irrigation and river basin management - agecon search

Irrigation and River Basin Management

Options for Governance and Institutions

1A4859 - Svendsen - Vouchers Proofs #P.prnZ:\Customer\CABI\A4831 - svendsen\A4859 - Svendsen - Vouchers Proofs #P.vpFriday, January 14, 2005 10:34:25 AM

Color profile: DisabledComposite 150 lpi at 45 degrees

Irrigation and River Basin Management

Options for Governance and Institutions

Edited by

Mark Svendsen

Consultant, Oregon, USAand

Fellow, International Water Management Institute

CABI Publishingin association with the

International Water Management Institute



CABI Publishing is a division of CAB International

CABI PublishingCAB InternationalWallingfordOxon OX10 8DEUK

Tel: +44 (0)1491 832111Fax: +44 (0)1491 833508E-mail: [email protected] site: www.cabi-publishing.org

CABI Publishing875 Massachusetts Avenue

7th FloorCambridge, MA 02139

USA

Tel: +1 617 395 4056Fax: +1 617 354 6875

E-mail: [email protected]

Published in association with:International Water Management InstitutePO Box 2075ColomboSri Lankawww.iwmi.cgiar.org

©CAB International 2005. All rights reserved. No part of this publication maybe reproduced in any form or by any means, electronically, mechanically, byphotocopying, recording or otherwise, without the prior permission of thecopyright owners.

A catalogue record for this book is available from the British Library, London,UK.

Library of Congress Cataloging-in-Publication DataIrrigation and river basin management : options for governance andinstitutions / edited by Mark Svendsen.

p. cm.Includes bibliographical references and index.ISBN 0-85199-672-8 (alk. paper)1. Watershed management. 2. Watershed management--Case studies.3. Irrigation. 4. Arid regions. 5. Integrated water development.I. Svendsen, Mark, 1945– II. Title.

TC405.I77 2004333.91′3--dc22 2004007807

ISBN 0 85199 672 8

Typeset by AMA DataSet Ltd, UK.Printed and bound in the UK by Biddles Ltd, King’s Lynn.



Contents

Contributors vii

Preface xi

1 Managing River Basins: an Institutional Perspective 1Mark Svendsen, Philippus Wester and François Molle

2 Phases of River Basin Development: the Need for Adaptive Institutions 19David Molden, R. Sakthivadivel, M. Samad and Martin Burton

3 Limits to Leapfrogging: Issues in Transposing Successful River BasinManagement Institutions in the Developing World 31Tushaar Shah, Ian Makin and R. Sakthivadivel

4 Making Sound Decisions: Information Needs for Basin Water Management 51Martin Burton and David Molden

5 Financing River Basin Organizations 75Charles L. Abernethy

6 Water Management for Irrigation and Environment in a Water-stressed Basinin Southwest France 93Henri Tardieu

7 Basin Management in a Mature Closed Basin: the Case of California’s CentralValley 109Mark Svendsen

8 River Basin Closure and Institutional Change in Mexico’s Lerma–ChapalaBasin 125Philippus Wester, Christopher A. Scott and Martin Burton

9 Water Resources Planning and Management in the Olifants Basin of SouthAfrica: Past, Present and Future 145M. de Lange, D.J. Merrey, H. Levite and M. Svendsen

v



10 Water Resource Management in the Dong Nai Basin: Current AllocationProcesses and Perspectives for the Future 169Mark Svendsen, Claudia Ringler and Nguyen Duy Son

11 Governing Closing Basins: the Case of the Gediz River in Turkey 193Mark Svendsen, D. Hammond Murray-Rust, Nilgün Harmanciohlu andNecdet Alpaslan

12 Managing River Basins: Lessons from Experience 215Mark Svendsen and Philippus Wester

13 Providing Irrigation Services in Water-scarce Basins: Representation andSupport 231Philippus Wester, Tushaar Shah and Douglas J. Merrey

Index 247

vi Contents



Contributors

Charles Abernethy is a consultant on irrigation and water resources management, basedin Colombo, Sri Lanka, where he was Senior Technical Advisor at the InternationalIrrigation Management Institute until 1994. He works mainly in Asian countries, asa resource person assisting training courses and workshops on institutional andmanagement questions related to water. E-mail: [email protected]

Necdet Alpaslan is a civil engineer and specializes in water quality, wastewater treatmentand management of solid wastes. He is a professor at Dokuz Eylul University, Izmir,Turkey, and the Director of Center for Environmental Research and Development(CEVMER). He has been working on various research and application projects, andgiving consultancy services in these fields for 25 years. E-mail: [email protected]

Martin Burton is a Director of ITAD~Water Ltd, and an IWMI Fellow. He specializes inwater management and capacity building, and is applying his recent MBA training tothe management of change within the irrigation and water resources sector. He haslived and worked in over 25 countries. E-mail: [email protected]

Nilgün Harmanciohlu is a professor of hydrology and water resources, affiliated withDokuz Eylul University (DEU) Faculty of Engineering in Izmir, Turkey, since 1976.Her areas of research are water resources management, watershed modelling, anddecision support systems as applied to water resources. She is currently the directorof the Water Resources Management Research & Application Center (SUMER) ofDEU and head of the Water Resources Department of the University. E-mail: [email protected]

Marna de Lange is a South African who trained as an engineer and works in rural develop-ment and poverty alleviation, with a specific focus on household food security.Her activities span hands-on implementation, training and capacity building ofdevelopment workers, research and policy brokering. E-mail: [email protected]

Hervé Lévite is a Researcher based at IWMI’s Africa office in Pretoria. He is an engineer bytraining and works on water resources management issues under the co-supervision ofCEMAGREF (France). E-mail: [email protected]

Ian Makin is IWMI Regional Director in Southeast Asia. He was trained as an irrigationengineer and has extensive research experience in water resources and irrigation sys-tem operations and management in South and Southeast Asia developed over 27 yearsworking in many countries of the region and in Africa. E-mail: [email protected]

Douglas J. Merrey is a social anthropologist who has worked on water and irrigationinstitutional issues since the mid-1970s. He is presently IWMI’s Director for Africa.His research is focused on the institutional arrangements at local, river basin and

vii



national levels for management of water resources, especially for irrigation. He haslived and worked in Pakistan, India, Sri Lanka, Indonesia, Egypt and South Africa,and has done shorter-term assignments in a number of other developing countries.E-mail: [email protected]

David Molden is a Principal Researcher at the International Water Management Institutebased in Sri Lanka. He has 25 years experience in water resources in the fields ofirrigation, groundwater, river basin analysis and institutions, with resident assign-ments in Botswana, Egypt, Lesotho, Nepal and Sri Lanka. E-mail: [email protected]

François Molle is a Senior Researcher at the Institut de Recherche pour le Développementin France, and currently serves at IWMI in Colombo. His main areas of work andpoints of interest are irrigation and river basin management and water policy.E-mail: [email protected]

Hammond Murray-Rust was team leader for IWMI in Turkey, and subsequently ThemeLeader for IWMI’s research on integrated water management for agriculture basedin Colombo, Sri Lanka, and then Hyderabad, India. He is now Senior Associate atARD Inc in Burlington, Vermont, USA. He specializes in integrated modelling forwater management, determinants of productivity of water, irrigation performanceassessment, and use of remote sensing for water management. E-mail: [email protected]

Claudia Ringler is a Research Fellow with the International Food Policy Research Institutein Washington, DC. Her research interests are in river basin management and naturalresource policies for developing countries. E-mail: [email protected]

Ramaswamy Sakthivadivel is an IWMI Senior Fellow stationed at Chennai, India, andworks half time with IWMI as consultant. Trained as a civil engineer, he specializes inirrigation and water management with special reference to groundwater and small-scale irrigation systems. He has been with IWMI for more than 15 years, working in adozen countries on his area of specialization. E-mail: [email protected]

Madar Samad is Theme Leader for Water Resources Institutions and Policy at theInternational Water Management Institute, Colombo, Sri Lanka. He is an agriculturaleconomist by training. His research interests are irrigation and water resourcesinstitutions, and agricultural development. He has carried out research in thesefields in several Asian and African countries during the last 10 years. E-mail:[email protected]

Christopher Scott is a hydrologist with over 15 years of research and applied hydrologyand water resources expertise, including 6 years research and management experiencewith IWMI in Mexico, USA and India. He is currently IWMI South Asia RegionalDirector, based in Hyderabad, India. E-mail: [email protected]

Tushaar Shah is a Principal Scientist at IWMI based in Gujarat, India. He leads IWMI’sglobal research on sustainable groundwater management. Trained as an economist andmanagement specialist, Shah’s work has focused on irrigation and water managementinstitutions in Asia and Africa. E-mail: [email protected]

Nguyen Duy Son is currently an Operations Officer with the World Bank in Hanoi. For-merly, he was a Land and Water Resources Planner with the Sub-Institute forWater Resources Planning in Ho Chi Minh City, Vietnam. Trained as an engineer, hespecializes in integrated land and water resources development and management, andhas worked on these issues throughout Vietnam over the course of 10 years. E-mail:[email protected]

Mark Svendsen is a consultant based in Oregon, USA, and an IWMI Fellow. Trained as anengineer, he specializes in irrigation and water resource institutions, and has workedwith these issues for a wide variety of clients in 25 countries in Asia, Africa, Europe,and North and South America over the course of 30 years. Prior to becoming an

viii Contributors



independent consultant, he was a Research Fellow at the International Food PolicyResearch Institute in Washington, DC, USA. E-mail: [email protected]

Henri Tardieu is General Manager of the Compagnie d’Aménagement des Coteauxde Gascogne, a semi-public irrigation and water resources service provider insouthwestern France, and a member of the board of a French basin agency. As anengineer, he is also an international consultant in the field of water management andinstitutional evolution. E-mail: [email protected]

Philippus Wester is Assistant Professor in the Water Reforms, Irrigation and WaterEngineering Group at Wageningen University in the Netherlands. Trained as a socio-technical irrigation engineer, he has worked as a water management researcher inSenegal, Pakistan, the Netherlands, Bangladesh and Mexico in the past 10 years,focusing on irrigation management transfer, participatory water management andriver basin management. His current research interests centre on the production andarticulation of water reforms and their effects. E-mail: [email protected]

Contributors ix



Preface

When ideas and discussions chain together, with one idea leading to studies and thenadditional discussion and more new ideas, it is difficult to pinpoint the origins of thediscussion thread. That is the case with this book. In the ‘post-IMT’ world of the late 1990s,questions arose about relationships between newly independent Water User Associations(WUAs) and the national water agencies that spawned them. While the WUAs needed togain experience with independent operation, national water agencies recognized that theymust continue to support the new WUAs with training, advice and other assistance if theywere to make the transition to independent operation successfully and permanently. Thisled to IWMI research on support services required by fledgling WUAs. Of course, thisline of inquiry itself grew directly out of the experience earlier in the decade with thechallenges of establishing and empowering the WUAs in the first place.

In addition, concern arose over the relationship between the irrigation commandsmanaged by the new WUAs and the hydrology of the basins in which they were situated.More specifically, questions concerned the impact that water scarcity and competitionin the basin would have on WUAs operating within the basin. No longer wards of the state,irrigation systems, and their WUA managers, were now more vulnerable to the competitivepressures affecting water use in the basin. In addition to managing their own internal affairs,WUAs also had to be concerned with the security of their water supply and the impacts of theactions of other water users in the basin on its quality and availability.

These concerns led, in 1999, to discussions among Doug Merrey, Tushaar Shah, MartinBurton, Hammond Murray-Rust, David Molden and myself about the possibility of develop-ing a methodology to assess basins simultaneously in terms of their hydrology and theirinstitutional make-up. The feeling was that there were important connections between thetwo that could be elaborated. At the time, we called this exercise ‘hydro-institutional map-ping’. This led to the commissioning of several case studies on river basin management andto several conceptual papers such as Tushaar Shah’s stimulating ‘Limits to Leapfrogging’.It also led to a matrix template, which allowed the mapping of essential basin functions onto basin actors in order to examine, in a structure way, who was doing what in a basin.

IWMI research on this whole set of issues had been funded by the German governmentfor several years, and this work culminated in a workshop on ‘Integrated Water Managementin Water-stressed River Basins in Developing Countries’, held at Loskop Dam in SouthAfrica in October of 2000. This workshop focused primarily on river basin managementinstitutions and organizations under conditions of water scarcity, but the concern with theimplications of scarcity for WUAs was always in the background. This book is an outgrowthof that workshop.

xi



Seven of the 13 chapters in this book began as papers presented to the workshop. Theycomprise three of the book’s thematic papers and four of the case studies. Most wereextensively modified following the workshop before being included in the book. To supple-ment the three thematic chapters, I worked with Flip Wester and François Molle to write aconceptual paper on river basin institutions to complement the Molden et al. chapter onbasin closure. I also asked Martin Burton to develop a chapter on the role of informationin basin management, information being nearly as important a resource as water itself inmanaging a basin. To fill out the set of case studies, Marna de Lange et al. prepared a study onSouth Africa’s experience with establishing a representative Catchment ManagementAgency in the Olifants Basin, and I contributed a case study on Vietnam’s Dong Nai RiverBasin to add a Southeast Asian dimension. The Vietnam case was originally developed forthe International Food Policy Research Institute with support from the Asian DevelopmentBank.

Because the earlier discussions on ‘hydro-institutional mapping’ had involved many ofthe case study authors, there was a good deal of consistency in the way the authorsapproached the case studies, which was enhanced further during the interactive editingprocess. As one result, all of the studies employed the ‘essential functions’ matrix as aheuristic device and all focused, at least in part, on critically important basin governanceissues. This lent a coherence to the cases that facilitated a cross-cutting look at them, a lookwhich took the form of a pair of analyses – one focusing on basin management itself andthe other looking at implications of basin closure for small-scale irrigators. These analysescomprise the final two chapters in the book.

Clearly there are important connections between water scarcity and the institutions thatevolve to manage water-scarce basins, but while scarcity may drive the institutions’ creation,it is local human, institutional and financial resources, economic forces, political dynamics,and, sometimes, international experience which shape them. The thematic chapters in thefirst part of the book develop the concepts used in the remainder of the book, particularly theideas relating to closing basins and institutional governance, and offer suggestions, insightsand qualifications related to those concepts. The case studies then describe some of thesharply differing forms that basin management institutions have taken in response to theinfluences mentioned above. The two analytic chapters attempt to understand what drivesthe creation of basin management institutions, describe how they differ from each otheralong key dimensions, and why, in some cases, they have taken the shape they have. Wehope you, readers, will find these chapters interesting and useful.

I must first and foremost thank the 19 chapter authors, who of course made the bookpossible. Their biosketches are included in the list of contributors. Doug Merrey and TushaarShah were originally to edit this volume with me, but graciously turned the project over tome when I turned out to be the least busy of us. Our discussions during the book’s formativemoments were invaluable. Michael Devlin and Nimal Fernando of the IWMI publicationsoffice took my edited manuscripts and worked with the publisher to transform them into abook and, in addition, were very patient with me and my sometimes elastic timetable. Manyother people were also involved in one way or another whom I will refrain from trying to listbecause I will inevitably forget someone important. Thank you all.

Mark SvendsenCorvallis, Oregon

March 2004

xii Preface



1 Managing River Basins: an InstitutionalPerspective

Mark Svendsen, Philippus Wester and François Molle

1.1 Introduction

Up until the beginning of the 19th century,human water use was largely confined tostreamside uses for drinking, stock wateringand water-powered mills, as well as in-stream use for navigation. There wereexceptions in the ancient hydrauliccivilizations in Mesopotamia, Egypt, China,India and a few other locations, whichabstracted large volumes of water frommajor rivers to irrigate extensive tracts ofriparian floodplain. However, whereas thelocal impacts of these abstractions weresignificant, on a global scale, river flowregimes were still largely dictated bynatural features and forces, and water userswere primarily natural biota.

As the world population grew, from lessthan 1 billion (109) in 1800, to 1.7 billion in1900, to more than 6 billion today, humandemands for water also expanded. Growingurban concentrations, often along rivers, ledto significant abstractions of water fromrivers for these settlements and to negativeimpacts on water quality. At the same time,the industrial revolution created newdemands for water, and new technology andthe growing demand for food gave rise to anexpanding irrigated agriculture throughoutthe world.

The latter half of the 19th century sawgreat strides in the development of hydrau-lic technology for controlling major rivers

and transporting water over long distancesfor irrigation and domestic purposes. Manyof these developments took place in theAsian subcontinent, and engineers fromthe USA and other countries pilgrimaged toBritish India to learn this new technology(e.g. Wilson, 1891). The following centurywitnessed a great remaking of river systemsacross the world, as humans manipulatedthe natural hydrology to meet the domesticsupply, sanitation, food, fibre and industrialneeds of growing populations and risingstandards of living. During much of the 20thcentury, expanding water supply was theeasiest and least costly way of satisfyingthese demands, since water was relativelyabundant and the harmful impacts on theenvironment were incremental, individuallymodest and at first little noticed.

From a situation of limited, low-impactand largely riparian uses of water, we havenow reached a point where, in manyparts of the world, cumulative uses of riverresources have not just local but basin-wideand regional impacts. The result is thatwater resources in many river basins arefully or almost fully committed to a varietyof purposes, both in-stream and remote;water quality is degraded; river-dependentecosystems are threatened; and still-expanding demand is leading to intensecompetition and, at times, to strife. Inresponse, there is growing interest in man-agement systems that can bring together

©CAB International 2005. Irrigation and River Basin Management(ed. M. Svendsen) 1



fragmented water uses, and water users, intoan integrated planning, allocation and man-agement framework. A common element ofthese approaches is that they do not justcover a single water use or an administrativejurisdiction, but deal with an entire riverbasin or sub-basin, such as the Colombia, theIndus or the Limpopo.

Integrated management frameworkspromises a number of important benefits:

• Greater utility from a given amount ofwater through adjusted allocations;

• Reduced groundwater mining throughconjunctive management of groundand surface water;

• More intensive reuse of water throughplanned sequencing of uses;

• Improved water quality through morecomprehensive data collection, moni-toring and enforcement;

• Incorporation of current social andenvironmental values into water alloca-tion and management decision making;

• Inclusion of a wider range of basinstakeholders into decision making;

• Reduced conflict among users.

Despite this promise, and although highlyfashionable of late in policy circles,integrated river basin management (RBM)is rather rare in practice. Reasons for thisinclude the following:

• It requires genuine collaborationamong administrative and sectoralunits;

• It usually involves reductions in dis-cretionary authority on the part ofexisting managing agencies;

• Managers who would gain influenceover basin decision making may cur-rently be bureaucratically and politi-cally weaker than current managers;

• Its costs can be significant;• It creates uncertainty for present

resource users;• It makes planning and decision making

more complex.

Given these potentially inhibiting factors, itis not so surprising that there are not morepractising examples of integrated manage-ment of water resources at the basin level.

In order to hurdle these constraints, dissat-isfaction with the current situation mustbe intense, and the prospective benefitsof a new management regime significant. Itis these conditions that we explore in thischapter.

This chapter defines the basic elementsand concepts comprising integrated basinmanagement and other key concepts andthen focuses on the process of analysinginstitutional arrangements for RBM tofurther our understanding of institutionaldesign. To do this, we first discuss institu-tions, organizations and policies in relationto water management. We then outline anessential functions and enabling conditionsframework for analysing basin managementregimes and discuss possible institutionalarrangements for RBM that meet the needs oflocally managed irrigation.

1.2 Terms and Concepts

1.2.1 River basin water resourcemanagement

There are a number of terms used to des-cribe an integrated process of assessing andmanaging water resources at the basin level.Most are variations on the terms integratedwater resource management or river basinmanagement (RBM).

Integrated water resource management(IWRM) is a newer term and is defined by theGlobal Water Partnership (TAC, 2000) in thefollowing way:

IWRM is a process which promotes theco-ordinated development and managementof water, land and related resources, inorder to maximize the resultant economicand social welfare in an equitable mannerwithout compromising the sustainability ofvital ecosystems.

IWRM tends to have a strong normativecontent, often referring to the DublinPrinciples and emphasizing such valuesas economic benefit, equity, sustainabilityand public participation. It is implicitlysuggested that all these values can bemade commensurate and compatible, but in

2 M. Svendsen et al.



practice, there are often trade-offs betweenthem, particularly between equity and eco-nomic efficiency, and between economicwelfare and sustainability.

RBM is a more traditional term and hasmore recently broadened its meaning toencompass many of the same features andvalues which characterize IWRM. RBM isdefined by Mostert et al. (2000) as follows:

RBM is the management of water systems aspart of the broader natural environment andin relation to their socio-economicenvironment.

This definition is simpler and less overtlyvalue-laden than the previous one. Bothdefinitions encompass both planningand management of water resources, eventhough the planning element is implicit. Akey element of both concepts is that plan-ning and management units almost alwayscut across other divisions more tradition-ally used to manage resources, such assectors, provinces or even nations, andherein lies both their great strength andtheir challenge. Some divisions acrosswhich the approach we are discussingextends are shown in Box 1.1.

Typically, the boundaries of the man-agement unit are defined hydrologically, asa basin or a sub-basin, defined by Mostertet al. (2000) as follows:

A river basin is the geographical area deter-mined by the watershed limits of a systemof waters, both ground and surface, flowingto a common terminus.

In a few areas, concepts do diverge some-what. Basin management can readily beextended to management of other relatedresources in a basin, especially landresources. IWRM focuses more tightlyon the water resource. At times RBMalso extends into the realm of river basindevelopment, in which case project-baseddevelopment of new basin infrastructureoften acquires a dominant position in theparadigm. Millington (2000) notes that pro-ject design or evaluation, construction andoperation is still a very strong role of

many river basin organizations. Experiencehas shown that where project-based designand construction is a major activity,resource management functions tend toreceive low priority. IWRM grows out of a‘post-construction’ milieu, where basins areclosing (Chapter 2) and new constructiontends not to be the dominant activity.

One important caveat to all this isthat integrated basin management doesnot imply or require a single basin manage-ment organization. There is an unfortunatetendency in some quarters to equate basinmanagement with a unitary basin manage-ment organization and to assume that in theabsence of such an organization, effectiveintegrated management is not possible. Thisis most certainly an incorrect assumption,as experience in the western USA demon-strates, and, in general, monolithic manage-ment organizations are the exception ratherthan the rule. The prototype for suchthinking is probably the American Tennes-see Valley Authority (TVA), which was,in its original form, an integrated basindevelopment authority established duringthe economic depression of the 1930s towork in a very underdeveloped region ofthe country.1 The TVA took on a wide rangeof development functions, including waterresource development, in an area in whichthere was a dearth of effective organizationalcoverage by other public entities. The failure

Managing River Basins 3

Box 1.1. Integrated basin management.

Although the term integrated most commonlyrefers to integration across use sectors, suchas agriculture and urban water supply, it canalso encompass a number of other divisions,including the following:

• Administrative jurisdictions• Ground and surface water• Upstream and downstream reaches• Environmental and human uses• Supply and demand management• Water quantity and quality• Land and water use• Trans-boundary uses

1 The TVA has since evolved into an energy production and management agency.



of most other attempts patterned on theTVA, such as the Damodar Valley Authorityin India in the 1950s, suggests the unique-ness of this model. More common amongeffective basin management set-ups is acoordinated model in which the efforts of anumber of different entities are articulated(see Section 1.3.4).

1.2.2 Institutional arrangements

How water is used in a river basin is deter-mined by the interactions between waterusers, technology and water availability,and hence is a sociotechnical process(Mollinga, 1998). Water management orga-nizations and institutions structure andmediate these interactions, and in turn arereshaped by water use in practice.

The terms ‘institutions’ and ‘organiza-tions’ are often used interchangeably, butit is useful to distinguish between them.In mainstream institutional theory, institu-tions are understood to be ‘the humanlydevised constraints that shape human inter-action’ (North, 1990, p. 3), and consist ofcomplexes of norms, values and behavioursthat persist over time and inform action(Uphoff, 1986). In this view, institutionsprovide structure and regularity to everydaylife by reducing uncertainty and providing aguide to human interaction. They are whatSir V.S. Naipaul calls ‘the contract betweenman and man’. A central tenet of this view isthat institutions work to reduce transactioncosts by reducing the costs of monitoringand responding to the behaviour of others.

Organizations, on the other hand, aredefined as ‘groups of individuals bound bysome common purpose to achieve objec-tives’ (North, 1990, p. 5). Other definitionshighlight the importance of seeing organiza-tions as ‘structures of recognized andaccepted roles’ instead of only groups ofindividuals, yielding a more realistic andaccurate definition (Uphoff, 1986). Organi-zations are created intentionally within anexisting web of institutions. Hence, whattypes of organizations exist and how theyevolve are fundamentally influenced by the

broader network of institutions in whichthey are embedded. Organizations, in turn,influence how institutions evolve overtime. Organizations constitute a subsetof institutions, which are distinguished bytheir purposive origin and maintenance andtheir hierarchically organized roles.

In this book, the combination ofthe institutions and organizations involvedin water management is termed the institu-tional arrangements for water manage-ment. Institutional arrangements for watermanagement thus include the following:

• The established policy and legal envi-ronment (policies, laws, rules, rights,regulations, conventions, and customs,both formal and informal);

• Water management organizations withresponsibilities in water management;

• Processes, mechanisms and proceduresfor decision making, coordination,negotiation and planning.

At this point, a chasm opens, which mustbe carefully negotiated. Policies, rules andregulations, as specified by public authori-ties, may differ substantially from the appli-cation of those rules in practice. Moreover,local rules, such as those governing theallocation of water, for example, maybe quite different from the formal set ofrules promulgated by state authorities. Thestudent of institutional arrangements mustthus be aware of both the formal rules, andthe set of rules-in-use, which operate on theground. The importance of the differencesthat often exist between the two has led oneprominent analyst to define institutionsexplicitly as rules-in-use, rather thansimply as rules (Ostrom, 1992).

To this point, we have treatedinstitutional arrangements as fixed and pre-existing. To animate this static view, itis necessary to consider how institutionsemerge and how changes in them takeplace. Alternative approaches to the studyof institutions, grounded in anthropologyand sociology, argue that institutions are notonly ‘the rules of the game’ or ‘sets of work-ing rules or rules-in-use’ (cf. North, 1990;Ostrom, 1990) but are reproduced, trans-formed and subverted through interactions




and negotiations between actors (Mosse,1997; Cleaver, 1999; Mehta et al., 1999). Thisapproach suggests that institutions emergehistorically from interactions, negotiationsand contests between heterogeneous actorshaving diverse goals, and that they only con-tinue to exist if they are invested in or prac-tised. Thus, institutions cannot be seen apartfrom what people do, and are constantlymade and remade through people’s prac-tices (Mehta et al., 1999). Cleaver suggeststhat, ‘[t]he institutions for the managementof water . . . are socially located and criticallydepend on the maintenance of a number ofgray areas and ambiguity regarding rightsof access, compliance and rules, [and] on acontinuous process of negotiation betweenall users’ (1999, p. 602). Such a notion ofinstitutions opens avenues to analyse howpower pervades institutional arrangementsand gives rise to differentiated access to andcontrol over water, and, more importantly,how to design processes to redressinequities.

Because water is essential to life andlivelihood security and has multiple usesand users, water management readily givesrise to intractable or ‘wicked’ problems,especially where competition for wateris acute. Wicked problems are clusters ofinterrelated problems characterized by highlevels of uncertainty and a diversity of com-peting values and decision stakes (Ackoff,1974; Rittel and Webber, 1973). The set ofproblems constituting a wicked problemcannot be solved in isolation from oneanother and are intractable since what con-stitutes a solution for one group of individu-als entails the generation of a new problemfor another. As wicked problems are charac-terized by competing perceptions andvalues, and often also involve powerdisparities, they enter the realm of politics,understood here broadly as the forum forchoosing among values and the processthrough which relations of power are consti-tuted, negotiated, reproduced or otherwiseshaped (cf. Mollinga, 2001). Water is

frequently a politically contested resource:a contest with unpredictable and unstableoutcomes and diverging pathways to alter-native futures (cf. Mosse, 1997; Mehta, 2000;Mollinga, 2001). Likewise, water manage-ment institutions and policies are frequentlycontested and the outcomes of politicalpractices.

1.2.3 Institutional effectiveness

From the brief discussion of institutionsand organizations above, it is apparent thatstudying effective institutional arrange-ments for water management is conceptu-ally challenging. If institutions differ inprinciple and in practice, are contested,beset with ambiguities and the outcomesof political practices, it follows that whatis defined as ‘effective’ by some will bedeemed ineffective by others. Nonetheless,at an intuitive level, it is clear that thestrong connection between institutions andhow water is managed is indisputable.Heathcote (1998, p. 7) suggests that institu-tional arrangements for water managementmay be considered effective if they:

1. Allow an adequate supply of water thatis sustainable over many years and provideequity in access to this water.2. Maintain water quality at levels thatmeet government standards and othersocietal water quality objectives.3. Allow sustained economic develop-ment over the short and long term.2

Thus, institutional arrangements for RBMare effective if they promote and achievesustainable water management. Sustain-ability can broadly be defined as a condi-tion in which natural and social systemssurvive and thrive together indefinitely(Euston and Gibson, 1995).

To be sustainable, water managementmust protect and restore natural systems,enhance the well-being of people and


2 A fourth condition relating to maintaining ecological systems could be added here, but in our treatment iscovered by an expanded definition of sustainability.



improve economic efficiency. These threeobjectives are often mutually exclusive, asthe partial attainment of one has negativeeffects for attaining the others. This con-tested nature of water makes the institu-tional arrangements for water managementof paramount importance, but also highlyproblematic. If institutions are viewed inmanagerial or interventionist terms, effec-tive institutions are seen as those thatcontribute to attaining sustainable watermanagement by reducing transaction costs,enhancing collective action and increasingcertainty. If a more process-oriented anddynamic view of institutions is adopted,emphasis is placed on how institutions areembedded in power relations, and equity,and not economic efficiency, is a centralconcern.

1.2.4 Policy

Although established policies are includedunder our definition of institutionalarrangements, they are often treated sepa-rately in discussions of ‘policy and institu-tional arrangements’. One reason for thisis that new policies not yet fully imple-mented do not necessarily rise to the levelof ‘institutions’, as they have not demon-strated an ability to persist. Consequently,they are discussed separately here.

Policies provide a direction and suggesta course of action intended to influencedecisions and actions in a particular realmof interest. Water resource policy thusgives overall guidance and direction todecisions and actions that determine theuses, protections and costs of water, andthe subsidies and prohibitions related toits use. In the face of changing conditions,needs, priorities and values, policies andresulting actions must also change. Policiesare important to water resource managementbecause they can serve as important entrypoints into the established cycle of watermanagement practices.

An instrumentalist view of governmentconceptualizes policy as a tool to regulatea population from the top down, throughincentives and sanctions. Shore and Wright(1997, p. 5) summarize the conventionaldefinition of policy as ‘an intrinsically tech-nical, rational, action-oriented instrumentthat decision makers use to solve problemsand affect change’. Although many wouldagree that policies frequently fail to functionas intended, there is a shared understandingthat a good policy is one that adheres tothe standard of rationality contained in theabove definition.

However, conceiving of policy develop-ment and application as an unproblematiclinear process that progresses from formula-tion to implementation to expected out-comes is dangerous, as it obscures howpolicies are produced through decidedlynon-linear and non-rational means, throughpublic and private negotiations, log-rolling,3

political pressure, media manipulation,legal action and other processes involving arange of actors within and outside of govern-ment circles. Moreover, there is seldom asimple progression from policy formulation,to legislation, to framing regulations, to exe-cution. The real process is much less tidy,with iterations, false starts and backtracking,where the lead role alternates betweenpolicy formulation and application.

An alternative perspective on analysinghow policies lead to changes in watermanagement practices conceives of policyformulation and implementation as politicalprocesses in which many interests are atstake. Premised on the notion that watermanagement is a politically contestedterrain (Mosse, 1997; Mollinga, 1998, 2001;Mehta et al., 1999), the ‘policy as process’approach attempts to understand how waterpolicies are ‘produced’ by the interactionsbetween water users, dominant paradigmsand the institutional arrangements thatmediate water control. Through theseinteractions, the content and compositionof policies are redefined and transformed,frequently leading to very different results


3 A process in which legislators trade concessions and support for sometimes unrelated objectives.



from those envisioned. A politicallyinformed analysis of policy processes helpsto understand how policies work in practiceto change control over water and watermanagement, thereby giving insight intowho gains and who loses.

1.2.5 Roles and actors

Individuals and organizations take action,and it is useful to define two other com-monly used descriptive terms. Roles aresets of expectations and tasks associatedwith a particular function (Coward, 1980).As such, roles can be played by individualsor by organizations. Actors are those indi-viduals or organizations who take actionsin a particular context, and thus playroles. Actors can play a number of rolessimultaneously, and roles can be splitamong different actors. Often actors, suchas a government ministry, will play rolesthat relate to water resource managementwhile playing other non-water-related rolesat the same time. Stakeholders are individu-als or groups which have a legitimate inter-est in the management of water resources ina basin but which may or may not play anactive role in basin planning and manage-ment processes. Actors are thus included inthe set of basin stakeholders, but do notcomprise the entire set.

1.3. Basin Management

1.3.1 Context

1.3.1.1 Phases of basin closure

In Chapter 2, Molden et al. posit that riverbasins pass through three phases as morewater is withdrawn by humans (develop-ment, utilization and reallocation), andargue the valid point that that institutionsneed to have the ability to adapt to changes.Keller et al. (1998) proposed a linear three-phase model of river basin maturation,with phases of exploitation, conservationand augmentation. In this model, the final

phase is a search for new water sources –from distant basins, or by desalinatingseawater, rather than reallocation.

Turton and Ohlsson (2000), expandthis general argument positing that waterscarcity per se is not the key issue, butrather whether a society has the adaptivecapacity to cope with the challenges thatwater scarcity poses. They argue that institu-tional transitions need to occur in the watersector as water becomes scarcer, the firstwhen water abundance turns to watershortage and the second when watershortage turns to water over-exploitation.In their argument, the ‘ability to cope’ withshortage is a critical societal attribute deter-mining the ‘pain’ a society will feel as a basincloses.

These models are inductive, attemptingto draw out an explanatory thread from abody of experience. As such, they are heuris-tic rather than predictive in particular cases.Keller et al. (1998) emphasize the economiclogic of the sequence of development. In thisframework, at any point in time the cheapestsolutions are selected, from simple flowdiversion on to desalinization. Molden et al.constrain the impulse to continuouslydevelop new sources of supply and insteadsuggest that attention will shift to reallocat-ing an ultimately fixed supply. In Turtonand Ohlsson’s approach, the logic of thesuccession is based on a scale of complexity,with the solution of water scarcity problemsdemanding ever-increasing levels of socialresources. This approach assumes thathydraulic development is the easiestresponse, and that its exhaustion leadsto conservation efforts, later followed byadjustments in allocations. The latter areregarded as much more sensitive and likelyto generate social conflicts. These threeanalytical grids are useful in makingconnections between the degree of waterexploitation and types of human responses– responses that are clearly related to thedegree of stress on the resource as well asother factors. At the same time, they cannotcapture important nuances found in variedconcrete situations. A number of interestingillustrations of this complexity are shownbelow (Molle, 2003).




• Sakthivadivel and Molden (2001) havecompared five basins said to be at dif-ferent stages of exploitation and havefound that the problems faced by thesebasins were different. However, severalof the problems encountered were notthose that would be typical of the phasein which each basin was classified. Forexample, East Rapti basin in Nepal isan open basin, with only 5% of waterresources being used by agriculture. Inspite of this, water pollution fromindustries and ‘intense competition’for river water during the dry seasonamong wildlife sanctuaries, touristrequirements, ecological requirementsand human use already appear assevere problems that ‘need immediateaction’, although they are normallyassociated with later phases of devel-opment. In the Singkarak-Ombilinbasin, Sumatra, considered to be atthe beginning of the utilization/conser-vation phase, water allocated to non-agricultural activities and trans-basindiversion threatens to throw the basindirectly into the last phase, wherewater rights need to be more formallyspecified and water reallocationbecomes paramount.

• Problems of pollution are generallyassociated with late phases in whichthe scarcity of the resource does notallow adequate mitigation by dilution,but it may also happen very early ifthere are significant pollution pointsources with little regulated water toensure dilution as with gold mines inSouth Africa.

• The need to design more complexand integrated forms of organization atthe basin level is associated with anultimate phase of reallocation of veryscarce water resources. However, in thecase of France in the 1960s, it wasthe problem of water quality and notquantity that was the driving force,despite both aspects being interlinked.

• Trans-basin diversion is considered intrajectory models as a way to ‘reopen’the basin after it has closed, but thisoption is sometimes observed at much

earlier stages of development, espe-cially in small and medium basins.In Sri Lanka, this was commonlyachieved as early as the 5th century(Mendis, 1993) and has remained abasic feature of water resource develop-ment ever since. Such transfers arealso typical of irrigation of mountain-ous interfluves, where irrigated areasstraddle the boundary of two adjacentbasins.

• In the later phases of basin closure, awide range of measures are sometimesundertaken to relieve pressure, and notsimply the reallocation strategy hypo-thesized. The case of California, as des-cribed by Turral (1998), clearly showsnot only that both efficiency and reallo-cative measures are sought in parallel,but that the gains they provide aremore limited than commonly believedand need to be accompanied with asubstantial amount of supply augmen-tation. Closure does not concludewith reallocation but, rather, elicitscontinuous improvements on all threefronts (conservation, reallocation andsupply).

• Not all trajectories are upward. Histori-cal examples of civilizations that havenot successfully maintained theirresource base and have collapsedcan easily be found. In Sri Lanka,for example, aerial photographs reveala high density of abandoned, silted anddestroyed small tanks in some basins.A classic case is that of ancientMesopotamia in the 9th century (cf.Pointing, 1991).

It seems clear that decisions must beunderstood also in terms of their politicaleconomy. That is, decisions must be under-stood not only on the basis of their actualcosts and social ‘pain’, but also in termsof the identity of the beneficiaries andthe increased power or financial gain thataccrues to different actors as a result ofthe decision taken. Costly solutions, suchas desalination, are sometimes justified andimplemented in lieu of less expensivedemand-management solutions because




they involve less political pain and can beaccommodated under the existing logic ofpork-barrel politics.

The difficulty of reforming managementvaries, depending on many cultural, socialand political factors, but it is recognized that‘regional politicians have a powerful intu-ition that economic principles and the allo-cative measures which follow logically fromthem must be avoided at all costs’ (Allan,1999). This largely explains the persistentgap between consultant’s rationality and theactual shape that policy measures take inthe real world. It also suggests why stricteconomic analysis is not always the bestframework to use to understand the succes-sion of state investments and responses.Resource capture can occur at any time,depending on the power balance within thesociety, and is perhaps more frequent thanrational allocation.

The basin closure models are usefulin outlining hydrologic changes which tendto take place in a generalized basin asits water resources are utilized ever moreintensively and the response strategieswhich tend to occupy the minds of basinmanagers as this unfolds. In practice,institutional responses are highly varied andfunctions of a diverse range of forces andinfluences.

1.3.1.2 The governance context

One of the most important, and least stud-ied, aspects of the environment controllinginstitutional change is the context of gover-nance in which water resource organiza-tions operate. Governance is defined asthe exercise of authority through formaland informal traditions and institutions forthe common good. Governance includes:(i) the process by which those in authorityare selected, monitored and replaced;(ii) the capacity of the government to effec-tively manage its resources and implementsound policies; and (iii) the respect ofcitizens and the State for the institutionsthat govern economic and social inter-actions among them (Kaufmann, 2000). InKaufmann’s framework, good governanceconsists of six interlinked components:

• Voice and accountability;• Political stability;• Government effectiveness;• Lack of regulatory burden;• Rule of law;• Control of corruption.

Problems stemming from poor governanceare numerous, well known and routinelygiven a blind eye. Examples includefavouritism in granting water use permits,kickbacks on construction and procurementcontracts, biased and inaccessible courtsystems, withholding of data and sale ofpublic data for personal profit, promotionof risk-averse bureaucrats and firingof innovators, flaunting of water-qualityregulations by well-connected industries,and bureaucratic red tape which strangleslocal initiative. Quality of governancepervades public decision making relatingto policy formulation, resource allocation,legislation, rule enforcement and adjudica-tion, making it the most importantsingle influence on the shape and paceof institutional change in the water sector.While difficult to change, improving thequality of governance is not impossible,and a variety of tools have recently beendeveloped to support such change, manyinvolving voice, transparency, informationand participation. Several specific ele-ments, which can support improvedgovernance and promote institutionalchange, are discussed in a subsequentsection.

1.3.2 Functions and actors

To analyse the institutional arrangementsfor water management in a river basinwe propose a framework of essentialfunctions and enabling conditions. Thegroundwork for this framework is providedby drawing up a water account of a basinas well as a basin profile that provides ananalytically rich description of the basin asa sociotechnical system. The next step is toidentify the water management actors in ariver basin and the essential functions theyexecute.




1.3.2.1 Actors

To portray the multitude of water manage-ment stakeholders in a river basin, it is use-ful to distinguish between the various wateruse sectors and the types of organizationsinvolved in water resource management.Combining sectors with actors yields amatrix of key organizations and stake-holders involved in water managementin a basin (Table 1.1). This matrix, afterindividual actors are identified under eachcategory, provides a basis for identifyingkey actors in a particular basin.

1.3.2.2 Essential functions

To analyse basin governance, it is necessaryto focus on the roles and functions of the

various actors engaged in water manage-ment in the basin, asking who does what,where, to what end and how well. Toguide the analysis, a set of essentialfunctions for RBM has been identified(Burton, 1999; Svendsen et al., 2001). Theseare defined in Table 1.2. How well func-tions are carried out, from whose perspec-tive and for whose benefit are empiricalquestions.

It is possible to construct alternativelists of basin functions. This one wasempirically and inductively developedfrom experience in a number of basins.Its functions subsume supporting functionssuch as data collection and resourcemobilization, which are not ends in them-selves, but rather facilitate the higher-levelfunctions listed.


Sector

Stakeholders Agriculture Domestic Industry Hydropower Environment Other

Multinational agenciesGovernment agenciesPrivate firmsAssociations/NGOsInformal groups

Note that government agencies may include national, sub-national and local entities, while private firmscan include both multinationals and local firms as well as regulated for-profit utilities. The list of sectors isfar from exhaustive and could also include fisheries, navigation, recreation, amenity value and others,depending on local importance.

Table 1.1. Illustrative matrix of key water management stakeholders.

Function Definition

1. Plan

2. Allocate water

3. Distribute water4. Monitor water quality5. Enforce water quality

6. Protect against waterdisasters

7. Protect ecology8. Construct facilities9. Maintain facilities

The formulation of medium to long-term plans for the management anddevelopment of water resources in the basin, by which the water demands ofdifferent sectors are brought in line with water supply.The mechanisms and criteria by which bulk water is apportioned among thedifferent use sectors.The activities executed to ensure that allocated water reaches its point of use.The activities executed to monitor water pollution and salinity levels.The activities executed to ensure that water pollution and salinity levelsremain below accepted standards.Activities executed concerning flood and drought warning, prevention offloods, emergency works and drought preparedness.Actions undertaken to protect associated ecosystems.Activities executed for the design and construction of hydraulic infrastructure.Activities executed to maintain the serviceability of the hydraulic infrastructurein the basin.

Table 1.2. Essential functions for river basin management.



1.3.2.3 Interactional analysis

Functions that should be performed in aclosing water basin for effective manage-ment may or may not be performed, infact. Moreover, some may be performedincompletely and some may even receivemore attention than they require. Sincemany organizations and stakeholders areinvolved with water management in ariver basin, a number that generally growswith closure, more than one organizationwill frequently be involved in performinga particular function. To structure andclarify the resulting patterns of activity,the essential functions of basin manage-ment can be crossed with key water man-agement actors as illustrated in Table 1.3.The essential functions are replicated, asappropriate, across three broad categories –surface water, groundwater and derivativewater. Because the perspective is basin-wide, functions are not separated by sector,e.g. irrigation or environment. Thus, thecategory derivative water includes irriga-tion return flows as well as municipalwastewater and industrial discharges. Cellsin the resulting matrix are coded to showwhether the actor is judged to play a majoror a minor role. One of the case studiesin this volume (Chapter 9) adds anotherdimension to the table and endeavoursto indicate the type of activity performedby an actor in addressing a particularfunction.

It is important to note that the matrixdepicts actual activity in practice and notnominal responsibilities according to legalframeworks or normative prescriptionsdefining what should be done. The matrix,together with description of the key actors,gives an indication of which essentialfunctions are being addressed and who isinvolved in their execution. The matrixexercise can shed light on a number ofimportant questions:

• The functions covered and a roughindication of the adequacy of coverage;

• The functions not covered;• The number of actors involved in each

function and the need for coordination;

• The stakeholders represented in per-forming particular functions, whichleads to conclusions about the repre-sentativeness of the basin governance.

Perhaps more importantly, the process isheuristic in that it produces insights andquestions that can be used to probe moredeeply into issues of functional perfor-mance, relationships among actors, and thepolitical dynamics of basin governance andmanagement.

The actors/functions matrix can alsobe used comparatively within a basin toexamine changes over time, the nature of atransition to a desired future state, or nomi-nal versus actual functional performance.Furthermore, it can be applied compara-tively to look for patterns among differentbasins and national contexts. This is done inChapter 13 of this volume for several of thecase studies presented.

The matrix can be generated in differentways. It can be filled in by expert observersafter study of a basin, as in the Turkey casestudy. It can be created on the basis ofquestionnaire survey results combined withexpert observation as was done in theMexico case, or it can result from focusgroup interactions of knowledgeable per-sons, as happened for the South Africancase. Although used here as a research tool,matrix generation can also be a useful under-standing and consensus building tool amongthe involved parties when used in focusgroups made up of key basin stakeholders.

1.3.3 Enabling conditions

Describing who executes the essential func-tions in a river basin and how effectively,while useful, does not constitute a suffi-cient methodology for understanding anddiagnosing problems affecting basin gover-nance. The essential functions and actors’roles depicted in Table 1.3 provide a staticview of responsibilities. Additional attrib-utes of well-functioning basin governancesystems relate to its dynamics.

In order for societies to reach decisionsconsistent with the public interest, several




conditions need to be satisfied. We termthese attributes enabling conditions (Box1.2). Enabling conditions are features of theinstitutional environment at the basin levelthat must be present, in some measure, toachieve good governance and managementof the basin. These attributes are not specificto any one actor, but apply to all actors andtheir interactions and comprise necessary(but not sufficient) normative conditionsfor good basin management. Most of themcontribute to good governance, as discussedin an earlier section. Some basic enablingconditions are shown in Box 1.2. A thoroughanalysis of these factors is well beyond thescope of this chapter, but a number of themare described briefly.

1.3.3.1 Political attributes

An important political attribute is the repre-sentation of interests. In most river basins,some water users will be well represented,while others will not, and in the arena ofpolitical give-and-take, those without repre-sentation become losers. Industrialists andcommercial farmers, for example, typicallyhave ample financial resources, are wellorganized and have ready access to politicaldecision makers. Poorer irrigators, onthe other hand, are likely to be less wellorganized and consequently will be weaklyrepresented. Their interests are often ratherfragile. Water users associations (WUAs)

are likely to be intermediate, particularlyif they are connected to the local politicalestablishment and collaborate informally,sharing information and coordinating activ-ities. Many times WUAs would benefit byestablishing more formal linkages amongthemselves to allow a single spokespersonto represent them collectively in discussionsover basin water allocation, water qualitystandards, potential irrigation return flowrestrictions, and so on.

A serious failure of representationwill frequently exist for the environment.Experience has shown that strong non-governmental organizations (NGOs) rootedin civil society are essential components of apolitical system making socially responsiblechoices about environmental issues. NGOscan serve as advocates for environmentalvalues and for unrepresented future generat-ions. At the same time, fund raisingrequirements may lead such mass-basedorganizations to take extreme and uncom-promising positions on issues that mustthen be moderated in the give-and-take ofpolitical debate and decision making.

Hence, fully as important as the exis-tence of representational bodies is the needfor a rough balance of political power andinfluence among various interests. Whenpower is one-sided, issues are not airedadequately and decisions are also one-sided.A key to the evolution of a suitable andbalanced governance regime is maturationof non-government organizations and asso-ciations based in civil society, which canadvocate for particular interests, coupledwith the informational attributes describedbelow.

1.3.3.2 Informational attributes

An essential enabling condition is thepresence in the public domain of accurateand up-to-date descriptive information onwater-related issues in the basin. Anotheris open public transactions, related topolicies, plans, regulations, violations andsanctions. The first of these stipulationsrequire that information on basin waterallocations, reservoir positions, ground-water elevations, water-quality conditions,


Box 1.2. Enabling conditions.

Political attributesRepresentation of interestsBalanced power

Informational attributesProcess transparencyInformation availabilityInformation accessibility

Legal authorityAppropriate institutionsAdequate powers

ResourcesHumanFinancialInstitutionalInfrastructural

A4859 - Svendsen - Vouchers Proofs #R pg 013.prnZ:\Customer\CABI\A4831 - svendsen\A4859 - Svendsen - Vouchers Proofs #R.vpMonday, January 24, 2005 2:46:41 PM

Color profile: Generic CMYK printer profileComposite Default screen

available resources, and so on be a partof the public record. Information collectedwith public funds should be availableto the general public at little or no chargein the interest of sound and democraticpublic decision making. This disclosurecondition applies to intra- and inter-departmental information relationships,as well as to those with the general pub-lic. The second stipulation, transparencyof public proceedings, is similarly essentialto fair democratic processes. Rent-seekingbehaviour requires darkness and privacyto thrive, and conducting regulatoryprocesses in full view of the public andthe press is an effective antidote to suchpractices.

1.3.3.3 Legal authority

Establishing appropriate organizationsrequires suitable legal authority. Thisauthority includes the right to exist, theright to a legal personality and suitable elec-toral procedures to ensure representativeleadership of the organization. A legalpersonality usually includes the right tohandle money and keep a bank account,enter into contracts, access the legal systemand represent the membership in dealingswith governmental agencies.

In some cases, existing legislation hasbeen adapted to allow new water-relatedorganizations to be established. The forma-tion of Irrigation Associations in Turkeytook this route. In other cases such asMexico, new legislation has been written atthe outset to facilitate establishment of neworganizations and relationships.

1.3.3.4 Resources

Clearly, all four types of resources listedin Box 1.2 are needed for effective imple-mentation of basin management activities.A potential problem is scattering of humanand financial resources among a number oforganizations, where each lacks a criticalmass to be effective. In a context of coopera-tion, it is not necessary that resources beconsolidated under a single administrativestructure for effective implementation.

However, cooperation and coordinationmust be effective if a decentralized strategyis to work.

1.3.4 Organizational configurations

The choice of a river basin as a unit of man-agement is based on a certain hydrologicimperative, controlled by gravity and topog-raphy. However, establishing basin bound-aries is by no means automatic. Thereare choices involved in subdividing largebasins into management units, in groupingsmall basins, and in deciding which naturalbasins and sub-basins are to receive priorityattention. Moreover, there are often differ-ences between surface water and ground-water divides and basins, where choiceshave to be made, and where water isimported from neighbouring basins, contro-versies have arisen over whether to includethe watershed of the transferred water inthe basin definition. Making such choiceshas an important bearing on basin manage-ment, as different boundaries imply differ-ent decision makers and possibly differentdecision outcomes. However, whetherdefining boundaries is straightforward orcontentious, the defined basin becomes apolitical unit as well as a hydrologic oneand questions immediately arise as to whowill make decisions, and how (cf. Westerand Warner, 2002).

Mostert et al. (2000) posit three differenttypes of organizational configurations forbasin management. One is an authoritarianmodel, in which management is organizedon hydrologic boundaries and a single orga-nization makes basin decisions. The secondis a coordinative model, in which the basinas a hydrologic unit is recognized, but manyfunctions remain in the hands of traditionalgovernmental units, and work is coordi-nated among these units. The third is man-agement by existing organizations withoutcoordination. The third model is, in reality,not a model of basin management at allbut rather the business-as-usual backdropagainst which the two other models can becontrasted.




We distinguish two basic organizationalpatterns for basin governance.4 The first isthe centralized (unicentric) model, in whicha single unified public organization isempowered to make decisions regardingmanagement of the basin. This centralizedorganization is not necessarily ‘authoritar-ian’, but does centralize authority under agovernance process that may be more or lessdemocratic. The second is the decentralized(polycentric, coordinative) model, in whichthe actions of existing organizations, layersof government and initiatives are coordi-nated to cover an entire river basin or sub-basin. While new structures may be created,the bulk of routine work is done by existingorganizations that are not specific to thebasin. Although both models are character-ized by separations among the three basicroles of management, regulation and serviceprovision, the firewalls between them aretypically stronger in the coordinative modelwhere separate organizations are involved.

In the real world, RBM structures areusually hybrids, relying to some extent onexisting government structures to providepolicy and direction, and perhaps executeparticular management functions, andbasin-specific organizations to collect data,and make certain circumscribed decisions.We describe the two hypothetical modelsbriefly to illustrate the two poles of thecontinuum.

1.3.4.1 Centralized (unicentric) RBMs

A strength of the centralized model is thatits operational span of control coincideswith the boundaries of the basin. Thisinternalizes upstream/downstream andother conflicts, making them easier to dealwith, and it concentrates the decision-making authority needed to resolvedisagreements. Disadvantages of the cen-tralized model include the following: (i) asthe organization will generally deal onlywith water, water will be isolated fromother relevant policy sectors such as

agriculture, environment and the economy;(ii) establishing a strong unified centralauthority presents a more challengingpolitical problem than securing agreementfor a coordinative body. The challenge herewill be even greater for international riverbasins; and (iii) governance of a centralizedorganization raises challenging questionsof broad stakeholder representation andaccountability.

The most prominent examples ofauthorities are those having the develop-ment of a river basin as a primary mandate.The classic example is the TVA, which wascreated during the economic depression ofthe 1930s to address problems of povertyand unemployment in a particular region ofthe USA. Other examples are the Rio SãoFrancisco Development Agency in Braziland the Mahaweli Development Authorityin Sri Lanka. When their primary develop-ment tasks are finished, these authoritiesoften try, with varying degrees of success, toassume a broader resource management role.

1.3.4.2 Decentralized (polycentric) RBMs

The decentralized model addresses someof the weaknesses of the centralized modelbut contains others. On the plus side, it pro-vides for a strong political base for action,since coordination involves voluntaryagreement among participating jurisdic-tions. The coordinative process also leadsto a more responsive governance process.Intersectoral linkages remain intact, ascoordination is among individual states,nations or other jurisdictions responsiblefor a range of policy sectors, and such aset-up provides a natural base for decentral-ization of responsibilities. On the otherhand, decision making can be cumber-some, coordinating costs may be high andpolitical changes in participating jurisdic-tions can upset agreements.

These two models represent polarextremes, and actual arrangements are oftenblends of the two. In the Murray–Darling


4 We focus here on organizations and governance and not on the way in which particular functions areexecuted. There are a range of options within each governance structure for providing services and executingother functions.



basin, for example, a cooperative MinisterialCouncil, made up of representatives of thefour involved states and the federal govern-ment, sets policy while under it an author-ity-like Commission supports the Counciland executes its decisions. A similar set-upexists in France, where a River Commissionmade up of local and national governmentrepresentatives and users sets water policy,which is implemented by an associatedWater Agency. Publicly held companiesmanage the distribution infrastructureand make bulk water deliveries to userassociations. In the USA, formal bodies formanaging river basins are rare, allowingsome exceptions such as the TVA and theDelaware Basin Commission. Policy-makingauthority is distributed among a variety offederal and state agencies and departments.Coordination is achieved through a plethoraof committees and working groups linkingstakeholders into discussion and decision-making forums. Legislation and negotiatedlegally binding agreements are importantinstruments for establishing policy andpractices, and the court system is routinelyinvoked to resolve disagreements and dis-putes. In California, a state water plan,updated every 5 years, provides a rollingframework for managing the state’s waterresources.

1.4. Basin Management and Irrigation

As basins close, irrigation systems withinthe basin are confronted with both internaland external challenges. Internal challengesrequire them to do more with less water,whereas the external ones require them toorganize and act effectively to protect theirinterests. Dealing with both at the sametime is difficult, and systems which addressthe internal challenges successfully beforehaving to tackle the external ones willgenerally be better off.

Closing basins, by definition, arebecoming water scarce, and newer rapidlygrowing sectors, typically urban and indus-trial users and the environment, will usuallydemand that irrigated agriculture, as the

largest traditional user, use less water tofree more of it for their growing needs. Thisputs pressure on irrigators to use water moreefficiently, and may lead to retirement ofless productive irrigated lands and transferof their water rights to other users, as ispresently happening in California’s CentralValley. The cost of water will generallyincrease to reflect its growing scarcity, lead-ing to pressure to grow higher-value crops tocover these costs. More efficient use of waterrequires that irrigation systems acquire newmeasurement and control technology, andmore professional management.

These same pressures may emerge aschallenges to agriculture’s right to use basinwater resources at all in legislative and legalarenas. Often agriculture began using waterat a time when rights were not formally spec-ified, which can make them less secure thanthe formalized rights allocated to industriesand larger corporate irrigators which camelater. Where water rights are merely implicitin the allocation priorities of a large publicirrigation agency, risks also arise. Regula-tion, service provision and other functionsof unitary public water agencies tend tobe split up among several new agenciesor departments as basins mature, givingbasin managers a broader constituencyand weakening their ability to defend theirformer clients. Basin-level decision-makingforums will tend to include more actors andcover a broader range of issues, requiringthat irrigators mobilize to represent theirown interests vigorously and become con-versant with a broader range of water-relatedconsiderations.

Irrigators located downstream of majorpopulation centres and industries will alsoneed to seek protection for the quality ofthe water they receive. Urban concentra-tions with inland locations, such as Cairo orNew Delhi, degrade significantly the qualityof water reaching downstream irrigators,with impacts on human health and contami-nation of produce and soil. Individually,irrigators and small systems will have littleor no ability to apply pressure for reducedpollutant loadings in their water supplies.Organized into a larger network, they canhave influence.




References

Ackoff, R.L. (1974) Redesigning the Future: aSystems Approach to Societal Problems.John Wiley and Sons, New York.

Allan, J.A. (1999) Productive efficiency andallocative efficiency: why better watermanagement may not solve the problem.Agricultural Water Management 40, 71–75.

Burton, M. (1999) Note on proposed frameworkand activities. Prepared for the IWMI/DSI/CEVMER Research Programme on Institu-tional Support Systems for SustainableManagement of Irrigation in Water-ShortBasins, Izmir.

Cleaver, F. (1999) Paradoxes of participation:questioning participatory approaches todevelopment. Journal of InternationalDevelopment 11, 597–612.

Coward Jr, E.W. (1980) Irrigation development:institutional and organizational issues.In: Coward, W.W. Jr (ed.) Irrigation andAgricultural Development in Asia. CornellUniversity Press, Ithaca, New York,pp. 15–27.

Euston, S.R. and Gibson, W.E. (1995) The ethic ofsustainability. Earth Ethics 6, 5–7.

Heathcote, I.W. (1998) Integrated WatershedManagement. Principles and Practice. JohnWiley and Sons, New York.

Kaufmann, D. (2000) Governance and anti-corruption. In: Thomas, V. (ed.) The Qualityof Growth. Oxford University Press, Oxford,pp. 135–168.

Keller, J., Keller, A. and Davids, G. (1998) Riverbasin development phases and implicationsof closure. Journal of Applied IrrigationScience 33, 145–163.

Mehta, L. (2000) Water for the twenty-firstcentury: challenges and misconceptions. IDSWorking Paper 111. IDS, Brighton.

Mehta, L., Leach, M., Newell, P., Scoones, I.,Sivaramakrishnan, K. and Way, S. (1999)Exploring understandings of institutionsand uncertainty: new directions in naturalresources management. IDS DiscussionPaper 372. IDS, Brighton.

Mendis, D.L.O. (1993) Irrigation systems in theWalawe Ganga Basin: an historical overview.In: The South-east Dry Zone of Sri Lanka,Proceedings of a Symposium, 29–30 April1992. ARTI, Colombo, Sri Lanka, pp. 37–54.

Millington, P. (2000) River basin management;its role in major water infrastructureprojects. Prepared as an input to the WorldCommission on Dams, Cape Town.

Molle, F. (2003) Development Trajectories of RiverBasins: a Conceptual Framework. ResearchReport No. 72. International Water Manage-ment Institute, Colombo.

Mollinga, P. (1998) On the waterfront. Waterdistribution, technology and agrarianchange in a South Indian canal irrigationsystem. PhD thesis, Wageningen AgriculturalUniversity, Wageningen.

Mollinga, P. (2001) Water and politics: levels,rational choice and South Indian canalirrigation. Futures 33, 733–752.

Mosse, D. (1997) The symbolic making of acommon property resource: history, ecologyand locality in a tank-irrigated landscape inSouth-India. Development and Change 28,467–504.

Mostert, E.N.W.M., Bouman, E., Savenije, H.H.G.and Thissen, W.A.H. (2000) River basinmanagement and planning. In: River BasinManagement, Proceedings of the Interna-tional Workshop, The Hague, 27–29 October1999. UNESCO, Paris.

North, D. (1990) Institutions, Institutional Changeand Economic Performance. CambridgeUniversity Press, Cambridge.

Ostrom, E. (1990) Governing the Commons: theEvolution of Institutions for Collective Action.Cambridge University Press, New York.

Ostrom, E. (1992) Crafting Institutions for Self-governing Irrigation Systems. Institute forContemporary Studies Press, San Francisco.

Pointing, C. (1991) A Green History of the World.Penguin Books, Harmondsworth.

Rittel, H.W.J. and Webber, M.M. (1973) Dilemmasin a general theory of planning. PolicySciences 4, 155–169.

Sakthivadivel, R. and Molden, D. (2001) Linkingwater accounting analysis to institutions:synthesis of five country studies. Paperpresented at the Regional Workshop onIntegrated Water Resources Management in aRiver Basin Context: Institutional Strategiesfor Improving the Productivity of Agricul-tural Water Management, Malang, Indonesia,15–19 January 2001. International WaterManagement Institute (IWMI).

Shore, C. and Wright, S. (1997) Policy: a newfield of anthropology. In: Shore, C. andWright, S. (eds) Anthropology of Policy:Critical Perspectives on Governance andPower. Routledge, London, pp. 3–39.

Svendsen, M., Murray-Rust, H., Harmancioglu, N.and Alpaslan, N. (2001) Governing closingbasins: the case of the Gediz River in Turkey.In: Abernethy, C. (ed.) Intersectoral Manage-ment of River Basin. International Water




Management Institute (IWMI), Colombo, SriLanka, pp. 183–214.

Technical Advisory Committee (2000) Integratedwater resources management. TAC back-ground papers no 4. Global WaterPartnership, Stockholm.

Turral, H. (1998) Hydro-logic?: Reform in WaterResources Management in DevelopedCountries with Major Agricultural WaterUse – Lessons for Developing Nations.Overseas Development Institute, London.

Turton, A.R. and Ohlsson, L. (2000) Water scarcityand social stability: towards a deeper under-standing of the key concepts neededto manage water scarcity in developingcountries. SOAS Water Issues Study Group,

Occasional Paper 17. SOAS, University ofLondon, London.

Uphoff, N. (1986) Local Institutional Develop-ment: An Analytical Sourcebook withCases. Kumarian Press, West Hartford,Connecticut.

Wester, P. and Warner, J. (2002) River basinmanagement reconsidered. In: Turton, A.and Henwood, R. (eds) Hydropolitics inthe Developing World. A SouthernAfrican Perspective. AWIRU, Pretoria,pp. 61–71.

Wilson, H.M. (1891) Irrigation in India. In: 12thAnnual Report of the United States Geologi-cal Survey, Part 2. USGS, Washington, DC,pp. 363–562.




2 Phases of River Basin Development:the Need for Adaptive Institutions

David Molden, R. Sakthivadivel, M. Samad and Martin Burton

2.1 Introduction

The last 50 years have witnessed enormouschanges in the way people use waterfor productive and economic purposes.Between 1950 and 2000, the world’s popu-lation increased from 2.5 to 6.0 billion (109).Concurrently, there was a tenfold increasein the number of large dams, from 4270dams to over 47,000 (ICOLD, 1998). Waterwithdrawals have jumped from 1360to 4000 km3 (Shiklomanov, 2000). Agri-cultural withdrawals have increased from1080 to 2600 km3 (Shiklomanov, 2000),whereas cultivated land expansion hasincreased at a much slower rate from 1063million ha to 1360 million ha. This clearlydemonstrates the important role additionalwater has played in agricultural produc-tion. Water resources development has notonly led to significant changes in land andwater systems, but also in the institutionsrequired to manage these resources.

Our working hypothesis is that chang-ing patterns of water use require adaptiveinstitutions for sustainable, equitable andproductive management of the waterresource basin wide. Institutional arrange-ments that are able to manage the presentsituation adequately are not likely to meetfuture needs unless they can adapt to changerapidly. This chapter presents hydraulic andinstitutional arguments to develop the con-cept of phases of river development. It uses

this concept to show why adaptive institu-tions are necessary, what issues may ariseduring various phases and what the keyareas of institutional focus are as river basinsdevelop.

2.2 Hydraulic and Hydrologic Responsesto Water Resources Development

To understand how institutions can moreeffectively manage water, it is essentialto understand how the physical waterresource base responds to interventions thatserve human needs. River basins provide alogical unit of analysis because water flowsstarting from precipitation can be tracedto understand the impact of these inter-ventions. A water accounting framework isused to understand changes of water usewithin a basin context (Boxes 2.1 and 2.2).

With the development of hydraulicinfrastructure, whether a large dam, water-harvesting structure or pump, the aim isto redirect water from its natural course tohousehold, agricultural, industrial or othercultural use. Similarly, changes in land-scapes such as from forests to farmland alterthe natural hydrologic condition and redi-rect water flow paths. Some uses depletewater – they render it unavailable for down-stream use – either by conversion of water toevaporation, or by the direction of flows intoa sink such as a sea or saline aquifer. Flowing




water can serve several purposes – hydro-power, in-stream fisheries and agriculture –and create considerable value before it isfinally depleted.

Relative to other uses, food productiondepletes huge quantities of water. To feedone person requires that between 2000 and5000 l of liquid water each day be convertedto vapour through the biophysical process ofevapotranspiration. In contrast, drinkingwater needs range from 2 to 5 l, and house-hold requirements range from 20 to 500 l perperson per day. As a result, food productionfor a growing population has significantlyaltered the hydrologic characteristics ofmany basins. Irrigation depletes waterthrough either evapotranspiration or direct-ing drainage flows to the sea. Rainfed agri-culture depletes water that would havesupported other ecosystems. Thus, there isa clear need to focus on water use in agricul-ture as it is a major factor in water resourceand land management.

2.3 Phases of River Basin Development

Changes in water use can be illustrated byconceptualizing phases of river basin devel-opment (see Keller et al., 1998; Ohlsson andTurton, 2000; Allan 2002; Molle, 2003, forsimilar discussions). In its most basic form,water scarcity is a situation where peoplehave difficulties accessing water for drink-ing or growing crops. As a reaction to waterscarcity, people tap basin water resources.Hydraulic structures, ranging from simplestone and wood diversion structures tocomplex dam and canal systems supplywater from streamflow for drinking andindustrial supplies, and for agriculture.Rainfed agriculture converts land use fromits previous cover (forest, grassland) tocropland, impacting the previous hydro-logic regime and ecosystems dependenton that regime.

At any particular time, the availablewater supply is limited by the installed

20 D. Molden et al.

Box 2.1. Definitions.

• Water balance and water accounting: Water accounting relies on water balance studies for a domainbounded in space and by time. For example, we may include a portion of a basin over a year’s time,bounded spatially so that runoff is captured by the sub-basin, and vertically to include the bottom of theaquifer up to the top of the vegetation canopy. A water balance analysis quantifies water flows acrossthe boundaries including rain, evaporation, surface and subsurface inflows and outflow. Changes instorage internal to the water balance domain, such as changes in reservoir levels or groundwater levelsmust be considered. Essentially, water accounting divides hydrological variables of discharge, rain andevaporation into water accounting categories of process (i.e. crop evapotranspiration, domestic andindustrial depletion), and non-process (evaporation and and evapotranspiration from non-agriculturalland cover, committed and uncommitted outflow, and utilizable and non-utilizable outflow).

• Diversions, depletions and recycling: Water is diverted to various uses. Water is depleted when it isrendered unavailable for further downstream use – either through evaporation or by directing the waterto sinks. Since not all water diverted to a use is depleted, some remains within the basin and is availablefor further use. Water recycling or reuse is prevalent in water resource systems. City effluents dischargedback into river systems are often used again downstream. It is common to underestimate how muchreuse exists in river systems, especially in those that are highly stressed.

• Accounting for precipitation: In many analyses of water resources, only the ‘developed’ or ‘blue’ watersupply is considered – supplies tapped from rivers by diversion structures. In IWMI’s water accountingframework, rain is considered a supply.

• Water commitments: All uses of water in a basin could be captured if the boundaries of a basin weredefined to extend to an ideal salt–freshwater interface. Most often it is practical and useful to consideronly part of a basin, but when we do this we have to make sure and define commitments of water todownstream uses to meet ecological or other human requirements downstream.

• Open and closed basins: When all available water has been allocated to various uses, we consider thebasin closed. When there is water remaining in the basin to develop and allocate, we say the basin isopen. In many basins, there is ample water during part of a year, and at other parts it is dry. We considerthese basins to be seasonally closed.



hydraulic structures and land placed undercultivation. When demand exceeds thisavailable supply, one response is to providemore supply either by expanding hydraulicinfrastructure or by expanding rainfed agri-culture. This supply approach is ultimatelylimited by the amount of land and waterresources within a basin, the technical andeconomic limits we have in abstracting thissupply (it would be difficult to divert theentire Amazon), ecological thresholdsbeyond which ecosystems cannot sustainland and water use practices, and societaldemands. The latter changes over time dueto the state of a country’s social developmentand economy.

Figure 2.1 represents a typical progres-sion of river basin development over timewith the original runoff and rainwatersources shown on the y-axis, and time on thex-axis. Over time, more water is made avail-able for human uses from stream flow orgroundwater by building structures (dams,diversions, groundwater pumps) yielding astair-step pattern. Larger dam or diversionstructures would yield a sudden jump in theamount of water made available. Smallerstructures like pumps or water-harvestingdevices also add more available water but inless dramatic steps. After a new dam is built,it takes time to deplete all available water byconverting it into evaporation. Populations

Phases of River Basin Development 21

Box 2.2. Water accounting categories.

• Gross inflow is the total amount of water flowing into the water balance domain from precipitation,surface, and subsurface sources.

• Net inflow is the gross inflow plus any changes in storage.• Water depletion is a use or removal of water from a water basin that renders it unavailable for further

use. Water depletion is a key concept for water accounting because of the interest in increasing thederived benefits per unit of water depleted. It is extremely important to distinguish water depletion fromwater diverted to a service or use as not all water diverted to a use is depleted. Water is depleted by fourgeneric processes:

• Evaporation: water is vaporized from surfaces or transpired by plants.• Flows to sinks: water flows into a sea, saline groundwater, or other location where it is not readily or

economically recovered for reuse.• Pollution: water quality gets degraded to an extent that it is unfit for certain uses.• Incorporation into a product: through an industrial, or agricultural process such as bottling water, or

incorporation of irrigation water into plant tissues.

• Process consumption is that amount of water diverted and depleted to produce a human intendedproduct.

• Non-process depletion occurs when water is depleted, but not by the process for which it was intended.Non-process depletion can be either beneficial or non-beneficial.

• Committed water is that part of outflow from the water balance domain that is required by other usessuch as downstream environmental requirements or downstream water rights.

• Uncommitted outflow is water that is not depleted or committed and is therefore available for a usewithin the domain, but flows out of the basin due to lack of storage or sufficient operational measures.Uncommitted outflow can be classified as utilizable or non-utilizable. Outflow is utilizable if byimproved management of existing facilities it could be consumptively used. Non-utilizable uncommit-ted outflow exists when the facilities are not sufficient to capture the otherwise utilizable outflow.

• Available water is the net inflow minus both the amount of water set aside for committed uses and thenon-utilizable uncommitted outflow. It represents the amount of water available for use at the basin,service or use levels. Available water includes process and non-process depletion, plus utilizableoutflows.

• A closed basin is one where all available water is depleted. An open basin is one where there is stillsome uncommitted utilizable outflow.

• In a fully committed basin, there are no uncommitted outflows. All inflowing water is committed tovarious uses.



grow, wealth rises and demand increases,until depletion reaches the available supply,when possibly another structure is built.Similarly, conversion of land to rainfedagriculture yields more water (directly fromrain) for agriculture. Rainfed expansion con-tinues until a limit is reached. Figure 2.1graphically illustrates three importantphases of river basin development implicitin the above discussion. The progressioncontinues until the threshold is reached.The three phases are:

1. Development. In this phase, the amountof naturally occurring water is not a con-straint. Rather, expansion in demand drivesthe construction of new infrastructure andexpansion of agricultural land. Institutionsare primarily engaged in expanding facili-ties for human use.2. Utilization. Significant constructionhas taken place, and now the goal is to makethe most out of these facilities. Water savingsand improved management of water deliver-ies are important objectives. A commonphenomenon during this phase is the growthin the amount of reuse through drains,

downstream diversions or groundwaterpumps, which ultimately deplete moreof the available supplies. Institutions tendto be concerned with sectoral issues suchas managing irrigation water or managingdrinking water supplies. The creation of theInternational Irrigation Management Insti-tute, for instance, was in response to theperceived need to improve management ofirrigation systems, especially new ones thatwere constructed in the 1950s to the 1980s.3. Reallocation. When depletionapproaches the potential utilizable water,there is limited scope for further develop-ment. We refer to this as a ‘closed basin’(Seckler, 1996). Efforts are placed on increas-ing the productivity or value of every drop ofwater. An important means of accomplish-ing this is to reallocate water from lower to‘higher-value’ uses. Reallocation of waterto achieve both sustainability and equityamong competing demands becomes a majorissue. Institutions are primarily involved inreallocation, conflict resolution and regula-tion. When resource depletion outstrips sus-tainable limits, environmental restorationbecomes an important agenda item.

22 D. Molden et al.

Fig. 2.1. Phases of river basin development.



Figure 2.1 illustrates changing depletion ofwater for human needs over time. Precipita-tion, the ultimate source, either contributesto runoff or groundwater, or is availablefor transpiration or evaporation providingwater for life support in landscapes (rainfedsources). Over time, people tend to depletemore water by making more water availablethrough hydraulic structures (illustrated bythe stair-step patterns) and using water foragriculture, municipal or industrial uses asillustrated by the upper curve. On land-scapes, evaporative water use changes bychanging landscapes such as expandingrainfed agriculture as illustrated by thelower curve. The dashed lines representthresholds, utilizable amounts that cannotbe exceeded over long periods. As humandepletion increases, river basins passthrough phases of development, utilizationand reallocation (Box 2.1).

In many closed or closing basins,supply side responses prevail, and waterdepletion exceeds the potentially availableresource – in the long run, a non-sustainablesituation. For example, in the North ChinaPlains (CCAP and CAAS, 2000; Kendy et al.,2003) and in northwest India (Molden et al.,2001), intensive groundwater use is lower-ing aquifers to unsustainable levels. Asanother example, people will ‘mine water’from important environmental reserves bydestroying wetlands, or depleting waterbefore it reaches its destination such as LakeChapala (Wester, this volume) or the AralSea.

In the case of closed or closing basins,inter-basin transfers can provide relief. Asouth to north diversion of water from theYangtze to the Yellow river basin could pro-vide some relief to the highly stressed NorthChina Plains. The final step of the stair-steppattern on the available resource and utiliz-able resource lines in Fig. 2.1 representsan inter-basin transfer. There are otherresponses though – such as limiting or man-aging demand for water, or reducing agricul-tural use of water. National policy responsesinclude providing other non-agriculturalemployment or importing ‘virtual’ water(Allan, 1998) in the form of trade incommodities into the basin. How societies

respond to the crisis of overexploitedresources is one of the critical water resourcemanagement questions of our times.

In overly stressed basins, restorationand sustainable use of river systemsbecomes a key agenda item. In the westernUSA, for example, dams are being disman-tled and an increasing amount of the avail-able supply is allocated to environmentaluses. In California, 46.5% of supplies wereallocated to environmental uses comparedto 11% for urban uses and 42.5% for agricul-tural uses in 1995 (Svendsen, this volume).In Australia, the Government of New SouthWales recently reduced allocations to irriga-tion by 10% so that allocations to theenvironment could be increased (HattonMacDonald and Young, 2000). In other situ-ations, such as Pakistan or the North ChinaPlains, concerns over water-based foodsecurity and rural employment would makea reallocation of water out of agriculture avery difficult political choice. The dangerthough is that the ecosystems that supportagriculture may collapse through saliniza-tion, loss of soil fertility or groundwaterdepletion before other rural incomegenerating and food security solutions arefound.

Pathways of basin development varyfrom basin to basin. For example, availablesupplies may actually decline over time,intersectoral reallocation may come intoplay early or interbasin transfers may occurearly in the development phase. Therefore,it should not be construed that following thepathway of Fig. 2.1 is ideal, or that a basin inthe reallocation phase is better than onein the development stage. To understandbasin dynamics and response options forthe future, it is important to understanddevelopment pathways.

Similarly to river basins, locally man-aged irrigation has its own developmentlife cycle, which has to adjust to theexternal environment within which it isset. Irrigation schemes that are transferredwhile the river basin is in an early develop-ment stage have fewer external issues todeal with. By contrast, those transferred ina latter stage of river basin developmentneed to focus both internally on the use of




water within the system, as well as exter-nally to ensure a continued share of theavailable scarce water resource. The mannerin which these locally managed organiza-tions react under these different conditionsis dealt with in Chapter 13. Nine essentialsupporting tasks are identified to supportlocally managed irrigation (see Chapter 1).The following is a discussion of how theseessential tasks take shape and gain or loseimportance according to the phase of riverbasin development.

2.4 Different Phases – Different Needs

Institutional concerns differ dependingon the phase of basin development. Theseconcerns may exist at all times, but theirimportance or emphasis may change overtime as illustrated in Table 2.1. Institutionshave to adapt to meet these changing con-cerns. For example, interest may shift fromconstructing facilities to provide supplyto better management of supplies. As basinclosure approaches, however, demandmanagement becomes a critical concern.At any phase, though, there could be over-lap. For example, even in the developmentphase, managing demand could be a con-cern especially when depletion approachesavailable supplies. The following sections

describe how essential institutional tasksvary as per phases of development.

2.5 Essential Institutional Tasks

2.5.1 Planning, construction andmaintenance

In the development phase, planning for andimplementing infrastructure constructionplays a dominant role. In modern times,organizations such as the United StatesBureau of Reclamation, the MahaweliAuthority and the DSI in Turkey were setup to build major dams, canals and, some-times, drinking water treatment and waste-water plants. Initially, these agencies weredominated by civil engineers who had theimportant job of getting high-quality workdone quickly. These tasks are most easilycarried out by single organizations insteadof several organizations because of prob-lems of coordination and high transactioncosts if many are involved.

During the utilization phase, a differentset of construction and maintenance activi-ties gain prominence. Maintenance becomesa concern to keep facilities in operatingorder, and rehabilitation is warranted tobring systems back to operating order. Mea-surement and regulation structures are often

24 D. Molden et al.

Characteristic Development Utilization Reallocation

Dominant activityFraction of utilizable

flow depletedValue of waterInfrastructure

GroundwaterPollutionConflictsWater scarcityData

Water-povertyconcerns

ConstructionLow (0–0.4)

Low value of waterInstalling new structures

Utilizing groundwaterDiluting pollutionFewer water conflictsEconomic water scarcityWater data – perceived

importance lessIncluding/excluding poor in

development of facilities

Managing suppliesMedium (0.4–0.7)

Increasing value of waterModernization/

rehabilitationConjunctive managementEmerging pollution/salinityWithin system conflictsInstitutional water scarcitySystem water delivery

data importantIncluding poor in O and M

decision making

Managing demandHigh (0.4–>1.0)

High value of waterMeasurement, regulating

Regulating groundwaterCleaning up pollutionCross sectoral conflictsPhysical water scarcityBasin water accounting

data importantLoss of access to water

by poor

Table 2.1. Various dominant characteristics and concerns at different phases of river basindevelopment.



added to assist the operation of facilities.Modernization is often pursued to meet thechanging expectations of users or societies.

During the reallocation phase, restora-tion of environmental damage, clean-up ofpollution, or placing procedures to limitor reduce demand tend to capture theattention of planners. Large and complexinfrastructure development projects, such asinter-basin water transfers, are often soughtto solve water problems. Decentralized con-struction efforts such as the installation ofpumps, small reservoirs or water-harvestingstructures play a significant role in thedevelopment and management of waterresources throughout all the phases ofdevelopment. Often these are not centrallyplanned but individual or communityresponses to water scarcity, with private orcommunity investments. One small struc-ture may not impact much on other uses, butthe cumulative sum of thousands of suchsmall works can quickly push a basin up thedevelopment curve. In China and India, forexample, the amount of area served bygroundwater irrigation has surpassed theland irrigated by large-scale surfaceschemes. Therefore, an institutional chal-lenge is to incorporate these efforts intobroader basin management schemes whilerecognizing the value of these locally driveninitiatives.

2.5.2 Water allocation and distribution

Important allocation decisions are maderight at the development phase. Thedecision of where to place a reservoir orwho gets subsidized support for pumps isbasically a water allocation decision. Somepeople will benefit, whereas others will beexcluded from access to water.

The utilization phase is characterizedby operationalizing within-system alloca-tion and distribution procedures. Oftenthese procedures are ill-conceived duringplanning and construction, or the institu-tions meant to carry out these tasks are ill-equipped to do so. Allocation and operatingdistribution systems takes a different set of

skills than building large infrastructure, yetit is often the same people initially in chargeof construction who make these operationaldecisions.

In the reallocation phase, cross-sectorand between-system allocation and realloca-tion become increasingly dominant con-cerns. For example, now a major concernof Turkey’s DSI is formally to allocate anddistribute bulk supplies of water across usesand sectors.

2.5.3 Monitoring and enforcing water qualityand protecting the environment

During the development phase, the devel-opment of water infrastructure and rain-fed land significantly change landscapes,affecting the environment. The changes areoften large and detrimental and includechanging the hydrologic regime of rivers,impounding large bodies of water anddrying up wetlands. During the utilizationphase, water use and depletion intensifyfurther, removing water that may haveenvironmental functions. Throughout theearly phases of development, dilution canbe sufficient to solve pollution problems.During reallocation phases, dilution is notan option, because there simply is notenough water. Clean-up at the sourcebecomes increasingly critical. In the devel-opment phase, institutions initially must beconcerned with the impact of large infra-structure development such as relocatingpeople, or removing large areas of habitat.Later, in utilization and reallocation phases,problems of pollution, salinization or over-withdrawal of resources become moreimportant.

2.5.4 Protecting against water disasters

Floods and droughts are key triggermechanisms in changing the developmentpathways of river basins. While long-termplanning and overall economic develop-ment of countries play an important rolein determining the path of river basin




development, extreme events have a veryimportant role in shaping water develop-ment trajectories. For example, a devastat-ing flood or drought can provide enoughpolitical backing for the construction ofa new reservoir. Protecting against thesedisasters is important at all phases ofdevelopment, but takes on different forms.Initially, people construct storage facilities.Later operation becomes more critical, andmore sophisticated early-warning systemsand operational systems developed. Some-times, droughts spur the development ofmore groundwater resources, pushing waterresource use beyond sustainable limits.

2.6 The Outcomes: Water Scarcity,Value of Water and Poverty

2.6.1 Water scarcity

Scarcity takes on different characteristicsduring various phases of development. Peo-ple feel water scarcity if they cannot obtainwater, even though it may be plentiful innature. Construction of facilities providesaccess to water, and relieves water scarcity.In the development phase, economic waterscarcity (IWMI 2000) is common. This isa condition where financial or humanresources limit access to water.

However, even with facilities to provideaccess, scarcity can exist. A condition ofinstitutional water scarcity1 exists whenlaws, customs, traditions and other socialbarriers, or organizations deny certaingroups of people easy access to water orprevent the equal distribution water to all,leaving some people water scarce. A persist-ing head–tail problem where overall water isadequate in an irrigation system is an exam-ple of this type of scarcity. Another exampleis when customs or laws restrict access by acertain group of people who are deemed tobe in the lower strata of the social hierarchy.

During the reallocation phase, theabsolute supply of the physical resource islimited. This is a condition of physical water

scarcity. In many closed or closing basins,such as those in the North China Plains, theGediz basin or the Colorado River, physicalscarcity limits the supply of additionalwater. It is important to note that even withphysical scarcity, it is possible to manageinstitutions so that individuals do not sufferfrom lack of access of adequate or good-quality water, though changes in water use,or even livelihoods, may be required.

