutions, as they are defined here, include the rules and rights that define people's relationships to resources. For example, water management has a technical and an institutional side. In ''core area'' rice production, secure and efficient access to irrigation water is a function of both the physical arrangement of the system and the operating rules for water distribution, system maintenance and enforcement. Improving irrigation water management is therefore recognized as being dependent as much on farmers' property rights in water and organizational arrangements for system operation as on technical efficiency in system design. Similarly, institutional rules governing access to watershed resources-forest products, grazing land, and water-are as important as sound technical solutions to forest, soil, and water conservation. Organizations are ordered groups of people such as family farms, firms, and government agencies. The legitimacy of organizations is based on the institutional rules society accepts. Understanding organizations for watershed management belongs in the field of rural development management. The concept of management used here is based on cybernetics and the guided transformation of inputs into outputs, which require (1) an explicit set of objectives; (2) information about actual outputs relative to the objectives; and (3) the ability to learn from feedback and respond to error. Knowing whether an organization is "managed" versus simply "administered" is signaled by the degree to which the organization knows the results of its behavior and responds to them (Seckler and Nobe 1983). For watershed management, questions about the appropriateness of different organizational forms, styles, and performance have rarely been asked. However, the existing lessons of management science obviously need to be adapted carefully to secondary regions in developing countries.

94

Christopher J.N. Gibbs

Institutions and Incentives for Individual Action Understanding institutions and property rights is essential to effective watershed management. Institutional questions arise about individuals and groups in the watershed, and between upstream and downstream populations. With respect to individuals, important property rights questions exist about land, grazing, and tree tenure. Perhaps the most important institutional factor affecting the watershed occupant is land tenure. Land tenure includes the legal, contractual, and customary arrangements whereby people gain access to productive opportunities in land (Dorner 1972). In watershed areas throughout Asia, customary or traditional land tenure systems persist despite the recent introduction of formal legislation. Frequently there is no relationship between what legislation requires and what actually occurs. Who has legal rights to own or allocate land and who, in fact, has possession are typically at odds. This may occur because recent and customary laws are in conflict, or because the modern state simply does not have the capacity to reach into secondary regions. When land tenure in the watershed is confused, it is important to understand who has effective possession. If tenure is insecure, watershed occupants are unlikely to invest in long-term or slow-maturing activities, or to be innovative or risk-taking. They may rationally opt for activities that deliver benefits predictably and fast. Even though they may be aware that tree crops, terracing, or live fences progressively stabilize and improve land, only those who control the land they farm will consider investing in these and similar improvements. The tenant who expects to be shifted or evicted by the landlord to prevent the establishment of title by prescription cannot plan for slow-growing perennial crops or soil-conserving infrastructure that pay off years in the future {Thomson 1981). Farmers without secure access to land cannot be faulted for hesitating to plant trees or terrace hillsides. For forests in watersheds, rights may be customarily shared among those who collect forage, those who gather fuelwood, and those who graze livestock, and land and tree tenure may not coincide. For example, deadwood and unprotected live trees may be unregulated, open access resources; that is, there are no formal or informal conventions establishing individual or group rights to wood. The first person to appropriate the wood keeps it. When land and trees are abundant, this presents no problem. When land and trees become scarce, open access can drive resources to extinction. When trees or grazing are managed collectively, conventions exist with respect to harvesting designed to keep the trees alive and maintain the pasture, that is, rules of common property apply (Wallace 1983). In Papua New Guinea where the tenure system ascribes rights in land to the clan or subclan, individual households retain specific rights to parcels of land as long as they contain economically valuable plants (Wein-

Institutional and Organizational Concerns

95

stock 1984). When trees are cultivated, individuals and households secure tenure rights over the trees, which can last for decades provided the trees remain productive. Where trees are private property, tree owners either protect them or bear the losses incurred by theft (Thomson 1981). Villagers' abilities to protect trees depend largely upon the pattern of settlement, the willingness of the community to police each other's trees, and the incentives of those who take fuelwood or forage, or graze seedlings. Privatization brings home the consequences of tree use to the individual directly, but privatization is not necessarily the answer to watershed management.

Institutions and Incentives for Collective Action In private markets for goods, you do not consume if you do not pay; however, watershed management provides numerous examples of the difficulty of denying benefits to persons who do not contribute. If a person can benefit without contributing, the person becomes a "free rider:' Examples of free-rider benefits include fuelwood or forage taken from common lands. The free-rider problem poses a major obstacle to expanding and maintaining protective land uses in watersheds that may be counteracted with collective action by users. Organizations for collective action in forestry must recognize both individual and group interests, as gainers and losers, and provide sufficient leadership and incentives to overcome individual resistance. Before participating in collective action, the individual will weigh in advance his or her expected costs, expected benefits, and the viability and skill of the project leadership (Popkin 1981). Collective action requires more than consensus or an intense need; it requires conditions under which villagers will find it in their individual interests to allocate resources to a common purpose and not be free riders. If an individual estimates that his or her contribution to a collective action has little or no impact on the contributions of others or the amount of the goods supplied, then special incentives may be needed to produce action in the group's interest. Agencies charged with administering watersheds in Asia have typically responded to the failure of collective action with formal regulations and police power. Regulations may prohibit entry into watershed areas; may outlaw practices such as cutting, burning, or grazing; or may seek to protect particular trees. Frequently regulations leave the responsibility for forest protection to the forest service alone, which ends up policing a diminishing forest base. While the legal boundaries around common and public lands have hardened, however, the gap between de jure and de facto control over land-use decisions has widened. Effective watershed management, on the other hand, requires greater compatibility between those with authority over land use and those who possess it and control use in practice (Romm 1979).

Christopher J.N. Gibbs When demand for fuel or forage exceeds supply and subsistence needs must be met, police power usually fails to protect the forest. In some cases, such as in Nepal, the introduction of national legislation to replace customary law resulted in an accelerated level of deforestation as the responsibility for forest management was transferred from the local community to the state. Nepal's forest was a managed common property resource, but villagers could turn the forest into a private resource by converting it to farmland. 96

Much land that is converted from forest to farmland is more valuable as forest. However, as forest land, no common owner can capture all of its value, so villagers convert it to private farmland, and the new private owners capture all of its reduced value (Wallace 1983). Following nationalization of the forests in 1957, Nepalese villagers saw their traditional rights of access to the forest curtailed and local responsibility for forest protection disappear. Communal responsibility for the forest was lost and the forests were converted from common property resources (group property) to open access resources (nobody's property), or to private farmland. Now Nepal has amended its forest legislation to recognize once again the responsibilities of communities as forest managers through more decentralized administration and panchayat forestry.

Institutions and Incentives for Resource Conservation Resource conservation has been defined as the redistribution of rates of resource use toward the future (Ciriacy-Wantrup 1968). Depletion is the economic opposite of conservation and means the redistribution of rates of resource use toward the present. Whether individuals or groups choose to conserve watershed resources will depend largely upon their estimates of the cost of delaying consumption and the level of certainty with which they will capture the future benefit. The cost of delaying consumption is dependent on the expected level of future costs, revenues, and interest rates. The level of certainty about capturing benefits is dependent to a large extent on the pattern and stability of property rights. Institutional uncertainty creates incentives to use watershed resources when they are available, for they may not be available later. Where property rights are insecure and organizational arrangements are weak, the incentive exists to deplete resources. Uncertainty flourishes when institutional arrangements are only suggestive and when rules are subject to frequent changes (Bromley 1982). The presence of institutional uncertainty means that benefits from watershed management are frequently captured before they mature by only a subset of forest users-those able to take advantage of the weak institutional structure. For watershed communities in Asia,

Institutional and Organizational Concerns

97

institutional uncertainty creates an environment that favors resource depletion and penalizes conservation. A distinguishing feature of resource conservation in a watershed context is that while the costs are borne upstream, many of the benefits accrue to populations downstream. Reforestation and slope stabilization may yield some benefits to watershed communities but the greatest benefit may be the prolonged life of a reservoir producing irrigation water, hydroelectric power, and flood control benefits for urban and agricultural communities in the lowlands. Watershed communities have little incentive to take account of these benefits, and investment in watershed management will be less than optimal from society's viewpoint as a whole. For this reason, the size and division of benefits and responsibilities between public and private interests upstream and downstream must be understood before institutional arrangements can be developed that help establish congruence between upstream and downstream interests.

Organization for Watershed Management Formal responsibility for upper watershed management in Asia typically rests with forestry, power generation, and soil conservation agencies. The roles of these agencies are usually defined in terms of biological or physical variables related to forests, forest lands, soil, and water for the benefit of downstream populations. Their jurisdiction is usually restricted to areas above a certain elevation or degree of slope, with no special relationship to agencies downstream. Analysis of the structure and performance of watershed management agencies has rarely been seen as a priority to be undertaken systematically. However, it should not be difficult to apply management theory to watershed management provided the variability and uncertainty of the production environment and the importance of people as the ultimate "producers" in the watershed are fully recognized. In most watershed situations, because of ecological and social heterogeneity, managers must adapt planning and implementation to fit the needs of particular localities. The "primary planning level" must be as close to the village or the farmer as an agency's management capacity permits (Bottrall 1981). At this primary level, planning and implementation must be "enabling" rather than "executive?' The extent to which agencies can effectively devolve power and responsibility to facilitate local initiative will determine, to a large extent, whether upper watersheds are managed sustainably for the benefit of both upstream and downstream populations. For foresters, engineers, and public administrators to work in this way will require new attitudes, new skills, and new procedures that must be learned. Building organizations with a capacity for learning is a necessary condition for watershed management since the basis for management from present experience typically does not exist. Furthermore, traditional ap-

98

Christopher J.N. Gibbs

proaches to program planning assume that the knowledge required to design programs comes from specialists and can be generated independently from the organizational capacity that will implement it. However, recent evaluations of rural resource management projects that have delivered benefits to target communities effectively have demonstrated the need for a learning approach that plans programs with the target beneficiaries and learns how to be effective from experimental action programs (Korten 1980). Successful development plans are not only "thought through" but also "acted out" (Johnston and Clark 1982).

Decentralization and Participation Upper watershed areas are typically remote from the centers of administration, frequently inaccessible, and generally full of growing numbers of people who subsist through application of stable but unproductive and unsustainable technologies. National agencies charged with watershed management, staffed by professional foresters and engineers, are typically highly centralized organizations with a limited reach. They lack the capacity for sustained local-level action, find adapting to local circumstances difficult, have difficulty dealing with people directly, and are generally not equipped with the appropriate technologies. For these reasons, watershed management agencies must recognize the need to develop local organizations and participatory approaches to management. Analysis of participatory approaches to irrigation water management suggests that water users organize to perform certain basic activities that are difficult to conduct independently. These activities are (1) decision making about design, construction, or operation of structures, water acquisition or distribution; (2) resource mobilization to acquire funds, material, manpower, or information to perform tasks; (3) communication about activities, decisions, or results; and (4) conflict management among water users and between water users and others (Uphoff 1985a). Benefits derived from community participation in irrigation management include the provision of resources (capital, labor, information, and organization) and the capacity to monitor the performance of national agencies at the local level. Small-scale, communally managed irrigation projects also show high positive returns on investment. Similarly, watershed users' organizational activities must be evaluated if national management agencies are to learn when and how to work collaboratively with watershed communities. Participatory approaches to development "fit projects to people" (Uphoff 1985b). Building national capacities for participatory management of rural resources has probably been taken furthest in irrigation management in the Philippines (Korten 1983). This experience suggests guidelines t~at may be transferable to upper watershed management. These guidelines suggest that strong local organizations with clear authority over

Institutional and Organizational Concerns

99

resource management are more effective than weak organizations; where local organizations for resource management exist, it is advisable to build upon them; local organizations must contribute to project costs, even if the contribution is minimal; local organizations must be involved in planning and implementation from the inception of the project; and national agencies must build their capacity to work purposefully with local organizations of resource users. Experience from irrigation also suggests that serious obstacles to developing participatory approaches to rural resource management exist at the agency, community, and societal levels (Korten 1981). Obstacles within implementing agencies include centralized decision making, values, and attitudes that overestimate the agency's influence and underestimate villager's skill and intelligence; inappropriate value systems that stress bureaucratic objectives; and frequent transfers of personnel that prevent villagers and officials from developing mutual understanding and respect. Even when it is most appropriate, however, participation cannot merely be ordered to take place. Instituting participatory approaches to management requires strong leadership at a high level, a good understanding of the changes needed, and time to implement them. Obstacles to participatory approaches to rural resource management within poor communities include lack of appropriate local organizations and leadership skills that have often failed to develop or been suppressed; poor community facilities; factionalism and differences between economic, tribal, or religious groups in the same community; and corruption. The structures, procedures, and skills needed by communities to implement participatory developments must be tailored to the program and the area if they are to be effective. Obstacles within society as a whole include political objections by the resource-rich to the transfer of resources to the resource-poor; inappropriate legislation; lack of incentives to reward appropriate action; and centralized control of programs and budgets. The combined result of these obstacles is failure in implementation even when participatory approaches to management are recognized as appropriate.

Success (in participatory approaches) requires transformations in the way an agency performs its task, in the way community members relate to each other and to the agency, and the way society views the poor, their needs, their capacities and their rights. Such transformations are inevitably slow and liable to setbacks but the reasons for seeking participation are compelling (Korten 1981 ). Unfortunately the systematic exploration of when participatory approaches to watershed management are appropriate and how they are introduced has barely begun.

100

Christopher J.N. Gibbs ---

--

--

---

CONCLUSIONS AND IMPLICATIONS FOR WATERSHED MANAGEMENT Many watershed management programs are aimed at upland and mountain areas, including remote and inaccessible parts of secondary regions. The focus of these programs is usually on reforestation and soil and water conservation. Interest in watershed management often begins when the economic life of investments in infrastructure, made for the benefit of core area populations downstream, is threatened. The threat may come from natural forces or from human action, but the response is typically aimed at controlling deforestation, runoff, and soil erosion through physical and biological means or through controlling the behavior of watershed occupants by police power. Even though the people most affected by mismanagement are watershed occupants themselves, their welfare is rarely the focus of the response. For purposes of development, the predominant characteristics of upper watersheds are their ecological and social diversity. Approaches to development, which have worked well in the simpler and more homogeneous lowlands, cannot be transplanted wholesale to the uplands. Improved technologies for watershed management must be tailored to the specific needs of particular settings. Because of their economic marginality, the level of investment in sustainable production systems for upper watersheds that can be justified economically must be carefully monitored. Although major emphasis in watershed management has been placed on techniques for soil and water conservation, our understanding of essential institutional and organizational arrangements is still rudimentary. However, based on this limited experience and lessons drawn from other sectors such as irrigation, many conclusions and implications for watershed management in Asia can be drawn. • Upper watersheds are frequently secondary areas-geographically, economically, politically, and administratively marginal and extremely diverse. Therefore, a fundamental challenge for management is to cope effectively with this marginality and diversity. • Upper watersheds are occupied by expanding numbers of people whose survival strategies condemn them to real poverty and may endanger the sustainability of the watershed and areas downstream. Watershed managers must learn to put the needs of watershed populations ahead of biophysical factors. • Creating institutional and organizational arrangements that complement techniques for watershed land management must become a priority. • In upper ~aters~ed areas, customary institutional arrangements frequently still dommate and managers must learn to recognize these and adapt their planning and implementation to reflect them.