2.6.2 The value of water

The value of water tends to increasethrough the different phases of develop-ment. Early, when water is plentiful, waterhas a relatively low value, but when a basincloses and demands for a scarce resourceintensify, the value of water can rise dra-matically. Accordingly, in the early phasesof development, concerns are centred ondeveloping a supply of low-valued water,while later in the development process,managing demand prevails. Agricultureand environmental uses are commonly per-ceived as low-valued uses, while industry,hydropower and domestic uses have ahigher economic value. A common phe-nomenon is the initial reallocation ofwater from environmental uses to agri-culture, then from agriculture to cities.Often wealthier countries place increasingvalue of water in environmental uses, andreallocate water back to these uses in thereallocation phase. When low-valued wateris plentiful, conflicts can be mitigatedwith more supplies. As supplies becomelimiting, the potential for conflict increases.

2.6.3 Water and poverty

In each phase of development, waterscarcity has important implications forpoverty (van Koppen, 2000). During thedevelopment phase, a key water and pov-erty concern is to target the poor in water

26 D. Molden et al.

1 First described by M. Samad.



development efforts. An important consid-eration is to identify the beneficiaries. Willinfrastructure benefit poor people? Willmore powerful people capture the benefits?Providing water to a limited number ofpeople in a society with pronounced socialdifferentiation will aggravate inequities.Poverty can be exacerbated or relieveddepending on the planning decision made.

During the utilization phase, having avoice in allocation and receiving a fair shareis a problem of the poor. Even thoughconveyance structures exist, the way wateris managed may not meet the needs of thepoor. If the rich or upstream users are able tocapture water more readily than tail-endersor the poor, the poor suffer. For example,in Kathmandu, homeowners often pumpwater from city pipelines sucking out cheapsupplies, and forcing the less fortunate topurchase high-priced tanker water.

During the reallocation phase, wateris reallocated amongst sectors and people.Maintaining access to water is a key problemfor the poor during this phase. People withfinancial or political power tend to receivemore benefits from water. If water movesaway from agriculture to higher-valuedcities and industries, will the poor and lesspowerful be able to maintain their right oraccess to water, or will they find employ-ment in other sectors? Will poor people beable to capture the economic gains whenwater moves to higher-valued uses?

2.7 Adaptive Institutions

If, early in the development phase, attemptsare made to design an institutional frame-work that deals with all these issues –pollution, poverty, allocation, regulation,construction – it is not likely that successwould be achieved across the board. At cer-tain phases of development, some of theseconcerns take precedence over others. Well-functioning construction agencies have animportant role in safe and sound construc-tion, but using an institutional set-updesigned to manage construction and alsoto manage water delivery is not likely to be

effective. Different sets of rules and skillsare required. Similarly, those who manageservice delivery are probably not equippedto regulate allocation and pollution ofresources when these problems emerge.

The implication is that water resourcemanagement institutions must adapt to meetdifferent challenges as patterns of water usechange. Common water problems are seenbecause agencies do not change fast enoughto adapt to changing needs. Institutionalset-ups then must have the means to recog-nize when important changes take place,and the means to adapt when necessary.

2.8 Development Pathways

At early phases of development, futurechoices about water are many more thanwhen a basin is at the reallocation phase.Near basin closure, many difficult-to-reverse decisions about allocation and infra-structure development have already beenmade. The agenda is to solve complex prob-lems of reallocation and degradation. Eco-system restoration is an important agendain some locations with controlled waterbeing reallocated back to natural uses.

In other basins, with high ruralpopulations and few off-farm employmentopportunities, maintaining intensive use ofwater in agriculture seems the only optionuntil further alternative economic develop-ment takes place. A solution for the problemof groundwater depletion in the NorthChina Plains would be simply to curtail theamount of agriculture. At the moment, this isalmost unthinkable as possibly millions ofpeople dependent on agriculture would losean important means of livelihood and foodsecurity. However, unless solutions in othereconomic sectors or within the water andagriculture sector can be found, the use ofwater for agriculture may cease to be anoption because of its unsustainable use. Theimpact on livelihoods would be the same.

Fortunately, at early phases, several keychoices remain: how much more rainfedand irrigated agriculture; how should thisbe done through dams, groundwater, water




harvesting; what are desirable landscapepatterns and what are associated water con-sequences; how can additional water sup-plies be allocated to the rural poor; and howmuch water to commit to ecological uses.Societies may choose a path of intensiveagricultural development in the hope of eco-nomic development and later restoration.The risk of this approach is ecological col-lapse, or very high costs due to poor health,and clean-up of degraded environment.Lundqvist (1998) refers to this situation as‘hydrocide’. Another development pathwaywould rely on using less water, particularlyin agriculture, and increasing the value orproductivity from each drop used. The asso-ciated risk is not using water to enhanceincome and food security for local poorpopulations.

All too often decisions are made inresponse to a natural or political crisis with-out consideration of long-term goals. Theavailable choices at early phases of develop-ment are not very well thought out. A ques-tion is whether institutional arrangementscan evolve and have the resilience tosolve short-term problems, in light of keylong-term growth and sustainability goals.

2.9 Summary and Conclusions

The growing recognition of a river basin asthe most appropriate unit for the develop-ment and management of water resourceshas prompted the search for appropriateinstitutional arrangements for river basinmanagement. This chapter has argued thatinstitutional requirements differ with thedifferent phases of development of the riverbasin. Institutional designs must be set upto adapt to these changes. Thus, an under-standing of basin development pathwaysis crucial in understanding or formulatinginstitutional arrangements for river basinmanagement. This chapter has outlineda framework to chart phases of riverbasin development based on hydraulic andhydrologic changes.

We demonstrated that as a river basinprogresses from an ‘open’ to a ‘closed’ basin,

three idealized phases can be distinguished:development, utilization and reallocation.These are not mutually exclusive and someoverlap of functions may occur. At the earlystages of development, institutional arrange-ments focus on a single or very limited set ofobjectives, typically involving developinginfrastructure to supply water. Later, moreconcern is placed on managing water withinvarious sectors. When demand outstripssupplies because of more utilization,competition increases, the value of waterincreases, and a host of other issues includ-ing environmental concerns, pollution andgroundwater overdraft may arise. Over time,there is an increasing need to deal withmultiple functions that require complexinstitutional arrangements that involveseveral organizations, and that function inthe realm of a broader and often conflictingset of national objectives. Thus, institutionsshould be conceived as dynamic entitiesthat need to cater to different managementdemands as water use changes over time. Akey feature of an effective institution will beits ability to adapt to changing needs.

Acknowledgements

The authors appreciate the constructivesuggestions by Mats Lannerstad, GinaCastillo and Mark Svendsen.

References

Allan, T. (1998) Moving water to satisfy unevenglobal needs: ‘trading’ water as an alternativeto engineering it. ICID Journal 47, 1–8.

Allan, T. (2002) Water resources in semi-aridregions: real deficits and economically invisi-ble and politically silent solutions. In Turton,A. and Henwood, R. (eds) Hydropolitics inthe developing world: a Southern Africanperspective. African Water Issues ResearchUnit, University of Pretoria, Pretoria.

CCAP and CAAS (2000) Water Accounting forFuyang River Basin. 2000. Draft InterimReport prepared by the Chinese Centerfor Agricultural Policy and the ChineseAcademy of Agricultural Sciences for

28 D. Molden et al.



the Five Country Regional Study onDevelopment of Effective Water ManagementInstitutions.

Hatton MacDonald, D. and Young, M. (2000) Insti-tutional arrangements in the Murray–DarlingRiver Basin. In Abernethy, C.L. (ed.) Inter-sectoral Management of River Basins:Proceedings of an International Workshopon Integrated Water Management in Water-Stressed River Basins in Developing Coun-tries: Strategies for Poverty Alleviation andAgricultural Growth, Loskop Dam, SouthAfrica, 16–21 October 2000. IWMI/DSE,Colombo, Sri Lanka.

ICOLD (International Commission on Large Dams)(1998) World Register of Dams. ICOLD, Paris.

IWMI (2000) World’s Water Supply and Demand1995–2025. IWMI, Colombo, Sri Lanka.

Keller, J., Keller, A. and Davids, G. (1998) Riverbasin development phases and implicationsof closure. Journal of Applied IrrigationScience 33, 145–163.

Kendy, E., Molden, D.J., Steenhuis, T.S. andLiu, C. (2003) Policies drain the North ChinaPlain: agricultural policy and ground-waterdepletion. In: Luancheng County, HebeiProvince, 1949–2000. IWMI Research Report,forthcoming.

Lundqvist, J. (1998) Avert looming hydrocide.Ambio 27, 428–433.

Molden, D., Sakthivadivel, R. and Samad, M.(2001) Accounting for changes in water useand the need for institutional adaptation. InAbernethy, C.L. (ed.) Intersectoral Manage-ment of River Basins: Proceedings of an inter-national workshop on Integrated Water Man-agement in Water-Stressed River Basins inDeveloping Countries: Strategies for PovertyAlleviation and Agricultural Growth, LoskopDam, South Africa, 16–21 October 2000.IWMI/DSE, Colombo, Sri Lanka, pp. 73–87.

Molle, F. (2003) Development Trajectories ofRiver Basins: a Conceptual Framework.IWMI Research Report, forthcoming.

Ohlsson, L. and Turton, A. (2000) The turningof a screw. Social resource scarcity as abottle-neck in adaptation to water scarcity.SAS Water Issues Study Group OccasionalPaper 19. University of London, London.

Seckler, D. (1996) The New Era of Water ResourcesManagement: From ‘Dry’ to ‘Wet’ WaterSavings. IWMI Research Report 1, Colombo,Sri Lanka.

Shiklomanov, I.A. (2000) Appraisal andassessment of world water resources. WaterInternational 25, 11–32.

van Koppen, B. (2000) From Bucket toBasin: Managing River Basins to AlleviateWater Deprivation. International IrrigationManagement Institute, Colombo, Sri Lanka.




3 Limits to Leapfrogging: Issues inTransposing Successful River Basin

Management Institutions in theDeveloping World

Tushaar Shah, Ian Makin and R. Sakthivadivel

3.1 Backdrop

Management becomes important as aproductive resource becomes scarce, andthere is hardly a situation in which this istruer than in the case of the water resource.For a long time now, water policies ofmany emerging nations have been focusedon developing the resource, and optimizingwas directed at the efficiency of waterinfrastructure rather than water itself. Aswater has become increasingly scarce,optimizing is now being increasinglydirected to improving the productivityof water. Increasingly, the river basin isemerging as the unit of management ofland, water and other natural resourcesin an integrated fashion. Many developedcountries such as the USA, France andAustralia have evolved highly advancedand resilient institutional regimes forintegrated river basin management (IRBM);but this has taken decades – in Europe andthe USA, centuries – of gradual change toevolve. An issue that has held great appealto policy makers and social researchersis this. Is it necessary that developingcountries in Asia and Africa take all thatlong in crafting such institutional regimes?Or might it be possible for them to doan ‘institutional leap-frog’, as it were, to

a stage at which developed-country basininstitutions are today?

A textbook case of institutional reformfor IRBM in recent times has been theMurray–Darling basin in Australia, wheresweeping changes have been made andenforced since 1990, and transferringthe lessons of success in IRBM – fromMurray–Darling to Mahaweli and Mekong– has emerged as a growth industry.

This chapter attempts a broad-brushapproach to understanding the materialdifferences between the contexts of thedeveloped-country river basins from whereinstitutional models emerge and the devel-oping-country river basins in which theseare applied. The idea is not to underminethe significance of the lessons from successbut to emphasize the need for sagacityand critical analysis in assessing what willwork and what will not, given the differ-ences in context. The phrase ‘institutionalchange’ is used to describe how communi-ties, government and society change recur-rent patterns of behaviour and interactionsin coping with water scarcity and its socio-ecological ill-effects. It involves understand-ing laws and rule making, roles, policiesand institutional arrangements at differentlevels. The overarching premise is that theeffectiveness of a pattern of institutional




development is determined by at least fourrealities of a river basin: hydrogeology,demography, socioeconomics and the orga-nization of the water sector. By implication,institutional arrangements that have provedeffective within one set of these realities mayrequire major adaptation before they becomeappropriate to the needs of a river basincontext defined by an alternative set ofrealities.

IRBM is a powerful idiom, and willincreasingly dominate natural resourcemanagement discussion in the developed aswell as developing world. In its broadestsense, a basin or catchment is visualized as

an inter-connected machine or systemwhich transforms natural inputs of solarenergy, atmospheric precipitation, nutrientsand other environmental factors, alongwith man-made inputs of labour, capital,materials and energy, into output productssuch as food, fibre, timber, building materi-als, fuels, minerals, natural vegetationand wildlife, recreational and aestheticamenities, buildings and developmentsites, as well as water in desirable qualityand quantity. (Burton, 1986, cited in Hu,1999, p. 324)

River basin management (RBM) as a notiongoes far beyond traditional land and watermanagement and

includes significant parts of land-useplanning, agricultural policy and erosioncontrol, environment management andother policy areas. It covers all humanactivities that use or affect fresh watersystems. To put it briefly, RBM is themanagement of water systems as part ofthe broader natural environment and inrelation to their socioeconomic environ-ment. (Mostert et al., 1999, p. 3)

Institutional discussions on IRBM havetended invariably – and probably errone-ously – to gravitate around three models ofstrategic organizations for managing riverbasins:

• The hydrological model, in which ariver basin organization/authority, cut-ting across administrative boundaries,takes over all charge of water resourcemanagement;

• The administrative model, prevailingin many developing countries, inwhich water management is theresponsibility of territorial organiza-tions unrelated to hydrologicalboundaries; and

• Coordinating mechanisms superim-posed on the administrative organiza-tions to achieve basin managementgoals.

Each has advantages and disadvantages.The hydrological model effectively dealswith upstream–downstream issues thatthe administrative organization is generallyunable to deal with. However, hydrologicalorganizations tend typically to focus onwater and overlook land managementissues. River basin commissions, as ahybrid, might combine the advantages ofboth, but at least in the developing-countrycontext, they often command little author-ity, and are therefore confined to lowest-common-denominator solutions (Mostertet al., 1999). In many developing countriestoday, institutional reform for RBM isconfined almost wholly to the creation ofthe basin-level organization – the implicitassumption being that mere formation ofthe appropriate organization will result inIRBM, an assumption whose validity hasbeen repeatedly refuted.

In the developed world, the discussionhas been much broader and has veeredaround initiatives in four aspects of naturalresource governance:

1. Some mechanism for basin levelnegotiation and coordination fortified withadequate authority and resources, and abroad mandate considered appropriate tothe basin’s context;2. Legal and regulatory reform;3. Redesigning economic instruments ofpolicy (transfer prices, taxes, subsidies) inharmony with national policy goals;4. Redesign of economic institutions(including utilities, service providers,property rights; water markets, irrigationmanagement transfer to user organizations).

Countries like the USA have achieved, overlong periods, high levels of integration even

32 T. Shah et al.



without a central basin organization (e.g.Svendsen, Chapter 7, this volume).

3.2 Applying the Lessons of the Murray–Darling to the Developing World

The Murray–Darling River system, as arecent case of accelerated institutionalreform, has emerged as a model of institu-tional structure for IRBM. The basin encom-passes over 75% of the state of New SouthWales, 56% of the state of Victoria, all ofthe Australian Capital Territory, and smallparts of Queensland and South Australia, avast region of the southeastern part of thecontinent. Already, several case studies ofthe Murray–Darling are available, and it isnot our intention to review these. In brief,the institutional innovations of the Murray–Darling basin management regime includethe following:

1. The Murray–Darling Ministerial Coun-cil as the top-level policy-making andcoordinating mechanism, the Murray–Darling Basin Commission as the operat-ing organization, and several CatchmentManagement Agencies that are responsiblefor day-to-day management of water;2. A system of permits for diversions thatencompasses all uses except the waterneeded for domestic use, livestock produc-tion and irrigation of up to 2 ha which arerecognized as prior rights (MacDonald andYoung, 2000, p. 10), and exempted fromlegal as well as permit systems;3. An effective cap on water diversions at1993–1994 levels of development to ensureadequate environmental supplies, accompa-nied by a system of volumetric licensing tousers that raises the scope for large-scalewater trade across states and sectors;4. Consumption based, full-cost-recoverypricing (MacDonald and Young, 2000,p. 14);5. A system of ‘salinity credits’ thatpermits trade in salinity;6. Explicit mechanisms for water alloca-tion for environmental needs;7. A legal regime that separates waterrights from land rights;

8. Privatization of service providers suchas Murray Irrigation Ltd and Victoria’s RuralWater Corporation (Malano et al., 1999).

The Murray–Darling RBM regime clearlyrepresents a highly evolved form ofinstitutional arrangement and effectivelyaddresses all major problems that a matureriver basin would face. As alluded to ear-lier, exploring whether developed-countrybasin institutions – particularly, Murray–Darling experience – can be replicated in adeveloping-country context has fascinatedmany researchers in recent years. An entireissue of Water International (vol. 24, no. 4,1999) was devoted to it in 1999. The resultsof these investigations have not been veryencouraging. For example, Hu explored theapplicability of Murray–Darling experiencein the Chinese context and concluded nega-tively because of: (i) difficulty of coordi-nating authorities at different levels; (ii)unclear ownership of resources; (iii) smallfarming scales; and (iv) poor education ofresource users (Hu, 1999, p. 323).

In a similar vein, Malano et al. (1999)ask: ‘Can Australian experiences be trans-ferred to Vietnam?’ Their conclusion is lessemphatic than Hu’s, but all their evidencesuggests that it will be long before Vietnambecomes really ready for the Murray–Darlingprescription, and that ‘context, hydrologicaland socioeconomic, defines the detail andbalance that is required . . .’ (p. 313). The newwater law of Vietnam contains provisions toadopt an integrated river basin approach.The Ministry of Agriculture and RuralDevelopment, which is at present in chargeof water, does not relish the responsibility ofIRBM. The progress in stakeholder partici-pation, another Murray–Darling prescrip-tion, has been slow to say the least. Farmersview irrigation provision as a governmentresponsibility; even so, irrigation charges inVietnam are high by Asian standards. Yet,presumably under donor pressure, thegovernment tried to eliminate irrigationsubsidies, but this was followed by massivepopular unrest in 1998, whereupon, theGovernment had to restore the subsidies.

Can the Australian success in enforcingthe ‘user pays’ principle be transferred to the

River Basin Management in the Developing World 33



Solomon Islands? Hunt explored this issuein a recent study and concluded that suchtransfer ‘is not sustainably viable’ onaccount of huge differences in politicalstructures, national priorities, living stan-dards, cultural traits, technological develop-ment, literacy levels, financial and infra-structure growth, and change-managementcompetency. All these differences result inthe absence of what Hunt calls a ‘contextualfit’ between the policy development and therespective policy application environment(Hunt, 1999, p. 302).

‘If there is any conclusion that springsfrom a comparative study of river systems,it is that no two are the same’ (GilbertWhite, cited in Jacobs, 1999). Each riverbasin must differ from any other in athousand respects, but that does notmean that lessons of success in one areof no value to another. It does meanthough that uncritical ‘copycat’ replicationof successful institutional models – either byenthusiastic national governments or at thebehest of enthusiastic donors – is a sure for-mula for failure. The history of institutionalreform in developing-country water sectorsis dotted with failures of such copycatreform.

IRBM is not a new idea, even indeveloping countries. India tried to trans-pose the TVA (Tennessee Valley Authority)model tried in the USA by constitutingthe Damodar Valley Authority, which was aresounding failure. Catchment managementcommittees were established in China wayback in the 1950s in some of the majorriver basins such as the Yangtse and YellowRiver, to plan and exploit water resources,generate electricity, mitigate flood damageand provide facilities for navigation (Hu,1999, p. 327). However, all these institutionsshed their broad agenda and ended up focus-ing on irrigation, the purpose that was mostcentral to their domains at those times.

In Sri Lanka, a Water Resources Boardwas established as early as 1964 to promoteintegrated water resources planning, riverbasin and trans-basin development and totackle water pollution. However, the Boardnever worked on its broad mandate and,instead, took to hydrological investigationsand drilling tubewells.1 Such examples canbe multiplied easily. The point is, in takinguseful lessons from success cases for makingmeaningful reform in developing countries,it is important to understand critical differ-ences between the two worlds that affectwhat will work and what will not. Wepropose that, in assessing the applicabilityof institutional innovations, it is criticalto take into account four types of materialdifferences between the developed- anddeveloping-country realities:

1. Hydrology and climate;2. Demographics;3. Organization of the water sector;4. Socioeconomics.

We briefly outline these material differ-ences in the following sections.

3.3 Hydrology

Historically, agriculture advanced early inarid climates such as those of Egypt andIraq, but industrial development beganearly in the temperate and humid climatesof Europe, North America and Japan. Aridareas where significant wealth creation andaccumulation has occurred, such as in WestAsia, are typically rich in mineral and oilresources. As of today, however, the bulk ofthe developing world, where rainfall tendsto be low and water scarcity is a majoremerging constraint to progress, is inthe arid or semiarid2 parts of the world.Figures 3.1 and 3.2, showing the globaldistributions of mean annual rainfall and

34 T. Shah et al.

1 Another round of reform has just begun in Sri Lanka. In 1990, a draft law made provision for bulk waterallocation and the establishment of a National Water Resources Council to do what the Water Resources Boardcould not. However, the law could be submitted to the Parliament only in 1995 for the lack of consensus withinthe cabinet as well as amongst the myriad agencies dealing with water (Birch and Taylor, 1999, p. 331).2 Referring to regions like India and West Africa, which are humid for a small part of the year but arid duringthe rest of the year.



potential evapotranspiration, help to illus-trate some major climatic differencesbetween developed countries (mostly inthe temperate latitudes) and developingcountries (mostly in the tropical andsubtropical regions).

Sutcliffe (1995) pointed out that devel-oping countries also happen to be concentra-ted in parts of the world with more extremeclimates when compared to the regionsoccupied by today’s developed countries.Figure 3.1 illustrates the variation in annual


Fig. 3.1. Global distribution of mean annual rainfall.

Fig. 3.2. Global distribution of mean annual potential evapotranspiration.



rainfall. India, for example, receives almostall of its annual rainfall in less than 100 h oftorrential downpours during June–October,and its challenge is to save enough of it fromevapotranspiration to last from October untilApril–May, the months that mark the periodof highest water stress. Botswana receives allof its 350–500 mm of rainfall during Novem-ber–March, the period which also coincideswith the highest evaporation, resulting inlittle or no runoff (Sutcliffe, 1995, p. 69).

Humid areas typically have higherstream densities than are found in the aridand semiarid areas, which means that,ceteris paribus, a higher proportion of pre-cipitation in the arid and semiarid areas runsoff in sheet flow before forming into streams,and is thereby subject to higher evapotrans-piration losses (Fig. 3.2). Other things alsoare not quite the same. The tropical develop-ing world – especially, South Asia and muchof Africa – has higher mean temperatures formore of the year than the developed world,and, for equivalent levels of precipitation,runoff and the need for irrigation tend to begreater in arid and semiarid areas than inhumid areas (Sutcliffe, 1995, p. 64).

Climate and hydrological conditions,combined with demography (discussed in

the following section), explain why decen-tralized institutions for water managementhave historically evolved in many partsof the developing world. The profusion ofsmall tanks in India’s southern peninsulaand Sri Lanka can be viewed as the responseof communities in the catchment areas tostake their claim on rainfall. Even today, onecollective maintenance task carried out bymany South Indian tank communities beforethe start of the monsoon is cleaning anddeepening the channels that feed rainwaterrunoff to their tanks. Village people hererecognize that if they do not capture runoffin artificial streams, most of it will be lostbefore it reaches their tanks.

3.4 Demographics

Many parts of the developed world haveextreme climates too; however, over time,population and urbanization in these partshave tended to concentrate in wet areasor on downstream reaches of rivers nearcoastal areas, where water can be suppliedthrough large-scale diversion structures. AsFig. 3.3 shows, except in Europe, most of

36 T. Shah et al.

Fig. 3.3. Global distribution of population density.



the developed countries have low popula-tion densities throughout, with urbanagglomerations near the coasts and ruralpopulation along rivers or irrigationsystems. Here, the competition is forlarge accumulated bodies of ‘divertedwater’. Since catchment areas have rela-tively sparse populations, the downstreamwater-harvesting structures have largecatchment areas that are virtually free fromcompetition.

However, this is not the case in someof the most densely populated regions ofthe world. In India, for instance, populationdensity is high – approaching 600 personsper km2 in the water-rich Ganga basin, andseldom less than 350–400 even in semiaridWestern India and hard-rock peninsularIndia. Population density is high bothupstream and downstream of dams. Thesame is true for much of China. The NorthChina plains have much less water thanSouth China, but their population densityis around the same. One might argue thatthe cause of intensive groundwater develop-ment in South Asia and China is that mostpeople here cannot be downstream oflarge dams. By sinking tubewells, peopleupstream are, in a sense, challenging thebasic inequity inherent in the pattern oflarge irrigation projects that usurp therainfall precipitation of populous upstreamcatchment areas to bequeath it to a smallnumber of canal irrigators.

All these factors have had implicationsfor the kind of water institutions that haveevolved historically in the developed anddeveloping world. For example, the systemof rights based on riparian doctrine and onthe doctrine of prior appropriation is alien tothe cultures of many developing countriesbecause the largest majority, by far, dependupon rainfall and local water-harvesting andstorage structures. Riparian rights or priorallocation become operative only alongthe streams and rivers, where the bulkof the irrigators and water users tend to be

concentrated in countries like the USAor Australia, but these make no sense,for example, for some 20 million personspumping groundwater in South Asia, or thecommunities that use over 300,000 tanks inSouth India or 7 million ponds in China.

Because large proportions of the popu-lation in the developing world depend uponrain and on local storage, the people’snotions of ownership and rights relate moreeasily to precipitation than large-scale pub-lic diversions. Egypt gets less than 10% of itswater from rainfall, yet Egyptians considerthe rainwater to be truly their own. In Asia,where population densities are commonlyas high in the catchment areas of the basin asalong the stream and river channels, theimplicit primacy of the right of communitiesover precipitation rather than over diver-sions is widely accepted. Indeed, in recentyears, a popular slogan in Western India is‘rain on your roof, stays in your house; rainon your field stays in your field; and rain inyour village stays in your village’. In theWestern countries, upstream–downstreamconflicts are important because most waterusers think of users upstream as their rivals.In the World Water Forum that met at TheHague in March 2000, the slogan that theDelhi-based Centre for Science and Environ-ment popularized was ‘everyone livesdownstream’, which is eminently sensible ifall or a majority of people in a basin dependfor their water needs directly upon rainfall.The IRBM discussion talks very little of theenormous amount of work on farming inthe semiarid tropics done by national andinternational centres such as ICRISAT.3 Asthe Global Water Partnership (2000, p. 25)notes, ‘Most water management, includingthe literature on IWRM, tends to focus onthe ‘blue water’,4 thus neglecting rain andsoil-water management. Management of‘green water’ flows holds significant poten-tial for water savings.’ This is because thereis little real ‘dryland farming’ of the Indianand West African variety in the developed


3 International Crop Research Institute for the Semi-Arid Tropics.4 ‘Blue water’ is water existing in bodies such as rivers or lakes, or pumped from aquifers. ‘Green water’ issoil water extracted and transpired by plants.



world; but making the best use of soilmoisture is a critical issue in many Africanand Asian countries. Europe, Canada, NewZealand and the USA do have rainfedfarming; but this is not quite the same asdry farming in western Rajasthan or sub-Saharan Africa; in many of these countries,favourable rainfall and climate conditionsresult in favourable soil-moisture regimesthat make irrigation unnecessary forgrowing good crops.

The conventional notion of irrigationis one of controlled supply of water to meetthe full scientific requirements of plantsprecisely when needed, but the irrigationthat is most widely practised in South Asiaand amongst smallholder farmers in Africais supplemental irrigation designed toincrease the productivity of ‘green water’.Green water is the precipitation useddirectly for crop production and thus ‘lost’in evaporation; blue water pumped out fromwells is as important in South Asia andNorth China as the part that flows into riversand canal systems.5 This is quite unlike thesituation in many developed-country riverbasins. In these, the bulk of economic waterdemands have been met from developmentof blue surface water and where, with theclosure of these basins, the focus of basinmanagement is on raising the productivity ofblue surface water, largely without regard togreen water.

Uniformly high population density,combined with unhelpful climate andhydrology, has a profound impact on theobjectives of water management in develop-ing-country river basins. In recent years,IWMI’s water accounting work (Molden andSakthivadivel, 1999, pp. 58–60) has mademuch contribution to understanding waterproductivity in the basin context.6 AlthoughIWMI’s focus has been on productivity ofwater in agriculture, the framework can be

easily generalized to develop a notion ofbasin-level water productivity in terms of asocial welfare function for all stakeholdersin a river basin constituting a basin commu-nity. Under this broad concept:

Basin welfare productivity of water = Basinwelfare/Available water.

Water productivity understood thus couldbe enhanced by:

1. Enhancing productivity in each use;and2. Constantly reallocating water amongstalternative uses – irrigation, domestic,industrial and environmental – so that themarginal contribution to overall welfare bywater allocated to all uses remains equal.

Using the IWMI water accounting frame-work, this welfare productivity measurecan be written in several alternative ways tohighlight the importance of different wateruse strategies,7 but for highlighting the dif-ference between developed and developingworld, a useful way to write the welfareproductivity ratio is:

Basin welfare/Available water = [Basinwelfare/Diversions] × [Diversions/Availablewater].

In relatively water-abundant humidregions, with low population density in thecatchment areas and dense human settle-ments near the coasts and along rivers, RBMseeks to maximize basin welfare productiv-ity by increasing the term [Basin welfare/Diversions]. Allocation of diverted wateramongst alternative uses is a crucial func-tion in basin-level water management insuch conditions. Here, reservoirs have largefree catchments, and evapotranspiration incatchment areas is often not high; therefore,the need for active human intervention tomaximize [Diversions/Available water] isnot great.

38 T. Shah et al.

5 This distinction between green and blue water is extremely important for developing countries in thesemiarid tropics. Terrestrial ecosystems are green-water dependent; aquatic ecosystems are blue-waterdependent (GWP, 2000, p. 24).6 Standard definitions used in IWMI water accounting work are found in Molden and Sakthivadivel (1999).7 For example, by writing [Basin welfare/Available water] = [Basin welfare/Total depletion] × [Totaldepletion/Total diversion] × [Total diversion/Water available], we can signify alternative routes to waterproductivity.



In water-scarce tropical countries withhigh population density everywhere, as inSouth Asia and China, maximizing basinwelfare involves working on both thecomponents. Increasing the productivity ofdiverted water is certainly important; butequally important is the need to maximizethe proportion of precipitation and inflowsinto a basin that can be diverted before theyare lost to non-beneficial depletion.

It is against this backdrop that we needto consider the growing mass movementfor rainwater harvesting and groundwaterrecharge in western India (Shah, 2000). Theregion has amongst the highest wind speedsencountered anywhere in the world, ithas high mean temperatures for 9 months,rainfall varies between 300 and 800 mm/year, and population density is 300–500persons per km2 in the catchment areasas well as in the downstream areas. Thegreatest challenge for rural communities issurviving the annual pre-monsoon droughtin April and May, which is made infinitelymore daunting by regular failure of monsoonrains. During the pre-monsoon months,leave-alone growing crops, ensuring ade-quate drinking water for humans and cattleis the great challenge, especially in thecatchment areas of river basins. Whilegovernment investment programmes con-centrated on building large reservoirs down-stream to support irrigation and municipalwater supplies to towns, the problems ofthe people living in the catchment areasremained unaddressed.

Disenchanted with government andpublic systems, non-governmental organiza-tions (NGOs) and communities are findingtheir own solutions. The past decade haswitnessed a massive popular awakening asthe result of the efforts of NGOs like TarunBharat Sangh, Pradan, and religious organi-zations such as the Swadhyaya Pariwar.This has taken the form of rainwater conser-vation and groundwater recharge work on ascale that governments or public agencieswould not be able to manage. The basic moti-vation that has been driving the movementis to ensure availability of domestic watersupply for 2 months before the monsoonand for one or two crop-saving waterings

from wells; there are indications that themovement may well meet this challenge.

Government agencies and scientists(hydrologists in particular) have been dubi-ous about this mass movement. Their argu-ment is that rainwater-harvesting structuresupstream merely transfer water, reducingthe input into the reservoirs downstream,thereby reducing their productivity. How-ever, this argument does not resonate withthe communities, especially in the upstreamareas, which fail to see why they cannotmeet their basic water domestic needsinstead of feeding reservoirs to irrigaterelatively small areas of paddy or cotton.In defence of this popular movement, theDelhi-based Centre for Science and Environ-ment has asked: what does India need more– irrigation or drought proofing? In reply,it has suggested that by a total rethink on‘appropriate’ RBM, India can trade droughtproofing over vast areas by sacrificingirrigation of small areas.

It has also adduced evidence to showthat diverting rainwater in a large number ofsmall water-harvesting structures in a catch-ment captures and stores more of the scarceprecipitation, closer to the communities,than having a large reservoir downstream(Agarwal, 2000). This is because watercollected over larger watersheds will have toflow over a larger area before it is collectedand a large part will be lost in small puddlesand depressions, as soil moisture andevaporation.

For instance, Agarwal cites evidencefrom the Negev desert in Israel to show that3000 micro-catchments of 0.1 ha capturefive times more water than a single catch-ment of 300 ha, and this multiple increasesin a drought period (p. 9). He also citesresults by Michael Evanari, an Israeli scien-tist that show that ‘While a 1 ha watershed inthe Negev yielded as much as 95 m3 of waterper ha per year, a 345 ha watershed yieldedonly 24 m3/ha/year. In other words, as muchas 75% of the water that could be collectedwas lost. This loss was even higher ina drought year.’ Agarwal cites Evenari:‘. . . during drought years with less than50 mm of rainfall, watersheds larger than50 ha will not produce any appreciable




water yield while small natural water-sheds will yield 20–40 m3/ha, and micro-catchments (smaller than 0.1 ha) as much as80–100 m3/ha.’

Before irrigated crop production,semiarid India needs drinking water for itsdispersed rural population during the 9months without rainfall. Many Indianobservers think that the answer is not pipedwater supply schemes but decentralizedrainwater harvesting.8

3.5 Organization of the Water Sector

Developed-country water sectors, whichhave evolved over decades of public inter-vention, tend to be highly organized andformalized, with most of the water deliv-ered – and most of the users served – byservice providers in the organized sector.In low-income countries, a vast majority ofwater users – the poorest ones – get theirwater directly from rain and from localprivate or community storage without anysignificant mediation from public agenciesor organized service providers. The notionof professional water service providers isalien to a majority of rural South Asians andAfricans.

As a society evolves and its economy,as well as its water sector, matures, the bulkof the water delivered to ultimate usersis produced, developed, planned and allo-cated – in general, managed – by formalorganizations, businesses or utilities. InIsrael, for example, 70% of the water supplyin the country is managed by Mekorot, aState-owned water company that operatesthe National Water Carrier, the pipelinesystem that moves water from Lake Galileeto the Negev Desert, and is in urban waterretail, desalination and sewerage treatmentbusinesses (Saleth and Dinar, 2000, p. 185).When the bulk of the users and uses areserved through the formal sector, resource

governance becomes feasible, even simple.If a basin management regime wants toincrease the water price to domestic usersby 5%, or make a law intended to changethe way business is done, it can do so withthe confidence that it will stick. However,this is not true when the bulk of the waterusers and uses are served by an informalsector where service providers are not evenregistered.

In comparing the Australian successwith containing agricultural pollution ofwater with the Chinese situation, Hu (1999,p. 327) laments that while the small numberof large Australian farmers are served by arange of local organizations, such as sugar,rice and cattle associations, which serve asvehicles not only for new knowledge andtechnical advice but also for implementingnew rules and laws, in China, ‘given thesmall scale of farming units and the largenumber of farmers, it is difficult to controlchemical and pesticide application, removalof vegetation, erosion and water resourceexploitation.’ In South Africa, over 90% ofwater is managed by formal organizationsbut 90% of rural people, the black irrigatorsin former homelands, are almost wholly inthe informal sector, far out of the reach of theorganized systems.

Ignoring the complexity of dealing withthe informal water sectors in the developingworld can lead to misleading analyses. Inthe perspective of Saleth and Dinar (2000,p. 186), for example, the institutional reformchallenge in South Africa ‘lies in translatingthe provision of its water law and waterpolicy without creating much uncertaintyamong private investors.’ In our view, theseare easily done. The real challenge withwhich the government of South Africa isstruggling is in reaching the reform to theblack communities in the former home-lands, who operate in the informal watersector. As hard as the government is trying,this is not proving to be easy. About theprocess of Catchment Management Agency

40 T. Shah et al.

8 The Centre for Science and Environment has estimated the average area needed per village to capturesufficient water to meet every household’s drinking and cooking water requirement in various regions withvarying climate, precipitation and demographic conditions. The average for India as a whole was all of 1.14 haper village in a normal year and 2.28 ha per village in a drought year.



(CMA) formation in Olifants, South Africa,Merrey (2000, p. 9) writes:

rural communities were unaware of theprovisions of the new water law and theCMA process, despite the efforts to informpeople and offer them opportunities toexpress their views. Small-scale farmershad not heard about the CMA . . . [But] theIrrigation Boards providing water to largecommercial farmers were participatingactively in the process . . .

A small number of large stakeholders iseasy to work with. The ballgame changesfundamentally once we have to deal with ahuge number of tiny stakeholders.

One way the informal sector can be‘formalized’ is through grassroots user orga-nizations, and the global irrigation manage-ment transfer (IMT) initiatives to organizeirrigators into water users associations ispartly motivated by the need to bring theminto the formal sector. In this too, small num-bers of large users in the developed worldhave an advantage over large numbersof small users in the developing world.All manner of user associations formspontaneously in countries like the USAand Australia. These institutional modelsare constantly being tried out in developingcountries but, here, these generally breakdown when faced with large numbers ofsmall stakeholders who face such diverseconstraints in their livelihood systems thatthey are at best apathetic towards each ofthem.

Thus, for example, IMT to wateruser organizations has unambiguouslysucceeded in the USA, New Zealand,Colombia, Turkey and Mexico, all situationsof medium to large commercial or exportfarmers who run their farms as wealth-creating enterprises. In contrast, nowhere inlow-income Asia, barring a few ‘islands ofexcellence’, including the much-researchedPhilippines, has IMT held out the promise oflong-term sustainability. White commercialfarmers in South Africa took to IrrigationBoards like ducks to water. In African small-holder black irrigation schemes, there seemslittle chance that IMT will take off at allunless it is preceded by a wide-ranging

intervention to make smallholder farmingitself viable (e.g. Shah et al., 2000).

One standard refrain of institutionaldiscussions in the water sector is get waterlaw and get it ‘right’. It is often the case, how-ever, that the problem is not passing a lawbut in enforcing it in a society with a largenumber of tiny stakeholders operating in theinformal sector with little or no linkage withmeso- and macro-level resource governancestructures. This is why many governmentsin Asia readily pass Acts but spend yearsbefore converting these into implementablelaws and regulations.

There are also cases of countries thathave passed laws which have come totallyunstuck. Sri Lanka has been debating a waterlaw – which has ‘all the right ingredients’(Saleth and Dinar, 2000) – since the early1980s, but is yet to enact it. This ispresumably because it is difficult to figurehow to make all the ‘right ingredients’ –water permit systems, full-cost pricing,water courts and explicit water policystatement – actually work in ways that makea significant difference to the managementof water resources in a country where50–70% of the rural people acquire theirwater not through water supply serviceutilities/companies but straight from natureor from local storage in small communitytanks.

India adopted a water policy in 1987,but nothing changed as a consequence, andit is now working on a new one. Many Indianstates have likewise been debating ground-water laws for 30 years and a dozen or sodrafts are in circulation. The legislativeassembly of Gujarat, a state with severegroundwater overdraft problems, passeda bill as far back as in 1974, but the ChiefMinister refused to make it into a law. Hisreasons were convincing. First, he wasunable to see how the law could be effec-tively enforced on a million small privatepumpers scattered throughout a huge coun-tryside, and second, he was certain thatit would become one more instrument ofrent-seeking for the local bureaucracy(Shah, 1993).

‘Get the price right’ is another old pre-scription to make water an economic good.




Now that water scarcity in many parts of theworld is real, it would be naïve to questionthe value of pricing, not so much for revenuecollection but to signal the scarcity valueof water to users. There can be no seriousdebate on whether the view of water asa ‘scarce but free’ resource is tenable intoday’s context. The real issue is making theprice of water stick in a situation where amajority of users are in the informal sectorand do not go to anyone except the rain-godsfor getting their water.

Even in canal irrigation systems inSouth Asia, which are in the formal sector,many political leaders and senior adminis-trators would become open to volumetricpricing of water to promote efficient use, ifonly the logistics of doing so were simpleand cost-effective, what with the large num-ber of small irrigators in the commands ofAsian systems. After all, paying high pricesfor high-quality irrigation service is commonfor millions of resource-poor buyers of pumpirrigation in India, Pakistan, Bangladesh andNepal. Most people would avoid paying thefull-cost price if not paying were an option,as is the case in many developing-countrywater sectors.

The high transaction cost of monitoringwater use and collecting water charge is thecentral issue in water pricing, rather than thepoliticians’ propensity towards giving awaylargesse. This will soon be evident in SouthAfrica, where a new water pricing policywill be easy to enforce on large commercialfarmers, since transaction cost of monitoringand collection will be low, but very difficultto apply in areas of black irrigation, domi-nated as they are by large numbers of smallusers.

Developed-country institutions havenot solved the problem of serving or regulat-ing large numbers of small users. Indeed,they have not yet found satisfactory waysof dealing with moderate numbers of largeusers. In New South Wales, Queensland andVictoria, the existing law confers on everyoccupier of land the right to take and usewater for domestic consumptive purposes,watering stock, irrigating home gardens andnon-commercial crops on a maximum of2 ha (MacDonald and Young, 2000, p. 24).

If this exemption were applied to India, itwould cover over 80% of all land and over90% of all people. In South Africa, it wouldcover 90% of all users though only 10% ofland and water. In South Asia, South-EastAsia and North China, groundwater isthe most valuable and threatened resource,and protecting groundwater from over-development is probably among the topthree priorities in this region. Yet doing so isproving to be a challenge precisely becausegroundwater is in the informal sector.

In the question of how best to deal withSouth Asia’s 20 million tubewell ownersin the informal sector, the experiences ofMurray–Darling or Mississippi do not havemany practical lessons to offer. Even in‘highly evolved’ river basins, sustainablemanagement of groundwater is at best prob-lematic, and at worst, as hopeless as inIndia and Pakistan. Murray–Darling hastried groundwater regulation but it is notcertain if it has worked. Access to ground-water in New South Wales is regulated bylicences under the Water AdministrationAct of 1986. However,

over much of New South Wales, undevel-oped licences were not cancelled. In retro-spect, this has proved an administrativedisaster as, in a number of areas, the totalvolume of licences issued is well in excessof estimated sustained yield. (MacDonaldand Young, 2000, p. 23)

In California’s Central Valley, groundwaterover-exploitation is a 60-year-old problem,yet in his case study of basin management,Svendsen (2000) concludes that:

Groundwater is the most lightly plannedand regulated segment of the state’s waterresources. There is little control overabstractions and, on average, the state isin a serious overdraft situation.

Even in middle-income countries, wheremajor institutional reforms have beeninitiated in recent years, groundwater over-exploitation has defied solution. Spain, oneof the European countries that suffer agri-cultural over-exploitation of groundwater,has instituted sweeping reforms that willaffect surface water but have little to dowith groundwater (Saleth and Dinar, 2000).

42 T. Shah et al.



Mexico’s aquifers too are amongst the mostover-developed. IWMI researchers based inGuanajuato state, one of Mexico’s agricul-turally dynamic regions, found water tablesin ten aquifers they studied declining ataverage annual rates of 1.79–3.3 m/yearduring recent years (Wester et al., 1999, p.9). An institutional solution is being triedhere through the establishment of AquiferManagement Councils called COTAS(Consejos Técnicos de Aguas). IWMIresearchers in Guanajuato are, however,sceptical: ‘. . . several factors bode illfor their (COTAS’) future effectiveness inarresting groundwater depletion . . .’.

Finally, for top echelons of nationaldecision makers, it is always easy to takehard decisions, which do not affect a largeproportion of a nation’s population in a seri-ously adverse manner. Political leaders andwater-sector leaders in emerging economiesconstantly face pressures to be myopic andadopt postures that are at odds with the idealof IRBM. The most powerful and compellingpressures emerge from their own internalsocial realities. In low-income agrariansocieties like in South Asia and much ofAfrica, food security and poverty alleviationwill continue to remain prime concerns fordecades to come.

When several poor states are involvedin a basin – such as India, Nepal and Bangla-desh in the Ganga–Meghana–Brahmaputrabasin, or the Central Asian states in theAral Sea – coordinating mechanisms tendto operate at suboptimal levels becausenational leaders are under pressure tomaximize their national interests. It hasbeen argued that the Aral Sea crisis isthe outcome of the compelling need ofthe political leaders in the Central Asianstates to ensure food security as wellas water-intensive cotton cultivation forexport, both at once. A major move to reversethe desiccation of the Aral Sea, the AmuDarya and the Syr Darya will have to waituntil something changes the dominantperception of the political leadershipin Turkmenistan and Uzbekistan thatcessation of cotton monoculture willhave politically and socially destabilizingconsequences.

3.6 Stage of SocioeconomicDevelopment

What factors might influence the pace ofinstitutional change in developing-countrywater sectors? Saleth and Dinar (2000)suggest that as water scarcity intensifies,opportunity costs imposed by missing ormalfunctioning institutions will increaseand transaction costs of institutionalchange will decline, which together willdetermine the pace of institutional changein developing countries. A competinghypothesis is offered by the application ofKuznets’ curve to natural resource manage-ment by societies. Recently, there have beenattempts to fit an environmental Kuznetscurve to deforestation using cross-countrydata (Bhattarai and Hammig, 2000).

The environmental Kuznets curve(EKC) poses an inverted U relationshipbetween economic growth and environ-mental degradation (Fig. 3.4). The corehypothesis is that, as economies grow,they use natural resources as a factorof wealth creation, but as per capita realincome grows, demand for environmentalamenity grows and there is greater demandand support for environmental protection.

Although the empirical results of someof this econometric work are far from con-clusive, intuitively, it seems compellingto suppose that the income elasticity ofdemand for environmental amenities islower at low per capita incomes (as inBangladesh and Burkina Faso) than at highper capita incomes. It follows, therefore, thathighly evolved economies of the Westernworld would have greater demand, capacityand collective will to fix the environmentalproblems resulting from natural resourcemismanagement than low-income emergingeconomies. In many Western countries,where per capita income growth to presentlevels took 200 years or more, the EKC effectalso took centuries to work out. Historicalevidence suggests deforestation in Europewas at its peak at the time of the IndustrialRevolution, and the area under forests beganto increase long after economic prosperityensued (Bhattarai and Hammig, 2000).




Much the same relationship seems tohold in the case of water resource manage-ment. Countries with highly developedwater institutions are also those that haveevolved industrially. In contrast, it is diffi-cult to find a low-income agrarian societythat has highly developed water institu-tions. Interestingly, some sketchy evidencesuggests that the period of decline followedby upswing gets telescoped in economieslike Japan and Taiwan that have grown theirindustrial output and employment rapidlyover a relatively short period.

In Table 3.1, we present the data set for57 countries organized around their percapita water and arable land availability.The figures alongside the country names aretheir respective per capita CO2 emissions,which is one of the best correlates of GDP percapita as well as the Human DevelopmentIndex. Mean per capita availability of waterand arable land, along with CO2 emission,are used to divide the countries into eightcategories.

Countries in categories B1, C1 and D1 arepoor in water and/or arable land resources;but these are rapidly becoming post-agrariansocieties where pressure on water and landfrom irrigated farming will rapidly ease. Thesocial and economic costs of fixing watermismanagement in these countries alreadyare or will soon be within acceptable limits.

It is notable that A1 represents the cate-gory of countries from which most models of

effective water institutions emerge, andthese are offered to countries in D2 categorywhich have the least water, land and CO2

emissions. Countries in A1 are amongst thebest endowed with both water and land. As aresult, despite being highly industrialized(as indicated by their high CO2 emissions),these still have large, wealth-creating agri-culture and agroindustry sectors that absorba very small proportion of their populations.

In the D2 category, poor land and waterresource endowments combine with highpopulation pressure, but ironically, theirmost critical problem is their low CO2 emis-sion. Industrial growth, urbanization andtransfer of people from agricultural tooff-farm livelihoods seems the only waypressure on land and water will ease. Manyof these countries will, over the comingdecades, more likely take the Kuznets-curveroute that Japan and Taiwan took than theone that Australia and the USA took.

In Taiwan, where rapid industrialgrowth and urbanization have resulted in a40% decline in irrigated areas over recentdecades, the popular outlook towardswater management issues has undergone afundamental transformation. Over 90% ofTaiwan’s irrigators have become part-timefarmers, and income from industrialemployment far outweighs agriculturalincomes. There have been major increasesin demand for environmental amenitiesand the touristic value of former irrigation

44 T. Shah et al.

Fig. 3.4. Relationship between level of economic growth and natural resource degradation.



structures. All these have resulted insubstantial private investments in improv-ing water quality and aquatic ecology.Taiwan has amongst the highest populationdensities we find anywhere in the world, yetits water institutions will soon approachthose in high-income Western countriesrather than low-income Asian countries,which share high population density withTaiwan.

The Kuznets curve hypothesis looksat the relationship only from the angle ofdemand for environmental amenities, butthere is also the supply side to it. Moreand higher-quality resources are appliedto natural resource management in high-income countries than in low-incomecountries. Consider the budget of the

water departments. The California StateDepartment of Water Resources has2000 employees, mostly professionals, whooperate an annual budget of US$1 billion(Svendsen, 2000). The Gujarat Departmentof Water Resources probably employs asmany engineers but operates a budget ofless than US$10 million. The upshot of thisdiscussion is that, over a decadal time frame,economic growth is probably both the causeof and the needed response to the problemof natural resource mismanagement, and, ifthe experience of Japan and Taiwan is anyguide, the period over which the interactionbetween the two plays out need not run intocenturies as it did in the case of Europe,but it can be telescoped from centuries todecades.


Per capita water > mean (10,460 m3);values of CO2 emissions/capita, countries

classified as high (1) or low (2)

Per capita water < mean (10,460 m3);values of CO2 emissions/capita, countries

classified as high (1) or low (2)

Per capitaarable land> mean(0.37 ha)

A1 A2 B1 B2

AustraliaCanadaUSANew ZealandHungary

4.613.735.372.181.59

ArgentinaBrazil

1.010.46

DenmarkGreeceKazakhstanLibyaSpain

2.952.092.892.181.6

SudanSyriaZambiaTurkeyAfghanistan

0.10.830.080.780.02

Per capitaarable land< mean(0.37 ha)

C1 C2 D1 D2

MalaysiaSwedenVietnam

1.581.671.77

CambodiaChileIndonesiaLaosMexico

0.010.920.330.021.02

SwitzerlandSouth AfricaSingaporeSaudi ArabiaOmanJapanItalyN. KoreaS. KoreaIsraelBelgiumFrance

0.682.15.323.881.852.531.922.463.072.52.861.69

ZimbabweTanzaniaSri LankaSenegalPhilippinesPeruPakistanNepalKenyaIndiaEthiopiaEgyptChinaBotswanaBangladeshIraqIranSyria

0.450.020.110.110.250.30.180.020.070.290.020.420.740.370.051.211.150.83

Source: Engelman et al. (2000).

Table 3.1. Natural resource availability and economic growth.



3.7 Conclusion

In this chapter, we have made anattempt to explore why efforts to transfer theinstitutional models of RBM from developedcountries to developing ones have not metwith the desired success. The contexts inwhich reforms are tried in developingcountries are vastly different – in theirhydrologic and climatic conditions, in theirdemographics, in their socioeconomic con-ditions and in the way their water sectors arecurrently organized – from the context of thecountries in which the models first suc-ceeded. Successful institutional reforms inthe water sector worldwide have tended tohave common over-arching patterns – theyhave focused largely on management of sur-face water bodies, they have aimed at improv-ing the productivity of publicly divertedlarge water bodies, they have largely ignoredgroundwater, and have not had to contendwith sizeable informal water sectors, andthey have focused on blue-water productiv-ity and largely ignored green water.

The problems that successful institu-tional models have resolved – poor waterquality, degraded wet lands, maintainingnavigation and dealing with occasionalfloods – are often not of paramount interestin developing-country contexts. The prob-lems that developing countries find criticaland insurmountable either have remainedunresolved in developed-country river bas-ins, e.g. groundwater over-exploitation, orare rendered irrelevant by their evolutionaryprocess, as with using irrigation as a meansto provide poor people with livelihoods andfood security. This does not by any meansimply that developed-country experiencehas no lessons to offer to the developingworld. Drawing such a conclusion would benaïve in the extreme. What it does mean,however, is that imposing institutionalmodels uncritically in vastly different socio-ecological contexts can be dysfunctionaland even counter-productive.

What it also means is that we need totake a broader view of institutional change.An extraordinary aspect of the institutionaldiscussion in the global water sector is howvery narrowly it has focused on things that

governments can do: make laws, set upregulatory organizations, turn over irrigationsystems, specify property rights. A recentreview of institutional changes in the globalwater sector in 11 countries by Saleth andDinar (2000), for example, treats water law,water policy and water administration asthe three pillars of institutional analysis.This makes water purely the government’sbusiness, quite contrary to the slogan popu-larized by the World Water Council to ‘makewater everyone’s business’. If institutionalchange is about how societies adapt to newdemands, its study has to deal with morethan what just the governments do. People,businesses, exchange institutions, civil soci-ety institutions, religions and movements– all these must be covered in the ambitof institutional analysis (e.g. Mestre, 1997,cited in Merrey, 2000, p. 5).

Which elements of the Murray–Darlingexperience can be sensibly applied in whichdeveloping-country context is certainly animportant and interesting analytical enter-prise, but equally, or even more important isthe need to listen to voices from the grass-roots. If people living, for example, in theDeduru Oya basin in Sri Lanka are facingwater scarcity, they are going to begin to dosomething about it. Likewise, if the govern-ment of South Africa withdraws from themanagement of smallholder irrigationschemes in the Olifants basin, the small-holders will soon respond in some way.What institutional reform makes best sensein Deduru Oya or Olifants should bestemerge from understanding the respectiverealities of these basins. A broad under-standing of what has worked elsewhereincluding in the developed world mightoffer a good backdrop to the design ofinstitutional interventions, but it might beunrealistic to expect much more. Copycatinstitutional reform would be outrightdisastrous.

In understanding how societies adapttheir institutions to changing demands,Nobel Laureate Oliver Williamson (1999)suggests the criticality of four levels of socialanalysis as outlined in Fig. 3.5. The top levelis referred to as the social embeddednesslevel where customs, traditions, mores and

46 T. Shah et al.



religion are located. Institutions at this levelchange very slowly because of the spontane-ous origin of these practices in which‘deliberative choice of a calculative kind isminimally implicated’. At the second level –where the institutional environment of asociety is involved – evolutionary processesplay a big role; but opportunities for designpresent themselves through formal rules,constitutions, laws, property rights. Thechallenge here is getting the rules of thegame right. The definition and enforcementof property rights and contract law arecritical features here. Also critical is under-standing how things actually work, ‘wartsand all’, in some settings but not in others.

However, it is one thing to get the rulesof the game (institutional environment)right. It is quite another to get the play ofthe game (enforcement of contracts/propertyrights) right. Which leads to the third levelof institutional analysis: transaction costsof enforcement of contracts and propertyrights, and the governance structuresthrough which this is done. Governance –through markets, hybrids, firms, bureaus– is an effort to craft order, thereby tomitigate conflict and realize mutualgains; good governance structures craftorder by reshaping incentives, which leadsto the fourth level of social analysis – gettingthe incentives right.


Embeddedness:informal institutions,customs, traditions,norms, religion

Governance: play of thegame-especially contractenforcement; aligninggovernance structureswith transactions

Resource allocation andemployment: incentivealignment

L1

L2

L3

L4

FrequencyLevel

Continuous

Non-calculative,spontaneous

Purpose

Get the institutionalenvironment right

Get the governancestructure right

Get the prices right

F1: Social theoryF2: Economics of property rights and positive political theoryF3: Transaction cost economicsF4: Neo-classical economics, principal/agent theory

Source: Oliver E Willamson. 1999. The new Institutional Economics: TakingStock/Looking Ahead , Business and Public Policy Working paper BPP-76, University ofCalifornia, Berkeley.

Fig. 3.5. Four levels of institutional change that explain how societies adapt to new demands.



Discussion of water policy and institu-tions in the developing-country context hasfocused a great deal on levels 2, 3 and 4and little on level 1. More, it has tended tounderplay the interactions between levels.Many populous developing countries willfeel a lot wiser about IRBM if we learn moreabout how level 1 operates in their respec-tive contexts and how the interactionbetween 2 and 3, and 3 and 4 can workbetter. How to create property rights thataffect users’ behaviour is more importantthan exhortations that clear property rightsbe created. Understanding how to enforce agroundwater law meaningfully on 20 mil-lion private pumpers scattered throughoutthe South Asian countryside is more helpfulthan pushing a groundwater law. How tomonitor water use and collect canal irriga-tion charges cost-effectively is more in orderthan discussing whether irrigation subsidiesshould be eliminated.

Acknowledgements

This work is supported by a Research Grantfrom BMZ (Germany’s Ministry of Eco-nomic Co-operation) to IWMI to studyInstitutional Support Systems for WaterManagement in Water-stressed River Basins.

The authors gratefully acknowledge thecontributions of colleagues in developingthese ideas: Matsuno for supplying litera-ture on water management institutions inJapan; Hilmy Sally for help with an Excelchart; David Molden for discussions onwater accounting; Randy Barker for drawingattention to Williamson’s excellent article;Madhusudan Bhattarai, from whose workthe ideas on the environmental Kuznetscurve are drawn; Frank Rijsberman for astimulating discussion in a car ride inHyderabad on what are the meaningfulissues in the debate on water pricing; TissaBandaragoda and M. Samad for many inter-esting discussions on basin managementresearch; and Doug Merrey for introducingthe author to the field of IRBM. The authorsown the responsibility for the argument,with all its limitations and flaws.

References

Agarwal, A. (2000) Drought? Try Capturing theRain. Centre for Science and Environment,Delhi.

Bhattarai, M. and Hammig, M. (2000) Institutionsand the environmental Kuznets Curve fordeforestation: a cross-country analysis forLatin America, Africa and Asia. WorldDevelopment, forthcoming.

Birch, A. and Taylor, P. (1999) Internationalmentoring; application of Australianexperience for Sri Lankan water sectorreforms under Technical Assistance ofthe Asian Development Bank. WaterInternational 24, 329–340.

Engelman, R., Richard, P., Bonnie Dye, C.,Gardner-Outlaw, T. and Wisenewski, J.(2000) People in the Balance: Populationand Natural Resources at the Turn ofthe Millennium. Population Action Inter-national, Washington.

Global Water Partnership (2000) Integrated WaterResources Management. Technical AdvisoryCommittee of GWP, Stockholm.

Hu, X. (1999) Integrated catchment managementin China: application of the Australianexperience.Water International 24, 323–328.

Hunt, C. (1999) Transposing of water policiesfrom developed to developing countries: thecase of user pays. Water International 24,293–306.

Jacobs, J.W. (1999) Comparing river basin develop-ment experiences in the Mississippi and theMekong. Water International 24, 196–203.

MacDonald, D.H. and Young, M. (2000) A casestudy of the Murray–Darling Basin: draft pre-liminary report. Prepared under commissionfor IWMI. CSIRO, Sydney.

Malano, H., Bryant, M. and Turral, H. (1999)Management of water resources: canAustralian experiences be transferred toVietnam? Water International 24, 307–315.

Merrey, D. (2000) Creating institutional arrange-ments for managing water-scarce riverbasins: emerging research results. Paperpresented at the session on ‘Enough Waterfor All’, at the Global Dialogue on the Roleof the Village in 21st Century: Crops, Jobs,Livelihoods, 15–17 August, Hanover.

Molden, D. and Sakthivadivel, R. (1999) Wateraccounting to assess uses and productivity ofwater. Water Resources Development 155,55–71.

Mostert, E., van Beek, E., Bouman, N.W.M., Hey,E., Savenije, H.H.G. and Thissen, W.A.H.

48 T. Shah et al.



(1999) River basin management andplanning. Keynote paper for InternationalWorkshop on River Basin Management, TheHague, 27–29 October, 1999.

Saleth, M. and Dinar, A. (2000) Institutionalchanges in global water sector: trends,patterns and implications. Water Policy 2,175–199.

Shah, T. (1993) Water Markets and IrrigationDevelopment: Political Economy and Practi-cal Policy. Oxford University Press, Bombay.

Shah, T. (2000) Mobilizing social energy againstenvironmental challenge: understanding thegroundwater recharge movement in WesternIndia. Natural Resource Forum, August.

Shah, T., van Koppen, B., Merrey, D., de Lange, M.and Samad, M. (2000) Institutional alter-natives in African small holder agriculture:lessons from international experience inirrigation management transfer. Paperpresented at National Policy Workshopon Irrigation Management Transfer andRehabilitation of Small Holder Irrigation

Schemes, 21 June, Hazy View, Mpumalanga,South Africa.

Sutcliffe, J.V. (1995) Hydrology in the developingworld. In: G.W. Kite (ed.) Time and theRiver: Essays by Eminent Hydrologists. WaterResources Publications, Saskatoon.

Svendsen, M. (2000) Basin management in amature closed basin: the case of California’sCentral Valley. In: C.L. Abernethy (ed.) Inter-sectoral Management of River Basins. IWMI,Colombo, Sri Lanka and DSE, Feldafing,Germany.

Wester, P., Pimentel, B.M. and Scott, C.(1999) Institutional responses to ground-water depletion: the aquifer managementcouncils in the State of Guanajuato. Paperpresented at the International Symposium onIntegratedWater Management in Agriculture,Gomez Palacio, Mexico, 16–18 June.

Williamson, O.E. (1999) The new institutionaleconomics: taking stock/looking ahead.Business and public policy. Working PaperBPP-76, University of California, Berkeley.




4 Making Sound Decisions: InformationNeeds for Basin Water Management

Martin Burton and David Molden

4.1 Introduction

This chapter presents an analysis of infor-mation needs in water-scarce river basinsfrom two perspectives – those of the basinmanager and of the manager of a locallycontrolled irrigation system (LCIS). Infor-mation is central to planning and managingthe water resource within a river basin.Information requirements develop andgrow over time as pressure on the waterresource increases with demographic andeconomic changes (see Chapters 1 and 2).Information plays a crucial role in quantify-ing the abundance and quality of the avail-able resource and the demands placed on it,and is central to decision making on waterallocation at both scheme and river basinlevel, and to formulation of policies andstrategies for river basin development.

As water becomes scarcer, data andinformation become increasingly importantfor irrigation system managers, since theneed grows to make the most productive useof available supplies. In addition to lookingat the internal allocation and management ofthe available water, managers of irrigationsystems must also look externally to ensurethat their systems acquire and retain theirshare of the available resource. As irrigationis often the largest consumptive user ofwater in a river basin, irrigation managersmust often justify the use, efficiency andproductivity of water in competition andcomparison with other uses and users. In

this context, data and information play acentral role.

4.2 Use and Users of Information

There is a wide variety of uses and users ofinformation related to water resourceswithin a river basin (Table 4.1). The riverbasin can be divided into water uses relatedto watershed land use, in-stream water use,extractive water use and environmentalwater use. The in-stream uses are thosewherein the water is used without beingremoved from the river, such as hydro-power or recreation. They generally resultin minimal depletion of available water.Extractive uses such as irrigation, on theother hand, involve removing water from ariver or aquifer, and generally result in largerdepletion of the resource from the basinthrough evapotranspiration. Removal ofwater from a river or aquifer is not absolute,however. For example, most of the waterused for domestic and industrial purposesis available for reuse, either after treatmentor through dilution in fresh water.

Information uses are many and varied,and can be categorized into developmentand master planning, operational manage-ment, water sharing and allocation, andresearch. Data collected for one purpose,such as operational management, may be ofuse for other purposes, such as development




52 M. Burton and D. Molden

Water usesWatershed water uses

In-stream water uses

Extractive water uses

Environmental wateruses

• Lakes/reservoirs• Forests• Natural vegetation• Hydropower• Recreation• Navigation• Fisheries• Irrigation (surface/groundwater)• Potable water (surface/groundwater)• Industrial water, including mining (surface/groundwater)• Aquatic, wetlands and floodplain environment and ecology• Drainage disposal• Waste dilution and disposal• Repelling salinity intrusions• Erosion control

Information usesDevelopment and

master planningWater sharing and

allocation

Operationalmanagement

Research

• Planning and forecasting• Decision making in relation to resource development and protection• Resource management and allocation• Allocation of water rights• Rule formulation• Pricing• Dialogue with, and amongst, users• Flow control and regulation• Flood control, protection and warning• Effluent control• Monitoring and evaluation (abstractions, effluent levels, environment, etc.)• Infrastructure asset management• Conflict resolution• Water resources, irrigation, environment, ecology, etc.

Information usersGovernment

Regulatory andmanagementauthorities

Companies, groupsand associations

Individuals

• Ministries of: Water Resources, Irrigation, Agriculture and Livestock,Energy, Hydrology and Meteorology, Health, Environment and NaturalResources, Fisheries, Forestry, Navigation and Marine Transport,Planning and Development

• Legislatures• State, regional or local government• Municipalities• River boards, river basin councils, drainage boards• Regulatory bodies (rivers, groundwater, environment, etc.)• Courts• Industry (manufacturing, services, mines, forestry, etc.)• Associations (irrigation, rural water supply, environmental lobbies, etc.)• Universities, research centres and training centres• Development agencies and agents• Non-governmental organizations• Domestic household users• Irrigation farmers• Livestock owners• Recreators

Table 4.1. Typical water uses, information uses and users within a river basin.



and master planning. Often, however,requirements in different applications differenough that data collected for one purposeare not usable in another.

Information users can be subdividedinto government ministries and agencies,authorities, groups of individuals (such as ina company or an association) and individu-als. The relationships that exist with a givenriver basin between government agencies,groups and individuals are a function of theinstitutional framework that has evolvedover time. The nature of these relationshipsgoverns the manner in which data are sharedand used.

Data needs evolve over time to suit thechanging needs of a developing basin. Thedrivers for change (population growth,agricultural and industrial development,concern for the environment, increasingdemocratization, etc.) lead to changes inwater use, water quality and institutions,which in turn govern the type and extentof data collected, processed, analysed anddisseminated, with new needs building onexisting practices. As a democratic societymatures, power and decision making aretransferred to a wider community. Informa-tion processes become more demand driven,with those responsible for collecting andprocessing data having to become moreresponsive to the data users. The nature ofthe transaction between the provider and thedata user may be hierarchical, commercial,informal or statutory, or it may not be clearlydefined. As time passes, these relationshipsneed to become more clearly and closelydefined if the best use is to be made of theavailable resource.

4.3 Information System Tasks

Information system tasks range from datacollection, through processing and analysis,to storage and retention (Fig. 4.1). A furtherstage, which is becoming increasinglyimportant, and which is relevant to LCISs,is that of disseminating both data and the

findings arising from that data. In somecountries, this stage has been reached. Inothers, data are not widely disseminatedoutside the agency that collects and pro-cesses the data. In these latter instances,access to data by third parties is sometimesimpaired.

The following sections provide a briefoverview of the information system tasksnecessary for both river basin managementand managing LCISs. A distinction is madebetween data and information, with databeing a series of observations or measure-ments, and information being the resultsderived from processing the data. Thus, thesame data can be processed in different waysto provide information for different uses orusers.

4.3.1 Formulation of an information strategyand management information system

It is essential for managers of any system, beit a river basin or an irrigation scheme, toformulate a management information strat-egy and from that to develop a managementinformation system. The information strat-egy is the pattern or plan that integratesthe organization’s information needs into acohesive whole to meet the organization’smanagement requirements. It deals withthe information needs, and the resourcesrequired for their collection, processingand use. The system is the collection ofprocedures that implement this strategy.The strategy and associated managementinformation system should address thefollowing questions:

• What are the key managementprocesses?

• What data and information are neededfor managing these processes?

• What are the key external factors affect-ing the enterprise, and what data needto be collected to monitor these overtime?

• How, and by whom, will data be col-lected, processed, analysed and used?

Information Needs for Basin Water Management 53



• How much will the data collection,processing, analysis and disseminationcost?

• What quality control measures will berequired and used?

• How transparent will the informationsystem be?

• To whom will data be disseminated,what data will they require and in whatform?

Key management processes in a river basingrow directly from the functions performed.Chapter 1 presented a set of generic riverbasin management functions and a method-ology for analysing their performance,

which can be used to guide a detailingof key management processes. Informationneeds for management will depend on suchfactors as the size of the basin, its stage ofdevelopment, the diversity of water usersand prevailing institutional arrangements.An important potential external factor is apolitical demand to transfer water outsidethe basin or to develop particular uses, suchas expanded irrigated agriculture, withinthe basin. The management informationsystem developed will depend, amongother things, on whether there is one riverbasin management authority, or severalseparate agencies with responsibility forthe management/oversight of the water


Fig. 4.1. Structure of a management information system.



resource. An information strategy and itsassociated management information systemare dynamic and must change over timeto match social and political changeswithin society (hence the question ‘Is theinformation system adequate?’ in Fig. 4.1).

For LCISs, the management processesare relatively straightforward, with databeing required in relation to water demands,cropped areas and irrigation fees paid. Thesedata are generally fairly easy to collect andare limited in geographical extent, com-pared with data for the basin as a whole.External data required relate to rainfalland river flows elsewhere in the riverbasin, and water abstractions by otherusers, especially those upstream. Such datamay be collected by a government agencyor by other irrigation scheme managers.However, gaining access to these data byLCIS managers may be difficult.

Information needs will vary with thestage of development of the water resource,and reflect the level of technology andexpertise available. It is crucial that the man-agement information system ‘fits together’from start (purpose) to finish (decision mak-ing), with systems designed to be congruentwith local capabilities and resources at allstages. It is preferable to establish a systemthat is robust and provides relatively simple,but useful, information, than to establishan ‘ideal’ one which is over-ambitious andunsustainable. The focus needs to be onthe data and information needs of all theactors involved in irrigation system or riverbasin management, leading to a demand-ledinformation system.

4.3.2 Data collection

Once data needs have been determined dur-ing the strategy formulation, attention shiftsto data collection. The two key dimensionsare time and space, and data collectionneed to be distributed throughout bothdimensions. Decisions on where to measureand the duration and frequency of measure-ment should be established when designingthe management information system.

Data-collection programmes can beregular, periodic or one-off dependingon needs. Regular data collection generallyincludes river flows, rainfall and water qual-ity measurements. Periodic data collectionmight include peak or low flow measure-ments, whilst one-off data collection mightrelate to a time-bound project where infor-mation is required to make decisions for theproject. Data-collection procedures can bedifferent in these cases. One feature of aregular data-collection programme is thatstandard pro forma are used to record thedata. Different degrees of certainty can beattached to these different types of data.Measured data are often more reliablethan reported or estimated data, whileobserved data may be more dependent onthe bias of the observer.

Types of data to be collected comprisethe following:

• Measured – data obtained from directmeasurements, e.g. discharge fromdepth gauge readings at a measuringstructure;

• Observed – information used in aqualitative mode based on fieldobservations, such as canal condition,activities of organizations, etc.;

• Estimated – data used in a quantitativemode which are derived indirectlyby calculation using combinations ofmeasured variables (such as potentialevapotranspiration using the Penman–Monteith equation), or which are esti-mated without direct measurement;

• Reported – data obtained from second-ary sources, using other peoples’observed, estimated or measured data.

Two major concerns in data collection arereliability and accuracy. The reliability ofthe data relates to the degree of assurancethat the data are measured and not madeup. The accuracy relates to how close mea-sured data are to the ‘true’ value of theparameter. If data are to be used for manage-ment and decision making, it is imperativethat the data be reliable. Data collection canbe a tedious and time-consuming task forthose collecting the data, and checks haveto be made to ensure that personnel




responsible for these tasks carry them outreliably. It is not uncommon for data to bemade up by data collectors to avoid thechore of travelling to a recording site. Accu-racy of data collection can be enhancedthrough trained personnel periodicallychecking measuring sites (river gauges,rainfall stations, climate stations, etc.) toensure that they are functioning correctly,and that data are being collected in the cor-rect manner, or by improved measurementtechnology. Generally, increased accuracyis associated with increased costs.

4.3.3 Storage, retention and retrieval of data

Often overlooked until data are requiredand found to be lost or missing are proce-dures for storage, retention and retrieval ofdata. The management information systemshould clearly indicate which data shouldbe stored, in what format and for how long.At higher management levels, summaries ofthe data can be stored, whilst at lower levelsthe complete data should be retained.

Data storage should be systematic andfiles clearly labelled. The data may be in theform of sheets of paper, in which case it mustbe kept in a secure, dry location. Computerdata must always be backed up on floppydisk or CD, and the disks kept in a securelocation separate from the computer.

Geographical information systems(GISs) are a valuable way of both present-ing and storing data, with the significantadvantage that the data can be retrievedin different combinations or formatsdepending on the need.

4.3.4 Data processing and analysis

The procedures for data processing andanalysis should be identified as part of theinformation strategy. It is the use to whichthe data will be put that drives the data-collection exercise; thus, how the datawill be processed and analysed should beestablished at the outset. For regular datacollection of river flows, rainfall, water

quality and the like, data processing andanalysis procedures are well established.For other processes, such as performanceassessment, procedures are still evolving.

The widespread availability of power-ful small computers and software packageshas transformed data handling, processingand analysis. Using spreadsheet programs,one is able, relatively simply, to set up pro-cedures for recording, storing, processingand presenting data. Particularly useful isthe ability to represent data graphicallyto show events, trends and relationships.Statistical analysis packages such as SPSScan often access data stored in spreadsheetformat and process it to present informationon trends and relationships among data.

An example of the value of trend analy-sis using data collected over several years isdiscussed in Chapter 8, drawn from study ofthe Lerma–Chapala river basin, showing thelevel of Lake Chapala, the receiving body atthe tail of this basin. As a consequence ofan extensive network of data-collectionstations throughout this basin, an importantlong-term set of data is available for analysisand presentation, leading to better-informeddecision making related to the managementand development of the basin. The role andimportance of these data has increased as theavailable water resource has become fullyutilized.

4.3.5 Reporting and dissemination of data

The reporting and use of data should beidentified as part of the information strat-egy, and tailored to the target audience. Forwater resources specialists, tables and fig-ures of river discharges and rainfall patternswill be of interest. Politicians, on the otherhand, will want summaries and a discus-sion of the implications of the data, withkey issues highlighted.

With the advent of computers andthe Internet, wider dissemination of databecomes possible, even with limited bud-gets. Data can be held on a central computerand accessed through the Internet, withpasswords allowing different levels of




access to different users. IWMI’s WorldAtlas is an example of such an Internet-based system (http://www.iwmi.cgiar.org/WAtlas/atlas.htm). In addition, the abilityto copy data on to CDs, which can then bedistributed to interested parties, means thatcheap access to data is possible for those thatdo not have ready access to the Internet.

There is an old adage that knowledgeis power, and in many countries, access todata is still restricted, either as a governmentpolicy, or as an organizational or individualpractice. In some locations, data are col-lected by government agencies and then soldto other users. This selling of data can beofficial, with clearly spelt out policies andprices, or unofficial with individuals usingtheir position to gain additional income.

4.3.6 Data quality control

Quality control on water resource datais often undervalued and inadequatelyimplemented. Quality control checks canand should be built into data-collection,

processing, analysis and reporting pro-cesses, and personnel responsible for theseprocesses trained in monitoring quality andidentifying suspect data. As mentionedpreviously, supervision of data collectionis important, as is explanation to data-collection personnel of the importance ofaccurate and reliable data, and the use towhich the data will be put.

4.4 Basin Stages and ChangingInformation Needs

In tandem with the development of ariver basin, the collection and use of datachanges. Figure 4.2 depicts the changingfocus for data and information needs asbasins mature.

As time passes, quality and regulatoryissues come to the fore, and information isrequired by a wider audience. The data setrequired for river basin managementbecomes both more extensive and moredetailed, with increased needs for support-ing processes, procedures, personnel and


Quantify how much water is available

Identify how the water resource can be used

Quantify how the water resource is being used

Allocate available resources between competing users/uses

Ascertain the quality and sources of degradation

Model and manage water at the basin level

Regulate use and prosecute over-abstraction and/or pollution

Expand models to encompass environmental and ecological needs and processes

Tim

e

Incr

easi

ng s

carc

ity o

f wat

er

Incr

easi

ng im

port

ance

Incr

easi

ng d

etai

l

Resolve conflicts over allocation and use

Form basin-wide consensus and agreement on allocation and use

Reallocate water amongst (established) users

Fig. 4.2. Changing focus for data and information over time within a river basin.



finances to match. As the basin develops,information is collected and used for model-ling – initially of the hydrological andhydraulic processes and later for environ-mental and ecological processes. Withincreasing water scarcity, conflicts overwater use increase, and data are increasinglyemployed in the resolution of conflicts.With increasing incidence or scale ofconflict, there comes a time when steps arerequired at the basin level to form a consen-sus on individual, corporate, societal andenvironmental rights to access and use thewater resource, following which there maybe a need for reallocation of water amongstuses and users.

As discussed in Chapter 2, three phasescan be identified in the developmentand growth of a river basin: development,utilization and reallocation. Informationand data requirements in each phase vary,and are summarized in Table 4.2. Thetable shows information needs, providesexamples of the data required and showsthe developments that typically take placein information processes during each phasein response to developments within theriver basin.

In the early development stages, muchof the data collection is project based, witheach project establishing data-collectionsystems to suit its particular needs. Thedata-collection needs will follow the needsof the project cycle: planning, design,construction, management, operation andmaintenance, and possibly later, rehabilita-tion. At this stage, there is relatively limitedhuman control over the natural environ-ment, with modest physical infrastructureand localized agreements on flood controland water use.

As time passes, these data systemsare integrated by one or another of thegovernment agencies, and the data networkbecomes more systematized. At this stage,data collection and processing may bedriven by established routines and notnecessarily by needs. As pressure on theavailable water resource increases, the needfor data grows, and the data system becomesmore refined with more data-collectionpoints being established and the data

collection, processing and analysis ofthe data becoming more sophisticated.Hydrologists develop computer-based waterresource models, which, in turn, begin todrive data collection so that the predictivecapabilities of the models can be fullyutilized.

Over time, the focus shifts from attempt-ing to supply water to match increasingdemand to demand management. During thedevelopment, utilization and reallocationphases, the approach to data changes, fromone in which the data are collected, pro-cessed and used by single agencies for theirown purposes to a situation where the dataare disseminated and made available to awide range of stakeholders. The power thatdata ownership holds is shared amongst thevarious stakeholders to achieve the widerobjective of consensus and buy-in to waterresources development and management.During the reallocation phase, the focus ison ensuring that the various users limit theirabstractions to the quantities licensed, andmaintain their effluent discharges withinthe specified water quality standards.

Potential interventions to addressincreasing water shortage vary accordingto the development stage of the river basin.At each stage, the solution revolves aroundinformation, as shown in Fig. 4.3. For eachproblem situation, the nature of the problemis formulated, and data collected and ana-lysed. Solutions can take many forms,involving alternatively, or in combination,construction, legislation, enforcement ofexisting legislation, improved managementor empowerment of organizations orindividuals.

4.5 Basin Level Information Needs

The types of basin level informationrequired change both with the phase ofdevelopment and with the perception ofwhat river basin management comprises.Early guidelines on river basin managementtend to emphasize the collection of techni-cal data (climate, river flow, catchmentarea, land use, etc.), whereas more recent





Pha

seD

ata

need

sT

ypic

alda

taco

llect

edD

evel

opm

ents

inin

form

atio

npr

oces

ses

Infa

ncy

Loca

lized

use

only

Rud

imen

tary

,lim

ited

tow

ater

leve

lsan

dex

tent

offlo

odin

gF

lood

wat

erle

vels

,flo

oded

area

s(t

hrou

ghex

perie

nce)

Dem

arca

tion

(and

avoi

danc

e)of

flood

edar

eas,

corr

elat

ion

offlo

odex

tent

and

flood

leve

ls

Dev

elop

men

t•

Wat

eral

loca

tion

issu

pply

focu

sed

•D

ata

colle

cted

and

used

bysm

all

num

ber

ofag

enci

esfo

rsp

ecifi

cus

esan

dpr

ojec

ts

•A

vaila

bilit

yof

wat

erdu

ring

the

year

and

exte

ntof

agric

ultu

rall

and

•M

ain

focu

sis

onsu

rfac

ew

ater

,th

ough

som

ein

tere

stin

grou

ndw

ater

for

urba

nan

dirr

igat

ion

deve

lopm

ent

•F

orin

itial

plan

ning

for

river

basi

nde

velo

pmen

t

•P

roje

ct-w

ise

colle

ctio

nof

river

flow

and

qual

ityda

ta•

Clim

atic

data

,par

ticul

arly

rain

fall

•La

ndus

ein

river

ine

plai

nsan

dex

tent

ofag

ricul

tura

llan

d•

Top

ogra

phic

surv

eys

•A

eria

lpho

togr

aphy

•La

ndow

ners

hip,

trad

ition

al/e

xist

ing

wat

errig

hts

•In

itial

data

colle

ctio

nsy

stem

ses

tabl

ishe

dfo

rin

divi

dual

proj

ects

;gra

dual

lyth

ese

are

linke

dup

and

coor

dina

ted

byth

ede

velo

pmen

tage

ncy(

s)•

Bas

in-w

ide

hydr

omet

ricst

atio

nses

tabl

ishe

dto

gath

erba

seda

ta

Util

izat

ion

•W

ater

allo

catio

nis

supp

lyfo

cuse

d•

Dat

are

late

dpr

oces

ses

and

proc

edur

esw

ell

esta

blis

hed

•D

etai

led

know

ledg

eof

the

avai

labl

ew

ater

reso

urce

s,bo

thsu

rfac

ean

dgr

ound

wat

er,p

artic

ular

lyov

er-y

ear

toes

tabl

ish

stor

age

patte

rns

for

rese

rvoi

rsan

dre

char

gepa

ttern

sfo

rgr

ound

wat

er.

•F

orriv

erba

sin

mas

ter

plan

ning

•R

iver

flow

data

thro

ugho

utth

eba

sin

•C

limat

icda

tath

roug

hout

the

basi

n•

Land

owne

rshi

pan

dtr

aditi

onal

/ex

istin

gw

ater

right

s•

Gro

undw

ater

leve

land

qual

ity•

Som

em

onito

ring

ofpo

llutio

nle

vels

•D

ata

colle

ctio

npr

oced

ures

stan

dard

ized

and

coor

dina

ted

•P

roce

dure

ses

tabl

ishe

dfo

rm

onito

ring

pollu

tion

leve

ls•

Pro

cedu

res

esta

blis

hed

for

mon

itorin

ggr

ound

wat

erde

pth

and

qual

ity•

Pub

licat

ion

ofw

ater

reso

urce

san

dcl

imat

icda

ta•

Dev

elop

men

tofs

impl

ew

ater

reso

urce

sm

odel

sfo

rriv

erba

sins

cont

inue

d

Tab

le4.

2.D

ata

requ

irem

ents

rela

ted

tode

velo

pmen

tpha

ses

ofriv

erba

sins

.




Pha

seD

ata

need

sT

ypic

alda

taco

llect

edD

evel

opm

ents

inin

form

atio

npr

oces

ses

Re-

allo

catio

nan

dre

stor

atio

n•

Dem

and

and

supp

lyfo

cuse

d•

Dat

are

late

dpr

oces

ses

and

proc

edur

esre

fined

and

mor

ew

idel

ydi

ssem

inat

ed

•T

oob

tain

deta

iled

know

ledg

eof

the

annu

alan

din

ter-

year

wat

erre

sour

cesi

tuat

ion

both

for

supp

lyan

dde

man

d•

To

mon

itor

and

cont

rolw

ater

abst

ract

ion

byus

ers

•T

om

ake

proj

ectio

nsof

supp

lyan

dde

man

d•

For

wat

erre

sour

ces

mod

ellin

g,us

ing

rem

ote

sens

ing

and

GIS

•F

orsc

enar

ioan

alys

is•

For

river

basi

nm

aste

rpl

anni

ng•

To

refin

ean

dup

date

supp

lyan

dde

man

dpr

ojec

tions

,sce

nario

anal

ysis

•T

ofo

rmul

ate

rule

sfo

ral

loca

tion

ofw

ater

durin

gdr

ough

ts/s

hort

ages

•R

iver

flow

and

wat

erqu

ality

data

thro

ugho

utth

eba

sin

•C

limat

icda

tath

roug

hout

the

basi

n•

Gro

undw

ater

leve

land

qual

ity•

Pol

lutio

nle

vels

•W

ater

abst

ract

ion

byal

luse

rs•

Dat

afo

rpr

osec

utio

nfo

rov

er-a

bstr

actio

nan

d/or

pollu

tion

•D

ata

anal

ysed

from

pers

pect

ive

ofdi

ffere

ntw

ater

user

s•

Wat

erne

eds

for

vario

usen

viro

nmen

talp

roce

sses

•H

ydro

met

ricne

twor

kex

tend

edan

dau

tom

ated

for

dire

cttr

ansm

issi

onto

data

colle

ctio

nst

atio

ns•

Gro

undw

ater

mon

itorin

gne

twor

kex

tend

ed•

Pol

lutio

nm

onito

ring

exte

nded

•F

urth

erco

mpu

teriz

atio

nof

data

colle

ctio

n,pr

oces

sing

and

anal

ysis

•D

evel

opm

ento

fsop

hist

icat

edw

ater

reso

urce

mod

els

for

river

basi

ns,w

ithre

finem

entt

obe

com

ean

oper

atio

nalt

ool

•R

emot

ese

nsin

gin

corp

orat

edin

tow

ater

man

agem

enta

ndde

cisi

onm

akin

g•

Pub

licat

ion

ofw

ater

reso

urce

ssu

pply

and

dem

and

info

rmat

ion

•A

naly

sis

and

pres

enta

tion

ofda

tafo

ra

wid

erra

nge

ofst

akeh

olde

rs•

Sce

nario

anal

ysis

toen

able

part

icip

atio

nin

deci

sion

mak

ing

Tab

le4.

2.C

ontin

ued.



publications recommend a broader collec-tion of data encompassing economic,institutional, social and environmentalfeatures of the basin and society in general.An example is found in the guidelinesdeveloped by the Asian DevelopmentBank following an extended process ofconsultation and discussion (ADB, 1996).The guidelines list seven key strategies forimproving water resources management.

• Prepare and adapt a national waterpolicy and action programme;

• Invest to manage the country’s priorityriver basins;

• Increase the autonomy and account-ability of service providers;

• Develop incentives, regulation andawareness for sustainable water use;

• Manage the use of shared waterresources and develop cooperation;

• Enhance water information, consulta-tion and partnerships;

• Invest in capacity building, monitoringand learning.

These seven strategies are clearly informa-tion and data intensive. A key process iscompiling National Water Sector Profiles.The broad content of these profiles is pre-sented in Box 4.1, and encompasses a rangeof items from general development goalsand water resources policy statements, tolegal framework, to the physical resourcebase, demonstrating the holistic approachthat is a central feature of integrated waterresources management. Whilst the ADB’swater sector profiles are country based,many of the profile categories and datarequirements are relevant at the river basinlevel as well.

Some key data requirements for riverbasin management are presented in Box 4.2.A list of leading indicators for describingriver basins is presented in Table 4.3. Collec-tion and presentation of these data allowcomparative assessment of different riverbasins, enabling the relative level of devel-opment to be assessed. As can be seen fromthe data presented in Table 4.3, the threebasins are at different stages of development.


Formulate nature of problem situation

Collect data

Construct

Monitor and evaluate

Implement

Make decision

Analyse data

Legislate Enforce Manage Empower

Fig. 4.3. Role of information in formulating solutions to increasing water scarcity.



The East Rapti basin has more than enoughwater resources to match current needs, theGediz basin is closing, with pollution amajor current issue, whereas the Lerma–Chapala basin is closed and over-exploited.

The heart of any basin managementexercise is a water budget for the basin(Molden, 1997). A water budget requirestime series data on water entering, beingutilized and leaving the basin, as well asinformation on storage and storage losses.

The budget must include both ground-and surface water. Preparing a basin waterbudget requires physical data from anetwork of hydrologic, meteorological andgroundwater stations within the basin.These networks often take years to establish,and a number of years of data are required toachieve a minimum level of reliability in theresults. In some cases, these networks willhave been established through externallyfunded projects, and finding adequate fundsto maintain the networks may have been aproblem. Whilst it takes time and effort toestablish these systems, particularly thetraining of staff to collect, process andanalyse data, they can deteriorate relativelyquickly if funding is inadequate or if soundmanagement is lacking.

Remote sensing is increasingly beingused for river basin management. Kite andDroogers (2000) used remotely sensed datato construct an integrated river basin modelfor the Gediz river basin in western Turkey.For this study, they compiled models atthree scales: the field scale, the irrigationscheme scale and the basin scale. Usingpublic domain datasets as well as localarea datasets, they constructed these modelsbased on two categories of data – areal andpoint data. Areal data include topography,land cover, leaf area index (LAI) and soilcharacteristics. Point data include climate,streamflow and operational rules for damsand regulators. Kite and Droogers point outthat two changes in methods of data accessand data collection can be observed in recentyears. First, data are increasingly availablefrom global datasets, and second, more andmore data are collected by remote sensing(RS) instead of conventional ground-basedtechniques (Fig. 4.4).

For their model, Kite and Droogerscollected and used data as summarized inTable 4.4.

Through the modelling, a wide range ofinformation was presented:

• Distribution of land cover;• Basin-wide soil water-holding capacity;• Annual basin-wide evapotranspiration;• Simulated streamflows at specified

locations within the basin;


Box 4.1. Main sections of a national watersector profile (ADB, 1996).

Section 1 Country overviewSection 2 National policy environment2.1 National development goals2.2 Water resources policy2.3 Transnational and subnational relationsSection 3 Capacity for water resourcesmanagement3.1 Legal base3.2 Institutional base3.3 Information base3.4 Training and human resources developmentSection 4 Water resources status4.1 Water resources and watersheds: the physicalbase4.2 Uses of water4.3 Community values of water4.4 Supply, demand and sustainability4.5 Summary: status and trendsSection 5 Financial resources5.1 Annual expenditure and revenues5.2 Return on investmentSection 6 Appraisal6.1 Water sector institutions6.2 Water resources and watersheds6.3 Uses of water6.4 Community values of water6.5 Sustainability of water resources and use6.6 Financial performance6.7 Consistency with external support agencyobjectivesSection 7 Agenda for action1 Appendices2 References3 Completed and current water-related projects4 Summary of lessons learnt from completedprojects5 Contact persons6 Maps of priority river basinsData Appendices



• Seasonal and annual crop transpirat-ion and soil evaporation in irrigatedareas;

• Simulated crop yields for actual sea-sonal climatic conditions in dry andwet years.

The model was used to analyse and assessthe current management of the waterresource within the basin and to assess theimpact, especially on crop production, ofpossible changes, such as a change inclimate.


Box 4.2. Summary of key data for river basin management.

Physical data• Latitude/longitude• Catchment area• River channel length• River slopes• Land use types and areas• Land slopes and areas• Soil types and areas• Aquifers (numbers and areas)

Demographic data• Total population (past, present

and projected)• Population densities• Population by location

(urban/rural)• Population by work type• Attainment levels for educa-

tion (by age and gender)

Institutional• Development policy• Water policy• Water law• Environmental law• Land tenure• Stakeholders – roles and

responsibilities• Water rights

Economic• National GNP• Regional or basin GNP• Average basin per capita GNP

Hydrometric data• River discharges• River water levels• River flood peak discharges• River base flows• River sediment load• River water quality• Lake/reservoir water levels• Lake/reservoir volumes• Lake/reservoir water quality• Lake/reservoir water

temperature

• Lake/reservoir surfaceevaporation

• Volume of water imported/exported to/from basin

Meteorological and climatic• Sunshine/radiation hours• Wind speed• Air temperature – average/

max/min• Humidity• Evaporation• Precipitation• Precipitation intensity

Groundwater• Groundwater levels• Groundwater quality• Aquifer yields and quality• Estimate annual groundwater

recharge

Agricultural• Cultivable area• Irrigable area• Irrigated area• Irrigation water abstractions

(surface/groundwater)• Drainage return flows –

quantity• Drainage return flows – quality• Number of landholders• Population dependent on

irrigated agriculture• Value of irrigated agricultural

production

Potable and wastewater• Abstraction quantity (surface/

groundwater)• Abstraction quality• Return flow – quantity• Return flow – quality• Number of people supplied

Industrial• Abstraction quantity (surface/

groundwater)

• Abstraction quality• Return flow – quantity• Return flow – quality• Number of people employed

Navigation• River water levels• River discharges• River channels and depths

Hydroelectric power• Generation capacity• Discharge requirements and

timing• Maximum discharge

requirements and timing• Minimum discharge

requirements and timing

Environmental• Minimum flow requirements• Critical flow periods and

demands• Protected areas and water

demands• Required water quality

standards

Recreational• Minimum flow requirements• Critical flow periods and



standards

Tourism• Minimum flow requirements• Critical flow periods and



standards




Indi

cato

rE

astR

apti

basi

n,N

epal

Ged

izba

sin,

Tur

key

Lerm

a–C

hapa

laba

sin,

Mex

ico

Clim

ate

clas

sific

atio

nT

ype

ofba

sin

(ope

n/cl

osin

g/cl

osed

)C

atch

men

tare

a(k

m2 )

Mai

nriv

erch

anne

llen

gth

(km

)La

ndus

e(%

byty

pe)

Num

ber

ofla

kes/

rese

rvoi

rsP

opul

atio

n(m

illio

ns)

Pop

ulat

ion

dens

ity(p

erso

ns/k

m2 )

Rat

ioof

urba

nto

rura

lpop

ulat

ion

Ann

ualh

ydro

pow

erge

nera

tion

Ave

rage

annu

alra

infa

ll(m

m)

Ave

rage

annu

alev

apot

rans

pira

tion

(mm

)R

ainf

allm

onth

san

dav

erag

eam

ount

falli

ng(%

)D

ryse

ason

mon

ths

and

aver

age

amou

ntfa

lling

(%)

Ave

rage

annu

alriv

erflo

wvo

lum

e(M

CM

)P

eak

flood

flow

(m3 /

s)M

inim

umba

seflo

w(m

3 /s)

Irrig

ated

area

(ha)

Tro

pica

lmon

soon

Ope

n3,

120

122

For

est(

65%

)in

clud

ing

Nat

iona

lP

ark;

rain

fed

agric

ultu

re(1

7%);

irrig

ated

agric

ultu

re(1

0%);

gras

san

dsh

rub

(1.5

%);

river

,str

eam

san

dsa

nd(5

.5%

),ot

hers

(1%

)N

oco

nstr

ucte

dda

ms/

rese

rvoi

rs0.

5416

625

:75

Nil

2008

987

(ET

a)6

mon

ths

(93%

)6

mon

ths

(7%

)5,

993

225

35 32,3

88(g

ross

)

Med

iterr

anea

nC

losi

ng17

,700

275

Mak

i(30

%);

fore

st(2

6%);

rain

fed

crop

s(2

5%);

irrig

ated

crop

s(8

%);

shru

blan

d(3

%);

barr

en(8

%)

2la

rge;

2sm

all

0.67

38 60:4

0

100

GW

h80

0m

m(u

pper

catc

hmen

t)45

0(lo

wer

catc

hmen

t)

1,25

0(E

Tp)

6m

onth

s(8

0%)

6m

onth

s(2

0%)

940

718

0 110,

000

(larg

esc

ale)

;25,

000

(sm

all)

Sem

iarid

tosu

bhum

idC

lose

d54

,300

750

Rai

nfed

culti

vatio

n(3

7%);

irrig

ated

agric

ultu

re(2

0%);

scru

b(1

2%);

gras

slan

d(1

4%);

fore

st(1

2%);

wat

erbo

dies

(3%

);ot

her

(2%

)

27(la

rge)

;1,5

00(s

mal

l)11 20

370

:30

(urb

ande

fined

asse

ttlem

ents

with

mor

eth

an25

00in

habi

tant

s)19

2G

Wh

712

(av.

)49

4(m

in.)

1,02

2(m

ax.)

1,90

06

mon

ths

(91%

)6

mon

ths

(9%

)5,

757

1,05

00 40

9,00

0(s

urfa

ce);

380,

000

(gro

undw

ater

);78

9,00

0(t

otal

)

Tab

le4.

3.E

xam

ples

ofke

yde

scrip

tors

and

indi

cato

rsfo

rriv

erba

sins

.



Information Needs for Basin Water Management 65A

vera

gedr

opof

grou

ndw

ater

leve

l(m

/yea

r)A

nnua

lvol

ume

ofw

ater

use

(MC

M,%

):Ir

rigat

ion

(sur

face

/gro

undw

ater

)

Pot

able

wat

er(s

urfa

ce/g

roun

dwat

er)

Indu

stria

l(su

rfac

e/gr

ound

wat

er)

Hyd

ropo

wer

Nav

igat

ion

Env

ironm

ent

For

estr

yF

ishe

ries

Wat

erac

coun

ting

data

(MC

M)b

Gro

ssin

flow

Net

inflo

wA

vaila

ble

wat

er–

pres

ent

Ava

ilabl

ew

ater

–po

tent

ial

Dep

lete

dfr

actio

nO

utflo

wP

roce

ssfr

actio

nN

on-p

roce

ssB

enef

icia

lN

on-b

enef

icia

lU

ncom

mitt

edC

omm

itted

Not

falli

ng

240

5.9

0.18

Nil

Non

e47

3(a

lloca

tion

for

Nat

iona

lP

ark

activ

ities

)1,

561

Non

esp

ecifi

ed

5,99

36,

120d

471

5,11

02,

007

3,57

524

91,

560

197

3,10

247

3

Fal

ling

660

(sur

face

,75%

);35

(gro

undw

ater

,4%

)13

4(1

4%)

50(6

%)

No

spec

ialr

elea

ses

Non

e4

(cur

rent

allo

catio

n,1%

,mor

ene

eded

)N

one

spec

ified

Non

esp

ecifi

ed

940

(sur

face

);16

0(g

roun

dwat

er)

1,10

01,

100

1,10

060

050

0(o

nly

inw

inte

r)50

075 25 0 0

2.1

6,58

4(6

8%)

1,35

1(1

4%)a

278

(3%

)39

(incl

uded

inev

apor

atio

nfr

omw

ater

bodi

es)

Non

e2,

270

(23%

)de

fined

asev

apor

atio

nfr

omw

ater

bodi

esN

one

Non

e

38,6

78c

39,3

1937

,319

36,4

191.

02(D

Fav

aila

ble)

1,10

00.

43(P

Fav

aila

ble)

22,0

41

0 −900

2,00

0

a Inc

lude

s79

1M

CM

used

with

inth

eba

sin

and

560

MC

Mtr

ansf

erre

dto

citie

sou

tsid

eth

eba

sin.

b Wat

erac

coun

ting

data

obta

ined

follo

win

gpr

oced

ures

seto

utin

Mol

den

(199

7).

c Fig

ures

for

the

Lerm

a–C

hapa

laba

sin

are

for

1999

.d I

nclu

des

tailw

ater

inflo

wfr

omH

EP

sche

me

onad

jace

ntriv

ersy

stem

.M

CM

,mill

ion

cubi

cm

etre

s.



In contrast to the above, informationfor analysing the institutional arrangementsgoverning the river basin can only becollected ‘on the ground’. Timelines andresponsibility charts are useful analyticaltools in this context. Timelines chartthe development of water resource-relatedevents within the river basin. These eventsare both physical, such as floods, droughts,disease outbreaks, etc., and institutional,such as enactment of water laws, establish-ment of water users associations, publicoutcries, etc. The case studies presented inChapters 6–11 offer examples of timelinesshowing how physical changes within theriver basin over time have been met with

institutional changes to help manage andalleviate the problems encountered.

Functional responsibility charts, as des-cribed in Chapter 1, can be used to identifykey actors in the water resource sector and toassess their role in relation to the nine keybasin management functions. These chartscan be used to structure an assessment of thedata and information processes within theriver basin by asking the following questionsrelated to each function:

• What data does each organizationcollect?

• What processing and analysis is doneon the data collected?


Operational rules

Soils characteristics

Climate

Streamflow

Land cover

Topography

Leaf area index

RS

Locally sourced

Non-RS

Global public domain

Fig. 4.4. Data required for integrated basin hydrological modelling as a function of availability andmethod of observation (Kite and Droogers, 2000).

Data type Data source

Topographicanalysis

Land coverclassification

Leaf area indexSoils

MeteorologicalStreamflowRegulation

Downloaded from the Internet a Digital Elevation Model (DEM) from the UnitedStates Geological Survey (USGS) dataset; calculated streamflow travel distancesusing the TOPAZ (Topographic PArameteriZation) programme

Base data of Normalized Difference Vegetation Index (NDVI) images taken from theNOAA-AVHRR satellite sensor at 1-km resolution

Derived from NDVI data for land cover classificationUsed a locally available soil map (1 : 200,000) scale as the FAO world soil map is

too coarse at 1:500,000 scaleUsed local climate station and global datasets.Used local river gauging station dataUsed records of actual system regulation rules

Table 4.4. Summary of data sources for Gediz river basin model.



• How are the data stored?• What reporting and dissemination is

carried out?• Who uses the data?• What action and decisions are made

using the data?

As with the institutional processes outlinedin Chapter 1, this analysis helps to identifygaps, both in the data collected, processedand analysed, and in its use and dissemina-tion. As will be seen in the case studies,there are significant differences in the effi-cacy of the data and information processesin the six river basins studied.

4.6 Irrigation System Level InformationNeeds

Data are required by the managers of theirrigation system to facilitate management,operation and maintenance of the system.Table 4.5 identifies some illustrative datarequirements and their use for LCIS in theirearly growth stages.

Data may also be required by externalorganizations, typically a governmentagency, which has the responsibility ofmonitoring the performance of the LCIS.This monitoring can be for three reasons:

1. To collect data for regional and/ornational level statistics;2. To ensure that the association is beingcorrectly managed financially;3. To ensure that the irrigation and drain-age system is being adequately maintained.

In the initial stages of irrigation manage-ment transfer (IMT), LCISs are oftenrequired to continue previous governmentagency practices of reporting cropped areas,yields and other data for preparation ofregional and/or national statistics. In addi-tion, the government has a duty of care tothe farmers to ensure that their systems areproperly managed, and that the fees they

pay are correctly used. For this purpose, thegovernment will usually carry out annualaudits of the organization’s accounts. Anadditional exercise that is not yet wellestablished is the periodic auditing of themaintenance of the system’s assets – thecanals, drains and structures. In many cases,even following IMT, the infrastructurestill belongs to government. Procedures aretherefore required to ensure that this infra-structure is not allowed to deteriorate to alevel where the system becomes unusable.

The extent and nature of the data col-lected will vary from scheme to scheme,depending on the general level of literacy,technical development of the irrigation sys-tems, and technological capacity of the farm-ing community. Farmer-managed schemesin Nepal, for example, generally have littleformal data collection, with an equitablesupply-orientated water allocation systembeing controlled through the use of propor-tional dividers. In locally controlled, profes-sionally managed systems in Turkey, on theother hand, where water is allocated ondemand and flow adjustments madeusing undershot sluice gates, relativelysophisticated data-collection and process-ing systems are required. In the Nepal case,the irrigation service fee is associated withthe irrigable area, whereas in Turkey the feeis associated with the area actually irrigatedand the crop type.

As pressure on the water resourceincreases, measurement of performancebecomes increasingly important. A popularversion of performance assessmentcurrently is benchmarking, whereby theperformance of different schemes iscompared, and internal performance overtime monitored. Table 4.6 presents perfor-mance indicators proposed by Malano andBurton (2001) to facilitate performancebenchmarking, enabling managers to com-pare the performance of their system withthat of other systems, in particular thosethat are performing well.1 By tracking these


1 Systems that are identified as performing well in relation to key indicators can be studied and the practicesand procedures contributing to this performance identified. These ‘best practice’ activities can then bereplicated by other systems in order to raise their level of performance.



indicators (such as total value of fees col-lected) over time, it is also possible to assessthe performance of a single system over time.Figures 4.5 and 4.6 present data collectedfrom 46 irrigation schemes in Australia,showing the range of values of waterdelivered and costs recovered. Both typesof comparisons give system managers the

information (and incentives) they need toraise their performance.

4.7 Issues

Some of the most important issues relatedto the collection, processing, analysis and


Data Purpose/use

ManagementNumber of farmers in

command areaNames and addresses of

association membersRevenue collected,

by sourceManagement costs

Command area serviced

Indicates the maximum potential number of members in the LCISorganization. Not all farmers will necessarily be members

Basic data for organization membership

Includes revenue from each member and non-members, plus othersources of income (rents, hire out of equipment, etc.)

Costs associated with running the association – salaries of office oradministrative staff, office rental, equipment purchase or hire,stationery, etc.

Potential area that can be irrigated. Provides base for analysis ofperformance (i.e. yield, kg/ha; fee recovery rates, $/ha; waterdelivery, m3/ha)

OperationTotal seasonal or annual

volume of water receivedTotal seasonal or annual

volume of water delivered

Operating costs

Total seasonal or annual volume of water received by the organization,based on daily measurements at the intake(s) to the irrigation system

Total seasonal or annual volume delivered by the organization basedon measurement of discharge at delivery points or estimates ofvolume delivered. Should correspond with irrigation service feesrecovered if fees are levied based on volume delivered. Relativelyfew LCIS have measurement at delivery points

Costs associated with operation – fees for water (from bulk seller),operation equipment (motorbikes, bicycles), salaries for watermasters, electricity costs for pumping, etc.

MaintenanceInventory of assets and

their condition

Maintenance requirements

Maintenance costs

Inventory of all irrigation and drainage infrastructure and its condition.The inventory can be presented in the form of tables and schematicdiagrams

Periodic or annual inspection of the irrigation and drainage system toidentify the maintenance needs, costs and priorities

Costs associated with maintenance – maintenance staff wages,maintenance contract costs, equipment hire or purchase, etc.

ProductionSeasonal and annual area

irrigated, by crop

Total value of cropproduction

Basic indicator of performance. If the area irrigated is low there may bepotential for increasing the area irrigated and thus the fee recovery.Shows performance over time

Useful indicator to assess the cost of irrigation relative to the productionbenefits

LCIS, locally controlled irrigation system.

Table 4.5. Typical data needs for locally managed irrigation systems.




Indi

cato

rD

efin

ition

Rem

arks

Fin

anci

alC

ostr

ecov

ery

ratio

Tot

alM

OM

cost

per

unit

area

($/h

a)

Rev

enue

colle

ctio

npe

rfor

man

ce

Pro

duct

ion

Irrig

ated

crop

area

ratio

Gro

ssre

venu

eco

llect

edT

otal

MO

Mco

st

Tot

alM

OM

cost

Tot

alco

mm

and

area

serv

iced

byth

esy

stem

Gro

ssre

venu

eco

llect

edG

ross

reve

nue

invo

iced

Tot

alan

nual

reco

rded

irrig

ated

crop

area

Tot

alco

mm

and

area

serv

iced

byth

esy

stem

Gro

ssre

venu

eco

llect

ed:T

otal

reve

nues

colle

cted

from

paym

ento

fser

vice

sby

wat

erus

ers

Tot

alM

OM

cost

:Tot

alm

anag

emen

t,op

erat

ion

and

mai

nten

ance

cost

ofpr

ovid

ing

the

irrig

atio

nan

ddr

aina

gese

rvic

eex

clud

ing

capi

tale

xpen

ditu

rean

dde

prec

iatio

n/re

new

als

Tot

alM

OM

cost

:Tot

alm

anag

emen

t,op

erat

ion

and

mai

nten

ance

cost

ofpr

ovid

ing

the

irrig

atio

nan

ddr

aina

gese

rvic

eex

clud

ing

capi

tale

xpen

ditu

rean

dde

prec

iatio

n/re

new

als

Tot

alco

mm

and

area

serv

iced

byth

esy

stem

:The

com

man

dar

eais

the

nom

inal

orde

sign

area

prov

ided

with

irrig

atio

nin

fras

truc

ture

that

can

beirr

igat

edG

ross

reve

nue

colle

cted

:Tot

alre

venu

esco

llect

edfr

ompa

ymen

tofs

ervi

ces

byw

ater

user

sG

ross

reve

nue

invo

iced

:Tot

alre

venu

edu

efo

rco

llect

ion

from

wat

erus

ers

for

prov

isio

nof

irrig

atio

nan

ddr

aina

gese

rvic

es

Tot

alan

nual

reco

rded

irrig

ated

crop

area

:The

tota

lirr

igat

edar

eafo

rw

hich

irrig

atio

nfe

esha

vebe

enpa

idor

invo

iced

durin

gth

eye

arT

otal

com

man

dar

ease

rvic

edby

the

syst

em:T

hear

eapr

ovid

edw

ithirr

igat

ion

infr

astr

uctu

reth

atca

nbe

irrig

ated

cont

inue

d

Tab

le4.

6.W

UA

and

Fed

erat

ion

perf

orm

ance

benc

hmar

king

indi

cato

rs(f

rom

Mal

ano

and

Bur

ton,

2001

).




Indi

cato

rD

efin

ition

Rem

arks

Wat

erm

anag

emen

tT

otal

annu

alvo

lum

eof

irrig

atio

nw

ater

deliv

ery

(m3 /

year

)A

nnua

lirr

igat

ion

wat

erde

liver

ype

run

itco

mm

and

area

(m3 /

ha)

Ann

uali

rrig

atio

nw

ater

deliv

ery

per

unit

irrig

ated

area

(m3 /

ha)

Mai

nsy

stem

wat

erde

liver

yef

ficie

ncy

Mai

nten

ance

Mai

nten

ance

cost

tore

venu

era

tio

Mai

nten

ance

cost

per

unit

com

man

dar

ea($

/ha)

Adm

inis

trat

ion

WU

Am

embe

rshi

pra

tio

Tot

alvo

lum

eof

wat

erde

liver

edto

wat

erus

ers

over

the

year

orse

ason

Tot

alan

nual

volu

me

ofirr

igat

ion

wat

erin

flow

Tot

alco

mm

a nd

area

serv

iced

byth

esy

stem

Tot

alan

nual

volu

me

ofirr

igat

ion

wat

erin

flow

Tot

alan

nual

reco

rded

irrig

ated

crop

area

Tot

alan

nual

volu

me

ofirr

igat

ion

wat

erde

liver

yT

otal

ann u

alvo

lum

eof

irrig

atio

nw

ater

inflo

w

Mai

nten

ance

cost

Gro

ssre

venu

eco

llect

ed

Mai

nten

ance

cost

Tot

alco

mm

and

area

serv

iced

byth

esy

stem

Tot

alnu

mbe

rof

WU

Am

embe

rsT

otal

num

ber

offa

rmer

sin

com

ma n

dar

ea

Mea

sure

dat

the

inte

rfac

ebe

twee

nth

eirr

igat

ion

serv

ice

prov

ider

and

wat

erus

ers

Tot

alan

nual

volu

me

ofirr

igat

ion

wat

erin

flow

:Tot

alan

nual

volu

me

ofw

ater

dive

rted

orpu

mpe

dfo

rirr

igat

ion

(not

incl

udin

gdi

vers

ion

ofin

tern

aldr

aina

ge)

Tot

alco

mm

and

area

serv

iced

byth

esy

stem

:The

area

prov

ided

with

irrig

atio

nin

fras

truc

ture

that

can

beirr

igat

edT

otal

annu

alvo

lum

eof

irrig

atio

nw

ater

inflo

w:T

otal

annu

alvo

lum

eof

wat

erdi

vert

edor

pum

ped

for

irrig

atio

n(n

otin

clud

ing

dive

rsio

nof

inte

rnal

drai

nage

)T

otal

annu

alre

cord

edirr

igat

edcr

opar

ea:T

heto

tali

rrig

ated

area

for

whi

chirr

igat

ion

fees

have

been

paid

orin

voic

eddu

ring

the

year

Tot

alan

nual

volu

me

ofirr

igat

ion

wat

erde

liver

y:T

otal

volu

me

ofw

ater

deliv

ered

tow

ater

user

sov

erth

eye

aror

seas

onT

otal

annu

alvo

lum

eof

irrig

atio

nw

ater

inflo

w:T

otal

annu

alvo

lum

eof

wat

erdi

vert

edor

pum

ped

for

irrig

atio

n(n

otin

clud

ing

dive

rsio

nof

inte

rnal

drai

nage

)

Mai

nten

ance

cost

:Tot

alan

nual

expe

nditu

reon

syst

emm

aint

enan

ceG

ross

reve

nue

colle

cted

:Tot

alre

venu

esco

llect

edfr

ompa

ymen

tofs

ervi

ces

byw

ater

user

sM

aint

enan

ceco

st:T

otal

annu

alex

pend

iture

onsy

stem

mai

nten

ance

Tot

alco

mm

and

area

serv

iced

byth

esy

stem

:The

com

man

dar

eais

the

nom

inal

orde

sign

area

prov

ided

with

irrig

atio

nin

fras

truc

ture

that

can

beirr

igat

ed

Tot

alnu

mbe

rof

WU

Am

embe

rs:T

otal

num

ber

offa

rmer

sre

gist

ered

asm

embe

rsw

ithth

eW

UA

Tot

alnu

mbe

rof

farm

ers

inco

mm

and

area

:Tot

alnu

mbe

rof

farm

ers

with

regi

ster

edla

ndho

ldin

gsw

ithin

the

WU

Aco

mm

and

area

WU

A,w

ater

user

sas

soci

atio

ns.

Tab

le4.

6.C

ontin

ued.



use of data and information in a river basincontext are summarized below.

4.7.1 Issues at river basin level

1. Coordination and sharing of data.As basins develop, different agenciesbecome involved to different degrees inwater resource data collection, processingand analysis. There is often a lack of

coordination, cooperation and sharing ofdata between these agencies. As thepressure on the available water resourcesincreases, improved cooperation and datasharing is needed between these agencies,and between collecting agencies andinterested third parties such as farmerorganizations. Measures to enhance co-ordination and sharing can include theformation of river basin councils, and thecreation of national data banks for waterresources.


0.000

0.200

0.400

0.600

0.800

1.000

1.200

1.400

1.600

1.800M

urra

y Ir

rigat

ion

Low

er

Mur

ray

Wim

me

ra-M

alle

e

Eto

n

War

rill

Bar

ker-

Bar

amba

h

Bun

dab

erg

Cre

ssy-

Long

ford

Boy

ne R

iver

G-M

W P

yram

id-B

oort

Jem

alon

g

We

rrib

ee

G-M

W S

hepp

arto

n

Em

eral

d

G-M

W C

ent

. Gou

lbur

n

G-M

W T

orru

mba

rry

G-M

W M

urra

y V

alle

y

St G

eorg

e

G-M

W N

yah

Bur

deki

n R

iver

Firs

t M

ildur

a

Cen

tral

Irrig

. (S

A)

Sun

land

s

Dep

th d

eliv

ered

(m

)

Fig. 4.5. Example of comparative plot of irrigation water delivery for different irrigation systems (ANCID,2000).

0

0.5

1

1.5

2

2.5

3

Col

eam

bally

Boy

ne R

iver

Bur

deki

n R

iver

Em

eral

d

Loga

n

Mar

eeba

-Dim

bula

h

Sou

th B

urde

kin

War

rill

Mac

alis

ter

Ord

Riv

er

Cre

ssy-

Long

ford

Car

narv

on

G-M

W S

hepp

arto

n

G-M

W R

oche

ster

G-M

W W

oorin

en

G-M

W T

orru

mba

rry

Jem

alon

g

G-M

W M

urra

y V

alle

y

G-M

W T

resc

o

Mur

ray

Irrig

atio

n

Wim

mer

a-M

alle

e

Cen

tral

Irrig

. (S

A)

Con

dam

ine

Cos

t rec

over

y ra

tio

Fig. 4.6. Example of comparative plot of cost recovery ratios (ANCID, 2000).



2. Adaptation to changing needs. Govern-ment organizations can be slow to adapt toa changing environment and to changingneeds. Change is often prompted by crisisevents, but managers should be alert to theneed for change before a crisis occurs. Risinglevels of pollution in rivers and fallinggroundwater table levels in river basins aretwo areas where, typically, action is nottaken in good time to alleviate loomingcrisis. In these circumstances, it is theresponsibility of managers to gather, presentand disseminate data to highlight the issueand stimulate a response.3. Computers, remote sensing and GISs.There is increasing use of computers for datastorage, processing, analysis and presenta-tion. They have a valuable role to play inpresenting data in ways that make themmeaningful to non-professionals, and acentral role in processing and analysingcomplex data, as in water resources model-ling. GISs are now widely used to store andpresent data, and such uses can be expectedto expand. Increasingly computers are beingused in association with remotely sensedimages to analyse and present complexrelationships, allowing analysis in formshitherto not possible.4. Water rights. When water supplies areabundant, it may not be necessary to specifywater rights or licences for specific uses. Aswater becomes increasingly scarce, societyhas to decide who should have access to theavailable water and in what quantity andquality. Users must be assigned water rightsso that they can plan for the future, particu-larly in relation to making investments.Information on water resource use andavailability over time is vital in allocatingwater rights and in determining proceduresfor real-time allocation of water in timesof shortage. Water right allocation andreallocation can be politically contentious,but become increasingly necessary as basinsmature and close.5. Pollution. Significant pollution is oftena feature of river basins in the middle stagesof their development, as water resourcesare taken up and begin to become scarce.Pollution is not only hazardous to human

health and the natural ecology; it requireslarge amounts of water to dilute it. Manyriver systems have had to die beforeadequate action has been taken to remedylethal levels of pollution. The River Thamesflowing through London was at one time sopolluted that the Members in the adjacentHouses of Parliament had to recess duringthe summer months to avoid the stench.Control of point and non-point sources ofpollution requires legislation, funding, andpolitical and social will to achieve.6. Over-abstraction of groundwater. Afeature of the latter stages of river basindevelopment, over-abstraction of ground-water, appears to be one of the mostdifficult issues for societies to address. Inthe case of the Lerma–Chapala (Chapter 8)and California’s Central Valley (Chapter 7),both well-developed river basins, over-exploitation of groundwater remains a majorunresolved issue. It is a classic ‘tragedy ofthe commons’ scenario, on which societyhas to form a consensus to legislate andabide by.7. Enforcement. Data and information arethe basis for establishing and enforcingregulations and standards related to waterresources management. In some societies,despite adequate legislation and data, thereis a lack of political or social will to enforcethe legislation. In such contexts, informa-tion and its wide dissemination are keyfactors in raising awareness of the problemsand building the political and social constit-uencies to support legal enforcement.

4.7.2 Issues at LCIS level

1. How much data to collect. The type andquantity of data collected will vary widelyamong countries and with size and type ofirrigation system. There is often a noticeabledifference between the kind and amountof data that water resources and irrigationprofessionals think is required for manage-ment purposes and that which farmers seeas being necessary. Systems large enoughto employ professional managers, as is




typically the case in Spain, Mexico andTurkey, for example, will generally havemore comprehensive management informa-tion systems than schemes that are managedby farmers without professional training inwater resources or irrigation management.Generally, larger schemes, especially thosegrowing principally for a market, requiremore comprehensive quantitative informa-tion systems than smaller, simpler schemes.2. Discharge measurement. Whether dis-charges are measured and the data used ornot marks a major divide among irrigationschemes. For smaller simpler schemes, therewill not be significant benefit to the addi-tional work involved in measuring andusing discharge data, while in schemeswhere water is scarce and returns to irriga-tion are high, significant benefits can arisethrough improved management based ondischarge measurement. On large irrigationschemes, measurement becomes essentialif water is to be distributed equitably. Areliable discharge measuring system is noteasy to establish and sustain, requiringmeasuring equipment or structures, calibra-tion, maintenance, regular data retrieval andprocessing, and trained staff.3. Data processing and analysis. There is amarked difference between data processingand analysis with and without a computer.Where LCIS managers do not have acomputer, data processing will necessarilybe more basic. Nevertheless, useful informa-tion can be generated from simple process-ing of the available data using a handcalculator. Once, again, the size of thescheme, the value of its output, and thepresence of professional managers willdetermine whether computer-based dataprocessing is appropriate.4. Transparency. Access to informationabout system operations is essential forgood governance and management. Whenfarmers’ access to information about wateravailability, rules, deliveries and finances isobstructed, corruption and misuse of fundsand resources becomes easier and morelikely. To ensure transparency, clearrequirements for information sharing needto be included in water legislation governing

LCISs. All farmers should be aware of theirright to inspect the association’s accountbooks, and information on the scheme’sphysical and financial performance shouldbe presented and discussed at the associa-tion’s annual general meeting. Governmentregulatory authorities should ensure that:(i) farmers are aware of their rights; (ii)annual financial audits are conducted; and(iii) the LCIS managers organize annual orsemi-annual general meetings and presentthe requisite performance information tofarmers.5. Access to external data. LCIS have animportant role to play in the management ofthe larger basin, defending irrigators’ inter-ests and reaching agreements with otherwater users over water sharing and waterquality issues. To play this role effectively,LCIS managers need access to basin-leveldata collected by government agencies andothers, and the ability to use and interpretthat data. This requires guarantees of accessto information, and the professional skillsto understand, discuss and debate thatinformation. Here a federation of LCIS in aparticular basin can be extremely valuablein pooling resources needed for effectiverepresentation.6. Training for LCIS personnel. Training isa crucial and underemphasized part of theturnover process. In project-funded transferprogrammes, training is often a part of theassistance package provided. Seldom, how-ever, do project sponsors or governmentsestablish an on-going training capability forLCIS following IMT. As elected leaders andmanagers rotate following elections, skillsfrom a one-off training effort are lost andmanagement quality degrades. Here too, afederation of LCIS can be important in pool-ing resources to support a national trainingprogramme for member organizations.

4.8 Conclusions

Data and information are fundamental tothe good governance and management ofa river basin. In keeping with changes in




society in general, decision making for riverbasins is increasingly being devolveddownwards from national to state orregional governments and local stake-holders. Associated with this trend is anincreasing awareness amongst the generalpublic of water-related issues, and theirrights and responsibilities in relation to theuse and management of increasingly scarcewater resources. Experience in many coun-tries is showing how the sharing of data andinformation by government agencies withthe wider public enables management tomove on to higher performance levelsthrough enlisting support and collaborationfrom water users. Societies and governmentorganizations that seek to hold on to powerthrough restricting the sharing and access towater resources data and information willfind that they are limiting their capabilityto make the most productive use of theavailable water resources.

In water-scarce situations, accountabil-ity and transparency through provision ofdata and information at both the river basinand irrigation scheme level enable manage-ment to gain the consensus and supportrequired from water users to address issuesrelated to scarcity. Formulation, agreement,compliance and enforcement of measures tomanage within the context of scarcity areempowered by the acquisition and sharingof data and information between stake-holders. For LCISs, the acquisition andsharing of data become central to theirneed to manage their external environmentto secure their water rights and gain a fairshare of the available resource. Within theirirrigation schemes, improved data and infor-mation processes are required as part of thepressure placed on the irrigation sector toimprove management and justify its share ofthe water resource.

Data and information needs changeover time, and those responsible for the

collection, processing and analysis of dataneed to review their information strategiesand information systems to reflect thedynamic environment in which they oper-ate. As river basins develop, data and infor-mation needs both broaden (to encompassmore diverse requirements) and deepen (toenable more detailed resource assessment,allocation and monitoring). The availabilityof microcomputers and associated softwareare dramatically changing the possibilitiesfor river basin managers, allowing them toaccess and process data to a level hithertoonly possible in research establishments ornational headquarters offices. Spatial repre-sentation systems, such as remote sensingand GISs, are becoming standard tools forsystem analysis, allowing more precisionin the analytical processes associated withriver basin management.

References

ADB (1996) Towards effective water policy in theAsian and Pacific Region. Arriëns, W., Bird,J., Berkoff, J. and Mosley, P. (eds) Proceedingsof the Regional Consultation Workshop,Manila, 10–14 May, Asian DevelopmentBank.

ANCID (2000) 1998/99 Australian IrrigationWater Provider – Benchmarking Report.Australian National Committee on Irrigationand Drainage, Victoria, Australia.

Kite, G. and Droogers, P. (2000) IntegratedBasin Modelling. IWMI Research Report 43,International Water Management Institute,Colombo, Sri Lanka.

Malano, H. and Burton, M. (2001) Guidelines forbenchmarking performance in the irrigationand drainage sector. International Pro-gramme for Technology and Research inIrrigation and Drainage (IPTRID), Rome.

Molden, D. (1997) Accounting for water useand productivity. SWIM Paper No. 1, Inter-national Irrigation Management Institute,Colombo, Sri Lanka.




5 Financing River Basin Organizations

Charles L. Abernethy

5.1 Introduction: Uses and Demandsfor Water

It has become quite widely acceptedthat countries should be aiming towardscomprehensive management of waterresources through organizations basedon river basins or aquifers. This chapteraddresses the situation where a country hasalready decided that it wants to assign anorganization for this management purpose.In most cases, this will mean establishinga new organization, rather than extendingthe functions of an existing organization.Such organizations will not develop effec-tively unless they can be provided withadequate financial resources. The questionsconsidered here relate to the ways ofsupplying these financial resources, andthe impacts that these financing processesmay have on the behaviour and effective-ness of the organization, and on the peoplewho are supposed to be helped by itsexistence.

In this respect, the experiences of thericher countries may not offer much usefulguidance to the developing countries, espe-cially the poorest ones. Their patterns ofwater use are radically different. Table 5.1shows the breakdown at the broadest level,among the three biggest user sectors. Thedata for this table are drawn mainly from theWorld Bank’s ‘World Development Report’for 1998–1999. The data on the broadsectoral distribution of water uses (amongagricultural, domestic, and industrial uses)are not up to date, since most of the informa-tion about this in the global referenceliterature dates from 1987; however, thesenumbers are adequate to illustrate somegeneral patterns.

The boundaries defining the fourwealth classes in Table 5.1 have been setat the following levels (1998 data): $2375,$5750 and $12,500. These are in terms ofgross national product per person per year,expressed in equivalent US dollars, at PPP(purchasing power parity).


Sector Agriculture Domestic Industry

Wealth categoryLow incomeLower middle incomeUpper middle incomeHigh income

89747340

48

1215

7181545

Source: Adapted from World Bank (1999).

Table 5.1. Sectoral use of water. Units: % of annual freshwater abstractions.



In the rich countries, industrial userspredominate. In the poorest countries, theindustrial category is not yet very signifi-cant, whereas as much as 89% of all thewater abstracted is used for agriculture.

These patterns of use illustrate twoobvious factors that have great influenceon the financing situation. First, in a basinwhere the majority of users are smallagriculturists, they are usually extremelynumerous, forming a large majority of thepeople in the poorest countries. Secondly,the productivity of water used in agricultureis usually very much lower than that ofwater used in industry. So when we dealwith industry we are usually dealing with arelatively small number of people who areengaged in relatively profitable activities,while when we deal with agricultural users(in developing countries), we are probablydealing with large numbers of people whosefinancial resources are very meagre.

We can also note (although it is farbeyond the scope of this chapter) that thelow productivity of irrigated agriculturehas well-known links with the agriculturalproduction and market-access policies ofrich countries, and other issues of globalscale which no developing country canmodify much by its own choice of policies.

Domestic water supply is differentagain. We cannot compare it with other

types of use on the basis of productivity.Domestic use is essential for human healthand indeed survival. So we supply water forsocial objectives as well as for productiveobjectives, and these are not really compara-ble in a financial sense.

That distinction between social andproductive objectives is not as clear as wemight like. The four basic human needsfor water – drinking, washing, cookingand sanitation – are certainly essential, butdomestic uses of water can include manynon-essential uses. When we compare theconsumption patterns of the rich and poorcountries, this becomes very evident. AsTable 5.2 shows, in the rich countries theamount of domestic water used per personis very much higher than in the poorestcountries. Also, within each country thereare similar variations of consumption,related to poverty or affluence.

In Table 5.2, we see that in the poorestcountries the abstractions for personal useare minimal. We may estimate the basicneeds at about 30 m3/person/year1, so thefigure of 16.4 indicates that many people areobliged to satisfy those needs in ways thatdo not reach the formal statistical system.These are people who have to bathe in openbodies of water, carry household water fromlocal streams, and in other such ways areomitted from the data.

76 C.L. Abernethy

Sector Agriculture Domestic Industry

Wealth categoryLow incomeLower middle incomeUpper middle incomeHigh income

332339332386

16.436.355.9

146.5

26.781.768.9

442.2

Source: Adapted from World Bank (1999).

Table 5.2. Abstractions per person. Units: m3/person/year abstracted from the natural systems.

1 This is of course an approximation, dependent on many circumstances. Various authors quote numbers inthe general range 20–35 m3/person/year. See for example Seckler et al. (1996: 236), citing Gleick (1996) for afigure of 20 m3/person/year, or Clarke (1993: 20), estimating 35 m3/person/year. The World Bank (1992: 99)notes, in regard to health and the fourth component of basic needs, sanitation, that ‘the use of water for personalhygiene usually increases only when availability rises to about 50 litres/person/day’ (18 m3/person/year), whichseems to imply that something less than this is the amount needed for the first three basic needs. The most basicneed of all, drinking, has an extremely small impact on demand, accounting for some 0.4–0.7 m3/person/yeardepending on location.



Table 5.2 also shows that, around theworld, the gross amounts of water extractedfor agriculture are quite similar. Here againwe see that, although irrigated agriculture isoften blamed for water scarcity in develop-ing countries, the amounts used for agricul-ture in poor countries are lower than in richones. The differences are small, and proba-bly not meaningful in view of the generaldoubts about data validity; but, as far as thedata go, they do not indicate that agriculturalusers are more wasteful of water in poorcountries than in rich ones.

We can also relate these figures to basichuman needs. At the minimal nutritionallevels required for sustaining human health,in a society where the basic food is a cerealcrop such as rice or wheat or maize, it takesa quantity of water in the order of 300 m3/person/year to grow that food, if the water isapplied very efficiently.

Of course that figure does not reflectdirectly the abstractions from the naturalriver systems, since much of the crop waterrequirement is supplied by rain, and alsomost of the irrigation water is not appliedat highest efficiency; nevertheless, the needto satisfy a certain basic food requirementapplies to us all, and it is useful to note thatthis need is in the order of ten times biggerthan our basic need for domestic water, andin the order of 500 times bigger than ourneed for water to drink.

These widely differing patterns of waterabstraction and use have various implica-tions for the effectiveness of financial poli-cies as instruments for influencing wateruses and demands. In the affluent countries,where personal domestic consumption isvery much more than the amount requiredfor satisfying the four basic needs, a chargingpolicy may have significant impacts onconsumption. A high charge may makepeople reduce frivolous or non-essentialuses of water, or may just make them moreconscious of water costs, and thereforeinduce them to change their behaviour, inways such as becoming quicker to attend toleakages.

But in countries, or families, wheredomestic consumption consists simply ofsatisfying basic needs, a charging policy

is not so likely to have those impacts.Supplying a basic need such as drinking isnot something about which the ‘user’ canexercise much choice. If the cost increases,the basic consumption will have to stayroughly the same. Any financial effecton that user will appear in some otherdirection, by not spending on some otheritem that must appear as less vital.

The nearer a country, or a village com-munity, is to poverty or to an elementarysubsistence level, the more this argumentwill apply to the case of agriculture. Peoplewill not feel that they have the optionto reduce their consumption, since foodis a basic need to just the same degree asdrinking water. So, while the potential effectof water charges in constraining consump-tion may be quite large in rich countries,it may be very small in poor countries.

5.2 Financial Productivity of Water

When we look at the productive applica-tions of water, we can find many illustra-tions of the relatively low financialproductivity of agricultural water. Forexample, Schiffler and others (1994)analysed the economics of water uses inJordan, a country with one of the world’slowest levels of water resources per person.They reported that the average productivityof water in industry was 11.2 dinars/m3

(about US$16.8/m3 at the bank exchangerate then) whereas the average productivityof water in agriculture was 0.28 dinars/m3,or 2.5% of the industrial level.

Within the agricultural sector, thereare further huge variations of productivity.The productivity of basic cereal crops in thedeveloping countries is usually around theequivalent of a few US cents per cubic metre,while fruit and vegetable crops, especiallythose for export, may show productivity asmuch as 100 times greater than the cereals.In the Jordan case, Schiffler et al. (1994)found that the productivity of grapes was130 times more than that of wheat.

So here again we have the problemof basic needs. In the poorer countries, or

Financing River Basin Organizations 77



poorer environments within any country,these low-productivity cereal crops – rice,wheat and maize especially – dominate theagricultural scene, and are not necessarilygrown for the market. In studies of five smallirrigation systems reported by PMI-BurkinaFaso (1997), it was found that the proportionof products marketed was 25.6%. In the sys-tem with least road access, this fell as low as5.3%. The rest was household consumption.

To the users in those villages, thefinancial productivity of water must seem anutterly irrelevant concept. The idea makessense (to the user of water) only in the con-text of marketable alternatives. Of coursethere is a governmental or ‘public interest’viewpoint that may suggest something else,concerning optimal uses of a scarce nationalresource. But if we feel that we may bemoving towards any kind of user-basedfinancing system, it seems that we have totry to understand how these things seemfrom the users’ perspective.

Briscoe (1997: 341–342) has emphasizedparticularly the idea that there can be ahigh opportunity cost associated with agri-cultural uses of water. The nub of this argu-ment is that, where water is not abundant,low-productivity applications of it, forexample to grow cereal crops, deny thatwater to higher-value potential users. Hepoints out that prices charged to agriculturalusers are, typically, around 10% of thosecharged to urban and industrial users forcomparable volumes delivered.

This is an argument that is easier toact upon in a mixed agricultural/industrialeconomy (or ‘middle-income’ economy),such as some of those in South America; butit seems to carry less weight in countrieswhere 89% of the utilization is agriculturaland the opportunities for transfers of useare correspondingly few. The concept ofopportunity cost depends on the existenceof such opportunities.

On the other hand, it is not safe forpoor countries to treat the opportunity costargument as irrelevant to their situation.In recent times, several countries in eastand South-east Asia (Thailand is an exam-ple) experienced rapid economic growth,continuous over more than a decade. A

consequence of this was the arrival ofmany new investment opportunities, someof which would depend on transferringwater from a traditional low-productivityuse into one of the new uses where its value(in economic productivity terms) wouldbe some orders of magnitude more. In theabsence of sound institutional mechanismsfor responding to these opportunitiesthrough orderly, voluntary transfers andcompensation of the prior users, thesechanges have occurred, but sometimes athigh social cost.

Perhaps the productivity dimensioncan be summed up somewhat as follows(referring in particular to countries in thelower parts of the national wealth tables):people are not ready to treat the major usesof water – domestic, agricultural, industrial– on a basis of simple financial equivalence.There is a quite widespread view thatagricultural and (especially) domestic usesshould not be compared against industrialuses in revenue-per-cubic-metre terms.Even if there is acceptance that agriculturalwater is often distributed and appliedinefficiently, the idea that water used agri-culturally should be transferable to indus-trial use, just through market pressures,does not seem to be generally acceptable atthis time. Political resistance to introductionof agricultural water charges, in democraticdeveloping countries such as Sri Lanka andThailand has provided evidence of this inrecent years. On the other hand, as agricul-ture moves from subsistence to commercialmodes of production, this attitude will nodoubt undergo some change.

5.3 Components of Costs

The financing question, for river-basinorganizations, depends of course on thetasks that each country may decide itsriver-basin organizations should perform.The scope of basin organizations falls intothree broad categories, which may overlapin some countries:

• Regulatory organizations: organizationswhich supervise, regulate access and

78 C.L. Abernethy



monitor the management of water, andmake rules which service-providingorganizations have to follow, but haveno direct role in service provision;

• Infrastructure managers: organizationswhich own the principal structures andfacilities for water supply, and investin new ones, but do not provide watersupply services directly;

• Service providers: organizations whichdeliver water supply services directlyto users or customers.

It may appear that the first of these, regula-tion only, is a relatively cheap alternative. Ifwe adopt that kind of organization, it mayseem that the budgetary issue will be smalland easily manageable. But that is notthe case. If regulation is to be done well,it needs a significant amount of finance.A short list of the primary regulatoryfunctions would include:

• monitoring of the quantity and qualityof water in all rivers and other naturalwater bodies in the basin;

• conserving and protecting thewatershed;

• making rules about abstraction, uses,disposal and pollution;

• supervising the application of a systemof water abstraction rights or licences;

• ensuring compliance with rulesthrough monitoring of activities,public information programmes, courtprocesses, etc.; and

• conducting transparency and account-ability programmes, to ensure wide andcontinuous understanding, acceptanceand support of basin managementprinciples.

These tasks amount to a quite formidablefinancial commitment. They are mosturgent in basins where water resources perperson are already low. On the other hand,an organization that has no service deliveryfunction does not have a direct customerbase from which a proportion of funds canbe sought. These considerations show thatthe design of basin organizations cannot beseparated from the question of how theywill be financed, at a level sufficient for

them to discharge the tasks that areassigned to them.

The movement towards establishingriver-basin management organizations iscoming at a time when governments, in bothrich and poor countries, have been trying toreduce the amounts they budget for provid-ing water services; so the idea that theseorganizations might be funded from thebudget of a central government ministry maynot be received well in many countries. Inthe developing countries, the main featureof this trend has been the numerous pro-grammes of irrigation management transfer,which began in a few countries such as thePhilippines and Colombia in the middle1970s and have since become very general,indicating a widespread perception thatcentral subsidizing of water services isdifficult to sustain.

The experience of irrigation manage-ment transfers in the past 20 years hashowever shown some of the difficultiesthat can occur when governments try totransfer the responsibility for certain tasksand their related expenditures from aservice-providing organization (such as agovernment irrigation department) to anorganization of service-receivers.

We can distinguish four kinds of coststhat are faced by water organizations whichare service providers:

• capital investment (constructing facili-ties for capture, conveyance and distri-bution of water; purchasing equipment;providing the buildings and other hard-ware of the management systems);

• major repairs and renewals of equip-ment and infrastructure;

• direct recurring costs (operation andmaintenance);

• overhead costs (sustaining an adminis-trative structure, including probablyhigher and remoter organizationallevels, national, regional, etc.).

An economist might say that the first twoof these belong together as one ‘capital’category. But it seems better to make themdistinct, as they usually happen far apartin time, and by the time that the need forrenewals becomes urgent, the fact of the




initial capital investment has usuallycaused great changes in the economiccondition of the users.

In irrigation management transfer,governments typically aim to transfer toorganizations of the service receivers theresponsibility for some or all of the third costcategory, operation and maintenance, butusually the first and fourth categories arenot transferred. The responsibility for thesecond category is often left unclear, and hasbeen a source of problems in a number ofsuch transfer programmes, because it createsdoubt about the borderlines between the twoparties.

The overhead costs of governmentalirrigation organizations are not often dis-cussed in the relevant literature. This couldbe because they are very large. Especiallyin Asia, government irrigation organizationsare among the strongest and most long-livedorganizations, and have developed largesuperstructures, often based in capital citiesfar away from their client populations. Thisseems to make the overhead cost a specialone, which is not likely to be transferable tothe individual end-users.

5.4 Regulation and Service-deliveryFunctions

Let us look now at the three different modesof basin organization which were identifiedat the beginning of the preceding section.In developing countries, we can usuallyfind existing organizations that exercisethe functions of service delivery for eachspecific use category. These are often quiteold organizations, which have developed avariety of specialist skills and have largeprofessional work forces. It seems unlikelythat governments will abolish them. Itseems unlikely, therefore, that basin man-agement organizations will evolve towardsdirect service provision to the ordinary citi-zens. A more probable path of evolutionwill be towards basin organizations takingup the regulatory functions, while directservice delivery will remain the task ofother organizations which manage urban

water supply, agricultural water supply,hydropower and other specific services topeople, to companies and to other userorganizations.

In theory, then, the service-providingorganizations should become more clearlyservice-oriented, should behave more com-mercially, should become more subject tocompliance with laws about pollution andother adverse social consequences of theiractivities, and (depending on the politics ofthe country) may be considered for privat-ization; while the regulatory organizationexists in the public domain to ensure goodlaws, allocation of resources by administer-ing a water-rights or licence system, conser-vation and protection of water sources, andcompliance with all of this.

That still leaves open the very difficultquestion of who should undertake new capi-tal investments for infrastructure provision.Will it be the service providers, or the basinorganizations? There are strong argumentsboth ways. But it seems clear that this issuewill be a vital one in determining the charac-ter of a basin organization, and its relation-ships with service-providing organizations.If basin organizations are going to beconstructors of major new facilities, theirfinancial requirements will be much heavierthan if they are purely regulators.

It seems that the primary reason why weneed basin organizations, as the prospectsof water deficits appear in an increasingnumber of countries, is for establishing com-pliance with a body of rules that will enablethe people at large, through institutions, toregain some kind of control over the dimin-ishing quality and quantity of water in theirrivers. If we take that view, then perhaps wewill think that this is a sufficiently huge andimportant task, and that we should not givethe same organization more conventionaltasks, such as construction of major facili-ties, or even ownership of facilities that existalready.

One of the problems of establishingbasin-management organizations is thatthere is clearly a potential conflict withexisting organs of local government, whichalmost everywhere have boundaries that aredifferent from the boundaries of river basins.

80 C.L. Abernethy



It is said sometimes that, since provinces orother local government units are responsiblefor achieving development within theirspecific boundaries, they must have controlover such a major development factor aswater. There is certainly much force inthat argument. However, the separationof regulatory and service-delivery functionsopens a way to escape from this problem. Itis possible to organize regulatory functionson a river basin (or aquifer) basis, whileservice-delivery functions can be organizedon a different basis which may conformmore closely to the boundaries of local orprovincial governments.

5.5 Sources of Funds

The Dublin and Rio de Janeiro Conferencesof 1992 enjoined us to regard water as aneconomic good (International Conferenceon Water and Environment, 1992). Thatseems to mean that users of it should payfor it according to the amount that theyuse. The way ahead, according to this view,seems to involve finance coming primarilyfrom users of water, paying to service-providing organizations. In that pattern,it would seem practicable to financeregulatory basin organizations throughsome system of levies on the income of theservice-providers.

On the other hand, it is difficult to seethat service-providing organizations, in thepoorer countries, are going to be able tobehave commercially, and at the same timeinvest substantially in new capital facilities.The low profitability of the prime user, irri-gated agriculture, together with the socio-political resistances to full-cost charges forbasic needs (discussed above in Section 5.2)indicate that, for many countries, this is notan immediate prospect.

Probably, too, the phrase ‘economicgood’ suggests that the prices we pay forwater services should somehow reflect thesort of factors that usually influence pricesof other economic goods. For example, ifwater is an economic good we might expectits price to rise in times and places of

scarcity, and to fall in circumstances of poorquality.

Concepts like this, however satisfactorythey are economically, face many difficul-ties from social and political angles. Waterhas been treated for long as an aspect ofwelfare provision, and in many places longperiods of provision of agricultural waterat zero price, or extremely low price, havepromoted high effective demand, which isnow very difficult to reduce.

However, there seem to be few alter-native routes available for financing theactivities. Either they must be financed fromuser charges, or they must be financed fromcentral government budgets, or they willprobably not happen effectively at all. Theproblem of central government funding, forthe poorer countries at least, is that there isvery little of it available, and there is strongcompetition for that little amount. We cansee from the fate of (for example) hydro-logical data-collecting organizations, whichin many countries have become weak andinadequate for their tasks, that centrallyfunded organizations which are doingthings that do not have direct popular appealare likely to be left on the sidelines inthe budget contest. Funding river-basinorganizations this way may well make themunstable, and unable to pursue consistentlong-range planning.

In Europe, there has been a trend inrecent years towards the use of abstractionlicences as a means of raising a significantproportion of the funds needed for sustain-ing regulatory organizations. This becomespossible when the regulatory function isclearly separated from the service-provisionfunction. Regulatory organizations assessthe available quantities of water and issuelicences accordingly. In that system, theservice-providing organization can be justanother holder of an abstraction licence.

Abstraction fees are not the same as userfees. A service-providing organization maypay abstraction fees to the regulatory organi-zation, and then sell the water to ordinarypeople or businesses, charging them a userfee, which exceeds the abstraction fee inorder to cover the costs and financial risksof delivering the water. In such systems,




abstraction licence fees may be gradedaccording to scarcity.

Buckland and Zabel (1998) describethe workings of these systems, and reportabstraction fees that are typically around theequivalent of 1–2 US cents/m3, but in somecases significantly more. In some countriesthe product of abstraction fees is sufficient tocover the cost of all regulatory functions.

In a licence fee system, there are twoways of charging the user. The charge may bebased on the measured actual consumptionof water, or it may be based on the amountallowed by the licence. The system of mea-suring actual amounts involves a higherlevel of regular metering of the users,whereas the licence-amount method can beimplemented with only occasional checks,to ensure that the conditions of the licenceare not exceeded. Dual systems, combininga fixed quota and a volumetric charge forexcess over this, are also used (see forexample, Tardieu, 2001).

If the cost of abstraction licences is sethigh enough, they can have an effect on theconsumption of water. In the German stateof Hamburg, for example, a relatively highabstraction fee for ground-water licencescaused about a third of the licences to berenounced, and handed back to the regula-tory organization, which could then re-issuethem to others (Buckland and Zabel, 1998:270).

There are other possible sources ofrevenue for a basin organization. We cannote three principal areas:

• waste-water disposal charges;• pollution charges or fines;• charges for permits for other water-

based activities, such as fishing,navigation, recreation, etc.

Many of the charging methods noted abovecan be levied by service-providers, butwould not easily be collected by a purelyregulatory organization, since it has noretail delivery function and is not in regulardirect contact with household and farmusers of water. Therefore, if we want toestablish a basin organization of theregulatory type, it seems that we shouldseek financing methods that are based

mainly on the relationship betweenthe regulator and the service-providingorganizations.

The problem of dealing with numerousindividual small users of water is a realsource of difficulty in many developingcountries. Licence systems, such as forexample licences to install and operateground-water wells, should therefore beintroduced only after careful preparation, asthey may not be respected if the capacity toensure compliance does not exist. However,licensing systems do not involve the sameamount of administrative work as volumet-ric use fees, and (subject to the condition ofadequate enforcement capacity) can have auseful place in the scheme of regulatoryfinance.

The principal types of mechanismfor funding a purely regulatory basinorganization therefore appear to be:

• abstraction licence fees,• waste-disposal permit fees,• pollution charges and fines,• permits for water-based activities,• government subsidy.

The above discussion has focused on thefunding of recurrent costs: regulatory andoperational costs. The question of sourcesof funding for capital costs presents otherissues which will not be analysed here.However it is worth noting the data pre-sented by Briscoe (1999: 307), who showedthat, in a sample of five developing coun-tries, the average share of the private sectorin water infrastructure investments wasonly about 6%, compared with an averageprivate share of about 49% in investmentsacross other infrastructure types: energy,telecommunications and transport. The lowlevel of private investment in water reflectsthe low financial returns that it produces,especially in its major developing-countryapplication, irrigated agriculture. For thisreason, we should expect that capitalinvestment in water is likely to remain pre-dominantly a public-sector responsibility,in the medium term at least. The problemsof bringing river-basin organizations intothis aspect are noted further below (Section5.8).

82 C.L. Abernethy



5.6 Methods of Assessment

Funding of regulatory organizations fromabstraction charges, direct user charges, orfrom a percentage levy on the user chargescollected by the service-providing organiza-tions does not necessarily mean that all cat-egories of users pay at similar rates. Whenwe examine current charging practices,world-wide, we find a tendency to chargeagricultural users much less, and industrialusers much more, than the average.

This leads us to the question, howshould charges be assessed? If basinorganizations are to draw their fundingultimately from user charges, how willthose charges be calculated?

This is related to other issues, about theimpacts that we may want a charging systemto have upon patterns of water consumption.It also brings in some very complicatedissues related to the quantity, quality andlocations of disposal flows, returning to thenatural system after use.

Industrial users are accustomed to payfor measured quantities of water delivered totheir premises. Urban users in the better-offsuburbs also probably pay on the basisof measured volumes, and poorer users,especially the very poor, also pay, thoughprobably not for measured flows but forvolumes brought by water-carriers.

But the biggest users in poor countries,the farmers, generally do not pay by volumeat all. In the countries where irrigationservice fees are levied, the overwhelmingmajority pay an amount that is based onland area. There are many variants of this,such as seasonal differences, crop differ-ences, and so on; but the central point aboutthe dominant current practice is that themarginal price, the cost to the farmer fortaking more water, is normally zero.

Countries vary in the way they accountfor water that flows back to the river systemsafter use. In virtually all the uses of water,there is some ‘return flow’, but the amount ofthis varies, and in many uses it is difficult tomeasure it. Briscoe (1997: 345) says that:

taking the US as an example, consumptiveuse as a percentage of withdrawals was56% for irrigation, compared with 17% for

urban water supplies, 16% for industry andjust 3% for thermoelectric power.

The UK, following the logic of these differ-ent levels of consumption and return,adopted a classification of use types intofour bands, according to their average pro-portions of return flow. In such a system,users who consume a large fraction of whatthey abstract (such as irrigation) are chargedmore heavily than users with a high returnpercentage (such as power generation).There are of course quality aspects in rela-tion to these return flows as well, which canbe dealt with by the different mechanism ofpollution charges.

In developing countries, water chargesdo not vary, generally, according to scarcityof the commodity. Water prices are usuallycalculated on some basis that is related to thecost of delivering it. That means that it staysthe same, and does not respond to variationsin available resources. In many countries,charging scales are centrally or provinciallydetermined for large sets of irrigation sys-tems, so that systems with water abundanceand systems with local water scarcity areobliged to charge their users the same price.

Indeed, when we look at inter-countrycomparisons, although the variations incharging practices are enormous, there aresigns of a correlation between scarcity andprice policy, but it is a correlation that isopposite to economic logic. Some of thelowest charges (even when comparisons arebased on purchasing power parity) are foundin dry countries such as Egypt or Iran, whilehigh charges can be found in much wetterplaces. This presumably represents a socio-political logic, which may well be strongerthan economics.

There are even cases where the cost oftaking water for irrigation becomes lower inthe driest, hottest time of the year. The mid-dle Niger river is such a case, simply becauseat that time the river level is comparativelyhigh and the cost of pumping water toadjacent land is therefore less. This kind ofanomaly results from basing charges only onthe cost of service delivery, which is effec-tively unrelated to scarcity, and often is onlyweakly related to quality, or to demand.




A river-basin organization could reducesome of these anomalies. In many countries,water has been made legally the property ofthe state. It is possible, therefore, for a river-basin organization to charge the service-providing organizations on the basis of themeasured amounts that they extract from thenatural system (as in the European examplesof abstraction licences, noted earlier). Eachbasin organization can devise its own levelof charge, related to the amounts of waterthat it has available for abstraction. It is pos-sible for those charges to be varied alongsome seasonal or even monthly scale. In thisway, a basin organization could exert somepressure on the service-delivery organiza-tions to look for ways of moderating theirrates of water use, while at the same timeimproving its own financial independence.

5.7 Collection

In the agricultural sector of developingcountries, the problems of how to financeirrigation services and how to collect irriga-tion service fees from users of agriculturalwater have been prominent issues through-out the 1980s and 1990s. It cannot yet besaid that the issues have been satisfactorilyresolved. This experience should make usaware that the establishment of new basinorganizations in developing countries isgoing to face similar difficulties.

Financing urban domestic water sup-plies is not any easier than financingirrigated agriculture. The World Bank (1993:126–127) reported its experience in lendingfor water projects in these terms (referringmainly to non-agricultural uses):

The Bank has maintained the policy thatcost recovery should be sufficient to payboth for operations and maintenance andfor a fair return on capital investment . .cost recovery was rated as unsatisfactory in80 of 114 projects. And, in 78% of countriesreceiving water supply and sanitationloans, financial covenants were notfulfilled. In 49 of the 120 water supply andsanitation projects, fees were not raisedenough to meet financial requirements dueto government constraints.

We can perhaps make a guess, that theseproblems, of investment projects whosecost-recovery conditions are not imple-mented in reality, happen because suchprojects are prepared in the bureaucraticdomain, and subsequently meet strongresistances in the political domain, dueto neglect (in project preparation) of theweight of the people’s views. This can onlyincrease, as more countries are incliningtowards democratic modes of government.

The compliance problem, in respect ofirrigation service fees, became famous dur-ing the 1980s. The Philippines, especially,made the ‘viability index’ a central feature ofits institutional reforms: this index is theratio between fees collected and the costs ofoperation and maintenance. Field officialsof the government agency could receivebonuses depending on the percentage offees actually collected. (Oorthuizen andKloezen, 1995: 18; Svendsen, 1992: 5). Thisstrategy addressed financial viability, buthas been criticised by various observersas having a negative effect on (especially)maintenance, since reduction of expendi-ture was one way of enhancing the viabilityindex, at least temporarily.

Studies of the costs of fee collectionshow that they can be a significant propor-tion of the total amount collected. ‘Passive’collection, meaning the kind of systemwhere each user is expected to bring thefee to the collection office, seems prone toabuses, or at least to long delays in payment,which present serious cash-flow difficultiesto service-providing organizations. ‘Active’fee collection, using paid collectors whovisit houses or farms, incurs a significantwage cost. Both methods need accountingstaff and certain facilities.

It is hard to make any general statementabout the proportion of fee income thathas to be used for fee-collection (that is, toensure that the organization’s income flowwill be stabilised) because of the inherentvariability of the process. For example,Oorthuizen and Kloezen (1995: 28) foundthat the average cost of fee collection in asmall Philippine system was 10.8% of theorganizational income, but even in that onesystem this percentage varied between 5%

84 C.L. Abernethy



and 15% over their 6 years of observations.These percentages are clearly linked to thedegrees of motivation and energy in thecollecting organization; this is doubtlesswhy the Philippines introduced collectionincentives. In other cases there may be indif-ference within the bureaucracies, and resis-tance among the water users; both of thesewould be likely to result in collection costsbeing substantially higher in proportion toincome. Cases have even been noted wherecollection cost is of the same order as theincome that the collection yields.

In the Niger systems passive collectionwas used. Costs of collection were stillsignificant, as there must be an accountingsystem to register the collections. Non-payment was avoided by careful recordingof each individual’s arrears and, ultimately,withdrawal of service from users in severearrears; but the level of arrears neverthelesscaused frequent cash flow crises (Abernethyet al., 2000).

How will river-basin organizations min-imize these linked problems of compliance,delayed payment, and collection cost? Theanswer to this seems to be (as for some of theother issues raised above) that the separationof regulatory from service-delivery func-tions should substantially reduce thisproblem. The service deliverer must have adirect relationship with the water users, andindeed the trend towards user-controlledservice organizations assists this. The regu-latory organization on the other hand hasdifferent duties, and should collect itsfees from a few major sources, principallythe service-delivery organizations, but alsoincluding any others to which it grants per-mission for abstractions, pollution permits,or other water-related activities. On thewhole, passive modes of collection may besufficient for this.

5.8 Control of Expenditure

As river-basin organizations come into exis-tence in an increasing number of countries,we will face another kind of issue: howwill their expenditure levels be controlled?

These organizations should become, asfar as possible, responsive to the interestsof their own stake-holders. But the stake-holders are very diverse. Everybody is awater user; and most of us are water usersin several different modes. Some may wantnew storage facilities to be built; otherswould prefer that costs be kept down as faras possible. The interests of birds, fish andother wildlife have to be accommodatedsomehow, along with other non-economicaspects of water, such as landscape beauty,waterfalls and the like. All of these thingstend to have cost connotations in some way.

These matters cannot be satisfactorilyresolved by creating river-basin organiza-tions that are firmly embedded in govern-mental bureaucracy. A different and moreresponsive kind of organization is needed,which will be accountable to some councilin which all principal stake-holders have avoice.

In these conditions, control of expendi-ture can be done transparently, with budgetsapproved in advance not by finance minis-tries, but by the people of the basin who willhave to bear most of the costs and receive theconsequent benefits.

The problems of expenditure controlprovide a strong reason why, at least in theinitial stages, new river-basin organizationsshould not be asked to develop functionsthat are too large. Stake-holder councils andother similar institutions are likely to berather meaningless and ineffective if they donot have the general right to formulate anannual budget, or at least to modify andapprove (or reject) budget plans proposedto them by some part of the governmentbureaucracy. On the other hand, it is likelythat a considerable time will pass beforesuch councils are trusted to formulate andsupervise a large capital budget effectively.Risks of incompetent or dishonest behaviourobviously exist. Capital expenditures alsomay frequently have impacts or benefitsoutside the basin where the works areconstructed, so their financial managemententails more complex procedures.

For such reasons it seems that it wouldbe better to move along the path of develop-ing the financing of regulatory functions




first, which can be done within the basin andshould be possible within the financialresources that can be generated from thekinds of funding sources described earlier.

To ensure public confidence in themanagement of such funds, we shouldexpect that river-basin organizations willsubmit to some process of public audit,according to whatever the relevant nationalsystems may be, perhaps sharing the sameprocesses as are applied to (for example)public auditing of the accounts of localgovernment bodies.

5.9 Impacts of Charging

The Economist (25 March 2000: 84), report-ing after the World Water Forum, said,

whether it is Australia or Rajasthan, oncepeople understand the true cost of waterservices, they will conserve water, and evenhelp to dig ditches if necessary. In return,they will want transparent prices and betterservice from both governments and privatefirms.

That is a clear statement of the standardeconomic view of the impacts of watercharges: consumption will reduce, capitalcosts will be partially taken up by users,and customer pressures will cause organi-zational behaviour to improve. Is it true?

It is quite difficult to reconcile thisoptimistic view of the power of economics,with the findings from the detailed work ofPMI-Niger (Abernethy et al., 2000). There, ina country that is at or near the bottom of theper capita wealth scale and other humandevelopment scales, irrigation service feesare among the highest in the world. On thewhole, fee collection rates are high. Farmerspay 20% of their gross crop value in fees. Ifthe foregoing quotation is true, water shouldbe used very efficiently in these circum-stances. But the water productivity wasfound to be equivalent only to 20 US cents/m3, in terms of gross product value, at pur-chasing power parity (less than 5 US cents/m3, at nominal bank exchange rate). No signsof reduction in water consumption could bedetected over ten seasons of monitoring.

There seem to be three sources of thedifference between these observations andthe view quoted earlier. First, The Econo-mist was drawing lessons primarily fromurban cases. Secondly, water is only oneinput to a production process, whether inagriculture or industry, and it is generallynot a replaceable input; so if the user thinksthat more water is necessary in order torealize the benefits of other inputs, that userwill probably apply the extra water. But itseems likely that the third reason is the mostinfluential. This is that farmers in Niger, asin most other developing countries, do notpay for the quantity of water they use. Theypay heavily, but the charge is area-based, sothe marginal cost of taking more water iszero. In the terms of the above quotation, theirrigation users may not come to ‘understandthe true cost of water services’. When themarginal price paid for additional supply iszero, the price system sends no signal to theusers concerning the true cost of their con-sumption. Yet area-based irrigation charges,which produce this ineffective result, are themost usual kind in developing countries.

Urban users, whose quantitative needsare smaller and more measurable, normallyare not in that situation. For them, themarginal cost of increasing their usageof water may be quite high.

This problem, that the major users ofwater (farmers) have no direct incentiveto reduce their consumption, is not likely tochange in the near future. Although we canmeasure pipe flow volumes acceptably, andthe equipment cost for doing so is quite tol-erable, devices for measuring flow volumes(as distinct from flow rates) in open chan-nels are not available at the scales and costsrequired for the small land units typical ofdeveloping countries, especially in Asia. Sothe impacts of charging in metered urbansystems can be quite different from those insmall-holder irrigation.

One proposed approach to this problemis by charging, not to individuals, but togroups, for example to all the farmers along asingle common channel. There is as yet littleevidence that this is effective. At the level ofthe individual, it does not alter the incentivemuch. If there are 50 farmers sharing a

86 C.L. Abernethy



metered source, each may calculate that, bytaking an extra cubic metre of water he or shewill obtain all the benefit of using it, but willpay only one-50th of its cost.

For a financial system to have a strongimpact on water abstractions, it must alsobe designed to give incentive to the service-delivery organizations to reduce the convey-ance losses in their systems. Both urbanwater-supply organizations and irrigationdepartments have until recently shown poorrecords and lack of concern about reducinglosses. This is another area where separationof river-basin management from servicedelivery is helpful.

If the service-delivery organization hasto buy the right to abstract water, it will bemore strongly motivated to ensure that asmuch of that water as possible reaches acustomer who pays for it. Water leaking froma canal or a pipe, in that system, means adirect financial loss to the service-deliveryorganization. Organizational separation alsomakes it easier to include in the financingarrangements some reflection of the valueof leakage water that can be recovered bypumping from aquifers, which varies greatlyaccording to factors like location, qualityand aquifer depth.

There is also the problem of equity.As we move more towards the principleof payment for water services, can we feelsure that the poorer sections of societywill have adequate access to water? Briscoe(1997: 349) put this problem clearly:

The inequities of existing command-and-control mechanisms for water allocation inirrigated agriculture have been widely doc-umented . . . Because water has rarely beenformally managed as an economic good indeveloping countries, however, there islittle information on the equity effects ofa market-oriented management system.

The problems of inequity are particularlyacute in urban water supply systems. Oneof the results of rapid urbanization in mod-ern times has been that cities have grownmuch faster than the administrative andfinancial capacities for providing them withcomprehensive public utilities. In many cit-ies, therefore, piped water is now provided

in only a minor proportion of households,and many of the other households, unlessthey can install their own wells, mustrely on small water vendors. Bhatia andFalkenmark (1992) gave data on the ratiobetween prices per litre charged by suchvendors, and prices charged by public utili-ties, in a large range of cities. Their datashow that, on average, the vendors’ priceswere about 16 times greater than the pricescharged by the public organizations; andthis ratio ranged up to 100 in very poorenvironments such as Port-au-Prince (Haiti)and Nouakchott (Mauritania). We mayguess too that the water supplied by thevendors is generally of substantially lowerquality.

This kind of information is not evidencethat a ‘market-oriented management system’will have necessarily bad effects for thepoor. The vendors may not be over-charging:they have costs to meet and transport andlabour to arrange, as well as the primary costof the water. These enormous price ratiosrather show the scale of under-investment inextending the public piped water systems.The under-investment in turn may belinked to inadequate prices charged forthe products of those systems.

We can, however, feel relativelyconfident that there are better chances forrestoring equity, under river-basin manage-ment organizations, than under the presentsystems of management. Traditional waterrights have been rapidly eroded by thepolitical and economic changes of the pasttwo or three decades, and relatively fewcountries have succeeded in supporting thetraditional systems, or in replacing them bymodern systems based on water rights thatare legally enforceable by their users. Basinmanagement offers a way of redressing thissituation, either through rights or licences. Itwould seem reasonable to accept the needfor some payment, or increase of existingpayments, in return for a better guarantee ofsupply or abstraction rights.

However we should also note the needfor good, transparent public informationprogrammes when such a policy change isunder consideration. If public opinion is notprepared for changes of traditional patterns,




and not informed about the benefits thatthey are intended to bring, they are likely tobe rejected.

Charges, or increases of charges, arenever going to be popular. It is futile to hopefor that. However sound our economic logic,however much we may feel that a charge sys-tem can reduce distributional inequities, orimprove water-use efficiencies, there willnot be demonstrators in the streets demand-ing the introduction of such charges. It isgood to keep this point in mind, as we thinkabout possible beneficial impacts of charg-ing policies. On the contrary, if charging sys-tems are introduced where there were nonebefore, the idea will almost certainly pro-voke a strong adverse reaction, and it musttherefore be preceded by transparent expla-nations to the people about the benefits thatare intended to be derived from the policy.

Yet there are strong arguments forsaying that the interests of poor people canbe better protected under a stronger systemof financial management, which can becombined with high charges for heavy uses,financial penalties for pollution and othertypes of social costs, stabilizing of accessand use rights, control of ground-waterdepletion, and a number of other desirableoutcomes which at present countries strug-gle to achieve. The interests of the poorcan be protected to the extent of some cross-subsidisation, introducing minimal pricesfor small domestic basic-needs quantities, oreven as far as the zero rate for the first 6 m3/person/month which is proposed in SouthAfrica.

We should note however that cross-sub-sidization (supplying to some customers atprices that are below true costs, or evenzero, and balancing the accounts by chargingprices above true costs to others) is not in thelong-term genuine equity, although it maybe justified occasionally in special circum-stances such as those South Africa faces.Non-equitable policies of this sort need togain the support of a consensus of the waterusers of a basin, just as much as other watermanagement policies.

The equity question can be regardedas yet another argument for keeping aseparation between the regulatory and

service-delivery functions. Research onequity effects will probably continue to benecessary for quite a long time. River-basinorganizations should take some responsibil-ity for monitoring these effects, and forencouraging the necessary research. Theyshould be in the position to adjust their regu-lations, and the constraints on the service-delivery organizations, so as to take accountof the need to limit the degree of inequitythat may exist, and particularly to ensureaccess for all up to a certain basic level.

5.10 Conclusions, and a PossibleWay Forward

River-basin organizations offer a promisingway towards better and more equitablemanagement of water resources. They needto be adequately financed, and it is betterthat their finance should be generatedlocally from among the users of their ser-vices, who should also have an effectivevoice in influencing their policies, thanthat their finance and policy should bedetermined centrally.

The ways of generating sufficientfinance, in the case of poorer countries, arenot yet certain, because of the weak financialsituation of agricultural users, who accountfor the overwhelming majority of waterconsumption in most countries. The lessonsthat can be learned from the financingsystems of richer countries are of limitedrelevance because of the different balanceof user types.

The financing system has a strongeffect on the behaviour of an organization, sothe financing system for a new river-basinorganization should be designed not only onaccounting principles, but also with a clearview of the behaviour we want. The desir-able characteristics of a river-basin organiza-tion could be summarized as follows:

• Its primary goal is to provide a frame-work of legality and security for all theuses of water, including environmentalas well as human uses.

• A river-basin organization shouldhave no function of delivering water

88 C.L. Abernethy



supplies or services, since one of itstasks will be to monitor and control theactivities of service-providers.

• Its primary funding mechanism shouldbe through licences for abstractionsand disposals of water, supported tosome extent by penalties for pollution,excessive use and other unwantedbehaviour, and permits for non-abstractive uses.

• Its charging structure should reflectthe scarcity and quality of the basin’swater.

• Its charges should be structured toensure that basic human needs can besatisfied cheaply.

The possible impacts of water charges onthe behaviour of water users include these:

• Consumption reduces;• Water-saving behaviour increases;• Organizations perform better;• Users contribute to the cost of capital

works;• Equity of access and of use changes.

As we have seen, the actual impactsdepend significantly on the precise mode ofapplication of the charging system, whichmust be designed carefully or it may notyield the desired outcomes.

Current systems of charging for waterservices have many defects. Irrigationcharges are usually area-based, not volume-based, so they give no incentive for watersaving, neither for the service-deliveryorganization to reduce leakages nor for theend-user to improve application efficiency.Charging rates are usually calculated on thebasis of the cost of providing water deliveryservice, and sustaining a supply organiza-tion, but do not often reflect the scarcityor abundance of the water resource, or thequality of the water. Often economic logicis reversed, as poor city-dwellers pay morefor low-quality domestic water, and farmerspay a marginal price of zero for waterapplications in excess of real need.

Systems of abstraction licences maybe the most easily implemented methodof addressing simultaneously these variousproblems, especially:

• to sustain the kind of river-basinorganization that is needed;

• to give firm legality to long-standingtraditional users;

• to protect principles of equity duringrapid socio-economic changes; and

• to make possible more flexible systemsof charging that will reflect scarcity andquality, and will follow some progres-sive scale so that basic needs (both per-sonal water use and food production)can be satisfied at rates that are lessthan those charged for levels of usewhich exceed the fundamental humanrequirements.

Figures 5.1 and 3.2 represent the kind ofpattern that is implied by these arguments.Figure 5.1 is what we may call a ‘traditionalsystem’ in many developing countries.A number of service-providing agencies,which each have a basic single servicefunction (irrigation department; electricityauthority; water supply board, and so on)exist, independent of each other and eachunder the patronage of a ministry. (Theremay also be large industrial companiesin the private sector, which are allowedto perform ‘self-service’, abstracting, usingand disposing of water for their own func-tions; there may also be companies whichare privatized water-service providers.) Theservice-providing entities are financed bysome combination of government subsidiesflowing down from the parent ministry andservice-related fees flowing up from theusers. The ratio between these downwardand upward components varies amongcountries, and is changing over time: 20years ago, the upward flows were generallymuch smaller than now, and the downwardsubsidy flows were much greater.

The pattern shown in Fig. 5.1 does notlend itself easily to integrated management,as many countries have found. The agenciesand their patron ministries tend to showrivalry rather than co-operation. Withoutintegrating the actions of these service-providing agencies, there can be little pros-pect of bringing about coherent managementand socially acceptable principles of alloca-tion of the water resource, and disposal of




it after use. Yet it is extremely difficult toenvisage dismantling these long-establishedstructures, and it is not at all sure that thedisruption involved would be beneficial.

Figure 5.2 indicates a ‘new’ system,leading towards a gradual integration. Theexisting system of organizations and finan-cial flows can be left in place. Alongside it isthe river-basin organization, which is givenauthority by the state to issue licences forwater abstractions and disposals, and tocharge appropriately for such licences.Initially the licence pattern will no doubtcorrespond closely to the existing pattern ofuses. The new system however changes theprocesses by which the existing situationwill evolve in future, and makes it morelikely that the allocations of water can bebrought into balance with future needs andwith environmental goals.

The financial flows to the new organiza-tion would, in this model, come initiallyfrom charges on the service-providing orga-nizations, including ‘self-serving’ industriesand irrigation schemes. The flows would beexpected to expand at a relatively early stageto include some flows from individuals,for example fees for licences to installnew ground-water wells, and gradually tobroaden their revenue sources by takingup some of the other financing optionsmentioned earlier, but leaving the primaryrevenue that can be derived from user fees tobe collected by service-providers.