Institutional and Organizational Concerns

101

• Institutional arrangements can be adapted to create incentives for individuals and groups in the watershed to practice management measures that are both productive in the short term and sustainable over the long term, but this requires creativity, experimentation, and participation. • In watershed areas the distributional economics of land and water management is as important as economic efficiency. Watershed management actions may create gainers and losers, both within the watershed and between upstream and downstream populations. If significant numbers of upper watershed occupants are losers, strong incentives to undo the work of the watershed manager will exist. • Formal organizations for upper watershed management in Asia are typically national bureaus charged with management of forests, hydroelectric power generation, or soil conservation. People in the watershed are often seen as obstacles to discharging these responsibilities. Watershed management organizations must learn how to work with people in the watershed to enable them to become effective implementors of managerial actions. • Local organizations for watershed management can take over planning and implementation for watershed management where national agencies do not have a sustainable reach. Lessons from irrigation water management can provide some guidance on why and how people organize to perform certain actions collectively. However, learning to devolve power to local levels will be difficult. REFERENCES Bottrall, A. 1981. Comparative study of the management and organization of irrigation projects. Staff Working Paper 458. Washington, D.C.: The World Bank. Bromley, DW. 1982. Land and water problems: An institutional perspective. American J. Agricultural Economics 64(December):834-844. Bromley, DW. 1985. Natural resources and agricultural development in the tropics: Is conflict inevitable? Invited paper for International Conference of Agricultural Economists, 24 August-4 September, Malaga, Spain. Ciriacy-Wantrup, S.V. 1968. Resource conservation: Economics and policies. Division of Agricultural Sciences, University of California, Berkeley. Dorner, P. 1972. Land Reform and Economic Development. London: Penguin Books. Johnston, B.F., and W.C. Clark. 1982. Redesigning Rural Development: A Strategic Approach. Baltimore: Johns Hopkins University Press.

102

Christopher JN. Gibbs

Koppel, B.M. 1981. Food policy options for secondary regions. Food Policy (February). Korten, D.C. 1980. Community organization and rural development: A learning process approach. Public Administration Review (September/October). Korten, F.F. 1981. Stimulating community participation: Obstacles and options at agency, community and societal levels. Rural Development Participation Review 2(3), Spring. Korten, F.F. 1983. Building national capacity to develop water users' association: Experience from the Philippines. Staff Working Paper 528. Washington, D.C.: The World Bank. Popkin, S.L. 1981. Public choice and rural development: Free riders, lemons, and institutional design. In Public Choice and Rural Development, eds. C.S. Russell and N.K. Nicholson, 43-80. Research Paper R-21. Washington, D.C.: Resources for the Future. Rambo, A.T., and P.E. Sajise. 1985. Developing a regional network for interdisciplinary research on rural ecology: The Southeast Asian Universities Agroecosystem Network experience. June. Romm, J. 1979. The uncultivated half of India. Ford Foundation, New Delhi. June. Seckler, D., and K. Nobe. 1983. The management factor in developing economics. In Issues in Third World Development, eds. K. Nobe and R. Sampath. Boulder, CO: Westview Press. Thomson, J.T. 1981. Public choice analysis of institutional constraints on firewood production strategies in the West African Sahel. In Public Choice and Rural Development, eds. C.S. Russell and N.K. Nicholson, 119-152. Research Paper R-21. Washington, D.C.: Resources for the Future. Uphoff, N. 1985a. Developing water management institutions in Sri Lanka. Paper for ASPA Panel, Bloomington, Indiana. March. Uphoff, N. 1985b. Fitting projects to people. In Putting People First: Sociological Variables in Rural Development, ed. M.M. Cernea. A World Bank publication. New York: Oxford University Press. Wallace, M. 1983. Managing resources that are common property: From Kathmandu to Capitol Hill. J Policy Analysis and Management 2(2). Weinstock, J. 1984. Tenure and forest lands in the Pacific. East-West Environment and Policy Institute Working Paper, Honolulu, Hawaii.

CHAPTER 8

Program Implementation K. William Easter

Implementation problems in watershed projects are evidenced by damagec or poorly constructed terraces, check dams in disrepair, excessive rates o erosion and sedimentation, and cropping systems that do not provide ~ basic subsistence for farm families. To determine the real problems, on< must look behind these physical and economic manifestations of failure The real reasons for failure often include little or no local participation inadequate information, exclusion of downstream interests, inadequate de velopment and testing of technology, conflicting viewpoints among vari ous interests, lack of adequate extension education and technical assistance delays in delivery of key inputs (including financial resources), and a frag mented government management structure. These factors have been dis cussed in the preceding five chapters of this book, and each has an impac on project implementation. This is not an exhaustive list of problems plaguing attempts to imple ment projects for improving watershed management. However, it is like!~ that for most watershed projects many of these problems will occur to som< degree. Since these are difficult problems to surmount, they are typicall~ ignored by governments.

Good ideas do not fully materialize unless proper care is taken at the implementation stage. The saddest thing ... is that, whilst so much time and attention was paid to the technological component in the earlier stages of pre-planning, planning and the feasibility study preparation, there was not much attention paid to problems of actual implementation of the watershed management project. It was assumed that the presence of the existing extension machinery in the district would take care of the situation. This expectation was proved wrong (Jayaraman 1982, 97). 103

K. William Easter

104

Another example is the exclusion of downstream interests in upland watershed development. Even the agencies concerned with irrigation, power, and agriculture will normally be excluded since the upland watersheds are mostly administered by forestry departments. Because normal administrative boundaries will exclude downstream interests, new administrative arrangements such as special watershed districts or watershed councils may have to be designed to overcome this typical organizational weakness. PROGRAM IMPLEMENTATION To help understand why many programs never achieve their full potential, this chapter focuses on the implementation dimension or the last three stages of watershed management process presented in Chapter 2 (T, 1 to T" in Table 8.1). The concern is with how plans are implemented during the project's installation, operation, and maintenance phases. What resource management actions, implementation tools, and institutional and organizational arrangements are required in a project, and how are the implementation tasks performed? As pointed out by Ziemer this "how" is often forgotten:

The "how" is often thought to be completed with planning. Although sound planning is a major and necessary step in minimizing erosion, its implementation is all too often underplayed. The on-the-ground operator is the key to success or failure of a plan. Commonly, little effort is expended to include operators in the planning process. In general, their skills have

Table 8.1 The watershed management process and the watershed management elements for program implementation Management System Elements

Management Process Planning Design Installation Operation Maintenance T = Tasks

Resource Management Actions

Implementation Tools

Institutional and Organizational Arrangements

Program Implementation

105

been developed through personal experience of what seems to work. Unfortunately, what works best for dragging a log or constructing a stream crossing may not be best for reducing erosion. An important part of reducing steep/and erosion is successful interactions between planners and operators (Ziemer 1984, 4).

Resource Management Actions As discussed in Chapter 3, a wide range of resource management actions are possible. However, one of the most important aspects of planning resource management actions is to build flexibility in the original plan. This will help overcome implementation difficulties later on. As Galvez states:

Project implementation must be provided with a variety of alternative species and the flexibility of changing the combinations must be left open .... A fixed plan on the hectarage of a given species will only magnify the difficult task of implementing a reforestation project .... Area estimates for a given plantation purpose should be established at the planning stage of the project to provide reasonably accurate estimates of costs and benefits. This should not, however, tie the hands of the implementers when some of the species later on are found to be unsuitable or display unsatisfactory performances in the field (Galvez 1984, 27-28). Maintaining flexibility means that a watershed management program should not be based only on a single practice such as terracing or on an individual tree species for reforestation. One wants a program that can adjust to changing socioeconomic conditions. In watershed management, a range of practices, trees, and crops should be offered so that packages can be designed for different resource and economic conditions found within each watershed. Program managers should anticipate and even expect that the original plan and design will need to be modified to adjust to the unforeseen conditions during implementation. User input can help in providing program flexibility. ''Changeability is of considerable importance for project preparation and implementation which needs to be flexible. Such flexibility is inherent to farmer-centered methods during project implementation" (Hoare 1984, 31 and 34). Villagers can make additional information available to planners concerning physical conditions and what practices might be acceptable. The users are an important source of information and political support who have been ignored all too often. User participation should eliminate situations where one arrives at the implementation stage only to find that the recommended practices are either physically inadequate or socially unacceptable. In other words, users can help plan-

106

K. William Easter

ners develop an acceptable set of resource management actions that will meet program objectives. An illustration of the failure to involve users at the planning stage was found in the Highland Agricultural and Social Development Project in Northern Thailand, which planned coffee production for large areas. During the implementation stage, however, two-thirds of the project area was found below the optimum altitude for Arabica coffee and that many soils were unsuitable for coffee (Hoare 1984). It took more than a year to get the plan changed and fruit trees introduced into the project plan.

Implementation Tools As pointed out in Chapter 2, an important part of determining how plans are to be implemented is to make a distinction between resource management actions---''things to be done''---and implementation tools'---!'ways of getting things done:' Too little time is spent in considering alternative packages of implementation tools. Many times no one asks the questions: What implementation tools would be the most effective in encouraging adoption of the desired resource management actions, and who should apply the tools? There are many implementation tools that can be used in watershed management. For ease of discussion, they are grouped into the following four general classes: (1) legal arrangements such as zoning, regulations, controls, permits, prohibitions, and licenses; (2) monetary incentives or disincentives including prices, taxes, subsidies, fines, and grants; (3) improvements in knowledge and information through technical assistance, research, moral suasion, and education; and (4) direct public installation and investment. A watershed management plan may include only one implementation tool or may recommend a combination, such as permits, with fines for those who do not comply with permit restrictions. There even may be a federal subsidy or cost-sharing for certain practices specified in the permit. For long-term practices, loan subsidies may be available to the permit holders. There are a wide array of potential/ega/ arrangements for getting users to follow certain practices. With many of these arrangements, particularly prohibitions and strict zoning, enforcement costs can be quite high, particularly in developing countries. Licenses and permits have been used for managing forest harvesting with mixed results. Many times environmental restrictions included in the permits or licenses have been ignored. Again, part of this problem relates to enforcement difficulties. Monetary incentives or disincentives would seem to be particularly appropriate implementation tools since sedimentation problems are partly I. Implementation tools are much the same as the policy instruments mentioned in Baumol and Oates (1979).

Program Implementation

107

the result of market failure (i.e., imperfect knowledge, distorted capital markets, and externalities). Still, monetary incentives and disincentives have had only limited success primarily because of collection problems and lack of information concerning target populations. The low level of fees collected from water users in developing countries is a prime example of the difficulties involved with collecting fees from farmers in developing countries. With a large number of small-scale users scattered over a wide area, it is difficult to implement an information or collection system that is inexpensive, yet effective. The potential for using modest subsidies (incentives) either through direct cost-sharing or low interest rate loans still needs to be explored. This is particularly true in areas where a significant amount of the benefits from on-site practices or land-use changes accrues to downstream farmers and other watershed residents. Thus, the subsidy would be used to encourage upstream farmers to apply practices that have social benefits greater than costs but where private benefits are less than costs (see Chapter 4 for more details). Yet many governments may find it hard to implement such a program due to severe budget constraints and the difficulties involved in distributing cost-sharing to the appropriate target population. Even the United States has had only limited success in targeting soil conservation costsharing funds to critical areas or farms. Technical assistance, extension education, and improved information appear to be an important part of a successful watershed management program as illustrated by the case studies in Chapters 11 and 12. Once an effective set of crops, trees, and management practices have been developed through research and testing under local conditions, getting the information to users is the next step. This may involve training extension staff to work specifically with people in the upper watersheds or the reallocation of existing extension personnel to the target watershed areas. As discussed in the case of Northern Thailand in Chapter 12, extension can play an important role in identifying the watershed management problems through meetings with local people. The information thus obtained should be an important input to the development of workable resource management practices. It should also make it easier to obtain rapid adoption of practices once they have proven effective. Thus, the extension service can act as a conveyor of information and a source of information concerning target group characteristics, location, and research needs. Many governments have found that direct installation has been the quickest short-run strategy to use. It works fairly well for off-site practices that are put in place to lessen the effects of erosion. For on-site practices, the potential is much lower. Direct installation is a high-cost approach, which will be constrained by budget considerations as well as limited staff to maintain the structures. If farmers have not been involved in installing the structures, it will be difficult to convince them that they should do the

108

K. William Easter

maintenance. Finally, in upland areas where tenure arrangements are uncertain (many may be illegal), governments may have to install the practices unless they are willing to establish firm tenure arrangements for the users. Thus, as shown in Chapter 7, institutional and organizational arrangements can be critical in determining effective implementation tools.

Institutional and Organizational Arrangements Many problems in watershed management can be traced back to inadequate institutional arrangements. Lipton (1968) describes an Indian village where the farms are situated down a long slope. Soil quality varies from top to bottom along the slope, but varies slightly along a contour of the slope. If farm fields were divided horizontally along the contours of the hill, plowing would be cheaper, erosion would be reduced, and average output would be higher. However, each father avoids the problem of giving the best quality land to one son by dividing the land into vertical strips: "This saddles each generation of sons with longer, thinner sloping strips, increasingly costly and inconvenient to plough properly, i.e., repeatedly and across the slope" (Lipton 1968, 339). Thus, one of the first steps in reducing erosion would be to help families develop a scheme for comparing plots of different quality. In addition, an insurance program may be necessary to eliminate the need to have plots at different locations down the hill slope (Popkin 1981). In watershed management one wants to develop institutions that foster collective action and internalizes off-site impacts (Olson 1965). These two aspects, along with tenure arrangements, can play a critical role in the success or failure of resource management actions. Farmers with long tenure arrangements are more likely to invest in practices that maintain long-term soil productivity. Institutions that encourage farmers to work together to minimize off-site damages and provide collective goods play an important role in upper watershed management. This is particularly true for watersheds farmed by a large number of small-scale farmers. Because of their small-scale farms and sloping landscape, these farmers are interdependent; thus, collective action is a necessity rather than a luxury. Cooperation is essential to limit livestock, to protect tree seedlings during reforestation, and to adopt soil conservation practices. On the organizational side, it is important to understand the capabilities of the various agencies involved in watershed management. Without adequate staff and resources to implement a watershed project, implementation is almost certain to be unsatisfactory.

An organization's capabilities are defined by its physical plant and equipment, its staff and the procedures it has developed to accomplish certain tasks routinely. The sad tales of implementation failures are frequently tales of insufficient foreknowledge of those assets and their limits (Kelman 1984, 78).

Program Implementation

109

A good example of this is the difficulty government agencies have in organizing to obtain farmer participation. Many times, officials from government agencies responsible for both watershed and irrigation management projects complain that the farmers will not participate in project maintenance. However, it is not surprising that farmers do not want to participate in maintenance after having been systematically excluded from the planning, design, construction, and operations phases. Farmers wonder why they should maintain a "government project?' The lesson here is to get users involved in planning and implementation as soon as possible. This will mean funds and staff for the specific purpose of organizing users so that they can provide useful inputs to the whole management process (Coward 1980; Bottrall 1981; Chambers 1980). However, to do this may require a basic change in a government agency's organization and incentive structure.

The improved layout resulting from farmer input in the design and construction stages causes better system performance ... and fewer operation and maintenance problems .... This would appear to be a potential incentive for the irrigation agency to incorporate farmers in the design process. But given the usual organization of irrigation agencies into separate divisions for design and construction on the one hand, and operation and maintenance on the other, the incentives to incorporate farmers may exist only at the very highest levels within the irrigation agency (Small 1982, 7). In watershed management the incentives to incorporate farmers may only exist at a government level above the agency or department. Forest departments usually want to exclude all farmers from the upper watersheds so that they can get on with the business of selling timber-growing trees. Only if other departments, such as public welfare or agricultural extension, are brought into the decision making will farmer participation be considered. For example, as discussed in Chapter I2, this required the decision making to be moved to the provincial governor level for a watershed project in Northern Thailand. EVALUATION OF IMPLEMENTATION The analytical framework for evaluating the implementation of watershed management projects involves the three management activities and three management system elements of the framework (see Table 2.2). The success of most watershed management programs depends on how well these activities and elements are understood and developed during project implementation. Implementation research can be used to identify watershed management problems and to evaluate watershed management programs (Figure 8.1).

110

K. William Easter

Institutional Milieu and Organizational Arrangements

Resource Management Actions

~

t

Implementation Research

Implementation Tools

Management Problems

Figure 8.1 Problems of implementation in watershed management and the role of implementation research.

To do this the three elements of management must be evaluated by the specific activities and tasks required to effectively complete a watershed program or project (see Table 2.2). A key part of the evaluation is to know the specific activities and tasks needed to obtain the planned outputs. "If eleven discrete activities were required to produce a given result, it would not be good enough to succeed at only ten. Failure at any step ... will cause the entire effort to fail" (Kelman 1984, 78). While Kelman's statement may be a little strong, it points out the importance of each activity and task. Implementation must be as carefully planned as the technical and physical aspects of the project and then put into practice. As Jayaraman (1982) so clearly stated, one cannot assume that the farmers or some government agency will do the proposed task.

Program Implementation

111

For example, soil conservation practices, which reduce short-term returns to tenant farmers, may be difficult to implement even when long-term benefits of land productivity are substantial. Also, construction of some types of structures such as check dams may be questionable if adequate administrative procedures and incentives do not exist for maintaining the structures. Incentive systems can be established for implementing soil conservation practices on rented land, and agencies can be set up with adequate funds and trained staff to maintain structures. But these "implementation incentives" and "institutional arrangements" must be planned for, along with planning for the physical control measures. Are there administrative procedures for offering incentives? Is there an existing sympathetic organization or agency that has standard operating procedures and the necessary equipment, or does the agency have people trained and motivated who can create the necessary capabilities? (Kelman 1984, 84). As pointed out in Chapter 11, this is a real problem when the program is targeted for the hill people living in upper watersheds.

Forestry Example For each management action, several candidate implementation tools should be considered and evaluated for feasibility, cost, and effectiveness. In a publicly owned forest harvested for commercial timber by a private contractor, the resource management actions should involve harvesting techniques including methods of timber removal and road design and maintenance. The implementation tools may then be a government license with specific requirements for harvesting techniques, accompanied by an inspection system and sanctions for failure to adhere to the requirements. In a governmental license-inspection approach to control private logging on public land, alternative combinations and levels of inspection and size of fines for noncompliance should be postulated, and their costs and effectiveness estimated. An alternative or addition to the fine would be to shut down the operation immediately after an infraction is discovered. The analysis could proceed by seeking answers to such questions as: How many unannounced inspections per year and what level of fines for noncompliance are necessary to induce at least a 90 percent probability of compliance? What would be the cost and administrative feasibility of such an inspection-fine system? Or, alternatively, should a government inspector be at the site 100 percent of the time, as is the case for dam construction? If the major cost of soil erosion is in downstream areas, then off-site activities may be easier to implement than on-site activities. For example, shelter belts could be established along all streams, or timber removal could be prohibited within 100 meters of a stream. These management activities might be easier for a forestry department to implement than on-site activities. The tools available to the forestry department for implementing these activities would then be evaluated to determine what would be the

112

K. William Easter

lowest cost procedure for controlling soil erosion. Part of the question would involve whether the forestry department would be sympathetic to the approach and if it had the necessary resources. By so analyzing each resource management action, implementation tool, and institutional arrangements in the implementation plans, detailed information on costs, effectiveness, and feasibility of implementation is provided. This information may lead to proposals to change the organizations and institutions or to modify specific resource management actions, which will increase project effectiveness and reduce costs of implementation. Such analysis could lead to formulation of alternative packages of resource management actions-implementation tools-institutional arrangements, and their ranking in terms of costs, benefits, and feasibility.

IMPLEMENTATION ISSUES Including program implementation as a key concern of watershed management plans has obvious implications for the planning process in that much more attention should be given to formulating detailed implementation plans and strategies than is now the case. The literature on policy implementation suggests that there are many common issues involved in effective implementation (Sabatier and Mazmanian 1983). These issues can be grouped under the two broad headings of political concerns and socioeconomic and technical concerns. A finer division is difficult because of the overlap among the social, economic, and technical issues.

Political Issues One of the key aspects of program implementation is the ability to deal effectively with many potential political problems (Blaikie 1985). Watershed management will almost always involve at least one level of government and many times three or four levels; thus, politics will be a major concern in program implementation. The issues are summarized in the following questions: l. Are objectives clear and consistent, and are the trade-offs among ob-

jectives understood? 2. Is implementation assigned to a sympathetic agency with adequate resources? 3. How can the number of clearances required and agencies involved in project or program implementation be minimized? 4. Is there cooperation among the various actors involved in the project or program? 5. Does the program or project have political support, and do implementing officials have good managerial and political skills?

Program Implementation

113

The first issue-clarity and consistency in objectives-is well understood, but many times it is difficult to do anything about. Because many watershed projects involve a number of potential beneficiaries with different objectives, compromises usually must be made. For example, how much weight should be given to the objective of reducing soil erosion and how much to raising the level of people's income in the region? Many watershed management programs have these two objectives that will be in conflict, at least in the short run (Sutadipradja 1985). The trade-off between these two objectives may not be easy to agree on since the people living in the watershed will give a higher weight to income than will the government or downstream interests. Second, it is important to assign implementation to a sympathetic agency with adequate resources. Most public bureaucracies develop a general policy orientation that can only be changed slowly and at considerable cost and delay. For example, a notable change in the regulation of pesticide safety occurred when enforcement was transferred from the U.S. Department of Agriculture to the Environmental Protection Agency (Sabatier and Mazmanian 1983). In addition, the sympathetic agency must have the necessary resources or be given new resources if the project is not to be delayed or neglected. For watershed management in developing countries, this organizational issue arises in assigning responsibility to the forestry, agricultural, public works, irrigation, land development, or electric power agencies. The third issue concerns the number of clearances or approvals required and the number of semiautonomous agencies involved in implementation. The more clearances required and agencies involved, the more difficult it will be to obtain agreement on basic objectives and to implement plans. Delays will occur as the lead agency attempts to negotiate a consensus.

In the absence of such goals consensus, there is every likelihood that opponents or lukewarm supporters of program objectives will be able to control sufficient clearance points to demand important concessions and potentially to scuttle the program .... This is particularly likely in intergovernmental programs .... The number of clearance points would not be so critical if central officials had sufficient sanctions and inducements at their disposal (Sabatier and Mazmanian 1983, 156). The fourth and closely related issue is the need to obtain cooperation between the lead central agency and the other participating organizations, particularly local government and village officials. A clear set of program

114

K. William Easter

goals and appropriate incentives and sanctions are important assets in obtaining this cooperation. Yet the central agency or agencies must be able to adjust to local conditions and needs if the program is to obtain local support and cooperation from village leaders. The integrated multidisciplinary approach necessary for effective watershed management does pose some special implementation problems related to issues 3 and 4. This approach almost guarantees that there will be a large number of clearances or approvals required and agencies involved (see Jayaraman [1982] for a good discussion of this problem in India). For example, the watershed management program in Indonesia involves no less than seven ministries (Sutadipradja 1985). Thus, special efforts will be needed to assure cooperation among the various organizations. The problem is magnified by watersheds that often cut across administrative and political boundaries. In addition, there will be five or six major land-use activities in these upland watersheds: forestry, grazing, agriculture, agroforestry, mining, and recreation. For some watersheds one would want to include transportation and housing as important associated land uses, which are a consequence of the preceding primary uses. Each of these land uses may be the responsibility of one or more different agencies, again making implementation of watershed management programs difficult. The fifth issue deals with political support as well as the management and political skills particularly important for an agency adjusting to changes in socioeconomic conditions, inconsistencies among program objectives, and weaknesses in program design. Most projects or programs will go through a period when the plan has to be adjusted to the specific conditions found in the local area. This certainly is true for watershed management programs. It is during these adjustment periods that effective management and political support is critical. As Jayaraman (1982) points out a key part of management in watersheds is to establish downward linkages (internal political support), horizontal linkages (external political support), and upward linkages to gain political support from higher executives. This will involve not only political support from various levels of government but also support from people living in the watershed. As pointed out previously, this will be a difficult task because most agencies are not well organized to obtain the support and participation of people living in the project area. In addition it may be difficult to get support from high-level government officials if they represent the interests of those who want to exploit the watershed resources. For example, government officials may get direct benefits from selling timber or mining concessions with minimal environmental regulations concerning extraction practices.

Program Implementation

115

Socioeconomic and Technical Issues Just as important as the politics involved in program implementation are the technical and socioeconomic questions. In fact, the program plan may have to be changed due to technical or socioeconomic problems that are not discovered until the program is being implemented. Included in these issues are the following: 1. Are there adequate financial and technical resources?

2. Is project or program flexibility adequate to adjust to possible changes in socioeconomic conditions, and is there an effective monitoring and information systems to identify the needed change? 3. How can the number of behavioral changes be minimized during project or program implementation? 4. Is there a valid causal theory with appropriate economic incentives underlying an economically feasible project plan? The first issue should not really need mentioning but it occurs again and again. Governments do not provide adequate financial and technical resources to install and operate water resource projects on a sustained basis (see Jayaraman 1982). Inadequate maintenance of irrigation systems is a classic example in developing countries (Easter 1985). Neither adequate funds nor technical resources are provided to perform the well-known maintenance tasks. This same problem is likely to arise with watershed practices now being installed. One strategy for dealing with the problem of inadequate financial and technical resources is to target resources in critical areas. In fact targeting financial and technical resources can result in a more cost-effective program. As pointed out by Ziemer,

Most steep/and erosion occurs in a few areas, and most of the remaining area produces only a small amount of erosion. To effectively minimize erosion in steeplands, it is more important to specify where land is to be treated than to be concerned with how much land is to be treated. A small amount of activity conducted in the wrong place can result in a great deal more erosion than a large amount of activity conducted in locations which are erosion resistant (Ziemer 1984, 14-15). The same conclusions can also be made about erosion from agricultural lands. Thus, part of our implementation task will be to identify these critical areas. However, to obtain broad political support for a watershed program to reduce soil erosion may require a program that offers services to areas outside the target areas. This may reduce cost-effectiveness but be necessary from a political standpoint.

116

K. William Easter

Second, changes in socioeconomic conditions may have a major impact on the implementation of a watershed project. For example, a change in prices for livestock or wood may make it more difficult to introduce practices to stabilize resource use in upper watersheds. Unanticipated population increases may require plan adjustments to include more food crop production in the watershed. These and other potential changes will require adjustments in the original plan. The program must have built-in flexibility to adjust fairly quickly to unexpected macrochanges and have monitoring and information systems to suggest when and how the changes should be made. Third, many watershed management programs may require major behavioral changes and therefore will be difficult to implement. Such changes could be minimized by developing resource management actions that use locally available materials, are easily understood by users, and do not deviate much from local practices. For example, in the Philippines, strip planting of Leucaena leucocephala fits these criteria much better than bench terraces as a means to control soil erosion and is preferred by upland farmers (Serrano 1984, 9). The use of pilot projects, the development of projects in stages to allow for learning, and the effective use of extension education programs can help overcome some of these difficulties. However, if project success depends on major behavioral changes judged unlikely to occur, the project should probably be rejected or modified at the planning stage. The fourth issue involves the basic problem of whether it is technically, economically, or socially feasible to achieve the program objectives given the type of program being implemented. For watershed management the question is whether the program can provide erosion control while meeting other objectives. Are incentives such that they will elicit the desired actions? This will depend partly on whether the project formulators understand the factors and causal linkages related to attaining the stated objectives and associated outputs. Even if the planners understand these relationships, is it possible to provide the necessary implementation tools and make adequate institutional and organizational arrangements to get the job done? Do farmers have incentives to apply the socially desirable watershed management practices? Finally, does the program or project meet the basic economic tests discussed in Chapter 4? The validity of the causal theory must include considerations of incentives and economic efficiency from the perspective of both society and the user.

CONCLUSION There are four key points to remember concerning project implementation. First, implementation must be planned with as much care as the tech-

Program Implementation

117

nical and physical aspects of the project. One cannot assume that the project will be operated, maintained, and used effectively. There are numerous examples of poorly performing rural development and irrigation projects caused by the lack of implementation planning. Second, user participation and flexibility must be built into project planning and implementation. This will probably require extra resources and a new way of approaching problems. Yet these extra expenditures will pay off in the long run through improved program performance and greater economic returns. Third, political concerns will always be an important aspect of project implementation. One can have all the technical and economic problems worked out, but the program can fail because of conflicts among agencies and the lack of political support from key individuals. Thus, project implementation is a complex process that has to be concerned with a wide range of political and socioeconomic issues. Finally, one should not take institutions and organizations as given. Institutional and organizational arrangements are an important management element and should be used to achieve the desired project outputs. This means that the effects of different institutional and organizational arrangements must be estimated and plans made to obtain the desired changes. Alternatively the implementation tools and resource management actions will have to be selected so that they fit the expected institutional and organizational setting. In either case, institutional and organizational arrangements will be important considerations in implementation planning.

REFERENCES Anderson, T.L. 1982. The new resource economics: Old ideas and new applications. American J Agricultural Economics (December):928-934. Baumol, W.J., and W.E. Oates. 1979. Economics, Environmental Policy, and the Quality of Life. Englewood Cliffs, NJ: Prentice-Hall. Blaikie, P. 1985. The Political Economy of Soil Erosion in Developing Countries. New York, NY: Longman Group. Bottrall, A. 1981. Comparative study of the management and organization of irrigation projects. Staff Working Paper 458. Washington, D.C.: The World Bank. Chambers, R. 1980. Basic concepts in the organization of irrigation. In Irrigation and Agricultural Development in Asia, ed. EW. Coward. Ithaca, NY: Cornell University Press. Coward, E.W. 1980. Irrigation development: Institutional and organizational issues. In Irrigation and Agricultural Development in Asia, ed. EW. Coward. Ithaca, NY: Cornell University Press.