To ensure that this system responds tothe general goals of the community, greatcare is needed in the design of its organiza-tion, its procedures of consultation, democ-racy, equity and so forth. Those importantfeatures are beyond the scope of this chapter;

90 C.L. Abernethy

Fig. 5.1. A common pattern of financial flows.



but the need for harmonizing the financialprocedures with those social dimensions isclear.

References

Abernethy, C.L., Sally, H., Lonsway, K. andMaman, C. (2000) Farmer-based Financingof Operations in the Niger Valley IrrigationSchemes. Research Report 37, InternationalWater Management Institute, Colombo,Sri Lanka.

Bhatia, R. and Falkenmark, M. (1992) Waterresource policies and the urban poor :innovative approaches and policyimperatives. International Conference onWater and Environment, Dublin, Ireland.

Briscoe, J. (1997) Managing water as an economicgood. In: Kay, M., Franks, T. and Smith, L.(eds) Water: Economics, Management andDemand. International Commission for

Irrigation and Drainage/E. and F. N. Spon,London.

Briscoe, J. (1999) The changing face of waterinfrastructure financing in developingcountries. Water Resources Development 15,301–308.

Buckland, J. and Zabel, T. (1998) Economicand financial aspects of water managementpolicies. In: Correia, F.N. (ed.) Selected Issuesin Water Resources Management in Europe,Vol. 2. Balkema, Dordrecht, pp. 261–317.

Clarke, R. (1993) Water: the International Crisis.Earthscan, London.

Gleick, P.H. (1996). Meeting basic human needsfor water. Stockholm Water Front 1, 6–7.

International Conference on Water and Environ-ment (1992) The Dublin Statement. WorldMeteorological Office, Geneva.

Oorthuizen, J. and Kloezen, W.H. (1995) The otherside of the coin: a case study of the impactof financial autonomy on irrigation manage-ment performance in the Philippines.Irrigation and Drainage Systems 9, 15–37.


Fig. 5.2. A future pattern of financial flows, after establishment of a river-basin organization.



Projet Management de l’Irrigation – Burkina Faso(1997). Analyse diagnostic et performancesde cinq périmètres irriguées autourde barrages au Burkina Faso. PMI – BFand International Irrigation ManagementInstitute, Ouagadougou.

Schiffler, M., Köppen, H., Lohmann, R.,Schmidt, A., Wächter, A. and Widmann, C.(1994) Water Demand Management in anArid Country : the Case of Jordan with SpecialReference to Industry. German DevelopmentInstitute, Berlin.

Seckler, D., de Silva, R. and Amarasinghe, U.(1996) The IIMI indicator of internationalwater scarcity. In: Richter, J., Wolff, P.,Franzen, H. and Heim, F. (eds) Strategiesfor Intersectoral Water Management inDeveloping Countries: Challenges andConsequences for Agriculture. DSE/ZEL,Feldafing/Zschortau, Germany.

Svendsen, M. (1992). Assessing Effects ofPolicy Change on Philippine Irrigation Per-formance. Irrigation Performance Paper 2,International Food Policy Research Institute(IFPRI), Washington DC.

Tardieu, H. (2001). Water management for irriga-tion and environment in a water stressedbasin in south-west France: charging is animportant tool, but is it sufficient? In: Aber-nethy, C.L. (ed.) Intersectoral Management ofRiver Basins. IWMI, Colombo, Sri Lanka.

World Bank (1992). Development and theEnvironment: World Development Report.Oxford University Press, New York.

World Bank (1993). Water Resources Manage-ment: a World Bank Policy Paper. WorldBank, Washington, DC.

World Bank (1999). Knowledge for Development:World Development Report, 1998–99. OxfordUniversity Press, New York.

92 C.L. Abernethy



6 Water Management for Irrigation andEnvironment in a Water-stressed Basin in

Southwest France

Henri Tardieu

6.1 Introduction

In France, the water management issue isno longer seen as a question of developingstakeholders’ participation or transferringState competence to user associations. Asfor the other countries with a completeand complex institutional framework, thechallenge now is to define clearly the roleof each water management stakeholder andto answer two remaining questions:

• How do we ensure sustainability ofthe investments by raising the price ofwater without discouraging economicdevelopment?

• How do we share water among userswhen resources are scarce?

The general answer to these questionsrelies on the two basic principles of goodwater management leading to sustainabledevelopment:

• Since it consumes more than 70% ofthe available water during low flowperiods, irrigated agriculture mustrespect the other uses by limiting itsdemand to the allocated volume;

• Since it involves large and long-termpublic investments, irrigated agri-culture must bear at least the ‘sustain-ability cost’ (Tardieu and Préfol, 2002)of the upstream water resources.

Such a general answer is of course largelycase-specific and should be adjusted toeach institutional framework.

France, like other Euro-Mediterraneancountries, has a long history of water devel-opment, born from water scarcity and a con-stant search for the best agricultural use andthe fairest sharing. A complex institutionalstructure has progressively been set up todevelop private initiatives within a publicservice framework.

During the last century, AuthorizedUser Associations (ASAs) were developed.They were formal institutions constitutedby landowners for sharing the constructionand management of irrigation systems.In the 1950s, the State created, within amore ambitious land-use planning frame-work, Regional Development Companies(SARs), which are public corporations witha ‘concession’ from the State, to developwater resources and manage irrigationschemes in the southern regions ofFrance. Well subsidized by the State atthe beginning, the SARs now cover theircosts from the payments of their customers.This management is now financially sus-tainable as it includes the provisionsnecessary to maintain the investmentsunder concession. It nevertheless keepsthe basic characteristics of a French publicservice: continuity, equity, sustainabilityand transparency.




Finally, Basin Organizations were setup more recently, with a widened mandateto include management and protection ofthe environment and seek a global consen-sus on water management by using dialogueand financial incentives, while the Statekeeps the role of regulation.

After a short discussion about thestakeholders in French irrigation and watermanagement, this paper addresses bothsocioeconomic questions stated above, witha specific discussion of the case of theNeste system, a water-stressed basin inthe southwest of France.

6.2 The Key Stakeholders in Irrigationand Water Management in France

6.2.1 The individual level: farmers

Farmers aim to satisfy the objectivesthey select for their household (to ensurea minimum revenue), their enterprise (tomaximize profits, to minimize risks andto improve the quality of the products)and their land (to be sustainable). Eachone freely chooses the crops to grow on thebasis of advice from his profession withdue consideration given to the market. Heconsequently optimizes the managementof the production factors, including theon-farm irrigation system.

The value of water in irrigated agri-culture varies, largely due to the hetero-geneity of the production systems. The costof irrigation water is generally relativelyhigh in the Mediterranean regions, implyinghigh performances with high value-addedcrops.

The constraints of agricultural competi-tiveness make the irrigator very sensitive tothe reliability of water supply and, of course,to its cost. For each crop in a given croppingpattern, water value can be assessed. Waterdemand can thus be represented by a graphof water value per unit volume. Such a graphsuggests how an irrigator reacts as the waterprice varies.

6.2.2 The small system level: ASAs

Gathering irrigators through an associationthat owns and/or manages common assetsis the first and the oldest way to managecollective irrigation (Lesbats et al., 1996).Such associations bring together the land-owners concerned with the irrigationsystem. They are self-managed organiza-tions, and, in France, are based on a legalframework developed in the 19th century.This framework provides all the authorityneeded to manage the irrigation system.

The statutes of ASAs are public andconfer the capacity to act for the public good,particularly in the matter of cost recoverywhere they follow the rules of publicaccounting. Costs are shared in proportionto the involvement of each owner in theproject area, generally as a function of hisirrigated area.

These associations have a very longlifetime, since the properties are irrevocablyengaged in the association. This long experi-ence provides valuable lessons. Initially, thecollective participation of members is exem-plary. They define their projects accordingto their needs and their means of fulfillingthem. They personally ensure operationand minor maintenance. The apparentcost, corresponding only to monetaryexpenditure, is thus largely below thecomprehensive cost of water.

However, it sometimes happens thatthe necessary solidarity decreases and theASA goes wrong by lack of involvement andlack of professionalism. Members are thenconcerned about the immediate balance ofthe accounts, and cut down the maintenanceexpenditure. This entails serious conse-quences in terms of quality and continuity ofservice.

To preserve or re-establish the durabil-ity of their systems, at the conception aswell as at the management stage, ASAs havecurrently been calling for the assistance ofthe State. The ASA statutes indeed foreseethat in case of bankruptcy of the ASA, theState representative must replace the ASAPresident. Considering the State’s other

94 H. Tardieu



involvements, ASAs are now looking forprofessional advice, particularly from theSARs (see below).

Such a complementary relationshipbetween ASAs and a technically competentbody can be organized to guarantee sustain-able management. This is the case withthe design and/or maintenance contractsoffered by Compagnie d’Aménagement desCoteaux de Gascogne (CACG), one of theSARs.

6.2.3 The large systems level: SARs

Created about 40 years ago in the southernregions of France where water was alimiting factor to development, the SARsare characterized by the originality of theirmission and statutes (Plantey et al., 1996).

Their mission, defined by the conces-sion contract with the State, deals with theimplementation and operation of hydraulicprojects necessary for the development oftheir region. Managing the conceded waterresources, they ensure the conveyance ofwater to the centres of urban and industrialconsumption, and the distribution of waterin rural irrigated areas. For this purpose,they have broad rights and obligations as‘owners’ of the works, but without the rightto sell them.

Their statutes are similar to those ofprivate companies, implying rules of soundmanagement and economic efficiency. Themajority of the shareholders are public,and so is the governance. Local authorities(Départements1 and Regions) therefore havecontrol of the strategic resource in thename of the public good for all water users.The agricultural users are represented on theBoard and participate in governance asthe SARs’ private shareholders.

In accordance with the terms of theirconcession contracts, the action of the SARsis guided by the principles of sustainablemanagement of a public service.

• Quality and continuity of waterservice;

• Equity when water is to be sharedbetween users;

• Sustainability with adequate provi-sions for long-term maintenance;

• Transparency of the management andaccountability to the Board.

The SARs have been relatively successful atbalancing resources and needs through thepractice of integrated water management.Despite the relative scarcity of the resourcein the French Mediterranean regions, watershortages and conflicts among users areno longer a major concern in the systemsmanaged by the SARs (Tardieu and Plantey,1999).

When a crisis situation arises, high-techequipment and tested methods of watersharing allow equitable management of theresource. This is typically the case for theNeste system managed by CACG.

6.2.4 The large catchment level

A Basin Committee, a sort of water parlia-ment where users, local authorities andgovernment are represented, is in charge ofconserving the water environment and ofwater management policy in one of the sixFrench large catchment basins. It develops,in collaboration with the State administra-tion, the long-term water policy plan.

Water Agencies are the executive bodiesof Basin Committees. Taxes, collected by theWater Agencies in accordance with deci-sions of the Basin Committee, discouragepolluters and consumers from polluting andconsuming. This incentive to behave in amore responsible manner is coupled with afinancial policy since the revenue from thetaxes is allocated to financial assistance forpollution abatement and for conservationand development of water resources. Forirrigation particularly, Water Agencies

Basin Management in a Water-stressed Basin 95

1 A Département is a local government unit, with population usually in the order of 500,000.



contribute to investments in modernizationand regulation, which are important sourcesof water savings.

After 30 years, this system based onthe principles of solidarity and equity (thepolluter pays and the consumer pays) is wellaccepted by the public. It should be under-stood that the French Water Agencies do nothave direct responsibilities in water systemmanagement, unlike the bodies describedabove.

6.2.5 The State level

According to the terms of the 1992 WaterLaw, it is not the State’s responsibility toensure directly the operational manage-ment of water resources, except for verylarge rivers. Its authority should guaranteethe respect of the necessary regulationof water uses, which are subject toprevious authorization. Elaborating andupdating the rules should be carried outin consensus with the members of thewater community to minimize the numberof rule-breakers.

Finally, the State is the owner oflarge hydraulic works used for irrigationpurposes and delegated to the SARs byconcession contract. As a consequence, itsupervises both the maintenance and thebest use of the assets in order to meet allwater demands.

Although this presentation of theFrench institutional framework in watermanagement is highly simplified, it doeshelp to clarify the respective roles of differ-ent stakeholders in different areas.

• Basin planning and financial policy:the Basin Committee seeks a consensusto reconcile all users, both amongthemselves and with the environment,in a global approach to water manage-ment using financial incentives.

• Operational management: the SARsmanage water resources by contractingwith users, and ensure the sustain-ability of the assets. The ASAs havealmost the same objectives but forsmaller irrigation systems.

• Regulation and law enforcement: theState sets up regulatory measures,keeping in mind both the need forconsensus and acceptance and its owncapability for applying them to allusers.

The French organizational set-up for wateris characterized by a ‘public/private’ mix.The freedom of the private initiative isbalanced by the responsibility for thepublic good. The economic efficiency ofprivate management is associated with thesustainability of the public service.

6.3 The Neste System: an Example ofControlled Water Management

Water management in the Neste systemprovides an important example of success-ful application of a set of rules, consultationmethods and high-tech controls, but alsorepresents a system facing a regulationproblem because water supply is currentlyexceeded by demand. The essential featuresof the water management agency, theCACG, that make success of ‘controlledwater management’ possible include: (i) theinstitutional originality of the SARs with apublic mission and private management;(ii) their joint experience in regional devel-opment and water management; (iii) theircapacity to carry out maintenance and assetconservation; and (iv) the practices theyfollow in pricing water in combination witha quota system. Economic analysis of thewater value in each use, particularly forirrigation, may clarify the allocation ofwater and validate the regulation toolsused in ‘controlled water management’.

6.3.1 Presentation of the Neste system

In 1990, after a serious crisis involving con-flicts between users due to water scarcity, anew management system was established(Tardieu, 1991). In operation for 10 years,it can be described as follows, with itssuccesses and limitations.

96 H. Tardieu



6.3.1.1 Location

• A 10,000-km2 basin located in thesouthwest of France with 650 mm ofrainfall, where irrigation is necessaryfor most kinds of agricultural produc-tion (Fig. 6.1);

• Surface water is the only resource forurban and industrial uses because of

lack of groundwater, and rechargedrivers (1300 km) are the commonresource for every user.

6.3.1.2 Water users

• Fish, wildlife and tourism need250 M m3/year to augment low flows;2


Fig. 6.1. Map of the Neste basin.



• 200,000 inhabitants consume 13 M m3/year;

• 51,000 irrigated ha (28,000 l/s sub-scribed by 3000 irrigators) consume(average) 70–95 M m3/year (dry years);

• A 10,000-ha waiting list for irrigationcontracts (equivalent to 6000 l/s) exists.

6.3.1.3 Water resources

• The Neste Canal (a State concession toCACG) diverts 250 M m3 of the naturalflow of the river Neste;

• Stored resources contribute 100 M m3,of which 48 M m3 are stored inmountain hydroelectric reservoirsand 52 M m3 in CACG lakes (also Stateconcessions).

6.3.1.4 Types of withdrawal

• Individual withdrawals (14,500 l/s)subscribed through ‘conventions derestitution’, or ‘pour-back contracts’;

• Collective withdrawals by ASAs orCACG irrigation networks (13,500 l/s).

6.3.1.5 Monitoring and remote control

• Supply monitored through 200 riverflow meters, 40 dam and canal gates,and 150 pumping stations underremote control;

• Demand monitored through 1500individual water meters (checked threeto four times a year), 6000 meters oncollective networks, and 150 pumpingstations under continuous monitoring.

6.3.1.6 Supply/demand balancing

• Currently the balance is ensured with afailure rate of 1 year in 4;

• For a more comfortable balance, anadditional resource of 43 M m3 isneeded for the waiting list and 7 M m3

to reduce the failure rate.

6.3.2 Management rules: the contract,individual and collective

Each user signs with CACG a water contractcalled convention de restitution guarantee-ing that his/her withdrawal is balanced outby an equivalent upstream replenishment.The contract states a maximum diversionflow and a subscribed volume, or quota,with a two-tiered price. The first price isa function of the allowed flow (?50 per l/s).The second price, the over-consumptionprice, is a function of the volume consumedabove the quota. (?0.10/m3 above the4000-m3 quota).

Thus, there are two limits on theabstraction of water by the user: a flow ratelimit, and a volume limit. If the authorizedflow rate per hectare is 1 l/s and the volumequota for the year is 4000 m3, the user may,in effect, abstract water for 1110 h beforethe quota is exceeded. The extra paymentrequired from those who exceed theirvolume quota is ?0.10/m3. The price step isthus large. By paying ?50, the user becomesentitled to take up to 4000 m3, so if the fullquota is taken its average price per m3 is?0.012. If the quota is exceeded by the user,the marginal price for taking more waterrises immediately to ?0.10, almost eighttimes the base rate.

In reality, the user will often take lessthan the quota, particularly during rainyyears. In that case the payment of ?50remains, so in effect the average pricepaid per cubic meter is more than ?0.012,which is the minimum possible. Over a longperiod, the average price actually paid isclose to ?0.02 per m3.

The contract also fixes penalties forthe user (in case of withdrawing above theallowed flow rate or the lack of a watermeter) and for CACG (in case of quota reduc-tion). As demand exceeds supply even whendams are full, the Neste Commission, whichbrings together all water stakeholders fromthe five Départements involved, decided tostart a waiting list of applicants. All applica-tions who are rejected because of supply

98 H. Tardieu

2 M m3 = million (106) cubic metres.



limitations are registered in an open accessfile which totals 6,000 l/s to date. Newlycreated resources and contract terminationsallow a few of these filed applications tobe accepted annually, according to priorityrules (young farmers) or to seniority on thewaiting list. All the yearly contracts get acollective withdrawal authorization in eachof the five Départements.

When dams are not full or when itis anticipated that the Neste river flowwill decrease, the Neste Commission meetsbefore the irrigation season to decide on aquota reduction. The choice of the meetingdate is the result of a compromise betweenthe possibility of making a sound hydrologi-cal forecast and the possibility for farmers toadjust cropping patterns or inputs.

During the irrigation period, watermeters are checked regularly. If a quotaseems likely to be exceeded, a warningletter is sent by CACG to the irrigator. Quotaoverruns are billed at the end of the sea-son. CACG is also in charge of resourcemanagement, which it carries out throughcomputerized remote control (RIO soft-ware), tactical water management in orderto save water transferred from the remote-controlled dams (checked every 3 h) andstrategic water management in order tooptimize water allocation between irrigationand river wildlife. The objective in the lattercase is to empty the reservoirs by the end ofthe low flow period with a failure rate of onein ten (weekly check).

After the irrigation season, water man-agement performance is assessed in termsof meeting or exceeding minimum wildlifeflows, supply of volumes subscribed byirrigators and water savings throughout thesystem.

• Since 1990, when the system was firstput in operation, failures to maintainwildlife minimum flows have beenrare, amounting to only 1–2 days/yearover a reach of a few kilometres, ascompared to the drying up of severaldozens of kilometres over severalweeks in 1989.

• However, irrigators’ quotas have beenreduced 4 years in 10 (though one of

these reductions was later cancelled),and in 3 years out of 10, users havebeen prevented from exceeding theirquotas regardless of extra payments.More extreme State intervention (bythe Prefect at the Département level),which constitutes the ultimate recoursewhen the crisis cannot be met by meansof quotas, has not taken place.

• After using the RIO software for 10years, experience has shown watersavings of more than 20% of themanaged volume.

6.3.3 Successes and problems of the system:can a limited supply be regulated?

In the Neste system, both principles of goodwater management are respected:

• Water is shared in such a way that fishand wildlife are preserved all alongthe 1300 km of controlled rivers andirrigators receive their contractedvolumes;

• CACG, on behalf of the ‘conceding’State, bills users for the cost of theservice and gets the financial meansto cover at least the system’s ‘sustain-ability cost’ assuring the maintenanceof assets worth ?540 million at currentprices.

There has clearly been progress in compari-son with the two ‘wrong practices’ of theprevious period: (i) daily interventions bythe Prefect which irrigators circumventedby over-investing in pumping capacities;and (ii) the inability to charge for the‘resource’ part of the water service, thusleading to asset jeopardy.

One direct positive consequence ofthe new arrangements is that irrigatorsare driven to save water and optimize theircropping patterns through a system of soundand sustainable incentives far more valuablethan any media campaign. It also induces arenewed interest on the part of research andextension networks in quota optimizationthrough revised cropping patterns and inputuse (Balas and Deumier, 1993).




However, a fundamental questionremains, which concerns continued spatialdevelopment: what about the waiting list’sdemand if water resource creation is ham-pered by procedural problems and limitedpublic funds? One solution sometimesenvisaged is to reduce the quotas of currentirrigators in order to admit the new appli-cants. This bad solution, which includesissues of equity, economic efficiency, socialacceptance and technical and agriculturalmanagement, has been discussed in aseparate paper (Tardieu, 1999) and issummarized in Section 6.6 below.

6.4 Key Actors and Essential BasinFunctions

In addition to the key actors describedin the first section of this chapter, othergovernment units also play important rolesin basin management. These include towns,the Ministry of the Environment and theMinistry of Agriculture. The interactionbetween key actors and essential basinfunctions is shown in Table 6.1. It shouldbe noted that almost all of the water used inthe basin comes from surface water sourcessince little groundwater is available, andthat surface water is used primarily fordomestic, agricultural and environmentalpurposes.

The point that stands out most sharplyfrom a review of Table 6.1 is the central roleplayed by the SARs in the French model. Inthe Neste, CACG plays major roles in con-structing and maintaining facilities and inallocating and distributing water to users.The roles played by the irrigation associa-tions are, by comparison, rather modest.Tools employed for allocating water arenot solely administrative. The tiered pricingsystem for water deliveries used by CACGand withdrawal taxes levied by the WaterAgency are important tools that help todetermine the volumes of water that will beallocated to various users. The Ministry ofEnvironment also plays an important role inproviding the ‘policing’ powers to enforceallocation and use rules.

Another important feature of the matrixis the important planning and monitoringrole played by the Basin Committee and theWater Agency, its executive arm. Becausethe Basin Committee includes most of theinterested basin stakeholders, their involve-ment at the individual level is subsumedinto the intersection between CACG and theplanning function in the table.

The Ministry of Agriculture, whichrepresents the State as the owner of theinfrastructure, plays an important regula-tory role in monitoring the care and mainte-nance given to the facilities by CACG, theconcessionaire.

6.5 How to Charge for IrrigationWithout Threatening Economic

Development?

Whichever institutional framework onechooses, a fundamental challenge is tocover the full cost of water used by raisingthe water price and collecting fees fromusers. For most irrigation systems managedby government agencies, public subsidiesare now limited by budget constraints. Suchsubsidies may consist of paying salaries ofoperating personnel, covering heavy main-tenance or rehabilitation costs, or under-pricing energy, etc. For us Europeans, theprinciple that ‘users pay for water’ willbe the basis of the new European WaterDirective. Some targeted and transparentsubsidies will still be acceptable, but onthe condition that they will be graduallyphased out. This objective of an irrigationsystem breaking even on the basis of its ownwater charge revenue is not impossible toreach, and is already the case in severalregions of France.

However, the economic and socialconsequences of water price rises can beserious, as shown by the following examplesof likely risks:

• Overall reduction in the country’sagricultural production, making itimpossible to reach the goal frequentlyassigned to irrigation of securing foodself-sufficiency. This consequence may

100 H. Tardieu



be accepted if the country can maintainits ‘food sovereignty’ (FAO, 1996). Aregular increase in the price of waterhas recently been started in Tunisia,except for cereals, for which watercharges have been kept constant.

• Higher food prices for urban consum-ers, which induces larger food importsand some losses of internal marketshares for irrigating farmers. Thishas already been verified in variousAfrican countries.

• Lower agricultural income, henceincreased rural poverty and populationmigration towards towns. Even if theirrigating farmers are not the mostvulnerable in economic terms – sincethey can use a wider range of crops– the economic development of ruralpopulations must remain irrigation’sfundamental objective.

On the other hand, the ‘true prices’ processcan also generate some benefits:

• A new respect for water, whichimproves management efficiency;

• An incentive to choose the mostprofitable crops and to maximizecomparative advantage;

• A means to know which assets have tobe maintained, and which investmentshave to be made.

The price adjustment process has to be con-ducted with great care, taking into accountthe economic consequences on production.This is done by analysing the value of waterfor the farmer, i.e. the additional addedvalue per water unit (m3) provided byirrigated crops as compared with rainfedones.

6.5.1 Full cost and ‘sustainability cost’

Before tackling this issue, it is worthrestating the definition of the full cost ofwater from the point of view of the agencyresponsible for water resource acquisitionand distribution.

The full cost of water includes thefollowing.

• Operating costs: staff, energy, dailyupkeep;

• Investment-linked costs: depreciation/maintenance/renewal, financial costsof the initial investment.

A water price set at this level secures abalanced budget for the managing agencywithout any subsidies. In France, this priceis about ?0.15/m3 for the large irrigationschemes, where water charges are basedon the full cost of water with the firstinvestment partially subsidized.

However, the cost of major headworks(reservoirs, transfer canals) is generally notincluded. The rationale for such under-charging is based on the consideration thatthese works are both strategic and multi-purpose and that they were created forthe sake of regional development at a timewhen economies were more State-backedand more protected. Today, countries wheresuch infrastructure is paid for by water usersinstead of taxpayers are rare. Nevertheless,it is the objective that has been set forirrigation, notably in France, with a transi-tion period allowing a smooth evolution ofproduction systems.

During that phase, water chargesare meant to cover what will be called thesustainability cost of water, something that,in the case of heavy, long-life investments, isvery different from the full cost.

• Sustainability cost = operating cost +maintenance and sustainable renewalcost; or

• Sustainability cost = full cost −financial cost of initial investment.

With a water price set at the sustain-ability cost level, no new investment ispossible; but budget constraints are met,and sustainable operation and maintenanceensured without having to resort to publicfunding.

Table 6.2 shows a simplified exampleof the costs generated by the water resourceinfrastructure, which, when added to thewater distribution cost, yield the full costof water supply. Values are drawn fromthe case of the Neste system, where actualannual costs of a storage reservoir feeding

102 H. Tardieu



the river amount to about ?1.5/m3 with aquasi-infinite reservoir lifetime.

A water price covering the sustain-ability cost of ?0.015 m3 is socially accept-able and, after the public funding of theinitial investment, prevents the need forfurther subsidies. In the transition phase,irrigation distribution costs are calculated ataround ?0.15/m3, i.e. at full cost pricing,while the irrigator’s share of the waterresource costs is charged more around?0.015/m3, i.e. at sustainability cost pricing.

6.5.2 Water strategic value

On the basis of the existing farminginfrastructure, the strategic value (Vs)corresponds to the optimum combinationbetween irrigated and non-irrigated crops,with a given cropping pattern:

VsVAI VANI

VI= −

where VAI = value added from irrigatedcrops (before deducting the cost of water);VANI = value added from non-irrigatedcrops (rainfed crops) which could becultivated instead of irrigated crops; andVI = volume of water allocated to irrigation.

This value3 reflects strategic choicesmade by the farmer at a point in time whenhe can still modify his cropping pattern, andadjust his irrigation practice to a variableallocated water volume. It is the result of adecision taken once or twice a year and itshould at least cover the cost of irrigation forit to be profitable.

Values calculated for each irrigatedcrop can then be plotted against the volumesof water required to produce the variouscrops, i.e. water demand curves showingthe decreasing returns of irrigated crops(Fig. 6.2).

The following remarks can be made onthe Vs formula presented above.

• Variations in crop prices (domestic orworld price) can lead to changes invalue that rule out irrigation or, on thecontrary, that result in a stronger waterdemand. This is particularly the casefor cereals, whose water valuation isrelatively feeble but which call for largevolumes of water.

• Changes in the yield or added value ofa given rainfed crop can paradoxically


ItemCents?/

m3Cents?/

m3

Operation and daily upkeepMaintenance/renewal

(0.5% × investment cost)Subtotal: sustainability costFinancial cost (long-term

interest rate: 4.6%)Total: full cost

0.750.75

6.9

___1.5

___8.4

Table 6.2. Sustainability cost and full cost in theNeste system.

0

1

2

3

4

5

6

7

8

0 2 4 6 8 10 12 14 16

Water demand (millions of m3)

Val

ue (

F/m

3 )

Fig. 6.2. Effect of water quantity on water value.

3 See also Molden et al. (1998).

115A4859 - Svendsen - 115.prnZ:\Customer\CABI\A4831 - svendsen\A4859 - Svendsen - Vouchers Proofs #P.vpFriday, January 14, 2005 10:55:20 AM


entail changes in irrigation waterdemand. For instance, a specific sub-sidy to rainfed durum (‘hard’ wheat)makes it an alternative to irrigatedmaize in the driest parts of southernFrance. On the other hand, the proba-ble diminishing profitability of cattlebreeding (as a result of the CommonAgricultural Policy of the EuropeanUnion) will increase water demand forirrigated cereals. The improvement ofrainfed crops in Sahelian Africa, lead-ing to lower import prices, may leadto reserving irrigation water for highadded-value crops such as vegetablesor fruits.

• Improving irrigation effectiveness, thusdiminishing the formula’s denomina-tor, increases water value across theboard.

6.5.3 Water’s strategic value, price andbudget constraints

By comparing the strategic value of waterfor the farmer and its cost, it is easy toknow the average price that will balance theirrigation manager’s budget. The problemfor the irrigation agency, and for the Statewhich backs it, is the following. Water pricerises, which help to balance its budget, havea negative effect on water sales and, hence,a tendency to raise the costs of each cubic

metre sold, since irrigation costs are mostlyfixed ones (depreciation, financing andmaintenance costs). This is a vicious circleleading inevitably to the collapse of thesystem. That is why, in a now-transparentmanagement environment, the State mayfind it advantageous to keep on financingintensification and modernization projects,thus boosting irrigated agriculture andincreasing its own chances of recoupingheavy sunk costs.

The concept of sustainability cost asdescribed above is essential, for it consti-tutes the lowest price the State can accept(Fig. 6.3). If the water price does not coverthe sustainability cost and exceeds thestrategic value of water for farmers (for atleast one given existing crop), this meansthat a long-term public subsidy throughwater charges will be necessary to maintainthat irrigated crop in the country or regionconsidered. The opening up of agriculturalmarkets and the new transparency in worldtrade will make this practice impossible inthe future.

On the other hand, it does not seem eco-nomically sound to dismantle entire sectorsof irrigated agriculture on the principle thatthe full cost of water should be covered . . .at all costs! This would mean that today’sirrigators would have to pay for investmentswhich will also be used by future genera-tions, which would be unfair, justifying acertain amount of public subsidy to helpstart the economic development process.

104 H. Tardieu

0

1

3

4

5

6

7

8

0 2 4 6 8 10 12 14 16

Water demand hm3

Va

lue

(F

/m3)

private development sustainable development

Thewater pricecovers the :

full cost

sustainability cost

NON-sustainable development

2

Fig. 6.3. Sustainable allocation of water.



So, when embarking on the true-priceprocess in irrigation, it is essential to havea good understanding of strategic value ofwater to be able to derive demand curves byfarm types and by region. Economic data onirrigated crops are not always available. Thisis one negative consequence of the disap-pearance of public irrigation agencies. Theoverall regulation of irrigation investmentsand of agricultural production requiresgovernments to allocate some funds to datacollection and processing that ‘privatized’agencies can no longer afford. It should notbe forgotten that economic issues to be docu-mented should also include rainfed crops, tobe in a position to correctly appraise watervalue by comparison between alternativeagroeconomic systems.

6.6 Is Water Pricing Useful forControlling Water Allocations?

The points made above assume that anessential prerequisite has been met – theclear identification of the economic agentswho measure, buy, and sell irrigation water.This is an often heavy but always decisivetask, which precedes and accompanies thetrue-price process in irrigation. It requiresmoving away from the idea that water is afree gift from the State towards the conceptthat irrigation is a service supplied by aprovider to a client farmer. The criticalquestion of how this transfer should beaccomplished is the subject of manyworkshops. Let it just be said that, whereverwater is scarce, it is very tempting to use thenewly established economic links between‘supplier’ and ‘customer’ to try to regulatewater use through prices.

Indeed, after the beneficial disengage-ment of the State from direct management ofirrigation schemes, some think that waterallocation can also be taken care of solely byprice mechanisms. To what extent is suchprice-based regulation reliable?

Water allocation regulation consistsof inducing each economic agent to respectthe volume of water allocated by the publicauthority. Is the pricing of water sufficientto avoid allocation crises in water-scarcesystems? Can it settle intersectoral disputesbetween competing uses? Can it improvewater distribution among farmers?

6.6.1 Quotas and pricing: instruments forallocation regulation

In many water-scarce regions, water quotasare allocated to farmers by sub-basin or byregion. For the public authority, the prob-lem is to ensure that these allocations arerespected. The answer is usually of the lawenforcement type, forbidding off-takes andpenalizing those who withdraw excessiveamounts. This type of regulation generateseconomic inefficiency and, sometimes, cor-ruption. Therefore, it is highly tempting touse the price of water to avoid disputesbetween users, provided all participantshave been identified and the service billedhas been clearly defined.

From analysis of the marginal valueof water,4 a method of water pricingcan be derived with a view to attemptingsuch regulation. It will necessarily be steppricing, i.e. a discontinuous series of pricelevels, increasing with water demand. Thehigher price step, which will counteractthe marginal value, must be higher than thelower price step, which itself is calculatedto cover at least the sustainability cost ofsupplying the water and also to secure thefarmer’s income.

The fairly simple system set up in theNeste system consists of the followingelements:

• An allocated quota, priced at a fixedamount which is the same whether theuser takes it all or takes less;

• An over-consumption price for eachunit of water above the quota.


4 Defined here as the value of the additional production that is brought about by one additional cubic metreof water applied during the irrigation period.



The overall volume quota must be compati-ble with the limited resources allocated toirrigation, as opposed to other competinguses. For a given existing irrigated area,there exists a corresponding volume quotaper hectare, which has to be regulatedwith a price step high enough to deterover-consumption.

However, efficient regulation is basedon understandable and practical watercharges within a freely negotiated contract.Charges are useless if they cannot be recov-ered. Excessive over-consumption pricescan only lead to jeopardized contracts andthen to legal prosecutions, which is pre-cisely the regulation mode to be avoided.

Experience from the Neste systemshows that a price step between the quotaprice and the over-consumption priceexceeding ?0.10/m3 would not be sociallyacceptable at present. The approach out-lined here provides useful strategic guide-lines for fixing the volume quota per hectare.In consideration of the marginal value graph(Fig. 6.3), the highest price step clarifiesthe concept of socially acceptable minimumquota. If the quota is too far below the opti-mum needs (less than 80% of those needs)the system does not work in a dry year, andcrises can be frequent, with stalled contractsand prosecution by the public authorities.

6.6.2 Best practices in water pricing andwater resources development

Stemming from this discussion, three basicideas can be emphasized:

• Water step pricing can help to regulatethe allocation system if the quota andthe over-consumption price are set inconsideration of the marginal value ofwater and the social acceptability of thewater charge.

• Quotas that are too low compared tothe crop need cannot be regulated bypricing, and lead to economic ineffi-ciency linked to enforcing inapplicablerules. The quest for equity at all costs ina system with limited resources leadsto the same result.

• Increasing water resources in a water-scarce system makes it possible, overand above the direct economic benefits,to rebuild collective regulation basedon a sound quota + price contract,which will leave each farmer freeto manage his irrigation efficientlyaccording to his own water valuefunction.

Such a water pricing strategy, togetherwith the investments needed to createnew water resources, are necessary tohelp farmers face open market competition.Other requirements are a guaranteed andclearly contracted water supply andfull supplier responsibility for irrigationmanagement without State intervention.

Nevertheless, it is clear that such asystem of price regulation can only worksmoothly within a narrow range of economicvariables – water price and water value – andit is the State’s responsibility not only toidentify this range but also to be ready to laydown rules on economically unacceptablebehaviours. Only this type of strong Statemakes it possible for the managing agency tomake efficient and economic use of priceregulation.

The main advantage of such a frame-work is to give back to farmers the freedomto optimize their choice of crops and theirmanagement of irrigated or non-irrigatedagriculture. This optimization becomeseven more complex in the context ofcompetitive world markets.

One prerequisite to the efficiency of thiseconomic approach lies in the identificationof the relevant agents (managing agency,individual farmers, water user groups), theclear and transparent content of their con-tract relations (water price, allocated watervolume), and the capacity to measure thetraded economic good (water meters). Thisindeed represents a move towards watermarkets (Kosciusko-Morizet et al., 1998), butanalysis of the value of irrigation water –particularly its marginal value – shows thatit would be unwise to go further alongthis line, especially when it comes to freebidding for water quotas, given, on theone hand, the disproportion between the

106 H. Tardieu



marginal value and the socially acceptableprice of water and, on the other hand, thenecessary equity in the sharing of a sociallyhighly valued good.

6.7 Conclusion and Lessons Learned

The role of service-oriented organizationsin irrigation and water management is nowlargely accepted as a prerequisite for imple-menting good control of water allocationand ensuring the sustainability of economicdevelopment (Malano and van Hofwegen,1999). The transfer of management to waterusers’ associations under control of an inte-grated basin authority is one possible solu-tion, frequently described but too recent tobe completely convincing. In fact, this typeof solution often leaves unanswered the twoimportant questions raised at the beginningof this paper, i.e. how to balance the budgetby raising the water price and how toreach a fair sharing of scarce water amongusers. If water management is transferred towater users’ associations, the process mustbe implemented carefully. The main ideais to promote ‘self-management withoutabandonment’ by transferring to sociallystrong users’ associations responsibilitiesadapted to their capacity while keeping atight partnership with a professional watermanager.

The French history of irrigation man-agement emphasizes the efficiency of someother solutions, such as management bySARs – mixed public-private companieslinked to the Government by concessioncontracts. With the experience of suchmanagement, we can propose two recom-mendations in the very difficult debate onwater pricing and allocation.

• Firstly, a cautious but firm movetowards sustainability cost pricing isrecommended. This involves chargingthe amount necessary to ensure the sus-tainability of the assets, i.e. operation,maintenance and renewal costs, but notthe full financial cost of the initialinvestment or of the most recentrehabilitation. To correspond exactly to

sustainability, the price charged mustcover all costs incurred in deliveringeach drop of water from the dam to thecrop. At this level of cost recovery,there is no further need of current sub-sidies for staff, for repairs, for energy orfor future rehabilitation. The subsidy’svicious circle is broken. Such develop-ment is sustainable, even though it isnot designed to recover the initialinvestment.

• Secondly, a step pricing system basedon water metering is recommended inorder to facilitate control of allocationin a fair and transparent water-sharingsystem. In case of water scarcity, theimplementation of a quota system isnecessary due to the high marginalvalue of water during the irrigationperiod. The collective regulation hasto be based on a sound quota + pricecontracts between the service providerand the clients, which will leave eachfarmer free to manage his irrigationefficiently according to his own watervalue function. However, in a closingbasin, the new resources must also bedeveloped to ensure the governabilityof the system. The success of controlledwater management, as developed inthe Neste system, has been foundedon joint management of demand andresources, with the construction ofsome new reservoirs during the last10 years. The shift to an organizationalset-up like the one described here cansucceed, if governance, after the periodof ‘decision without consultation’,avoids the current tendency towards‘consultation without decision’.

References

Balas, B. and Deumier, J.M. (1993) De l’entréeparcellaire à une stratégie globale dans lagestion de l’irrigation. Aménagement etNature No. 3.

FAO (1996) NGO consultation on the world foodsummit. FAO, Rome.

Kosciusko-Morizet, N., Lamotte, H. and Richard,V. (1998) Que peut-on attendre de la mise en




place de quotas individuels échangeablesde prélèvement sur la ressource en eau enFrance? L’exemple de l’agriculture irriguée.Colloque SFER-CEMAGREF-ORSTOM,Montpellier.

Lesbats, R., Tardieu, H. and Heritier, M. (1996)Associations d’irrigants dans le sud ouestde la France, et leur relations avec la CACG.Question 46–1, paper R-107, ICID Congress,Cairo.

Malano, H. and Van Hofwegen, P. (1999) Manage-ment of Irrigation and Drainage Systems.A Service Approach. A.A. Balkema,Rotterdam/Brookfield.

Molden D.J., Sakthivadival, R., Perry, C.J., deFraiture, C. and Kloezen, W.H. (1998) Indica-tors for Comparing Performance of IrrigatedAgricultural Systems. Research Report 20.International Water Management Institute,Colombo, Sri Lanka.

Plantey, J., Tardieu, H., Mesny, M., Nico, J.P., Rieu,Th. and Verdier, J. (1996) Gestion de l’eau

pour l’agriculture en France: durabilitésocio-économique et implication desusagers. www.afeid.montpellier.cemagref.fr

Tardieu, H. (1991) Gestion de l’eau en Gascogne:la CACG, instrument technique et lieude concertation. POUR No. 127–128.L’Harmattan, Paris.

Tardieu, H. (1999) La valeur de l’eau en agricultureirriguée: une information économiquenécessaire pour mieux réguler la gestion del’eau et des productions agricoles dans unmarché ouvert. Symposium, ICID Congress,Granada.

Tardieu, H. and Plantey, J. (1999) Balanced andsustainable water management: the uniqueexperience of the regional development agen-cies in Southern France. ICID Journal, no. 1.

Tardieu, H. and Préfol, B. (2002) Full cost or‘sustainability cost’ pricing in irrigatedagriculture. Charging for water can beeffective, but is it sufficient? Irrigation andDrainage 51, 97–107.

108 H. Tardieu



7 Basin Management in a MatureClosed Basin: the Case of California’s

Central Valley

Mark Svendsen

7.1 Introduction

7.1.1 Basin management

River basins are managed at two differentlevels. At the higher level, the basin level,overall policies and plans are set, resourcesare allocated and regulations written andenforced. At the use level, regulated waterdeliveries are made to users of water, whichmay be irrigators, urban residents, indus-tries, wetlands or natural river reaches. Thispaper focuses on the first level of manage-ment, the basin level, and examines the wayin which basin level management functionsare performed in the large interior CentralValley of California. The Central Valley com-prises what T. Bandaragoda (unpublishednote, 1999, International Water Manage-ment Institute (IWMI), Colombo, Sri Lanka)terms an advanced river basin, one which isalready well developed in terms of physicalinfrastructure and effective institutions forintegrated water resource management.

Issues of particular interest here are theinterplay of political forces which supportalternative water uses, the currently chang-ing priorities accorded to alternative wateruses, and processes and institutionswhereby allocational and regulatory activi-ties at the basin level are directed andcoordinated. One central issue of global sig-nificance is the extent to which one apex

organization must be in control of thehighest level of decision making in a basin.Berkoff (1997), for example, has asserted that‘if water is to be managed holistically, allaspects must be coordinated by one . . .agency’. The present study suggests that thisassertion does not apply universally andraises questions about the conditions underwhich different models of basin-levelmanagement would be most effective.

7.1.2 California’s Central Valley

California’s Central Valley is home to mil-lions and is one of the premiere agriculturalregions in the USA, containing six of thetop ten agricultural counties in the country.California itself has 33 million residentsand is the most populous state in thenation. An overwhelming 97% of thepopulation live in urban areas.

The state as a whole has abundantrenewable water resources, which, inaddition to meeting environmental, urbanand agricultural needs, generate 42% ofthe utility-produced electricity in the state.Irrigated agriculture generates 81% ofCalifornia’s total agricultural revenue on30% of the state’s farmland. Agriculture alsoprovides 14.4% of the state’s employment,though only 2.1% of that is engaged in directproduction activities. The remaining 12.3%




works in input production, marketing andprocessing, and wholesale and retail sales.The state is also blessed with a magnificentand varied natural environment – the Pacificcoast, the Sierra Nevada mountains, broadinland valleys, wetlands and the southerndeserts. All of these features – the environ-ment, urban concentrations, power genera-tion and agriculture – require water for theirsustenance and operation.

Several features of California’s situationmake it especially valuable as a case forstudy.

• First, California comprises a sophisti-cated economic environment in whichwater is used for a wide variety of pur-poses and is treated more as a commod-ity than as a common pool resource.

• Second, intense competition overwater has emerged in what Seckler(1996) would call a closed water system– one in which there is little new waterleft to develop. This competitionincludes agricultural, municipal andindustrial (M&I), and environmentalinterests and is driving rapid change inthe institutions that allocate, regulate,convey and use water.

• Third, the responses to changing publicpriorities have been characterized bypragmatic problem-solving behaviour.This has made California a virtuallaboratory for innovative solutions toproblems of water reallocation andmanagement, environmental quality,efficient water use and water qualitymanagement.

7.2. Basin Hydrology1

7.2.1 Supply

California possesses abundant waterresources, receiving nearly 250 billion (109)

m3 of precipitation annually in averageyears. Of this amount, about 65% is used bytrees and other natural vegetation. An addi-tional 10% flows to the Pacific Ocean orother salt sinks unchecked and unallocated.The remaining runoff is available asa renewable water supply for urban,agricultural and environmental uses.2

Developed surface water resources inthe state total about 80 billion m3, ofwhich nearly half are set aside as requiredenvironmental flows.3 About 12% of thetotal has been developed under FederalGovernment projects, 5% by the State ofCalifornia and 17% by local governmententities. An additional 8% comprises waterimported from the Colorado River Basinunder a multi-state water sharing agreementusing facilities also constructed by theFederal Government.

In addition to surface water sources,an additional 15 billion m3 is availableas renewable groundwater. Present with-drawal rates are higher than this, resulting inan overdraft of about 1.8 billion m3 annually,some 12% of the renewable total. Further-more, the rate of overdraft is increasing andwas 10% greater in 1995 than it was in 1990.To some extent, this overdrafting is aconsequence of 1992 federal legislationreallocating water away from irrigators toenvironmental uses. This has led to supplydeficiencies of up to 50% for some CentralValley irrigators and caused them toturn to lightly regulated groundwater asa replacement supply.

Most of California’s precipitation fallsas snow in the mountains of northernCalifornia and in the Sierra Nevada range,which comprises the high backbone of thestate running from north to south along itseastern flank (Fig. 7.1). A second rangeof much smaller hills, the Coastal Range,fronts the narrow coastal plane in the West,creating a broad alluvial valley betweenthe two ranges. This Central Valley is an

110 M. Svendsen

1 Data for this section are drawn largely from DWR (1998).2 A portion of the water specifically designated for in-stream environmental use also flows to the PacificOcean.3 The total of developed surface and groundwater is greater than the 25% of precipitation designated asavailable runoff because of reuse.



area of rich soils and favourable growingconditions for a wide variety of crops and isthe heart of California agriculture. In it, morethan 200 types of crops are grown and fromit comes 45% of the nation’s fruits andvegetables. Two major river systems drainthe Central Valley and some 158,000 km2

of watershed, the Sacramento River inthe north, and the San Joaquin River in thesouth. The two rivers meet in the

Sacramento–San Joaquin Delta (the Delta),just inland of San Francisco Bay, from wherethey flow into the Bay and out to the PacificOcean.

The fact that two-thirds of California’swater is in the north, while the bulk ofagricultural land and the largest populationcentres are in the south, has led to twomassive engineering projects designed totransport water from north to south. These

Basin Management in a Mature Closed Basin 111

Fig. 7.1. California’s Central Valley.



are the federal Central Valley Project (CVP)and the State Water Project (SWP).

The CVP was constructed in the 1940sby the Bureau of Reclamation, the federalirrigation development agency. Constructionwas begun in 1935 as a part of a massivedepression-era public works programme.The project is anchored by Shasta dam in theCascade Mountains in northern California,which stores water for use in the south.Water from Shasta and several smaller damsis routed down the Sacramento River to theDelta, which it crosses in a network of natu-ral and artificial channels. Some of the wateris used to irrigate land along the SacramentoRiver to the north, but most crosses the Deltato be lifted 60 m into the Delta–MendotaCanal (DMC). The DMC supplies 32 irriga-tion districts in the San Joaquin Valley withwater.

The second project, the SWP, wasdeveloped in the 1960s by the State ofCalifornia. Its backbone, the CaliforniaAqueduct, parallels the DMC south fromthe Delta before continuing on to southernCalifornia. Its primary purpose is to conveyM&I water to desert cities in the south (70%),principally the greater Los Angeles area,though it does supply irrigation water (30%)as well. Together these two projects deliverabout 7.3 billion m3 of water annually to thesouth.

7.2.2 Demand

7.2.2.1 Current patterns

Overall demand for developed sourcesof water is dominated by environmentalreservations (46.5%) and by irrigatedagriculture (42.5%) (Table 7.1). Municipaldemand currently makes up 11.0% of thetotal.

7.2.2.2 Changing patterns of demand

Projections for 2020 (DWR, 1998) anticipateonly a very modest expansion in availablesupply (1%), but with important shifts inthe composition of use. While environmen-tal uses of water are expected to remain

constant, urban demand will expand by37% and agricultural water use willshrink by nearly 7% to accommodate thisgrowth. Additional developed supplies willbe devoted entirely to urban use.

The federal Endangered Species Act,passed in 1973, established the legalframework for protecting species ofplants and animals listed as threatenedor endangered and the allocation of waterfor their preservation where necessary. Thelisting of winter run Sacramento Riversalmon as endangered under this act inthe early 1990s was the first importantapplication of the law in California, whichhad a significant impact on water allocationin the Sacramento–San Joaquin Valley. A farmore sweeping change was wrought by theCentral Valley Improvement Act, passed byCongress in 1992. This act reallocated a por-tion of the water the federal government hadcontracted to deliver to irrigation districtsto the ecosystems of the Sacramento–SanJoaquin Delta. This reallocation has resultedin significant shortfalls in supplies to manyof the irrigation districts in the San JoaquinValley.

7.2.2.3 Urban use

Driving the growth in urban water useis projected growth in the Californiapopulation of nearly 50% between 1995and 2020, as a result of continuing in-migration from others regions of thecountry and from abroad. The demand forwater caused by this growth completelyovershadows modest potential reductionsin per capita water use of about 6% if

112 M. Svendsen

1995 2020

Volume[109 m3] Share

Volume[109 m3] Share

UrbanAgriculturalEnvironmentalTotal

10.841.745.698.0

11.0%42.5%46.5%

100.0%

14.838.945.699.3

14.9%39.1%45.9%

100.0%

Table 7.1. Average water year water use, 1995and 2020.



household level best management practicesare fully implemented.5

7.2.2.4 Agricultural use

California has more than 3.6 million haof agricultural land under irrigation, 80%of it in the Central Valley. Projectionsfor 2020 indicate a modest reduction inthe total irrigated area of about 130,000 ha(3.6%) resulting mainly from urbanencroachment, land retirement due todrainage problems, and more competitiveeconomic markets for agricultural products.6

In addition, changes in cropping patternsand irrigation technology and practices willyield small reductions in the rate of perhectare water use (an estimated 2.4% of1995 use levels).

7.2.2.5 Environmental use

Environmental water use comprises severalcategories of flows that have been set asidefor environmental purposes. These are:

• Dedicated flows in designated ‘wildand scenic’ rivers (64%);

• Instream flow requirements in otherrivers established by water right per-mits, court actions, agreements or otherregulatory actions (17%);

• Required Sacramento–San JoaquinDelta outflows (15%);

• Wetlands freshwater requirements(4%).

Note that while there are other environ-mental uses of water, the above usesare distinguished by being managed andquantifiable. Most of this environmentalwater allocation is brought about bylegislative and regulatory processes ratherthan through the water right permittingprocess, which authorizes agricultural andmunicipal uses.

7.2.3 Summary

California is well endowed with renewablewater resources. Of the 250 billion m3

received as precipitation annually, aboutone-quarter is available for various allo-cated uses. About half of this allocatedwater is set aside for instream environmen-tal uses. The remainder (just over 50 billionm3) is available for withdrawal for agricul-tural and urban uses. Groundwater, thoughabundant, is currently overdrafted by about12% of the renewable total and exploitationcontinues to expand.

Two major plumbing projects, onefederal and the other state, transfer waterfrom the wet north to the arid south of thestate. Water moving through both of thesesystems must transit the Sacramento–SanJoaquin Delta ‘in the open’, where it mixeswith water in the Delta and contributes to it.The Delta is also important environmen-tally, and it serves as the nexus of the debateover the future of California’s water.

According to the most recent version ofthe California Water Plan, urban demandis expected to grow by 37% over the nextquarter-century, while agricultural wateruse shrinks by 7% and environmentaluse holds constant. Additional allocationsto environmental uses are being promoted,however, and if they are adopted, additionalreallocation of agricultural water will be thelikely outcome.

7.3 Legal, Policy and InstitutionalEnvironment

7.3.1 Water rights

Water in California, as in the USA in gen-eral, is regarded as a good belonging to alland held in trust by the State.7 Managementof water, and allocation of rights to use


5 If explicit conservation practices are not implemented, per capita urban demand will increase by about6%.6 There is potential for much more significant reductions if major proposed conversions of agricultural landto wildlife habitat are implemented.7 This is the Public Trust Doctrine, derived from Roman Law.



water, are the responsibility of the individ-ual states. Rights to use water in Californiacomprise a complicated mixture of types,priorities and levels of security. Groundand surface water rights are treated sepa-rately, and surface water rights, which arethe most important, include both riparianand appropriative rights. Underlying andarticulating the various elements of theallocation scheme are a number of stateand federal laws and numerous courtcases, each of which establishes precedentsupon which subsequent cases build.

Riparian rights to surface streamsare available, under common law, to theowners of property abutting streams. Waterabstracted under a riparian right cannot beapplied to plots of land that do not abut thestream, and cannot be transferred to otheruses removed from the riparian land. Theycomprise about 14% of rights to non-imported surface water in California.

Appropriative rights to surface waterare more flexible and comprise the remain-ing 86% of non-imported surface waterrights. Appropriative rights are grantedthrough a permitting process managed bythe State of California. Appropriative rightscan be for use at points removed from thestream of origin and are subject to transferand change of purpose. Maintenance of anappropriative water right requires continu-ous beneficial use, and the courts have heldthat appropriative rights can be lost after5 years of non-use. Riparian rights areneither created by use nor lost by non-use.

Groundwater use is only lightly regu-lated. There is no permitting processfor groundwater exploitation, which isavailable, in the first instance, to owners ofoverlying land for reasonable beneficial useon those lands. Groundwater users establishrights simply by use. Rights are correlativewith the rights of other owners, meaning thatif the water supply is insufficient, the supplymust be equitably apportioned. Subject tofuture requirements on overlying lands,‘surplus’ groundwater may be appropriatedfor use on non-overlying lands. Again, nopermit is required.

This very vague and permissivespecification of rights to groundwater has

two important implications. First, as pres-sure on nearly fully allocated surface watersources continues to build, users turn togroundwater to make up deficits, leadingto a serious and growing problem of over-drafting in many portions of the state.Second, groundwater is a magnet for litiga-tion as water users joust over such termsas ‘surplus’, ‘sufficient’, ‘reasonable’, ‘equi-table’ and ‘beneficial’. Development ofa suitable institutional framework formanaging groundwater in the state isurgently needed but proceeding slowly.

7.3.2 Actors

There are seven important groups of actorsinvolved in basin-level water managementin California, in addition to the generalpublic. These are the managers, the serviceproviders, the users, the regulators, advo-cacy groups, elected officials and thecourts. Some groups, such as regulators,service providers, the courts and electedofficials, consist of both federal and state-level actors, while others are purely local.The main actors in each category arediscussed briefly below.

7.3.2.1 Managers

The most important managing organiza-tions are two California state organizations– the Department of Water Resources(DWR) and the State Water ResourcesControl Board (SWRCB). The DWR replacedthe Office of the State Engineer in 1956,assuming responsibility for planning andguiding development of the state’s waterresources. Over the past 45 years, it hasgrown from 450 staff to a level ten times thatin 1967 before dropping back, and nowemploys about 2000. DWR operates on anannual budget of about $1 billion and is adivision of the state public administrationunder a director who is accountable tothe state governor. Its responsibilities areprimarily technical and operational, but doinclude some regulatory functions. Majorresponsibilities include the following.

114 M. Svendsen



• Preparing and updating the CaliforniaWater Plan every 5 years;

• Operating and maintaining the SWP;• Protecting and restoring the

Sacramento–San Joaquin Delta;• Dam regulation and flood protection;• Public education;• Providing technical assistance to local

communities.

The SWRCB performs functions that aremanagerial, regulatory, and quasi-judicialin nature. It thus occupies a special nichein the overall set-up. Among its importantresponsibilities are the following:

• Allocating rights to appropriate (use)surface water;

• Adjudicating disputes over rights towater bodies, such as the Sacramento–San Joaquin Delta;

• Establishing water quality standards;• Guiding and overseeing the nine

Regional Water Quality Control Boards.

Board appointments are made by thegovernor and the makeup of the Board isas described in Box 7.1.

Regional Water Boards under theSWRCB do not allocate water rights butmanage and regulate water quality throughthe following kinds of actions:

• Writing waste discharge permits;• Implementing contamination clean up

operations;• Monitoring quality and use of regional

ground and surface waters;• Inspecting dischargers and enforcing

state and federal water quality laws.

Regional Boards consist of five memberswho are also appointed by the governor.

7.3.2.2 Service providers

At the basin level, the most importantwater service providers are the UnitedStates Bureau of Reclamation (USBR or ‘theBureau’) and DWR. The USBR is an agencyof the Federal Government housed in theDepartment of the Interior. The Bureauconstructed most of the federally financedwater conveyance and control facilities in

the state, including the pivotal CVP, andoperates the storage and delivery facilities ithas constructed. However, while it retainsoperating responsibility for the upstreamportions of the CVP, it has recentlytransferred operating responsibilities forthe portions of the system lying south of theDelta to an organization established andcontrolled by San Joaquin Valley waterusers, the San Luis–Delta Mendota WaterAuthority (SLDMWA). Users have pro-posed that they assume responsibility forthe upstream portions as well, but action onthat step is more controversial.

The other major water storage andconveyance project in the state, the SWP, isoperated by DWR, which constructed thefacilities using state resources.

7.3.2.3 Users

Principal water users are the various dis-tricts that purchase water and deliver it tothe members or residents in the district.Districts are generally organized to supplyirrigation water to farmers or municipalwater to urban residents. Districts areincorporated as non-profit entities understate law and are self-governing. The largestshare of managed surface water is deliveredto agricultural users, most of whom are inthe Central Valley. Other users includethe state Department of Fish and Game,conservation districts, hydropower facilityoperators, and DWR and the USBR for floodcontrol operations. Freshwater navigation,though significant in the past, is of minorimportance today.

7.3.2.4 Regulators

Water-related regulation centres aroundprovisions of federal and state laws protect-ing endangered species and maintainingdrinking water quality. The federal Endan-gered Species Act of 1973 is the mostimportant of these and is enforced by thenational Environmental Protection Agency(EPA). Of the endangered species affectingwater use in California, the most critical arethe listed runs of salmon. Technical regula-tions and certifications relating to salmon




are made by the National Marine FisheriesService (NMFS), whereas criteria for otheranimal species are set and supervised bythe federal Fish and Wildlife Service.

State environmental regulators also listendangered species, and this list includessome that are not on the federal list. The stateFish and Game Department supervisesenforcement of water quality and quantityrequirements relating to state-listed species.

The SWRCB and its subtended regionalboards bear overall responsibility for surfaceand groundwater quality in the state. The

Federal Clean Water Act and the CaliforniaWater Quality Control Act, both aimed atpollution control, are enforced by theseboards.

7.3.2.5 Advocates

One of the most dramatic recent changes inthe cast of characters in the water drama inCalifornia, and in the USA, is the emer-gence of environmental advocacy groupsas potent political actors. Most groups aremembership based and supported and may

116 M. Svendsen

Box 7.1. Evolution of a water control agency.

1940s Serious water-quality problems emerged in California, including outbreaks of water-bornediseases and degradation of fishing and recreational waters. In 1949, a fact-finding committeehighlighted cumbersome and unreasonable laws and administrative procedures, multiplejurisdictions, limited and conflicting interests, and overlapping authorities as roots of the evidentproblems. The committee concluded that the state’s limited water resources could only beextended through planning to maintain water quality while at the same time allowing maximumeconomic use and reuse. It recommended a central focus point at the state level to coordinatewater pollution control activities.

1949 Legislation created a State Water Pollution Control Board consisting of nine gubernatorialappointees representing specific interests and four ex officio state officials. Its duties includedformulating statewide policy for pollution control and coordinating the actions of various stateagencies and political subdivisions of the state in controlling water pollution. The same legisla-tion created nine Regional Water Pollution Control Boards in major watersheds. These regionalboards had responsibility for administration, investigation and enforcement of the state’spollution abatement program. Five gubernatorial appointees, representing water supply, irrigatedagriculture, industry, and municipal and county government in the region, served on eachregional board.

1959 The 1949 law was revised and broadened on the basis of 10 years of experience. State ex officiomembers were removed from the board, increasing its separation from the state administrativemachinery.

1963 The state board was renamed the State Water Quality Control Board and given the broadermandate of water quality control, replacing the more limited earlier focus on sewage andindustrial waste control.

1967 A proposal to consolidate water-related functions, including water quality control functions,within the Department of Water Resources (DWR) was rejected on the grounds that this wouldcreate conflicts of interest internal to DWR. Instead quantity and quality management functionswere consolidated external to DWR by merging the State Water Quality Control Board and theState Water Rights Board into the State Water Resources Control Board. The ‘State Water Board’consists of five full-time members mandated to protect water quality and to determine rights tosurface water use. Members are appointed by the governor and fill specialized roles on the Board,i.e. an attorney versed in water law, two civil engineers with expertise in water rights and watersupply, a water quality member, and a public member.

1969 A new Water Quality Control Act was passed which retained the basic structure of state andregional boards but provided a new regulatory framework for waste discharges to both surfaceand ground waters. This act served as a model for the federal Clean Water Act, passed 3 yearslater.

Source: http://www.dwr.water.ca.gov



draw on grants from charitable foundations.Some focus on a single issue – a resource orspecies – while others have a broader rangeof interests. The group Friends of the Riveris an example of a resource-focused groupwhich is largely concerned with restoringfree-flowing rivers in California, while theSierra Club, based in California but nationalin scope, is an example of a group with awide range of conservation interests beyondwater. There are about 20 environmentalgroups in California interested in waterissues. These groups are linked throughan Environmental Water Caucus, whichmeets every couple of weeks. Accompany-ing expanded federal and state environ-mental regulation over the past 25 yearshas been greatly strengthened requirementsfor transparency in regulatory processes.

7.3.2.6 Elected officials

Legislators at both the federal and state lev-els write the laws providing the frameworkfor water resource management in the state.Although establishing systems for allocat-ing water resources is the purview of thestate legislature, the federal governmentexerts a powerful influence on water alloca-tion by applying the terms of the federalEndangered Species Act. This act con-strains water-related construction projectsin various ways, and can require increasedin-stream allocations of water for fish spe-cies classed as threatened or endangered.The governor is a particularly importantfigure in the state water resource manage-ment picture, controlling appointments tothe State Water Board and the regionalWater Quality Boards and as the head ofthe state administrative apparatus, whichincludes the important DWR. The US Con-gress also influences allocation throughits ability to mandate changes in waterpermits, which are held on behalf ofthe US Government by the US Bureauof Reclamation. Water quality is regulatedby both federal and state statutes.

7.3.2.7 Courts

Both state and federal courts hear casesrelating to water. Where the US Govern-ment is a party to the litigation, a federalcourt must be the venue, as state courtscannot have jurisdiction over the FederalGovernment. Almost all of the cases heardare civil cases rather than criminal cases,involving disputes between parties ratherthan a violation of state or federal law.

A prominent federal judge in theCentral Valley estimates that about 20% ofhis caseload consisted of water cases, andthat the volume of water-related cases hadincreased considerably over the past 9 years.Cases have also increased in complexity.The introduction of the Federal ESA and thelisting of a number of fish species in impor-tant California rivers has played a major rolein this increased complexity. The integra-tion of Public Trust Doctrine into Californiawater law8 has also made decisions morecomplicated. Throughout this period ofchange, the non-governmental (NGO) sectorhas increasingly become a ‘third presence’in nearly every significant civil case, seekingto include the environmentalist viewpointinto the deliberation.

Major drawbacks to the heavy relianceon the court system for dispute resolut-ion are the often drawn-out nature ofproceedings, their expense and the diffi-culty of reaching sound decisions throughadversarial proceedings. A federal judgeinterviewed cited approvingly an old adage,‘hard cases make bad law’. Increasinglyattention is shifting to various modes ofalternative dispute resolution.

7.3.3 Essential functions

Burton (1999) has identified 11 essentialfunctions of basin management. A some-what modified listing of these functions isshown in Table 7.2, crossed with the keyactors identified in the previous section.


8 Accomplished by a decision of the state Supreme Court in 1983 in a suit filed by the National AudubonSociety.



118 M. Svendsen



These functions are replicated, as appropri-ate, across four broad categories – surfacewater, groundwater, wastewater disposaland agricultural return flows. Cells aremarked to indicate an actor active in aparticular functional area. Information isdrawn from interviews, printed materialsand Internet postings. A number of interest-ing points emerge from an examination ofTable 7.2.

• A comprehensive planning functionrests with the state DWR. This respon-sibility covers surface and ground-water and both quantity and quality.Although technical analyses andmodelling are done by DWR, extensiveinteraction with a variety of stake-holders in the planning process makesplanning a widely shared activity. Theprimary planning document is theState Water Plan, which is updatedevery 5 years in a process led by DWR.

• Surface water allocation and waterquality assurance are assigned to asingle state agency, which is independ-ent of the other state agencies engagedin planning or system operations. TheWRCB is autonomous, though it ispolitical to the extent that the membersare appointed by the governor of thestate. The US congress assumed acertain amount of de facto allocationalauthority in passing a 1992 law, whichdirected the USBR to reallocate waterfrom agricultural users with whom itheld contracts to environmental uses.Federal and state courts also playimportant roles in the allocationalprocess by resolving disputes overallocation.

• Enforcement of water quality standardsrests with nine regional boards withstrong local ties but under the overallguidance of the state-level WRCB. Thecourts also play significant roles ininterpreting disputes related to waterquality.

• Retail water delivery services are, forthe most part, in the hands of user-controlled districts. Such irrigationand municipal water supply districts

are financially autonomous and self-regulating. They usually obtain waterfrom wholesale suppliers throughlegally enforceable contracts.

• Groundwater is the most lightlyplanned and regulated segment of thestate’s water resources. There is littlecontrol over abstractions, and the stateis in a serious overdraft situation.

• Advocacy groups (environmentalNGOs) make up an important thirdpresence in most important disputesinvolving water. This is a relativelyrecent development, but has pro-foundly changed the way in whichdecisions are made, and modified theiroutcomes. These groups also playimportant roles in joint consensualprocesses, such as CALFED and theAmerican River Water Forum, whichare being used increasingly to developmutually acceptable plans and agree-ments over contentious water-relatedissues.

• There is a certain internal conflictwithin the DWR with respect to itsmultiple roles as wholesale supplierof water, water resource planner andregulator. Transparency of processappears to keep these potential con-flicts in check. Although not includedin the table, there is a significantconflict of interests internal to theUS Army Corps of Engineers, which ischarged with wetland permitting andprotection, but is, at the same time, aconstruction and operating agencywith close ties to the congressionalappropriations process.

7.3.4 Coordinating processes

Managing an important publicly held natu-ral resource will always involve multipleactors, differing interests and perspectives,and relational dynamics. This is true evenin situations where a single agency isresponsible for all aspects of basin watermanagement, as there will be winners andlosers among users of basin water resources




and factions within the managing agencyhaving differing perspectives and interests.

In California, where there are many dis-crete actors in the water resource allocationand management picture, coordination anddecision making have long been importantand contentious functions. Traditionally,the courts, both federal and state, have pro-vided a critical dispute resolution function.As the Sacramento–San Joaquin basin hasclosed and water becomes relatively scarcer,disputes have become more frequent andthe number of interested parties has grown,making proceedings more complex. There ispresently growing interest in various formsof alternative dispute resolution, includingthe use of mediation, arbitration and specialmasters.

There is also growing reliance onprocesses of shared consensual decisionmaking to replace the more typical two-stageprocess of a technical decision made by agovernment agency, followed by extensiveand lengthy litigation initiated by unsatis-fied parties. The most prominent example isthe ongoing CALFED process, which tacklessome of the most contentious water-relatedproblem in the state, as shown in Box 7.2.

CALFED is a consortium of federal andstate government agencies with manage-ment and regulatory responsibilities in theBay–Delta system. It was formed in 1994with the mission of developing a long-termcomprehensive plan to restore ecologicalhealth and improve water management forbeneficial uses of the Bay–Delta system, theheart of the Central Valley hydraulic system.CALFED spent its first 2 years identifyingand defining problems and a further 4 yearsassessing the environmental implications ofvarious actions that might be taken. It isabout to begin an implementation phase thatcould last 30 years and cost $10 billion.

What sets CALFED apart from otherprogrammes is the fact that problems andsolutions are being discussed from the out-set in an open forum with participation thatspans the entire range of water-related

interests, and that it is proposing an entirebasket of measures which will address thefour problem areas in an integrated, comple-mentary, sustainable way. Fundamentalprinciples guiding the process are shown inBox 7.3. What is striking is the commitmentof all participating parties to make theCALFED approach work. This commitmentarises in part from the fear that if the processfails, years of litigation will follow in a farmore adversarial and expensive process ofdispute resolution.


The essential functions and actors’ rolesdepicted in Table 7.2 provide a static viewof responsibilities. Additional attributes ofwell-functioning basin governance9 systemsrelate to their dynamics. We term theseattributes, which provide the contextfor functional performance, enablingconditions.

120 M. Svendsen

9 The term governance is used in a somewhat different sense here than in Burton’s list of essential attributes,of which it is one. Here the term refers to the rules providing the context for multi-actor basin management andthe processes and activities engaged in by those actors operating within this set of rules.

Box 7.3. CALFED solution principles.

Affordable – solution can be implemented andmaintained with the foreseeable resources of theCALFED stakeholdersEquitable – solution will focus on resolvingproblems in all problem areasImplementable – solution has broad publicacceptance, legal feasibility and will be timelyand relatively simple compared with alternativesDurable – solution will have political andeconomic staying powerReduce conflicts – solution will reduce majorconflicts among beneficial users of water

Box 7.2. Bay–Delta problem areas.

• Ecosystem restoration• Water quality assurance• Levee system improvement• Water supply reliability



Enabling conditions are features of theinstitutional environment at the basin levelthat must be present, in some measure, toachieve good governance and managementof the basin. These attributes are not specificto any one actor, but apply to all actors andtheir interactions and comprise necessary(but not sufficient) normative conditionsfor success. Basic enabling conditions areshown in Box 1.2 in Chapter 1. While a fullanalysis of these factors is well beyondthe scope of this chapter, a brief sketch ofeach, in the context of California, is given toillustrate the concepts and indicate broadstrengths and weaknesses.


Representation is generally well developed,with groups having similar interests alliedinto various associations. These associa-tions are supported by their memberswho provide funds for representationand litigation. Environmental concerns arerepresented by NGOs, which have grownover the past 25 years in number, resourcesand influence. Supported by protectionsprovided by federal and state endangeredspecies laws, they now enjoy power com-mensurate with the other major players.


The availability of information and trans-parency of decision-making processes inthe USA, and in California, has alsoexpanded over the past quarter-century.These changes have been driven by require-ments in environmental protection laws, bythe existence of the World Wide Web, andby growing public demand for informationand openness. It is now a rare deci-sion-making process that is not character-ized by ready availability of technical infor-mation, public hearings and extensiveopportunities for public comment.


The system of surface water rights, thoughcomplex, is relatively well specified in lawand through cumulative court decisions.

Rights to groundwater are not well speci-fied and comprise an area where a strongerand more appropriate legal basis is urgentlyrequired.

There is a sound legal frameworkunderlying user-based districts, which pro-vide such services as irrigation, domesticwater supply, groundwater managementand wetland conservation. Districts are self-financing and self-governing and generallywork effectively.

7.3.5.4 Resources

Though participants always feel thatfinancial resources are inadequate, bothfinancing and human resources within thebasin management system appear generallyadequate. There is a well-developed physi-cal infrastructure for transferring wateraround the state, and from neighbouringbasins, and a steady stream of additionsand improvements to it. Environmentalrestrictions and concerns, however, makeinfrastructural design a far more demandingprocess than it previously was, and havecompletely stymied some proposed pro-jects, such as the peripheral canal aroundthe Delta. New institutional forms (alongwith a legal basis for them) will likely berequired in the future to legitimize andimplement consensual agreements reachedby ad hoc bodies such as CALFED, but theneed for these is still evolving.

7.4 Salient Characteristics of CaliforniaBasin Management

A number of important features character-ize basin water management in California.These are summarized below:

• Multiple sources of authority andpower – no single public agencymanages water resources in California’sriver basins. Instead, decisions aremade and enforced by a number of stateand federal agencies. Integration is pro-vided by the State Water Plan, variousregional plans and processes such as




CALFED, the centralized system of sur-face water rights and the court system.

• Dynamic interplay of competing inter-ests – an even broader group of actorsparticipate in and influence decisionmaking. These actors are from bothpublic and private sectors. They debatein a variety of fora to assert their pointsof view. These include public hearings,the media and the courts. Extensivelobbying of public officials also takesplace behind the scenes. Decisionsemerge from this interplay.

• Adequate representation of all inter-ested parties – major parties in thewater debate are well represented andfinanced. These include municipalwater districts, agricultural water dis-tricts, public water supply agencies,state and federal environmentalregulators, and environmental NGOs.

• Heavy reliance on legally enforceablecontracts and agreements – many of thewater-related decisions made takethe form of contracts or agreementsbetween two or more parties, ratherthan administrative decree. Thisrequires confidence on all sides in theenforceability of the agreements.

• Separation of operating and regulatoryfunctions – regulatory functions aregenerally handled by organizationswhich are independent of federal, stateand user-controlled operating agencies.

• Adequate databases on hydrologicprocesses and capacity to researchnew issues – extensive measurementand data collection programmes havecreated a large database of informationon California water resources and theiruses and impacts. Equally importantly,a strong technical capacity exists in theprivate sector to conduct additionalassessments, on a consulting basis, asneeds arise.

• Open access to information andgenerally transparent decision-making

processes – information on waterflows, water quality, wastewater qual-ity, water rights and so on is availableto the public and is generally accessiblethrough the World Wide Web, andin publications and public records.Decision-making processes are gener-ally conducted in the open and includepublic hearings.10 Moreover, decisionsreached are subject to challenge incourt and decisions over controversialissues often are so challenged.

• Self-financing autonomous districts asretail service providers – retail waterservice delivery is typically handled byirrigation or water districts, whichare user-controlled, self-financing,non-profit quasi-municipal entitiesincorporated under state law. Thisvastly simplifies the service deliveryproblem by reducing the number ofmajor ‘users’ to several hundred fromtens of thousands.

• Important role of an impartial courtsystem in resolving disputes – federaland state courts are regularly calledupon to settle disputes brought to themas civil suits. Without this service, thewater resource management system inthe state would be unworkable.

• Well-defined system of water rights(except groundwater) – there is a clearsystem of allocating and protectingrights to surface water, which providesreasonable security to users. Protectionof groundwater is presently moreproblematic.

References

Berkoff, J. (1997) Water Resources FunctionalAnalysis. Report prepared for the Sri LankaWater Resources Council, Colombo, SriLanka.

Burton, M. (1999) Note on proposedframework and activities. Prepared for the

122 M. Svendsen

10 As is always the case, real compromises are often hammered out in private by a smaller group ofparticipants. Nevertheless, the compromises reached must be capable of standing up to public and interestgroup scrutiny when they are announced.



IWMI/DSI/CEVMER Research Programme onInstitutional Support Systems for Sustain-able Management of Irrigation in Water-ShortBasins, Izmir.

Department of Water Resources (1998) CaliforniaWater Plan Update Bulletin 160–98.

Department of Water Resources, Sacramento,California.

Seckler, D. (1996) The New Era of Water ResourcesManagement: From ‘Dry’ to ‘Wet’ WaterSavings. IWMI Research Report 1. IWMI,Colombo, Sri Lanka.




8 River Basin Closure and InstitutionalChange in Mexico’s Lerma–Chapala Basin

Philippus Wester, Christopher A. Scott and Martin Burton

8.1 Introduction

The Lerma–Chapala basin in central Mex-ico is a telling example of the institutionaland political challenges that river basinclosure poses, especially for locally man-aged irrigation. Although rainfall in the past10 years has been slightly above average,total water depletion in this basin exceedssupply by 9% on average. Groundwater isbeing mined, with sustained declines inaquifer levels of 1.00–2.58 m/year (Scottand Garcés-Restrepo, 2001), while surfacewater depletion exceeds supply in all butthe wettest years. As a result, Lake Chapala,the receiving water body of the basin, isdrying up. In early 2001, the volume ofwater stored in the Lake was around 20% ofcapacity, the second lowest level recordedsince systematic data collection began in1934. This lake is the largest in Mexico,giving it a high symbolic value, and itgenerates significant tourism revenue.

Approximately 68% of the annualrenewable water in the basin is used toirrigate around 700,000 ha, all of which islocally managed. Since the basin’s wateris fully committed, there is no scope forirrigation expansion, and the drilling of newwells and the construction of new dams hasbeen prohibited. Moreover, water pollutionis serious, with significant wastewater reusefor irrigation within the basin. Lastly, wateris being transferred from agriculture tothe urban and industrial sectors, without

due compensation to farmers. In sum, thesustainability of locally managed irrigationin the basin is under serious threat, carryingwith it grave implications for social equity.Due to basin closure, the poorer segment ofthe farming community is losing its accessto primary water sources as it is divertedto economically higher-valued uses. Conse-quently, poor farmers increasingly vie forderivative water such as wastewater anddrainage effluent (cf. Buechler and Scott,2000).

In response to the water crisis in thebasin, several institutional changes haveoccurred in the basin since 1989, includingthe signing of a river basin coordinationagreement (1989), the creation of a RiverBasin Council (1993) and the establishmentof aquifer management councils (1995onwards). Water reforms at the nationallevel, such as the creation of a nationalwater agency (1989), the decentralizationof domestic water supply and sanitation tostate and municipality levels (starting in1983), the transfer of government irrigationdistricts to users (1989 to present), thecreation of state water commissions (from1991 onwards), and a new water law (1992),have also significantly altered institutionalarrangements for water management in thebasin (González-Villareal and Garduño,1994). The water reforms in Mexico aredriven by the need to deal with increasingwater over-exploitation, and influencedby the vested interests of the hydraulic




bureaucracy and the neo-liberal economicpolicies pursued by the Mexican govern-ment (Rap et al., 2004).

This chapter assesses the institutionalarrangements for water management in theLerma–Chapala basin and how well theyare dealing with basin closure. It does so bypresenting an overview of the basin’s waterresources and uses, followed by a sectionexploring the legal, policy and institutionalconditions that influence how the basin isgoverned and managed. This basin profileprovides the backdrop for an analysis of theessential functions for river basin manage-ment, after which the key challenges facingthe basin, namely surface and groundwaterallocation mechanisms and the managementof derivative water, are reviewed andconclusions for the future direction ofinstitutional change in the basin are drawn.

8.2 The Lerma–Chapala Water Balance

The Lerma–Chapala basin covers some54,300 km2 and crosses five states: Querétaro

(5%), Guanajuato (44%), Michoacán (28%),Mexico (10%) and Jalisco (13%).1 Thebasin is home to a dynamic agriculturalsector and a growing industrial sector andaccounts for 9% of Mexico’s GNP. It isthe source of water for around 15 millionpeople (11 million in the basin and 2million each in Guadalajara and MexicoCity) and contains 13% of the irrigatedarea in the country. The average annualrunoff in the basin from 1940 to 1995 was5,757 million cubic metres (MCM), a littleover 1% of Mexico’s total runoff (CNA,1999a).

The headwaters of the Lerma River risein the east of the basin near the city of Tolucaat an elevation of 2600 m a.s.l. (metresabove sea level) to discharge into LakeChapala in the west at an elevation of1500 m a.s.l. The total length of the LermaRiver is 750 km, and eight major tributariesdischarge into it (Fig. 8.1). Lake Chapala,with a length of 77 km and a width of 23 km,is Mexico’s largest natural lake, storing8125 MCM and covering 111,000 ha whenfull. The shallow depth of the lake (average7.2 m)2 results in the loss of around

126 P. Wester et al.

Lake Chapala

Rìo Lerma

Rìo Turbio

Rìo Lerma

Guadalajara

Leûn

RìoSilao

RìoGuanajuato

Rìo Laja

Rìo Querètaro

Rìo Duero

Rìo Grandede Morelia

Querètaro

Toluca

MexicoCity

Elevation (m)

1600

3200

0 50 100 km

N

Morelia

Fig. 8.1. Topography and stream network of the Lerma–Chapala basin.

1 Percentages indicate the area of the basin that falls in each state.2 On a scale of 1 : 10,000 the dimensions of the lake are 7.7 m by 2.3 m and less than 1 mm deep.



1440 MCM (25% of the annual averagerunoff in the basin) of its storage toevaporation each year (de Anda et al.,1998). At times of high water levels, LakeChapala discharges into the Santiago River,which flows in a northwesterly directionand then drops to the Pacific Ocean after524 km.

The climate in the basin is semiarid tosubhumid, with 90% of the rain fallingbetween May and October. Rainfall ishighly variable, with an average over the1945–1997 period of 712 mm/year, a mini-mum of 494 mm in 1999 and a maximum of1022 mm in 1958 (CNA, 1999e). Averagemonthly temperatures vary from 14.6°C inJanuary to 21.3°C in May; thus a range ofcrops can be grown throughout the year.The potential evapotranspiration mirrorsthe temperature variation, with a peak inApril/May, and an annual total of some1900 mm. In every month except July andAugust there is a net deficit between rainfalland potential evapotranspiration, indicatingthe importance of irrigation.

Forty aquifers have been identified inthe basin (CNA/MW, 1999). The upper layerof these aquifers is generally 50–150 m thickand composed of alluvial and lacustrinematerials, while the lower layers, severalhundred metres in depth, are composedprimarily of basaltic rocks and rhyolitetuff. The aquifers are recharged throughrainfall infiltration, surface runoff and,

importantly, deep percolation from surfaceirrigation. Various sources report differentdata on annual extraction and recharge rates,making it hard to portray with any precisionthe groundwater situation in the basin. Whatis clear is that 30 of the 40 aquifers are indeficit, with static water levels droppingat 2.1 m/year on average (Scott andGarcés-Restrepo, 2001). Recent data fromthe Comisión Nacional del Agua (CNA;National Water Commission) indicatethat average annual recharge is 3980 MCM,while average annual extractions are placedat 4621 MCM, giving a deficit of 641 MCM,some 71% of the total water deficit in thebasin (CNA, 1999a).

Table 8.1 presents current averageconsumptive water use for different sectorsin the basin compared to average annualrenewable water, and shows a deficit of900 MCM. The percentage of available waterthat is developed and put to use in thebasin is 109%, showing its degree of over-commitment. The out-of-basin transfers areto Guadalajara (surface water) and MexicoCity (groundwater) for urban water supply.To portray basin closure in the Lerma–Chapala basin, it is instructive to analysefluctuations in the water levels of LakeChapala. Figure 8.2 shows these fluctuationsfrom 1934 to 2000 and relates them todevelopments in the basin.

Starting in 1945, water levels in the lakedeclined sharply, from around 97 m on

River Basin Closure and Institutional Change 127

Surface water Groundwater Total

MCM % MCM % MCM %

Runoff/rechargeTotal depletion

Irrigated agricultureUrbanOut-of-basin transferIndustryOther

Total consumptive useEvaporation from water bodies

Balance

5,7576,0163,4243, 403,2373, 393, 63,7462,270,−259

100104

59> 1

4> 1> 16539−4

3,9804,6213,1603,7513,3233,2393,1484,621

–,−641

100116

7919

864

116–

−16

9,73710,637

6,58410,79110,56010,27810,154

8,3672,270

0,−900

100109

688632

8623−9

Source: CNA (1999a).

Table 8.1. Water balance of the Lerma–Chapala Basin.



average to 90.8 m in 1954,3 due to a pro-longed drought combined with significantabstractions from the lake for hydroelectric-ity generation. During this period, around200,000 ha were irrigated in the basin,mainly with surface water, and the con-structed storage capacity in the basin was1628 MCM. This period was the first timethe basin headed towards closure as far assurface water is concerned. However, thanksto good rains towards the end of the 1950s,the lake recuperated, and levels fluctuatedbetween 95.5 and 98.5 m from 1960 to 1979.

In 1979, a second period of declineset in, leading to basin closure in the

mid-1980s. By this time, constructed storagecapacity in the basin had increased to4499 MCM and the average irrigated areahad grown to around 670,000 ha, with a sig-nificant increase in groundwater irrigation.Although abstractions from the lake forhydropower generation had ceased, thecombination of these factors resulted indeclines in the lake level, from around 95 mat the start of 1980 to 92 m in 1990. Aftera modest recuperation in the early 1990s,lake levels declined again after 1994, andby February 2001 had fallen to 91.5 m, thelowest level measured since 1954, due tocontinued over-exploitation of surface and


��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

Fig. 8.2. Lake Chapala water levels and basin developments.