118

K. William Easter

Easter, KW. 1985. Recurring cost of irrigation in Asia: Operation and maintenance. Report for U.S. Agency for International Development. Galvez, J.A. 1984. Management and cost of watershed reforestation: The Pantabangan and Magat case. Paper for Seminar on Economic Policies for Forest Resources Management at Club Solviente Tagana. Hoare, P.W.C. 1984. Improving the effectiveness of agricultural development: A case study from North Thailand. Manchester Papers on Development, no. 10:13-43. Jayaraman, T.K. 1982. Evaluation of implementations phase of rural development projects: A case study from Gujarat, India. Agricultural Administration 10:55-100. Kelman, S. 1984. Using implementation research to solve implementation problems: The case of energy emergency assistance. J. Policy Analysis and Management 4(1):75-91. Lipton, M. 1968. The theory of the optimizing peasant. J. Development Studies 4(April):327-351. Olson, M. 1965. The Logic of Collective Action. Cambridge, MA: Harvard University Press. Popkin, S.L. 1981. A public choice and rural development: Free riders, lemons, and institutional design. In Public Choice and Rural Development, eds. C.S. Russell and N.R. Nicholson, 43-75. Research Paper R-21. Washington, D.C.: Resources for the Future. Sabatier, P.A., and D.A. Mazmanian. 1983. Policy implementation. In Encyclopedia of Policy Studies, ed. S. Nagel, 143-163. New York: Marcel Dekker. Serrano, R.C. 1984. Project monitoring and evaluation with emphasis to watershed management and reforestation. Paper presented at Nepalese Training on Watershed Management and Reforestation Session, AERD, Los Banos, Philippines. Small, L.E. 1982. Investment decisions for the development and utilization of irrigation resources in Southeast Asia. Teaching and Research Forum Workshop Report 26. New York: Agricultural Development Council. Sutadipradja, Engkah. 1985. Status of watershed management research and identification of needed research in Indonesia. Paper presented at Workshop on Integrated Watershed Management Research for Developing Countries, 7-11 January, East-West Environment and Policy Institute, Honolulu, Hawaii. Ziemer, R.R. 1984. The link between upper watershed sediment processes and river sedimentation. Paper presented at Workshop on the Management of River and Reservoir Sedimentation in Asian Countries, 14-19 May, East-West Environment and Policy Institute, Honolulu, Hawaii.

CHAPTER 9

The Potential Role of Agroforestry in Watershed Management Napoleon T. Vergara

One of the foremost challenges of watershed management is to develop management plans to achieve diverse and conflicting goals. The preceding chapters have identified a range of factors that will affect the success or failure of any watershed management plan. In more specific terms, however, the task is often to develop a watershed management strategy that will enable upland farmers to produce their food and wood requirements on a sustained basis without impairing the capacity of the catchment to yield water of the required quality, quantity, and continuity. In exploring various watershed management options, there are several land-use systems from which to choose. Some of these land-use options completely ignore the human inhabitants of the watershed while other options explicitly try to meet conflicting goals. The land-use options include: • protection forestry, completely withdrawn from any use and depended upon solely to provide protective forest cover for the watershed; • production forestry (natural or plantations) from which timber, pulpwood, poles, or fuelwood are extracted; • combined protection/production forestry; • recreation forestry (such as parks) for outdoor recreation, wildlife refuges, and gene pools; • combined protection and recreation forestry (as in wilderness areas); • food crop farming; • combined forestry and food crop farming (agroforestry); • pasture on open grass areas; and • pasture under trees (silvipastoralism). 119

120

Napoleon T. Vergara

Depending on watershed size, topography, soil, climate, population density, and management objectives, one or more of these land uses may be employed. Systematic land capability classification and land suitability assessment should be the first task to help determine the land uses appropriate for the entire watershed. However, since many watersheds are already inhabited, the "protective" functions of a watershed have to be balanced with the "productive" needs of the inhabitants. Agroforestry is one of the most promising land-use options when farmers are present in upland watersheds. AGROFORESTRY: A POTENTIAL FOR UPLAND WATERSHEDS Agroforestry, by definition, is a land-use system that tries to enhance productivity and ensure sustainability by combining annual food crops and woody perennial plants under either a rotational or intercropping arrangement. At times, a livestock component may be added. Under this system, the food crops' principal role is to help meet the dietary needs of the farmers. The trees (either fruit trees or forest trees) help stabilize slopes, reduce erosion, maintain the productive capacity of hilly areas, and meet needs for fuelwood, poles, small timber, tree fruits and nuts, and fertilizer and fodder. Agroforestry is, therefore, a subset of social forestry and usually implies intercropping of trees and annual crops. Agroforestry, although an age-old land-use practice, is a new field of scientific research (Huxley 1980; ICRAF 1979, 1983). Only a limited number of scientific studies in the tropics have been completed. Thus, discussions of beneficial impacts of agroforestry upon productivity, sustainability, and ecological protection have to be based upon these limited studies and on data extrapolated from agricultural cropping, mixed horticultural cropping, and forest ecosystems, and may require validation prior to use in actual agroforestry projects. Nonetheless, the empirical evidence indicates good potential roles for agroforestry in many upper watersheds.

Agrojorestry and Slope Stabilization/Erosion Control Studies of forest ecosystems have highlighted the protective role of trees in slope stabilization. Ziemer (1981) and O'Loughlin (1974), for example, have shown that the woody and extensive root systems of trees bind and anchor the weaker upper horizon of soil to fractures and crevices of bedrock and provide interlocking soil binders to reduce slips and slides. Thus, as long as there are sufficient numbers of strategically located trees in watersheds to perform this stabilization role, it is possible to reduce shallow mass wasting and sedimentation. In an unexploited natural forest, the "triple protective armor" provided by the crown layer, the undergrowth, and the litter mass, as well as the reduction of velocity and quantity of surface runoff by vegetative impediments, combines to stabilize slopes and minimize surface erosion. In an

The Potential Role of Agroforestry

121

/Slope ·.:.~~ ..~:~· \'''

~

'•'-'•

. ~-~:~:~:~~. Figure 9.1 Natural terraces formed over time by contour hedges or strips in an agroforestry system (Source: Vergara 1982).

agroforestry situation, however, the threat of accelerated surface erosion is serious because of soil cultivation (Wiersum 1984). Spaces between trees are cultivated for annual crops, and the disturbance of the soil surface can substantially increase soil loss. Placement of trees in relation to the topography becomes a critical management consideration. The primary roles of the trees are to minimize natural erosion and to arrest the accelerated erosion brought about by intensive food cultivation intercropping activity. The random tree arrangement found in natural forests may no longer be adequate and effective. Instead, the trees in agroforestry need to beartificially arranged in contour rows and spaced closely along the rows so that they may serve as effective erosion barriers and nutrient filters. Depending on the degree and length of slope, erodibility of the soil, and intensity of rainfall, either contour rows or contour strips (consisting of two or more rows) may be used as erosion bars. The spaces between contour strips, ranging from 2-5m, may be allocated to annual crops under a system referred to as "alley cropping" (Nair 1984). Over time, natural terraces form with the contour strips acting as terrace runners (Figure 9.1). Man-made terraces are a well-known approach to erosion control. Their effectiveness is demonstrated by the continued use of centuries-old irrigated

122

Napoleon T Vergara

rice terraces in southern China, in Bali (Indonesia), and in Ifugao (northern Philippines). Although the population and cultivated land in the uplands are both increasing, no corresponding significant expansion in terraced areas has occurred. Upland farmers apparently are unwilling to invest heavy inputs of labor and capital into such permanent land improvements, partly because of unsettled tenure questions and the expense of terracing difficult terrain. ''Alley cropping'' type of agroforestry, with trees deliberately arranged in contour strips, is a feasible alternative approach to joint food production and watershed protection in the Asian tropics. If markets are readily available for horticultural products, the agricultural crop component of an agroforestry system could be changed from subsistence-type annual crops to market-oriented horticultural perennials. The combination of agricultural perennials with forest crops, as exemplified by coffee/ Leucaena and coffee/Casuarina systems in Papua New Guinea; Lanziumlcoconut and cocoa/ Leucaena in the Philippines; coffee!Gliricidia and cocoa/Gliricidia/coconut in the Solomons; and coffee/Gmelina in Thailand, provides much more soil protection than an annual/perennial crop combination. The frequent soil disturbance brought about by site preparation, planting, and harvesting of annual crops is avoided. A possible objection farmers may have to perennial/perennial crop combinations is that perennials are cash crops and not food crops that can be directly consumed. In addition, perennials do not usually begin to bear for several years; once harvesting begins, however, production is continuous and there is no need for replanting each year. To fill the gap between planting of perennials and the first yield, farmers may need to interplant annual food crops to tide them over. For example, tree farmers in the southern Philippines started out as agroforestry farmers intercropping annuals such as corn, taro, and rice with N-fixing Albizia. The food crops were for domestic consumption, while the tree crops were marketed at a nearby pulp and paper plant. As soon as it became evident to the farmers that they would earn more revenue by tree cropping exclusively rather than food crop/tree farming, they shifted from agroforestry to specialized tree farming (Domingo 1981). Their economic situation has improved as they moved from a subsistence- to a market-oriented economy, and their tree farming has contributed to greater site stability.

Agrojorestry and Sustained Site Productivity On hilly terrains such as those found in most watersheds, nutrients and, in turn, productivity are lost through three principal ways: erosion, leaching and runoff, and biomass harvest and removal (Figure 9.2). A fourth source of loss, percolation, may be added, although it does not really represent a loss from the system. Percolation merely transfers nutrients to soil depths below the reach of annual plant roots.

The Potential Role of Agrojorestry TRANSFERS (within the system)

Litterfall

Uptake

Percolation

I

l

123

LOSSES (output from the system) Agroforestry Vegetation Surface Soil

Biomass Harvest (food crops and tree products) Erosion Leaching/Runoff

Subsoil

Figure 9.2 Nutrient losses from, and transfers within, an agroforestry system.

In natural systems where no biomass is harvested, as in an "untouched" watershed protection forest, the natural nutrient losses from normal erosion and leaching are more or less equal to the natural inputs from the atmosphere and from nutrient cycling so that a balanced nutrient budget is maintained (Christanty et al. 1982). This is the reason for the well-known and much-admired stability of undisturbed natural systems. In an agroforestry system, however, nutrient losses are significantly increased by accelerated erosion and by the harvest and removal of plant biomass (grains, fruits, leaves, flowers for food; tree stems for fuelwood; timber or other uses) (Figure 9.2). Unless something is done to replace these lost nutrients, productivity of the site will decline. Four remedial steps associated with agroforestry can be taken to avoid site degradation: • Reduce nutrient loss. This can be achieved in agroforestry by removing only the usable (edible) parts of the food and tree crops and recycling the rest of the biomass as plant residue. For example, only the woody stems of trees would be extracted while the roots, stumps, tops, and branches would be left behind. • Minimize erosion and runoff losses. This can be done with the use of contour tree strips as discussed previously. • Increase the rate of natural nutrient input. The most promising method to achieve this is through the introduction of trees and plants that fix nitrogen and produce a steady supply of N-rich litter and green manure (Figure 9.3). • Apply chemical fertilizer. The economics of fertilizer use should be carefully evaluated since it may be expensive for upland farmers to purchase.

124

Napoleon T. Vergara Nutrients from Rainfall

Organic Ferti lizer (agricultural and forestry residue) Chemical Fertilizer

Nitrogen from Atmosphere through N-Fixing Plants

Agroforestry Vegetation

Surface Soil

Subsoil

Figure 9.3 Natural and man-induced nutrient inputs into an agroforestry system. Nitrogen is one the major elements required for plant growth and output. When properly used in combination with other nutrients (such as phosphorus and potassium), one kg of elemental nitrogen (N) can increase food production by as much as 10 kg (Ahmed 1982). Thus, one possible route to enhanced productivity and sustainability is in supplying sufficient nitrogen and other nutrients cheaply to cash-poor upland farmers; this may require some form of subsidy. The supply of chemical nitrogen can be supplemented through the introduction of N-fixing trees, especially N-fixing legume trees, which number about 2,500 species (Brewbaker et a!. 1982). The amount of nitrogen fixed by five of the more common and oftenused legume trees in agroforestry is substantial (Table 9.1). When judiciously Table 9.1 Nitrogen yields of selected tree legumes N Yield (kg/ha/yr)

Source

Leucaena

500-600

National Academy of Sciences 1977

Acacia mearnsii

240-285

National Academy of Sciences 1979

Species

Sesbania

500+

Wiersum 1981

Calliandra

105+

Wiersum 1981

Gliricidia

84

Nair 1984

The Potential Role of Agroforestry

125

applied to crops, theN inputs can increase agricultural outputs. Nair (1984) indicates, for example, that corn yields could be maintained at reasonable levels with periodic Leucaena green manure application. Pacardo (1981) showed that, with Leucaena contour strips 4 m apart and corn crops planted between strips, grain yields could be almost doubled with application of Leucaena herbage from the strips at 6-to-8 week intervals. Besides their N-fixing ability, most legume trees possess three other characteristics that make them potentially important components of agroforestry systems in watersheds. • When cut down for timber or foliage harvest, legume trees do not die but quickly regenerate by coppice. This is important to farmers because they do not need to spend labor and capital to replant and replace the harvested trees. It is also important for watershed protection because the live trees maintain their root systems to prevent mass soil movement, minimize surface soil erosion, and help sustain productivity. • The leaves, flowers, and fruits of legume trees are generally palatable to livestock and edible by humans. • The N-rich leaves are good organic fertilizers-whether allowed to fall as litter, or regularly cut and used as mulch and green manure-thus contributing toward raising and maintaining crop yields.

Agroforestry and Stabilization of Watershed Communities Attempts at relocating farmers from watershed areas are usually unsuccessful. Some forest managers have tried confining farmers to small selected spots to stabilize land use and minimize damage to the catchment. This effort has been largely unsuccessful for several reasons including: • Ethnic minorities are commonly found in the upper watershed. If they are traditionally nomadic, they will usually refuse to be confined to reservation-like areas. • Farmers, displaced from the lowlands to the upper watershed, may intend to stay permanently on one site but have to move after inappropriate tillage techniques deplete their hilly farms or when weeds become uncontrollable. Sustainable agroforestry systems, by making possible sustained production, could eliminate the reasons for frequent shifting. When farm production is stabilized, farmers are more likely to settle in more permanent communities and further encroachment into the forests will be minimized; this is assuming land tenure is not a problem. Traditionally nomadic ethnic groups are not easily converted to sedentary farming. Where fruit trees are used as the perennial component of an agroforestry system, however, farmers become strongly attached to the

126

Napoleon T Vergara

crops and stay near the farmsite to watch over and harvest the fruits. Coconut trees, mangoes, jackfruit, and citrus trees in the Philippines have served to induce mountain farmers to locate their homes permanently at the farmsites (Payuan, pers. com.). A similar phenomenon was observed with tea, coffee, durian, jackfruit, and other fruit trees in Thailand (Hoare 1983).