1930s 1940s 1950s 1960s 1970s 1980s 1990s

Rainfall (mm)a

Population (millions)b

Storage capacity dams (MCM)c

Irrigation (ha)d

Lake inflow from Lerma (MCM)e

Lake extractions (MCM)e

2,670.52,672.52,747.5

n.a.2,864.52,638.5

175,648.5175,643.0

1,628.5175,843.5

1,652.51,049.5

175,685.517568,3.6

1,817.5250,500.5

1,692.5175,674.5

175,757.517575,4.5

3,269.5408,746.5

1,773.51,350.5

175,740.5175,745.9

3,840.5681,668.5

1,931.51,817.5

175,668.517566,8.7

4,499.5657,734.5175,590.5175,308.5

175,720.5175,711.0

4,499.5689,743.5

n.a.175,293.5

Sources for lake levels: de P. Sandoval (1994) and CNA (1991, 1992, 1993a, 1994, 1995, 1996, 1997,1998, 1999b, 2000b).aDecade average for de P. Sandoval (1994) for 1934–1949 and CNA/MW (1999) for 1950–1999.bPopulation data for the end of the decade. Source: de P. Sandoval (1994) for extimates for 1939, 1949,1959, 1969 and 1979, and CNA/MW (1999) for actual figures for 1989 and 1999.cConstructed storage capacity at end of decade. Source: de P. Sandoval (1994) and CNA (1999e).dAverage actual irrigated area over the decade. Source: CNA (1999e).eDecade average, excluding evaporation for de P. Sandoval (1994) and CNA (1991, 1992, 1993a, 1994,1995, 1996, 1997, 1998, 1999b).

3 Lake elevations are measured relative to a locally defined benchmark where 100 m is set as the highshoreline (de Anda et al., 1998).



groundwater. It is unlikely that the lake willrecover without exceptional runoff, as mightbe generated by a major hurricane. While theriver basin was closing in water quantityterms, water quality also deterioratedseverely, with increased effluent dischargesand hardly any treatment of urban andindustrial wastewater before 1989. Cur-rently, the Lerma River and most of itstributaries are classified as contaminated.Two of its tributaries, the Turbio River andGuanajuato River, are classified as highlycontaminated (CNA, 1999a).

8.2.1 Irrigated agriculture

The main water user in the basin isirrigation, depleting 59% of total availablesurface water and 79% of renewablegroundwater (Table 8.1). Nine large-scalecanal irrigation systems (termed irrigationdistricts in Mexico) command around285,000 ha and some 16,000 farmer-managed and private irrigation systems(termed unidades de riego in Mexico) cover510,000 ha. Twenty-seven reservoirs witha storage capacity of 250 MCM provide235,000 ha in the irrigation districts withsurface water, whereas around 1500 smallerreservoirs serve 180,000 ha in the unidades.An estimated 26,000 deep tubewellsprovide around 380,000 ha in the basinwith groundwater, of which 47,000 ha arelocated in irrigation districts (CNA, 1993b;CNA/MW, 1999). In the irrigation districts,there are an estimated 88,000 water users(70,000 ejidatarios and 18,000 pequeñospropietarios4) compared to 100,000 waterusers in the unidades (84,000 ejidatariosand 16,000 pequeños propietarios) (CNA/MW, 1999). Detailed data on croppingpatterns and productivity for the wholebasin are not available, although studieson selected irrigation systems are available(e.g. Kloezen and Garcés-Restrepo, 1998;

CNA/MW, 1999; Flores-López and Scott,2000; Silva-Ochao, 2000). These studiesshow that the main crops irrigated in thebasin are lucerne, wheat, sorghum andmaize, whereas vegetables and fruits areincreasing in importance.

In the early 1990s, the Mexican govern-ment transferred the government-managedirrigation districts to water users associa-tions (WUAs) (cf. Rap et al., 2004). Transferwas part of a major reform of the agriculturalsector initiated at the highest level of govern-ment. Reform was driven by market-orientedeconomic and political imperatives andresulted in the following.

• Removal/reduction of direct and indi-rect subsidies to agricultural production;

• Privatization/elimination of public sec-tor input supply and crop marketingbodies;

• Removal/reduction of tariffs andbarriers to agricultural trade;

• Reform of the Mexican Constitution topermit the sale and renting out of ejidoland.

The intent of these reforms was to stimulateeconomic growth through private invest-ment in agriculture. The main objective ofthe Mexican irrigation management transfer(IMT) programme was to reduce publicexpenditure on irrigation by creatingfinancially self-sufficient WUAs that wouldshoulder the full operation and mainte-nance (O&M) costs of the irrigation systems(Gorriz et al., 1995; Johnson, 1997). Farmerswere initially resistant to these changes, buthave now generally come to accept them.

In the Lerma–Chapala basin, nineirrigation districts were transferred to farmermanagement in the early 1990s. WUAs nowmanage secondary canal units varying insize from 1500 to 30,000 ha. The WUAs wereformed as legally recognized non-profitassociations to which CNA granted 30-yearconcessions for the use of water and the


4 Ejidatarios are members of ejidos, land reform communities created after the Mexican Revolution of 1910.Land holdings per ejidatario are typically less than 5 ha. Pequeños propietarios are private farmers with a limiton land ownership of 100 ha; however, holdings may be managed in much larger blocks, with nominalownership in the hands of family members, friends and others.



irrigation infrastructure. In all the districts,CNA continues to manage the dams andheadworks, and in most districts, it managesthe main canals and delivers water in bulkto the WUAs as well. However, in the AltoRío Lerma irrigation district, a federation ofWUAs has been formed to manage the mainsystem (Kloezen, 2000).

Research carried out by Kloezen et al.(1997) shows that the WUAs in the Alto RíoLerma irrigation district have been effectivein improving the provision of servicesand recovering costs from users.5 Morerecent work in the district raises questionsabout the WUAs’ long-term sustainability(Monsalvo, 1999; Kloezen, 2000). It is impor-tant to note that IMT was externally imposedand occurred at a time when the Lerma–Chapala basin was already closed. This hasplaced a double strain on the newly createdWUAs. On the one hand, they have hadto learn how to manage sizeable secondarycanal units, at which they have been rela-tively successful, and how to interact withthe CNA. On the other hand, they needto organize themselves by establishingrelations with WUAs in other irrigationdistricts to ensure their voice is heard inriver basin management discussions.

The management structures in theunidades are much more diverse, and mayconsist of informal WUAs, government rec-ognized WUAs, water judges, pump groupsor commercial management (Silva-Ochoa,2000). As state intervention in the unidadeshas been piecemeal in comparison to thedistricts, and has usually only consisted ofassistance in construction and the concess-ioning of water rights, government controlover water use in the unidades is muchweaker than in the irrigation districts. Asa corollary, the representation of unidadesin basin-wide decision-making forums isweak.

8.2.2 Urban water supply

Domestic water supply in the basinmainly depends on groundwater (95%),with total consumptive use at 791 MCM.In addition, water is transferred out of thebasin to provide Guadalajara (237 MCMsurface water) and Mexico City (323 MCMgroundwater) with urban water. The popu-lation in the basin has increased sig-nificantly, doubling from 2.1 millioninhabitants in 1930 to 4.5 million in1970 and then more than doublingagain to 11 million in 2000 (CNA/MW,1999). The population of the basin willagain double in the next 30 years if thepopulation growth rate of 2.16% remainsthe same (CNA/MW, 1999). Besides afivefold increase in population in thepast 70 years, the basin’s population hasbecome strongly urbanized. Populationin the seven largest cities in the basinincreased from 12.7% to 40.9% of thebasin’s total population between 1930 and2000, while the rural population droppedfrom above 75% to less than 25% duringthe same period (CNA/MW, 1999). Popula-tion growth has led to increasing pressureson the basin’s water resources. Scott et al.(2001) project that urban water demand inthe medium term will increase by some4.1% per year.

Starting in 1983, domestic watersupply, wastewater collection, and, morerecently, the operational costs of wastewatertreatment, were decentralized to municipal-ities. The creation of water utilities has beenpromoted to separate these activities fromother municipal responsibilities. However,according to CNA (1999d, p. 8) ‘most ofthe water utilities have a poor perfor-mance and need to be greatly improvedto achieve technical and economicalsufficiency.’


5 Based on an extensive survey of farmers’ perceptions on changes in irrigation management after transfer,Kloezen et al. (1997) report that 36% of the farmers perceived that water adequacy at field level had improvedwith IMT, 26% perceived no change and 23% reported it had become worse. According to 64% of theejidatarios and 47% of the private farmers the condition of the irrigation network had improved, while 54% ofthe ejidatarios and 38% of the private farmers stated the drainage network had improved. Cost recovery went upfrom 50% to more than 100%.



8.2.3 Industry

Although industry only uses a smallamount of water (278 MCM or 3% of annualrenewable water in the basin), it generates35% of Mexico’s industrial GNP and paysaround US$42 million in water taxes6 to thefederal government (CNA/MW, 1999). The6400 registered industrial firms in the basinare a major source of water pollution (fig-ures are not available), although officiallythey must have a permit from CNA indicat-ing effluent standards to discharge waste-water. Fees for discharging wastewaterabove effluent standards were establishedin 1991 and updated annually. Accordingto Tortajada (1999), most polluters do notpay these fees, while industries and citiesthat have submitted wastewater treatmentproposals to CNA are exempted frompayment.

8.3 The Hydro/Institutional Landscape inthe Lerma–Chapala Basin

A watershed year for water managementin Mexico was 1989. Whereas the previous100 years were characterized by increas-ing federal control over water, since1989 decentralization has been thenorm. Currently states, municipalities andwater users have a larger say in watermanagement decision making. The currentdistribution of competencies in watermanagement in the Lerma–Chapala basinis set out below.

8.3.1 Water rights

In Mexico, surface water is defined in theConstitution as national property placed

in the trust of the federal government. Asthe trustee of the nation’s water, the federalgovernment has the right to concessionsurface water-use rights to users for periodsranging from 5 to 50 years (Kloezen, 1998).In irrigation districts and unidades, conces-sions are granted to WUAs and not individ-ual water users. The concession titlesset out the volumes of water concessionholders are entitled to, although CNA mayadjust the quantity each receives annuallyto reflect water availability, with prioritybeing given to domestic water users (CNA,1999b). Thus for allocating surface water,Mexico follows the proportional appropria-tion doctrine and, in theory, all concessionholders share proportionally in any short-ages or surpluses of water.7 Once issued,water concessions need to be recordedin the Public Register of Water Rightsmaintained by CNA. After registration, theconcessions become fully tradeable withinriver basins, although the CNA needs tobe notified of trades and needs to approvethem (Kloezen, 1998).

The situation surrounding groundwateris more complex, as the Constitution doesnot define it as national property but ratherstates that overlying landowners may bringgroundwater to the surface as long as thisdoes not affect other users. In 1946, theConstitution was amended to enable thefederal government to intervene in aquifersin overdraft by issuing pump permitsor declaring that new pumps may not beinstalled. Based on a ruling of the SupremeCourt in 1983, groundwater is now consid-ered national property, although this isnot reflected in the Constitution or the1992 water law (Palacios and Martínez,1999). Groundwater concessions in Mexicoare granted on a volumetric basis, with amaximum extraction or pumping ratespecified.


6 Under the Federal Law on Excise Taxes, updated annually, water users have to pay a tax for the benefitsderived from using water, because it is national property. For agriculture a zero rate was established; farmersonly have to pay irrigation service fees to WUAs (Tortajada, 1999).7 This contrasts with the prior appropriation system, where first rights have seniority.



8.3.2 Water management organizationsand stakeholders

Numerous stakeholders are involved inwater management in the Lerma–Chapalabasin. The government agency responsiblefor water management in the basin is theCNA, a semi-autonomous federal agencyfalling under the Ministry of Environment.Created in January 1989, the CNA ischarged with defining water policy,granting water concessions and wastewaterdischarge permits, establishing norms forwater use and water quality and integratingregional and national water managementplans. The aim of unifying all governmentresponsibilities related to water in the CNAwas to create the necessary conditions formoving towards sustainable water manage-ment (CNA, 1999d). To complement thismove, a modern and comprehensive waterlaw was promulgated in 1992 (CNA, 1999c).This law defines an integral approach formanaging surface and groundwater in thecontext of river basins, which it considersthe ideal unit for water management. Italso promotes decentralization, stakeholderparticipation, better control over waterwithdrawals and wastewater discharges,and full-cost pricing.

To discharge its mandate, CNA consistsof four levels: federal headquarters, regionaloffices, state offices and irrigation districtoffices. Specific responsibilities of thefederal headquarters include the following:

• Manage the nation’s water and act as itscustodian;

• Formulate and update the NationalWater Plan and ensure its execution;

• Maintain the Public Register of WaterRights;

• Monitor water taxes payment and sendreports to fiscal authorities;

• Facilitate the resolution of, or arbitratein, water conflicts;

• Promote conservation and efficient useof water;

• Support the development of urban andrural domestic water supply and drain-age and sanitation networks, includingthe treatment and reuse of wastewater;

• Support the development and manage-ment of irrigation and drainage systemsand storm and flood protection works.

To facilitate river basin managementand interaction with stakeholders, CNAhas divided the country into 13 hydrologicregions based on river basin boundaries andestablished an office in each region. Theseregional offices have been delegated res-ponsibilities from the national level andare relatively autonomous. Their mainresponsibilities include the following:

• Organize and manage CNA’s actionsat the regional level concerning theplanning, execution and evaluation ofthe Regional Water Plan;

• Integrate and validate requests forwater concessions (users), allocations(municipalities) and permits (ground-water and wastewater), issuing thosethat fall in its competency and forward-ing to HQ those that do not;

• Supervise the Public Register forWater Rights offices at the state level,and consolidate and send to HQ allinformation necessary to keep theRegister updated at the national level;

• Assist CNA state offices in theircollection of water taxes from users;

• Integrate and update programmesfor the operation, maintenance andrehabilitation of irrigation districts andwater treatment plants;

• Supervise and assist in the operationof climate and river flow measuringnetworks;

• Enforce legal standards concerningwater pollution and wastewaterdischarges.

Responsibilities for water management atthe state level are more diffuse. CNA hasestablished offices in Mexico’s 31 statesthat function under the supervision ofthe regional offices. The role of stategovernments, as opposed to the federalgovernment, in water management has beenlimited to regulating municipal water utili-ties and supporting utilities showing poortechnical and economic performance. Aspart of the ‘new federalism’ policy during




the Zedillo administration (1995–2000),the federal government promoted delega-tion of responsibilities and programmesto the states, but, notably, not financialresources. Although the federal governmenthas encouraged the modification of statelaws to promote the participation of stategovernments in water management throughthe creation of State Water Commissions,the response has been lukewarm. This isnot the case in Guanajuato, where CEAG(Comisión Estatal del Agua de Guanajuato;Guanajuato State Water Commission) hastaken on its new role with vigour (Guerrero-Reynoso, 2000). The relationship betweenthe State Water Commissions and the CNAstill needs to be defined.

At the irrigation district level, CNAand WUAs share responsibility for watermanagement. According to the 1992 waterlaw and the WUAs’ concession titles, CNAremains the authority in the irrigationdistricts. The CNA Irrigation District Officeretains the following responsibilities andpowers:

• Notify the WUAs on the volume ofsurface water they have been allocatedfor the coming year;

• Operate and maintain the dams andheadworks of the irrigation districtand also the main system if afederation of WUAs has not beenestablished;

• Approve irrigation service fee levels,determined by the WUAs according tothe procedures outlined by the CNA inthe concession title;

• Establish and periodically revise theinstructions for the operation andmaintenance of the secondary canalunits managed by WUAs;

• Approve the WUAs annual mainte-nance plan and ensure that it is carriedout satisfactorily;

• Participate in the General Assembly ofthe WUAs with the right to speak butnot to vote;

• Cancel or refuse to renew concessiontitles if WUAs perform unsatisfactory.

Based on the concession granted to themby the CNA, WUAs legally assume the

responsibility to operate, maintain andadminister their secondary canal unit.Their specific responsibilities to the CNAare the following:

• Develop and enforce bylaws that detailprocedures for water distribution, sys-tem maintenance and investment, costrecovery, and dealing with complaintsand sanctions;

• Collect irrigation service fees that fullycover the O&M and administrationcosts of the WUA;

• Pay CNA a percentage of the revenuesfrom fee collection for CNA servicesrelated to O&M of the dams, headworksand main canal system;

• Prepare annual operation and mainte-nance plans and budgets and send toCNA for approval.

At the river basin level, an important inno-vation has been the creation of river basincouncils. The water law stipulates thatstakeholder participation is mandatory inwater management at the river basin level.To this end river basin councils, definedin the water law as coordinating and con-sensus-building bodies between the CNA,federal, state and municipal governments,and water user representatives (CNA,1999c), have been established by CNA in 25river basins (CNA, 2000a). The stated goalof the councils is to foster the integral man-agement of water in their respective riverbasins through proposing and promotingprogrammes to improve water management,develop hydraulic infrastructure andpreserve the basin’s resources. Formally,the river basin councils have littledecision-making power, as CNA remainsresponsible for water concessions, the col-lection of water taxes and water investmentprogrammes. The role of the councils is toassist CNA in the execution of its vestedpowers and to ensure that CNA takesstakeholders’ opinions into account (CNA,2000a).

Mexico’s first river basin council wasestablished in the Lerma–Chapala basin inresponse to the dropping level of LakeChapala in the 1980s and the severe contam-ination of the Lerma River (Mestre, 1997). In




April 1989, the federal government and thefive state governments signed a coordinationagreement to improve water management inthe basin by: (i) allocating surface andgroundwater fairly among users and regulat-ing water use; (ii) improving water quality bytreating effluents; (iii) increasing water-useefficiency; and (iv) conserving the riverbasin ecosystem and watersheds. Based onthis agreement, a formal Consultative Coun-cil was formed in September 1989 to followup on these objectives. Based on the 1992water law the Consultative Council becamethe Lerma–Chapala River Basin Council on28 January, 1993.

In the past 10 years, the River BasinCouncil has been in flux, and only in August2000 was its structure formalized (CNA,2000a). It now consists of a GoverningCouncil, a Monitoring and Evaluation Group(MEG), a Basin Level User Assembly andSpecial Working Groups, while the CNA’sregional office forms the Council’s secre-tariat (Fig. 8.3). The Governing Council ischaired by CNA, while user representativesfrom six sectors (agriculture, fisheries, ser-vices, industry, livestock and urban) and thegovernors of the five states falling in thebasin are its members, yielding a total of 12

members. Although agriculture uses 68% ofthe annual renewable water in the basin, itonly has one vote out of 12 on the Council.The decision-making body of the RiverBasin Council is the MEG, which is a carboncopy of the Governing Council except thatstate governors send representatives in theirstead, while CNA is represented by the headof its regional office. The MEG meets on aregular basis to prepare and convene Coun-cil meetings and more importantly to draftagreements to be signed at formal Councilmeetings.

The structure of the River Basin Councilis complemented by a stepped form of userrepresentation, consisting of water usercommittees for the six water use sectors rep-resented on the Council. These committeescan be formed at the regional, state or locallevel, building on already existing WUAs orother legally recognized water managementgroups where possible. The water user com-mittees form the Basin Level User Assembly,which elects the six user representatives onthe Council. In addition, forums at the sub-basin level, such as watershed commissionsand aquifer management councils, form partof the structure of the River Basin Council(Fig. 8.3).


��

��

��

��

!� ��

"��

��"�#

$��

��

��%��

��&��' �� $��

��(��)*�� +��

��+ ��,

��-�.��/ �

'��& ' ��0

�� %

�1 ��2 (

$��,��& ��1�

��+ � "��

��/��'��

� ��

��1�� ''�� &��

�� 3445367 ��( 368

��+ ��' ��

��

�� (

�9� ��:��

��''��

��''��/�� ,

��1�� "��

��

�**��

Fig. 8.3. Structure of the Lerma–Chapala River Basin Council.



8.4 Essential Functions for River BasinManagement

As set out in Chapter 1, a set of essentialfunctions has been posited to analyse theinstitutional arrangements for river basinmanagement. A list of these functions asthey apply in the Lerma–Chapala basin,crossed with the key actors identifiedin the previous section, is shown inTable 8.2. Cells are marked to indicate ifan actor is active in a particular functionalarea, while action verbs such as Execute(E), Supervise (S), Advise (A), Authorize(Aut), Regulate (Reg) and Represent (Rep)are used to describe the nature of an actor’sactivities or responsibilities. The table,together with the description of the actors,gives an indication of which essentialfunctions are accounted for and who isinvolved in their execution. In itself, this isuseful to identify whether certain essentialfunctions are not being executed. A numberof interesting points emerge from Table 8.2:

• Most of the essential functions relatingto surface water are covered.

• The withdrawal and distribution ofgroundwater is weakly regulated, andthe construction of new facilitiescontinues although this has beenprohibited.

• Monitoring and ensuring the qualityof primary water sources are relativelyweak functions. Stopping groundwaterpollution is a function that is notcovered at all.

• The monitoring of wastewater qualityand enforcing wastewater dischargenorms are also functions that receivelittle attention, while the allocationand distribution of wastewater in thebasin is not regulated at all and left toactors on the ground.

• Industries and non-governmental orga-nizations (NGOs) are hardly involvedin water management in the basin.

Another, perhaps more interesting, ques-tion is how efficiently the functions areexecuted and if they are effective. Toevaluate how well the functions are being

performed, an evaluation matrix was devel-oped and applied to the Lerma–Chapalabasin. This matrix consisted of eightvariables for each essential function, fourrelated to effectiveness and four relatedto efficiency. For each variable, threequestions were formulated that could beanswered on a scale of 1 to 5, giving a maxi-mum score of 60 points for effectivenessand 60 points for efficiency per function.The scores per function can be set outagainst each other, with effectiveness on thex-axis and efficiency on the y-axis, yieldinga graphical representation of how well thefunctions are executed (Fig. 8.4). The scorespresented in Fig. 8.4 are the averages of theresponses given by 15 key water manage-ment stakeholders in the basin. Although adrawback of the evaluation matrix is that itdid not distinguish between surface andgroundwater or between primary and deriv-ative water sources, it is presented here asit reflects overall impressions of key watermanagement actors in the Lerma–Chapalabasin regarding the execution of the essen-tial functions in their basin. An interestingpoint that emerges from Fig. 8.4 is that noneof the essential functions are executed effi-ciently according to key water managers inthe basin. What also stands out is that, apartfrom protecting the environment, the effec-tiveness of the execution of the essentialtasks is moderate. The next section presentsmore detail on the overarching issues facingthe basin.

8.5 Overarching Issues

Through steady changes in the institutionalarrangements in the Lerma–Chapala basinin recent years, progress has been madetowards improved water management.This progress is significant, in light of thecomplicated transition from highly central-ized water management to one in whichstates, municipalities and water users havea larger say. None the less, from a water per-spective the Lerma–Chapala basin is still incrisis. The efforts of the Council over thepast 10 years need to be redoubled to tackle




the significant challenges lying ahead ofit. Three challenges stand out, namelythe allocation of surface water and ground-water, and the management of derivativewater.

8.5.1 Surface water allocation

To allocate surface water in the basin, thegovernors of the five states in the basin andthe federal government signed a treaty inAugust 1991 (CCCLC, 1991). An importantobjective of the treaty is to maintain ade-quate water levels in Lake Chapala and toensure Guadalajara’s domestic water sup-ply. To preserve Lake Chapala, the treatysets out three allocation policies, namelycritical, average and abundant, based onwhether the volume of water in the lake isless than 3300 MCM, between 3300 and6000 MCM, and more than 6000 MCM,respectively. Each year the Council verifiesthe volume stored in Lake Chapala to deter-mine the allocation policy to be followedfor the next year. For each allocation policy,formulas have been drawn up to calculate

water allocations to the irrigation systemsin the basin, based on the surface runoffgenerated in each of the five states in theprevious year. Table 8.3 indicates how thisworks for the Alto Río Lerma irrigation dis-trict. Based on extensive modelling of theseformulas, it was concluded that the result-ing water allocation would not impinge onthe 1440 MCM needed by Lake Chapala forevaporation. Thus, as shown in Table 8.3, ifthe surface runoff generated is below a cer-tain threshold, a fixed volume is deductedfrom the irrigation district’s allocation sothat this can be passed on to the lake.

Since 1991, the MEG of the Council hasmet each year and has applied the waterallocation rules set out in the treaty. Figure8.5 sets out the volumes of water allocatedand used from 1992 to 2000, as well as thevolume of water stored in Lake Chapala.This shows that the 1991 treaty has beenenforced, as actual use has never been higherthan the allocated values. A caveat here isthat only the extractions by irrigation dis-tricts are accurately measured, thus actualwithdrawals may have been higher as theamount of water going to the unidades deriego is unknown.


Effi

cien

cy

Effectiveness

1

4765

32

8

Ineffective and inefficient Effective but inefficient

Efficient but ineffective Efficient and effective

1. Protect ecology 2. Construct facilities 3. Flood protection 4. M/E Water quality 5. Maintain facilities 6. Plan (basin-level) 7. Distribute water 8. Allocate water

Fig. 8.4. Effectiveness and efficiency of essential functions in the Lerma–Chapala basin.



Despite the apparent compliance withthe surface water treaty, Lake Chapala’svolume has halved in the past 8 years. Thisis so, in part, because the treaty takes thesurface runoff generated in the previousyear to determine water allocations. In 1997rainfall was only 645 mm (against theaverage annual value of 712 mm) and damstorage (used here as a proxy of surfacerunoff) was consequently low. Combinedwith a lake volume below 3300 MCM, thecritical allocation policy was followed for1998, leading to the lowest allocations sincethe treaty was signed. However, rainfall in1998 was exceptionally good, at 810 mmsome 100 mm above average, leading to a

recuperation of the volume of water storedbehind dams and a slight increase in thevolume of Lake Chapala to 3361 MCM. Asa result, the average allocation policy wasfollowed for 1999 and 3664 MCM were allo-cated to water users, the highest level sincethe signing of the treaty. Unfortunately, rain-fall in 1999 was a historic low of 494 mm.These two factors resulted in Lake Chapaladropping to its lowest level since the signingof the treaty and point to inadequateprovisions in the treaty for inter-annualplanning of water availability and dealingwith contingencies.

The members of the Council have recog-nized the shortcomings of the surface water


Lake Chapala volumeSurface runoff generated (SRG) inthe State of Guanajuato (MCM)

Volume allocated (VA) to irrigationdistrict (MCM)

Critical

Average

Abundant

if SRG between 280 and 1260if SRG > 1260if SRG between 144 and 1125if SRG between 1125 and 1400if SRG > 1400if SRG between 19 and 1000if SRG between 1000 and 1200if SRG > 1200

then VA = 94.2% of SRG–262.8then VA = 924then VA = 94.2% of SRG–135.6then VA = 924then VA = 955then VA = 94.2% of SRG–17.9then VA = 924then VA = 955

Source: CCCLC (1991).

Table 8.3. Water allocation principles for the Alto Río Lerma irrigation district.

0

1.000

2.000

3.000

4.000

5.000

6.000

1992 1993 1994 1995 1996 1997 1998 1999 2000

Vol

ume

(MC

M)

Volume allocated Volume used

Lake Chapala volume on 1 Nov. Dam storage on 1 Nov.

Fig. 8.5. Surface water allocated and used in the Lerma–Chapala basin. Sources: CNA (1991, 1992,1993a, 1994, 1995, 1996, 1997, 1998, 1999b, 2000b).

150A4859 - Svendsen - 150.prnZ:\Customer\CABI\A4831 - svendsen\A4859 - Svendsen - Vouchers Proofs #P.vpFriday, January 14, 2005 10:59:33 AM


treaty and in 1999 decided to revise thetreaty, as it was clear that it was not rescuingLake Chapala. In 1999 and 2000, detailedstudies were carried out with hydrologicaldata from 1945 to 1997 (an improvementover the 1950–1979 data used for theprevious treaty) to develop a new model forcalculating surface runoff (CNA, 1999e). TheCouncil signed the amendment of the 1991surface treaty on 24 August, 2000 (Consejode Cuenca Lerma–Chapala, 2000). However,various states in the basin feel that they didnot have sufficient input in the design of thesurface runoff model and that CNA imposedthe treaty on them. In addition, consultationwith water users concerning the new treatyhas been minimal, although user representa-tives on the Council voted in favour.Although the signing of the new treatyshows the adaptability of the Council andthe commitment of its members to construe awater allocation policy that meets urban andagricultural needs while safeguarding theenvironment, the process by which it wasarrived at is contested.

An issue that the Council has not yetstarted to consider is how to compensatefarmers for water transferred out of agricul-ture for urban and environmental demands.Scott et al. (2001) calculate that the benefitsforgone for farmers in the Alto Río Lermairrigation district as a result of a reducedwater allocation to the district for 1999/2000amounted to US$14 million. Although suffi-cient water was stored in the district’s reser-voirs to cover its full allocation (955 MCM)the district was allocated only 648.2 MCMunder the treaty, due to the criticallylow volume of water in Lake Chapala andthe minimal surface runoff generated inGuanajuato in 1999. To shore up waterlevels in Lake Chapala, the CNA released240 MCM from the district’s main reservoirinto the Lerma River in October 1999,thereby transferring surface water fromthe agricultural sector to the urban andenvironmental sector.

The reduced allocation to the Alto RíoLerma irrigation district in 1999 resulted insome 20,000 ha out of a total of 77,000 ha notbeing irrigated with surface water in thewinter season. Four of the 11 WUAs in the

district went without irrigation altogether,whereas in the other WUAs, irrigation wasrestricted to a maximum of 3 ha per land-owner. For better off farmers who couldswitch to groundwater, this was not tooproblematic, but for many poorer farmerswho mainly rely on surface water, the resultswere disastrous. In addition, many poorfarmers who traditionally pumped returnflows from the Lerma River were hard hit asthe use of this precarious source of waterwas prohibited and enforced through armypatrols along the river. The surface waterallocations for all the irrigation districts inthe basin for the 2000/01 winter season wereso low that the WUAs decided to let200,000 ha (of a total of 235,000 ha in theirrigation districts normally irrigated withsurface water) lie idle. Once again, poorfarmers were hardest hit. On the other hand,Lake Chapala has dropped to its lowestlevels in 50 years, and environmentalNGOs and the Jalisco state government aredemanding the transfer of water storedbehind the dams of the irrigation districts tothe lake.

The lack of water in the past 2 years hasgalvanized WUA leaders to take action. InMay 2000, the presidents of WUAs locatedin Jalisco, Guanajuato and Michoacán meteach other for the first time to discuss waysto strengthen their representation in theRiver Basin Council. Until then, WUAs of aparticular irrigation district had dealt onlywith the CNA, and there were no horizontallinkages between WUAs from different dis-tricts. In 2001, WUAs from irrigation dis-tricts in Querétaro and Mexico joined thediscussions, and the combined WUAs estab-lished a new working group in the RiverBasin Council focusing on agriculture. Theyhave vowed that not a single drop of waterwill be passed to Lake Chapala and havethreatened civil disobedience if CNA trans-fers water as in October 1999. It is clear that aserious conflict is brewing in the basin sur-rounding surface water and that the Councilneeds to act fast to come up with a workablesolution. What is more worrisome is that thelack of surface water has led many farmersto increase their use of groundwater, whileaquifers are already in perilous decline.




8.5.2 Groundwater management

The serious overdraft of the basin’s aquifersis arguably a more pressing issue than sur-face water allocation. Although the Councilsigned a coordination agreement to regulategroundwater extraction in the basin in1993, progress on the ground has been slow(CCCLC, 1993). A key problem is that theCouncil, through the CNA, does not physi-cally control the water extraction infra-structure (the wells), as it does in the case ofsurface water (the dams). Although the Con-stitution mandates the federal governmentto intervene in aquifers in overdraft by plac-ing them under veda, prohibiting sinkingof new wells without permission from thefederal government, the experience withvedas has been disappointing (Arreguín,1998). The reality of groundwater extractionin Guanajuato clearly shows how ground-water regulation by the federal governmenthas run aground. According to Vázquez(1999) 10 vedas were issued in Guanajuatobetween 1948 and 1964 prohibiting thedrilling of new wells in large parts of thestate, and in 1983, the remainder of the statewas placed under veda. Notwithstandingthese legal restrictions, the number ofwells in Guanajuato increased fromapproximately 2000 in 1958 to 16,500 in1997 (Guerrero-Reynoso, 1998).

Based on the recognition that vedashave not worked and to counter thecontinued depletion of groundwater in thebasin, CNA started promoting the formationof Comités Técnicos de Aguas Subterráneas(COTAS; aquifer management councils)in selected aquifers in 1995, as an out-growth of the 1993 agreement. Through theestablishment of COTAS, which fall underthe River Basin Council, the CNA is seekingto stimulate the organized interaction ofaquifer users with the aim to establishmutual agreements for reversing ground-water depletion, in keeping with Article 76of the water law regulations (CNA, 1999c).Based on recent developments in the State ofGuanajuato, where CEAG enthusiasticallypromoted the creation of COTAS (Westeret al., 1999; Guerrero-Reynoso, 2000), thestructure of the COTAS has been defined at

the national level in the rules and regula-tions for river basin councils (CNA, 2000a).In these rules, the COTAS are definedas fully fledged user organizations, whosemembership consists of all the water users ofan aquifer. They are to serve as mechanismsfor reaching agreement on aquifer manage-ment taking into consideration the needsof the various sectors using groundwater(CNA, 2000a).

To date, 14 COTAS have been formedin the basin. However, none of them hasyet started to devise ways to reduce ground-water extraction. Considering that some380,000 ha in the basin are irrigated withgroundwater and that industrial and domes-tic uses depend almost entirely on ground-water, it is fair to say that groundwater is thestrategic resource in the basin, particularlyfrom a water productivity perspective. Thelong-term consequences of its continueddepletion easily overshadow those of LakeChapala drying up. Although the COTASare a timely institutional response to thepressing need for innovative approaches tomanaging aquifers in the basin, it remainsto be seen if they will succeed in reducingaquifer over-exploitation. Current discus-sions in the COTAS focus on installingsprinkle and drip irrigation systems to savegroundwater, but the tough issue of howto reach agreement on an across the boardreduction in pumping has not yet beenbroached. In addition, new pumps continueto be installed and regularized throughextra-legal means. The reluctance of thegovernment to impose strict pumping limitsand the continued race to the pumphouse byfarmers bode ill for the COTAS.

8.5.3 Management of derivative water

The effects of the new institutional arrange-ments in the basin on surface and ground-water management were briefly describedabove. The picture that emerges is that,although some progress has been made,the basin’s primary water sources arestill seriously overdrawn, to the particulardetriment of poorer farmers. For them,




derivative water is becoming a critical, andfrequently the only, available source ofwater (cf. Buechler and Scott, 2000). In1993, the municipal wastewater flowgenerated in the basin was estimated at12,700 l/s, and by 1997, this had risen to17,000 l/s, which is equal to around536 MCM per year (Mestre, 1997). Thereturn flows from agriculture are notmeasured, but the fact that hardly anywater flows into Lake Chapala suggests that,whatever the magnitude of the return flows,they are used upstream. The extent of waste-water irrigation in the basin has not beenaccurately assessed, but estimates rangefrom 20,000 to 40,000 ha (CNA/MW, 1999).The use of return flows and wastewater iscurrently a free-for-all and is not formallyregulated by the institutional arrangementsfor water management in the basin. How-ever, as Buechler and Scott (2000) outline,farmers at times need to obtain informalpermission from municipalities or WUAsto use wastewater. These claims are beingthreatened by the construction of waste-water treatment plants and the de factoreallocation of treated water to other uses.

In 1989, the River Basin Councillaunched an investment programme toclean the Lerma River through the plannedconstruction of 48 treatment plants with acapacity of 3700 l/s. Priority was given tocities with more than 10,000 inhabitants thatdischarged directly into the Lerma River orLake Chapala. This was essential, as up to1989 hardly any of the municipal and indus-trial wastewater in the basin was treated. Asa result, large stretches of the Lerma Riverwere heavily contaminated, as were twoof its tributaries, the Turbio and QuerétaroRivers, while Lake Chapala was classified ascontaminated. On 28 January, 1993, the riverbasin council agreed on a second water treat-ment programme, entailing the constructionof 52 new plants and the enlargement of fiveexisting ones, with a combined treatmentcapacity of 10,835 l/s (Mestre, 1997). It wasforeseen that the two investment program-mes would create a total treatment capacityof 14,561 l/s, representing 85% of municipalwastewater generated in the basin. At theend of 1999, 48 plants had been constructed

with a treatment capacity of 6037 l/s, whilesix additional large plants are still underconstruction, with a capacity of 3513 l/s(CNA, 1999a). Of the installed capacity of6037 l/s, only 3155 l/s provides full treat-ment. Although the River Basin Council hasnot succeeded in attaining the ambitiousgoals it set for itself, the installed capacity of6037 l/s, up from 0 l/s in 1989, is an achieve-ment in itself. The Council readily admitsthat more needs to be done, although thelarge investments required and the lack offinancial resources at the basin level makesthis difficult (CNA, 1999a).

Scott et al. (2000) place the value ofwastewater irrigation for 140 ha of irrigatedland that depend on the City of Guanajuatofor their water at some US$252,000 per year.As part of the Council’s water treatmentprogramme, Guanajuato City is in the pro-cess of contracting for an activated sludgewastewater treatment plant and plans to sellthe treated water to commercial interests. Asa result, farmers will lose their access towastewater as well as the nutrient benefits ofthat water. This raises the question how a defacto right to derivative water should beaddressed when water treatment redirectswastewater outflows. Although the CNA,the River Basin Council and the State WaterCommissions are actively pursuing the con-struction of wastewater treatment plants,the allocation and distribution of derivativewater has received no attention. In closingriver basins, where primary water sourceshave already been captured by the better offand renegotiating water rights is extremelycomplicated, derivative water rapidlybecomes the poor farmer’s last resortand should be recognized as such.

8.6 Conclusions

This chapter has explored the dynamicinterplay between trends in water use andinstitutional changes for water managementin the Lerma–Chapala basin. Drought andwater shortages between 1945 and 1954resulted in a doubling of the reservoircapacity within the basin. Pressures on




available primary water, both surface andground, continued to increase with the irri-gated area increasing almost fourfold up to1989 and the population increasing almostthreefold. After primary water resourceshad become over-committed, institutionalreforms were introduced from 1989 onwardsdevolving responsibility for water manage-ment in irrigation systems to water usersand initiating participatory water manage-ment bodies at the basin level for high-leveldecision making on water allocation.

A central component in this reform hasbeen the 1992 water law, though of equalimportance have been the institutionalcapability to put the law into practice andinstitutions’ ability to adapt to a dynamicsituation with additional measures for con-trolling and managing available surface andgroundwater resources. The assessment ofthe institutional arrangements for watermanagement in the Lerma–Chapala basinbrings out clearly the need for coordinatingmechanisms at the basin level in river basinsfacing closure. Through the Lerma–ChapalaRiver Basin Council, progress has been madetowards improved water management in thebasin. This progress is encouraging, in lightof the complicated transition from a highlycentralized management of water resourcesto one in which states, municipalities andwater users have a larger say. Though theinstitutional measures have had a signifi-cant impact in restructuring water manage-ment, the basin is still in crisis, with thelevel of Lake Chapala in perilous declineand aquifers running dry. The efforts ofthe River Basin Council over the past tenyears will need to be redoubled to tackle thesignificant challenges lying ahead of it.

Although surface water allocationmechanisms are working, and the revisionof the 1991 treaty may lead to increasedinflows to Lake Chapala, compensatingfarmers for water transferred out of agri-culture needs to be considered. In closedbasins, inter-sectoral transfers are inevitableand it will invariably be the irrigation sectorthat will need to cede water. A key institu-tional challenge in closed river basinsis dealing with these transfers in a justand equitable manner, such that the

sustainability of locally managed irrigationis ensured. The Lerma–Chapala River BasinCouncil could be a forum for drawing up andenforcing compensation mechanisms forsurface water transferred out of agriculture,but to date it has been reluctant to considerthis issue due to the costs involved. To becredible and effective partners in basinmanagement, WUAs and farmers need tomake concerted efforts to reduce water use,for which there is real scope, and to joinforces to argue their case more strongly inthe River Basin Council.

A much more serious challenge thatthe Council and other water managementstakeholders in the basin need to deal withurgently is the overdraft of the basin’saquifers. The aquifer management councilsare a step in the right direction, but theirrole in groundwater management should gobeyond mere consultation. Bundling extrac-tion rights in an aquifer and concessioningthis to a COTAS is feasible under theMexican water law and should be seriouslyconsidered. Placing aquifer management inthe hands of the aquifer users, under thesupervision of the River Basin Council, StateWater Commissions and the CNA, showsmore promise of reducing extractions thanthe current system of vedas and federalregulation.

With basin closure, poorer farmersin particular are losing, or have lost, theiraccess to primary water due to reductions insurface irrigation and increased costs forgroundwater irrigation. Hence derivativewater, both wastewater and agriculturaldrainage effluent, is increasingly becoming acritical resource for poor farmers. However,its allocation in the basin is currently a free-for-all. Although progress has been made inconstructing wastewater treatment plants,the management of derivative water hasreceived little attention. More generally,meeting the water needs of poor people, andincluding poor women and men at all levelsof water management decision making, isnot a priority of the Council, nor is it a strongfeature of the larger set of institutionalarrangements for water management inMexico. The Council needs to seriouslyconsider how to safeguard and improve the




access of the poor to water, and how to com-bat the current de facto concentration ofwater rights in the hands of the few. Overall,a coherent approach for tackling the contin-ued pressures on the basin’s primary watersources, the significant re-use of wastewater(derivative water) and the deterioration ofwater quality needs to be developed.

References

Arreguín, J. (1998) Aportes a la historia dela geohidrología en México, 1890–1995.CIESAS-Asociación Geohidrológica Mexi-cana, Mexico City.

Buechler, S.J. and Scott, C.A. (2000) ‘Paranosotros, esta agua es vida.’ El riego encondiciones adversas: Los usuarios de aguasresiduales en Irapuato, México. In: Scott,C.A., Wester, P. and Marañón-Pimentel, B.(eds) Asignación, productividad y manejo derecursos hídricos en cuencas. InternationalWater Management Institute (IWMI), SerieLatinoamericanaNo. 20. IWMI, Mexico City.

CNA (Comisión Nacional del Agua) (1991, 1992,1993a, 1994, 1995, 1996, 1997, 1998, 1999b,2000b) Volumenes maximos de extracción deagua superficial para los sistemas de usuariosde la cuenca Lerma-Chapala. Boletin No.1–10. CNA, Mexico City.

CNA (1993b) Plan maestro de la cuencaLerma-Chapala. Documento de Referencia.CNA, Mexico City.

CNA (1999a) El consejo de cuenca Lerma–Chapala 1989–1999. 10 años de trabajo enfavor de la gestión integral y manejosustentable del agua y de los recursosnaturales de la cuenca. CNA, Guadalajara.

CNA (1999c) Ley de aguas nacionales y sureglamento. CNA, Mexico City.

CNA (1999d) Regional Vision North America(Mexico). CNA, Mexico City.

CNA (1999e) Estudio de disponibilidad y balancehidráulico de aguas superficiales de lacuenca Lerma-Chapala y cuencas cerradasde Pátzcuaro. CNA, Guadalajara.

CNA (2000a) Reglas de organización y funcion-amiento de los consejos de cuenca. CNA,Mexico City.

CNA/MW (1999) Proyecto lineamientosestratégicos para el desarrollo hidráulicode la Región Lerma–Santiago–Pacifico. Diag-nostico regional. CNA/Montgomery Watson,Guadalajara.

CCCLC (Consejo Consultivo de Evaluación ySeguimiento del Programa de Ordenación ySaneamiento de la Cuenca Lerma-Chapala).(1991) Acuerdo de coordinación de aguassuperficiales. Colección Lerma-Chapala,vol. 1, no. 5. CNA, Queretaro.

CCCLC (1993) Propuesta de acciones parareglementar la distribución, uso y aprove-chamiento de las aguas del subsuelo enlos acuiferos que se ubican en la cuencaLerma-Chapala. CNA, Queretaro.

Consejo de Cuenca Lerma-Chapala (2000)Actualización de las bases y procedimientospara el cálculo de disponibilidad y distri-bución de las aguas superficiales. Consejo deCuenca Lerma-Chapala, Guadalajara.

de Anda, J., Quiñones-Cisneros, S.E., French, R.H.and Guzmán, M. (1998) Hydrologic balanceof Lake Chapala (Mexico). Journal of theAmerican Water Resources Association 34,1319–1331.

de P. Sandoval, F. (1994) Pasado y futuro del LagoChapala. Gobierno de Jalisco, Guadalajara.

Flores-López, F.J. and Scott, C.A. (2000) Superficieagrícola estimada mediante análisis deimágenes de satélite en Guanajuato, México.IWMI, Serie Latinoamericana No. 15. IWMI,Mexico City.

González-Villareal, F. and Garduño, H. (1994)Water resources planning and managementin Mexico. Water Resources Development 10,239–255.

Gorriz, C.M., Subramanian, A. and Simas, J. (1995)Irrigation Management Transfer in Mexico.Process and Progress. World Bank TechnicalPaper No. 292. World Bank, Washington,DC.

Guerrero-Reynoso, V. (1998) Participación socialen el aprovechamiento sustentable de lasaguas subterráneas – el caso de Guanajuato.In: Memoria del Simposio Internacional deAguas Subterráneas, 7–9 December. León,Guanajuato, Mexico, pp. 33–42.

Guerrero-Reynoso, V. (2000) Towards a new watermanagement practice: experiences and pro-posals from Guanajuato State for a participa-tory and decentralized water managementstructure in Mexico. Water ResourcesDevelopment 16, 571–588.

Johnson III, S.H. (1997) Irrigation ManagementTransfer in Mexico: a Strategy to AchieveIrrigation District Sustainability. IIMIResearch Report 16. IWMI, Colombo, SriLanka.

Kloezen, W. (1998) Water markets betweenMexican water user associations. WaterPolicy 1, 437–455.




Kloezen, W. (2000) Viabilidad de los arreglosinstitucionales para el riego después dela transferencia del manejo en el distritode riego Alto Río Lerma, México. IWMISerie Latinoamericana No. 13. IWMI, MexicoCity.

Kloezen, W.H., Garcés-Restrepo, C. and JohnsonIII, S.H. (1997) Impact Assessment ofIrrigation Management Transfer in the AltoRio Lerma Irrigation District, Mexico. IIMIResearch Report 15. International IrrigationManagement Institute (IIMI), Colombo, SriLanka.

Kloezen, W. and Garcés-Restrepo, C. (1998)Assessing Irrigation Performance withComparative Indicators. The Case of theAlto Rio Lerma Irrigation District, Mexico.IWMI Research Report 22. IWMI, Colombo,Sri Lanka.

Mestre, E. (1997) Integrated approach to riverbasin management: Lerma-Chapala casestudy – attributions and experiences in watermanagement in Mexico. Water International22, 140–152.

Monsalvo, G. (1999) Sostenibilidad institucionalde las asociaciones de riego en México.IWMI, Serie Latinoamericana 9. IWMI,Mexico City.

Palacios, E. and Martínez, R. (1999) Aspectosconstitucionales, legales e institucionales.In: Politicas opcionales para el manejo dela sobrexplotación de acuiferos en México.Estudio sectoral. World Bank/CAN, MexicoCity, pp. 6.1–6.41.

Rap, E., Wester, P. and Pérez-Prado, L.N. (2004)The politics of creating commitment:Irrigation reforms and the reconstitutionof the hydraulic bureaucracy in Mexico.In: Mollinga, P.P. and Bolding, A. (eds)

The Politics of Irrigation Reform. AshgatePublishers, Aldershot, pp. 57–94.

Scott, C.A., Zarazúa, J.A. and Levine, G. (2000)Urban-wastewater Reuse for Crop Pro-duction in the Sater-short Guanajuato RiverBasin, Mexico. IWMI Research Report 41.IWMI, Colombo, Sri Lanka.

Scott, C.A. and Garcés-Restrepo, C. (2001) Con-junctive management of surface water andgroundwater in the Middle Río Lerma Basin,Mexico. In: Biswas, A.K. and Tortajada, C.(eds) Integrated River Basin Management:The Latin American Experience. OxfordUniversity Press, New Delhi.

Scott, C.A., Silva-Ochoa, P., Florencio-Cruz, V.and Wester, P. (2001) Competition for waterin the Lerma-Chapala Basin. Economic andpolicy implications of water transfers fromagricultural to urban uses. In: Hansen, A. andvan Afferden, M. (eds) The Lerma–ChapalaWatershed: Evaluation and Management.Kluwer Academic, Dordrecht, pp. 291–323.

Silva-Ochoa, P. (ed.) (2000) Unidades de Riego:La otra mitad del sector agrícola bajo riego.IWMI, Serie Latinoamericana No. 19. IWMI,Mexico City.

Tortajada, C. (1999) Legal and regulatory regimefor water management in Mexico and itspossible use in other Latin Americancountries. Water International 24, 316–322.

Vázquez, M.L. (1999) Decretos de veda en el estadode Guanajuato. Aqua Forum 4, 20–21.

Wester, P., Marañón-Pimentel, B. and Scott, C.A.(1999) Institutional responses to groundwa-ter depletion: The aquifer management coun-cils in the State of Guanajuato, Mexico. Paperpresented at the International Symposium onIntegratedWater Management in Agriculture,16–18 June, Gómez Palacio, Mexico.




9 Water Resources Planning andManagement in the Olifants Basin of South

Africa: Past, Present and Future

M. de Lange, D.J. Merrey, H. Levite and M. Svendsen

9.1 Introduction

The singular history of South Africa hascreated huge development challenges,along with important opportunities toaddress them. During apartheid, rights toproductive resources such as land, waterand minerals were controlled by the domi-nant minority, while the remainder of thepopulation was concentrated into isolated‘bantustans’ or ‘homelands’. Although for-mal barriers to residence and economicactivity were quickly dismantled in thewake of democratic elections in 1994, thelegacy of earlier policies persists in ruralareas. This manifests itself in huge incomedisparities; dense concentrations of blackSouth Africans in the regions of the former‘homelands’; a skewed distribution of basicinfrastructure such as roads, electricity andpiped water supplies; extremely high levelsof unemployment; and a dualistic agricul-tural economy.

In irrigation, this dualism is reflected inthe disparity between the commercial andthe subsistence or small-scale irrigation sec-tor. Only about 100,000 ha of the nationaltotal of 1.3 million ha is currently worked bysmall-scale farmers. A sophisticated irriga-tion industry with extensive manufacturingand equipment distribution capacity servesthe commercial sector, while small-scalefarmers remain isolated from this capacity

by insecure land tenure, and lack of formaleducation, capital, market access and man-agement experience. At the same time, thepresence of such a sophisticated industryprobably inhibits the development of a moreinformal irrigation services sector. One ofthe results of these disparities is that, in thenew government, there is an almost exclu-sive policy focus on the historically disad-vantaged sector, in contrast to the previouspredominant focus on the commercialsector. The historically disadvantagedsector is receiving a spectrum of supportto both deracialize commercial agricultureand create sustainable livelihoods for the12 million people residing in the formerbantustans. Although agriculture directlygenerates less than 5% of South Africa’sgross domestic product (GDP), it is only nowgaining recognition for its importance incombating widespread rural poverty and asa stabilizing factor in the national economy.Taking a broader perspective on the contri-bution of agriculture to GDP and includingassociated support services and agro-industries, agriculture actually accounts formore than 14% of the total. The GDP multi-plier of agriculture is 1.51 overall. Further,out of an economically active population of13.8 million people, at least 35% is directlyor indirectly dependent on agriculture.About 10% of total export earnings ofthe country is from agriculture. Irrigation




produces a quarter of the agricultural outputon 11% of the cultivated land (Hirschowitz,2000; Mullins, 2002).

One of the early actions of the post-apartheid government was to formulatea new and progressive water policy thatmandated, among other things, integratedmanagement of water resources at the basinlevel. The vehicle for basin level manage-ment is the Catchment Management Agency(CMA), which is intended to be the primarypolicy-making and management entity atthe basin level. The country is presentlyengaged in implementing this policy, and inthe process, confronting a number of verychallenging issues. These include the task ofdeveloping integrated representative gover-nance of the CMA in a bipolar social and eco-nomic environment, sharing costs amongwater-using sectors, and formalizing andreallocating water use entitlements in acontext of growing water scarcity.

The present study analyses manage-ment of a South African river basin underthe old regime and policies, and that envi-sioned under the new set-up. The analysisemploys a matrix of essential functions andkey actors to identify the extent of coverage,gaps and overlaps, extent of participation,and needs for coordination. It focuses on oneof the pilot basins in which the new

approach is being developed and tested – theOlifants.

9.1.1 Overview

9.1.2 The Olifants basin

The Olifants River basin is located in thenortheastern corner of South Africa andsouthern Mozambique (Fig. 9.1).

The bulk of the South African part of thebasin lies in Limpopo and MpumalangaProvinces, with a small portion in GautengProvince. The river flows from southwest tonortheast and, upon leaving South Africa,enters Mozambique and joins the Limpoporiver before discharging into the IndianOcean about 200 km north of the capitalMaputo.

The Olifants River is about 770 km longand, with its tributaries, drains 73,534 km2.Under the new National Water Act, the CMAexcludes the two northernmost tributaries,bringing the total drainage area under thefuture CMA to about 54,000 km2, an area thesize of Slovakia or Croatia.

The climate is semiarid, with rain fall-ing primarily during the summer (Novemberto March). Precipitation averages 630 mmand potential evaporation is 1700 mm.

146 M. de Lange et al.

Fig. 9.1. Location of the basin in the region.



Although situated only 24° south of theequator, much of the basin is located atrelatively high elevations (300–2300 mabove sea level). This explains its coolwinters and the wide annual temperaturevariation, which ranges from −4 to 45°C.

The population of the basin was 3.4million in 2000. Population densities varyconsiderably, depending on whether a par-ticular area was a former black ‘homeland’(bantustan) or a part of the former whitearea, with densities ranging from 100 to 350people/km2 in former black areas and from50 to 100 people/km2 in former white areas.Whites currently comprise about 7% ofthe basin population. Ninety per cent of theblack population lives in rural areas in thebasin. The rate of illiteracy in the basin ismore than 50%.

The basin is divided into five hydro-logic areas generally regrouped in fourecological regions (Fig. 9.2).

• The Upper Olifants which correspondto the Highveld region;

• The Upper Middle and Lower MiddleOlifants represent the Middleveldregion;

• The Steelpoort basin assimilated to theMountain area;

• The Lower Olifants situated in theLowveld region.

9.1.2.1 Highveld region

The upper basin or Highveld region has aslightly higher rainfall than the Middleveld,and is characterized by extensive rainfedcropping and stock farming, coal miningand coal-fired power generation. Althoughthe pollution impact of these activitiesis significant, the strategic importance isclear: 55% of South Africa’s electricityis produced here. Annually, around 200million m3 of water is imported, mainlyfrom the neighbouring Vaal River basin, forwater-cooling the power plants.

Another interesting characteristic of theupper basin is that the landscape is dottedwith natural pans and farm dams. Users

Planning and Management in the Olifants Basin 147

Fig. 9.2. Main hydrological regions.



in the Middleveld point out that theyexperience prolonged effects of droughts asa result, since these pans and dams in theupper basin have to fill up before significantrunoff reaches the Middleveld.

9.1.2.2 Middleveld region

The Middle Olifants stretches for about300 km from below Loskop dam to thedramatic drop down the escarpment nearHoedspruit. This escarpment area is famousfor its natural beauty and exciting riverrafting when the river flows full. However,this beauty and pleasure is in stark contrastto the plight of the 1 million peopleliving in the old ‘homeland’ area in theMiddleveld. This is one of the poorestregions in the country, where 75% of thepopulation cannot read or write, and infantmortality, hunger and crime are rife.

The area directly below Loskop dam isquite different from the barren situation inthe old homeland. Loskop dam was builtby poor white labour as a public worksprogramme after the Second World War.Below the dam are 28,800 ha of intensivelyirrigated land growing a variety of crops,with a trend towards permanent high-valuecrops including large citrus plantations andexport table grape production under hailnetting.

9.1.2.3 Mountain region

There is intensive irrigation in Steelpoortsub-basin and extensive mining activity inthe valley. Expansion of platinum miningnear the Steelpoort–Olifants confluencemay result in the construction of anothermajor dam on the Olifants River. Thisregion is a heavy contributor to the basinwater supply because of its high rainfalland steep slopes.

9.1.2.4. Lowveld region

Except for the upper part of this area (Blyderiver sub-basin), there is little irrigationalong the Olifants River below the escarp-ment, possibly because of poor soils. Thisregion is characterized by game farms and

industrial activity concentrated at the townof Phalaborwa, on the border of the famousKruger National Park conservation area.The minimum dry season flow and theimpact of the industrial effluent on thequality of the water entering the KrugerPark are of major concern to Park officialsand conservationists.

9.2 Basin Hydrology

9.2.1 Supply

Basin water resources either originate fromprecipitation, which is transformed intorunoff, evaporates or flows in the ground, orare imported from neighbouring basins.

9.2.1.1 Rainfall

Rainfall has an average value of 630 mm/year in the basin, and does not vary greatlyat the level of the water management region(Table 9.1). At a smaller scale, however, therange is significant, varying from 500 to1000 mm with values of up to 2000 mmalong the Drakensberg escarpment.

There is a distinct rainy season betweenOctober to April, with the heaviest raingenerally occurring in December andJanuary, but the main characteristic ofthe precipitation is its high degree ofunevenness, spatially, throughout the yearand inter-annually. Midsummer dry spellsare a common occurrence, often in Januarywhen dryland crops are at their mostvulnerable.


Water managementregion

Mean annualprecipitation (mm)

Upper Olifants RiverUpper MiddleMountainLower MiddleLower

682621679550631

Source: BKS (2002, Appendix C, p. 4-2 [Table11]).

Table 9.1. Precipitation by region.



9.2.1.2 Surface water

The mean annual runoff (MAR) calculatedunder ‘virgin conditions’, that is, withoutthe introduction of exotic vegetation (suchas afforestation with eucalyptus species)and without the impact of human activities,would be 1992 million cubic metres (MCM)(Table 9.2). The average value of observedrunoff with these influences present isaround 1235 MCM at the mouth of theSouth African part of the basin, on theborder with Mozambique.

A recent study of surface waterresources (Midgley et al., 1994) presentsrelationships between rainfall and runoff bylocation in the basin. To generate significantrunoff, rainfall has to be sizeable. In thisregard, the mountainous regions play themost significant role. The Mountain region(Steelpoort sub-basin) as well as Blyde River(part of the Lowveld) provide about 42% oftotal runoff. It appears also that groundwateris of major importance in maintaining baseflow in the river. The wide range observedin MAR follows the high inter-annualvariability of the precipitation.

Typically, high flows occur fromDecember to February and then decline untilSeptember. At the end of this period, theOlifants River can experience periods of noflow in the Kruger Park on the border withMozambique. A major feature of the basin isalso its capacity to generate extreme flows,yielding dreadful floods especially affectingMozambique. During the last flooding

period in February 2000, the flow in theOlifants peaked at 3800 m3/s at the mouth.

9.2.1.3 Groundwater

Total groundwater recharge in the basinis estimated by the Department of WaterAffairs and Forestry (DWAF) to be 3–6% ofthe mean annual precipitation. This wouldamount to approximately 1800 MCM. Mostof this recharge occurs during periods ofheavy precipitation and it is suspected thatthe majority of water reaching the watertable does so via macro pores (cracks,fissures, etc.) in the soil rather than throughthe actual soil body (Ashton, 2000).

Groundwater is an important source ofwater supply for many small towns, villagesand small-scale farms, where it is usedfor stock watering and some irrigation,especially on the Springbok Flats in theMiddleveld region. The largest share of thecatchment’s exploitable groundwater existsin a relatively shallow weathered aquiferthat gives average yields in the vicinity of1 l/s. Areas of higher potential (5 l/s andmore) do occur in the area of the SteelpoortRiver, while roughly half of the catchment tothe west of the Drakensberg Mountains maybe classified as having moderate potential(1–3 l/s).

9.2.1.4 Storage, imports, exports andreturn flows

There are approximately 2500 dams in thebasin, including 31 major ones defined asthose storing more than 2 MCM. The totalstorage of major dams is 1100 MCM witha firm yield of 645 MCM per year. Smalland minor dams supply additional storagecapacity of 193 MCM. This storage capacityalso represents an important source of lossby evaporation, estimated for all dams to bearound 159 MCM.

Figure 9.3 shows the historical patternof water resource development in the basin.As seen, periods of rapid growth took placein the late 1930s and again between 1970and 1990. Plans for several additional damsin the basin exist, but await approval andfunding.


Watermanagementregion

Meanannualrunoff(MCM)

Range ofmean annual

runoff(MCM)

Upper Olifants RiverUpper MiddleMountainLower MiddleLowerTotal

466200397107822

1992

134–123386–538

147–76923–555

255–2351

Source: BKS (2002, Appendix C, p. 4-2 [Table11]).

Table 9.2. Runoff per sub-region.



Imports of water occur in the upper partof the basin, primarily for coal-fired powerstations. Imports total around 200 MCMannually, and are derived from the adjacentKomati, Usutu and Vaal River basins. Com-parable volumes of water are not reliablyavailable in the upper reaches of the Olifantsbasin itself and the quality of the availablewater is often unsuitable. Small amountsof water are exported from the basin tothe Limpopo provincial capital, Polokwane(5 MCM), and these exports are slated toexpand in the future due to growing demandfor domestic water supply there.

Return flows are not well estimated.BKS (2002, appendix C, Table 10) gives thefigure of 42.3 MCM, mainly generated byurban areas in the Upper and Lower partof the basin. Return flows from miningactivities are relatively small, varying from0% to 20%, due to recycling processes.However, the high pollution levels in theriver indicate that return flows that do occurare often heavily polluted. Little is knownabout return flows from irrigation, thoughthey are undoubtedly significant, and thisportion of the basin water budget requiresfurther study.

9.2.2 Demand

There are various estimates for total waterdemand, even in the same report, suggest-ing the estimates are approximate at best.

BKS (2002, Table 3.2) provides total waterdemand figures of 1125.1 MCM/year in1995 and a prediction it will rise to1356.5 MCM by 2010. These figures includeuse of imported water and can be brokendown among users as shown in Table 9.3.It should be noted the ecological and basicneeds reserve, and the obligation to Mozam-bique as the downstream riparian, are notincluded.

9.2.3 Water balance

The basin is a water-stressed catchment andimports high-quality water from neighbour-ing catchments for economically importantuses in the upper basin.

Nevertheless, in the Upper region thereis a net surplus of 218 MCM per yearcurrently, which is expected to drop to anestimated 168 MCM by 2010. In the UpperMiddle, there is already a deficit of 60 MCMin 1995, growing to 79 MCM/year by 2010.The Lower Middle region shows a smallsurplus. Taking into consideration theneed to meet ecological and downstreamrequirements, it is expected that waterresources will be fully utilized by 2010 inthe Upper, Upper Middle and Lower Middleregions.

For the Mountain region, the waterbalance shows surplus water availablefor further development. The Lower regionappears to be in the same situation, with


1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000

1200

1000

800

600

400

200

0

Years

Vol

ume

(mill

ion

cubi

c m

etre

s)

Fig. 9.3. Development of water resources in the Olifants basin (data from Midgley et al., 1994).



surplus water flows from the Blyde Riversub-basin and the escarpment. The SelatiRiver near Phalaborwa is experiencing watershortages, and transfers are planned fromthe neighbouring Letaba River basin to solvethe problem (BKS, 2002, pp. 6-1–6-2).

Thus over the medium term, water inthe basin appears to be adequate in terms ofquantity, but there are serious distributionalissues, related to both regions and usergroups, that need to be addressed. Solutionscontemplated comprise both construction ofnew storage and transfer facilities and real-location of available supplies among uses.

9.2.4 Water quality

The basin faces significant water qualityproblems, due to mining activities, indus-tries, power generation and agriculturaluse of water. The impacts of pollutionfrom these activities (high salinity, high

concentrations of metals, low pH) areprobably multiple with serious ecologicalimpacts. Of particular concern is the down-stream Kruger National Park, which is a majortourist attraction, as well as very worrisomehealth impacts, since many people rely onuntreated surface water for drinking.


9.3.1 Water policy and water law

9.3.1.1 Pre-1994

With the establishment of a Dutch settle-ment at the Cape of Good Hope in 1652,customary principles governing access towater for stock watering and domestic usewere supplemented and gradually replacedwith ideas of European origin as the CapeColony expanded. Over time, a pastiche of


Use 1995 2010 Remarks

Irrigation

Livestock

Urban(including smallerindustries)Mining

Afforestation

Reserve

Internationalrequirements

540.3

27.8

117.8

94.3

55.4

None

?

593.0

33.1

221.8

93.8

63.6

460(?)

?

107,000 ha, 95% white commercial farming9% increase projected mostly in the Upper Middle, Lower Middleand Lowveld regionsMajor crops: cotton, grapes, citrus, vegetables and grainTotal of 2.3 million large stock units by 2010. Mainly in theHighveld and watered from boreholes (70%)A 55% increase is likely to result from improving living conditionsand population expansion, especially in the former homelands

In the Highveld region, open cast coal mining uses both waterfrom the catchment and imports from the Komati, Usutu andVaal Rivers. In the Mountain region, mines are a major user ofboth surface and borehole water. The Lowveld part of the basincontains numerous mines, but most of these are gold mines andhave low water requirements, though they may have appreciablepollution potentialIndigenous forests will probably remain at 11,500 ha and permitsfor 15,000 ha of additional commercial afforestation will havebeen granted. It is envisaged that the existing 60,000 ha ofexotic afforestation will only increase by a further 7000 haDetermination underway but first figure of 460 MCM for theecological Reserve is proposed in the Proposed National WaterResource Strategy (PNWRS, August 2002)Not determined. The Olifants is the largest contributor of the flowto Massingir Dam in Mozambique

Source: BKS (2002, Appendix C, Figs 2–7).

Table 9.3. Trends in water demand in the Olifants basin.



principles with roots in Roman, Dutch andEnglish water law emerged to govern waterallocation in South Africa (Thompson et al.,2001). Initially, allocation policy employedthe public trust doctrine, which gave theState (or the Dutch East India Company) theright to control and allocate use. Later astrong riparian component was added,giving individual landowners adjacent tonatural watercourses the right to use wateron those lands, subject to the rights of othersimilar landowners. Generally, the riparianprinciple was employed by white commer-cial large-scale farmers to secure water forirrigation. Appropriative rights to abstractwater and use it elsewhere were also grantedon a case-by-case basis. Water was classedas private or public, with water arising on alandowner’s property entirely his to use.Flows in public streams were apportionedinto ‘normal’ and ‘surplus’ flows, with dif-ferent rules applying to each. These rulesdeveloped as the need arose from intensify-ing economic activity to regulate use amongthe commercial farms, mines and urbanconcentrations of the minority Europeanpopulation. In sum, during the apartheidera the white government, commercialfarmers, mining firms and other interestsestablished a complex system of formalizedlaws and institutions on a largely ad hocbasis, which excluded a large majority ofthe population (van Koppen et al., 2003).

9.3.1.2 Post-1994

The introduction of democratic governmentin 1994 demanded changes in the skeweddistribution of access to water for both basicand productive purposes. It offered a rareopportunity for comprehensive review ofall water-related legislation to develop amodern water policy appropriate to a water-scarce region. This was done in a verydeliberate and conscientious way underthe farsighted Minister of Water Affairs andForestry, Professor Kader Asmal (Box 9.1),and resulted in the new National Water Actof 1998 that is widely regarded as a modelpiece of legislation.

Two fundamental provisions in the1996 constitutional bill of rights give ‘basic

rights’ status to access to water to supportlife and for personal hygiene and a healthyenvironment, both now and for future gener-ations. These constitutional rights under-gird two fundamental pieces of new waterlegislation, the National Water Act (NWA)and the Water Services Act (WSA). The newwater policy, as embodied in the WhitePaper on National Water Policy for SouthAfrica, marks a radical departure from theprevious legal regime. Basic principles areshown in Box 9.2.

South Africa’s new water law reservescommitted outflows to neighbouring coun-tries. South Africa is a signatory to a SADCProtocol on Shared Watercourse Systemsbased on the Helsinki Rules, the Dublinprinciples and Agenda 21 of the Rio EarthSummit. The SADC protocol strives tomaintain a balance between developmentalneeds of the member countries and theneed for environmental protection andconservation. Signatories to the protocolcommit themselves to seeking peacefulsolutions to disputes. The protocol providesfor the formation of basin-wide commis-sions, such as those for the Okavango andthe Limpopo.

The new water law principles declareall water in the hydrological cycle to bean indivisible national asset, held in trustby national government (Box 9.2). Highlyfragmented resource management acrossapartheid borders is being replaced by acatchment-based management system thatfollows natural, rather than internal politi-cal boundaries. Privatization of services pro-vision is possible, but strictly governed bynational legislation, administered throughnew local government structures.


Box 9.1. Water policy timeline.

May 1994 Review of all water relatedlegislation initiated

Nov 1994 Water Supply white paper publishedNov 1996 Fundamental Principles paper

approved by cabinetApril 1997 Water Policy white paper publishedSept 1997 Water Services Act promulgatedAug 1998 National Water Act promulgated



Since promulgation of the NWA, thewater users have by-and-large remained thesame, but there is a much greater emphasison basic service provision to historicallymarginalized communities. The Governmenthas embarked on a major investment pro-gramme to provide basic water supplies tothe poor. New legislation also enables theestablishment of water users’ associations(WUAs) for user-management of shared localinfrastructure – an option previously open tocommercial farmers only. Furthermore, theNWA establishes mechanisms for public par-ticipation that enable reallocation of water toredress past racial and gender discrimination.

9.3.1 Actors

9.3.2.1 Government agencies

DWAF During the apartheid era, thenational DWAF was responsible for water

resources management in the white areas ofthe Republic, while similar departments inthe former homelands and self-governingterritories concentrated largely on basicservices provision. This fragmentationcreated severe problems in water resourcesmanagement. In the white areas, waterservices were rendered by well-resourcedmunicipalities while few resources wereavailable to homeland water departments,leading to huge disparities in service levelsbetween the white and homeland areas.

The 1996 constitution establishedwater as a ‘national competency’, vestingresponsibility for water resources and ser-vices in the DWAF. The character of theDepartment was fundamentally transformedin terms of its functions and staff profile(i.e. race, gender, discipline) to respondto its new mandate. DWAF embarked onan aggressive programme to speed upbasic water and sanitation service deliveryto the marginalized areas and changedthe resources management paradigmfrom a supply-driven to demand-drivenapproach.

The NWA requires DWAF to develop aNational Water Resources Strategy (NWRS),to create CMAs in the 19 Water ManagementAreas and to delegate catchment-basedwater resources management functions tothese CMAs. Each CMA is to develop its owncatchment management strategy through aparticipative process with water users in itsarea.

Through the WSA, DWAF designateslocal government structures as water ser-vices authorities (WSAs) and local govern-ment, water boards or private sector entitiesas Water Services Providers (WSPs).

NATIONAL DEPARTMENT OF AGRICULTURE AND

PROVINCIAL DEPARTMENTS OF AGRICULTURE

After 1994, nine provinces were createdintegrating the former four white provinces,six ‘self-governing’ territories and four‘independent’ states. In the 1996 constitu-tion, agriculture is a ‘concurrent function’,meaning it is a shared responsibilitybetween the new National Department ofAgriculture (NDA) and the nine ProvincialDepartments of Agriculture (PDAs).


Box 9.2. Principles of the National WaterPolicy.

• All water in the water cycle is a part of anindivisible national asset

• This asset is held in trust for society by thenational government

• Water to meet basic human needs, to sustainthe environment, and to meet legitimate needsof neighbouring countries is reserved

• All other water uses must be beneficial in thepublic interest

• The riparian system of allocation is abolished• Allocations will no longer be permanent but

for a reasonable period, e.g. 40 years, and canbe traded

• Water resources will be managed on acatchment basis by specialized bodies

• All water use in the water cycle is subject toone or more charges intended to reflect the fullfinancial costs of protecting and managing thewater resource

• Water-based waste disposal is subject toappropriate charges

• Charges for water for basic human needs andfor small-scale productive purposes may bewaived for disadvantaged groups

Source: White Paper on a National Water Policyfor South Africa (1997)



PDAs still retain significant responsibil-ity for irrigation infrastructure created previ-ously. In the Limpopo Province, for exam-ple, the PDA has inherited over a hundredirrigation schemes of various types. Manyare no longer functioning because the subsi-dies previously provided have been with-drawn and no new farmer-based structuresare yet in place. The Province has launched avery ambitious ‘revitalization’ programme toorganize farmers into WUAs and help themachieve the capacity to take over and managetheir irrigation schemes profitably.

The former homeland departments ofagriculture had played a significant rolein creating irrigation infrastructure (oftenthrough parastatal corporations) and someof the PDAs have inherited responsibilityfor the operation, maintenance and refur-bishment of these schemes. National policycurrently promotes Irrigation ManagementTransfer (IMT) to the users, though progressin implementing this policy to date has beenslow.

The homeland governments’ role inwater allocation for agriculture resultedfrom the link between land and water con-tained in the Water Act of 1956. Extensionofficers, together with tribal authorities,controlled the issuing and repeal of ‘Permis-sion to Occupy’ (PTO) certificates to small-holder farmers, which conveyed rights toboth land and water. Through the repeal ofthe riparian principle under the new NWA,the direct link between land and waterallocation has been broken.

OTHER GOVERNMENT DEPARTMENTS Before1994, several departments played strong rolesin water resources management, while morecontrol now vests in DWAF through the newNWA. However, the Department of Mineraland Energy Affairs and the Department ofEnvironmental Affairs and Tourism stillplay important roles in the protection andrestoration of water resources through theirrequirements for Environmental Planningand Management Reports and for Environ-mental Impact Assessments, respectively.

FORMER HOMELAND GOVERNMENTS In pursuitof the goal of separate institutions for each

racial and ethnic group, apartheid-era gov-ernments attempted to divide the countrygradually into various independent andautonomous racially homogeneous States.National States had a large degree of inde-pendence, while self-governing territorieswere to be gradually led to independentState status. Virtually no country other thanSouth Africa recognized this system. TheOlifants basin encompassed small portionsof the national State of Bophuthatswana andincluded sections of three self-governingterritories – Lebowa, KwaNdebele andGazankulu (Thompson et al., 2001). Therewere various departments in these home-land governments that dealt with water-related matters, but because only certainfunctions were transferred to the homelandadministrations, others being retained bythe Republic of South Africa, there wasconsiderable confusion over some basicfunctions, such as the issuance of wateruse permits. Homelands were abolishedin 1996 with the adoption of the nationalconstitution.

SOUTH AFRICAN DEVELOPMENT TRUST AND OTHER

PARASTATALS The Development Trust, cre-ated originally in 1936 as the South AfricanNative Trust, was responsible, among otherthings, for supplying basic infrastructure inthe homelands, including water-relatedinfrastructure. It did this directly and actingthrough the various homeland DevelopmentCorporations which it was responsible forestablishing, financing and maintaining. Itwas dissolved with the implementation ofthe new constitution in 1996 and its water-related functions transferred to DWAF.

TRADITIONAL LEADERS Historic tribal leader-ship in the Olifants basin was seriouslyundermined during the apartheid period.For instance, in an area that would tradition-ally be the domain of four tribal chiefs, theapartheid government ordained more than100 new chiefs, mostly drawn from amonglower ranking leaders who were willing tocooperate with the apartheid regime. Thisfragmentation, together with an influx ofpeople through forced removals from whiteareas, created serious conflicts over tribal




boundaries. These new ‘traditional leaders’served in homeland government structuresand areas of resistance to the regime wereneglected in development initiatives. Theyhad considerable authority over access toresources such as water and land (vanKoppen et al., 2003).

In the new set-up, tribal leadership isagain at the heart of deep conflict, this timeover shared local governance with new dem-ocratically elected structures. In the deeprural areas, members of the older generation,especially, swear allegiance to their chiefsand believe that the spiralling crime rate is atleast in part related to the disrespect for thetraditional systems and leadership. In turn,the new democratic structures are battling toestablish themselves with limited resourcesand virtually no experience or systems forservice delivery. In an attempt at integration,the tribal heads are automatically membersof the new local councils, while other mem-bers are elected. As one community in theOlifants basin recently pointed out to theirchief in a public meeting: ‘these youngsterswill come and go depending on electionresults, but you will stay, therefore we needyou to take the lead for the sake of ourdevelopment.’

LOCAL GOVERNMENT/MUNICIPALITIES Localgovernment is responsible for water servicesdelivery, while the new CMAs will be res-ponsible for water resources management.Local government structures, in turn, will berepresented in the Olifants basin CMA.

Before 1994, ‘municipalities’ wereestablished for service delivery in all citiesand white towns of 200 houses or more.Their functions were limited to white urbanconcentrations, while rural areas were theresponsibility of regional services councils.Separate structures were responsible forservice delivery to blacks in the homelands.

After 1994, these boundaries were abol-ished and a period of restructuring followed.Most recently, a system has been adoptedthrough which the entire country is dividedinto three types of local government struc-tures: metropoles (class A) and shoulder-to-shoulder local municipalities (B), whichare grouped together into 42 District

Municipalities (C). In effect, the formermunicipalities have seen their boundariesredrawn to include the rural hinterland.

WATER COURT/TRIBUNAL Under the 1956Water Act, a specialized Water Court ruledon disputes between rights holders. In thenew legal framework, water use entitle-ments will be issued by the CMAs accordingto a water allocation plan that will formpart of the catchment management strategy.People can appeal to the Minister on issuesof unjust administrative process, followedby appeal to a newly established WaterTribunal.

9.3.2.2 Boards and agencies

CMAS The NWA mandates the Ministerof Water Affairs and Forestry to developcatchment-based water resources manage-ment through the creation of CMAs in the19 Water Management Areas. Each CMAwill develop and review, on a 5-yearly basis,its own catchment management strategythrough a participative process with waterusers in its area. As the CMA develops thenecessary capacity, DWAF will progres-sively delegate functions to it. This is nottrue devolution of power, but rather a dele-gation, since the Minister has the right torevoke authority if he is convinced that aCMA is falling short of its commitments.

The CMA board must be representativeof the range of water users in the catchment.The Minister appoints board membersthrough the following process. First (s)heappoints an advisory council to identifywater user groups and relevant institutionsin the basin. The Minister invites theseinstitutions or organizations to nominatecandidates for appointment to the CMAboard. If (s)he is not satisfied with thesuitability of a nominee, (s)he can ask foran alternative. The Minister then appointsthe nominees and if necessary, ‘tops up’ theboard with additional members to achieveracial and gender balance.

The CMA board appoints a Chief Execu-tive Officer to develop and manage the CMAoffice or operational structure. This officemay develop its own capacity to perform the




full range of water resources functions, orit may contract out some or all of thesefunctions. The NWA authorizes the CMAto collect a catchment management chargefrom water users to cover its costs.

As of early 2003, no CMA has yet beenestablished, though proposals for several,including for the Olifants basin (BKS, 2002),are nearly ready to be submitted to the Min-ister. In the case of the Olifants, there hasbeen a long period of attempting to consultwith stakeholders in the basin as a basisfor preparing a proposal. While the well-organized commercial sectors have partici-pated actively, it has been relatively difficultto find effective means to consult the poorpeople scattered throughout the basin,particularly those residing in the formerhomeland areas (Wester et al., 2003). Suchconsultation will be a continuous process,which will hopefully be done more effec-tively through the CMA when it is estab-lished. The proposed name of the CMA is theLepelle Catchment Management Agency.

WATER BOARDS AND OTHER WSPS Waterboards were developed to provide servicesand perform water management functionsbeyond the reach of the old municipalities.Increasingly, they engaged in regional watersupply schemes primarily for industrial anddomestic use. Water boards are now regu-lated, together with other WSPs, under theWSA where their role in water resourcesmanagement is considered incidental ratherthan intentional.

IRRIGATION BOARDS AND WUAS The 1956Water Act provided for the establishmentof membership-based organizations calledirrigation boards, through which groups offarmers could join forces to develop infra-structure and jointly manage their watersupply – essentially a type of WUA. Theseirrigation boards were eligible for a one-third capital subsidy on shared water supplyinfrastructure, but membership was legallyrestricted to people who had title to the landreceiving services from the irrigation board.This effectively excluded black membershipin white irrigation boards, since blacks wereexcluded from land ownership in ‘white

areas’, but there was also no similar institu-tion available to groups of black farmerswith similar needs. Irrigation services in thehomelands were supplied by governmentor parastatal corporations and participatingfarmers were passive recipients with nopower to demand adequate services.

The NWA of 1998 calls for the trans-formation of existing irrigation boards intoWUAs and removes title deed as a member-ship requirement. It thus also enables theestablishment of WUAs on communallyowned tribal or State land. Indeed, the NWAauthorizes the issuance of water use entitle-ments – and by extension membership inWUAs – to water users rather than landown-ers. This is of particular importance in tribalareas where Permission To Occupy (PTO)certificates have traditionally been issued tomen, but where women are predominantlythe users of the land and water.

As of early 2003, only a few irrigationboards nationally have been officially trans-formed into WUAs, and only one formalWUA has been established on a small-scaleirrigation scheme in the Olifants basin.

9.3.2.3 Farmers

COMMERCIAL IRRIGATORS Irrigation farmersaccount for more than half of national wateruse and a quarter of the agricultural outputin South Africa, irrigating in all 1.3 millionha. The commercial irrigation sector pro-duces a wide range of crops for export andlocal use. It is supported by a sophisticatedirrigation equipment manufacturing indus-try – indeed, micro irrigation technologyoriginated in South Africa before it wasdeveloped into a major industry in Israel andthe USA. Backward and forward linkages toinput suppliers, service providers, process-ing, value-adding and export industries arenot accurately quantified but are significant,both in terms of the national economy andemployment.

Commercial irrigators fall roughly intothree categories in terms of their access towater. About one-third of the irrigated areafalls under irrigation boards or WUAs as dis-cussed above, while another third is servedfrom government water schemes, most of




which are at some stage of handover to usermanagement. The balance derived theirwater rights from the riparian principleand withdrew water directly from rivers andstreams. The latter group had no need toparticipate in user management groups,a situation likely to change with theimplementation of catchment-based waterresources management.

Commercial irrigation in the OlifantsRiver basin is almost a microcosm of thenational situation, with farmers in govern-ment schemes and irrigation board schemesand unassociated farmers, as discussedabove, growing a wide variety of crops. Basinfarmers also represent a wide cross-sectionof attitudes to the new political reality, fromthe most conservative to the most liberal.Commercial farmers are well organized, butoften overwhelmed and uncertain aboutthe implications of the NWA. Many are alsoconcerned about the affordability to waterusers of the establishment and operation of aCMA.

EMERGING IRRIGATORS AND OTHER VILLAGE-LEVEL

ECONOMIC ACTORS Some of the poorestrural areas in the country are found in theOlifants basin. Smallholder irrigation hashad a troubled history of imposed develop-ment and State-managed irrigation infra-structure in a context of serious mistrustbetween the ‘benefactor’ and ‘beneficiaries’.The 1600-ha Arabie–Olifants and othersmallholder irrigation schemes in the basinhave suffered in varying degree from thedependency created by this approach. Onlya small portion of the irrigated area in theOlifants basin is occupied by smallholders,but a relatively large number of familiesderived at least part of their livelihoods fromirrigation, either on the formal governmentschemes or on much smaller communal veg-etable gardens or homestead food gardens.Not surprisingly, recent withdrawal of gov-ernment management and subsidies for theschemes resulted in the collapse of most ofthe production here, with serious effects onfood security in the area. However, amidstthe collapse of the large systems, communalgardens and informal food producers havelargely continued their activities as before.

In these poverty-stricken areas, theopportunities for expansion of both foodproduction and income generation fromagriculture and other small-scale village-based economic activities (such as brickmaking, ice making, poultry rearing, foodprocessing) are largely dependent on theavailability of water.

While their need for access to water isdesperate, this sector is probably the mostdisorganized and under-represented of allwater user sectors in the Olifants andmany other South African river basins. Theconcept of small-scale water users forums(SWUFs) with representation in representa-tive, decision-making and operational struc-tures within the CMA framework has beenproposed in the draft CMA proposal toaddress this problem (BKS, 2002).

9.3.2.4 Industrial firms

MINING COMPANIES Coal mining dominatesin the upper or Highveld region of theOlifants basin, while the mining andprocessing of platinum and related mineralsis a growth sector in the Middleveld. Copperand phosphates mining and associatedindustries are the most important economicactors in the Lowveld area.

POWER GENERATOR (ESKOM) The Olifantsbasin is of national strategic importancesince more than half of the country’selectricity is generated in coal-fired powerstations in the coal-rich Highveld region ofthis basin. The electricity provider, ESKOM,depends on the transfer of high-qualitywater from other basins for its coolingtowers.

FOREST PRODUCTS FIRMS Two national for-estry and forest products companies areactive in the Olifants basin. Although theforestry activity in the Olifants is limitedcompared to other basins, the industry hasbeen well represented in public consul-tations. Through the NWA, commercialforestry has been declared a StreamflowReduction Activity (SFRA), which is viewedas a water use under the Act and thereforesubject to all provisions in the Act that




applies to other water uses, includingallocation rules, water charges and measuresfor the protection of water resources.