Economic Feasibility of Agrojorestry Agroforestry is technically feasible as a land-use system for hilly terrain, including upland watershed areas. What is not clear is whether a financial analysis (conducted from the standpoint of the farmer) or an economic analysis (from the societal viewpoint) will also find agroforestry systems ''profitable:' Some studies have been conducted to determine the economic benefits of agroforestry to the farmer. Thus far, preliminary conclusions appear positive. For example, Mendoza in the Philippines (1977), Rachie in Colombia (1981), and Nair in Africa (1984) have shown that the Leucaena-cornvegetables combination yielded higher aggregate physical outputs than annual monocropping. These studies examined yields in physical terms and stopped short of converting outputs and inputs into financial terms, thereby precluding the calculation of net financial returns to the farmer. Other researchers, such as PCARRD (1983) and Corpuz (1984), quantified inputs and outputs in monetary terms and were able to demonstrate that agroforestry is financially attractive. They did not take into account, however, the other streams of off-site costs and benefits that occur over a longer period. As a result, the net discounted values they calculated are probably an understatement of the long-term productive potential of agroforestry. Economic analysis of agroforestry is more problematic from the societal standpoint. Analytical methodologies are available, but there is a dearth of quantitative information, particularly about the environmental externalities (both positive and negative) of agroforestry. This information is essential to an accurate calculation of net social benefits. Therefore, the data gap needs to be closed quickly through research and close monitoring of ongoing agroforestry development activities. Despite the absence of detailed economic analysis, decisions to employ agroforestry in place of annual cropping in densely populated and heavily cultivated catchment areas have been, or are being, made in some countries. Such decisions are in part based on the readily visible capacity of "contour-strip" type agroforestry to minimize erosion on steep slopes. AGROFORESTRY lN UPLAND WATERSHEDS: CONSTRAINTS AND OPPORTUNITIES The foregoing discussion indicated that agroforestry may be acceptable and desirable as a land-use system for integrated watershed management

The Potential Role of Agrojorestry

127

in densely populated developing countries. The next task is to assess the constraints that limit its acceptance and to identify the factors that could enhance its adoption.

Traditional Forest Management Practices Controlled and managed use, rather than a complete ban of human activity in temperate zone countries is a good policy for maximizing net societal benefits from watersheds (Hamilton 1983). In most tropical countries, particularly in Asia, a backlash against accelerated deforestation due to loggers, ranchers, and shifting cultivators, and the simplistic belief that forest removal causes floods and droughts, have hardened the attitudes of many watershed managers against most forms of human activity, including logging and crop cultivation. Police action has been continuing against those who use the watersheds, and frequent confrontations have occurred between forest managers and settlers. Policies prohibiting nonforestry use, including agroforestry in watersheds, have been formulated. Under these conditions, introducing agroforestry as an element of integrated management in water catchment areas is extremely difficult. Forest administrators are beginning to realize that police action cannot stem the inflow of migrants in search of farm land. Furthermore, watershed managers are starting to recognize the need to consider and understand both people and forests, rather than forests alone, in planning resource conservation programs. Enlightened watershed management practices, such as those in Northern Thailand which are discussed in Chapter 12, now allow limited terracing and the planting of annual food crops with woody perennials such as coffee and tea (Jones 1984; Hoare 1983).

Institutional Constraints Certain obstacles to the adoption of new management strategies are more institutional in nature rather than biophysical or socioeconomic. A good example of this is land tenure. Assuming that watershed managers are convinced that contour-strip or alley-cropping type of agroforestry is indeed an acceptable and desirable land use, the next problem to be faced is how shifting cultivators can be encouraged to adopt these practices. One important reason for farmers' reluctance to practice agroforestry systems is their lack of secure tenure over their farmsites. The long-term nature of the woody perennials makes it difficult, perhaps even impossible, to induce farmers to incorporate them with their annual crops without a corresponding assurance that they will be around to harvest the fruits of their labor. Governments often reject the notion of putting public watersheds under permanent control of private individuals. Given this position, it is doubtful that upland farmers will ever acquire the land rights that are a necessary precondition to widespread adoption of agroforestry. Certain

128

Napoleon T. Vergara

approaches being tried in Thailand and in the Philippines may yet provide a way out of this dilemma. The Thai Royal Forest Department and UNDP have tried, in a limited area in Northeast Thailand, a usufruct lease agreement with upland farmers. Under this arrangement, farmers will never own the land, but they can use it for life as long as they employ agroforestry systems. The Philippine Bureau of Forest Development grants long-term land leases of 25 years, renewable for another 25, if farmers agree to practice agroforestry or tree farming on government-owned hilly land. Preliminary information points to a favorable response from farmers. Long-term, secure land leases could be an answer to the problem of insecure land tenure.

Access to Markets If, as predicted by its proponents, agroforestry succeeds in enhancing the productivity and sustainability of the farm system, there will be a need for marketing of surplus output to increase farm cash incomes. There are three main problems: (1) most agroforestry sites are in remote hilly areas with minimal transport facilities; farmers face difficulties in marketing surplus products; (2) the marketable surplus often does not reach sufficient volume to justify the development of new access roads to the uplands; and (3) agroforestry crops may be seasonal and need to be harvested and marketed simultaneously, thereby aggravating marketing problems, flooding the market, and forcing down prices. ln addition, many of the products, especially fruits, have an extremely short storage life. Because of severe limitations in resources, most developing countries concentrate infrastructure development programs on the heavily populated lowlands. Given this situation, the most feasible solution to the problem of marketing agroforestry products may lie in selecting appropriate crops rather than in government action to improve access. Upland farmers facing access problems and marketing difficulties should limit their output of highly perishable fruits and vegetables to the amounts demanded for local consumption. Other farm production should be commodities that have a long storage life, require little or no added processing, and can withstand rough handling. Appropriate tree crops include nuts, spices, resins, and fibers; annual crops such as grains may also be suitable. Another promising type of product is one that does not need to be harvested at a particular time; many root crops belong to this category. For example, cassava is an important crop in Northeast Thailand and upland Indonesia. Strong export demand for dried cassava has also been an important factor in making the crop attractive to farmers.

Attitudes of Upland Farmers to Changes in Agrotechnology As mentioned earlier, many upland farmers originally came from the lowlands and continue to use lowland agricultural techniques of intensive monocropping of annual crops. Such techniques are often inappropriate

The Potential Role of Agroforestry

129

to the fragile upland ecosystems. Converting farmers to polycultural farming methods that incorporate the use of woody perennials with annual crops cannot be accomplished overnight. Farmers may resist the new technology due to lack of familiarity and experience. Demonstrating to farmers that agroforestry systems are feasible and profitable is a difficult task, because verbal explanations are insufficient and physical demonstrations require a long time to produce results. Therefore, a combination of both approaches is necessary. Although a slow process, trial demonstrations on selected farms can effectively show the value of the system to farmers. Technical support assistance is also necessary to ensure that the technology is properly employed and to help achieve the expected benefits. Financial and material assistance will be required when farmers begin to introduce agroforestry systems. Other inputs will be needed, including seeds of N-fixing trees and seeds of shade-tolerant food crops. If the extension organization can help provide these many inputs at the appropriate time, it will help ensure that the new agroforestry system and techniques will be successfully introduced. REFERENCES Ahmed, S. 1982. Projected nitrogen needs in the year 2000 and alternative supply sources. Paper presented at the Workshop on Biological Nitrogen Fixation, 13-14 January 1983, East-West Resource Systems Institute, Honolulu, Hawaii. Brewbaker, J.L., R. Van Den Beldt, and K. MacDicken. 1982. Nitrogenfixing tree resources: Potentials and limitations. In BNF Technology for Tropical Agriculture. Colombia: CIAT. Christanty, L., et a!. 1982. Soil fertility and nutrient cycling in traditional agricultural systems in West Java. East-West Environment and Policy Institute Working Paper, Honolulu, Hawaii. Corpuz, E.B. 1984. A comparative economic study of traditional kaingin, modified cropping patterns and tree farming in Mt. Makiling. Master's thesis, University of the Philippines at Los Banos. Domingo, l.L. 1981. Agroforestry and Albizia fa/cataria in PI COP. Paper presented at the Workshop on Agroforestry, East-West Environment and Policy Institute, Honolulu, Hawaii. Hamilton, L.S. (with P.N. King). 1983. Tropical Forested Watersheds: Hydrologic and Soils Response to Major Uses or Conversions. Boulder, CO: Westview Press. Hoare, P. 1983. Techniques for encouraging community participation in agroforestry projects. Paper for Workshop on Agroforestry, East-West Center/FAO, Thailand.

130

Napoleon T Vergara

Huxley, P. 1980. Research for the development of agroforestry land use systems. Paper presented at a Research Conference, Sudan. ICRAF. 1979. First Report, 1978179. Nairobi, Kenya. ICRAF. 1983. An account of the activities of ICRAF. Nairobi, Kenya. Jones, P. 1984. Soil conservation practices for shifting cultivators in Northern Thailand: Lessons in extension. Report to the Australian Development Assistance Bureau and The World Bank. Mendoza, R.C. 1977. Notes on corn/ipil-ipil intercropping trials: Research results. University of the Philippines at Los Banos. Nair, P.K. 1984. Soil productivity aspects of agroforestry. ICRAF, Nairobi, Kenya. National Academy of Sciences (NAS). 1977. Leucaena: Promising Forage and Tree Crop for the Tropics. Washington, D.C. National Academy of Sciences (NAS). 1979. Tropical Legumes: Resources for the Future. Washington, D.C. O'Loughlin, C.L. 1974. The effect of timber removal on the stability of forest soils. Hydrology 13(2):121-134. Pacardo, E. 1981. Water relations and nutrition of some leguminous trees. Paper presented at the Workshop on Agroforestry, East-West Environment and Policy Institute, Honolulu, Hawaii. PCARRD. 1983. Philippines recommends for agroforestry. Philippine Council for Agriculture and Resources Research and Development, College, Laguna. Rachie, K.O. 1981. Intercropping tree legumes with annual crops. Cali, Colombia: CIAT. Vergara, N.T., ed. 1982. New Directions in Agroforestry: The Potential of Tropical Legume Trees. Honolulu, HI: East-West Environment and Policy Institute. Wiersum, K.F., ed. 1981. Observations on Agroforestry on Java, Indonesia. Agricultural University, The Netherlands. Wiersum, K.F. 1984. Surface erosion under various agroforestry systems. Paper presented at Symposium on the Effects of Forest Land Use on Erosion and Slope Stability, 7-11 May, East-West Environment and Policy Institute, Honolulu, Hawaii. Ziemer, R.R. 1981. Roots and stability of forested slopes. IAHS Publication 132. Christchurch, New Zealand.

PART II

APPLICATIONS

CHAPTER 10

Watersheds in Hawaii: An Historical Example of Integrated Management Joseph R. Morgan

Hawaii, as an island state of the United States, has some unique problems in managing its natural and human resources. With a population of slightly more than 1 million people unevenly distributed among seven inhabited islands, statewide policies need to consider both the differences among islands and their similarities. Almost 80 percent of the people are on Oahu, which has less than 10 percent of the land area of the state. In general, however, the islands have many common features: high rainfall on windward slopes, steep slopes in many areas, and few perennial streams of any great length. Watersheds of the relatively few important streams are small and well delineated by the complex topography of the eroded volcanic peaks that are such an important feature of the physical geography. Vegetation ranges from tropical rainforests, some still relatively undisturbed in rainy areas to virtual deserts in the much drier leeward regions. Land-use regulations in the state of Hawaii provide for protection of upper watershed areas by prohibitions against both agricultural or construction activities on steep slopes and the clearing of vegetation in regions designated for "conservation?' Rules are designed to prevent pollution of streams and coastal waters. Problems of control of flash flooding, always a danger in the small watersheds, have been recognized. Many of the streams have been canalized in their lower reaches as a flood control measure. In a few cases decisions had to be made regarding allowable use of surface water, allocating supplies to taro farmers and those desiring water for urban development. But the management of water and the associated problems of preventing erosion and pollution do not specifically recognize the essential unity of a watershed as a functional region. Watershed districts are not designated, and political regions do not coincide with natural physical regions, typified by watersheds. Such was not the case in 133

134

Joseph R. Morgan

ancient Polynesian Hawaii, where watersheds were clearly recognized as functional regions and administered accordingly. In a sense, functional regions are natural in that physical processes operate within them, and topographic boundaries form divides between adjacent regions. As discussed in Chapters 1 and 3, a river system that forms a drainage basin or watershed is a good example of a functional region. Each watershed functions as a region, within which are interrelationships among the water, soils, vegetation, topography, and, if occupied, activities of people. Management-for wise utilization of resources, conservation of important plant and animal species, control over pollution, and apportionment of the water-is facilitated if the political authority operates within a functional region and, as mentioned in earlier chapters, problems arise when they do not. Ideally, watershed boundaries and political boundaries should coincide. In many so-called traditional societies, this was the case; perhaps one of the best studied of these land-use systems operated in Hawaii before and in the early years of Western contact. ANCIENT HAWAIIAN SOCIETY AND LAND USE The Polynesians who settled in the Hawaiian Islands, after several hundred years of adaptation to their environment, established an integrated system of government, economy, and religion based on feudal principles. Important chiefs (alii nui) 1 were people who could trace their ancestry back to the original mythical gods who created the islands. Genealogies were of utmost importance, since the clearer and more direct the line of succession, the greater the degree of spiritual power (mana) the alii possessed. All land was owned by the gods, but alii were allocated portions of it for their sustenance. Wars among the important chiefs were frequent, as alii nui vied for control over the larger, more desirable land sections. Commoners (makaainana) worked the land allocated to them by alii. Although they did not own the land they occupied, the makaainana could raise crops on it, provided they paid a tax to the alii and participated in required communal labor. An elaborate system of laws regulated political affairs, economy, and religion. These rules, the kapu, were designed to ensure the proper degree of respect for various grades of alii by the makaainana, proper and conservative use of the resources, and suitable veneration of the gods. Thus, the kapu system was political, economic, and religious at the same time. The society also included two other classes: kahuna and konohiki. Kahuna were of two types: priests who supervised the ritual observances to the gods, and experts in fields such as medicine and canoe building. Konohiki were land agents or supervisors for the alii. Usually they were of the 1. The Hawaiian language uses the same form for singular and plural nouns. This usage will be followed in this chapter.

Watershed Management in Hawaii: An Historical Example

135

alii class themselves, but with a lower degree of mana, and frequently were related to the alii for whom they worked. The Ahupuaa Land units were configured to make maximum use of resources. The largest unit was the moku or district under the control of an alii nui. The moku, however, was usually too large a unit to be managed efficiently; it was subdivided into ahupuaa (the basis for this study), and many smaller units occupied by makaainana. Whenever possible an ahupuaa extended from the highest mountain peaks in the vicinity to the coast and beyond, usually to the outer edge of the offshore coral reef. The ahupuaa was thus designed to be a selfcontained economic and environmental unit, providing for the alii and the makaainana. The high forests provided wood, bird feathers, and various useful plants. The moderate slopes produced upland crops, while the lowlands were planted in taro. From the shore and reef, the products of the ocean could be harvested. Included within the ahupuaa were freshwater and marine fish ponds, irrigation systems, and the homes and ceremonial buildings of the community. The alii, assisted by his konohiki, maintained both political and economic control of the land and its resources. Political boundaries were coincident with agricultural and natural resource boundaries. According to Handy and Handy:

The Hawaiian political geography was patterned in terms of agricultural districts topographically determined by the stream systems upon which taro plantations were dependent for irrigation. The boundaries of the ahupuaa, or agricultural subdistricts, were determined by the watershed (emphasis added), where there were streams. Where there were not stream systems, arbitrary bounds were drawn from the mountains to the sea to give an equivalent subdivision into districts for levying taxes on produce (Handy and Handy 1972, 77). Despite copious annual rainfall in the Hawaiian Islands, many of the geologically younger islands have few perennial streams. Porous lava flows permit the water to infiltrate into a groundwater system, which was of little use to the early Polynesian settlers. On the other hand, the geologically older islands have more surface water, as there has been more time for soils to form on the lava parent material. The wettest and the oldest of the inhabited Hawaiian Islands is Kauai; consequently it is better endowed with streams and well-defined watersheds. The land-use system characterized by the ahupuaa was well developed on the island, with the boundaries of the wedge-shaped land sections following the natural boundaries of watersheds. On other islands ahupuaa land patterns were different. Earle describes the contrasting types as follows:

136

Joseph R. Morgan Aboriginally, there were two general patterns on the Hawaiian Islands, reflecting contrasting geographic conditions. The first, or valley ahupuaa, was located in geologically old areas where major streams had developed through erosion. An ahupuaa ideally encompassed a complete catchment area-one major valley, one ahupuaa The second, or upland ahupuaa, was found in geologically young areas where major streams were absent. Its boundaries, running perpendicular to the contours, defined an arbitrary strip which ran from the mountains to the sea (Earle 1978, 155).

Both ahupuaa types were configured to provide the alii and his makaainana with a unit of land that would completely sustain them with a variety of natural and agricultural products. Because the valley ahupuaa is the most relevant subject to this book, it will be further described in this chapter. Water was a key feature of ahupuaa, with boundaries following ridgelines to separate the watershed or drainage basin of one ahupuaa from its neighbors.

The central idea of land division in the Hawaiian Islands was radial, running from the seashore up into the mountains, thus including fishing rights, cultivable lands, upland timber and planting zones, and areas of bird-catching privileges in the higher mountains (Handy and Handy 1972, 48-49). Management Considerations Neither the alii nui rtor the alii ai ahupuaa (literally "the chief who eats the ahupuaa:' a reference to the fact that the chief was entitled to the produce of the land) was an absolute ruler. Although both had great powers they were regulated by the kapu, many of which were designed for regulating and preserving the environment. For instance, kapu forbade the taking of various fish species at certain times of the year, regulations that prevented overfishing of spawning and young fish. In the ahupuaa many kapu were associated with water rights and regulation of irrigation practices. Since irrigation ditches were fairly numerous and elaborate, the general rule was that rights of those taking water from below the dam had to be respected, and those upstream had to allow those downstream their fair share. No ditch was by itself permitted to take more than half the flow of a stream. A konohiki controlling the water was appointed by the alii; the luna wai (literally "water boss") regulated irrigation in accordance with the kapu. In times of drought the luna wai had the authority to adjust the sharing of water to meet exigencies (Handy and Handy 1972, 59). Hawaiian native water rights laws generally provided more water to the makaainana who had bigger plots of land, provided they were reasonably

Watershed Management in Hawaii: An Historical Example

137

productive in their land use. Farmers who were more than usually diligent in their raising of crops, a portion of which the alii took as a tax, were allotted a larger acreage and consequently more water. This general rule tended to increase productivity. At the same time, however, the aim of the konohiki was to ensure equal water rights for all, thus avoiding controversies. But Nakuina (1893, 79-84), describing the communal efforts organized in constructing irrigation ditches (auwai), stated water was sometimes allocated in accordance with the amount of labor contributed by taro planters to the construction of the auwai and not necessarily to the acreage they had under cultivation. Clearly, it was to the makaainana 's advantage to be industrious both in raising taro and in working on irrigation ditches. Court decisions, after Hawaii became a territory of the United States, later upheld these general water rights principles; and a 1916 decision of the Supreme Court of Hawaii held that ''a title to a water right is a title to real estate'' (Thayer 1916, 770).

The fact that before the Mahele (the land division of 1848 which substantially altered the ahupuaa system) disputes concerning water were extremely rare undoubtedly stemmed from the Hawaiians' acceptance of fresh water as sacred. No believing Hawaiian would tamper with or pilfer that which was identified with Kane (the Hawaiian god of creation), the source of life. Great care was taken not to pollute streams. There was a place for bathing (auau) low down in the stream, a place farther along the stream for washing utensils or soaking calabashes; still farther up were the dams for auwai (irrigation ditches); and above the dams was the place where drinking water was taken (Handy and Handy, 1972, 61). Taro was the most important agricultural product and required the most elaborate and careful cultivation of the land. Leveled plots, loi, were constructed so the plant could be grown under flooded conditions. Irrigation water was brought to the /oi by ditches, some of which were marvels of engineering skill, considering the level of technology available: "In Hawaii the farmer was also a civil engineer" (Handy and Handy 1972, 16). The most impressive of the ancient Hawaiian irrigation structures is the Menehune Ditch on the island of Kauai. Vancouver provides a detailed description of the cultivation practices in the lower Waimea Valley and describes the ditch as "an exceedingly well constructed wall of stones and clay about twenty-four feet high" which served "as an aqueduct to convey the water brought thither by great labor from a considerable distance" (1798, 376-377). Some examples of the importance of water to the Hawaiians can be found in their language. Wai is the word for fresh water, and when reduplicated as waiwai, literally "water-water;' its importance is stressed. Waiwai means

138

Joseph R. Morgan

wealth, prosperity, ownership, possession or importance, referring to the fact that the one who had control of much water was a person of great status in the community. Such a person was wealthy, important, and influential. Control of land was not nearly as important. Evidently with a population estimated at about 250,000 land was generally in plentiful supply, but in many areas water was scarce. The word kanawai, literally "belonging to the waters;' is the Hawaiian term for "law?' Water rights did not imply water ownership, for water like the land itself did not belong to any individual. The alii nui owned neither the land nor the water; he administered it for the principal god, Kane, for the benefit of all the people. In general, the alii and their konohiki did the job well, for the estimated 250,000 Hawaiians, occupying a small percentage of the land in the islands, sustained themselves and developed a society where food scarcities were rare and disputes over land and water were almost unknown. Ahupuaa existed on all the settled islands but were most effective in those areas where perennial streams occurred. The Halelea district of Kauai was one such region (Figure 10.1). Kauai, the northernmost of the inhabited Hawaiian Islands, gets the most annual rainfall. Kauai is in the path of prevailing trade winds, which produce large amounts of orographic precipitation, and of some north Pacific cyclonic storms, which pass to the north of the more southerly islands. The summit peaks of the island receive more than 10,160 mm of rain per year, feeding many rivers that flow radially from the center of the island to the coasts. Archaeological evidence indicates there were nine separate ahupuaa in Halelea, each comprising the catchment area of one or more principal streams (Figure 10.1).

The ahupuaa boundaries were laid out to include, within a single territorial unit, all the resources necessary for a generalized subsistence economy (Earle 1978, 30). In all the ahupuaa the boundaries run along the ridges from the mountains to the sea to include a permanent water source, alluvial soils, access to the open sea and reefs or shallow water areas. The hydrology of the stream dominates the economy of the ahupuaa, and the management and conservation of the stream waters and the adjacent soil are prime considerations of the alii ai ahupuaa and his konohiki. The successful management of the environment, characterized by the ahupuaa system, was due to several factors. First, boundaries were wisely delineated and land was apportioned to the principal alii. They each had enough land with the required range of resources to sustain themselves and the makaainana who worked for them. Second, an elaborate set of kapu was designed to protect environmental resources and fairly apportion critical ones to all. Third, economic and political boundaries coincided. The alii ai ahupuaa was both political ruler and economic and environmental manager. Political leaders were not working at cross purposes with those

Watershed Management in Hawaii: An Historical Example

139

Kalihikai

- - - - Ahupuaa Boundary

~Stream

Figure 10.1 The radial draining pattern of streams in Kauai and the location of Halelea District with its nine historical ahupuaa (Adapted from Earle 1978).

who depended on the natural resources of the land for their sustenance. Finally, and perhaps most important, the watershed as a natural functional region was clearly recognized. Boundaries, formally established, described, and recognized by all, were not contrived without reference to the natural features of the land. They delineated basic land units shaped by topography and hydrology.

140

Joseph R. Morgan

Although the overall unity of the ahupuaa was the most important concept, alii recognized, in effect, four separate subzones suitable for management purposes. Accordingly, most ahupuaa had a konohiki for the upland forest, unirrigated agriculture on the moderate slopes, irrigated agriculture in the lowlands, and marine fisheries in the offshore coral reef areas. In the forest, trees were cut for construction of canoes and houses, and bird feathers were collected principally to make the elaborate capes prized by alii as a symbol of wealth and importance. But wholesale harvesting or clear-cutting, which might lead to serious erosion or siltation of streams, did not occur. Dryland agriculture was practiced on sloping land, with sweet potatoes and some dry taro species as the principal crops. In the irrigated lowlands taro was the chief crop, and the konohiki in charge was concerned primarily with irrigation practices and water rights. At times group labor had to be recruited and organized for maintenance and repair of irrigation systems. The konohiki handled this, under the overall authority of the alii. Fresh- and saltwater fishponds were maintained and regulated by a konohiki, who may also have been in general charge of offshore reef fisheries. Konohiki reef fishery rights exist today and have been upheld by modern courts. There are numerous descriptions of Hawaiian agricultural practices in the two decades after Cook's discovery of the islands. Invariably the writers are impressed with the irrigation systems and the orderly development of the small plots of land within the ahupuaa. Menzies, the naturalist and surgeon accompanying Vancouver's expedition, described the environs of the village of Lahaina:

We would not indeed but admire the laudable ingenuity of these people in cultivating their soil with so much economy. The indefatigable labor in making these little fields in so rugged a situation, the care and industry with which they were transplanted, watered and kept in order, surpassed anything of the kind we had ever seen before (Menzies 1920, 105). Portlock visited Kaneohe, Oahu, and provided the following description:

The bay all round has a very beautiful appearance, the low land and valleys being in a high state of cultivation, and crowded with plantations of taro, sweet potatoes, sugarcane, etc., interspersed with a great number of coconut trees, which renders the prospect truly delightful (Portlock 1789, 74). And Campbell, who visited Oahu in 1809, wrote:

The flat land along shore is highly cultivated; taro root, yams, and sweet potatoes are the most common crops; but taro forms the chief object of their husbandry, being the principal article of food amongst every class of inhabitants.

Watershed Management in Hawaii: An Historical Example

141

The mode of culture is extremely laborious, as it is necessary to have the whole field laid under water; it is raised in small patches, which are seldom above a hundred yards square; they are surrounded by embankments, generally about six feet high, the sides of which are planted with sugar-canes, with a walk on top; the fields are intersected by drains or aqueducts, constructed with great labour and ingenuity, for the purpose of supplying the water necessary to cover them (Campbell 1967, 115-116).

Each year an elaborate and lengthy festival, the makahiki, was celebrated, which lasted for about four months, beginning at the start of the rainy season in late October or early November. During this time, taxes were collected in the form of produce of the individual makaainana; the god of the harvest and fertility, Lono, was worshipped with a series of complex rituals; sports and games were held for the enjoyment of all; and the land and sea were put under a kapu, that is, all work involving farming and fishing was forbidden. The length and timing of the makahiki lead one to surmise that the ahupuaa were generally productive, since they produced enough in an eight-month cycle of planting, growing, and harvesting to sustain the people for an entire year with apparently a considerable surplus for the chiefs, konohiki, and a separate priestly class, the kahuna. The fact that the land and sea were under kapu during the four months of the makihiki might logically be construed as an enforced period of fallow to ensure that the land would remain productive indefinitely. The timing of the festival indicates that land preparation was not to take place during periods when heavy rains were likely, thus protecting the land against the threat of soil erosion.

The Beginning of the End What caused the formal end of the admirable system of land use based on the concept of the watershed as a natural functional region? It was not a realization that the boundaries of the ahupuaa somehow became less suitable, but rather the gradual loss of faith in the kapu and the increasingly greedy nature of the alii. The system of kapu began to erode in 1778 when Captain Cook discovered the islands and brought his ships into Kealakekua Bay, Hawaii, for supplies. The Hawaiians initially were quite willing to exchange agricultural surpluses for trinkets, bits of iron, and other trade goods, while still rigidly adhering to their ancient laws and procedures. But subsequently, as more and more Westerners came to the islands and a trading economy began to develop, Western ways and laws began to become more important than the old kapu. Westerners were always considered exempt from the kapu, while Hawaiians had to obey them. The fact that no harm came

142

Joseph R. Morgan

to the disobedient foreigners became evident to the Hawaiians, and their faith in the kapu system began to weaken. The system was formally abolished in 1819, but the control of the land by the alii remained in force. Some ahupuaa continued to be carefully tended, as Arago described for Lahaina:

It would be difficult to find a soil more fertile, or a people who can turn it to better advantage; little pathways sufficiently raised, and kept in excellent condition, serve as communications between the different estates. These are frequently divided by trenches, through which afresh and limpid stream flows tranquilly, giving life to the plantations, the sole riches of the country (Arago 1823, 119). Moreover, with the first introduction into the islands of trade goods by early explorers and settlers, many of the alii became less interested in their land and its management and more interested in short-term profits that the trading of agricultural and other products to ~hip captains could gain them. Mismanagement of some ahupuaa was most pronounced during the sandalwood trade era. Sandalwood was a native species in the islands, prized for its scent, when Western traders discovered that it was highly coveted by the Chinese. Alii set their makaainana to work cutting sandalwood, at the expense of tending their taro crops. The wood was sold to traders who carried it to China on their vessels and sold it at a high profit. The alii reaped the benefits in the form of trade goods, while the makaainana suffered with longer, harder labor coupled with less food. Moreover, wholesale cutting of sandalwood forests denuded the land, resulting in largescale erosion.

Exposed to cold, badly fed, and obliged to bear painful burdens, they died in great numbers, so that it was a blessing to the Islanders when the wood became scarce. Again, supplies of food were sold by the chiefs to the ships, and this necessitated unusual labor from the people. One famous chief for years used his retainers to tow ships into the narrow harbor of Honolulu, sending them out on the reef, where up to their middle in water, they shouldered the tow line (Nordhoff 1974, 84).