TOURISM FIRMS Several tourism companiesoperate river rafting and other recreationalattractions, especially in the escarpmentarea. River rafting and fishing is entirelydependent on the flow in the river andtherefore on dam operations, and high waterquality is also important. Because of thesmall size of this sector, however, it isunlikely that tourism firms will significantlyinfluence water resources managementdecisions Like many other users, though,these firms stand to benefit from improvedand timely information on water releasesfrom the major dams in the system.

9.3.2.5 Environmental advocacy groups

The Olifants is one of six significant riverscrossing the Kruger National Park. Pressureon mining companies and other industriesbecause of their impact on Olifants waterquality as it enters the Park, resulted in thefounding of the Olifants River Forum (ORF)in the early 1990s. In the early stages of con-sultation to form a CMA, there was a vagueexpectation that the ORF might become theCMA, but it quickly became apparent thatits history and limited membership baseexcluded this possibility. ORF now sees itsrole as that of an environmental watchdogand remains very active.

9.3.2.6 Development banks

Water infrastructure in the past has beenfunded almost exclusively by government.In the case of the former homelands, fundsfor irrigation development were adminis-tered through the Development Bank ofSouthern Africa. The new economic policypromotes private financing and requirestransparency and sustainable financing, notonly of infrastructure development, but alsoof its operation, maintenance and futurereplacement. It is therefore expected thatcommercial financing institutions willincreasingly play a role in the constructionand rehabilitation of infrastructure.

9.4 Essential Functions for River BasinManagement

Chapter 1 identified 11 essential functionsof basin management. A slightly modifiedlisting of these functions is used in Tables9.4 and 9.5, crossed with the key actorsidentified in the previous section. Thesefunctions are replicated, as appropriate,across four broad categories – surface water,groundwater, wastewater disposal and dif-fuse returns, or agricultural return flows.Cells are marked to indicate an actor’s levelof activity in a particular functional area.Information is drawn from interviews,printed materials and Internet postings. Anumber of interesting points emerge from acomparison of Table 9.4, which reflects thesituation during apartheid (pre-1994) andTable 9.5, which conjectures about thepossible scenario 5 years after the expectedestablishment of a CMA for the Olifants.This projection is made on the basis ofa group discussion with key officials inDWAF in July 2001.

A few observations emerge from anexamination of Tables 9.4 and 9.5.

• Prior to 1994, DWAF was the pre-eminent organization in managingsurface water and wastewater, playinga major role in planning, allocating,supplying and protecting surface waterresources.

• Roles of many key actors are changing,none so dramatically as that of DWAF.In combination with the CMAs, most ofthe same functions previously coveredby DWAF will be covered in future.However, DWAF will delegate con-siderable authority to the CMAs andwill assume new responsibilities inenvironmental protection.

• The important flood protection func-tion, previously handled by DWAFand provincial and local governmentauthorities, appears to have no primarylocus of responsibility in the future.The CMA role is uncertain.

• Many uncertainties remain regardingwho will do what in the future. Inparticular, the roles to be played by





Sur

face

wat

erG

roun

dwat

erW

aste

wat

erD

iffus

ere

turn

s

Key

acto

rsD

WA

Fr

rr

rr

rr

rq

qq

rr

rN

DA

/PD

As

rr

Oth

ergo

vern

men

tdep

artm

ents

qr

rW

ater

trib

unal

rq

SA

Dev

elop

men

tTru

stan

dot

her

para

stat

als

rr

q

Hom

elan

dgo

vern

men

tsq

qr

rq

qr

rq

qq

qq

Mun

icip

aliti

esr

rr

rr

rr

rr

rr

Tra

ditio

nall

eade

rsq

qC

MA

Wat

erbo

ards

rr

rW

ater

serv

ice

prov

ider

sIr

rigat

ion

boar

dsr

rr

qq

WU

As

(um

brel

la)

WU

As

(gra

ssro

ots)

cont

inue

d

Tab

le9.

4.P

astb

asin

man

agem

entf

unct

ions

and

key

acto

rs–

pre-

1994

.

Protectagainstflooding

Protectecology

Ensurequality

Monitorquality

Maintainfacilities

Distributewater

Constructfacilities

Allocatewater

Plan(basinlevel)

Plan(basin-level)

Allocatewater

Constructfacilities

Withdraw/distributewater

Maintainfacilities

Monitorquality

Ensurequality

Constructfacilities

Operatefacilities

Maintainfacilities

Monitorquality

Enforcequality

Authorizedischarges

Maintainfacilities

Monitorquality

Enforcequality

Authorizedischarges

Constructfacilities




Sur

face

wat

erG

roun

dwat

erW

aste

wat

erD

iffus

ere

turn

s

Key

acto

rsE

mer

ging

irrig

ator

sC

omm

erci

alirr

igat

ors

qq

rr

qq

Oth

ersm

all-s

cale

econ

omic

acto

rsN

on-ir

rigat

ing

farm

ers

qq

qq

Min

ing

firm

sq

qq

rq

qq

qP

ower

gene

rato

r(E

SK

OM

)q

qq

qT

ouris

mfir

ms

For

estp

rodu

cts

firm

sO

ther

man

ufac

ture

rsE

nviro

nmen

tala

dvoc

acy

grou

psq

Dev

elop

men

tban

ksr

r

r,I

ndic

ates

activ

ity;q

,ind

icat

eslim

ited

activ

ity.

DW

AF

,Dep

artm

ento

fWat

erA

ffairs

and

For

estr

y;N

DA

,Nat

iona

lDep

artm

ento

fAgr

icul

ture

;PD

As,

Pro

vinc

ialD

epar

tmen

tsof

Agr

icul

ture

;CM

A,C

atch

men

tM

anag

emen

tAge

ncy;

WU

A,W

ater

Use

rs’A

ssoc

iatio

n.

Tab

le9.

4.C

ontin

ued.


Protectecology

Ensurequality

Monitorquality

Maintainfacilities

Distributewater

Constructfacilities

Allocatewater

Plan(basinlevel)

Plan(basin-level)

Allocatewater

Constructfacilities


Maintainfacilities

Monitorquality

Ensurequality

Constructfacilities

Operatefacilities

Maintainfacilities

Monitorquality

Enforcequality

Authorizedischarges

Maintainfacilities

Monitorquality

Enforcequality

Authorizedischarges

Constructfacilities




Sur

face

wat

erG

roun

dwat

erW

aste

wat

erD

iffus

ere

turn

s

Key

acto

rsD

WA

Fq

qr

??

qr

qr

qr

ND

A/P

DA

sq

qq

qO

ther

gove

rnm

entd

epar

tmen

tsr

rq

Wat

ertr

ibun

alr

qr

rq

SA

Dev

elop

men

tTru

stan

dot

her

para

stat

als

Hom

elan

dgo

vern

men

tsM

unic

ipal

ities

rr

rq

rr

rr

rr

rr

Tra

ditio

nall

eade

rs?

?C

MA

rr

??

rr

?r

rr

rr

rr

rW

ater

boar

dsr

rr

rr

rr

rW

ater

serv

ice

prov

ider

sr

rr

qr

rr

rIr

rigat

ion

boar

dsW

UA

s(u

mbr

ella

)q

qr

rq

rW

UA

s(g

rass

root

s)r

rr

?r

rr

cont

inue

d

Tab

le9.

5.P

roje

cted

basi

nm

anag

emen

tfun

ctio

nsan

dke

yac

tors

–5

year

saf

ter

CM

Aes

tabl

ishm

ent.


Protectecology

Ensurequality

Monitorquality

Maintainfacilities

Distributewater

Constructfacilities

Allocatewater

Plan(basinlevel)

Plan(basin-level)

Allocatewater

Constructfacilities


Maintainfacilities

Monitorquality

Ensurequality

Constructfacilities

Operatefacilities

Maintainfacilities

Monitorquality

Enforcequality

Authorizedischarges

Maintainfacilities

Monitorquality

Enforcequality

Authorizedischarges

Constructfacilities




Sur

face

wat

erG

roun

dwat

erW

aste

wat

erD

iffus

ere

turn

s

Key

acto

rsE

mer

ging

irrig

ator

sC

omm

erci

alirr

igat

ors

qq

rr

rO

ther

smal

l-sca

leec

onom

icac

tors

Non

-irrig

atin

gfa

rmer

sq

qq

qq

Min

ing

firm

sq

qq

rr

rr

rr

Pow

erge

nera

tor

(ES

KO

M)

rr

rr

Tou

rism

firm

sF

ores

tpro

duct

sfir

ms

Oth

erm

anuf

actu

rers

Env

ironm

enta

ladv

ocac

ygr

oups

rD

evel

opm

entb

anks

rr

r,I

ndic

ates

activ

ity;q

,ind

icat

eslim

ited

activ

ity.

DW

AF

,Dep

artm

ento

fWat

erA

ffairs

and

For

estr

y;N

DA

,Nat

iona

lDep

artm

ento

fAgr

icul

ture

;PD

As,

Pro

vinc

ialD

epar

tmen

tsof

Agr

icul

ture

;CM

A,C

atch

men

tM

anag

emen

tAge

ncy;

WU

A,W

ater

Use

rs’A

ssoc

iatio

n.

Tab

le9.

5.C

ontin

ued.


Protectecology

Ensurequality

Monitorquality

Maintainfacilities

Distributewater

Constructfacilities

Allocatewater

Plan(basinlevel)

Plan(basin-level)

Allocatewater

Constructfacilities


Maintainfacilities

Monitorquality

Ensurequality

Constructfacilities

Operatefacilities

Maintainfacilities

Monitorquality

Enforcequality

Authorizedischarges

Maintainfacilities

Monitorquality

Enforcequality

Authorizedischarges

Constructfacilities



DWAF and the CMA in operatingand maintaining facilities that deliversurface water are undefined.

• Some institutions from the previousregime have now disappeared (suchas the South African DevelopmentTrust and homeland governments)and new ones are still in the process offormation. The first CMAs should beestablished in 2003; WUAs began beingestablished in 2002 as DWAF workedout procedures and principles.

• Prior to 1994, planning was largelya technocratic and political processwith little participation by otherplayers. Planning under the CMA isexpected to be much more inclusiveand participatory.

• Prior to 1994, there was virtually noregulation of groundwater allocationand use, or of groundwater quality.In future, both the CMA and the watertribunal are expected to be involvedin water allocation, while DWAF,the CMA and other government depart-ments are expected to ensure quality.

• The roles envisioned for the CMA spana wide range, including planning andallocation, wastewater regulation, andmonitoring and enforcement of waterquality standards. It remains to be seenwhether all of these functions, particu-larly the enforcement ones, are appro-priate to such a pluralistic institution.

• Virtually no formal activity in regulat-ing the quality of agricultural returnflows is envisioned in the near future.This reflects both the existence ofother higher priority concerns and thedifficulty of addressing these issues.

The following sections provide additionalobservations based on the consultationwith the stakeholders around the essentialfunctions framework.

9.4.1 Water resources management

Before 1994, DWAF dominated the manage-ment and regulation of surface waterresources and water quality, but with

limited jurisdiction in the homeland areas.There was limited activity in the planningand allocation or control of groundwaterand diffuse returns.

The intention is that the Ministerof Water Affairs and Forestry will progres-sively delegate these functions to CMAs,as the CMAs achieve adequate capacity toassume such functions. DWAF will developand maintain a national resource classifica-tion system against which the CMA willmanage its resources in accordance withagreed resource quality objectives. In effect,DWAF will be transformed from a planningand implementation agency to one that doesnational level planning, provides technicalsupport to CMAs and local water serviceagencies, and regulates all of these. Thestructure and personnel of DWAF are beingmodified to meet these new responsibilities.

CMAs are intended to be more partici-patory and sensitive vehicles for managingbasin water resources. They incorporate amix of public and private characteristics butare still a work in progress and the exact mixof characteristics that will emerge remainsuncertain. On the one hand, CMAs areintended to be self-financing and semi-autonomous. On the other, they are expectedto represent the broader public interest inprotecting water resources and the naturalenvironment. How compatible these func-tions will prove to be, and whether they willprove affordable for a self-financing CMAremains to be seen.

9.4.2 Water services provision – domesticand industrial use

Before 1994, DWAF played a dominant rolein bulk water supply, but was generally notinvolved in service provision to end users.For domestic and industrial use, this wasthe domain of the homeland governments,water boards and urban municipalities.

Several changes in the South Africangovernment system affect the way inwhich water services will be providedin future. Most importantly, the new localgovernment operates in newly demarcated




shoulder-to-shoulder municipalities thaterase the fragmentation along racial bound-aries and consolidate jurisdiction overfunctions formerly held by the home-land structures, tribal authorities, urban-focused municipal areas and regionalservices councils. These new municipalitiesserve as the new WSAs under the WSAof 1997, and can act as WSPs themselvesor contract this function out to the privatesector, water boards or public sector WSPs.

9.4.3 Water services provision –agricultural use

The dominant water users in agricultureare the white commercial irrigation farmers,accounting for more than half of nationalwater use. Before 1994, roughly one-thirdof the 1.3 million irrigated ha was servicedthrough irrigation boards. Irrigation boardsacted as a type of WUA, providing servicesto its members, occasionally also supply-ing water to the municipality of thenearby rural town. However, only proper-ties with full title deed were eligible formembership, which excluded blacks, whowere prohibited from owning titled land inwhite areas.

All of this is now changing. Land andwater rights have been separated by law,and there are some black water users thatmust be accommodated in the transforma-tion of irrigation boards to WUAs. In fact,most applications for transforming boardsto WUAs in 2002 were rejected by DWAFbecause they were not sufficiently inclusive(Faysse, 2003). Small-scale irrigation farm-ers are getting organized into WUAs, albeitnot rapidly. In stressed river basins, thelicensing process could very well lead toreductions in water allocations to commer-cial farmers, though this remains to be seen.

The tables reflect a new idea that hasrecently emerged with regard to WUAs: theconcept of distinguishing between ‘grass-roots’ and ‘umbrella’ WUAs. Initial inter-pretations of the NWA held that wherethere were both commercial and small-scaleemerging farmers in one area, they must all

be members of one WUA. However, based onexperiences in the Olifants, it was realizedthat it may often be useful to enable smallergroups of water users sharing a sub-schemeto organize around their water infrastruc-ture, and the various smaller WUAs couldbe federated into a larger WUA to addressmutual problems.

9.5 Enabling Conditions

The essential functions and actors’ rolesdepicted in Tables 9.4 and 9.5 provide astatic view of responsibilities. Additionalattributes of well-functioning basin gover-nance systems relate to its dynamics. Chap-ter 1 suggested a list of attributes termedenabling conditions to analyse these. Herewe discuss a slightly modified version ofthis list with respect to the Olifants basin(see also van Koppen et al., 2003; Westeret al., 2003).

9.5.1 Political attributes

South Africa’s negotiated transition fromwhite minority rule to democracy is legend-ary and is embodied in its 1996 Constitu-tion. This early success firmly establishednegotiation as the preferred modus operandiand representativeness, legitimacy, equityand sustainability became requirements ofthe new political environment. The intro-duction of democratic governance gaveSouth Africa’s post-1994 government anunprecedented mandate for change. Thismandate for major change broughtuncertainty that dislodged vested interestssufficiently to enable fundamental policyreview. In reviewing the water law, Profes-sor Kader Asmal’s strong political leader-ship resulted in this window of opportunitybeing seized to introduce a new systemof inclusive representation and balancedpower in water resources decision making.

The new system has to create mecha-nisms through which water users acrossthe board can make themselves heardand understood, enabling gradual and




systematic redress of racial and genderinequities, whilst ensuring a secure base foreconomic growth. This is a major challengein policy and implementation. Lack ofaccess – through both lack of rights and lackof infrastructure – is a priority issue of therural poor in the Olifants basin. Their ade-quate representation is hard to achieve: mostlive in remote areas, excluded from partici-pation through the cost of transportationand a lack of organization. Industrialists andcommercial irrigation farmers, by contrast,are well organized and have larger financialresources, better enabling them to partici-pate in consultation processes. The pre-eminent challenge to the CMA process isto find ways to ensure representative partici-pation of major stakeholding groups and lev-elling the playing field to allow the variousstakeholders to interact with equivalentknowledge, authority and legitimacy.

The environment is nominally pro-tected through the strong provisions in theNWA for priority allocations to maintainthe integrity of the resource base. However,the Environmental Affairs Ministry and theenvironmental non-governmental commu-nity may lack the organized political clout toensure that such provisions are adequatelyimplemented.

The proposed CMAs ignore internalpolitical boundaries, playing to both thestrengths and weaknesses of the newnational political scheme. The SouthAfrican situation is in contrast to countrieslike Turkey, where a long history of strongand stable local government provided idealvehicles for the establishment of their newWUAs. In South Africa, local governmentstructures were deeply problematic: well-resourced white municipalities and regionalcouncils historically excluded black repre-sentation, while structures in the formerhomeland areas were weak and now have tograpple with major ideological differencesbetween traditional tribal leadership and thenew democratically elected representation.Apart from the advantages from a naturalresource perspective of managing along nat-ural boundaries, the thinking was that thisdeliberate disregard for political boundarieswould enable the CMAs to continue their

business throughout successive politicalchanges. Nevertheless, the Constitutionprovides for ‘cooperative governance’ wherethere are overlapping functions. The CMAswill have to find ways to engage effectivelywith the local governments, who bothrepresent the citizens’ interests and haveresponsibilities to provide water andsanitation services.

9.5.2 Informational attributes

It has been argued that, on the one hand, theconsultation process in the Olifants basinfor establishing the CMA was not fullyeffective in reaching the large majority ofpoor stakeholders; and on the other hand,the long consultation process is delayingaction to address some of the serious issuesand problems of the basin (Wester et al.,2003). One of the lessons of the water lawconsultations was the importance of trustedinformation as a basis for consultation andnegotiation. Good information is crucial todelineate areas of agreement and disagree-ment to structure and inform debate, but oflittle use if the source is doubted. Equally,good information becomes useful in negoti-ation and decision making only when itis accessible by all interested and affectedparties. South Africa is well equipped touse the most modern techniques for datagathering, storage and knowledge creation,but faces a major challenge in presentinginformation in a meaningful way to thewide range of interests in the sector. Thosemost in need of water for basic and produc-tive uses are poorly equipped to access andinterpret information from the nationalsystems.

Indeed, good information has thepower to defuse unnecessary tensions,while misinformation and lack of informa-tion are powerful instigators of conflict. Forexample, during public consultation onSouth Africa’s largest inter-basin transferscheme, from the Orange River system to theeconomic heartland around Johannesburgthrough the Lesotho Highlands, irrigatingfarmers along the lower Orange were




concerned that their future was in jeopardy.A rumour spread quickly that the water cri-sis in Johannesburg was due to the estimated6 million illegal immigrants in and aroundthe city. A quick back-of-the-envelope cal-culation put an end to the rumour: the basicwater needs of 6 million people would onlyirrigate about 5000 ha, a fraction of the irri-gation along the Orange and less than 0.5%of the irrigated area nationally. However,information is often more complex than inthis example, and open to interpretation.

The language of consultation is a partic-ularly important tool in processes of inclu-sion and exclusion. Preparatory consulta-tions in the local languages are a powerfultechnique to prepare the rural poor to engagewith other water user sectors. This problemis recognized but not yet fully addressed.

9.5.3 Legal authority

South Africa possesses a strong legal frame-work for water resources management.Water is viewed as an indivisible nationalasset, held in custody by the State. Conse-quently, significant powers vest in theMinister of Water Affairs and Forestry.Legal authority is not expected to be aconstraining factor in the implementationof basin management. Similarly, the CMAsand WUAs are provided a strong legalbase in the NWA. Indeed the NWA pro-motes the establishment of CMAs, at theinitiative either of water users in the WaterManagement Area or of the Minister.

9.5.4 Resources

It is often said by South Africa’s neighboursthat they lack the resources to conductpublic consultation processes as compre-hensive as those characterizing SouthAfrican policy making. The ongoing costof consultation during implementation wasthe subject of exhaustive debate during thewater law review. Compromise was reachedby toning down the specifics around con-sultation requirements in the NWA. Still

there are important unresolved questionsregarding the potential for financial auton-omy of the CMAs and how they will befunded.

Water resources management functionswill be delegated to CMAs in 19 Water Man-agement Areas. Currently, some functionsare performed at DWAF head-office in Preto-ria, while others are the responsibility of itsregional offices. Demand may well outstripsupply of human, financial, institutionaland infrastructural resources to service sucha large number of water management offices.Modern water resources management isdependent on a range of specialist skills thatmay be unaffordable and unattainable byall 19 CMAs in the short to medium term.This possibility has led to discussions overthe possible sharing of specialist expertiseacross CMAs. In the new political frame-work, another challenge would be to findleadership with command of the technical,social and political skills required by thejob. Water managers of the previous erawere highly skilled technically, but were notobliged to possess the wider range of skillsnecessary in the new framework.

9.6 Conclusion

Speculation on the future of integratedwater resources management in the Olifantsbasin needs to be informed by the currentpolicy and legislative framework, as well ashistoric factors and issues of water scarcity,and emerging critical issues. Some keythemes which will characterize the futureare outlined below.

• There is a single source of authorityand power. The Constitution and theNWA vest the custodianship of SouthAfrica’s water in the Minister of WaterAffairs and Forestry. While this enablesDWAF to keep tight control on theCMAs’ adherence to national policy asexpressed in the NWRS, it implies thatthe process is externally induced anddriven. At the same time, the CMA maybe tempted to inflate its empire withfunctions better addressed at either




national or local levels, exacerbatingproblems of sustainable financing. Theministry will have to proceed carefullyto empower the emerging CMAs, andto refrain from casual interventionsonce they are empowered, if they are todevelop a sense of responsibility andlocal accountability. Another distinctadvantage of this single source ofauthority is that water allocations,called ‘water use entitlements’ in theNWA, are well defined.

• Poverty and poor representation of theneediest sectors. It is still unclear towhat extent the CMA will have a devel-opmental agenda focused on address-ing poverty, though the latest draft ofthe CMA proposal (BKS, 2002) doesrecognize this role. Schreiner et al.(2002) suggest two possible scenarios:(i) ‘public participation’ that is cap-tured by the powerful, thus reinforcingmarginalization of the disadvantaged,and (ii) a scenario that builds povertyeradication and achieving genderequity into the initial design of theCMA. Even if the second scenarioprevails, however, ultimately a balancemust be secured by the disadvantagedthemselves developing the organiza-tion and political muscle to protecttheir rights and privileges. The right towater for ‘basic human needs’ is a well-protected right in both the Constitutionand the NWA, but lack of access towater for domestic use is still a veryserious problem in the former home-lands in the Olifants. Development ofdomestic water services, which is thedomain of local government, is of theutmost importance to rural residentsand is largely independent of the CMA.However, in the realm of small-scaleeconomic activity, CMA activitiescould affect tremendously the ability ofthe poor to improve their livelihoodsby making available additional alloca-tions of water to these sectors. Informa-tion flows need to be designed to reachall water user sectors, but especiallythose with limited access to modernmedia.

• From administrative to public proces-ses for water allocation. Water alloca-tion was historically an administrativeprocess with little public involvement.The NWA introduces mechanisms forpublic interaction at a relatively local-ized level, through the developmentof an ‘allocation plan’ in the CMA’s5-yearly catchment management strat-egy. This implies a shift from user-to-official interaction to a much moredirect negotiation among user groups.In a context of growing water scarcity,conflict management is likely tobecome increasingly important. Thisis a vitally important issue. Powerfulinterests, such as mining and large-scale commercial sectors, still havevery strong voices, and of course havestrong arguments for protecting theiraccess to water: mining and agriculturein the Olifants basin are major earnersof foreign exchange. On the other hand,there are concerns as to whether thisstatus quo is best for the millions ofpoor people living in the basin, withlittle or no access to water for theirown productive uses. Further, there arelarge uncertainties regarding the degreeto which there will be unmet waterdemands in the future in this basin andthe extent to which these can be metby better demand management (Leviteet al., 2002).

• Growing water quality issues. Intensifi-cation of economic activities like min-ing, industry and farming, as well asgrowing population densities will putincreasing pressure on the quality ofwater resources. Innovations like theexisting agreements on quality manage-ment through ‘controlled releases’ ofpolluted water from coal mines duringhigh-flow periods will gain impor-tance, but remediation will eventuallyneed to enter the next, more costlyphase of treatment. Water qualityimpacts of dense settlements andagricultural outflows will demandattention, especially in the light ofnational standards, requirements forthe maintenance of an ecological




reserve, and achievement of resourcequality objectives.

In sum, South Africa is at an early stage ofits long journey to effective integrated man-agement of its water resources. While itslegislation and policies are ahead of thoseof many countries, it faces special majorchallenges. Both nationally and in many ofits basins, it faces potentially serious gapsbetween growing demand for water and theavailable resources. It needs to focus farmore than it has on how to use thewater productively while also reducingthe current gross inequities and achievingthe aspirations of its poor majority tobenefit from this and other resources.

References

Ashton, J. (2000) Conceptual Overview of theOlifants River Basin’s Groundwater, SouthAfrica. African Water Issues Research Unit,University of Pretoria.

BKS (2002) Proposal for the establishment ofa Catchment Management Agency for theOlifants water management area. Departmentof Water Affairs and Forestry. Draft,November 2002. Appendix A, B, C, D, E.

Department of Water Affairs and Forestry.Proposed National Water Resource Strategy,first edition, August (2002) (also websitehttp://www.dwaf.gov.za/Documents/Policies/NWRS/Default.htm) and itsAppendix 9 D4, concerning the Olifantsriver basin.

Faysse, N. (2003) Possible outcomes of small-holders’ participation in water resourcemanagement institutions in South Africa.

Draft paper under review for publication.IWMI, Pretoria.

Hirschowitz, R. (2000) Measuring Poverty in SouthAfrica. Statistics South Africa, Pretoria.

Levite, H., Sally, H. and Cour, J. (2002) Waterdemand management scenarios in a water-stressed basin in South Africa. Paperpresented at the 3rd WARFSA/WaternetSymposium, Dar-es-Salaam, Tanzania,30–31 October.

Midgley, D.C., Pitman, W.V. and Middleton, D.(1994) Surface Water Resources of SouthAfrica 1990. Water Research Commission,Pretoria.

Mullins, D. (2002) Calculation of Sectoral Multi-pliers. Conningarth Economists, Pretoria.

Schreiner, B., van Koppen, B. and Khumbane, T.(2002) From bucket to basin: a new paradigmfor water management, poverty eradicationand gender equity. In: Turton, A. andHenwood, R. (eds) Hydropolitics in theDeveloping World: a Southern AfricanPerspective. African Water Issues ResearchUnit (University of Pretoria), Pretoria.

Thompson, H., Stimie, C.M., Richters, E. andPerret, S. (2001) Policies, Legislation andOrganizations Related to Water in SouthAfrica, with Special Reference to the OlifantsBasin. IWMI Working Paper 19 (SouthAfrica Working Paper 7). IWMI, Colombo,Sri Lanka.

van Koppen, B., Jha, N. and Merrey, D.J. (2003)Redressing Racial Inequities Through WaterLaw in South Africa: Revisiting Old Contra-dictions? Comprehensive Assessment ofWater Management in Agriculture ResearchPaper 3. IWMI, Colombo, Sri Lanka.

Wester, P., Merrey, D.J. and de Lange, M. (2003)Boundaries of consent: stakeholder represen-tation in river basin management in Mexicoand South Africa. World Development 31,797–812.




10 Water Resource Management in theDong Nai Basin: Current Allocation Processes

and Perspectives for the Future

Mark Svendsen, Claudia Ringler and Nguyen Duy Son

10.1 Introduction

Vietnam is emerging slowly from an eraof strong, but compartmentalized, centralcontrol of all-important functions to onewhere limited authority is being delegatedand more effective linkages among sectorsand ministries are being established. Onesector in which this is taking place is thewater resource sector.

This chapter examines the policy andinstitutional environment governing waterallocation and other management functionsin the Dong Nai River basin, together withthe rules-in-use, which currently controlallocation decisions. In doing this, itemploys the interactional analysis matrixdeveloped in Chapter 1 and the concept ofstages of basin closure, developed in Chap-ter 2. The chapter will define the basin andoverview basin hydrology and water uses. Itwill then assess the present level of stress onbasin water resources and its implications.Following this, it will describe the legalframework for basin management, identifythe important stakeholders in basin watermanagement, and examine the processesby which allocation and water control deci-sions are currently made. A final section

discusses possible future changes and theimplications for agriculture.

10.2 Basin Hydrology

10.2.1 Boundaries

Vietnam is formally divided into ninebasins for water resource planning. Oneof these is the basin entitled Dong Nai-SaiGon and Surrounding Areas. It covers about15% of the total surface area of the country,not including the roughly 10% of thetotal Dong Nai watershed, which lies inneighbouring Cambodia. The main stem ofthe Dong Nai River extends for 628 km, allwithin Vietnam.

While based primarily on watershedboundaries of the Dong Nai and tributaries,this planning unit includes a number ofshort coastal rivers that flow out of themountains and across a narrow plaindirectly to the South China Sea. In addition,the West Vam Co River adjacent to theMekong Delta is not included in theplanning basin, though it is a tributary tothe Dong Nai River near its mouth.1


1 A primary reason for this is that the West Vam Co receives substantial transfers from the Mekong Riverthrough a connecting channel.



While the basin defined in this way isappropriate as a planning unit, for manage-ment purposes the Dong Nai basin is moreproperly defined as the hydraulic unitdraining into the South China Sea throughthe Dong Nai River. This definition wouldexclude the coastal rivers in Ninh Thuan,Binh Thuan and Ba Ria-Vung Tau Provincesand include the West Vam Co River in LongAn Province. The rationale for this is that themanagement unit should cover a hydrauli-cally integrated area where water-relateddecisions in one part of the unit have thepotential to affect other parts of the unit. Itshould exclude unaffected and unrelatedareas. The southeastward-flowing coastalrivers generally do not satisfy this criterionand should thus be excluded from themanagement unit.

Management decision making relatingto these small external watersheds shouldrest with local stakeholders and not withthe larger basin council. This has twocomplementary advantages. First, it avoidsunnecessarily expanding the basin counciland complicating its decision-making pro-cesses. Second, it allows control of thesmaller coastal basins to remain with localstakeholders, rather than transferring influ-ence and control to the larger multi-partybody.

Complicating this picture, however, isan existing transfer from the upstream DongNai River to the Cai River in the coastal area.Additional diversions are planned. Clearlyplanning for future inter-basin diversionsneeds to include all of the affected areas,justifying the current planning boundaries.For management purposes, however, unlessthese inter-basin diversions are the domi-nant source of water for the receivingbasin, it is more appropriate to managetransfers through bilateral interactionsbetween managers of the two basins. If thetransfers do comprise the majority of flow inthe receiving basin, then it may be reason-able to include the whole of the receivingbasin in the purview of the managementbody for the larger basin and give receiving

basin stakeholders representation in thisbody.

Defined as a management unit, as above,the Dong Nai basin covers about 40,200 km2

and comprises four major branches. Theseare the Dong Nai main stem, the Be, theSaigon and the two branches of the Vam Co(Fig. 10.1). The basin covers all or nearly allof five provinces2 and a part of five others.

10.2.2 Characteristics

The Dong Nai basin has several distincthydrogeological regions, ranging from thelowland areas in the Vam Co Dong Riversystem that are inundated from Mekongfloods during the rainy season, to theCentral Highland areas of up to 1600 m.The lower basin reaches are subject to tidalinfluences, particularly during the dry sea-son, with substantial saltwater intrusion.Precipitation averages 2000 mm, rangingfrom 1200 mm in the lowlands to 2800 mmin the highlands. The basin exhibits markedseasonal variations in flow with 87% oftotal precipitation concentrated duringthe rainy-season months of April/May toOctober/November. In addition, there aremodest temporal variations in flow, withlow inflows of 28 billion (109) cubic metres(BCM) in 1998 compared to high inflows of40 BCM in 1997.

10.2.3 Basin water balance

10.2.3.1 Supply

The natural yield of the basin, averagedover the 1978–1998 period, is around36.26 BCM annually. Imports from theMekong to the Vam Co tributary contributean estimated additional 6.22 BCM per year,while power generation diversions to theCai basin reduce the basin water avail-ability by about 0.64 BCM.

The current project inventory forthe basin includes additional storage of


2 Ho Chi Minh City is considered as a province here.



Water Resource Management in the Dong Nai Basin 171

Fig.

10.1

.D

ong

Nai

Bas

in.



4.23 BCM, with 1.95 BCM of that consideredlikely to be in service by 2010. By that year,basin storage would thus total about7.56 BCM (Fig. 10.2). Still, the optionsavailable to system managers will remainlimited as a consequence of the limitedstorage. Limited storage also means thatseasonal variations in discharge anddemand become extremely important, andthat short-term water balances, rather thanannual averages, will drive allocation andmanagement decision making.

Significant development of storagebegan in 1963 with the construction ofthe Da Nhim dam and is continuing. Totalstorage in the basin is relatively small at5.61 BCM, and comprises only about 13%of average annual discharge. All basins aredifferent and have different characteristicsand storage requirements for effective man-agement. In general, however, storage is anessential element of water control and man-agement and it is informative to contraststorage ratios in some other basins with thatof the Dong Nai. Indonesia’s Brantas basin,for example, lacks reservoir sites and canstore only about 4% of annual discharge. Onthe other hand, Australia’s Murray–Darlingbasin storage is about 250% of annual

discharge, and on the Colorado River inthe USA storage is about 400% of annualdischarge. Other things being equal, thissuggests that additional storage in theDong Nai would enhance the scope ofmanagement options available to managersthere.

10.2.3.2 Demand

Current demand for water in the basin is notknown with certainty, but is estimated inTable 10.1. As can be seen, while domesticand industrial demands are important, irri-gation and salinity control requirements arethe dominant items. Estimates of the levelof sustained dry-season reservoir releasesrequired for salinity control in the lowerriver vary widely. Values used here are flowsof 85 m3/s at Binh An on the Dong Nai Riverfrom Tri An Reservoir and 25 m3/s at thesite of the future Ben Than intake on the SaiGon River from Dau Tieng Reservoir. Theaggregate salinity control requirement shownis intended to control the tidal salinity frontat the points of withdrawal of the Ho ChiMinh City (HCMC) urban water supply, anddoes not directly consider any additionalenvironmental or waste dilution needs.3


Year

Bill

ion

cubi

c m

etre

s

Cumulative storage

Average annual discharge

Fig. 10.2. Cumulative basin storage capacity, 1950–2010.

3 HCMC does receive some controlled deliveries for waste dilution.



One difficulty in estimating demand,implicitly for surface water, is the lackof current knowledge about ground andsurface water interactions in the basin.This uncertainty affects both irrigation anddomestic/industrial demands, but is mostimportant for irrigated agriculture. Whileabout one-fifth of domestic/industrial use iscurrently drawn from groundwater, futuregrowth in this use sector is expected tobe met almost entirely from surface watersources, making it reasonable to allocatedomestic and industrial use to surface waterdemand.

Groundwater use for irrigation, how-ever, is growing rapidly, and much of thenew high-value crop production, such ascoffee, fruit and pepper now being estab-lished relies on private pumping of ground-water. The extent to which groundwater

aquifers and surface water flows are inter-connected in a given region is thusimportant in determining the share ofdemand that should be allocated to surfacewater resources. In Tay Ninh Province, forexample, it is known that seepage fromthe Dau Tieng Reservoir and associatedirrigation facilities have caused a rise in theregional water table from 10–12 m below thesurface to just 4–5 m, facilitating private irri-gation development and allowing extractionwith cheaper centrifugal pumps. In otherareas, private pumping is drawing down thewater table. Shallow water tables are typi-cally hydraulically connected with adjacentstreams and rivers, and it is reasonable toattribute this private well demand to surfacesupplies until more detailed information isavailable which would permit partitioninginto surface and groundwater supplies.

10.2.3.3 Balance

Viewed on an annual basin, there is stilllittle stress in the basin, and seeminglylittle need to incur significant expensefor coordination and management. Totalcurrent demand is only about 30% ofannual supply, even assuming no reuseof municipal and industrial return flows.

The picture changes significantly,however, when supply and demand are


Month

With storage

Without storage

DemandBill

ion

cubi

c m

etre

s

Fig. 10.3. Hydrographs of Dong Nai Basin supply and demand.

Source Current demand (BCM)

IrrigationDomestic water supplyIndustrySalinity controlHydropower

4.570.500.723.470.64

Total 9.90

Table 10.1. Gross water demand in the DongNai basin.



examined on a monthly basis. Figure 10.3shows monthly supply and demand aver-aged over 3 years, 1996–1998.

The without storage curve depicts thevolume of synthesized inflows to streamsand rivers throughout the basin. The withstorage curve shows supply as reservoirreleases from the three major storage reser-voirs in the basin combined with naturalinflows below each reservoir. The secondcurve thus shows the same volume ofinflows as the first, but distributed overtime according to the pattern of reservoiroperations prevailing during the subjectperiod. Existing storage is able to take themonsoon peak off the inflow hydrographand provide nearly four times the dry-seasonflow in March and April than would haveresulted without storage.

More important to observe, however, isthat during the first 4 months of the year,supply and demand are nearly in balance,and with small increases in demand overtime, or in poor rainfall years, the basin willbe seasonally in deficit. This will place thedomestic water supply for HCMC, located atthe lower end of the basin, at risk, both interms of absolute volume of available waterand as a consequence of inadequate flowspushing the tidal salinity front below theHCMC water intake on the Dong Nai River atBinh An. This constraint can be relaxed in anumber of ways, including shifting morehydropower production to the dry monthswhen reservoir releases can do double duty.However, each alternative solution has itsown set of benefits and costs.

The important point here is that, from amonthly and seasonal perspective, the basinis already approaching a stressed conditionand pressure will be increasing to find thelowest-cost approaches that will relievestress and ease constraints on the mostaffected sectors.

10.2.4 Water use sectors

10.2.4.1 Hydropower

Five reservoirs currently produce electricityin the basin (Table 10.2). The Dong Nai

basin is one of two major hydropowerproducers in the country, the other beingthe Da River basin in the north. Installedcapacity at the five Dong Nai hydropowerstations is about 1.2 GW. The basin thusrepresents about 25% of the nationalhydroelectric capacity and about 18% oftotal national installed generating capacityof around 6.5 GW. It is thus an importantpower producer, but not the major one inthe country. A 500-kV north–south inter-tiemakes it possible to transfer power backand forth between the north and the southof the country in response to demand.

Staff at each of the power stationsare employed by Electricity of Vietnam(EVN), a State-owned company establishedunder the Ministry of Industry (MoI).Generation is controlled by an EVN unit,the National Regulation Center (NRC), inHanoi. Under the NRC are two RegionalRegulation Centers – one for the north, alsoin Hanoi, and one for the south, locatedin HCMC. Prior to the completion of thenorth–south inter-tie in 1994, Dong Naibasin power stations were under the controlof the Southern Regional Center in HCMC.Since completion of the inter-tie, all powerstations in the country are controlled fromthe national centre in Hanoi. The NRCprepares yearly, quarterly and monthlyplans for power production for eachpower station in the country, thermaland hydro, and directs operations on areal-time basis. Power station personnelhave little discretionary authority withrespect to operations.


Dam

Designcapacity

(MW)

Annualoutput(GWh)

Activestorage(MCM)

Tri AnDa NhimHam ThuanDa MiThac Mo

400160300175150

17001025

957595590

2542156523

171260

1185 4867 4498

Table 10.2. Hydropower generation in the DongNai basin.



Because thermal stations generally havehigher operating costs, operators attemptto maximize the use of the hydro stations.Moreover, generation from hydro stationsis generally increased during daily periodsof peak demand to take advantage oftheir flexibility. Plants are thus operatedfor a combination of base load and peakingpower. Daily and hourly changes in generat-ing levels are ordered by the national centrethrough a computer network linking thestations and the centre. Data on actualreleases and generation are also fed into thissystem by the power stations. These data areclosely held by EVN.

In addition to generation targets, thereis an operating curve for each reservoirgiving maximum allowable reservoirelevations throughout the year. When areservoir level exceeds that specified by thecurve, spill is ordered. Spills are generallyavoided whenever possible because theybypass turbines and result in a loss ofpotential energy generation. These curvesare regarded as a part of the dam design, andare rarely changed. There is thus no dynamicallocation of flood storage based on long-range weather and climate patterns. Rulecurves for Tri An Dam are shown inFig. 10.4.4

10.2.4.2 Irrigated agriculture

The Dong Nai River basin boasts a highlydynamic agricultural sector, with productsranging from basic staples like rice and maizeto raw materials for local industry includ-ing rubber, cashew (both non-irrigated) andsugarcane and to high-valued crops likecoffee, fruit, grapes, pepper, tea and vegeta-bles. It produces about 15% of the total nat-ional agricultural output (Ringler et al., 2001).

Low rainfall during the dry season(as little as 10–50 mm/month) and dry spellsin the rainy season make irrigation indis-pensable for the cultivation of many crops.According to estimates by the Sub-Institutefor Water Resources Planning (SIWRP), thegross irrigated area in the Dong Nai basin(excluding the area irrigated by the WestVam Co River) is 630,000 ha. The net areairrigated during the dry season is anestimated 293,086 ha or 24% of the totalagricultural area.

Estimates of the share of groundwaterirrigation in total irrigated area vary widely,ranging upward from a conservative10–15% of the total to a third of all the cropsin the basin. Groundwater-irrigated cropsinclude the larger share of the coffee area,pepper and fruit trees, some upland crops,


Month

Ele

vatio

n (m

)

Fig. 10.4. Example of a rule curve for a basin reservoir.

4 Elevations shown for each month are for the first day of the month.



and even a small share of the area underrice. The higher estimate is supported by therapid increase in the number of pumps inuse in the basin. In 1999, pumps numberedan estimated 248,000, having grown at a rateof 16.1% per year during 1991–1999. Thisrate of increase is second only to the MekongDelta, where private pumping from canals isprevalent.

Over the 1999–2010 period, Vietnameseplanning authorities project growth in grossirrigated area in the basin of 3.85% per year,with a rapid increase in area under uplandand industrial crops and perennial crops,and slowing growth in areas under rice(SIWRP, 2001). As there are no water quan-tity measurements in the basin’s irrigationsystems, irrigation water demand must beestimated. Estimates based on the FAO cropwater demand methodology yield irrigationdemands of about 4.6 BCM. Estimates basedon a detailed farm production survey yieldwater demands of 3.8 BCM. The differencecan be accounted, in part, by the large cropwater demand of perennial crops indicatedby the FAO methodology that, in effect, isbeing supplied from shallow groundwaterdue to the plant’s deep root system instead ofirrigation applications. Based on the SIWRPirrigation expansion planning for 2010,irrigation water demand will increase to5.4 BCM by 2010, at a rate of 3.3%/year.5

10.2.4.3 Flood control

The Dong Nai basin is subject to flooding inboth upstream and downstream areas. Indownstream areas, flooding from rainfallreceived in the upper watershed can beexacerbated by flooding from the adjacentMekong River, local rainfall and tidal influ-ences. The official flood season is specifiedas 1 July to 30 November (MARD, 2000).

The three large upstream reservoirs inthe system afford some flood control poten-tial, but the level of protection provideddepends on the way in which the reservoirsare operated. Presently, two of the three

largest reservoirs (Tri An and Thac Mo) areoperated primarily for hydropower produc-tion, while the third (Dau Tieng) is notequipped for power production and is oper-ated for irrigation, domestic and industrialwater supply, and downstream salinitycontrol. The operating regulations for DauTieng describe operation for flood controland require consideration of Tri An andThac Mo operating plans. In all three cases,however, emphasis appears to be on opera-tion to fulfil the reservoir’s primary purpose– irrigation and hydropower generation,respectively. Even releases from Dau Tiengthat are well below spillway design capacitycan cause flooding and damage in down-stream Binh Duong Province, where habita-tion and economic development is presenton the flood plain below the dam.

The hazards of this single-purposeoperating regime were demonstrated inOctober of 2000, when continuous heavyrains over much of the watershed resulted inthe need to spill water simultaneously fromall three reservoirs (Ngoc Anh, 2000). Therains occurred near the end of the rainyseason when reservoirs were nearly full andspill was required to protect the dams. Majorspills occurred for 10 days on 10–20 Octo-ber. Downstream flooding in the basinresulted in 18 deaths, more than 37,000dwellings damaged or destroyed, whileaffecting more than 60,000 ha of crops.

In addition to calling for constructionof additional reservoirs and river dyking, areport on the 2000 flooding prepared bySIWRP suggests a number of new basinmanagement initiatives. These include thefollowing:

• Operations studies on the three largereservoirs to develop multipurposeoperating criteria and assigning toppriority to flood control;

• More catchment rainfall gauging sta-tions and improved flood forecasting;

• Flood control planning;• Surveys and research on flash flooding;• Flood warning and rescue systems.


5 This value was estimated based on the average irrigation application values from the farm householdsurvey applied to the projected crop mix in 2010.



10.2.4.4 Domestic and industrial water supply

The Dong Nai basin is the economic centreof the country, and domestic and industrialwater use has increased rapidly over thelast decade driven by increased invest-ments in industrial zones and water supplyinfrastructure. Despite this development,less than half of the population in thebasin is served with public connections.For example, in 2000, the HCMC watersupply company had a capacity of837,000 m3/day (10% of which was fromgroundwater). Domestic deliveries in thesame year reached about 417,000 m3/dayfor an urban population of 4.2 millionpeople. Currently HCMC employs two largetreatment plants utilizing water from theDong Nai River. A third plant, on the SaiGon River was completed in 2004. In addi-tion, private wells accounted for estimatedadditional 400,000 m3/day. In general,urban areas such as Tay Ninh Town, ThuDau Mot, Da Lat, Bien Hoa and HCMC havepiped distribution systems for treated waterdrawn primarily from surface sources.Groundwater supplements these sourcesand provides the only source of supply forsome smaller towns and many individualhouseholds.

Overall, domestic water demand hasbeen estimated at 0.5 BCM, based on astandard of 160 l/day per person for urbanareas and 60 l/day per person for rural areas.This probably exceeds current actual supplyand use rates.

Industries in the basin are concentratedin the lower basin provinces, including BinhDuong, Dong Nai, and HCMC. Estimates forindustrial use can vary by more thanan order of magnitude, ranging from162,000 m3/day based on delivery informa-tion to 2,000,000 m3/day based on an esti-mate of water use in the production of mainindustrial products in the basin. Industrialsupplies come both from municipal-treatedwater supply systems and from individualriver intakes or private wells. Organizedindustrial estates in some areas provideindustrial water supplies from wells toconcerns within the estates. Current indus-trial demand is estimated at just under

three-quarters of a billion cubic metresannually and is expected to continue to growrapidly.

10.2.4.5 Environment

Environmental uses of water in the basinappear, in practice, to have a lower prioritythan the extractive uses of irrigation andmunicipal and industrial water supply.Guidelines exist which suggest minimumflows of 25 m3/s at the future Ben Thanintake on the Sai Gon River and 85–100 m3/sat the Binh An water supply intake on theDong Nai River. These flows typicallyrequire releases of 25 m3/s from Dau Tieng,100 m3/s from Tri An and 70 m3/s from thefuture Can Don reservoir. The Dau Tiengrelease satisfies needs of local irrigation onabout 5000 ha in Binh Duong Provincealong the Sai Gon River below the reservoirand to control salinity, and the Tri Anreleases are to prevent the tidal salinityfront from migrating upstream past the pointof the HCMC municipal water intakes. Atthe same time, however, these releases havethe effect of putting a small assured flowinto the rivers below the reservoirs.

10.2.4.6 Other uses

Fisheries and aquaculture are relativelysmall activities in the freshwater portions ofthe basin, but are important contributors tolocal incomes and local diets. Cage aqua-culture contributes importantly to wasteloads in reservoirs, particularly Tri An.

The Sai Gon River is navigated byocean-going freighters visiting the HCMCport. The rivers upstream from HCMC areused only by much smaller craft, and waterflows and channel depths for navigationhave not been a significant issue for basinmanagement or reservoir releases.

10.3 Institutions and Actors

10.3.1 Legal framework

In 1999 and 2000, Vietnam initiated a seriesof major reforms in the country’s water


A4859 - Svendsen - Vouchers Proofs #R pg 177a.prnZ:\Customer\CABI\A4831 - svendsen\A4859 - Svendsen - Vouchers Proofs #R.vpMonday, January 24, 2005 3:35:55 PM


sector – including the framework Law onWater Resources (hereafter called the WaterLaw) of 1999, and the Decision on theEstablishment of the National WaterResources Council (NWRC) in June of 2000.Among other things, the Water Law enablesthe establishment of river basin committeesor organizations in the country.

These measures are meant to bring morecoherence into a sector currently comprisedof highly fragmented water authorities withsometimes overlapping responsibilities andlittle coordination. They follow the 1995merger of the Ministry of Water Resourcesand the Ministry of Agriculture, whichtended to align water resource managementwith one of its most important user sectors,but at the cost of potential for a broaderresource-based perspective. As a result, thecountry is currently transitioning from awater sector that is fragmented and weaklycoordinated to a more holistic, decentral-ized and integrated management of thecountry’s water resources at the river basinlevel.6

The Water Law was adopted on 20 May1998 and went into force on 1 January 1999.According to the law, water resourcesbelong to the people under the managementof the State, and organizations and individu-als have a right to exploit and use theresources (Official Gazette, 1998). Waterallocation is carried out from a river basinperspective adhering to the principles offairness and reasonability. Priority in use isaccorded to drinking water in both qualityand quantity (Article 20).

According to the Water Law, the Minis-try of Agriculture and Rural Development(MARD) is in charge of overall managementof the country’s water resources, but the gov-ernment may delegate authority for specificwater uses to other ministries. Water man-agement is to be carried out in river basinsthat follow hydrologic catchment bound-aries, and not administrative ones. MARD,together with provincial governments, is

in charge of establishing both flood anddrought plans for the country’s river basins.Both water use and wastewater dischargeare to be licensed by provincial governmentauthorities (People’s Committees) under theguidance of MARD. Decree 179/1999/ND-CPof 30 December 1999 assigns specific dutiesto MARD, other ministries and ProvincialPeople’s Committees (PPCs) related towater resources management. Additionalregulations are currently being drafted toimplement the framework Water Law.

In addition to the Water Law, severalother laws and regulations are important forwater resources management in Vietnam.They include the Environmental ProtectionLaw (27 December 1993) and the MinisterialInstruction from the Ministry of Science,Technology and Environment (MOSTE)entitled Guiding Environmental ImpactAssessment for Operating Units (InstructionNo. 1420/QD-MTg).

In June 2000, a national-level umbrellaorganization for the water sector, the NWRC,was established, based on Article 63 of theWater Law (Government Decision No. 67/2000/QD-TTg). The NWRC has an office inMARD and a number of permanent memberswho represent the range of ministries andorganizations that are involved in waterresources management in the country.

In analysing water resource institutions,it is important to understand one salientconcept pervading governance in Vietnam– the concept of State management. Statemanagement gives pre-eminence to the Stateacross a wide range of public decisions. Thekey actor in this is the government, which isembodied in the office of the Prime Minister.Ultimate State management responsibilityresides here, and the most important deci-sions are made by this office. Statemanagement authority for lesser decisionsis delegated to ministries at the nationallevel and to PPCs at the provincial level.For example, for water resources, Article57 of the Water Law assigns a number of


6 In August 2002, after this chapter was written, water resources was again separated out from agricultureand placed under the newly created Ministry of Natural Resources and Environment (MONRE). Hence most ofthe functions ascribed below to MARD are now functions of the MONRE.



important State management functionsto MARD. At the same time, Article 58reiterates the supremacy of the governmentin exercising these State managementfunctions and makes MARD answerableto the government in carrying out theresponsibilities assigned to it.

A variety of other organizations, whichdo not possess State management authority,are placed under the direct supervision ofan organ of government which does haveit. Notable among these is the plethora ofpublic ‘companies’, which have been cre-ated by various government units to provideparticular services, such as infrastructuredesign and construction, or irrigationsystem management. These companiestypically have strictly limited autonomy andare directly supervised by an entity withState management authority. This supervi-sion goes beyond what is generally consid-ered ‘regulation’ and blurs the distinctionbetween government and nominally autono-mous State corporations. While the conceptof State management, as applied to Statecompanies, appears to be evolving slowly toallow more autonomy, it is still a strong andintegral part of their governance.

10.3.2 Basin actors and stakeholders

Actors can be defined as organizations andindividuals who take part in the processof managing basin water resources. Stake-holders can be defined as those organiza-tions and individuals having some directinterest affected by the outcomes of basinmanagement decision making. The set ofstakeholders thus includes actors, but isbroader, in that it can include groups suchas coastal fishermen, urban environmental-ists and small farmers who may not have adirect role in basin management.

In Vietnam, actors in basin managementtend to be government organizations ratherthan associations of individuals, trade orga-nizations and other civil society groups. Inthe case of multi-province basins, govern-ment organizations at national, regional andprovincial levels are typically involved,

with international agreements coming intoplay for the Mekong River. There are nointernational agreements for the Red River,despite its being an important trans-bound-ary river. In the case of the Dong Nai, thesmall portion of upper watershed locatedin Cambodia has so far not been causefor international consultations on basinmanagement.

There is a strong sense of hierarchyin Vietnamese government structures, withbonds within ministerial groups of units andorganizations being significantly strongerthan those among units across differentministries. This is true even where theseunits are addressing a common set ofissues. Within ministerial groups, there areoften extensive consultations and upwardrequests for clearance before decisions aremade. This can slow decision taking andlimit cross-ministerial discussion. At thesame time, it leads to coherence and confor-mity with ministerial policy directives.

The general form of public administra-tion in Vietnam employs a matrix model,in which sub-national administrative unitsare responsible in two directions (Fig. 10.5).At the provincial and district level, sector-based units such as Agriculture and RuralDevelopment or Public Works and Trans-portation are accountable administrativelyto the Provincial or District People’s Com-mittee, while in technical matters they areresponsible to the corresponding ministry inHanoi. Budgets flow through the People’sCommittee, and staff are employed by theprovince but receive technical directionfrom Hanoi.

This picture is complicated by thepresence of a large number of ‘companies’,which have been created by governmentministries and departments at both nationaland provincial levels. These companies areState-owned and typically closely tied totheir parent ministry or department, oftensharing staff, functions and budgetaryresources. Most exist to do business withthe government or with other State-ownedcompanies.

In the water sector, important examplesare the MARD engineering design compa-nies, of which there are two at the national




level – Hydraulic Engineering Companies1 and 2 (HEC 1, HEC 2) – and numerous onesat provincial levels, and water resourcesconstruction companies at both national andprovincial levels. Similar sets of design andconstruction companies are found underthe Ministry of Construction (MoC) andprovincial construction departments fordomestic water supply and under the MoIfor hydropower development.

In addition, MARD and many provinceshave set up irrigation management compa-nies (IMCs) to supply water for irrigationand maintain irrigation facilities. Three ofthese are directly under MARD – the DauTieng Management Company, which oper-ates Dau Tieng Reservoir on the Saigon River,and two others, which operate large irrigationsystems in the north. Most provinces havealso created one or more IMCs to operateirrigation systems within their boundaries.Provincial IMCs are units of the ProvincialAgricultural and Rural Development Ser-vice (PARDS). The IMCs are generallyintertwined with the PARDS themselves.

The government is moving slowly toprivatize these engineering design and

construction companies, and to put theminto competition with a growing number ofprivate firms in the same field. Currently,however, private firms are largely limited tosubcontractor roles and many functions arestill reserved for State firms.

Because IMCs operate in a naturalmonopoly setting, such competitive reme-dies are not available for the managementproblems affecting them. Moreover, someIMCs have also become involved in theconstruction business, creating further ques-tions of conflicting interests and privilegedcompetition against private firms. Futureremedies lie in creating a much clearerseparation between the parent governmentunit and the IMC, providing for usergovernance control of IMCs, transparentsubsidies (if present) and other measures toenhance corporate behaviour and operatingefficiency.

10.3.2.1 National and regional actors7

Important national and regional actors inthe Dong Nai basin are shown in Box 10.1.The government, as discussed in an earlier


��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�� !""!#$

Fig. 10.5. Generalized management structure in Vietnam. Source: DWRHWM (2002).

7 Descriptions of line government ministries and departments draw on DWRHWM (2002) and Ringler et al.(2001).



section, exercises State management func-tions, in part through delegation to minis-tries and Provincial and District People’sCommittees.

In addition to MARD, the lead ministry,other ministries are also involved in thewater sector, as delegated by the govern-ment. Currently, the water sector in Vietnamis in a transition period and water resourcesare still largely administrated on a sectoralbasis. Thus, for example, the MoI is respon-sible for the National Hydropower Plan; theMoC is responsible for urban water supplyplanning; MARD is largely focused on irriga-tion sector development and flood control;and MOSTE is responsible for water quality.

NWRC The recently established NWRC isthe apex organization for water resources inthe country. The Council is chaired by a VicePrime Minister and includes the Ministerof MARD as a standing member and 15other permanent members comprisingvice ministers from Agriculture and RuralDevelopment; Planning and Investment;Construction; Science, Technology andEnvironment; Finance; Transportation;

Fisheries; and Industry as well as the head ofthe General Department of Hydro-meteorol-ogy; the Chairman of the NWRC Office; andfour specialists working in the water sector.The Chairman of NWRC may also appointnon-permanent members representing cen-tral and local agencies relating to specificissues on the agenda for each session of theCouncil.

The Council has the objective offacilitating coordination among the variousministries and agencies involved in waterresources management. Its role is to advisethe government on important decisionsrelated to water resources management,including: (i) strategies and policies onnational water resources; (ii) major riverbasin plans; (iii) plans for major inter-basindiversions; (iv) projects for protection,exploitation and utilization of waterresources and projects for flood controland water damage control; (v) management,protection, exploitation and utilization ofinternational water sources and disputesettlement; and (vi) conflict resolutionbetween ministries and branches, andbetween ministries and provinces and citiesunder central control.

MARD MARD, established in 1995 outof the three former ministries of waterresources, agriculture and food industry,and forestry (Decree 73/CP of 1 November1995), is the State agency in charge of waterresources management and reports directlyto the government. The 1999 Water Lawreaffirmed this role.

MARD manages water resources in thecountry, especially in the fields of irrigation,drainage, flood control and bulk water sup-ply. It is responsible for the overall policyframework, the planning and prioritizationof new development, and the allocation ofinter-provincial water resources. Fundingfor large capital projects, including maincanals of large irrigation and flood controlprojects, is largely provided by the centralgovernment and includes significant foreignassistance. Secondary works and localprojects are designed and funded by theprovincial government with assistance fromthe central government.


Box 10.1. National actors in water resourcemanagement of the Dong Nai Basin.

• Government of Vietnam• NWRC• MARD

• DWRHWM• Various construction-related departments• SIWRP• Southern WR Research Inst• HEC 1, HEC 2• Construction companies• Dau Tieng Management Company

• MoI• EVN

• NRC• PECC no. 2

• Groundwater Department• MOSTE• GDHM• MoC

• Design Company for WSS• WSS Companies 1 and 2

• MoPH• Ministry of Fisheries



Key water resources departments are: (i)Water Resources and Hydraulic Works Man-agement (DWRHWM); (ii) Flood Control andDyke Management; (iii) the Centre for RuralWater Supply and Sanitation (CERWASS);and (iv) Plan and Planning. Other associatedunits comprise design companies, construc-tion companies, the Institute of WaterResources Planning, the Water ResourcesUniversity and the Institute of WaterResources Research.

MARD is responsible for constructingdykes, headwork and canals down to150 ha, and provinces are responsible forthe areas below 150 ha. Operation andmaintenance of existing irrigation systemsgenerally rests with the provinces in whichthe systems are located. MARD controlsthree large inter-provincial irrigationsystems – Bac Hung Hai, Bac Nam Ha andDau Tieng.

MINISTRY OF PLANNING AND INVESTMENT (MPI)

The MPI is the central agency for allocatingresources among sectors. It receives sectoralsubmissions and prepares consolidatedpublic investment plans. The MPI coordi-nates Official Development Assistance(ODA) and direct foreign investment.

MOC The MoC is responsible for urbanwater supply, drainage and sanitation. Itsets regulations, and plans, designs andconstructs water supply and sanitation facil-ities, working through associated designand construction companies. Followingconstruction, management is transferred toa State-owned water supply company. TheMoC is also responsible for water supply insmall rural towns.

MINISTRY OF PUBLIC HEALTH (MOPH) TheMoPH is responsible for monitoring waterquality and for sanitation.

EVN EVN is responsible for hydropowerdevelopment under the direction of itsparent ministry, the MoI. It identifieshydropower development and establishesoperating rules for reservoirs. EVN designs,constructs and operates schemes throughassociated companies.

MOSTE The MOSTE was created inconjunction with the enactment of theenvironmental protection law in 1993.It sets water quality standards, conductsresearch and manages the environmentalimpact assessment (EIA) process for newwater resource projects. Its primary actionarm with respect to environmental manage-ment is the National Environmental Agency(NEA).

GENERAL DEPARTMENT OF HYDROMETEOROLOGY

(GDHM) The GDHM reports directly to thegovernment. It collects primary level hydro-meteorological data through its hydrologyand hydrometeorology stations networkacross the country, conducts surveys, andprovides analyses and forecasts based on thedata it collects.

10.3.2.2 Provincial actors

The PPC is the paramount authority at theprovincial level. Beneath it is a structure ofdepartments that mirrors that existing atthe national level. As indicated earlier,there are technical accountability linkagesfrom these provincial departments back tocorresponding ministries in Hanoi.

In the water resource sector, the PARDSis responsible for planning and implement-ing smaller water resource projects (lessthan 150 ha) and for irrigation systemorganization and management throughassociated IMCs. Other water-related res-ponsibilities are carried out by departmentsgenerally corresponding to the ministriesdescribed above.

A number of national agencies havebranches that are responsible for differentregions of the country. However, branches ofdifferent departments and ministries havedifferent geographic mandates. The resultis that particular provinces may fall into a‘southern’ region for one function and intoa ‘central’ region for another. Table 10.3shows the way that Dong Nai basinprovinces are allocated among differentsub-national jurisdictions for selectedagencies. The result is an even greaternumber of actors involved in basin planningand management functions.




10.3.2.3 Civil society

Civil society organizations, such astrade and professional groups, commodityproducers, environmental and other issue-based non-governmental organizations(NGOs), and landowner groups are non-existent or relatively unimportant inVietnam, at least in so far as water resourcesand agriculture are concerned. Basin plan-ning and management organizations inclosing (water-short) basins in other nationsoften include extensive participation ofsuch civil society groups. These groups givevoice to the interests of particular groupsof stakeholders such as farmers, urbanwater supply utilities or the environment.Stakeholder interest groups are typicallysupported by member contributions andmembership cuts across administrativeboundaries. The relative absence of suchstakeholder representation in Vietnam is afactor that must be taken into account indesigning basin management organizationsand considering their future development.Also to be considered is the political imbal-ance which may arise as powerful interestswhich are sophisticated and with relativelyfew members, such as industry groups,emerge in the absence of groups represent-ing more numerous but less powerful andsophisticated stakeholders, such as agricul-tural water users. The natural environmentrepresents a special case of particularimportance in that it has no direct represen-tatives and its interests must be advancedby public agencies such as MOSTE, or by

nascent civil society groups such asenvironmental NGOs.

10.3.2.4 Dong Nai Basin Council

According to the 1999 Water Law, waterresources in Vietnam are to be managed atthe basin level. A Planning ManagementCouncil for the Dong Nai River Basin wasestablished by a 9 April 2001 Decisionissued by MARD. The Council consists of12 standing and ten non-standing members,with the Vice Minister, MARD as chairmanand the Director of the MARD DWRHWMas vice chairman. Standing membershipalso includes the Director of the SIWRP inHCMC, who is designated as Head of theCouncil Office. The Office functions as asecretariat for the Council and is housedat SIWRP in HCMC. Though the Office isnot yet operating, two SIWRP staff membershave been designated to staff it, and it hasreceived a small allocation of VND 80M(US$5,333) to purchase furniture and officeequipment. At the same time, councils wereestablished for the Red River basin andthe Mekong Delta. These three basins are toserve as pilot sites for implementing thebasin planning and management conceptstipulated in the Water Law.

Of the three possible models for basinorganizations posited by Millington (1997),(i) a coordinating committee or council; (ii)a river basin commission; and (iii) a riverbasin authority, the council fits the first, andsimplest, model, though it also contains a


GW S GW C IWRP SIWRP FIPI S FIPI N NIAPP S NIAPP C

Binh DuongBinh PhuocDak LakDong NaiHCMCLam DongLong AnTay Ninh

XX

XX

XX

X

X

X

XX

XXXXX

XX

XXXXX

X

XX

XXXXX

X

GW, Division of Hydrogeology and Engineering (working on groundwater); FIPI, Forestry Inventory andPlanning Institute; NIAPP, National Institute of Agricultural Planning and Projections. S, south; N, north;C, centre.

Table 10.3. Regional responsibility breakout by selected agencies.



secretariat, which is characteristic of a basincommission in Millington’s framework.

Significant features of the Council, asestablished, include the following:

• All 12 standing members of the Councilare representatives of departments andministries of the central government.Representation is generally at thedepartmental Vice-Director level. Tenof the 12 are based in Hanoi, whiletwo are located in HCMC but directlyconnected with central governmentorganizations.

• The ten non-standing members of theCouncil represent the ten provincesincluded in the planning basin (seeBasin boundaries above).8 Representa-tion is at the relatively junior levelof Vice-Director of the PARDS. Assuch, provincial representation doesnot include sectors of public health,domestic water supply, wastewaterhandling and treatment, and environ-mental quality. Non-standing membersparticipate in the Council uponindividual invitation.

Given their administrative level and statusas non-standing members, provinces willlikely have limited influence in the Councilrelative to representatives of the centralgovernment.

A second separate basin-level effortrelated to water quality management is alsounderway. This effort grows from a UnitedNations Development Programme (UNDP)-sponsored programme, which broughttogether representatives of Departmentsof Science, Technology and Environment(DOSTEs) in five key basin provinces in thelate 1990s to discuss basin water planningand management. Although the originalUNDP support for this activity has lapsed,provinces have continued to discuss abasin-wide effort focused on water qualitymanagement. A meeting in April 2002resulted in a formal proposal for an interim

committee comprising ten basin provincesand HCMC, which was presented to theOffice of the Prime Minister for decision.The interim committee would focus on:

• Developing a common informationsystem for the basin;

• Increasing cooperation among provin-cial leaders for water qualitymanagement;

• Establishing communication channelsamong provinces for this purpose.

The proposal envisions primary representa-tion by the provincial DOSTEs, a secretariatthat rotates among the provinces, and thedesignation of an existing institution tomanage a monitoring system for the basin.

10.3.3 Functions and actors

Chapter 1 described a set of essential func-tions of river basin management. A listingof these functions, crossed with the keyactors identified in the previous section, isshown in Table 10.4. These functions arereplicated, as appropriate, across threebroad categories – surface water, ground-water and return flows (drainage). Thereturn flow category includes both urbanand industrial wastewater flows and agri-cultural return flows. Cells are marked toindicate an actor active in a particular func-tional area and whether the actor plays amajor or a minor role in performing thefunction. Information is drawn primarilyfrom informal focus group sessions withknowledgeable personnel at SIWRP com-bined with the experience of the authors.It is very important to note that the classifi-cation is based on assessments of actualeffective performance rather than nominalor official responsibility. Thus, where a par-ticular agency is given a clear official man-date to perform a function, but is ineffectivein carrying out this mandate, its scoring


8 Long An, which – from a hydraulic perspective – has about 95% of its area within the Dong Nai basin, hasnot been included as a non-standing member, compared with Dak Lak province, which has only about 15%within the Dong Nai basin or Ninh Thuan and Binh Thuan with even lower shares.




Sur

face

wat

erG

roun

dwat

erD

rain

age

Key

acto

rsN

atio

nal

Gov

ernm

ento

fVie

tnam

rr

qq

Nat

iona

lWat

erR

esou

rces

Cou

ncil

qM

AR

DD

epto

fWR

and

Hyd

raul

icW

orks

Man

agem

ent

rr

qq

rr

qq

HE

C2

rq

qq

Con

stru

ctio

nC

ompa

nies

rq

rq

Dau

Tie

ngM

gtC

ompa

nyq

qr

rq

qr

Dep

tofD

yke

Mgt

and

Flo

odC

ontr

olq

rD

epto

fInv

estm

enta

ndC

onst

ruct

ion

rq

qD

epto

fPla

nnin

gr

qD

epto

fPol

icy

rq

Con

stru

ctio

nM

gtB

oard

rq

qIn

tern

atio

nalC

oope

ratio

nD

epar

tmen

tq

Sub

-Ins

tfor

WR

Pla

nnin

gr

qr

qq

qq

Sou

ther

nIn

stitu

tefo

rW

ater

Res

ourc

esR

esea

rch

rq

qq

qq

qM

inis

try

ofIn

dust

ryG

roun

dwat

erD

epar

tmen

tr

r

cont

inue

d

Tab

le10

.4.

Key

acto

rsan

dfu

nctio

nsm

atrix

.

Planandprepareinvestments

Allocatewater

Constructfacilities

Distributewater

Maintainfacilities

Monitorquality

Ensurequality


Protectecology


Allocatewater

Constructfacilities

Withdraw/distribute

Maintainfacilities

Monitorquality

Ensurequality


Constructfacilities

Operatefacilities

Maintainfacilities

Monitorquality

Enforcequality




Sur

face

wat

erG

roun

dwat

erD

rain

age

Key

acto

rsE

VN Pow

erE

ngan

dC

onsu

lting

Com

pany

no.2

rr

qN

atio

nalR

egul

atio

nC

entr

er

rC

onst

ruct

ion

Com

pany

rr

Equ

ipm

entI

nsta

llatio

nC

ompa

nyr

rM

OS

TE

Nat

iona

lEnv

ironm

enta

lAge

ncy

qq

rq

qq

qG

enD

epto

fHyd

rom

etno

.2r

qq

Min

istr

yof

Con

stru

ctio

nW

SD

esig

nC

ompa

nyno

.2r

rq

qq

rW

SC

onst

ruct

ion

Com

pany

rr

rr

rr

Min

istr

yof

Pub

licH

ealth

qq

Min

istr

yof

Fis

herie

sq

Min

istr

yof

Tra

nspo

rtat

ion

qM

inis

try

ofD

efen

ce(E

nviro

nP

rote

ctio

nC

entr

e)q

q

Tab

le10

.4.

Con

tinue

d.


Allocatewater

Constructfacilities

Distributewater

Maintainfacilities

Monitorquality

Ensurequality


Protectecology


Allocatewater

Constructfacilities

Withdraw/distribute

Maintainfacilities

Monitorquality

Ensurequality


Constructfacilities

Operatefacilities

Maintainfacilities

Monitorquality

Enforcequality




Pro

vinc

ial

PP

Cs

rr

rq

qP

AR

DS

sS

ub-D

epto

fWR

and

HW

Mgt

rr

qq

qr

rq

qq

qIr

rigat

ion

Mgt

Com

pani

esq

rr

qq

qr

WR

Con

sulti

ngC

ompa

nies

ar

qq

WR

Con

stru

ctio

nC

ompa

nies

ar

qq

qD

epto

fDyk

esan

dF

lood

Pro

tect

ion

qr

DO

ST

Es

rr

qq

qr

qD

epar

tmen

tofT

rans

port

atio

nr

qD

epto

fCon

stru

ctio

nr

rr

rr

rr

rW

ater

Sup

ply

Com

pani

esb

rr

qr

rq

Dra

inag

eC

ompa

nies

bq

rr

rR

egio

nal&

Spe

cial

HC

MC

Exp

ortP

roce

ssin

gan

dIn

dust

rialZ

ones

Aut

horit

yr

qq

qq

qq

qq

q

Don

gN

aiB

asin

Com

mitt

eec

rq

qq

qP

rivat

ein

dust

ries

qq

qP

vtco

nsul

ting

and

cons

truc

tion

firm

sq

qr

Uni

vers

ities

and

othe

rin

stitu

tes

qIn

tern

atio

nalc

onsu

lting

firm

sq

q

a Dire

ctly

unde

rth

eP

PC

inso

me

prov

ince

sw

ithbr

oade

rm

anda

te.

b Com

bine

din

som

epr

ovin

ces.

c Not

yetf

unct

iona

l;an

ticip

ated

initi

alro

les.

r,M

ajor

role

;q,m

inor

role

.M

AR

D,M

inis

try

ofA

gric

ultu

rean

dR

ural

Dev

elop

men

t;H

EC

,Hyd

raul

icE

ngin

eerin

gC

ompa

ny;E

VN

,Ele

ctric

ityof

Vie

tnam

;MO

ST

E,M

inis

try

ofS

cien

ce,

Tec

hnol

ogy

and

Env

ironm

ent;

PP

C,P

rovi

ncia

lPeo

ple’

sC

omm

ittee

;PA

RD

S,P

rovi

ncia

lAgr

icul

tura

land

Rur

alD

evel

opm

entS

ervi

ce;D

OS

TE

,Dep

artm

ento

fS

cien

ce,T

echn

olog

yan

dE

nviro

nmen

t;H

CM

C,H

oC

hiM

inh

City

.

A4859 - Svendsen - pg 187a.prnZ:\Customer\CABI\A4831 - svendsen\A4859 - Svendsen - Vouchers Proofs #R.vpTuesday, January 25, 2005 10:27:11 AM


may indicate that it plays a minor role, orno role at all. Different observers can obvi-ously score the matrix in different ways,depending on their assessments of perfor-mance and effectiveness and the weightsthey attach to different components of per-formance.9 Nevertheless, the matrix givesa broad indication of which functions arebeing covered effectively, and by whom,and which are not being addressed or areaddressed by a multitude of actors. It couldreadily be refined – by developing separatetables for nominal and actual responsibil-ity, for example – but is introduced here toserve as a base for discussion and a broadindication of involvement and responsi-bility. Descriptions of the functions shownin the text are given in Chapter 1.

All functions represent the ‘action edge’of functional performance. Underlying eachare a set of supporting functions such as datacollection and management, resource mobi-lization, consultation and so on, that arerequired to carry out the top-level functionshown. For simplicity, these have not beenshown separately.

10.3.3.1 Key functions

Three key functions are of greatest impor-tance for basin level water management inthe Dong Nai basin. These are resourcedevelopment, water allocation and waterquality management, and are discussedbelow. Other functions, particularly thoserelating to water service provision, are alsoimportant, but are beyond the scope of thischapter (World Bank, 1996; Ringler et al.,2001; Chapter 1 of this book).

RESOURCE DEVELOPMENT Resource develop-ment remains the primary focus of waterresource management in Vietnam. Impor-tant functions involved are planning andpreparing investments, and constructingfacilities. A very large number of ministries,

departments and institutes are involved inthe planning and investment preparationprocess, particularly for surface waterresources. Many of these entities lie withinMARD itself. Elaborate procedures for thissurvive from the earlier era of comprehen-sive central planning. Many of theseare based on old Soviet standards andprocedures, but are being slowly updatedto achieve compatibility with standardsand procedures used by multilaterallending institutions and other internationallenders.

Construction of water resource facilitiesis carried out principally by State-ownedconstruction companies, and a number ofState agencies are involved in monitoringand supervising construction. For smallerprojects, planning and construction pro-cesses employed at the national level arereplicated, in somewhat simpler form, at theprovincial level.

Important resource development deci-sions are ultimately made by the govern-ment or PPCs.10 In addition to their signifi-cance for water resource development, thesedecisions also have important implicationsfor water allocation, as discussed in thefollowing section.

WATER ALLOCATION Water allocation canbe accomplished in different ways. Theserange from a permitting or rights allocationsystem covering all uses, to an open accesssystem where water is available for thetaking. The 1999 Water Law calls for a per-mitting system for both surface and ground-water to be operated by MARD, granting userights for water to larger-scale users. Rightsto groundwater use would be granted for 15years and surface water use rights for 20years. Small quantities of both ground andsurface water may be extracted and usedwithout permission by individual familiesfor domestic use and for small-scale agricul-ture, forestry, aquaculture and other home


9 This makes the matrix a useful tool in group planning processes to illuminate strengths and weaknesses, aswell as areas of agreement and disagreement among stakeholder participants.10 The MPI plays an important role in allocating investment funds among projects, but is not separated outhere for simplicity.



enterprises. During periods of shortage,priority is accorded to domestic water sup-plies over all other uses. MARD is currentlydeveloping regulations for implementingthis important provision of the law, butthe permitting system is not expected to beoperational until 2003.

Allocation of access to surface water useis currently controlled implicitly throughthe investment decision-making process.Since major users of water are almostexclusively public agencies and companies,public investments in new water-usingfacilities determine, de facto, the basicallocation of surface water among users.Such uses include irrigation schemes,hydropower stations, riverine port facilitiesand municipal water supply systems. Obvi-ously, the State would not construct thesefacilities unless it intended that they haveaccess to a reliable supply of water – ade facto allocation of water use rights.Hydrologic studies conducted during pro-ject preparation determine the availabilityof water in the source to be tapped, andmay also detail the expected return flowcontribution to receiving waters. By prepar-ing plans for such facilities, planning agen-cies such as SIWRP, HEC 2 and PECC 2play important supporting roles in waterallocation. Ultimately, water-allocatinginvestment decisions are taken by thegovernment, on the advice of MARD and itsDWRHWM.

Surface water can easily be over-allocated in this way as basins approachclosure; hence the need for a more rigorousand explicit allocation system as envisionedin the Water Law. Conflicts among wateruses can also arise over the shorter term,during periods of seasonal low flow andduring drought years. Similarly, the needto control flooding creates the need tomake allocation decisions – between hydro-power generation and increased floodstorage capacity, for example – over theshorter term.

In addition to the allocation of longer-term access to the water resource, short-termallocation decisions are required to dealwith changes in water availability anddemand throughout the year and to allocate

shortages in low flow years. In these cases,where capacity has been created to use morewater than is available at a given time, someprocess is needed to allocate the scarcityamong existing users. The Water Law indi-cates that in such cases, after giving priorityto domestic water supplies, water shallbe distributed according to ‘the percentagedefined in the planning of the river basinand the principle of ensuring fairness andreasonability’. EVN is an extremely impor-tant actor in short-term allocation throughits decisions about when and where to gen-erate power. The timing and magnitude ofpower releases have the effect of controllingthe amount of water available to down-stream users, particularly during the dryseason.

Presently decisions regarding theallocation of shortages among users duringdroughts and seasonal low-flow periods restwith the Southern Disaster and Flood Com-mittee. However, this committee is mostlyconcerned with flood prevention. In prac-tice, shortages are allocated by PPCs for res-ervoirs located within a single provincialjurisdiction. At present, there does notappear to be a comprehensive functionalsystem for dealing with shortage conditions.

With respect to flood control allocationdecisions, dam managers, i.e. the Dau TiengManagement Company and EVN, makerelease decisions on the basis of unchangingrule curves as discussed elsewhere in thischapter. These allocation rules are thuslargely fixed at the time of dam design,though operators have some operationalflexibility over the short run within theconfines of the upper and lower boundarycurves. In the case of severe floods, theSouthern Disaster and Flood Committeealso meets and has the power to changeallocations.

A more formal allocation procedureis currently in place on an interim basisfor groundwater. These interim procedureswill be supplanted when implementingregulations for permitting under the 1999Water Law are finished. Present rules callfor groundwater abstractions greater than1000 m3/day to be licensed by MARD,while smaller abstractions are licensed by




provincial authorities. Permits are notrequired for minor abstractions.11

WATER QUALITY MANAGEMENT Water qualitymanagement consists of two basic functions– quality monitoring and quality assuranceand enforcement. The former consists ofdata collection, processing, analysis, dis-semination and storage. The latter is a morecomplex task consisting of standard setting,problem recognition, problem diagnosisand problem solution through provision ofincentives, enforcement of standards andregulations, and mitigation of damage.

Both monitoring and quality assurancefunctions can be applied to ambient waters,both ground and surface, and to drainagedischarges from potentially polluting wateruses, including agriculture, municipalitiesand industry. In all cases, it is important toaddress both the simple monitoring functionand the more complex and difficult qualityassurance and enforcement function.

In the Dong Nai, a number of organiza-tions from both national and provincial lev-els are active in monitoring the quality ofsurface water in basin rivers. These includethe NEA, MoPH, SIWRP, Southern Institutefor Water Resources Research (SIWRR),GDHM, DOSTEs and water supply compa-nies. Groundwater quality is also monitoredby several agencies and departments,though less extensively than surface water.Monitoring of drainage water quality, partic-ularly municipal and industrial point sourcereturn flows, is more limited still. Overall,there are problems with inconsistent datasets, lack of continuous time series data,and a profound reluctance to share data andinformation with other agencies and withthe general public. Monitoring is takingplace, but it is fragmented and sometimesduplicative, and information is difficult toaccess. Most basin provinces feel the needfor an integrated water-quality monitoringnetwork for the basin.

There is far less activity in the areaof ensuring quality of receiving watersand enforcing water quality standards. The

provincial DOSTEs are the primary enforc-ing actors, but they often lack resources andclout to provide aggressive enforcement.At its present stage of economic and waterresource utilization in the country, empha-sis is on resource development, economicgrowth, expanding industrial output andemployment, and enhancing agriculturalexports. These objectives tend to take prece-dence over those related to water qualityassurance. Concomitantly, the influence ofnational ministries and municipalities andprovinces which host major industrialgrowth zones is generally greater thanthat of offices charged with protecting theenvironment. This results in a situation inwhich enforcement of existing quality anddischarge standards is incomplete.

Fortunately, the general abundance ofwater in the Dong Nai basin has preventedpollutants in wastewater discharges fromcausing major problems thus far. Despite thefact that HCMC, a city of 5 million people,currently has no wastewater treatment facil-ities whatsoever, pollution in the coastalestuary of the Saigon River, which receivesthis effluent load, is said not to be a signifi-cant problem. Still, this is an issue likely togrow to significant proportions in the nearfuture and will require cross-sectoral andinter-provincial cooperation to address.

10.4 Perspectives for the Future

Vietnam in general and the Dong Nai basinin particular remain in an evolutionarystage of water resource development. Thereis still relatively little pressure on the over-all water resources in the Dong Nai basin,and water quality problems have yet toreach crisis proportions. Concomitantly,primary emphasis remains on increasingstorage and developing new use potentialin the basin rather than on improvingefficiency in individual use sectors or onre-allocating water from lower to higherproductivity uses.


11 In Binh Duong province, for example, this is interpreted to apply to wells smaller than 60 mm.



There are both historical and resource-based reasons for this. Vietnam did notparticipate in the reservoir constructionboom, which took place in Asia duringthe 1970s and 1980s. As a result, there ispresently storage for only about 13% of theannual discharge of the Dong Nai Riversystem, and major portions of its largemonsoonal flow reach the South ChinaSea unused. This limits dry-season wateravailability for all uses, as well as theoptions available for controlling and allo-cating water in the basin. Hence, while onan annual basis the basin is still amplysupplied with water, seasonally a constraintis rapidly being approached.

A water quality problem is emergingtogether with growing dry-season scarcity.Burgeoning industrial growth and anabsence of municipal wastewater treatmentare degrading dry-season river water qualitywith consequences for users of high qualitywater such as HCMC and other expandinglowland population centres, and certainkinds of industries such as food processing.

Authorities are moving to address prob-lems on several fronts. These include regu-larizing the permitting system for water userights, monitoring water quality and estab-lishing a basin planning committee. Overall,however, primary emphasis appears to reston ambitious storage development plans forthe basin, which call for a 35% increase insurface water storage over the next 8 years.However, the fact that much of this storage isbeing developed for hydropower productionleads to important potential conflictsin operational goals among hydropoweroperators, irrigators, municipalities and theenvironment.

Presently, surface water allocation isimplicit in the basin investment planningprocess. Individual projects are evaluatedwith respect to water supply availabilityand, once approved and constructed,possess a de facto right to use water. MARDis currently working on implementingregulations for a formal system of use rightsallocation as called for in the 1999 WaterLaw. Groundwater regulations are beingtackled first, to be followed by surface waterpermitting rules. Principles articulated in

the Water Law establish priority for domes-tic water use and provide for a proportionalsystem of sharing shortages, though detailsof this procedure remain to be worked out.

A basin planning management councilfor the Dong Nai was established in 2001, butoperating regulations for the council are stillbeing worked out. Those regulations willprovide more insight into the functionsthe council will be performing, which, atpresent, are very broadly defined.

Based on the make-up of the council,the resource planning mandate of its secre-tariat, and the pattern of public prioritiesand decisions observed to date, it is possibleto speculate that the primary role of thecouncil, at least at the outset, will be inplanning new basin development activities.In this role, the council will serve as animportant consultative body for SIWRP inthe planning process.