LESSONS FROM THE PAST The feudal system of land tenure came to a formal end with the land division (mahele) of 1848. However, long after the ahupuaa system was formally abolished by the mahele, Hawaiians maintained elements of the old land-use methods and practices. Nordhoff, writing in 1874, provides an illuminating description:

Watershed Management in Hawaii: An Historical Example ----

----

-----

143

But it is along and in the streams which rush through the bottoms of these narrow gorges that the Hawaiian is most at home. Go into any of these valleys, and you will see a surprising sight: along the whole narrow bottom, and climbing often in terraces the steep hillsides, you will see the little taro patches, skillfully laid out so as to catch the water, either directly from the main stream or from canals taking water out above. Such a taro patch oftenest contains a sixteenth, less frequently, an eighth of an acre. It consists of soil painfully brought down from above, and secured by means of substantial stone walls, plastered with mud and covered with grass, strong enough to withstand the force of the torrent. Each little patch or flat is so laid that a part of the stream shall flow over it without carrying away the soil,· indeed, it is expected to leave some sediment. And as you look up such a valley you see terrace after terrace of taro rising before you, the patches often fifty or sixty feet above the brawling stream, but each receiving its proper proportion of the water (Nordhoff 1974, 77). But eventually the ecologically sound ahupuaa system, which had served the Hawaiians so well when their economy was primarily one of subsistence, succumbed to a modern land management system based on the desire for immediate profits and less concern for the longer term sustainability of natural resources. But, while it lasted, it was a model of efficient, conservative use of the land and its resources. An elaborate system of institutional arrangements had been developed over a period of time to guide the use of effective resource management actions. Very strict implementation tools were used such as outright prohibition, and the system was organized to enforce these regulations effectively. Although times have changed and subsistence economies rarely exist in the modern world, there is much, even today, that we can learn from the ahupuaa system.

REFERENCES Arago, J. 1823. Narrative of a Voyage Round the World During the Year 1817 to 1820. Vol. 1. London: Trenttel and Wurtz. Campbell, A. 1967. A Voyage Round the World from 1806 to 1812. Honolulu, HI: University of Hawaii Press. Earle, T.K. 1978. Economic and Social Organization of a Complex Chiefdom: The Halelea District, Kauai, Hawaii. Museum of Anthropology, University of Michigan. Handy, E.S.C., and E.G. Handy. 1972. Native Planters in Old Hawaii: Their Life, Lore, and Environment. Honolulu, HI: Bishop Museum Press.

144

Joseph R. Morgan

Menzies, A. 1920. Hawaii Nei 128 Years Ago, ed. and publisher W.F. Wilson. Honolulu, HI. Nakuina, E. 1893. The Punahou spring: A legend. In Throm's Hawaiian Annual, Honolulu, HI. Nordhoff, C. 1974. Northern California, Oregon, and the Sandwich Islands. Centennial edition. Berkeley, CA: Ten Speed Press. Portlock, N. 1789. A Voyage Round the World in 1785-1789. London: Stockdale. Thayer, W.W. 1916. A Digest of the Decisions of the Supreme Court of Hawaii. Honolulu, HI. Vancouver, G. 1798. A Voyage of Discovery to the North Pacific Ocean and Round the World Performed in 1790-1795. Vol. 1. London: Robinson.

CHAPTER 11

Annexation, Alienation, and Underdevelopment of the Watershed Community Anis A. Dani

It is well understood that the biophysical elements of a watershed or subwatershed are interrelated. It is also accepted that the resources of a watershed are utilized both by those communities living within the watershed and by those living downstream. What has not been emphasized in the past is that, by virtue of residence, the social groups living within a watershed form a special interest group in relation to the resources of that watershed. These groups manage, conserve, and also initially utilize those resources unless external agencies intervene. It is proposed that upland populations residing within a watershed be dealt with as a unit in the context of watershed management and be called the watershed community. This term is used in a generic sense to refer to watershed communities as a generalized collectivity in relation to lowlanders. The watershed community is comprised of small and dispersed settlements in upland watersheds. Some cultural and ethnic groups tend to live in different ecological niches, resulting in further scial fragmentation. The dispersed and inaccessible nature of the terrain results in the watershed community being relatively underdeveloped and largely unaffected by development programs. Members of the community eke out a bare subsistence from intensive (terracing) or extensive (shifting) cultivation. The community contributes little to the total economic sector of the country (except in Bhutan). Finally the biophysical limits of the upper watersheds compel families to share resources within the community watershed, often leading to the development of a symbiotic relationship among its constituent segments. Historically, the watershed community has never been involved in affairs of the nation state and the macro world. Those in the community fail to understand how they have suddenly been included in the larger social universe and why they are being asked to pay the price for this inclu-

145

146

Anis A. Dani

sion by being asked to conserve their natural resources for the apparent benefit of the lowlanders. The watershed community is perceived by the lowlanders as being parochial and incapable of providing for their economic needs or looking after their environment. Ironically, those previously labeled optimists have had the time to assess the reality. Messerli (1984, 48) asserts that "A lot of farmers know more about ecology than any scientist:' He argues that if scientists take the trouble of asking farmers instead of telling them what to do, they will see the social or political constraints, such as the land tenure or the size of landholdings that prevent farmers from improving their farming practices. Alternatively, intensification of agriculture on marginal, impoverished land under the pressure of increasing population can also eliminate the possibility of improved conservation practices unless alternative means of livelihood are available. Unfortunately, the complexities involved in resource utilization and conservation at the watershed community level are rarely understood by the decision makers. This lack of understanding has led to legislation and administrative interventions to control resource utilization, which, in fact, displace the watershed community to more marginal areas where ecological conditions are even more precarious. More often than not, the watershed community again is blamed for indulging in careless utilization of resources. When major decisions affecting the lives of the watershed community are made, the community is not consulted and may not even be informed until after the fact. When reservoirs, dams, or roads are built, the watershed community rarely has any say in the matter. Every one of these decisions, however, has a significant impact on their welfare. They often have to pay a price in the form of loss of whatever meager resources they have for ''the benefit of the larger good" of the country. This may take the form of resettlement or requisition of resources such as land, water, or trees for which they may or may not be compensated. Even when the actual task of resource conservation is done by project staff from the outside, the watershed community is inevitably involved either directly through loss of resources or indirectly by the unplanned side effects of development. The lopsided nature of the relationship between lowlanders and the watershed community was discussed in Chapter 7; these relationships are illustrated in Figure 11.1. For both decision making and benefit sharing, the balance is heavily tilted toward the low landers. On the side of implementation and resource conservation, however, inputs are almost completely from the watershed community. To understand the cause of the unequal relationship between the watershed community and the decision makers (who almost always are lowlanders), an analysis of the process of incorporation of the watershed community within modern nation-states is essential.

The Watershed Community

147

THE ECOLOGY OF THE HINDU KUSH-HIMALAYA In a geological context, the Hindu Kush-Himalaya is in a stage of infancy. The tectonic movements, which led to Himalayan orogeny, are still continuing and, as a result, earthquakes and landslips are common in the area. The extraordinary height of the mountain ranges in combination with the monsoons-which unleash from 1,000-5,000 mm of rain during the short span of June to August-results in landslides, extensive surface and gully erosion, and sediment-laden waters flooding the Indus, Gangetic, and Brahmaputra plains of Pakistan, India, and Bangladesh, respectively. During the past century a matrix of political, economic, and demographic changes has been superimposed on this inherently unstable landscape. The region has been agglomerated and carved up into a series of political and administrative units that often bear little relevance to ecological, cultural, or ethnic boundaries. Geopolitical events in the wake of World War II resulted in an increase in tension and hostility within the Hindu KushHimalayan region. This has prevented the development of a concerted regional effort to understand and cope with resource management problems of the region. Simultaneously, the economic penetration of the hinterlands and demographic growth have increased pressure on the existing resource base. Goods that were once abundant-forests for fuelwood and fodder, land for cultivation, pastures for grazing, and, in some places, even water for drinking and irrigation-are now felt insufficient to support even existing populations. The continuing depletion of the resource base is causing increasing anxiety. The problem of environmental degradation, particularly deforestation, has received considerable attention from scholars and researchers who have been interested in the Himalayan region. The alarmists (Eckholm 1975, 1979; Mauch 1976; Maddie 1981; Rieger 1981) seem to have had an edge over the optimists (Ives 1981, 1984; Messerli 1984) in the debate. While dealing with the worldwide process of environmental stress, Eckholm (1979) identified land clearing for agriculture and wood gathering for fuel as the two major causes of deforestation. Bajracharya (1983a) has argued that the primary cause of deforestation is the necessity to clear forests for increasing agriculture and fodder production rather than for fuelwood. However, when this process of land conversion reduces the sustainable supply of fuelwood, additional fuelwood extraction also begins to cause deforestation (Bajracharya 1983b ). While accepting the seriousness of loss of fertile agricultural land and forest on the Himalayan slopes, Ives (1981) suggests that considerable geomorphological and paleoecological data are required before a viable longterm response to the problem can be agreed upon. However, local scholars seem to be impatient because of the urgent situation. Gurung (1981) has

148

Anis A. Dani

criticized lves for expressing a myth about ecological balance between humans and mountains. He goes on to emphasize that the precarious ecology of the Nepalese hills cannot be improved without first tackling the economic poverty of the people. Whether humans are the prime cause of environmental degradation or merely an added factor contributing to this process, if this process is to be checked and ultimately reversed, only humans can make a conscious effort toward this end. And human beings will not do so unless their needs and priorities are taken into account. This self-evident truth is often neglected by professionals who formulate programs for resource conservation and resource management. This truth must be accounted for in the very conceptualization of watershed management programs if they are to have any long-term impact on a self-sustained basis. Thus, the focus of this chapter is on the watershed communities in the Hindu Kush-Himalayan region and on the necessity for them to play an integral part in the planning and implementation of watershed management programs.

THE PROCESS OF INCORPORATION The recent history of many Third World countries is one of the incorporation of smaller communities within larger nation-states. The watershed community of the Hindu Kush-Himalayan region is no exception. Although many studies of incorporation processes have been conducted (Bailey 1960; Cohen and Middleton 1970; Bujra 1971), they do not appear to have been applied to the watershed community as a social group or to have focused on resource management. Consequently, the conceptual tools developed are inadequate for the task. Three concepts have been developed as analytic tools to deal with these complexities in the development process. They are annexation, alienation, and underdevelopment. The concepts embedded in these tools overlap considerably; yet, each is distinct in its emphasis. They have been separated for heuristic purposes but should be seen as parts of a single process. The literal definition of annexation is ''the addition of something as a subordinate part, or the attachment as an attribute or consequence:' By definition, annexation implies a secondary function for the annexed part slightly removed from the primary entity-in this case, mainstream society. It is precisely in this sense that the concept of annexation is found more useful than that of incorporation, which simply means to combine with or unite to form a legal corporation of entity, implicitly as equal partners. However, it should be clarified that annexation is not being used in the colloquial sense of territorial annexation but in a much broader sense including the social, economic, and cultural dimensions of being appended to a larger, previously external entity.

The Watershed Community UPLANDS

149

LOWLANDS

.

Resource Management Actions

Watershed Infrastructures

• Contour stripping • Afforestation • Check dams

• Roads • Dams • Irrigation schemes

• Gully control • Terracing

• Hydroelectric plants

yvatershed Management Benefits

I

I

• Erosion control

I

• Slope stab[lization

I I

• Reduced sedimentation • Regulated waterflow • Flood control • Energy production

• Protection of forest resources I

ltt:t1'1IO

Decision making

k'''"""l

Resource input in implementation

-

Benefit flow

Figure 11.1 Highland-lowland relationships in watershed management. Note: Width of arrow indicates relative magnitudes.

Alienation literally means estrangement. Simultaneously it also means the transfer of ownership. In this chapter alienation refers both to the practical, material alienation of communities from their resources as well as to the cognitive sense of helplessness, impotence, and withdrawal stemming from the alienation of the resource base.

150

Anis A. Dani

Material alienation can take the form of nationalization of common property resources such as forests, pastures, or minerals formerly considered free goods or, at least, were accessible to the watershed community. It can also take the form of formalization of property rights through ownership deeds, which have to be registered in the land revenue office. These title deeds then take priority over traditional rights of usufruct, but the title deeds themselves are not directly related to the realities of cultivation or resource utilization. They are alien to the real act and are easily amenable to manipulation and expropriation by the literate, money-lending elite. Cognitive alienation stems from the realization that an external entity now has more authority over their lives and their resources than have the members of the watershed community themselves. Initial resistance and bravado soon give way to frustration and helplessness, leading to the feeling of impotence. More critical is the manifestation of material alienation in the cognitive sphere. The inevitable result of cognitive awareness of alienation from the resource base is the alienation of responsibility for maintenance of that base. Once the local people realize that they have lost control over their resources, they have no interest in maintaining those resources even if they still retain some rights of usufruct. Traditional resource management systems break down or become redundant, and the sole responsibility for maintenance of the resource base is believed to be vested in the external authority. Underdevelopment literally means "not fully developed; below its potential economic level:' The term, however, has been adopted by dependency theorists for situations where specific interventions occurred. Colonialism, for example, disrupted the gradual process of evolutionary development and replaced it with a form of development that catered to the needs and priorities of the interveners. As a result, the subservient communities had to undergo a distorted form of development (Rodney 1972; Leys 1975). lf we bear in mind the lopsided relationship between the watershed community and the lowlanders, the relevance of this concept of underdevelopment emerges. Having been incorporated into larger entities-nationstates-the watershed communities are subjected to development plans that may not benefit them directly. The negative consequences of these development programs are the manifestations of underdevelopment. When the total process of incorporation is broken down into its three components of annexation, alienation, and underdevelopment, a clearer picture emerges. INCORPORATION OF THE WATERSHED COMMUNITY Although in the biophysical sense upland watersheds have always been integrated with their respective downstream areas, the incorporation of watershed communities within larger national entities is a relatively recent

The Watershed Community

151

phenomenon. Without going into all the historical details, one can generalize to say that in the Hindu Kush-Himalayan region the major centers of power have rested in the lower plains. These centers were content to ignore the watershed community which happened to be located in the periphery on both the Chinese and the Indo-Gangetic side of the mountains. Improvements in communication and transport in the nineteenth century made these peripheral areas more accessible. The rise of nationalism in the twentieth century and the search for national resources accelerated the incorporation process. The watershed community had neither the means nor the incentive to incorporate the societies of the plains. The latter, however, started reaching out to the watershed community.

Annexation The first manifestation of this incorporation process was in the annexation of the watershed community. The power centers downstream (mainstream society) started absorbing entire communities into their domain. It is important to understand that this annexation was not territorial annexation, although that also occurred in some cases, but the annexation of entire communities. The watershed community, which was until now ignored as tribals, suddenly found themselves part of larger units. For example, the Hunzukut, Kohistanis, and Nuristanis were absorbed into Pakistani and Afghani societies. This annexation not only strengthened the manpower base of the nations but also opened up the resources of the mountain areas to mainstream society. Economic penetration of the hinterlands, both by government fiat and by market forces, was inevitable. At the same time considerable migration from the hills to the plains took place as the watershed community sought a share of the economic development taking place downstream. Nevertheless, the distinct identity of the watershed community was lost in the tidal wave of the new national identity, much as first- and second-order streams lose their identity as they merge with third- and fourth-order streams and then with the larger river. In the Hindu Kush-Himalayan region, two countries-Nepal and Bhutan-stand out as exceptions to this scenario. These exceptions, however, tend to prove the rule rather than refute it. Both countries have survived largely as buffer zones between India and China. Although both Nepal and Bhutan can justifiably claim distinct national identities of their own, they have not completely escaped annexation in the economic and cultural sense. Being landlocked and industrially undeveloped, both countries are dependent on India for most of their imports. Migration of a significant scale between these countries dates back to the beginning of the twentieth century. Nepal has provided Gurkha soldiers for the British and subsequently the Indian army as well. The number of Nepalese abroad increased steadily from an estimated 82,000 in 1941

152

Anis A. Dani

to 403,000 in 1981. Simultaneously, the open border with India has led to reverse migration with an estimated 400,000 Indians now settled in Nepal's southern plains called the Terai (Karki 1985). A sizable Nepalese population, who are descendants of early settlers brought in during the late nineteenth and early twentieth centuries to work the foothills, reside in southern Bhutan. Their numbers have been swelled by immigrants who spontaneously followed the direction of labor flow. People of Nepalese origin are estimated to constitute 20-30 percent of the total population of Bhutan (The World Bank 1984). Thus the annexing role of dominant countries has had an impact on Nepal and Bhutan as well.

Alienation Among the watershed community, alienation from resources had farreaching consequences. The Nationalization Act of 1957 in Nepal brought all forested land, as well as trees planted on private lands, under government ownership. This act raised insecurity regarding use-rights of trees. People cut down trees by the thousands, fearing that the government planned to do so. Nationalization resulted in deforestation by the same people who previously had maintained the forests. In other countries, such as Pakistan, subsurface resources including minerals and natural gas were declared to be the dominion of the federal government. Extraction of these minerals was done by government departments or by contractors who were granted leases for this purpose. Members of the watershed community rarely had the access and resources necessary to acquire such lucrative lease options. For resources such as natural gas, which could not be exploited by the local population, the immediate impact was not apparent. But where other resources such as coal were at stake, many local people lost their livelihood and migrated to other areas. Thus, resource alienation in watersheds has led to subsequent problems downstream. Just as important as material alienation is cognitive alienation. The mores and cultural strategies that the watershed community rely on to maintain its ~ocial existence often become the subject of ridicule. Conscious attempts are sometimes made to pressure the watershed community into accepting the cultural norms of mainstream society which, in some cases, may go to the extreme of political indoctrination or imposition of religious orthodoxy. The cognitive alienation from responsibility for maintenance of the resource base is probably the most critical form of alienation. As discussed in the case of Hawaii, the watershed community has often relied on indigenous management systems to maintain their resources at a sustainable level (Messerschmidt 1981). Material alienation will also result in removal of this sense of responsibility. Sometimes it even renders the continuation of responsibility for maintenance impossible. The traditional resource management system then becomes defunct.

The Watershed Community

153

When responsibility for maintenance lies with an authority not physically present on site, effective maintenance is difficult. In a few years after alienation occurs, encroachments, misuse, and theft may result. When trees are gone, it is only a matter of time before resultant erosion affects slope stability and productivity.

Underdevelopment The most common manifestation of underdevelopment in upland watersheds is the negative impact of roads. While roads may be constructed to facilitate transport for the watershed community, more often than not, roads are constructed because the government wants to link certain parts of the country for economic, political, or military reasons, or even for tourism. That the watershed community may also benefit from road construction is often incidental. Roads have been identified as the single most hazardous form of development in upland watersheds (Easter and Brooks 1985). Carelessly constructed roads act as catalysts for destabilization. Landslides may be precipitated by the debris dumped on the lower side of newly constructed roads. Considerable forest destruction can also be caused by use of available trees as fuelwood to heat the bitumen for road construction and to meet personal fuel requirements of construction laborers. Not all road construction need be environmentally destructive. The Lamosangu-Jiri Road Project in Nepal is an example of what can be accomplished with foresight and ecologial awareness. The implementors used a special bitumen emulsifier that required no heating, thus saving a sizable forest area in constructing a 110 km road. They also took remedial measures such as afforestation and cleaning up of debris wherever necessary to protect against landslides. The net cost involved in such measures for ecological stability was I percent of the total cost of road construction. This marginal cost should be viewed as an investment because this road will be less prone to damage by landslides. In a few years the investment is likely to be paid off by lowered maintenance costs. Road construction is not the only activity that results in underdevelopment of the watershed community. Afforestation programs can fence off livestock from vital pasture. The construction of dams and barrages can flood large tracts of highly productive land. Dams also affect marine life by preventing fish from traveling upstream to their spawning grounds. Underdevelopment is not restricted to the watershed community; the rural population downstream can also be affected, as was observed in Sind, Pakistan. The construction of Ghulam Mohzmmad Barrage, the artificial expansion of Kinjhar Lake, and the rerouting of the waters of Haleji Lake to provide drinking water to the Karachi metropolis had an impact on the silt content and water quality in canals. This changed the natural vegetation and consequently depleted fodder crops in what formerly was

154

Anis A. Dani

predominantly a livestock area in the Thatta District. When the rural population realized that forces beyond their control had changed the resource base, they shifted to rice cultivation and vegetables as cash crops. Not knowing the technology of these crops, yields were poor. The rural population also found themselves completely at the mercy of the middlemen in the urban markets. When the Sind project started in 1982, the rural population exhibited signs of both underdevelopment and alienation. They reacted with pessimism and apathy toward the entire process of development. They felt they were not in control of their lives and, since the government was responsible for this situation, the same government would have to come in to bail them out. The only kind of development activity they could conceive of was handouts from the government. They were skeptical about the impact of development efforts and therefore were extremely reluctant to take a more active interest in the project. It took several months of dialogue to break this barrier and to initiate a more meaningful development process. The Sind experience indicated some of the difficulties of involving rural communities in resource management activities. These difficulties are exacerbated when dealing with the watershed community, which may not even benefit from resource management activities in the short run although their cooperation is vital for their success (see Figure 11.1 ). RESTORING THE BALANCE The foregoing analysis of the watershed community was intended to establish the nature of three vital dimensions of the incorporation process that result in an unequal relationship between the watershed community and mainstream society. This suggests that the balance, which needs to be rectified in the Hindu Kush-Himalayan region, is not just ecological but also social, political, and economic, as observed by Huntings (1961) more than two decades ago. The ecological balance cannot be restored without also restoring the balance in human relationships. The attention being paid to highland-lowland interactive systems and participation of the watershed community may be wasted unless the fundamental relationship is stabilized. In order to conduct meaningful research in the field of integrated watershed management, the field of inquiry has to be widened to include the fundamental relationships under which the watershed community coexists with lowlanders. The watershed community, being based physically in the midst of the resources of concern, should have been the logical choice for bearing the responsibility of maintaining these resources. As mentioned above, their links with these resources have been distorted and, in some cases, even severed.

The Watershed Community

155

Many decision makers, on the other hand, may never have had strong links to these resources because their priorities have often been quite different. In most countries of the Hindu Kush-Himalayan region, these priorities have been constrained by the imperatives of development. Watershed resources, however, are reusable only as long as they are allowed to regenerate. Once the downward spiral of overexploitation starts, it is extremely difficult to reverse, thus creating a sense of urgency. Existing efforts at watershed management need to take account of the differential access to decision making, resource inputs, and benefit sharing. Perhaps the most appropriate and effective way will be for the lowlanders to subsidize resource conservation and watershed management efforts by the watershed community, as suggested by Easter and Brooks (1985). Such subsidies from the urban to the rural highland sectors are already in vogue in Switzerland where the mountain regions are maintained primarily for tourism and as winter sports areas. Implementation of a subsidy scheme in the Hindu Kush-Himalayan region will certainly not be an easy task. For countries such as China, India, and Pakistan, the highland-lowland relationship is an internal one; whereas, for countries such as Nepal, Bhutan, and Bangladesh, subsidies would involve international transactions-something unlikely in the foreseeable future. Subsidies alone will not work. They need to be complemented by concerted efforts directed specifically at the watershed community to redress the situation caused by the incorporation process. The watershed community's relationship with the resources of the watershed has to be restored if it is to share the responsibility of maintaining those resources. It can best be motivated by initiating activities that provide immediate relief to the community's pressing problems (Panday 1978-79). The incorporation process is irreversible, but it can certainly be modified to make it more positive. At the same time, if the watershed community is expected to share its resources with the nation or the region, an effort has to be made to provide them with some environmental education, enabling them to better understand their role. The most obvious step is to break the monopoly of information that lies with the urban-lowland elite. Mobilization of human resources for the environmentalist movements in the Euro-American sphere was possible only when information on the subject became widely known. In fact, a major portion of the resources of environmentalist groups is still devoted to information dissemination. Past public awareness concerning environmental issues has been rather limited in the Hindu Kush-Himalayan region. Relatively greater attention is being paid to these issues in Bangladesh, Nepal, and India than in other countries of the region. If the aim is to have a serious long-term impact on improving the watershed environment, much more information dissemi-

156

Anis A. Dani

nation is needed to improve the public's understanding of land and water conservation issues. CONCLUSION The term watershed community has been used specifically to refer to the communities residing in upland watersheds as opposed to lowlanders and decision makers in the metropolis, who also tend to be lowlanders. Since it is the watershed community that lives closest to the watershed's resource base, they should have a more active role in maintaining those resources, not only for themselves but for the entire Hindu Kush-Himalayan region. However, the watershed community will not fulfill this obligation unless their needs and priorities are taken into account. While a watershed is integrated ecologically through the interrelationship of its biophysical resources, it is also integrated socially by the utilization patterns of those resources. This social integration often manifests itself in the form of symbiotic relations among the constituent segments of the watershed community. However, this symbiosis is not essential for the watershed community to be treated as a unit. The relationship of the watershed community to its resource base is marred when the community is subordinated to an authority external to the watershed. Three analytic tools-annexation, alienation, and underdevelopmentwhich represent different faces of the incorporation process are useful in understanding the subordination of the watershed community. The relationship of the watershed community to the lowlanders is clearly one of subordination. The watershed community gradually becomes divorced from its resource base, which falls under the sway of the low landers. Yet the active involvement of the watershed community is crucial if the watersheds are to be maintained in a state of ecological stability. Awareness of the unity of the ecosystem and the social system does exist among some professionals. What is needed is the development of suitable programs that account for this unity and are conceptualized with a thorough understanding of the development of the relationships among the actors involved. Remedial measures will then be needed to correct the human imbalance in order to achieve the desired ecological ends. REFERENCES Bailey, F.G. 1960. Tribe, Caste and Nation. Manchester: Manchester University Press. Bajracharya, D. 1983a. Deforestation in the food/fuel context: Historical and political perspectives from Nepal. Mountain Research and Development 3(3):227-240.

The Watershed Community

- - - - -

157

Bajracharya, D. 1983b. Fuel, food and forest? Dilemmas in a Nepali village. World Development 11(12):1057-1074. Bujra, A.S. 1971. The Politics of Stratification. Oxford: Clarendon Press. Cohen, R., and J. Middleton. 1970. From Tribe to Nation in Africa: Studies in Incorporation Processes. San Francisco: Chandler. Easter, KW., and K.N. Brooks. 1985. Integrated watershed management in Asia and the Pacific. A background paper for the Workshop on Integrated Watershed Management, 7-11 January, East-West Environment and Policy Institute, Honolulu, Hawaii. Eckholm, E.P. 1975. The deterioration of mountain environments. Science 189:764-770. Eckholm, E.P. 1979. Losing Ground: Environmental Stress and World Food Prospects. New York: WW. Norton. Gurung, H. 1981. Ecological change in Nepal: A native interpretation. New ERA Occasional Paper 001. Kathmandu. Buntings Technical Services. 1961. Mangla Watershed Management Study. Vols. 1 and 2. Borehamwood, Herts, England. Ives, J.D. 1981. Applied mountain geoecology: Can the scientist assist in the preservation of the mountains? In The Himalaya: Aspects of Change, ed. J.S. Lall, 377-402. Delhi: Oxford University Press. Ives, J.D. 1984. Current approaches to research and development in the Hindu Kush-Himalayan Region. In Proc. First International Symposium and Inauguration of the International Center for Integrated Mountain Development (ICIMOD), 54-56, 1-5 December 1983, Kathmandu. Karki, Y.B. 1985. Country Demographic Profile: Nepal. OFEG Working Paper 2. Kathmandu: ICIMOD. Leys, C. 1975. Underdevelopment in Kenya: The Political Economy of Neocolonialism. Berkeley: University of California Press. Mauch, S.P. 1976. The energy situation in the hills: Imperative for development strategies? In Mountain Environment and Development. Kathmandu, SATA (Swiss Association for Technical Assistance in Nepal). Messerli, B. 1984. Highland-lowland interactive system on a local, national, and international level. In Proc. First International Symposium and Inauguration of the International Center for Integrated Mountain Development (ICIMOD), 47-53, 1-5 December 1983, Kathmandu. Messerschmidt, D. 1981. Nogar and other traditional forms of cooperation in Nepal: Significance for development. Human Organization 40(1 ):40-4 7. Maddie, A.D. 1981. Himalayan environment. In The Himalaya: Aspects of Change, ed. J.S. Lall, 341-350. Delhi: Oxford University Press.

158

Anis A. Dani

Panday, K. 1978-79. Participation of extension population and soil conservation sciences. J. Nepal Research Center 2/3:195-211. Rieger, H.C. 1981. Man versus mountain: The destruction of the Himalayan ecosystem. In The Himalaya: Aspects of Change, ed. J.S. Lall, 351-376. Delhi: Oxford University Press. Rodney, W. 1972. How Europe Underdeveloped Africa? Dar-es-Salaam, Tanzania: Tanzania Publishing House. Reprinted by Howard University Press, Washington, D.C., 1974. The World Bank. 1984. Bhutan: Development in a Himalayan Kingdom. A World Bank Country Study, Washington, D.C.

CHAPTER 12

The Role of Extension: A Northern Thailand Watershed Case Study Peter W.C. Hoare 1

This chapter focuses on the role of extension and its importance in integrated watershed management. Highlighted are some of the problems facing governments when they implement complex projects requiring coordination among numerous government agencies. The discussion is divided into two parts: • a description of the physical, social, and institutional/organizational setting for watershed management in Northern Thailand; and • the role of extension in integrated watershed management, which is illustrated by the extension program being used in Northern Thailand to enable communities to participate in watershed plan formulation and implementation. NORTHERN THAILAND SETTING Northern Thailand is bordered to the east by Laos and to the north and the west by Burma (see Figure 12.1). The Far North part consists of eight provinces covering an area of 89,484 km', with a population of about 4.8 million. Sixty-two percent of the area (55,500 km') is within the Chao Phraya River watershed, 20 percent in the Mekong River watershed, and the 18 percent bordering Burma is within the Salween River catchment (Donner 1978). The Chao Phraya River is the major Thai river system. It feeds the Central Plain where 5 million ha of rice are planted annually

I. Acknowledgment is made to the Australian Development Assistance Bureau, the Public Welfare Department of the Kingdom of Thailand, and the Australian Agricultural, Consulting, and Management Company for the opportunity to do this work. Also, I gratefully acknowledge the assistance of my former colleagues of the Highland Agricultural and Social Development Project.

159

160 ggo

1'/ Loh Pa

BURMA

(

/

I

I1

I

I

Mae Hong

LAOS

/ * 8,,--... on ,)

/

Mae

Chiang / * Mai / \

~\

Lamphun:(

,'~ *

Mae

I

Chaem

/

/

e

\ \

: \

,

\

-

:

I

I

t\

\ ,,...- ,_,

\

."'

.

/

I

1

I

\

........

~ I

\

,

//

I

/I

'...._,

-,

1 I I

I

//-