Other roles may be added subsequentlyas basin infrastructure becomes more fullydeveloped, stress on the available resourceincreases, and water quality deteriorates.A primary requirement for more compre-hensive basin management will be thedevolution of greater power over resourcemanagement to provincial authorities. Thisis happening, but at a very deliberatepace. Provincial, municipal and other localauthorities are closer to the needs and prob-lems of the water users in the basin andare in a better position to represent thoseinterests. They also have more at stake in sat-isfying local constituencies. Ultimately, onewould expect to see direct representationemerging, as interest groups in civil societydevelop their own structure and voices.Over the short-term, it is likely that impor-tant decisions with respect to basin manage-ment will continue to rest with MARD andthe MoC and EVN in Hanoi, and that deci-sions will be more grounded in nationalinterests than in regional or local ones.

A parallel basin-wide coordinatingeffort has been ongoing for several yearsunder the sponsorship of MOSTE andsupport by the UNDP. This effort differsfrom the MARD-based planning manage-ment council in that it is based in provincialgovernments and in its focus on water




quality issues. Ultimately, these two effortswill need to be integrated or closely coordi-nated. For the time being, they appearpositioned to evolve on parallel tracks. Insome ways, this is a positive development asit aids the emergence of a provincially basedcapacity to engage in basin managementactivities. Coordination will be requiredand, at some point, the two mechanisms maybe merged, but this should not be done at theexpense of the seemingly more democraticand representative process emerging at theregional level.

What are the implications for irrigatedagriculture of this unfolding scenario? Aprincipal one is that a management strategybased on expanding basin storage is unlikelyto stress agriculture significantly, or requireit to relinquish water supplies it is currentlyusing, or even to curtail growth in the agri-cultural sector. Additional storage shouldenhance dry-season water supplies, whichcan be used by urban and industrial usersdownstream of the turbines of the hydro-power dams. In addition, the position ofMARD as both water resource manager andagricultural supervisor should assure a priv-ileged place for allocations to agriculture inthe basin.

Within the irrigated agricultural sector,strongest growth is expected in the cash-crop sector, where shallow groundwater isthe most important source of supply. Thishas the important advantage of being widelyavailable and being a demand-driven supplyunder the control of the well-owning farmer.On the other hand, shallow groundwatersources are vulnerable to drought-inducedshortages, since the shallow aquiferreservoir volume is often small, and toover-exploitation. This leads to the needto consider strategies to replace relianceon groundwater with farmer-controlled

reservoir-backed surface water supplies,and for artificial groundwater rechargeto buffer the effect of seasonal droughtson high-input cash crops, particularlyperennial crops.

References

DWRHWM (2002) Water Resources Developmentand Management in the Dong Nai RiverBasin. Report prepared for the ADB-IFPRIproject Irrigation Investment, Fiscal Policy,and Water Resource Allocation in Indonesiaand Viet Nam. MARD, Hanoi.

MARD (2000) Decision of Minister on DauTieng Reservoir Temporary OperationalRegulation. No. 127/2000/QD-BNN-QLN, 18December.

Millington, P. (1997) Institutional developmentoptions for the water management of theDong Nai Basin. Unpublished report.

Ngoc Anh, N. (2000) Preliminary report. Floodsituation in Dong Nai River system in2000 and flood control plan (Bao Cao So Bo –Tinh Hinh Lu Lut Nam 2000 O He ThongSong Dong Nai Va Nhung De Xuat Ve ChienLuoc Kiem Soat Lu). Sub-Institute for WaterResources Planning (Mimeo).

Official Gazette (1998) The Law on Water Resource(No. 8/1998/QH10 of May 20, 1998). EnglishTranslation. Official Gazette No. 21 (31 July,1998), pp. 32–47.

Ringler, C., Chi Cong, N. and Vu Huy, N. (2001)Water allocation and use in the Dong NaiRiver Basin in the context of water institutionstrengthening. Paper prepared for theworkshop on Integrated Water ResourceManagement in a River Basin Context,Malang, Indonesia, 15–19 January. IFPRI,Washington, DC.

SIWRP (2001) Data provided to the ADB/IFPRIRETA 5866 project. SIWRP, Ho Chi MinhCity.

World Bank (1996) Vietnam Water ResourcesSector Review. World Bank, Washington,DC.




11 Governing Closing Basins: the Case ofthe Gediz River in Turkey

Mark Svendsen, D. Hammond Murray-Rust, Nilgün Harmanciohluand Necdet Alpaslan

11.1 Introduction

11.1.1 National context

Turkey is a rapidly modernizing country ofabout 66 million people with a populationgrowth rate of 1.5% per year. At the begin-ning of the 1980s, national economicstrategy switched from a policy of industri-alization based on import substitution to apolicy aimed at allowing a greater role formarkets. Between 1979 and 1993, the Turk-ish economy expanded at an average rate ofaround 5% per year, with growth in recentyears being even faster. The GNP in 1998was US$200 billion, similar to that of Aus-tria or Sweden. However, per capita incomewas a more modest US$3160, slightlyless than those of Malaysia or Venezuela.Turkey is a very dynamic but still emergingcountry, displaying characteristics of bothdeveloping and developed nations.

The service sector dominates the econ-omy, contributing 62% of GDP in 1999, whileagriculture provided just 15%. However,agriculture still employs a very significant45% of the nation’s workforce. Agriculturalincomes, though, average less than one-quarter of those in other sectors. Regionaldisparities in income and other measures ofdevelopment also remain significant.

As a result, extensive internal migrationis leading to rapid urbanization. At present,65% of the population lives in urban areas,and the urban population is growing at 2.8%per year while the rural population shrinksat an annual rate of 0.7%.

Turkey clearly sees itself joining theranks of economically developed Westerndemocracies and hopes to join the EuropeanUnion by 2023, the 100th anniversary of itsindependence. This is a powerful motiveacross all sectors for harmonizing Turkishpolicies, practices and standards with thoseof the EU.

At a national level, Turkey is a parlia-mentary democracy. At the sub-nationallevel, the country is governed by a mixedsystem employing both local elections andcentral government appointments. Popula-tion centres (cities, towns and villages)are governed by locally elected assembliesor councils with administrations headed bylocally elected mayors. Provinces and dis-tricts, while having locally elected assem-blies, are headed by senior civil servantsappointed by the Ministry of the Interior.National-level policy guidance and instruc-tion is important at all levels and centralizedrevenue collection makes revenue transfersfrom national to local levels important andenhances the power of the centre.




11.1.2 The Gediz basin

The Gediz basin is located in WesternTurkey (Fig. 11.1) near Turkey’s thirdlargest city, Izmir, which is, however,located just outside the basin.

The Gediz River rises in the mountainsof Western Turkey and enters the AegeanSea to the north of Izmir. It covers a distanceof about 400 km and drains an area of17,000 km2, about 8% of which is irrigated.The basin had a population of 2.33 millionin 1997.

The Gediz basin has changed consider-ably in the past decade, shifting from a com-paratively water-rich basin to one that isnow closing.1 Change has been driven by anabove average increase in urban and indus-trial demand, rapidly growing concern forissues of water quality and environmentalprotection, and reduced rainfall over thecatchment. Paralleling these hydrologicchanges has been a much slower institu-tional response that has not kept up with therequirements for changes in the way water isallocated and managed.

11.1.3 The case study

Characteristics of the nation and the basintogether make the case an interesting onefor study for the following reasons:

• Turkey is a dynamic middle-incomecountry with rapidly growing demands

for industrial and municipal watersupplies;

• Agriculture is an employment sourcefor nearly half of the Turkish popula-tion and the country’s most importantcurrent user of water;

• The Gediz basin represents thesetrends well, containing both significantagricultural water use and rapidlyexpanding municipal and industrialdemands; moreover, its water resourcesare nearly fully allocated.

Although the drought of the early 1990s hasnow passed, the legacy is seen in a numberof important issues that continue to lieat the core of the debates surrounding themanagement of water in the Gediz basin.Several of these are highlighted below anddiscussed in more detail in the remainder ofthe chapter:

• The increasingly apparent need fora unified coordinating mechanismfor water allocation among irrigation,urban areas, industries and the envi-ronment to replace existing bilateralprocesses;

• The continuing struggle between olderlong-established institutions dealingwith water resource development andwater allocation and emerging institu-tions concerned primarily with waterquality and environmental issues;

• The need to represent and protectthe interests of certain water users,such as the Gediz delta ecology and


Fig. 11.1. Location of Gediz basin.

1 A closed basin is one in which there is no unused water left to be allocated.



the Irrigation Associations (IAs)established during the past 5 yearsto assume O&M responsibility for110,000 ha of large-scale irrigation sys-tems, within the wider debate of waterresources allocation and management;

• The need for: (i) clear rules assigningresponsibility for setting water qualityand quantity standards and monitoringactual conditions; and (ii) sufficientpolitical power and will to sanctionviolators of the standards.

In the remainder of the chapter, we dealfirst with the hydrology and water use pat-terns within the basin and then turn to thelegal, policy and institutional conditionswhich influence how the basin is governedand managed. Finally, we combine the twoassessments to summarize the problemsfacing the basin and the challenges it mustmeet to overcome them.

11.2 Water Resources of the Gediz Basin

11.2.1 Hydrology

11.2.1.1 Surface water

The hydrology of the Gediz basin is typi-cally Mediterranean. Precipitation fallsbetween November and April, and peakriver flows occur in February or March.Annual precipitation varies from 800 mmin higher inland areas to about 450 mm nearthe coast, with about 80% falling in thewinter months. Under natural conditions,there is a steady decline in stream dischargeuntil May when many of the smaller streamsdry up. Summer flows are only present inthe Gediz River and its largest tributaries,and even they may be negligible in the peaksummer months. Following the irrigationseason, the only flows in the Gediz Riverare from the few larger tributaries plusresidual return flows from irrigated areasand industrial and municipal wastewaterdischarges to the river (Fig. 11.2).

Following the construction of Demir-kopru Dam in the 1960s, net annual surfacewater availability in the main basin and the

delta is estimated to have been approxi-mately 1900 million cubic metres (MCM) peryear. During the period 1989–1994, a severedrought affected the basin, and since 1990there has been a persistent decline in surfacewater flows into Demirkopru. Annual wateravailability since that time has averaged just940 MCM. As some of this flow occurs inwinter and is derived from tributaries wherethere is no storage, there is little differencebetween annual surface water availabilityand current demand of about 660 MCM.

11.2.1.2 Groundwater

Groundwater resources are able to make upsome of the potential shortfall in overallwater availability. The central part of thebasin is an alluvial plain whose ground-water reserve is replenished in most years.

In the alluvial fan areas on either sideof the main valley and in the Nif Valley, thesituation is more critical. Tubewell-basedfarmers in the Akhisar area complain of asteady and long-term decline in water tables,and in the Nif Valley, the water table isreported to have dropped by 5–8 m in thepast 10 years as industrial extractions haveburgeoned. Springs in the limestone areasare also reported to have declined in the pastdecade.

The estimated safe annual yield forgroundwater in the main part of the valleyis 160 MCM/year, which is about one-thirdless than the 219 MCM estimated as beingextracted from the main and Nif valleys.Despite the absence of definitive figures,it appears that groundwater use presentlyexceeds, by a sizeable margin, thesustainable limit.

In the Gediz delta there is little ground-water utilization, and extraction near thecoast is prohibited to prevent saltwaterintrusion. The groundwater is deep andtherefore there are no shallow tubewells.

11.2.2 Basin water use

The Gediz basin contains a typical range ofwater users, although the balance amongthem has been changing during the past

Governing Closing Basins 195



couple of decades. Each user category isdescribed briefly.

IRRIGATED AGRICULTURE Traditionally, thelargest user of water has been irrigatedagriculture, originally deriving from smallrun-of-the-river diversions from the Gedizand its tributaries dating back some 3000years. Since 1945, the development of large-scale systems and groundwater exploitationhas transformed irrigated agriculture.

LARGE-SCALE SURFACE IRRIGATION The firstinvestments in modern irrigated agriculturebegan in 1945 with the construction oftwo large regulators to tap the flow of theGediz River. Adala regulator serves some20,000 ha of land in the middle portionof the basin, whereas Emiralem regulatorcommands 22,000 ha in the Gediz delta(Fig. 11.2). In the 1960s, a second set ofinvestments were made that included theconstruction of Demirkopru Reservoir afew kilometres upstream of Adala, a thirdregulator at Ahmetli, and the regulationand raising of the natural lake of GolMarmara. Ahmetli Regulator commandssome 45,000 ha of land. The final surface

water developments took place in theAlasehir Valley with the construction of twosmall reservoirs. The total command area ofthe large-scale surface systems is approxi-mately 110,000 ha. The predominant cropsare cotton (50%), grapes (35%), maize, fruitorchards and vegetables.

At present, surface water issues fromDemirkopru are limited to the intervalbetween mid-June and mid-September,which is focused on the cotton-growing sea-son. Natural stream flows from tributariescan be used for land preparation for cotton orfor early irrigation of grapes and fruit trees,but there are no releases made into the GedizRiver from Demirkopru outside this period.

Water use in the 90,000 ha of the centraland delta zones is limited to 75 m3/s fromDemirkopru and 15 m3/s from Gol Marmarafor a release period of approximately 60days, or a total of some 550 MCM during theyear. This is equivalent to some 450 mm ofirrigation water for the growing season.

In the Alasehir Valley in the east of thebasin, irrigation is almost exclusively forgrapes. Application rates are approximately350 mm/season and during the summerthere is no significant net outflow into


Irrigation diversions

Drainage return flows

1. Adala regulator2. Ahmetli regulator3. Emiralem regulator

Elevationin metres

metres

Fig. 11.2. Irrigation and drainage flow patterns, Gediz basin.



the main part of the basin. It is estimatedthat, through a combination of surfaceapplication and some pumping of theshallow aquifer, approximately 60 MCM areconsumed during the summer season.

SMALL-SCALE SURFACE IRRIGATION In manytributary valleys into the Gediz, there aresmall-scale surface water diversions thattake advantage of winter runoff and springsnow melt. Typical crops are fruit trees,winter wheat and vegetables, becausethese only require water in spring and earlysummer before the streams dry up.

There are no accurate records of the totalarea involved, but it almost certainly is morethan 25,000 ha, since almost every villagesituated on the valley fringe has someirrigated area (Kayam and Svendsen, 1999).Because the number of irrigations is low,normally two to four irrigations of about50 mm each for the entire season, total wateruse is also low and is estimated at 50 MCM.

GROUNDWATER IRRIGATION There are twodifferent categories of groundwater users:those who are members of village or pumpcooperatives and those who make privateinvestments.

Starting in the 1960s, but increasinglyin the 1970s, the government has fosteredcommunity-based irrigation from deep tube-wells. Most deep tubewells have dischargesin the range of 50–150 l/s and are often tap-ping groundwater at least 100 m below thesurface. The majority of wells are outside theboundaries of the surface irrigation systemsand are concentrated in the Akhisar andNif Valleys (Fig. 11.2). Typically, crops arehigh-value, and include tobacco, vegetablesand fruit trees. Total water extractions areestimated at 30 MCM/year on the basis of100 wells having a typical discharge of 75 l/sand operating for 40–50 days/year.

Private groundwater exploitationstarted during the drought of 1989–1994.Many individuals purchased centrifugalpumps to exploit shallow groundwaterwithin the boundaries of surface irrigationsystems, and in some cases neighboursformed informal pump groups to purchase apump and well and then share operating

costs at the end of the season. Farmers havecontinued using these pumps and therehas been a recent small increase as somefarmers adopt trickle irrigation systems forhigh-value fruits and vegetables.

The vast majority of private pump own-ers are within the boundaries of the surfaceirrigation systems. As such, they rely on theseepage and management losses from thesurface irrigation system. While there issome evidence that the shallow ground-water table dropped during the drought, ithas since recovered and it is assumed thatthey do not mine groundwater but merelyre-use surface water. Their net water use istherefore included in the total surface irriga-tion volumes. However, official records ofsurface irrigation shows that only about 70%of farmers use surface water and some ofthose also pump. It is estimated that some40,000 ha of land in the command areas ofthe surface irrigation systems are actuallypump-based with only 70,000 ha relyingprimarily on canal water.

A few private pump owners are situatedon the fringes of the surface irrigation systemor in the area between the Alasehir Valleyand the main Gediz Valley. These areestimated to use some 5 MCM/year that isnot direct recharge from surface irrigationsystems.

11.2.2.1 Municipal water supply

The Gediz basin has two separate classes ofurban and municipal water users: the townsand villages within the basin itself and asubstantial transbasin diversion of drinkingwater to Izmir.

WITHIN-BASIN USE There are no accuraterecords of total water extractions for urbanand municipal water consumption in theGediz basin. All municipal extractions arefrom groundwater. Based on estimates pro-vided by the different municipalities, itappears that extractions are in the orderof 130 MCM/year. However, much of thisreturns in the form of wastewater as eitherpercolation into the groundwater or dis-charge in surface water. Allowing for 20%actual consumption, the net municipal




extraction within the basin is estimated to be26 MCM/year.

Some municipalities also have showninterest in using good-quality spring waterfor their water supply. In a few cases, munic-ipalities have arranged with villages to usea portion of their spring water and agreeto compensate them by improving villageirrigation systems.

IZMIR USE OF GEDIZ WATER Izmir has had along-standing claim on groundwater withinthe Gediz basin and there are two main wellfields, at Sarikiz in the north of the basin andGoksu near Manisa. Actual consumptiondata are not available, but Izmir hasextracted as much as 108 MCM/year fromthese well fields. Because the water is trans-ferred out of the basin, there is no returnflow.2 An important potential source ofadditional water for in-basin use is theestimated 50% of the water entering theIzmir municipal system that is lost tounderground leakage. Since Izmir is on thesea, no reuse of these losses is possible. If theconveyance efficiency of the Izmir pipedsystem were improved, up to 50 MCM ofhigh-quality groundwater could be left inthe Gediz basin annually and used for otherpurposes.

11.2.2.2 Industry

There are two important industrial areas inthe basin. The largest is in the Nif Valleyimmediately east of Izmir in Kemalpasamunicipality (Fig. 11.2). There is also agrowing industrial estate on the westernedge of Manisa. Industries includedceramics, leather, food processing, metalworks and assembly plants.

In both areas, groundwater is usedfor the industries, and each industry mustobtain a permit from the General Directorateof State Hydraulic Works (DSI). However,there are no records of how much water

is consumed and it is difficult to make anestimate of total water use.

11.2.2.3 Hydropower

Between 1970 and 1988, DemirkopruReservoir was used for hydropower genera-tion throughout the year. Since the drought,however, power generation has beenrestricted to periods when water is releasedfor irrigation and no special releases forhydropower are made.

11.2.2.4 Environmental consumption

The seaward fringe of the Gediz deltais an important nature reserve and hasrecently been designated as a Ramsar siteto protect rare bird species. Originally, thearea received excess water from the GedizRiver for much of the year, but since 1990,with restrictions on irrigation releases, thereserve suffers from water shortages. Thesummer months are the critical time forproviding water specifically for the naturereserve, since during the winter water isavailable from the Gediz River beforeflowing to the sea.

In response to demands to preserve birdhabitat, one small channel with a capacity of0.7 m3/s does now extend from the irrigationsystem into the nature reserve. However, thechannel does not always flow at the maxi-mum rate during the 60-day irrigation sea-son, and so the potential volume of justunder 4 MCM for the season is not normallyprovided. One preliminary estimate sug-gests that to maintain appropriate condi-tions for freshwater bird habitat, as much as1.5 m3/s is required during the 120 days ofthe summer season, a total of about 15 MCM.Two other options are available for supply-ing water to the reserve. A pair of pumps ona nearby drainage channel can supply0.5 m3/s and a well is available which canprovide a modest 0.05 m3/s. These pumps


2 Plans are in place for Izmir to supply irrigators in lower portions of the Gediz basin with treatedwastewater. Irrigators are enthusiastic about this because of the very poor quality of the surface water theycurrently receive. There are concerns, however, about possible high salinity levels in the treated effluent. Atdesign output, the treatment plant would produce about 880 MCM/year, roughly equivalent to the entirecurrent use in the basin.



have not been operated in recent yearsbecause reserve managers have notrequested additional water.

A second component of environmentaldemand is the water needed for waste con-veyance from points of origin within thebasin to the sea. In transporting wastes, theflow must provide sufficient velocity tokeep organic compounds and heavy metalsadsorbed on to soil particles from settlingout before reaching the sea and sufficientdilution to avoid in-stream environmentalharm. Obviously, reducing the pollutantloads that must be carried would reduce thequantity of water needed for this purpose.

11.2.3 Summary

The total estimated water extraction by thedifferent users is shown in Table 11.1.

Irrigation currently uses a large share ofthe surface water resources of the basin.Withdrawals total about 660 MCM, with83% of that going to large-scale irrigationsystems. Current surface water allocationpractices are primarily aimed at providingreliable water deliveries to the IAs in thelarge irrigation systems, and this has beenachieved with considerable success. Hydro-power generation has no priority of its ownand uses only water that is to be released forirrigation. A small and probably inadequate

allocation of poor-quality surface water iscurrently made for the wetlands in the Gedizdelta.

Heavily polluted wastewater dischargesfrom urban areas and industries within thebasin seriously degrade the quality of sur-face water in natural channels, particularlyin the low-flow summer months. Sincewater use for these purposes is growing atan estimated rate of 6–8% per year, thisdegradation can be expected to worsenunless major efforts at control are madesuccessfully.

Groundwater supplies roughly a quarterof basin water use, of which about 16% is forirrigation and the remainder for urban andindustrial use. Groundwater supplies nearlyall of the water used for these latter twopurposes. Irrigation use of groundwater islargely static or declining as less-waterintensive crops replace cotton and improvedwater application technology gains a foot-hold. Municipal and, particularly, indus-trial use is expanding rapidly, however. Atpresent as much as one-quarter of ground-water withdrawal in the basin may beunsustainable overdrafting, and pressureon these aquifers is expected to increase asindustrial demand continues to grow.

Much of the water withdrawn formunicipal and industrial use within thebasin is returned to surface waterways, butin seriously degraded condition. This, inturn, gives rise to a need for additional


Estimated consumption

Water user MCM Share of total Notes

Surface waterLarge-scale irrigationSmall-scale irrigationHydropowerBird reserve

GroundwaterPump irrigation groupsPrivate irrigatorsUrban within the basinTransfer to Izmir CityIndustry

5506050

4

305

26108

50

62%7%6%–

3%1%2%

12%6%

From Demirkopru and Gol MarmaraAlasehir ValleyNo priority for hydropowerCurrent releases only; needs more

Only those outside surface irrigation area18% of extraction, remainder is return flowTrans-basin transfer, no return flowEstimated by DSI

MCM, million cubic metres.

Table 11.1. Estimated water use by sector.



allocations of surface water for waste loadtransport and dilution, water which is sim-ply not available at present. The alternativeis to improve quality of wastewater dis-charges significantly at their sources.

11.2.3.1 Changing patterns of demand

Patterns of agricultural demand are in flux.Rice used to be grown in poorly drainedcentral parts of the basin but has beenreplaced by cotton, and there has been asteady increase in grape and fruit tree areasas agroindustrial enterprises have grown upto support cash crop agriculture. The trendtoward grape cultivation, partly a responseto the growing market for raisins, resultedin decreased demand for irrigation water,and total irrigation deliveries are now onlyabout 70% of the pre-drought situation.With a recent surge in interest in dripirrigation by fruit, vegetable and seed maizegrowers, demand is likely to continue todecline.

In contrast, non-agricultural demand isgrowing rapidly. The area has a higher thanaverage growth rate because of in-migrationfrom poorer parts of Turkey, and Izmir’spromotion of industrial developmentto complement agricultural production.Domestic demand for water has beengrowing by 2–3% a year and industrialdemand by as much as 10% per year. Giventhat most non-agricultural consumption ofwater is from groundwater rather than

surface water, aquifer management requirescloser attention than surface water withrespect to available volumes.

However, an additional demand isarising which is associated with growingconcern over water quality, particularlyduring the peak of the summer season whensurface water supplies are limited. Figure11.3 shows actual and estimated growth ofbasin population between 1970 and 2010,along with estimated organic load from bothdomestic and industrial sources. As seen,although domestic load increases modestlyalong with population growth, industrialload grows exponentially. Note that thechart shows only potential loads createdand does not take into account the effect oftreatment facilities that may subsequentlybe built.


11.3.1 Water rights

All natural water resources, except somesmall privately owned springs, are vestedin the State by the Turkish Constitution(Yavuz and Cakmak, 1996). The basic prin-ciple governing surface water use rights inTurkey provides that water is a public goodwhich everyone is entitled to use, subject tothe rights of prior users. Surface water use


,

,

,

,

,

,

,

, ,

,,

,,

,

,

,

Fig. 11.3. Estimated Gediz basin population and organic waste load generation, 1970–2010.



is normally free of any obligation to obtainprior authorization. Conflicts are resolvedby first referring to local customary rulesand regulations. If the dispute cannot beresolved in this way, rights are settled bycourt decision. There is no registrationsystem for surface water rights or wateruse. In large basins where impacts of newdiversions are diffuse, this system is gener-ally unable to prevent or resolve conflictsbetween new and existing claims, and thisis leading to serious problems of over-allocation in some basins (Svendsen andNott, 2000).

Groundwater also is State property.Its management is governed by a 1960 lawgiving sole authority over the use and pro-tection of groundwater to DSI. Drilling a welldeeper than 10 m requires prior approvalfrom DSI, while constructing shallowerwells requires only that DSI be notified.Shallow groundwater is thus an open accessresource, while deeper aquifers are subjectto some controls.

According to the groundwater law,when abstractions ‘approach the safeoutput level’ of the aquifer, a committee ofrepresentatives of ‘relevant ministries’ is tobe formed to decide on pending and futureapplications for groundwater utilization.Frederiksen and Vissia (1998) conclude thatenforcement of the groundwater law isweak, implying that, in practice, the systemof groundwater rights created by the currentgroundwater law is not as effective as itcould be.

Rights to both ground and surface wateruse are thus not formalized. Although theyfollow roughly the appropriative doctrine ofallocation, there are no guarantees of contin-ued access. The principles of the system ofwater rights outlined above apply in theGediz basin.

11.3.2 Actors

In an earlier section, five categories of waterusers in the Gediz basin were identified andtheir respective water uses outlined. Someof these water users are able to represent

their own interests (industries and munici-palities), while others may be either manyand disorganized (small system irrigators)or unable for other reasons to representthemselves (ecosystems). In addition, thereare other State actors such as DSI involvedin Gediz basin water management, which,while not water users, are important play-ers. The range of basin stakeholders is thusdifferent, and broader, than the group ofactual water users. The major ones of theseare described below.

11.3.2.1 Public agencies

DSI The DSI is the main executive agencyof the Government of Turkey for the coun-try’s overall water resource planning, execu-tion and operation. It was established in1954 and is currently a part of the Ministry ofEnergy and Natural Resources. The mandateof DSI is ‘to develop water and landresources in Turkey’ (DSI, 1995). It isresponsible for major irrigation, floodcontrol, drainage, hydropower developmentand supplying water to cities with a popula-tion over 100,000. DSI centralizes most ofthe State functions involved in planning anddeveloping large-scale water resources.

Until recently, DSI’s policy has been tomanage the irrigation schemes it designsand constructs. Current policy and practiceis to transfer schemes to locally based IAsto manage. DSI also transfers hydropowerand municipal water supply schemes that itdesigns and constructs to other agencies tooperate.

DSI is also responsible for managingand allocating groundwater to prospectiveusers. It does this through the permittingsystem described in the previous section. Itsresponsibilities for groundwater quality arelimited to monitoring.

DSI maintains 26 regional offices acrossthe country, organized along watershedlines. The Gediz basin lies entirely withinone of these regions and is serviced by theregional office in Izmir.

MINISTRY OF ENVIRONMENT The Ministry ofEnvironment (MoE) is the public agencywith overall responsible for surface water




quality. In spite of this general status, how-ever, its mandate and capacities cover onlysome of the functions that implementingthis responsibility entail. Its major responsi-bilities include coordinating plans amongthe various public and private agenciesinvolved with protecting the environment,commissioning environmental impactassessments of major water resourcesprojects, and setting standards for andmonitoring surface water quality. Actualmonitoring and reporting of water qualityand wastewater discharges are carried out byprovincial offices of MoE. The explicitmandate of the MoE does not extend togroundwater quality. Neither the nationalnor the provincial offices of MoE possessdirect enforcement powers.

MUNICIPALITIES AND VILLAGES Towns andvillages play three important roles in thewater resource arena. First, they are waterusers and dischargers of wastewater. Thereare 19 settlements in the basin with a com-bined population of 1.35 million. All drawtheir domestic water supplies from ground-water. Of the 19, only three have completedwastewater sewage systems and treatmentplants. The remainder discharge untreatedwastewater into the Gediz and tributaries.

The second important role played bytowns and villages is that of representingirrigation water users in their areas. They dothis: (i) through their statutory dominance ofthe boards of large-scale IAs; (ii) as ownersand operators of municipal irrigationwellfields; and (iii) as representatives of theinterests of otherwise disorganized farmersirrigating from private wells or small surfacewater sources who make up parts of theirconstituencies.

The third role is that of environmentalregulation. Municipal and village adminis-trations are responsible for operating waterand wastewater treatment plants withintheir jurisdictions and monitoring the qual-ity of domestic water supplies.3 They alsohave some authority to monitor industrial

wastewater discharges, although most arenot active in this area.

PROVINCES AND DISTRICTS Provincial anddistrict governors, appointed by the Minis-try of the Interior in Ankara, are the onlyauthorities with the power to assess finesor issue and enforce prohibitions againstviolators of water quality regulations. Allother actors, including MoE, MoH, DSI andmunicipalities, may only report cases thatcontradict laws for which they are responsi-ble to the provincial or district governor.District governors must secure approvalfrom the provincial governor before takingaction. Provincial governors are thussingularly responsible for water qualityenforcement proceeding.

STATE PLANNING ORGANIZATION The StatePlanning Organization is an arm of thePrime Ministry that prepares a rolling5-year investment plan for the nation. It isresponsible for planning all public capitalinvestment in the country, including invest-ments for water resource development,wastewater treatment and environmentalproblem mitigation.

GENERAL DIRECTORATE OF RURAL SERVICES TheGeneral Directorate of Rural Services(GDRS), a part of the Prime Ministry, isresponsible for developing small-scalegroundwater resources for irrigation, devel-oping surface water sources with flows ofless than 500 l/s for irrigation, on-farm irri-gation development, and the construction ofrural roads and village water supply sys-tems. GDRS’s minor irrigation schemes aretransferred to farmers’ cooperatives or localgovernments upon completion. GDRS doesnot have an operation and maintenancecapacity.

11.3.2.2 Semi-public or private groups

IAS Thirteen Irrigation Associations (IAs)were established in the seven large canal


3 The Ministry of Health (MoH) may monitor the quality of drinking water in piped distribution systems at therequest of a municipality or village.



irrigation commands in the basin in 1995under the accelerated irrigation manage-ment transfer programme of DSI and haveassumed operational control of canal irriga-tion in those areas. DSI continues to operatethe main reservoir and river diversion struc-tures, but operational management belowthat level is now in the hands of the IAs.

The legal basis for forming IAs is a lawallowing the establishment of associationsof local governments, and the present gover-nance structure of the IAs is dominated byelected village headpersons, town mayorsand elected members of local municipalcouncils. IAs are public bodies that enjoy taxexemptions and are non-profit, but are notbound by standard government civil serviceregulations and financial procedures.Although this system has drawbacks, it doesprovide valuable links with local govern-ment structures. IAs operate the canal sys-tems within their areas, employing hiredstaff and financing operations and mainte-nance through fees collected from waterusers. The 13 IAs collaborate extensively onan informal basis and have discussed thepossibility of forming a more permanentassociation to represent their commoninterests. They are the most important waterusers in the basin and retain a strong func-tional tie to DSI, which provides their bulkwater supply and serves as the basin waterallocation authority in the absence of a moreexplicit system of water right allocation.

OTHER IRRIGATORS Other irrigation watersuppliers and users not encompassed byIAs include towns and villages that havedeveloped well-fields for irrigation supplyin their areas, individual farmers and groupsof farmers who have invested in irrigationwells, and farmers who employ small sur-face water diversions in upper parts of theGediz catchment to irrigate crops. There isno formal organization tying these waterusers together, though their number is sig-nificant. To some extent, local village headsand town mayors are able and generallywilling to represent the interests of these

irrigators when a need arises. Such represen-tation is not coordinated among villages,however, and in general would not beexpected to be particularly potent in compe-tition with larger, better organized interests.

ENVIRONMENTAL NON-GOVERNMENTAL ORGANIZAT-

IONS (NGOS) There are many NGOs activein the field of environmental conservation inTurkey. A 1995 directory lists 98 of them,and there are others not included in thedirectory. With respect to water-relatedissues in general, and the Gediz in par-ticular, the following are among the mostimportant:

• Turkish Erosion Control, Reforestation,and Environment Foundation (TEMA).TEMA was established in 1992 withstrong business community support. Itcurrently has about 50,000 membersand in 1997 operated on a budget ofUS$2 million. TEMA publishes amonthly bulletin on environmentalissues and every 2 years publishesan Environmental Profile of Turkey,which is also available in English.It enjoys good contacts with the MoE,and has been instrumental in shapingthe new national environmental lawsand regulations.4 It is the most influen-tial of the national environmentalNGOs.

• Gediz Basin Erosion Control Reforesta-tion and Environment Foundation(GEMA). This NGO has interests simi-lar to those of TEMA but is concernedspecifically with the Gediz basin.

• Society for the Protection of Wildlife.This society was established in 1975and works to raise awareness of shrink-ing populations of various wildlifespecies, with a special focus on birds.The society works extensively with ele-mentary school children, publishing anewsletter and guidebooks for schoolsand others. It collaborates with theWorld-Wide Fund for Nature and otherinternational organizations.


4 DSI has an agreement with TEMA for reforestation of certain catchment areas above DSI reservoirs.



Although concerned with water, none ofthese organizations places a priority focuson it. Most NGO activities to date havebeen concerned with education, awarenessraising and lobbying, with little independ-ent scientific or information collectioneffort evidenced.

INDUSTRIES Although industrial plants arescattered throughout the Gediz basin, thelargest concentration is in two OrganizedIndustrial Districts, one in Kemalpasa inIzmir Province with about 180 enterprises,and the second near Manisa in Manisa Prov-ince with about 50 enterprises. The ownersof these industries are organized into severalassociations, which wield considerablepolitical power. These include the AegeanChamber of Industry, the Businessmen’sAssociation and the Young Businessmen’sAssociation.

ENVIRONMENTAL ASSEMBLIES Local environ-mental assemblies (Mahalli Cevre Kurulu)have recently been formed in several basinareas. Authorized under the EnvironmentalLaw, assemblies are broadly constituted,comprising mayors, DSI, the Chamber ofIndustry, and so on, and are typicallychaired by the provincial governor. Theymeet monthly and are authorized to makefairly influential decisions on issues relatingto urban environmental quality. Such anauthority, chaired by the district governor,exists also for the Kemalpasa OrganizedIndustrial District. A similar association wasestablished for the Gediz delta bird reservein 2002.

WATER AND THE POOR As always, the poor,particularly those living in makeshift andillegal housing on the urban fringe, have theworst access to safe drinking water and sani-tation services. Because of more limitedmobility, they are also the ones most affectedby pollution of the Gediz, which they are

more likely to use for recreational purposes.Access to irrigation water is determined byaccess to land, which in turn is also relatedto wealth, both as a cause and as an effect.Many smallholders do practise very produc-tive agriculture, however, often growinghigh-value horticultural crops.

11.3.3 Essential functions: gaps and overlap

Essential functions of basin managementare described and defined in Chapter 1. Anessential functions matrix for the Gedizbasin, crossing functions with key actors, isshown in Table 11.2. These functions arereplicated, as appropriate, across four broadcategories – surface water, groundwater,wastewater disposal and agricultural returnflows. Cells are marked to indicate anactor active in a particular functional area.Information is drawn largely from richlydetailed reports prepared by Harmanciohluand Alpaslan (1999, 2000a,b).5

A number of interesting points emergefrom an examination of Table 11.2, supple-mented with background observations:

• There has been very limited planningat the basin level with respect tosurface water and virtually none forgroundwater and waste disposal.6

There is no integrated plan that consid-ers both ground and surface wateravailability, nor does existing planningconsider water quality, wastewater dis-posal, current and projected land use,anticipated future demand and returnflows, or projected future quantity andquality of water resources.

• Water is allocated, in practice, by avariety of agencies and users operatingindependently of each other. Theseinclude DSI, private surface andgroundwater irrigators, municipalitiesand industries. There is no national


5 Note that the activity indications contained in the table refer to actual activity in practice, and not nominalresponsibility as assigned in statutes. Open circles indicate limited activity, while filled circles indicate moreextensive activity. Situations where there is only minor activity might not be indicated in the table. Theindications are the collective judgements of the study authors and do not represent formal or official judgementsby any of the collaborating organizations.




Sur

face

wat

erG

roun

dwat

erW

aste

wat

erA

g.re

turn

s

Key

acto

rsD

SI

qr

rr

rr

qq

rr

rr

qIr

rigat

ion

asso

ciat

ions

qr

rq

Oth

erirr

igat

ors

qq

rr

qr

rr

GD

RS

rq

qM

oEq

qq

Loca

lgov

ernm

ents

rr

rq

rr

Indu

strie

sq

rr

rr

rr

Pro

vinc

es(M

oI)

qN

GO

sB

ank

ofth

eP

rovi

nces

(IB

)r

r,I

ndic

ates

activ

ity;q

,ind

icat

eslim

ited

activ

ity.

DS

I,S

tate

Hyd

raul

ics

Wor

ksO

rgan

izat

ion;

GD

RS

,Gen

eral

Dire

ctor

ate

for

Rur

alS

ervi

ces;

MoE

,Min

istr

yof

Env

ironm

ent;

loca

lgov

ernm

ents

,loc

ally

elec

ted

urba

ngo

vern

men

ts(m

unic

ipal

ities

and

villa

ges)

;MoI

,Min

istr

yof

Inte

rior;

NG

O,(

envi

ronm

enta

l)no

n-go

vern

men

talo

rgan

izat

ions

.N

ote:

Sur

face

wat

eris

used

only

for

irrig

atio

nan

den

viro

nmen

talp

urpo

ses.

Tab

le11

.2.

Key

acto

rsan

des

sent

ialb

asin

man

agem

entf

unct

ions

inth

eG

ediz

basi

n.

Allocatewater

Withdraw/distribute

Constructfacilities

Maintainfacilities

Monitorquality

Ensurequality

Plan(basin-level)

Constructfacilities

Operatefacilities

Maintainfacilities

Monitorquality

Enforcequality

Constructfacilities

Maintainfacilities

Monitorquality

Enforcequality

Plan(basin-level)


Ensurequality

Monitorquality

Maintainfacilities

Distributewater

Constructfacilities

Allocatewatera

Plan(basin-level)

Protectecology



legal framework for surface water rightsand only a rudimentary system of allo-cating access to groundwater, and, asa result, both are largely open-accessresources. Although nominal controlis stronger for groundwater than forsurface water, groundwater is presentlythe most stressed of the two resources.It is also the more desirable of thetwo, in part because of the poor qualityof Gediz surface water in the lowerreaches, but also because of the relativeease of access provided by ground-water. The current system of registeringgroundwater withdrawals does notappear to be effective at limiting over-drafting, which is occurring in certainsub-basins.

• Water quality monitoring takes place,but information is often not available inuseful forms to interested parties. DSIoperates 14 water-quality samplinglocations within the Gediz basin,sampling about 35 parameters on amonthly or bimonthly basis. The infor-mation collected, however, remainsas raw data in DSI files and is notgenerally used as a basis for policymaking or decision making for basinmanagement.

• A single actor, the provincial gover-nor’s office, is empowered to authorizeenforcement of breaches of wastewaterdischarge regulations by banningoffending practices or imposing fines.In practice, attempts to process finesor prohibit industrial activities oftenleads to confrontation between indus-trialists and public administrators,with the administrators generallylacking the political will and power tomake penalties stick. It is very commonfor files of violation reports to remainunprocessed in the offices of districtand provincial governors.

• Ensuring surface and groundwaterquality is not actively practised in theGediz basin. Ensuring water qualityinvolves conducting follow-up investi-gations of observed substandard waterquality to identify its sources andproposing remedies.

• No attention is currently paid to thequality of agricultural return flows. Itis sometimes presumed that theseflows contribute nitrates to ground-water and nitrates, phosphorus andorganic chemical residues, e.g. frompesticides, to surface water, but there islittle hard information on this, nor isany responsible party actively monitor-ing or assessing the quality and impactof agricultural return flows.

• Agricultural drainage infrastructure isinadequately maintained at present.Drain maintenance receives a lowerpriority from IAs compared withdelivery channel maintenance andDSI does not have an adequate budgetand equipment to fully maintain largerdrains. In addition, responsibility formaintaining main drains, which servemore than one IA, is under dispute byDSI and IAs.

• NGOs have no specific role in perform-ing essential management functions,but clearly have an important role toplay in overall basin governance.7 Thisleads to the discussion of enablingconditions in the following section.

11.3.4 Enabling conditions: whereproblems lie

The essential functions and actors’ rolesdepicted in Table 11.2 provide a static viewof responsibilities. Additional attributes ofwell-functioning basin governance systems


6 DSI is currently anticipating a new Gediz basin planning exercise. The previous plan was prepared 35years ago and updated in 1982, but covered only surface water. The groundwater section of the DSI regionaloffice is also planning a new groundwater survey in the near future.7 The term governance is used refering to the rules providing the context for multi-actor basinmanagement and the processes and activities engaged in by those actors operating within this set ofrules.



relate to its dynamics. These attributes,which we term enabling conditions, weredefined in Chapter 1 and are shown in Box1.2. A full analysis of these factors is wellbeyond the scope of this paper. A briefsketch of each in the context of the Gedizbasin will be attempted to illustrate theconcepts and indicate broad strengths andweaknesses.


This is perhaps the most important gap inthe current set of enabling conditions.Although some water users are well repre-sented, others are not, and in the arena ofpolitical give and take, those without repre-sentation become losers. Industrialists, forexample, have ample financial resources,are well organized and have ready access topolitical decision makers. Other irrigators,on the other hand, are not organized andenjoy representation only through theirlocal village heads. IAs are intermediate.They enjoy multiple connections to thelocal political establishment by virtue ofhaving a number of village heads and townmayors on their managing committees.In addition, they collaborate informally,sharing information and coordinatingactivities. IAs would benefit by establishingmore formal linkages among themselvesto allow a single spokesperson to representthem collectively in discussions over basinwater allocation, water quality standards,irrigation return flow restrictions, and soon. Other irrigators could affiliate with suchan association and, contributing financiallyto it, participate in its representationalbenefits.

The most serious failure of representa-tion currently relates to the environment.Although nominally represented by theMoE, the Ministry is still relativelyyoung and has yet to establish presenceand capacity in many areas. For example,it currently has provincial offices in onlytwo of the four provinces covered by theGediz basin. It also lacks sufficient budgetto perform its many duties fully. Moreover,as a government agency, it will always besubject to political pressures and pulls that

encourage or inhibit vigorous pursuit ofparticular water quality issues. Experiencefrom other countries has shown that strongNGOs rooted in civil society are essentialcomponents of the political system sur-rounding environmental issues. TheseNGOs serve as advocates for environ-mental values and for unrepresented futuregenerations. There are several groups withpotential to fulfil this role, but they pres-ently provide an ineffective counterweightto other interests.

Just as important as the existence ofrepresentational bodies is the need for arough balance of political power and influ-ence among various interests. When poweris one-sided, issues are not aired adequately,and decisions are also one-sided. A key tothe evolution of a suitable and balancedgovernance regime for the Gediz basin isfurther maturation of non-governmentorganizations and associations based incivil society which can advocate forenvironmental interests.


Stakeholders in the basin need to haveaccess to accurate and up-to-date descrip-tive information on water-related issuesin the basin, and open access to decision-making processes related to plans, regu-lations, violations and sanctions toparticipate effectively in basin governance.Data collection in the Gediz basin isincomplete, in that it falls short of the setof data required for competent professionalmanagement of the basin’s water resources.Data that are collected, however, areoften not widely disseminated or, worse,are sequestered and jealously guarded bythe collectors. This retentiveness may stemfrom feelings that the data are a source ofpower if they are not widely available, orperhaps from doubts about the reliability ofthe data. In a larger sense, though, it reflectsthe lack of a strong demand for the datafor use in basin planning and management.Once active planning and management pro-cesses are mounted, the pressures for accessto data will grow and push much of thisdata into the open.





Establishing appropriate institutions requiressuitable legal authority. This does notappear to be the most serious current prob-lem constraining the emergence of an effec-tive governance regime in the Gediz basin.Although improvements can be made, and anumber of legislative changes are proposedin the Eighth Five-Year Plan, the mostimportant short-term constraints appear tolie in other areas, such as balanced politicalpower and providing adequate resources.Over the longer run, however, the legisla-tion of a new legal basis for an effective sys-tem of water rights allocation, protectionand transfer will be essential.

11.3.4.4 Resources

Clearly, all four types of resources listedin Box 1.2 are needed for effective imple-mentation of basin management activities.In a number of the responsible public orga-nizations, they are inadequate. In some, thisconstraint may be relaxed by reassigningstaff positions from functions that have lostimportance to those that are increasingly so.Another potential problem is scattering ofresources among a variety of institutions,where each lacks a critical mass to be effec-tive. In a context of cooperation, it is notnecessary that resources be consolidatedunder a single administrative structure foreffective implementation. However, cooper-ation and coordination must be effective if adecentralized strategy is to be effective.

11.4 Toward Solutions

11.4.1 Problem recap

In this section, the hydrologic, policyand institutional problems identified inearlier sections are brought together andsummarized.

11.4.1.1 Poor surface water quality

The most pressing water-related problemfacing the Gediz basin presently is the poor

and deteriorating quality of its surfacewater. The deterioration results primarilyfrom the basin’s recent rapid growth inpopulation and the even more rapid growthin local industry, coupled with the wide-spread use of agricultural chemicals in ahighly productive agriculture. Failure tocontrol this growing problem at its severalsources leads to large requirements forin-stream flows for dilution – flows thatare then unavailable for other uses. Theproblem stems from several sources:

• Weak enforcement – first and foremost,it is the inability of the provincial anddistrict governors, appointed by theMinistry of Interior, to apply andenforce sanctions and penalties onviolators of wastewater dischargestandards that is responsible for thegrowing pollution problem. Althoughmonitoring could be improved andstandards tightened, the failure toeffectively enforce existing standards,based on existing information, sendsa powerful signal to polluters thatcompliance is unnecessary.

• Weak coordination – a second causeof deteriorating surface water qualityis poor coordination and cooperationamong the three separate agenciesresponsible for: (i) surface waterquality monitoring; (ii) wastewaterdischarge monitoring; and (iii) enforce-ment of standards. To some extent, thisis driven by bureaucratic tussling overturf. In addition, the failure of anyof the three parties to come forwardwith effective, inclusive and forward-looking leadership is a cause.

• Limited availability of data – becauseof restricted access, the debate onwater quality is poorly informed andemotional rather than scientific,making development of acceptableremedial measures difficult andcontentious.

• Haphazard monitoring of wastewaterdischarges – the most readily identifi-able and correctible causes of Gedizpollution are untreated or inadequatelytreated wastewater discharges from




industries, cities and towns. It is theresponsibility of the MoE to monitorthese discharges and report breaches ofstandards to the provincial governor.Limited staff, laboratory facilities andfunds currently forestall an adequatemonitoring programme.

• Inadequate funding for wastewatertreatment plants – inadequate fundinghas two components, capital andoperating expenses. Several fundingwindows are available to industriesand municipalities for investmentcapital but a shortage remains. Formunicipalities, there is little privatesector involvement in constructingand operating treatment facilities, incontrast to the case in many othercountries. A considerable share of thefunds made available by the State formunicipal treatment plant constructionhave come through the Bank of theProvinces, but with little expectation ofrepayment. This makes the disciplineof mobilizing private capital for suchinvestment difficult or impossible.Inadequate investment in wastewatertreatment by the industrial sectorrelates, in part, to the weak enforce-ment record of provincial governors.As long as the costs of complianceexceed the costs of non-compliance,this situation is likely to continue.

• Limited public awareness of theproblem – negative effects of surfacewater pollution include harm to publichealth, increased costs to other waterusers and negative environmentaleffects, particularly in the Gediz delta.Limited public awareness of theproblem and its impacts results inlimited public pressure and supportfor reforms, which in turn affects everysingle one of the factors outlined above.Causes of limited public awarenessinclude restricted public access towater quality data collected by govern-ment agencies, inadequate MoE effortsto publicize water quality problems,and the failure, to date, of environmen-tal NGOs to become effective advocatesand spokespersons.

11.4.1.2 Unknown groundwater quality

The extent of possible groundwatercontamination, particularly in the NefCreek watershed, is unknown. Due tocoarse alluvial soils and the extensive useof in-ground holding pits, some wastewaterfrom both urban and industrial sourcesmay go into groundwater rather than beingdisposed of as surface water effluent.Groundwater quality monitoring is notwidespread and the results not publiclyavailable, making it difficult to know ifsignificant degradation of groundwaterquality is occurring.

This is a significant gap because aquiferpollution is often more difficult and expen-sive to mitigate than pollution of surfacewaters. The possibility of contaminationgains added significance as a result ofthe almost total dependence of the basin’spopulation on groundwater for domesticsupplies.

11.4.1.3 Loosely controlled allocationamong users

Shallow groundwater is an open accessresource in the Gediz basin, meaning thatanyone with physical access to such watercan withdraw and use it. Deep groundwaterand surface water, once released fromDemirkopru Reservoir, share this openaccess characteristic, in part, as well.The result is that some legitimate needs,especially environmental needs, are inade-quately met, access of existing users isinsecure and it is difficult to transfer waterallocations among users in a rational way.Among the causes are the following:

• Inadequate representation – interestsof some users are not well representedin allocational planning and decisionmaking. The most salient example isthe environment, and, in particular, theneeds of the Gediz delta and its richcomplement of wildlife.

• Inadequate specification – while waterneeds in the basin for large-scale irriga-tion and urban use are generallyknown, present use and future require-ments for small-scale irrigation, the




burgeoning industrial sector, and theenvironment are not well specified interms of quantity, timing and qualityrequirements. This makes allocationdecision making difficult.

• Ineffective reporting and record keep-ing – while municipal extractions arereasonably well documented, indus-trial extractions from groundwaterremain largely a matter of conjecture.This makes evaluation of new requestsfor withdrawal permits difficult. Thereis likewise no cumulative inventory ofthe total number of permits issued, theagreed extraction rates and the depthfrom which water is extracted, render-ing this process even less rigorous andreliable.

11.4.1.4 Overarching future problems

In addition to the current problems affect-ing water allocation and basin governance,there are longer-term problems that willrequire more fundamental changes in laws,policies, institutions and practices. Two ofthe most significant are the following:

• Rudimentary water rights system – thecurrent national system of recordingand harmonizing rights to use waterdates from an earlier simpler time andis not well adapted to a water-shortenvironment. It does not provide secu-rity for present users, does not allow foror adequately protect environmentaluses of water, and does not provideincentives for economy of use or fororderly transfers among sectors.

• Lack of integrated planning –assessment of basin water resourcesis currently separated into ground andsurface water components. Becausethese interact in practice, the basinneeds to be understood as an integratedwater resource system. Also, becausewater quality influences the uses towhich water can be put, and givesrise to its own quantitative demands

for dilution flows, quantity and qualitymust also receive joint consideration.

11.4.2 Recent strides

The current situation in the Gediz basin,and at the national level, is dynamic andsomewhat fluid. Locally, a number of stepshave been or are being taken to improve theenforcement of water quality standards andprotect the natural environment. Many ofthese steps began with a Franco-Turkishbasin study of the Gediz in the mid-1990s.Although this study was intended to leadto various action programmes, the latterfailed to materialize because of lack of co-operation among the different institutionsinvolved.8 The study did raise awareness ofproblems and stimulate other initiatives,however.

In 1996, the provincial MoE officeconducted a study of polluting industries inIzmir province, which resulted in sanctionson 14 firms. In that same year, MoE officesin Izmir and Manisa began a 2-year surfacewater-quality sampling exercise in theLower Gediz.

In early 1998, three provincial gover-nors from the Gediz basin, together withMoE and other parties, convened a ‘coordi-nation meeting’ for the basin, which led tothe establishment of a coordinating commit-tee consisting of the directors of the threeprovincial MoE offices. More recently, thiscoordinating committee was transformedinto a permanent body named the Environ-mental Protection Service Association ofGediz Basin Provinces. This Associationwas officially authorized by the cabinet ofthe national government in December 1999,giving it a legal persona. This association hasa broader base than earlier initiatives, and,in principle, has considerable power. Sinceits creation, however, it has lain largelydormant due to lack of resources and theongoing bottleneck in the enforcement ofexisting standards and regulations.


8 The study was never formally ‘accepted’ by DSI or MoE.



At the national level, sentiment forchange is reflected in the recently publishedEighth Five-Year Plan covering the period2001–2005. The Plan recognizes the need forchange in the way water is allocated andmanaged. There is a commitment to intro-duce new water legislation that will coversuch issues as water rights, responsibilitiesfor water allocation, setting and enforcingenvironmental standards, and consolidatingthe position of the IAs. However, the Plandoes not specifically mention basin-levelwater management nor is there any provi-sion for establishing basin-level entities thatcould implement or coordination variousbasin-level activities. Reduction in supportfor utilization of fertilizers and agriculturalchemicals is a preliminary step towardaddressing non-point source pollutionproblems. Revising water and wastewaterstandards to comply with EU standards willraise the bar for existing polluters and thosein compliance alike.

The Plan does indicate that the privatesector will become more involved in variousaspects of water, adding an additional setof regulatory challenges to the existingsituation and raising the question of thesecurity of rights to water use by less well-represented groups. Until these new initia-tives are defined and brought into place,however, current institutional arrangementswill continue within the context of rapidgrowth in demand for water and increasingpressure on water resources from waste-water disposal from urban, industrial andagricultural users.

11.4.3 Strengths to build on

Although the problems faced are formida-ble, Turkey and the Gediz basin have anumber of strengths on which to build aneffective basin governance regime. Theseinclude the following:

• The premiere water resource agency inthe country, DSI, is responsible for bothground and surface water, providing astrong base for integrated treatment inthe future. This is not the case in many

other countries, where separate organi-zations are responsible. Moreover,handing over irrigation managementresponsibilities to IAs positions it wellto take on the role of basin planner andwater quality monitor for both groundand surface waters.

• Water quality is squarely identified asan important problem in the Gedizbasin. Moreover, while serious, it hasnot yet reached catastrophic propor-tions, offering a grace period in whichaction can be taken. Some actors in thebasin appear to be responding to thewarning signals.

• There is recognition that a numberof different actors must be involved insolving water quality problems in theGediz. It is important to transformthis recognition into effective waysof working together, rather thansquandering energy and resources inintra-governmental squabbles overbureaucratic turf.

• Likewise, there is recognition thatthere are multiple dimensions to waterresource management problems – dif-ferent disciplines, different interests,different uses, ground and surfacewater, quantity and quality, and so on.Recognizing this provides opportunityto develop an integrated approach tobasin water resource planning andmanagement.

• A new water law is under consider-ation, offering an opportunity to laylegal groundwork for effective basinmanagement and protection for theGediz and other water-short basins inthe country.

• There is a strong university-based sci-entific community, e.g. CEVMER andothers in or near the basin, providingcapability for applied problem-solvingresearch and, where needed, independ-ent scientific assessment.

• There are linkages with internationalinstitutions such as IWMI, which pro-vide access to international experiencewith basin governance problems. More-over, there is a healthy willingnessto look outside the country to the




experiences of others and a stronginterest in harmonizing standards,practices and procedures with those ofthe EU.

11.4.4 Challenges

A number of immediate solutions to prob-lems affecting governance and managementof water resources in the Gediz basin areself-evident from the listing of problems inSection 11.4.1. In this concluding section,we indicate four important longer-rangechallenges facing the basin. Addressingthese challenges effectively would go a longway toward putting into place a strong,dynamic and flexible system of basingovernance.

11.4.4.1 Systematize water rights

The current rudimentary rights systemcannot provide security and flexibility in anera of growing water scarcity. A new systemwhich is fair, flexible, and effective needsto be designed, based on both Turkish andinternational experience.

11.4.4.2 Build representational presenceand political muscle

Some basin water users, such as the naturalenvironment, are not well represented inwater-related discussions at present, andthere are severe imbalances of politicalpower among the various water users. Fairand equitable governance of basin waterresources requires that users and interestsbe represented in discussion and decision-making fora in a balanced way. NGOsrooted in civil society provide an importantvoice and advocacy presence for the envi-ronment, supplementing the efforts of theMoE. Their emergence as a political forcewill add balance to decision making and

pressure for effective enforcement ofsanctions for water quality violations.

11.4.4.3 Develop coordinating mechanisms

Alternative models for water basin gover-nance exist. At one pole is a comprehensivebasin authority, which concentrates power,responsibility and capacity to implementdirectly many basin management tasks. Atthe other pole is a coordinating committee,which simply provides a forum for discus-sion and voluntary coordination. Betweenthese poles, many variations are possible.One thing that is clear is that the presentsystem of compartmentalizing water quan-tity and quality, ground and surface water,and fresh water and wastewater is not aneffective base for the future. Mechanismshave to be developed for bringing thesecomponents together in a functional inte-grated system for planning, governance andmanagement.9

A useful first step would be thecompletion of an integrated assessment ofbasin water resources of all types and ofthe present and predicted demands onthose resources. High-level political com-mitment to such an undertaking and strongleadership would be essential. It is equallyimportant that this exercise not be carriedout by a single organization, but that itinvolves the various agencies, and thedifferent departments within agencies,which have mandates to address water-related issues in the basin. The reportproduced is only half of the desiredoutput of such a process. The other halfis the experience of joint action amongagencies and groups to implement thestudy and the creation of ‘ownership’ ofthe result by the various stakeholders.

11.4.4.4 Involve the private sector

In many countries, the private sector playsimportant roles in water resource manage-ment. Turkey is well embarked on the


9 One reason that Turkey is not pursuing the concept of basin-wide authorities at present is because of thedifficult issues posed by important trans-national river basins shared by Turkey and several of its neighbours. It issaid that the next 5-year plan may address this option.



devolution of responsibility for managingpreviously State-operated irrigation systemsto locally based associations. The privatesector can also play a major role in provid-ing safe drinking water and effective sanita-tion services to urban areas. Private sectorinvolvement has a number of advantages,including operational efficiency, ability tomobilize private capital and access to newtechnology. To attract such involvementwithout State guarantees of repayment,firms must have confidence that the princi-ple of payment-for-service will be honouredand supported by the involved govern-mental entities. Bringing in such privateparticipation would provide needed capitalfor wastewater collection and treatmentsystems and provide wider access to theseessential services.

References

DSI (1995) DSI in brief. Ankara.Fredericksen, H. and Vissia, R. (1998) Consider-

ations in Formulating the Transfer of

Services in the Water Sector. IWMI, Colombo,Sri Lanka.

Harmanciohlu, N. and Alpaslan, M.N. (1999)Institutional Support Systems Project, TheTurkey Activity, Interim Report I. CEVMER,Dokuz Eylul University, Izmir.

Harmanciohlu, N. and Alpaslan, M.N. (2000a)Institutional Support Systems Project, TheTurkey Activity, Interim Report II. CEVMER,Dokuz Eylul University, Izmir.

Harmanciohlu, N. and Alpaslan, M.N. (2000b)Institutional Support Systems Project, TheTurkey Activity, Final Report. CEVMER,Dokuz Eylul University, Izmir.

Kayam, Y. and Svendsen, M. (1999) Small ScaleIrrigation in the Gediz Basin. Report preparedfor the IWMI/GDRS joint research projecton Irrigation in the Basin Context. IWMI,Colombo, Sri Lanka.

Svendsen, M. and Nott, G. (2000) Irrigationmanagement transfer in Turkey: processand outcomes. In: Groenfeldt, D. andSvendsen, M. (eds) Case Studies inParticipatory Irrigation Management. WorldBank Institute, Washington, DC. Wall StreetJournal, 4 May.

Yavuz, H. and Cekmak, E. (1996) Water PolicyReform in Turkey. Report, DSI and IstanbulUniversity.




12 Managing River Basins: Lessons fromExperience

Mark Svendsen and Philippus Wester

12.1 Introduction

The case studies presented in the precedingchapters provide a rich vein in which toprospect for common threads and mean-ings. The six cases vary along a numberof dimensions and, at the same time, eachis, in some respects, unique. Moreover,the case studies employ similar, thoughnot identical, formats. This chapter willcompare the cases and attempt to draw con-clusions that will aid those establishingand strengthening basin managementcapabilities elsewhere.

We first ask what drives countriesto establish and maintain mechanisms forbasin coordination and management. Next,we describe the mechanisms that haveemerged in the six case study basins, con-trasting selected characteristics. Then welook at the conditions which enabled thesemechanisms to emerge, or which preventedthem from emerging. Finally, we summarizelessons learned.

12.2 What Drives Basin Management?

In a matter as complex as the emergence ofhuman institutions, it would be astonishingif a single variable explained the emergenceconclusively. In this case, we are not to beastonished, as a number of factors appearto come into play, while other plausible

influences do not seem to have a strongeffect. If the sample of cases were wider, nodoubt other factors would rise to the surfaceas well. Table 12.1 shows the six casesarrayed against a number of dimensionsrelated to the possible emergence of basinmanagement mechanisms.

At this point, we make the summaryjudgement that the first three basins – thosein France, California and Mexico – have putin place mechanisms to manage water at thebasin level, while the other three have notyet done so. The mechanisms in the firstthree basins are very different from eachother and operate with different degrees ofeffectiveness, but they do exist and operate.In the study basins in South Africa and Viet-nam, governments intend to develop suchmechanisms and have taken preliminarysteps in this direction but neither is yet oper-ating. In Turkey, there are as yet no clearplans to introduce formal basin managementpractices. We will examine the nature ofthese mechanisms subsequently, but for themoment it is interesting to explore some ofthe possible driving forces relative to thissimple ‘with’ and ‘without’ dichotomy.

The first dimension, the size of the basinin question, seems to have little or no effecton its propensity to develop managementmechanisms. It might be supposed thatlarger basins would have a stronger need fororganized management. Conversely, it couldbe thought that smaller basins would beeasier to manage and therefore more likely to




216 M. Svendsen and P. Wester

Bas

inC

ount

ry

Wat

ersh

edar

ea(k

m2 )

Nat

iona

lP

Cin

com

eaQ

uant

ityco

nstr

aint

Qua

lity

prob

lem

Gro

undw

ater

over

draf

tb

Dow

nstr

eam

envi

ronm

enta

las

sets

cS

tora

ged

Nes

teC

entr

alV

alle

yLe

rma-

Cha

pala

Olif

ants

Don

gN

aiG

ediz

Fra

nce

Cal

iforn

ia(U

SA

)M

exic

oS

outh

Afr

ica

Vie

tnam

Tur

key

10,0

0015

8,00

054

,000

74,0

0040

,000

17,0

00

$24,

080

$34,

280

$8,2

40$1

0,91

0$2

,070

$5,8

30

r r r q q

q r

–11

2%11

6% – –13

7%

? r r r q r

<40

%S

ubst

antia

l78

%10

5% 13%

128%

r,M

ajor

;q,m

inor

.a P

C,p

erca

pita

inco

me;

2001

PP

Pba

sed

onW

orld

Ban

kW

orld

Dev

elop

men

tInd

icat

ors.

b With

draw

alas

ape

rcen

tage

ofav

erag

ean

nual

rech

arge

.c S

ensi

tive

envi

ronm

enta

lare

aat

dow

nstr

eam

end

ofth

eba

sin.

d Con

trol

labl

est

orag

eas

ape

rcen

tage

ofav

erag

ean

nual

disc

harg

e.

Tab

le12

.1.

Sel

ecte

dba

sin

char

acte

ristic

s.



develop a management system. However,both the smallest watershed and the largestin the sample fall into the ‘with manage-ment’ category. Moreover, the two smallestwatersheds – those in France and Turkey –clearly are of great importance in theirrespective countries, and yet one has a verysophisticated management set-up while theother has virtually none.

Per capita income, which would indi-cate the ability of an economy to pay forbasin management services, is a potentialenabler of formal basin management. In thiscase, the sample of case studies offers somesupport for this hypothesis, since the twowealthiest countries clearly have the mostsophisticated basin management set-ups.However, the strong steps being taken byMexico contrast sharply with the virtualabsence of action on the part of Turkey,though both countries have similar incomelevels, and if regional income measureswere compared, the Izmir/Gediz area woulddoubtless show a considerably higher percapita income than both the Turkey andMexico averages.

A second connection between higherlevels of per capita income and basin devel-opment is that per capita income serves as arough proxy for a higher level of institutionalspecialization and diversification. This hasimplications for a variety of things, from datacollection and analysis capacity to the avail-ability of a large pool of trained enterprisemanagers. It is likely that this effect isalso operating here, though lower nationalincomes are certainly not an absoluteconstraint on the emergence of basinmanagement mechanisms.

It is probably significant that all threeof the ‘with management’ basins are experi-encing severe surface water availability con-straints, while in the three ‘without’ basinsthe constraint is less severe. The judgementmade here on ‘quantity constraint’ is not asrigorous as is the designation of a ‘closedbasin’ in the depletion methodology out-lined in Chapter 3. In particular, the ‘closedbasin’ definition effectively requires thatthe temporal variability in water supply hasalready been smoothed out through thecreation of in-basin storage, since this is the

heart of the stair-stepped water availabilityfunction in that concept. In this sample ofcases, the Lerma–Chapala has storage fornearly four-fifths of the annual runoff of thebasin, and the construction of additionalstorage in both the Neste and in California isseverely constrained by public policy andenvironmental considerations. It is thus rea-sonable to regard these basins as essentiallyclosed. The Oliphants and Gediz basinsboth have the capacity to store more thanthe annual basin runoff and have utilizedmost of the available supplies, but effectivedemand has not yet risen to the point wherethere is serious conflict over existingresources. As historical inequities areredressed in the Olifants, however, demandwill doubtless expand to exceed supply. Inthe Gediz, the agricultural demand for wateris declining as cropping patterns shift undereconomic pressure to less water-intensivecrops, reducing the pressure on the resourcefrom this source. The Dong Nai is lightlyused at present and has ample scope foradditional storage to smooth out seasonalpeaks in water supply. While seasonal con-straints on availability are approaching inthe Dong Nai, availability has not begun toapproach the utilizable limit.

So in all three ‘with’ basins, seriouscompetition over available basin water sup-plies appears to drive basin managementefforts, supporting the fundamental premisedeveloped in Chapter 2.

Water quality concerns have not, bythemselves, been sufficiently strong todrive the creation of basin managementmechanisms in the two basins where theyare most prominent – Lerma–Chapala andGediz. Similarly, groundwater overdrafting,which is serious in three of the basins, doesnot figure prominently in basin managementplans, where they exist. In fact, in the Gedizbasin where overdrafting is the most severe,basin management has received the leastattention among the six cases.

Four of the basins have importantenvironmental assets at their lower end.In California, the Sacramento–San JoaquinDelta is the focal point of environmentalconcern and controversy in the state, and inMexico the level of Lake Chapala is seen as a

Lessons from Experience 217



primary indicator of the success of manage-ment efforts. In the Olifants the worldrenowned Kruger National Park lies astridethe river on the South African border withMozambique, and in the Gediz the estuarywhere the river enters the Mediterraneanis an important bird sanctuary recognizedinternationally as a Ramsar site. These foursites are evenly divided between the ‘with’and ‘without’ basins.

From this discussion we can hypothe-size that absolute limits on available watersupply in practice provide the strongestimpetus for creating formal basin manage-ment mechanisms and that wealth andinstitutional resources in the larger societyare important enabling factors for their emer-gence. Where water supply expansion isfeasible, it is likely to be a preferred option tomore intensive management and conserva-tion of existing supplies, as seems to be thecase in the Dong Nai. Water quality and envi-ronmental protection problems do not seemto be sufficient motives, by themselves, tocause the emergence of comprehensive man-agement mechanisms. Equally, overdraftedgroundwater, while worrisome, is not astrong driver of management efforts. In part,this may be because groundwater over-drafting is a devilishly difficult collectiveproblem to get a grip on, and also because theproblem is ‘invisible’ and its consequenceslie primarily in the future.

12.3 Management Mechanisms

12.3.1 Configurations

12.3.1.1 Neste

The Neste is managed by a carefully struc-tured set of interlinked organizations thatcover all major basin management func-tions. It was first put into place about40 years ago. Management is focused onsurface water, since groundwater is unim-portant in this basin. The centrepiece of thestructure is a Regional Development Com-pany (RDC) – a state-owned firm holding anoperating concession for public water rights

and facilities that plays major roles in con-structing and maintaining facilities and inallocating and distributing water to users.Water users comprise towns and munici-palities, individual agricultural water users,and a few water users’ associations (WUAs).

A Basin Committee or ‘water parliament’made up of local and national authoritiesand water users develops long-term policiesand plans for the basin. An executive armof the Basin Committee, the Water Agency,collects water and pollution-based taxes andmonitors water quality, but does not take adirect role in water management.

Regulatory oversight and ‘policingpowers’ for water quality and environmentalprotection are provided by the Ministry ofEnvironment. The Ministry of Agriculture,which represents the State as the owner ofthe infrastructure, plays a regulatory role inmonitoring the care and maintenance givento the facilities by the concessionaire, theRDC.

The RDC is not involved in managing orregulating derivative water from municipali-ties and agriculture, which is handled bytowns and individual farmers, supervisedby the Water Authority and the Ministry ofEnvironment.

The organizational set-up thus contains:(i) a broadly based representative body, theBasin Committee, supported by its executivearm, the Water Agency, to make plans andset policy directions; (ii) a government-owned company to build, operate andmaintain hydraulic facilities; and (iii) publicregulatory agencies comprising the Minis-tries of Environment and Agriculture. Nota-ble by their relative absence are irrigationassociations providing irrigation service totheir members and representing their inter-ests and other pressure groups representingenvironmental interests. The State, on theother hand, has a strong presence in all of theimportant organizations in the managementset-up.

12.3.1.2 Central Valley

Institutional arrangements for managingCalifornia’s Central Valley water resourcesstand in marked contrast to the planned and




orderly set-up in the Neste. California’sinstitutions have grown up by accretionover more than 100 years of resource devel-opment into a boisterous confluence ofactors, laws, policies, plans and influence.Here two important categories of actors notfiguring importantly in the Neste arrange-ments, interest groups and the courts, areinvolved, and in each of the major catego-ries of involved actors – managers, serviceproviders, users, regulators, advocacygroups, elected officials and the courts– there are multiple actors.

Major water resource developmentinvestments made in California over the past100 years have blurred basin boundaries bymaking it possible to transport water acrossbasins from north to south, and from theColorado River basin west to southernCalifornia. This infrastructural develop-ment has multiplied the options available tomanagers, but at the same time, it has madetraditional watershed boundaries somewhatobsolete as delimiters of responsibility andauthority. Thus, while California has neverhad a basin management authority in thetraditional sense, such an authority mightwell have become obsolete by now had onebeen established.

Instead of a formally structured author-ity to manage the waters of the Sacramentoand San Joaquin Rivers, what controls israther a body of laws, rules and plans thatgovern water allocation and reallocation,water quality and return flows, and whichprotect the natural ecology. Changes inexisting practices and responses to changingconditions are contested by a variety ofinterest groups before decisions are reached.Federal and state courts usually serve as thefinal arbiters in these contestations.

The state Department of WaterResources (DWR) is the primary waterresource manager, planning water develop-ment and use in the basin, constructing andmaintaining major facilities, and protectingagainst floods. The United States Bureauof Reclamation (USBR) also constructs,operates and maintains major watercontrol and supply facilities as does alarge private consortium of irrigation waterusers.

The users of water in the Central Valleyconsist almost entirely of user-controlledwater districts. These districts are non-profit, quasi-municipal entities governed byfarmers, in the case of irrigation districts,and by urban residents in the case of muni-cipal water districts. The districts receivewater from the DWR or the USBR and, inturn, provide water services to their memberfarmers or households. These water districtshave been features of the institutional land-scape in the Central Valley for more than100 years, and continue to operate reliably.They are linked together in associations andare thus influential actors in basin watermanagement decision making.

Environmental protection and miti-gation is a critically important issue inCalifornia and nearly half of the state’sdeveloped water resources is allocated toenvironmental purposes. Regulating all usesand protecting environmental resources areseveral powerful federal and state agencies.The important state Water Resources Con-trol Board and the Regional Water QualityControl Boards are independent of the DWRand are insulated to some extent from otherorgans of state government as well. In addi-tion, influential mass-based environmentalaction groups actively lobby for particularcauses, join court cases, publicize andprovide their own analyses of water/environment interactions.

Groundwater is far more lightly regu-lated than is surface water in the CentralValley and is seriously overdrafted. No stateagency takes effective responsibility forregulating groundwater use and the legalbasis for such regulation is deficient.

The regional water quality controlboards enforce quality of return flows fromboth municipal uses and from agriculturerigorously, with support from the courts asneeded. Municipal return flows are becom-ing increasingly important as water sourcesfor agriculture, landscape and turf irrigation,aquifer recharge, and even new municipalsupplies.

In sum, institutions are in place to con-struct, maintain and operate water facilitiesin the Central Valley, to protect water qualityand the environment, to distribute water to




end users efficiently, to reallocate wateramong uses using both market and adminis-trative mechanisms, and to plan for futurewater supply and use. However, specializedbasin-based management organizations donot exist and are unlikely to emerge. User-governed water supply organizations pro-vide services to end users in most cases, andthey remain healthy and effective. The insti-tutional glue that holds the managementsystem together consists of an accumulatedbody of legislation and case law, legallyenforceable contracts, environmental rulesand regulations, and an inclusive planningprocess led by the state DWR every 5 years.Market mechanisms play an increasinglyimportant role in reallocating water to newand more highly valued uses.

12.3.1.3 Lerma–Chapala

Water resource management in Mexicooperates under the strong centralizedauthority of the National Water Commis-sion (CNA). CNA is a young organization,created in 1989, and operates under a com-prehensive water law promulgated in 1992.It is empowered to manage national waterresources, plan development and manage-ment, assign and register water rights,collect water-related taxes, and developirrigation and urban water supply systems.It operates through a set of 13 regional(basin) and 31 state offices.

Day to day management of irrigationsystems has been handed over to irrigationdistricts, similar to those in California, ina highly successful management transferprogramme dating from the early 1990s.

In the Lerma–Chapala basin, a BasinCouncil was established under the newwater law in 1993. A Governing Councilchaired by the CNA Director General is thetop level decision-making body, and a Moni-toring and Evaluation Group (MEG) acts asits executive organ. The regional CNA officeserves as a secretariat for the Council. Abasin-level users assembly brings togetherthe users of the basin and selects representa-tives from the various sectors to the Council.Council working groups address particularproblems. Although the Basin Council

provides a prototype vehicle to managebasin water resources, it has little decision-making power, since CNA remains responsi-ble for major decisions, including waterconcessions, the collection of water taxesand water investment programmes.

The Lerma–Chapala Basin Councildisplays some of the features of the FrenchNeste model, such as a users assembly, atop-level governing body, and strong Stateinvolvement and direction. At the sametime, the set-up in Lerma–Chapala lacks theprofessional management expertise of theFrench Regional Development Company, itscouncils and assemblies have little of thedecision-making authority of their Frenchcounterparts and they are heavily depend-ent on the CNA for authority, funds andtechnical expertise.

An analysis of actors and functionsshows that, as in California, most of theessential functions related to surface waterare covered, while groundwater is onlyweakly regulated. Planning, water alloca-tion and water distribution are done reason-ably effectively, while other functions arenot. Demand management and reallocationto Lake Chapala are particular weaknesses.Monitoring and ensuring the quality ofprimary water sources are relatively weakfunctions, and controlling groundwaterpollution is not covered at all, as is thecase in the Gediz basin in Turkey. Advocacygroups do not have an important voice inbasin management decisions.

In a backlash against restrictionsimposed by the Council on agriculturalwater use, agricultural interests haverecently coalesced to strengthen their voiceand influence. WUA presidents from fivebasin states have met together to coordinatetheir actions and have formed a new Work-ing Group for agriculture within the BasinCouncil. Interestingly, as in California andTurkey, limitations on surface water accesshave led irrigators to increase exploitation ofless regulated groundwater, exacerbating agroundwater overdraft problem in all threecountries.

To try to regulate this overdrafting,member states of the Council signed acoordinating agreement in 1993. To date,




though, there is little to show. While theState has prohibited the sinking of new wellsin the basin, groundwater development con-tinues apace, and overdrafting accelerates.

Another issue to be faced in thefuture is the right of access to derivativewater. With a major programme of con-structing wastewater treatment plants nowshowing results, raw wastewater streams,which were available for reuse in rivers here-tofore, are now controlled and improved inquality and can be diverted to new userswho are willing to pay for the right touse them. Complicated issues of rights toderivative water and compensation for lostuse rights await.

12.3.1.4 Olifants

Prior to majority rule in 1994, waterresources in South Africa were managed forthe benefit of the white minority under ahybrid system of riparian and appropriativerights. The national Department of WaterAffairs (DWA) was the pre-eminent organi-zation managing surface water and waste-water, playing a major role in planning,allocating, supplying and protecting surfacewater resources.

After 1994, water institutions, alongwith most other public institutions, under-went massive fundamental changes. The1996 constitution gave ‘basic rights’ status toaccess to water to support life and for per-sonal hygiene and a healthy environment.There followed two new water acts in1997 and 1998, which remapped the waterlandscape. Water was made a public trustand the Department of Water Affairs andForestry (DWAF) was made their custodian,with wide-reaching powers similar to thoseof the CNA in Mexico.

Following 1994 and in accord withthe new water legislation, institutions thatformerly served white farmers and commu-nities were modified and expanded to coverall citizens. Thus, for example, the formerirrigation boards are being converted intoWUAs and established wherever irrigationwill be practised across the country, arrange-ments similar to those existing in the USA,Mexico and Turkey.

In a major break with tradition, the man-agement of water resources was assignedto a set of 19 semi-autonomous CatchmentManagement Agencies (CMAs) to be createdin major water basins in the country. TheOlifants is one of these. The CMA mustplan a 5-year management strategy andwill progressively assume managementand monitoring functions delegated to it byDWAF. It will collect fees from water usersto cover its costs. The roles envisioned forthe CMA span a wide range, includingplanning and water allocation, wastewaterregulation, and monitoring and enforcementof water quality standards. It remains to beseen whether all of these functions, particu-larly the enforcement ones, are appropriateto such a pluralistic institution.

The process of forming a CMA in theOlifants basin has been deliberate, withmuch attention paid to obtaining wide-spread participation of basin stakeholders inthe formative process. However, with morethan 3 million people in the basin, securingthe vitally important participation of blackrural residents has been difficult. Otherinterests, such as industrial users, powergenerators and white commercial farmersare well organized, articulate (in English andAfrikaans) and politically adept, and tend toovershadow smallholder farmers and otherrural residents.

The process of creating locally account-able basin management organizations is inan early stage. It is underlain by an exem-plary body of legislation, a wealth of goodintentions and capable technical support,and faces an enormous challenge in bringingstakeholders together to govern the CMAin a fair and balanced way. The technicalchallenges of information collection andmanagement, planning, and reallocationalso lie ahead.

12.3.1.5 Dong Nai

Like Mexico and South Africa, Vietnamalso has a new and comprehensive waterlaw. Promulgated in 1999, the new lawprovides a forward-looking structure for thesector and mandates that water resourcesbe managed on the basis of hydrologic




boundaries rather than administrative ones.An apex-level National Water ResourcesCouncil was established in Vietnam in 2000to anchor the new set-up. While logical andconsistent with international thinking onthe topic, the notion of management at thebasin level across administrative boundariesruns counter to the administrative modelcurrently in place, wherein administrativeboundaries form the basis of a complex,interlocking and hierarchical administra-tive set-up extending downward from thenational level to provinces, districts andcommunes. Administrators at all of theselevels are currently involved in waterresource management, and it makes for avery complex situation.1 Unfortunately, too,the tendency is to see basin-level manage-ment as an overlay to this existing structurerather than as a new model that cuts acrossand supplants existing governmental unitsand authority, and this has the potential tomake the set-up even more complex.

As in South Africa, the government isonly just starting to implement this newmodel. A Planning Management Council forthe Dong Nai River basin was establishedin 2001 under the new water law. The 12permanent members of the council are allrepresentatives of national level ministriesand agencies, while the involved provincesare represented by non-standing membersdrawn from more junior administrativeranks. Moreover, provincial representationis drawn from the provincial Agriculturaland Rural Development Service and doesnot include the sectors of public health,domestic water supply, wastewater handlingand treatment, and environmental quality.It seems likely that the initial role of theCouncil will be planning new basin devel-opment projects, serving as an extensionof the Sub-Institute for Water ResourcesPlanning, which also acts as the Council’ssecretariat.

A parallel basin-wide coordinatingeffort has been ongoing for several yearsunder the sponsorship of the Ministry of

Science Technology and Environment.Participants are trying to establish a ten-province-plus-HCMC committee to monitorand manage water quality in the basin. Thiseffort differs from the MARD-based Plan-ning Management Council in that its pri-mary members are provincial governmentsand in its focus on water quality issues. Forthe time being, these two basin-level organi-zations appear positioned to evolve on par-allel tracks. In some ways, this is a positivedevelopment as it aids the emergence ofa provincially based capacity to engage inbasin management activities in addition tothe nationally based one. Coordination willbe required, and at some point, the twomechanisms may be merged, but this shouldnot be done at the expense of the more demo-cratic and representative process emergingat the regional level.

12.3.1.6 Gediz

The Gediz basin represents the negativeimage of basin management – obvious in itsabsence. The national water agency, DSI, issimilar to the Mexican CNA in its broad andcomprehensive powers to develop, manageand regulate the nation’s water resources.It has been in existence for 50 years andis a capable and professional organization.As with the CNA in Mexico, DSI has imple-mented a very successful managementtransfer programme over the past 10 years,shifting management responsibility for 80%of the irrigated area under its charge tolocally controlled Irrigation Unions.

However, despite its professionalcompetence and its success in devolvingoperational control of most of its irrigationsystems to water users, DSI has not yet re-interpreted its mandate to include basin-level water resource management. As aresult, there is a sizeable gap in institutionalcoverage of basin management functions.

This is manifested in very limitedplanning of surface water at the basin leveland virtually none for groundwater. There


1 The actors and functions matrix for the Dong Nai basin is considerably larger than those for each of theother cases studied.



is no integrated plan which considers bothground and surface water availability, nordoes existing planning consider water qual-ity, wastewater disposal, current and pro-jected land use, anticipated future demandand return flows, or projected futurequantity and quality of water resources.

Water is allocated, in practice, by avariety of agencies and users operatingindependently of each other. There is nonational legal framework for surface waterrights and only a rudimentary system of allo-cating access to groundwater. As a result,both are largely open-access resources.Although nominal control is stronger forgroundwater than for surface water, ground-water is presently the more stressed of thetwo resources. It is also the more desirable ofthe two, in part because of the poor quality ofGediz surface water in the lower reaches, butalso because of the relative ease of accessprovided by groundwater. The current sys-tem of registering groundwater withdrawalshas been no more effective at limiting over-drafting than practices in California andMexico.

Water quality monitoring takes place,but information is often not available inuseful forms to interested parties. Theinformation collected remains as raw data inagency files and is not generally used as abasis for policy making or decision makingfor basin management.

Despite serious quality degradation ofsurface waters in the basin, there is littleenforcement of water quality regulationsbecause a single official, the provincial gov-ernor, is empowered to authorize enforce-ment. Although a number of environmentalNGOs exist in Turkey, these have not yetdeveloped sufficient confidence and inde-pendence to lobby effectively for enhancedenvironmental protections.

In sum, the need for basin-level manage-ment in the Gediz is acute, and adequateauthority and professional competenceexists within DSI to begin this process.Bringing the process to completion willrequire a new national system of water rightsto safeguard existing users and prevent over-allocation; however, there are many inter-mediate steps related to integrated planning

and data collection and dissemination thatcould be taken now. This has not been ahigh priority for DSI, perhaps because ofa primary focus on extensive irrigationconstruction activities elsewhere in thecountry.


Additional attributes of well-functioningbasin management systems relate to dynam-ics. We term these attributes enabling con-ditions (see Chapter 1). Enabling conditionsare features of the institutional environ-ment at the basin level that must be present,in some measure, to achieve good gover-nance and management of the basin. Theseattributes are not specific to any one actor,but apply to all actors and their interactionsand comprise necessary, but not sufficientnormative conditions for success.


An important political attribute of an effec-tive basin management system is adequaterepresentation of interests. Equally impor-tant is the need for a rough balance of politi-cal power and influence among the variousinterests. When power is one-sided, issuesare not aired adequately, and decisions arealso one-sided.

In two of the study cases, a wide array ofinterests represent themselves in basin man-agement discussions and decision making.In California, the various principal sectors –agricultural, municipalities and the envi-ronment – are well organized and fundedand represent their members in a varietyof different forums. Involved entitiesinclude associations of irrigation districts,associations of municipal water districts,federations of farmers, state and federalmanagement and regulatory agencies, andenvironmental NGOs. Following debate,decisions are made by the state legislature,the US Congress and regulatory authorities.Disputes are adjudicated by state and federalcourts. Voluntary agreements among partiesare often reached after negotiation andformalized in legally enforceable contracts.




In France, in a simpler set-up, the BasinCommission or ‘water parliament’ serves asa discussion and decision-making foruminvolving the major actors.

In three other cases, those of Mexico,South Africa and Turkey, some interests arewell organized and well represented, whileothers are not. For example, in the Gediz,local industrialists form associations, whichpromote their interests with DSI and the gov-ernment. Farmers, though more numerous,are less well organized and funded, and donot speak with a unified voice. However,Irrigation Unions recognize their disadvan-tage in this regard and have begun to discusscoordination and possible federation amongthemselves. In Mexico also, the success ofWUAs in five Lerma–Chapala basin states informing an irrigation working group withinthe Basin Council and in taking a commonposition against further reductions in theirwater allocations shows their increasingsophistication and an evolution towardgreater balance in representation. In bothcountries, the environment is poorly repre-sented in debates over allocation and pollu-tion because of the relatively weak status ofthe national environmental agencies and therelative absence of effective environmentaladvocacy groups based in civil society.Neither country appears to be taking aproactive stance to rectify these imbalances.

In South Africa, the cleavages are quitedifferent. Here, the dualism resulting fromthe apartheid legacy has created a situationwhere wealthier white segments of all sec-tors of the water economy are well organizedand represented, while the generally poorerblack segments of the same sectors are not.Industrialists and commercial farmers arebetter endowed to participate in consulta-tion processes and are concerned about con-tinued access and water quality issues. Theenvironment is well represented and sup-ported by strong provisions in the NationalWater Act mandating priority allocations tomeet environmental needs.

Administrative structures serving blackSouth Africans in the former homelandareas were extremely weak and have beendisbanded. These populations currentlyhave to grapple with major ideological

differences between traditional triballeadership and the new democraticallyelected representation. This disparity in rep-resentation and influence is rightly recog-nized by policy makers as the pre-eminentchallenge in developing democratic andrepresentative basin management agencies.

They recognize that the new systemmust create mechanisms through whichwater users across the board can makethemselves heard and understood, enablinggradual and systematic redress of racial andgender inequities while ensuring a securebase for economic growth. Lack of access –both through lack of rights and lack of infra-structure – are priority issues for the ruralpoor in the Olifants basin. At the same time,however, adequate representation is hard toachieve: large numbers of people live inremote areas, excluded through the costof transportation and an absence oforganization. The government, throughDWAF, is experimenting with a variety ofapproaches to achieving better balance inrepresentation.

In Vietnam, the model is quite different.Power rests overwhelmingly with the Stateand the Party, and there is scant scope forrepresentational associations based in civilsociety. It is assumed that the various organsof government – ministries of agriculture,environment and energy, and the multitudeof other involved government bodies – canadequately represent the interests of allinvolved stakeholders. Moreover, nationallevel agencies tend to have much strongerrepresentation than provincial and locallevel ones. As a result, basin forums tendto involve only public sector actors andare dominated by those representing State(national) agencies. The result is an adminis-trative technocratic approach to planningand management which does not necessar-ily reflect the values of provincial authori-ties or private economic actors active in thevarious sectors.

In sum, in the five multi-party democra-cies, representation of the variety of publicand private interests involved in basinplanning and management is either presentor evolving. The modes and mechanismsfor this vary widely, as do the present




balances among interests, but the involvedstakeholders themselves, and, in most casesthe governments, recognize the importanceof this process. In the sixth case, State organsrepresent all of the involved interests,resulting in a more centralized, publicpolicy-oriented and technocratic approachto planning and management.


Another essential enabling condition is thepresence in the public domain of accurateand up-to-date descriptive information onwater in the basin, along with open andaccessible decision-making processes relatedto plans, regulations, violations and sanc-tions. The first of these stipulations requiresthat information on basin water allocations,reservoir positions, groundwater elevations,water quality conditions, available resourcesand so on be a part of the public record.This disclosure condition applies to intra-and interdepartmental information relation-ships as well as to those with the generalpublic. The second stipulation, transpar-ency of public proceedings, is similarlyessential to fair democratic processes.Rent-seeking behaviour requires darknessand privacy to thrive, and conducting regu-latory processes in full view of the public isan effective antidote to such practices.

With regard to data openness, the con-trast between California on the one hand,and Turkey and Vietnam on the other isstrong. In California, law requires suchopenness and the many players in basinmanagement demand it. Information isreadily available in paper form and on theInternet. In the other two cases, informationis often regarded as a source of power andis jealously guarded by the collecting unit.There may be a reluctance to share evenwithin the same agency or ministry. Suchsequestering of information may be rational-ized in terms of state security, but legitimatejustification for this is usually lacking, andthe real reasons generally relate to control,power and perhaps to avoid conflict orembarrassment.

Generating improved information onbasin water resources and making it

available to all interested parties can bean early step in establishing basin planningand management mechanisms, and can beapplied in both democratic and technocraticenvironments. Improved information gath-ering and dissemination can help to easetensions between involved parties, evenwhere distrust is extreme. In the Kura–Araksbasin of the South Caucasus, for example,where the three riparians of Armenia,Azerbaijan and Georgia agree on little else,they have recently been willing to collabo-rate on data collection and sharing relatingto river hydrology.

In South Africa, the problem isdifferent. One of the lessons of the water lawconsultations was the importance of trustedinformation as a basis for consultation andnegotiation. Good information is crucial todelineate areas of agreement and disagree-ment and to structure and inform debate, butof little use if the source is doubted. Equally,good information becomes useful in negotia-tion and decision making only when it isaccessible by all interested and affected par-ties. South Africa is well equipped to use themost modern techniques for data gathering,storage and knowledge creation, but faces amajor challenge in presenting informationin a meaningful way to the wide range ofinterests in the sector. Those most in needof water for basic and productive uses arepoorly equipped to access and interpretinformation from the national data collec-tion systems. Good information has thepower to defuse unnecessary tensions,while misinformation and lack of informa-tion foster conflict. In Mexico also, goodinformation is available, but other actors inaddition to CNA, which collects most of it,need to learn to assess, analyse and employit. Ultimately, private stakeholders maybegin their own programmes to collectand analyse key information as a check oninformation from official public sources.


Mexico, South Africa and Vietnam all haverecent comprehensive water laws to serveas a basis for water resource management.The other three case study countries rely on




a pastiche of older laws to guide and struc-ture the sector. There does not seem to beany particular relationship between theexistence of a modern comprehensive waterlaw and successful basin managementpractices. In California, for example, somelegislation governing the sector, such asthat governing water rights, is more than100 years old, while laws dealing withother aspects are quite recent. In addition,court decisions in California may have theeffect of modifying law and creating newprinciples. For example, the public trustdoctrine was added to the body of Califor-nia law by a court decision in 1983. SouthAfrica achieved this same effect throughprovisions in the new water law.

In Turkey, the 1954 law establishing thenational water resource agency consolidatedbroad authority over water resources in thatagency, which has allowed it to respond tomany of the changing needs it has encoun-tered over the past 50 years. For example, itwas able to use another existing law autho-rizing local governments to form consortia todeal with issues cutting across their bound-aries to establish effective Irrigation Unionsin the 1990s. However, the absence of alegally sanctioned system of water rights,poorly designed enforcement powers forwater quality standards, and reform of thestructure of the Irrigation Unions all requirenew legislative action, and the absenceof such reform legislation hamstrings thedevelopment of basin water managementmechanisms.

In Vietnam, the principles embodied inthe new water law still await implementingrules and regulations to translate them intoactionable practices. Effective implementa-tion also requires establishing administra-tive systems for managing water permits,collecting and consolidating data, and thelike. In a larger sense, too, to be fully imple-mented, the tension between principles inthe new law and the prevailing administra-tion culture will have to be resolved.

The important lesson in this is thatdeveloping and promulgating a new andcomprehensive water law to subsume andreplace existing laws is not a necessary con-dition for effective basin management. This

step may be useful when a major transforma-tion is contemplated, as it was in SouthAfrica and Mexico, but it does not appearessential. What is important, however, isthat the legislative basis for water resourcemanagement be kept current through regularlegal amendments, updated administrativerules and reinterpretations of current rulesand practices. Where this has not happened,as in Turkey, emergence of new basinmanagement institutions will certainly beconstrained.

12.3.2.4 Resources

Resources needed to practise basinmanagement can be classified as human,financial, institutional and infrastructural.The human resources needed are thoseindividuals skilled in engineering, hydrol-ogy, water chemistry, economics, massorganization, management and otherdisciplines needed to plan and managethe use of basin water resources. In all ofthe cases studied, professional skills in thebasic disciplines are probably available,though particular skills, professionalmanagement expertise for example, may bein short supply in particular countries. Inaddition, specialized training will often benecessary to build on basic disciplinaryexpertise. This will be particularly true incountries just beginning to practise basin-level water management. To a degree also,the availability of these skills to the basinorganizations will depend on the financialresources available to hire them, and tocompete with other potential employers incases where particular skills are scarce.

As noted earlier in this chapter, thewealth of a society will help determinethe financial resources available for basinmanagement tasks. Most funding is usuallydrawn from the public treasury, since taskssuch as data gathering and analysis, issu-ance and recording of water and wastewaterpermits, monitoring, and enforcement ofrules and standards are public sector tasks.However, it is also important that someresources be available from private usergroups as well to preserve their independ-ence. In France and California, for example,




representational resources for different usersectors are drawn from the members of theseorganizations. In Mexico and South Africa,this is happening for some, but not all, usersectors. Within the public sector as well,it is important that resources be availablethrough the budgets of different sectoralministries and agencies, rather than beingrouted through a single ministry budget, sothat potentially divergent interests such ashydropower generation and flood controlor agriculture and environment canbe represented with some independence.On the other hand, a potential problem isscattering of financial and human resourcesamong too many organizations, where eachlacks a critical mass to be effective, as may bethe case in Vietnam at present.

Institutional resources are those neededto translate policy and law into action.They include public organizations taskedwith collecting and processing data andmonitoring and enforcing standards, userorganizations to represent sectoral interests,and coordinative and adjudicatory bodiesto knit sectors together and reach decisions.In a broader sense, institutional resourcesextend to cover basic social institutions suchas rule of law, enforceability of contracts,right to organize, and so on. These are lessamenable to change than more specificinstitutions are, such as a system of waterrights, but are important determinants ofthe type of basin management set-up chosenand the expectations held for it. The stock ofinstitutional resources varies widely amongthe case study countries.

Infrastructural resources are those thatfacilitate control of the water resource. Theyinclude plumbing such as dams, diversions,canals and reservoirs, as well as measure-ment and data management and processingtechnology. One of the most important ofthese characteristics is the amount of con-trollable storage available in a basin, bothabove and below ground. This characteristicinteracts with basin hydrology, since snow-fed rivers such as those of the Central Valleywill need less storage, other things beingequal, than those fed by monsoonal rainfallsuch as Dong Nai. Without storage, optionsavailable to managers are much more

limited. The storage estimates shown inTable 12.1 for the various basins show thewide range of controllable water supplyavailable to managers in these basins andsuggest different strategies for dealing withbasin development and management.

In addition to the ability to store water,infrastructural resources also allow transfer-ring water from one basin to another inresponse to politically expressed demandsand relative water availability. This capacityis considerable in the Central Valley andNeste, and is also present in Olifants andDong Nai. This capacity often, but notalways, comes in later stages of basindevelopment.

12.4 Conclusions

Conclusions from any set of case studiesalways depend on the cases making up theset, the information included in them andthe frame of reference employed in analys-ing them. Recognizing these limitations, weoffer the following conclusions drawn fromour interpretation of this particular set ofcases and tempered by our experience inother basins around the world:

• Constrained surface water availability,coupled with an absence of readilyaccessible groundwater to take upslack, is the most powerful driver ofconcerted water resource managementat the basin level. Other needs whichconstitute potential drivers of basinlevel management – overdraftedgroundwater, water pollution and envi-ronmental degradation – are usuallyless potent than constrained wateravailability in fostering the necessaryinstitutions. In part, this is becauseexcess surface water can compen-sate for water quality degradation bydiluting waste streams and substitutefor constrained groundwater, thoughperhaps at higher cost.

• High-quality basin management ser-vices are expensive, and wealthier soci-eties are better able to bear these coststhan poorer ones. Moreover, wealthier




societies often have a broader rangeof human and institutional resourcesavailable internally that are needed inbasin planning and management thando poorer ones. This does not preventthe practice of basin management inany society, but may influence theform that management takes and theobjectives established for it.

• Basin management, as conceived here,is better described as a political processthan a technical one. This observationstems both from the concepts of ‘rulesin use’ and ‘water as a contestedresource’ developed in Chapter 1, andfrom practices, organizational struc-tures, and interactions described in thesix case studies. Going a step further,basin management is, in essence, ademocratic process, rather than anautocratic or administrative one.Although technical processes of watermeasurement, demand estimation,modelling, water quality sampling, andso on are involved, the core processesof basin management involving makingfundamental decisions on water alloca-tion and quality on the basis of particu-lar economic interests, social values,priorities and expectations – decisionmaking that is inherently political. Inthis conception of basin management,pride of place belongs to the gover-nance of basin management processes,in turn placing high value on accessto information, transparency, account-ability, balanced representation, andrule of law. Introducing basin manage-ment processes into a social contextwhich also values these characteristicsof good governance is consequentlymuch easier than introducing theminto one which does not.

• Management institutions do not haveto be embodied in a unitary organiza-tional structure to be effective. A moredispersed set of organizations can alsomanage a basin effectively if they areknit together with suitable processes,rules and other institutions, and if theyprovide for revision and updating. Thatsaid, when a country chooses to make a

radical change in its organizationalset-up for managing water resources,a comprehensive reassessment andrestructuring may be an appropriateway to understand and introduce theinterlinked set of changes that such amajor reorganization implies.

Finally, we offer for consideration a set offour broad institutional changes that seemto take place as a basin closes. The first ofthese is expansion in the number of publicand private sector actors involved in basinplanning and management. This expansionincludes service providers for water usesof growing importance, such as municipali-ties, industries, environmental agencies;citizens groups; and other regulatory bod-ies. In addition to growing scarcity, thisexpansion is occasioned by shifts in thecomposition of demand for water, deterio-rating water quality, growing public con-cern over environmental quality, andgrowth in public wealth and increased pro-fessionalism in resource management as aresult of overall economic development.

A second change is that organizationsassociated with basin planning and manage-ment become more specialized and differen-tiated. One important shift that typicallytakes place is the separation of basinfunctions into three broad categories:

• Regulation and standard setting;• Resource management;• Service provision.

Regulation and standard setting are carriedout in the public interest and are necessar-ily functions of government. Resourcemanagement tasks include allocating andprotecting water resources, and must bedone even-handedly and responsively toboth clients and regulators. Resourcemanagement is generally done by publicagencies, or by hybrid public/privateorganizations. Water-related services suchas irrigation and drainage or municipalwater supply are generally provided bya special-purpose organization operatingunder private-sector-like incentives toprovide services to clients effectively andefficiently. Experienced observers argue




convincingly, in fact, that such differentia-tion is a necessary precondition of afunctional and efficient basin manage-ment structure as closure is approached(Millington, 2000; van Hofwegen, 2001).

The third shift is greater involvementof civil society in basin planning and man-agement. With rising standards of living,urbanization, and in the face of progressiveenvironmental deterioration, elements ofcivil society increasingly mobilize to protectthe environment and reverse cumulativeadverse effects of resource exploitation andindustrialization on it. By organizing intoassociations, societies and other groups, thisperspective gains entry into water resourcedecision-making processes.

The fourth shift is toward a broaderrange of disciplines playing roles in basinplanning and management. A typicalprogression begins with engineers andhydrologists and then expands to includeeconomists, agronomists, management spe-cialists, ecologists, public health specialists,water chemists and others. This placesa growing premium on communicationand sharing of information, both amongspecialists and with civil society as a whole.

These changes seem to occur in a varietyof contexts as water resource exploitationprogresses, but they are in no way ordainedby basin closure. Basin closure is just one ofa set of changes that proceed simultaneouslyas societies grow and develop. Thesechanges are interlinked, but not rigidly so.Corollary changes in a society include

expanding exploitation of all availableresources (including water); rising agri-cultural productivity; expanding industrialand service sectors; a growing population;urbanization, rising per capita incomes;rising educational levels; burgeoning wastedischarges to land, air and water; andnumerous others. Growing scarcity of watercreates pressures and incentives for institu-tional change, but the actual manner andpace at which these institutional changestake place are particular to the prevailingsocial, political and cultural environment.Moreover, in the absence of leadership anddirection, countries may ‘sleepwalk’ intopotential disaster as basin closure progres-ses, but without appropriate institutionalresponses. Moreover, the pace of institu-tional change will accelerate as closureapproaches, i.e. more rapid change will tendto occur between stages 2 and 3, utilizationand reallocation, than between stages 1 and2, development and utilization.

References

Millington, P. (2000) River basin management;its role in major water infrastructureprojects. Prepared as an input to the WorldCommission on Dams, Cape Town.

van Hofwegen, P. (2001) Framework for assessmentof institutional frameworks for integratedwater resources management. In: Abernethy,C. (ed.) Intersectoral Management of RiverBasins. IWMI, Colombo, Sri Lanka.




13 Providing Irrigation Services inWater-scarce Basins: Representation

and Support

Philippus Wester, Tushaar Shah and Douglas J. Merrey

13.1 Introduction

As water becomes scarcer, competitionintensifies and its value rises; irrigatedagriculture must change profoundly to usewater more productively. Compoundingthis challenge is the widespread poverty inriver basins in developing countries and thepressure this creates to reallocate water tothe poor for productive uses. This chapterdraws from the river basin studies in thisbook, along with supplementary materials,to outline the challenges facing irrigationservice provision in closing river basinsin developing countries and the implica-tions this has for institutional supportsystems. It also addresses the adequaterepresentation of the large numbers ofsmall-scale water users characteristic ofdeveloping country river basins, alongwith other interests, in basin-level decision-making institutions.

In semiarid regions, irrigation is gener-ally the largest water user in the basin. In thepast 50 years, vast sums of money have beeninvested, by both governments and farmers,to construct new irrigation systems or reha-bilitate deteriorating ones. The irrigatedarea nearly doubled from 140 million ha in1960 to 275 million ha in 2000. In manycountries, this occurred under the banner ofintegrated river basin development, aptlysummarized as ‘the orderly marshalling of

water resources of river basins . . . to promotehuman welfare’ (UN, 1970, p. 1, quoted inBarrows 1998, p. 172). In the past 15 years,governments have shifted away from irriga-tion infrastructure development for povertyreduction to improving irrigation manage-ment, frequently through institutionalreforms such as irrigation managementtransfer (IMT). An overriding concern inirrigation reforms has been to reducegovernment subsidies to the operation andmaintenance of irrigation systems, whilethe focus on poverty reduction through thedevelopment of large-scale irrigation haswaned. Only recently has a paradigm shiftstarted to occur in water management, basedon the recognition that providing access towater for drinking and growing food, eradi-cating poverty, and stopping groundwateroverexploitation are central challenges inriver basins in developing countries thatrequire new ways of thinking.

In the developing world, institutionalarrangements for basin management are onlystarting to take shape and are still ratherweak at a time when the management ofriver basins is becoming more difficult andchallenging (Vermillion and Merrey, 1998).In part, this is due to the widespread trendto decentralize the management of waterservices, such as with IMT or to privatizeurban water supply, but also becauselocal institutions – such as water users




associations (WUAs) – are young, weak anddependent on agencies for resources as wellas technical support. While having mixedoutcomes in terms of accountability and costeffectiveness at local levels, decentraliza-tion leads to a multiplicity of organizations,often with competing interests, and makeswater management in river basins reachingclosure more challenging. Although a largebody of literature has developed on riverbasin management, relatively little attentionhas been given to the needs of locally man-aged irrigation and how to address waterdeprivation in closing river basins in devel-oping countries. The closure of river basins(see Chapter 2) in semiarid regions posessignificant challenges for pro-poor waterpolicies. The following issues stand out:

• The over-exploitation of primary watersources (waters tapped from rivers,lakes and aquifers, i.e. ‘blue water’)leads to environmental degradationthrough the destruction of aquaticecosystems, depletion of aquifers andgeneration of polluted wastewaterflows (both industrial/urban effluentsand agricultural drainage effluents). Inclosed river basins, the only way toreverse these trends is to consume lessprimary water and to make judicioususe of derivative water (municipalwastewater, industrial discharges andagricultural return flows).

• Alleviating poverty through the creationof new hydraulic property (Coward,1986) becomes very difficult as primarywater sources are already fully commit-ted, and frequently under the controlof the relatively better off. Creatingnew entitlements for the poor musttherefore be sought in renegotiatingrights to primary water. Equallyimportant is the need to increase theproductivity of ‘green water’ (waterstored in the soil profile) throughrainwater harvesting and anti-erosionmeasures.

• The lack of possibilities to develop newwater supplies and perceptions thatagriculture is a ‘low-value’ use of waterleads to increasing inter-sectoral water

transfers. These are frequently one-waytransfers from agriculture to industryand domestic use, as well as intra-sectoral transfers in agriculture tohigher-value crops, usually grown bycommercial farmers.

• Without clear water rights and effectiveenforcement, it is relatively easy forpoor people, such as smallholder irri-gation farmers, to lose access to waterfor production due to these transfers.Consequently, poor farmers increas-ingly have to turn to derivative water astheir only source of water.

• Before IMT, most irrigation schemeswere under government managementand to a degree, protection. After IMT,this protection may disappear, and thelocal irrigation management entity mayfind itself under-represented at higherlevels (Svendsen et al., 2000).

To address these issues, concerted changeat different levels, building on thestrengths of government, civil societygroups, popular movements and communi-ties, is necessary. Such change needs tofocus on the creation of interlocking institu-tional arrangements at the local, meso andmacro level to manage water in a sociallyjust and equitable manner that meets theneeds of the poor and ensures the sustain-ability of locally managed irrigation. Ofspecial concern is the need to represent andprotect the interests of water users that areat risk, such as the large number of small-scale water users that fall outside the ambitof formal water management. Furthermore,the ability of groundwater irrigators to makemincemeat of any effort towards orderlyresource management at the basin levelneeds to be taken into account. Previouschapters of this book have analysed theinstitutional challenges that arise when theutilization of water nears or exceeds theannual renewable water in a river basinand how various countries are dealing withthese challenges. In this final chapter, weendeavour to draw lessons from these expe-riences as they relate to irrigation serviceprovision and poverty alleviation. Onefundamental premise of the chapter should




be stated explicitly: in poor developingcountries, it is desirable to find ways topreserve and improve irrigated agriculture,as this is a major source of livelihoodsand food security for which there are fewsubstitutes.

13.2 Challenges Facing Locally ManagedIrrigation in Closing River Basins

Between the 1950s and 1980s, vast invest-ments were made in poor countries toincrease the area served by large-scaleirrigation systems dependent on surfacewater. A central purpose of this effort wasto reduce poverty and to attain nationalfood self-sufficiency. However, starting inthe 1980s, serious concerns were raised thatpublic investments in large surface irriga-tion were not sufficiently benefiting poorfarmers and that the government agenciescharged with irrigation management wereperforming poorly, especially in recoveringcosts from farmers. To reduce the burden onthe public purse, and under the influenceof the structural adjustment programmes ofthe 1980s, the role of government inirrigation management in many developingcountries started to change, with an empha-sis placed on transferring managementresponsibilities and financial obligations tofarmers. Much of this reform is captured inthe phrase ‘irrigation management transfer’,though the actual reforms are broader than‘simply’ transferring water managementresponsibility from government to waterusers. The widespread trend to devolve irri-gation management to farmer organizations,coupled with the rapid increase in ground-water irrigation and water harvesting andlocal storage means that currently a largemajority of the world’s irrigation is locallymanaged. A distinguishing characteristic oflarge portions of locally managed irrigationis that it has thin or no contact with formalresource governance structures. Examplesinclude the 20 million people pumpinggroundwater in South Asia and the commu-nities that depend on South India’s 300,000tanks or China’s 7 million ponds.

The chapters in this book haveshown that there are clear stages to riverbasin development related to the changingpattern of demand for water over time.A fundamental point for understanding thechallenges facing locally managed irrigationin water-scarce basins is that irrigationreforms such as IMT were not enacted todeal with river basin closure but ratherto reduce government expenditure. Thus,management transfer in many countries isoccurring at a time when water resourcesin river basins are becoming increasinglyscarce and the subject of inter-sectoral com-petition. This places an extra burden onfarmers, who, while having had little timeto develop their associations, immediatelyneed to start focusing on the river basin levelto secure and retain an adequate share ofwater. As many governments have beentransferring irrigation management to userorganizations rather hastily, the world’sirrigation is increasingly likely to bemanaged by ill-formed and ill-prepareduser organizations. Where systematic effortsare not made to support newly establishedWUAs and secure their access to water, thesustainability of smallholder irrigation itselfmay come into question (cf. Shah et al.,2002). At the same time, burgeoning infor-mal irrigation economies, unrestrained bynational policies and government regulatorystructures, defy initiatives for more orderlybasin-level water allocation and manage-ment, and are instead driven by their ownrules and internal logic.

The challenges facing locally managedirrigation in water-scarce basins are thustwofold: internal and external. This chaptermainly deals with the external challenges, asprevious studies have adequately identifiedthe internal support needs of locallymanaged irrigation (Yoder, 1994; IIMI, 1997,1998; Frederiksen and Vissia, 1998; Huppertand Urban, 1998; Svendsen et al., 2000;Huppert et al., 2001). Local irrigationmanagement entities need to focus inter-nally on improving irrigation water manage-ment, while at the same time negotiatingexternally with policy makers, river basinauthorities and other water users to protecttheir water allocation at the basin level.

Irrigation Services in Water-scarce Basins 233



They must become more outward-looking togain or maintain access to water suppliesthat other sectors may try to capture. Asthe entities managing local irrigation arespatially dispersed and mainly focused ontheir own irrigation system, they need toconfederate to lobby and compete at thebasin level. Compounding this challengeis the declining share of the agriculturalsector in many countries, and theperception that other uses of water arehigher value, while the persistentdependence of a large share of the popula-tion, especially of the poor, on farming forlivelihoods continues. There is growingpressure on local irrigation managemententities and farmers to increase waterproductivity and to make the case that theyare not wasting water.

The above challenges apply to bothsurface water and groundwater managementorganizations, and locally managed irriga-tion in upper catchments as well as newlyformed water users’ organizations in large-scale irrigation systems under IMT pro-grammes. Although this is necessarily abroad-brush approach, and the specificchallenges to the myriad irrigation organiza-tions that exist depends on many morefactors than river basin closure, severalcommon threads emerge from the analysisof the basins studied in this book.

13.3 Irrigation Service Provision inWater-scarce Basins

The challenges facing locally managedirrigation in water-scarce basins outlinedabove directly affect the institutionalrequirements for irrigation service provi-sion. A service provision perspective onirrigation management entails focusing onthe different service roles or functionsperformed by the multiple actors involvedin water management in a river basin andthe mechanisms that govern the exchangeof services between them. Huppert et al.(2001, p. 41) distinguish between primaryservices (the provision of irrigation infra-structure and water delivery), secondary

services (maintenance and operation ofinfrastructure) and supporting services. Ofimportance in irrigation service provision isthe large number of actors involved, as theprovision of primary services (such as waterdelivery to a farmer) is the result of a net-work of (supporting) service providers andreceivers. Thus, when analysing servicenetworks, it is necessary to look at theservice relationships between actors byidentifying the laws, procedures, contractsand/or common practices that are the basisfor the relationship, i.e. the governancemechanisms between service providers andreceivers. Analysing service arrangements,whether highly formalized in contracts orbased on customary understandings, entailsstudying the following:

• Which services are being provided, bywhom and to whom;

• What is being exchanged in return foreach service;

• Which governance mechanisms struc-ture the service delivery and whetherthis results in the provision of servicesin a way that suits those concerned.

Our concern here is not with the servicearrangements between farmers in the caseof self-provision of services, which isthe most common form in much of thedeveloping world, or with the arrangementsbetween farmers, local irrigation manage-ment entities and others governing theprovision of primary and secondaryirrigation services. While important fora clear understanding of the internalworkings of locally managed irrigationand identifying areas for improvement (seeHuppert et al., 2001, for different strategiesto strengthen locally managed irrigation),we focus here on the external threats to thesustainability of locally managed irrigation.Several essential support needs for locallymanaged irrigation in water-scarce basinsstand out, namely water rights systemsand water allocation mechanisms, compen-sation mechanisms for water transfers,and increasing water productivity. Thesethree support needs are briefly outlinedbelow.




13.3.1 Water rights systems and waterallocation mechanisms

The basins studied in this book demon-strate the increasing pressure being placedon irrigated agriculture to relinquish water,primarily for environmental and urban/industrial uses. It follows that in water-scarce basins with increasing competitionfor water, the need for effective mechanismsfor allocating water becomes critical. In theabsence of such mechanisms higher-valueuses will tend to out-compete lower-valueuses (with value being defined politically aswell as economically), depriving them ofwater and leading to unregulated transfersof water out of agriculture. A crucial sup-port need of locally managed irrigation isbasin-level water allocation mechanisms forboth primary and derivative water, whichare based on defined water rights that pro-vide security of tenure. The call for clear,secure and transferable water rights hasoften been made, but how to create propertyrights to water that are just, equitable andfeasible (both technically and politically)and how to make them stick is neither clearnor straightforward. Nonetheless, severalprinciples can be defined.

Water rights form part of propertyregimes, which define ownership andconsist of principles and rules to resolvedisputes over property. They may be definedas ‘authorized demands to use (part of) aflow of water, including certain privileges,restrictions, obligations and sanctions’(Beccar et al., 2002, p. 3). Ideally, waterrights delimit the amount of water a right-holder is entitled to, defined either volumet-rically or in terms of shares of availablesupply, as well as the duties of right-holdersrelative to one another and to society atlarge, such as quality and quantity of returnflows. Water rights form the foundationof water allocation mechanisms. An idealwater allocation system is characterizedby flexibility in the allocation of watersupplies, security of tenure for establishedwater users, predictability of the outcome ofthe allocation process, equitability and fair-ness. Mechanisms to allocate water may take

a variety of institutional forms, and includemarginal cost pricing, public (administra-tive) allocation, water markets and user-based allocation (Dinar et al., 1997). Theprevailing system of water rights signifi-cantly influences the specific allocationmechanisms available and the effectivenessof their application. However, withoutinfrastructure in place to withdraw anddistribute water, water rights remain anempty shell. Conversely the creation ofhydraulic property can lead to the de factocreation of water rights (cf. Coward, 1986;Chapter 11).

Due to the type of infrastructureinvolved and the feasibility of transparentand reliable measurements, surface watermay be subject to more effective allocationthan groundwater. Almost everywherein the developing world, groundwater istreated as an apertinent right to privatelyowned land, and where groundwater rightssignificantly different from these have beentried – as in Mexico – they have defiedenforcement. For surface water, three broadwater rights systems have developed histori-cally in different parts of the world basedon either the riparian or the appropriationdoctrines (Simpson and Ringskog, 1997). Allthree are ‘administered’ systems in the sensethat an authority (government or commu-nity) plays an important role in defining,allocating and enforcing rights. Other prop-erty regimes for water are also conceivable,such as a market-based system in whichwater rights are awarded to the highestbidder, but are much less common inpractice.

Under the riparian doctrine, a user hasthe right to extract water from a river systemfor use on land adjacent to the river as long asthe water is returned to the river undimin-ished in quantity or quality and in a mannerthat does not impair downstream use. Fromthis stringent definition, it is clear that inpractice the pure riparian doctrine does notexist, and that it is also not very suited toconditions of water scarcity. Nonetheless,this doctrine is used in various formsthroughout the world, especially in coun-tries with humid climates, and generally




operates without any form of permit orregulatory administration.

Under the appropriation doctrine,water is regarded as a good belonging to alland held in trust by the State or a communalauthority. Water use concessions or licencesare issued to users by the state or the commu-nal authority, which ensures their right todivert or store and use a certain quantity ofwater. Many variations on the appropriationdoctrine exist, the most common two beingprior appropriation and proportional appro-priation. The most significant differencebetween these two is how they treat watershortages. Under prior appropriation, thefirst rights issued on a river have priority orseniority implying that rights issued laterare the first to be curtailed in times of short-age (last in, first out). Under the proportionalappropriation system, concessions (usuallytime bound) are issued to either individualusers or organizations for the use of a maxi-mum quantity of water, although the actualquantity of water that may be used in anyyear is adjusted to reflect water availabilitywithin the river basin. Thus, all concessionholders share in any shortages or surplusesof water proportionally. For this to workwell, an administrative water allocationmechanism needs to be in place that candetermine annual water availability in atimely and transparent manner. Surfacewater rights in the basins studied in thisbook are based on some variation of theproportional appropriation doctrine, withthe exception of California, which has a mixof riparian rights and prior appropriation.

The closure of river basins calls for areassessment of the usefulness of the appro-priation doctrine, be it prior or proportional.The prior appropriation system, whileproviding security of tenure, is problematicin water-scarce basins as it does not providefor sharing shortages among right-holders,nor does it allow for new entrants (Huffakeret al., 2000). Furthermore, due to the senior-ity principle, the actual transfer of waterrights is problematic, as all appropriatorsmore senior than the buyer will need toapprove the sale, while selling water mayalso be interpreted as proof of non-beneficialuse, constituting grounds for revocation

of the water right (cf. Rosegrant and Bins-wanger, 1994; Bolding et al., 1999; Haddad,2000). The flexibility of the proportionalappropriation system would appear to per-mit the application of various allocationmechanisms, including market-based ones.However, regulating large numbers of smallusers is exceptionally difficult and placesstrong demands on enforcement.

In the basins studied in this book – andothers in the developing world – existingwater rights systems and allocation mecha-nisms are not well tailored to deal with basinclosure. This is clearly the case in Turkeyand Mexico, whose water rights systems aregenerally unable to prevent or to resolveconflicts between new and existing claims,or prevent the over-allocation of water. Thisis even more so in stressed basins in poorercountries in South Asia and Africa. In addi-tion, most water rights systems primarilydeal with surface water, with much lessattention given to groundwater or derivativewater. Thus, in the case of California,groundwater is only lightly regulated andthe permissive specification of rights togroundwater has led to increasing problemswith aquifer overdraft. In the Gediz basin,shallow groundwater is an open-accessresource, meaning that anyone with theinfrastructure to tap that water can do so.In the Lerma–Chapala basin, groundwateris in effect also an open-access resource,although formally groundwater users arerequired to obtain a pumping permit fromfederal authorities indicating a maximumextraction rate. This is by far the best thatany middle- or low-income country with asubstantial irrigated agriculture sector hasdone to bring a modicum of order in privateappropriation of groundwater for irrigation.The reform has helped register all ground-water users; but restricting their with-drawals to their permitted quotas hasproven to be nearly impossible.

In closing river basins with significantgroundwater extractions, surface irrigationsystems, both large and small, play anincreasingly important – often the sole – roleas cheap and effective groundwater rechargesystems. In effect, groundwater users pumpwater previously paid for by farmers




irrigating with surface water, without com-pensating these surface irrigators. As long asthis is conjunctive use by the same farmerthis is not a problem, but in areas wheregroundwater levels have fallen sharply andonly the better off can afford to continuepumping groundwater, this does become anissue for poorer farmers only using surfacewater. Both in Gediz and Lerma–Chapala,there are strong hydrologic linkages betweensurface water flows, surface irrigation andgroundwater recharge. For Lerma–Chapala,Scott and Garcés-Restrepo (2001) found thatapproximately 50% of canal water appliedto crops recharges the underlying aquifersand subsequently becomes available forpumping. Because of this recycling, half ofeach unit of canal water provides subse-quent additional benefit as groundwater. Byway of example, assuming marginal valuesof Mex$1.80/m3 for canal water andMex$2.40/m3 for groundwater, with 50%recharge the aggregate value of surface wateris Mex$3.00/m3. However, groundwaterusers do not pay canal water users for therecharge function they perform, while canalwater users do have to pay for the fullvolume of water they receive, includingthe 50% that goes to deep percolation.

The situation surrounding derivativewaters (agricultural return flows andmunicipal/industrial wastewater) is evenless regulated. In all the basins studied,quality standards exist for return flows andwastewater, but access to this water forproductive purposes would appear to be afree-for-all, with the exception of California.In the delta of the Gediz basin and in theLerma–Chapala basin, derivative water isbecoming a critical source of water for farm-ers. Although no allocation mechanismsexist for derivative water, in Mexico, farmersdo need to go through a fair amount oftrouble to establish a tenuous claim onwastewater by reaching local agreementswith municipalities or WUAs (Buechler andScott, 2000). These claims are being threat-ened by the construction of wastewatertreatment plants and the de facto realloca-tion of treated water to other uses, such asgolf courses. In closing river basins, wherethe better off have already captured primary

water sources and renegotiating water rightsis extremely complicated, derivative waterrapidly becomes the poor farmer’s last resortand should be recognized as such.

Without secure water rights and clearlydefined water allocation mechanisms, indi-vidual farmers and locally managed irriga-tion systems in water-scarce basins face anuncertain future. The hydrological inter-actions between surface water, groundwaterand derivative water make it necessary toarrive at a coherent and feasible system ofwater rights and water allocation mecha-nisms in water-scarce basins. Several issuesthat such a coherent system of water rightswould need to deal with stand out, namelysurface–groundwater interactions, returnflows, water quantity and quality inter-actions, provisions for basic human needsand environmental flows, and lastly provi-sions for new entrants. The new water rightssystem and water allocation mechanisms inSouth Africa appear to meet all these needs,and could serve as an example for theother countries studied in this book ifthey can be implemented successfully. Theestablishment, modification or clarificationof water rights systems requires action atthe national level (legislative and executivebranches of government), but ideally shouldconsist of a process in which all interestshave adequate representation. The supportservices required by local irrigation manage-ment entities in this regard are legal adviceand representation as well as lobbyingcapabilities at the basin and national level.

13.3.2 Water transfers and compensationmechanisms

A second critical support need of locallymanaged irrigation in water-scarce basins iscompensation mechanisms for water trans-ferred out of agriculture. This is closelytied to water rights and water allocationmechanisms, but is sufficiently importantto warrant separate consideration. In recentyears, there has been an increase in interestin the feasibility of establishing watermarkets (Rosegrant and Binswanger, 1994;




Lee and Jouravlev, 1998; Haddad, 2000). Inprinciple, water markets enable allocationof water to high-value uses and faircompensation of those who sell their water.Especially in the case of inter-sectoral watermarkets, e.g. between the urban and agri-cultural sectors, several advantages areapparent, including more efficient use ofwater in agriculture and compensationto farmers for the transfer of water outof agriculture. Rosegrant and Binswanger(1994) outline the following benefits ofwell-designed water markets:

• Empowerment of water users by requir-ing their consent to water reallocationand compensation;

• Incentive for users to invest in water-saving technology, if well-definedrights are established;

• Incentive for efficient use and incomethrough sale by considering full oppor-tunity costs of water;

• Greater flexibility in responding tochanges in the production environment(crop prices, demand patterns andcomparative advantage).

While the economic virtues of efficiencyand productivity are invoked as beneficialoutcomes of water markets, the very realobstacles to establishing inter-sectoralwater markets are often overlooked(Haddad, 2000; Young, 1986). Chief amongthese are issues of access to water and thelack of infrastructure to enable physicalexchange and effective measurement ofwater (Dinar et al., 1997; Perry et al., 1997;Simpson and Ringskog, 1997). Markettransactions assume that effective meansexist for the exchange of commodities.Groundwater may be transferred fromexisting low-value uses to competinghigh-value uses only within an aquifer, notwithin a larger river basin unless consider-able investment is made in infrastructurefor water conveyance. The same is true forsurface water, although the presence ofrivers makes transfers from upstream todownstream uses feasible. Transferringwater in the opposite direction requiressubstantial investments in infrastructureand control mechanisms.

Much of the claims for the ‘success’ ofsurface water markets, especially outsideof the USA, rest more on political and ideo-logical beliefs than on rigorous empiricalstudies (Kloezen, 1999). In the case of Chile,Bauer (1997) convincingly demonstratesthat water markets are much thinner on theground and hence less effective in waterallocation than often assumed in policypapers. For groundwater, the situation isdifferent, with active markets reported on inSouth Asia (cf. Shah, 1993; van Koppen,1998). While groundwater markets have alsohad positive impacts on the poor (Shah,1993), there is widespread concern thatsurface water markets will result in theconcentration of water rights in the handsof the few, at the expense of small farmers(Bauer, 1997; Zwarteveen, 1997). The chal-lenge this poses is to design regulated watermarkets that are pro-poor, recognize thesocial and environmental values of water,and facilitate the resolution of inter-sectoralconflicts through compensation.

If appropriately designed, water marketsmay provide a good mechanism to compen-sate farmers for water transferred out ofagriculture. Pre-conditions for the effectiveoperation of water markets include definedwater rights, demand in excess of supply,legal frameworks that indicate how tradesshould take place, physical infrastructurefor conveyance and measurement of water,and provisions for the protection of third-party interests (Perry et al., 1997; Simpsonand Ringskog, 1997; Lee and Jouravlev,1998). In combination with the proportionalappropriation of water rights and adminis-trative allocation, socially just water marketsare conceivable. Such a system could consistof three tiers, namely a reserve for basichuman needs and the environment, areserve for productive water for the poor anda third tier consisting of water for productiveuse and additional water for urban use andthe environment. Only the third tier wouldenter the water market, after proportionaladministrative allocation on an annual basishas determined the water available for pro-ductive use. Special provisions would needto be made to enable the poor to lease theirproductive water if they so decide, while




protecting their ownership of the water right.This system demands a strong regulatoryand administrative framework, but is theo-retically conceivable in Turkey, Mexico andSouth Africa, although water laws in allthree countries are notably vague on thepossibility of such market-based transfers.

Water rights transfers in the third tiercan take a variety of forms, depending onfactors such as the structure of the market,legal and third-party considerations, thecharacteristics of the water rights, trans-action costs and above all, the needs of theparties to the transaction. Three types oftransactions can be identified, namely sales,lease contracts and option contracts (Leeand Jouravlev, 1998). Sales consist of thepermanent transfer of title, including allbenefits, costs, risks and obligations associ-ated with the water right. Sales are mosttypical of inter-sectoral transfers, with irri-gated agriculture being the dominant waterseller and urban users the principal buyers.The leasing of water rights involves the saleof water, but not of the water right. Leasestend to be short term, consisting of thetemporary exchange of a quantity of waterfor monetary or other remuneration, and arefrequently used for water transfers withinirrigation systems.

Option contracts are long-term agree-ments to lease, but not to sell, a water rightwhen a given contingency arises, e.g. adrought. Under option contracts, the receiv-ing party holds an option to buy water at aspecified price under specified conditionsfrom the seller, who guarantees future deliv-ery if the conditions apply. In exchangefor this guarantee, the holder of the optionalso pays a premium to the granting party,usually at the onset of the option contract.Option contracts are most commonly usedto transfer water from irrigated agricultureto non-agricultural uses such as theenvironment during periods of low streamflow (Landry, 1998). This means that waterusers can continue to use their water duringyears of normal water availability, and hence

option contracts are a more attractive alter-native than the outright sale of water rightsor long-term leases. Although option con-tracts are quite complex and require clearlyestablished water rights, initial experienceswith them in Chile (Thobani, 1997) andColorado (Michelsen and Young, 1993) havebeen quite positive. In our assessment, anoption contract arrangement may be suitableto compensate farmers for agriculture-to-urban and agriculture-to-environmentwater transfers. Crucial to the design ofappropriate option contracts are:

• Definition of the contingency condi-tions under which water will betransferred;

• Duration of the contract and/orconditions for its renegotiation priorto expiration;

• Volumes to be ceded (by whom in thecase of multiple parties or states);

• Specification of compensation (bothmonetary amount and process) for lostincome resulting from the watertransfer;

• Mechanisms to redress grievances onthe part of ceding, receiving or thirdparties.

13.3.3 Increasing water productivity

The third critical support need for locallymanaged irrigation is assistance in increas-ing the productivity of water. While theview that irrigation uses water inefficientlymay often be erroneous, to be credible part-ners in basin management local irrigationmanagement organizations and farmers willneed to show that they are using waterproductively. This can be measured interms of the biomass or income producedper cubic metre of evapotranspiration.1

Both elements of water productivity areimportant but place different demands onthe support needs for locally managed irri-gation. Higher biomass production can be


1 Not per cubic metre withdrawn, as irrigation water that is not used in the production of biomass will eitherrecharge the aquifer or become available for re-use as return flows (see Kijne et al., 2003).



achieved through improvements in waterdelivery practices, crop varietal improve-ment and improved cultural practices (suchas mulching, crop planting dates, etc). Thebreeding of new crop varieties, the intro-duction of improved cultural practices andsimilar actions require agricultural supportservices, whereas improving water deliveryis the primary responsibility of the locallymanaged irrigation entity. However, exter-nal support remains important to improveirrigation infrastructure and managementto enhance reliable water delivery and torealize real water savings.

Improving the productivity of water ineconomic terms is crucial for sustainablelocally managed irrigation. Perhaps one ofthe biggest threats to locally managed irriga-tion is the low profitability of agricultureand the high costs of agricultural produc-tion. Solving problems related to the levelsand collection of irrigation service fees aswell as the quality of irrigation service deliv-ery is an internal responsibility of locallymanaged irrigation entities, but the willing-ness and ability of farmers to pay fees isclearly related to crop market prices and eco-nomic returns to agricultural production.Where appropriate, shifting to less-water-consuming crops that provide a higher eco-nomic return needs to be supported. Theefforts to do so in the Lerma–Chapala Basin,with farmers and government agenciesworking together to shift from low-valuegrain crops to crops that use less water andsell at a higher price, show how complicatedit is to achieve this for substantial areas.Although increasing the profitability ofagriculture touches on much wider issuesthan support needs for locally managedirrigation, such as trade regimes and agri-cultural subsidies in affluent countries,it lies at the heart of sustainable locallymanaged irrigation.

13.4 Poverty and Representation in RiverBasin Management

The case studies in this book show thatensuring the sustainability of locally

managed irrigation and meeting the waterneeds of the poor in closing river basins isa serious challenge to current institutionalarrangements. Although a service provisionperspective on water management high-lights where changes are necessary in thewider institutional environment to moreeffectively support locally managed irriga-tion, it is relatively silent on two criticalissues in closing river basins in low- andmiddle-income countries, namely povertyreduction and stakeholder representation.Cross-cutting these two issues is the questionof how access to water and water manage-ment decision making is gendered. A sus-tainable livelihoods perspective, with itsemphasis on rights and entitlements and theinstitutional arrangements through whichthese are provided and reproduced, holdsmore promise for defining pro-poor waterpolicies in water-scarce basins (cf. vanKoppen, 2000).

13.4.1 Productive water and pro-poorwater policy

Although river basin closure poses signifi-cant challenges to the sustainability oflocally managed irrigation, generally thisaffects people who already have access towater for productive purposes. A character-istic of many stressed river basins in devel-oping countries is the large number of poorpeople who do not have access to water forproductive purposes. The degree of waterdeprivation in many river basins around theworld is well documented, with more than1 billion people lacking access to water ofsufficient quality and quantity to meetminimum standards of living, let alone forproductive purposes. The processes thatcome into play as river basins mature canhave serious consequences for perpetuatingpoverty.

In analogy to low prices for basic grains,which hurt poor farmers but benefit thepoor urban population, the reallocationof water from agriculture to urban suppliesmay benefit the urban poor. Whether this isthe case depends on how the institutional




and financial resources for urban waterinfrastructure development are targeted atthe poor. As many maturing river basinsexhibit a rapid pace of urbanization causedby rural poor moving to the cities, on balancethe transfer of water out of agriculturecould benefit more poor people than arehurt. Likewise, as agriculture maturesand consolidates, taking water out ofagriculture may be hurting the better offrather than the poorest. If farmers invest toimprove the productivity – and profitability– of water there may not be any major dam-age to their livelihoods. Understanding howthese scenarios work out requires empiricalresearch.

However, other processes act in theopposite direction. These relate to the devel-opment of the physical means to abstractand convey water and the distribution ofland and water rights. If water infrastructuredevelopment and the allocation of landand water rights is not specifically targetedat poor women and men, the danger ofresource capture by the better off is everpresent. While freshwater supplies areclearly limited, for most people waterscarcity is caused by political, technologicaland economic barriers that limit their accessto water and by competition between wateruse(r)s (Falkenmark and Lundqvist, 1998).Water scarcity is not only a naturally occur-ring phenomena, but has also been createdthrough the development of water resourcesin the past, the selective entitlement of waterrights and incidental and structural resourcecapture by the better off (cf. Homer-Dixon,1999; GWP, 2000). These processes makeit very difficult to increase and protectthe water security of the poor, especially inclosing river basins where the consumptionof water by one literally deprives another ofthat water.

While the processes that come intoplay as river basins close have multipleand uncertain consequences for perpetuat-ing poverty, it is increasingly recognizedthat access to water for productive usesis very important for the poor to buildsustainable livelihoods (van Koppen, 2000;Schreiner and van Koppen, 2001). We main-tain that pro-poor water allocation policies

can transform irrigation into a powerfulinstrument for creating sustainable liveli-hoods (cf. Hussain and Hanjra, 2004; Shah,2000). The challenge this poses in water-scarce basins is balancing the allocationof productive water for poverty reductionwith allocations designed to meet the needsof proven productive capacity (i.e. industry,commercial agriculture, mining) and theenvironment (cf. Perret, 2002). This maymake it necessary to redistribute waterrights in favour of the poor, but alsocalls for a judicious use of water andinnovations in land and water managementtechnologies.

To craft pro-poor water policies, anunderstanding of the processes that createpoverty is needed. While individuals expe-rience poverty and can work their way out ofpoverty, there is also truth in the statementthat societies produce poverty through pro-cesses of exclusion. Culturally embeddednotions of entitlements, ownership, access,control and participation underlie theconcept of exclusion (Bhalla and Lapeyre,1997, p. 417). The deprivation commonlyassociated with exclusion is not only relatedto a lack of economic resources but alsoa lack of recognition and entitlements. Aspointed out by Sen, ‘economic resourcesenable access not only to economic goodsand services but also to political goodslike freedom and the ability to influencepolicies’ (1975; cited in Bhalla and Lapeyre,1997, p. 418).

In this sense, access to water can beviewed as a potential vehicle to achieve eco-nomic and political rights. These are prereq-uisites for full citizenship, which in turnopen opportunities for political partici-pation. This interpretation brings out thestate’s role in exclusion. Through theirstructures, procedures and legal frame-works, governments can exclude somegroups from fully attaining their economicrights, while including others. A case inpoint is the systematic exclusion of womenin government irrigation programmesthroughout the world, convincingly docu-mented in numerous case studies (for adiscussion of the literature, see Zwarteveen,1994, and van Koppen, 1998).




A defining feature of poverty is thatthe poor have very little influence on theways in which governments and economiesallocate rights and resources in society. Inclosing river basins, pro-poor water policiesthat focus on redressing imbalances throughthe reallocation of water rights will chal-lenge the existing distribution of rights andresources as there is no extra water to goaround. This creates the political dilemmaof confronting vested interests in society,and requires that government has boththe organizational and political ability toovercome resistance to redistribution. Animportant first step is the formulation ofwater legislation that sets out procedures forthe creation of a reserve of productive waterfor the poor and how new institutions suchas river basin management entities can workin a redistributive manner (cf. van Koppen,2000).

Of the five basins reviewed in this book,South Africa is the only country formallyplacing emphasis on redressing imbalancesand achieving equity in water management.In the Government of South Africa’s WhitePaper on national water policy (DWAF,1997), three components of equity aredefined: equity in access to water services(drinking water and sanitation), equity inaccess to water resources (water for produc-tive purposes) and equity in the access tobenefits from the use of water resources(allocative efficiency). However, even withthe commitment of the government, reallo-cating water to new entrants is provingvery difficult (van Koppen et al., 2003).A way forward could be to impose watersavings on commercial agriculture and otherproductive uses, and then allocating thesesavings to a water reserve for the poor. Underthe system of compulsory water licensingincluded in the 1998 National Water Act,this is possible, but it will require deter-mined resolve from the government to carryit through (cf. Perret, 2002). Compulsorywater licensing is currently being pilot-tested in one river basin, making itpremature to draw conclusions. New infra-structure will also need to be developed, toensure that the water rights of the poor donot remain paper rights.

13.4.2 Representation in river basinmanagement

An important challenge in river basin man-agement is ensuring that all stakeholdershave a voice in basin governance. Althoughfrequently advocated as a key to achievingeffective water management, stakeholderparticipation in river basin management isnot straightforward, and actually includingthe poor and achieving substantive stake-holder representation has proven elusive inpractice (Wester et al., 2003). The questionof how greater equity in water managementand representation in river basin manage-ment can be achieved in highly stratifiedsocieties with significant gender and socialinequalities remains. As poverty is an out-come of how societies are structured, it isevident that marginal groups are excludedfrom decision making. A danger of theemphasis on participation in river basinmanagement is that attention is drawn awayfrom the very real social and economic dif-ferences between people and the need forthe redistribution of resources, entitlementsand opportunities. In the long term, mar-ginal groups will only gain a voice in riverbasin management when they are no longermarginal. This entails fundamental changesin the way societies are structured, suchthat they no longer produce poverty butwealth that is fairly distributed. As it isunlikely that this will happen in thenear future, in the short run mechanismsneed to be devised that strengthen therepresentation of marginal groups in riverbasin management.

It is clear that the size of the populationin most river basins is such that it precludesthe direct participation of all stakeholders inbasin-level decision making. Thus, as deci-sion making moves to the river basin level,serious thought needs to be given to howhard-won democratic rights in conventionalsocial and political domains are assured inthe river basin domain (cf. Barham, 2001).As Green and Warner (2000) point out, inte-grated water management and participationpull in opposite directions. While thecomplexity of integrated management ofsizeable river basins invites centralization




and technocracy, participation suggestssubsidiarity and small-scale operations,engaging people to think creatively aboutissues with which their lives are intimatelylinked. Thus, in any basin of some size, riverbasin management would entail a layeredsystem of representation. The questionthen becomes who gets to represent groupsof stakeholders in river basin management.This question strikes at the heart of basingovernance, and revolves around whichtype of democracy is implied, liberal orsocial.

Liberal democratic theory is premisedon a notion of abstract individualism andassumes that all people are equal in thepublic sphere (Held, 1995; Luckham et al.,2000). In water reforms informed by liberaldemocracy, it is assumed that it is possiblefor water management stakeholders tobracket status differentials and powerinequalities and to deliberate ‘as if’ theywere equals in water management forumssuch as WUAs or river basin councils. Socialdemocracy, on the other hand, departs fromsocial inequalities and attempts to increasecitizen involvement in the affairs of govern-ment and expand the concept of citizenshipto cover economic and social rights as wellas political rights. Thus, it aims at a redistri-bution of power and resources to enablecitizens to participate in the decisions thataffect their lives (Luckham et al., 2000). Inwater reforms informed by social democ-racy, water is seen as a basic human rightand a politically contested resource (Gleick,1998; Mehta, 2000).

On the face of it, stakeholder platforms,be they river basin councils, catchmentmanagement agencies or watershed coun-cils, democratize river basin management bygiving voice to a multiplicity of interestedactors. However, much depends on theexisting institutional arrangements fromwhich stakeholder platforms for river basinmanagement emerge, as many roles, rightsand certainly the technologies and physicalinfrastructure for controlling water arealready in place. In river basins, watermanagement stakeholders may have differ-ent levels and kinds of education, speak dif-ferent languages, differ in access to politics

and hold different beliefs about how natureand society function (cf. Edmunds andWollenberg, 2001). If this is not taken intoaccount when creating new rules, roles andrights, the institutional outcome can easilyprivilege those who are literate and haveaccess to the legal system. If doneunreflectively and without an emphasis onthe redistribution of power and resources,new institutional arrangements for riverbasin management will institutionalizeinequality and power differentials instead ofgiving voice to marginal groups (Wester andWarner, 2002). Without firm land and waterrights and livelihood security, there is verylittle incentive for the poor to participate inriver basin management.

Having argued for social democracy inriver basin management, the issue of stake-holder representation remains. The relation-ship of the representatives participating inriver basin management to their constituentsis problematic, especially when third partiesare involved. It is a nostrum of developmentwork that third-party facilitators are neededto help identify, mobilize, organize andinform stakeholder groups. However, aspointed out by Edmunds and Wollenberg

the relationship of a representative to his/her constituency is perhaps most politicallycharged when representatives of a group aredesignated by outsiders or are accountableto them, as is often the case in multi-stakeholder negotiations. From the start,outside conveners and facilitators influencerepresentation by the selection of stake-holder groups, the people to represent eachgroup and how the expression of interests isfacilitated in the meeting. (2001, p. 240)

This points to the need for a broad andinclusive process of invitation, consultationand consolidation of interested stakeholders,assumed to represent all interested parties.Whereas stakeholder processes and repre-sentation in river basin management areimportant, they need to be twinned with afocus on securing water entitlements for thepoor. This points to an important role forgovernment, both in drawing up and enforc-ing water laws that explicitly safeguard cus-tomary water rights and contain provisionsfor reallocating water rights to the poor.




13.5 Conclusion

Reflecting on the challenges facing small-holder irrigation in water-scarce basins, itbecomes clear that where poverty is wide-spread, river basin management needs tohave a strong developmental dimension.However, where the extra water for pro-ductive purposes that is needed for povertyalleviation will come from is less clear.If a country is rapidly industrializing anddiversifying its economy, perhaps this isnot an issue. However, most poor countriesare not creating alternative employment ata sufficient rate to provide the rural poorwith attractive alternatives to agriculture. Insuch cases trying to intensify smallholderagriculture, which often requires irrigation,is the only feasible strategy in the shortto medium term. Thus, central concernsbecome the mechanisms in place to allocatewater rights, the regulatory entity responsi-ble for water allocation, and how conflictsabout water rights and water distributionare mediated. Finding the right mixbetween the state, the market and theempowerment of the marginal points to theneed to move beyond token consultationtowards partnerships, negotiations and con-flict resolution. At the minimum, strategiesfor river basin management should detailmechanisms for redressing imbalances inaccess to water for productive purposes andestablishing clear and secure water rightsfor the poor.

While much can be learned from insti-tutional arrangements for river basin man-agement in affluent countries, it is crucialto understand that these do not operatein the conditions of low-income countries:dominance of smallholder agriculture, weakinstitutions, insufficient financial andhuman resources, marked social inequityand extreme poverty. While water develop-ment and management can only partlyaddress these issues, they must explicitlyform the points of departure in the reformof institutional arrangements for river basinmanagement in developing countries. Thischapter does not hold the answers to howthis should be done, but it does offer

elements of a strategy that could be followedto address water deprivation in closing riverbasins in developing countries. Such a strat-egy consists of a fine balancing act betweenallocating water for poverty reduction andallocations designed to meet the needs ofproven productive capacity.

References

Barham, E. (2001) Ecological boundaries ascommunity boundaries: the politics of water-sheds. Society and Natural Resources 14,181–191.

Barrows, C.J. (1998) River basin developmentplanning and management: a critical review.World Development 26, 171–186.

Bauer, J.C. (1997) Bringing water markets downto earth: the political economy of water rightsin Chile, 1967–95. World Development 25,639–659.

Beccar, L., Boelens, R. and Hoogendam, P. (2002)Water rights and collective action incommunity irrigation. In: Boelens, R. andHoogendam, P. (eds) Water Rights andEmpowerment. Koninklijke van Gorcum,Assen, pp. 1–19.

Bhalla, A. and Lapeyre, F. (1997) Social exclusion:towards an analytical and operational frame-work. Developmentand Change 28, 413–433.

Bolding, A., Manzungu, E. and van der Zaagm, P.(1999) A realistic approach to water reform inZimbabwe. In: Manzungu, E., Senzanje, A.and van der Zaag, P. (eds) Water for Agricul-ture in Zimbabwe: Policy and ManagementOptions for the Smallholder Sector. Univer-sity of Zimbabwe Publications, Harare,pp. 225–253.

Buechler, S. and Scott, C. (2000) ‘Para nosotros,esta agua es vida.’ El riego en condicionesadversas: Los usuarios de aguas residuales enIrapuato, México. In: Scott, C.A., Wester, P.and Marañón-Pimentel, B. (eds) Asignación,productividad y manejo de recursos hídricosen cuencas. IWMI, Serie LatinoamericanaNo. 20. IWMI, Mexico City.

Coward, E.W. Jr (1986) State and locality in Asianirrigation development: the property factor.In: Nobe, K.C. and Sampath, R.K. (eds) Irriga-tion Management in Developing Countries:Current Issues and Approaches. Studies inwater policy and management no. 8. West-view Press, Boulder, Colorado, pp. 491–508.




Dinar, A., Rosegrant, M.W. and Meinzen-Dick, R.(1997) Water Allocation Mechanisms.Principles and Examples. World Bank PolicyResearch Working Paper 1779. World Bankand International Food Policy ResearchInstitute, Washington, DC.

DWAF (1997) White Paper on a National WaterPolicy for South Africa. Department of WaterAffairs and Forestry, Pretoria.

Edmunds, D. and Wollenberg, E. (2001) A strategicapproach to multistakeholder negotiations.Development and Change 32, 231–253.

Falkenmark, M. and Lundqvist, J. (1998) Towardswater security: political determination andhuman adaptationcrucial.Natural ResourcesForum 21, 37–51.

Frederiksen, H.D. and Vissia, R.J. (1998) Con-siderations in Formulating the Transfer ofServices in the Water Sector. IWMI, Colombo,Sri Lanka.

Gleick, P. (1998) The human right to water. WaterPolicy 1, 487–503.

Global Water Partnership (GWP) (2000) TowardsWater Security: A Framework for Action.GWP, Stockholm.

Green, C. and Warner, J. (2000) Flood hazard man-agement: a systems perspective. Paperpresented at the 9th Stockholm WaterSymposium, SIWI, 9–12 August.

Haddad, B.M. (2000) Rivers of Gold. DesigningMarkets to Allocate Water in California.Island Press, Washington, DC.

Held, D. (1995) Democracy and the Global Order:From the Modern State to CosmopolitanGovernance. Polity Press, Cambridge.

Homer-Dixon, T.F. (1999) Environment, Scarcityand Conflict. Princeton University Press,Princeton, New Jersey.

Huffaker, R., Whittlesey, N. and Hamilton, J.R.(2000) The role of prior appropriation inallocating water resources in the 21stcentury. International Journal of WaterResources Development 16, 265–273.

Huppert, W. and Urban, K. (1998) AnalysingService Provision. Instruments for Develop-ment Cooperation Illustrated by Examplesfrom Irrigation. Schriftenreihe der GTZ,No. 263. GTZ, Wiesbaden.

Huppert, W., Svendsen, M. and Vermillion, D.(2001) Governing Maintenance ServiceProvision in Irrigation – A Guide to Institu-tionally Viable Maintenance Strategies. GTZ,Eschborn.

Hussain, I. and Hanjra, M.A. (2004) Irrigation andpoverty alleviation: review of the empiricalevidence. Irrigation and Drainage 53, 1–15.

International Irrigation Management Institute(IIMI) (1997) Program of Support Systems forLocal Management in Irrigation. Final Reportsubmitted to BMZ and GTZ, November 1997.IWMI, Colombo, Sri Lanka.

IIMI (1998) Privatization and Self Management ofIrrigation, Phase II. Final Report submitted toBMZ and GTZ, March. IWMI, Colombo.

Kijne, J.W., Barker, R. and Molden, D. (eds) (2003)Water productivity in africulture: limits andopportunities for improvement. Comprehen-sive Assessment of Water Management inAgriculture Series. No. 1. CAB International,Wallingford, UK.

Kloezen, W. (1999) Water markets betweenMexican water user associations. WaterPolicy 1, 437–455.

Landry, C. (1998) Market transfers of water forenvironmental protection in the westernUnited States. Water Policy 1, 457–469.

Lee, T.R. and Jouravlev, A.S. (1998) Prices,Property and Markets in Water Allocation.United Nations Economic Commission forLatin America and the Caribbean, Santiago.

Luckham, R., Goetz, A. and Kaldor, M. (2000)Democratic Institutions and Politics inContexts of Inequality, Poverty, and Conflict.A Conceptual Framework. IDS, Brighton, UK.

Mehta, L. (2000) Water for the Twenty-firstCentury: Challenges and Misconceptions.IDS, Brighton, UK.

Michelsen, A.M. and Young, R.A. (1993)Optioning agricultural water rights for urbanwater supplies during droughts. AmericanJournal of Agricultural Economics 75,1010–1020.

Perret, S.R. (2002) Water policies and small-holding irrigation schemes in South Africa:a history and new institutional challenges.Water Policy 4, 283–300.

Perry, C., Rock, M. and Seckler, D. (1997) Water asan Economic Good: A Solution, or a Problem?IIMI Research Report 14. InternationalIrrigation Management Institute, Colombo,Sri Lanka.

Rosegrant, M.W. and Binswanger, H.P. (1994)Markets in tradable water rights: potential forefficiency gains in developing country waterresource allocation. World Development 22,1613–1625.

Schreiner, B. and van Koppen, B. (2001) Frombucket to basin: poverty, gender andintegrated water management in SouthAfrica. In: Abernethy, C. (ed.) IntersectoralManagement of River Basins: Proceedings ofan International Workshop on ‘Integrated




Water Management in Water-stressed RiverBasins in Developing Countries: Strategiesfor Poverty Alleviation and AgriculturalGrowth’. IWMI and DSE, Colombo, Sri Lanka,pp. 15–69.

Scott, C.A. and Garcés-Restrepo, C. (2001) Con-junctive management of surface water andgroundwater in the middle Río Lerma Basin,Mexico. In: Biswas, A.K. and Tortajada, C.(eds) Integrated River Basin Management:The Latin American Experience. OxfordUniversity Press, New Delhi, pp. 176–198.

Shah, T. (1993) Groundwater Markets andIrrigation Development: Political Economyand Practical Policy. Oxford UniversityPress, Bombay.

Shah, T. (2000) Water against poverty: livelihood-oriented water resource management. In:Mollinga, P.P. (ed.) Water for Food and RuralDevelopment: Approaches and Initiatives inSouth Asia. Sage, New Delhi, pp. 38–68.

Shah, T., van Koppen, B., Merrey, D.J., de Lange,M. and Samad, M. (2002) InstitutionalAlternatives in African Smallholder Irriga-tion: Lessons from International Experienceswith Irrigation Management Transfer. IWMIResearch Report 60. IWMI, Colombo, SriLanka.

Simpson, L. and Ringskog, K. (1997) WaterMarkets in the Americas. The World Bank,Washington, DC.

Svendsen, M., Trava, J. and Johnson III, S.H.(2000) A synthesis of benefits and second-generation problems. In: Groenfeldt, D.and Svendsen, M. (eds) Case Studies inParticipatory Irrigation Management. WorldBank Institute Learning Resources Series.World Bank Institute, Washington, DC,pp. 139–157.

Thobani, M. (1997) Formal water markets: why,when and how to introduce tradable rightsin developing countries. In: Proceedings ofSeminar on Economic Instruments forIntegrated Water Resources Management:Privatization, Water Markets and TradableWater Rights. Inter-American DevelopmentBank, Washington, DC.

van Koppen, B. (1998) More Jobs per Drop.Targeting Irrigation to Poor Women and Men.Royal Tropical Institute, Amsterdam.

van Koppen, B. (2000) From Bucket toBasin: Managing River Basins to AlleviateWater Deprivation. IWMI, Colombo, SriLanka.

van Koppen, B., Jha, N. and Merrey, D.J. (2003)(forthcoming) Redressing Racial Inequitiesthrough Water Law in South Africa: Revisit-ing Old Contradictions? ComprehensiveAssessment Research Paper 3. Comprehen-sive Assessment Secretariat, Colombo, SriLanka.

Vermillion, D. and Merrey, D.J. (1998) Whatthe 21st century will demand of watermanagement institutions. Journal of AppliedIrrigation Science 33, 165–187.

Wester, P. and Warner, J. (2002) River basinmanagement reconsidered. In: Turton, A.and Henwood, R. (eds) Hydropolitics inthe Developing World: A Southern AfricaPerspective. African Water Issues ResearchUnit, Pretoria, pp. 61–71.

Wester, P., Merrey, D.J. and de Lange, M. (2003)Boundaries of consent: stakeholder represen-tation in river basin management in Mexicoand South Africa. World Development 31,797–812.

Yoder, R. (1994) Locally Managed IrrigationSystems: Essential Tasks and Implicationsfor Assistance, Management Transferand Turnover Programs. InternationalIrrigation Management Institute, Colombo,Sri Lanka.

Young, R.A. (1986) Why are there so few trans-actions among water users? AmericanJournal of Agricultural Economics 68,1143–1151.

Zwarteveen, M. (1994) Gender Issues, WaterIssues: A Gender Perspective to IrrigationManagement. Working Paper No. 32. IIMI,Colombo, Sri Lanka.

Zwarteveen, M. (1997) Water: from basic needto commodity. A discussion on gender andwater rights in the context of irrigation. WorldDevelopment 25, 1335–1351.




Index

Note: page numbers in italics refer to figures, tables and boxes.

abstractionagriculture 76, 77charges 87licences 81, 82, 89limits in Neste system (France) 98patterns 77uses 76

actors 7, 9–10Central Valley (California) 114–117Dong Nai River basin (Vietnam)

177–184, 185–187, 188–190Gediz River basin (Turkey) 201–204,

205, 206key 100, 101Lerma–Chapala basin (Mexico) 220Olifants River basin (South Africa)

152–153roles 7

advocacyCentral Valley (California) 116–117,

119Olifants River basin (South Africa)

158agriculture

abstractions 76, 77Central Valley (California) 111, 113charging impacts 86–87Dong Nai River basin (Vietnam)

175–176, 192drainage 206Gediz River basin (Turkey) 196–197,

199, 200

income 102Lerma–Chapala basin (Mexico)

129–130, 220productivity 76, 77–78rainfed 20, 22, 23return flows 206South Africa 145–146water productivity increase 240water service providers in South

Africa 164water use 27–28, 75, 76

unsustainable 27annual runoff, mean 149appropriative rights 114, 236aquaculture 177aquifer management councils 140Aral Sea crisis 43arid areas, precipitation 36Asian Development Bank guidelines 61authoritarian model 14Authorized User Associations (ASAs,

France) 93, 94–95, 96

banks see development banksbenchmarking indicators 67–68, 69–70

CALFED process 119, 120California Aqueduct (USA) 112canal construction in Vietnam 182capital investment 79, 82

247



Catchment Management Agency (SouthAfrica) 146, 155–156, 158, 163,165, 166, 221

Central Valley (California) 109–122basin hydrology 110–113basin management 109, 121–122competition for water 110configuration 218–220enabling conditions 120–121information attributes 225institutional environment 113–117,

118, 119–121legal authority 121legal issues 113–117, 118, 119–121policy issues 113–117, 118, 119–121politics 120–121, 223–224public priorities 110

Central Valley Improvement Act (1992,USA) 112

Central Valley Project (California, USA)112

citizenship 243climate 34–36communities

Gediz River basin (Turkey) 202water management 39

compensation mechanisms 237–239computers 72coordinative model 14cost recovery ratios 68, 71costs

capital 82full 102–103overhead 79, 80recurring 79, 82river basin organizations 78–80sustainability 102–103, 104, 107

courtsCentral Valley (California) 117Olifants River basin (South Africa)

155

dams 20, 21dismantling 23

dataaccess to external 73accuracy 55–56analysis 56, 73collection 55–56coordination 71dissemination 53, 56–57, 58

irrigation 67–68key for river basins 61, 63, 64–65processing 56, 73quality control 57reliability 55–56remote sensing 62, 66reporting 56–57retention 56retrieval 56river basin management 57–58,

59–60sharing 71storage 56types 55

datasets, global 62decision-making, Central Valley

(California) 120–121Delta–Mendota Canal (California) 112demographics 36–40Department of Water Affairs and Forestry

(South Africa) 149, 153, 154, 158,163, 221

Department of Water Resources (California)114–115

internal conflicts 119departments of agriculture, South African

153–154derivative water 11, 237

Lerma–Chapala basin (Mexico) 221Mexico 141

desalination 8–9developing countries 33–34

basin management 34, 46–48,231–232

climate 35–36informal water sector 40–43institutional models of river basin

management 46–48integrated river basin management

34, 48socioeconomic development 43–45water charges 83water resources management 44

development banksAsian Development Bank guidelines


158Directorate of State Hydraulic Works

(Turkey) 198, 201, 202disasters, protection against 25–26discharge measurement 73

248 Index



domestic water supply 76consumption 77Dong Nai River basin (Vietnam) 177fee collection 84Gediz River basin (Turkey) 200

Dong Nai Basin Council 183–184Dong Nai River basin (Vietnam) 169–192

actors 177–184, 184, 185–187,188–190

boundaries 169–170, 171characteristics 170configuration 221–222flood control 176hydrology 169–170, 171, 172–177institutions 177–184, 185–187,

188–190inter-basin transfers 170key functions 184, 185–187,

188–190legal framework 177–179water balance 170, 172–174

drainage, Gediz River basin (Turkey) 206drinking water 20, 77dykes

construction in Vietnam 182Dong Nai River basin (Vietnam) 176

economic good 81Endangered Species Act (US, 1973) 112environment

Central Valley (California) 113, 219protection 25

environmental advocacy groupsCentral Valley (California) 116–117,


158environmental assemblies 204environmental assets 216, 217–218environmental Kuznets curve 43, 44, 45environmental regulation 202environmental water use 23

Dong Nai River basin (Vietnam) 177Gediz River basin (Turkey) 198–199

equipment repairs/renewals 79equity

payment for water by poor 87river basin organizations 87, 88water management 242

evapotranspirationloss 36

potential global 35expenditure control 85–86

farmerscommercial 13irrigation in France 94Olifants River basin (South Africa)

156–157water pricing 106see also agriculture

fees, collection 42, 84–85finance, river basin organizations 81–82financial productivity 77–78fish, endangered/threatened species 117fisheries 177floods

control in Dong Nai River basin(Vietnam) 176, 189

droughts 25–26flowing water 19–20food

prices 102production 20security 43

France, water management 93–94Authorized User Associations 93,

94–95, 96Basin Committees 94, 95–96, 100,

101controlled 96–100costs

full 102–103sustainability 102–103, 104, 107

essential basin functions 100, 101irrigation charges 100, 102–105key actors 100, 101pricing in allocation control

105–107quotas 98–99

allocation regulation 105–107Regional Development Companies

(SARs) 95, 96, 100, 101State level 96strategic value of water 103–104see also Neste system (France)

functional responsibility charts 66–67

Gediz River basin (Turkey) 62–63, 66,193–213

basin water use 195–199

Index 249



Gediz River basin (Turkey) continuedcase study 194–195closing 194configuration 222–223enabling conditions 206–208hydrology 195institutional environment 200–204,

205, 206–208legal issues 200–204, 205, 206–208location 194nature reserve 198–199policies 200–204, 205, 206–208, 224solutions to issues 208–213water resources 195–200

General Directorate of Rural Services(Turkey) 202, 206, 211, 222, 223

geographical information systems (GISs)56, 72

governance 9, 10, 15centralized 15decentralized 15–16Gediz River basin (Turkey) 212natural resources 33

government agencies, South Africa153–155

green water 37–38groundwater 11

Central Valley (California) 219depletion 27Dong Nai River basin (Vietnam) 173,

189–190irrigation 175–176

Gediz River basin (Turkey) 195, 197,198, 199, 201, 209, 222–223

irrigation 175–176Lerma–Chapala basin (Mexico) 140,

221, 236management 140markets 238Olifants River basin (South Africa)

149open-access resource 236over-abstraction 72over-exploitation 42–43private exploitation 197protection 42recharging 39, 237renewable 110rights in Central Valley (California)

121use regulation in Central Valley

(California) 114, 119

headwork construction in Vietnam 182human needs for water 76, 77hydraulic development 7, 19–20hydraulic structures 20hydraulic technology 1hydrocide 28hydrologic development 19–20hydrologic model 33hydrology 34–36

Central Valley (California) 110–113Dong Nai River basin (Vietnam)

169–170, 171, 172–177Gediz River basin (Turkey) 195Olifants Basin (South Africa)

148–151hydropower

Dong Nai River basin (Vietnam)174–175, 189

Gediz River basin (Turkey) 198

income 216, 217indicators for river basins 61, 63, 64–65

benchmarking 67–68, 69–70industrial growth 44–45

Gediz River basin (Turkey) 200industrialists 13industry

Dong Nai River basin (Vietnam)177

Gediz River basin (Turkey) 198,199–200, 204, 207

Lerma–Chapala basin (Mexico) 131Olifants River basin (South Africa)

157–158payment for water 83water service providers in South

Africa 163–164water use 75, 76

informationattributes 207, 225guidelines 61needs 55

basin level 58, 61–63, 64–65,66–67

changing 57–58, 59–60irrigation system level 67–68,

69–70strategy formulation 53–55transparency 73, 87–88use/users 51, 52, 53

information systems 53–57

250 Index



infrastructuremanagers 79provision 80repairs/renewals 79resources 227water rights 235

institutional change 46–47river basin closure 228–229

institutionsaccelerated reform 33adaptive 27arrangements 4–5Central Valley (California) 113–117,

118, 119–121decentralized in developing countries

36demographic factors 37Dong Nai River basin (Vietnam)

177–184, 185–187, 188–190effectiveness 5–6Gediz River basin (Turkey) 200–204,

205, 206–208management 228Olifants River basin (South Africa)

151–158resources 227river basin development 24stakeholder platforms 243tasks 24–26

integrated management frameworks 2complexity 243gradual 90

integrated river basin management (IRBM)2, 31–33

developing countries 34, 48Murray–Darling River system

33–34integrated river basin model 62–63, 66integrated water resource management

(IWRM) 2–3investment

capital 79, 82irrigation 104

irrigation 16–17agricultural productivity 76basin management 16–17charges 89, 100, 102–105costs 83decline in Taiwan 44–45developing countries 38Dong Nai River basin (Vietnam) 173,

175–176, 189, 192

efficiency 16fee collection 84Gediz River basin (Turkey) 196–197,

199, 202–203, 207, 222, 237groundwater 175–176investment regulation 104Lerma–Chapala basin (Mexico)

129–130, 220–221, 237locally managed 23–24

closing river basins 233–234sustainability 234–235, 240water productivity increase

239–240Neste system (France) 99semi-arid areas 231services in water-scarce basins

231–244smallholder 157South Africa 145–146stakeholders in France 94–96strategic value of water 103–105surface systems 236–237viability index 84water allocation 139

reduction 23water depletion 20water quality 16water-scarce river basins 231–244see also locally controlled irrigation

system (LCIS)irrigation boards, South Africa 156irrigation management companies 180irrigation management transfer 67, 79, 80,

231, 232initiatives 41Mexico 129–130

irrigatorscommercial 156–157poor 13, 157

Kuznets curve hypothesis 43, 44, 45

land rights 33legislation/legal authority 14, 41,

225–226Central Valley (California) 226Dong Nai River basin (Vietnam)

177–179, 221–222, 226Gediz River basin (Turkey) 200–204,

205, 206–208, 208, 211, 226

Index 251



legislation/legal authority continuedOlifants River basin (South Africa)

151–158, 221South Africa 166, 225–226

Lerma–Chapala basin (Mexico) 125–143aquifers 127area 126–127Basin Council 220climate 127closure 128configuration 220–221derivative water 140–141groundwater management 140industry 131irrigated agriculture 129–130, 224,

237lake flows 141lake levels 127–129lake volume 138–139river basin council 134–135, 141river basin management functions

135, 136stakeholders 132–135surface water allocation 137–140urban water supply 130–131wastewater 141water allocation for irrigation 139water balance 126–131water consumption 127water crisis 125–126water management organizations

132–135water rights 131–132

liberal democratic theory 243local government, Olifants River basin

(South Africa) 155locally controlled irrigation system (LCIS)

51data 72–73

dissemination 53information needs 67–68, 69–70,

72–73management processes 55personnel training 73

managementdiscipline range 229institutions 228key processes 54–55public involvement 229reform 9

resources 228see also water management

management information systemformulation 53–55

markets, inter-sectoral 238–239Ministry of Agriculture and Rural

Development (Vietnam) 178–179,180–181, 182

Ministry of Environment (Turkey)201–202, 207, 210–211

municipal water supply, Gediz River basin(Turkey) 197–198, 199–200

municipalitiesGediz River basin (Turkey) 202Olifants River basin (South Africa)

155Murray–Darling basin (Australia) 31,

33–34

National Water Act (1996, South Africa)152, 224, 242

national water sector profiles 61, 62natural resources

availability 44, 45governance 33

nature reserve, Gediz River basin (Turkey)198–199

navigation, river 177Neste system (France) 96–100

Basin Committee 100, 101configuration 218costs 102–103irrigation 99location 97management rules 98–99quotas 98–99, 105–106Regional Development Companies

(SARs) 100, 101users 97–98water management 99–100

non-governmental organizations (NGOs)13, 39

Gediz River basin (Turkey) 203–204,206, 207, 212

officials, elected 117Olifants River basin (South Africa)

145–168apartheid legacy 224climate 146–147

252 Index



configuration 221development banks 158enabling conditions 164–166environmental advocacy groups 158hydrological areas 147hydrology 148–151industry 157–158informational attributes 165–166institutional environment 151–158legal issues 151–158location 146–147policy issues 151–158, 224regions 147–148river basin management functions

158, 159–162, 163–164water resources management 163

option contracts 239organizations see river basin organizations

planning function 119, 229discipline range 229Gediz River basin (Turkey) 204, 210,

211policies 6–7

Dong Nai River basin (Vietnam) 222,224

Gediz River basin (Turkey) 200–204,205, 206–208, 212, 224

Lerma–Chapala basin (Mexico) 224pro-poor 240–242

politics 223–225Gediz River basin (Turkey) 207, 212,


164–165river basin management 228

pollution 8, 25, 72Dong Nai River basin (Vietnam) 190

population density 36–37, 38tropical countries 39

poverty 26–27, 77alleviation 43, 232equity in payment for water 87financial productivity of water 77–78Gediz River basin (Turkey) 204Olifants River basin (South Africa)

157, 167river basin management 240–242rural 102structural changes in society 242

power redistribution 243

precipitation 23arid areas 36Central Valley (California) 110–111harvesting 39rights 36semi-arid areas 36

pricing 33, 41–42adjustment process 102best practice 106–107over-consumption 106regulation 106step 106, 107water allocation control 105–107

private sector, Gediz River basin (Turkey)212–213

privatizationengineering design and construction

companies 180Olifants River basin (South Africa)

152–153service providers 33, 228

problem-solving behaviour 110property regimes 235Provincial Agricultural and Rural

Development Service (Vietnam)180

public agencies, Gediz River basin(Turkey) 201

public awareness in Turkey 209public procedures 14public transactions 14pump cooperatives 197pumps 21

quotas 98–99allocation regulation 105–107

rainfalldeveloping countries 35–36global distribution 34–35Olifants Basin (South Africa) 148

rainwater diversion/harvesting 39record keeping 210Regional Development Companies (SARs,

France) 95, 96, 100, 101regional impacts of water use 1–2regulation 228remote sensing 72

data collection 62, 66rent-seeking behaviour 14

Index 253



reporting 210reservoirs

Dong Nai River basin (Vietnam) 176Gediz River basin (Turkey) 198

resourcesdevelopment in Dong Nai River basin

(Vietnam) 188Gediz River basin (Turkey) 208management 228redistribution 242, 243river basin management 226–227South Africa 166see also natural resources; water

resourcesresponsibility charts 66–67return flows 237

Gediz River basin (Turkey) 206Olifants River basin (South Africa)

149–150riparian doctrine 235–236riparian rights 37, 114river basin(s)

adaptive institutions 27basin-wide impacts of water use 1–2closure 8, 9, 23, 194

appropriation doctrine 236institutional changes 228–229locally managed irrigation

233–234pro-poor policies 242water rights 236

development 22, 24, 25, 26–27pathways 23, 27–28progression 21–22

development phases 20–24, 24–25,26–27

information needs 57–58, 59–60disaster protection 25–26exploitation 8functions 10–11

Dong Nai River basin (Vietnam)184, 185–187, 188–190

essential 100, 101Gediz River basin (Turkey) 204,

205, 206institutional concerns 24institutional essential tasks 24–26maturation model 7quantity constraints 216, 217reallocation 8, 22, 25, 26, 27size 215, 216, 217utilization 22, 24–25, 27

river basin management (RBM) 3, 7–11,12, 13–16, 33, 215, 216, 217–229

administrative model 33Central Valley (California) 109,

121–122centralized 15complexity 7–8context 7–9coordinating mechanisms 33decentralized 15–16democratization 243developing countries 231–232dimensions 215, 216, 217–218Dong Nai River basin (Vietnam) 191drivers 215, 216, 217–218enabling conditions 11, 13–14,

223–227essential functions 117, 118, 119fee collection 84–85functions 10–11

Lerma–Chapala basin (Mexico)135, 136

governance 9, 10, 15–16hydrological model 33informational attributes 13–14institutional effectiveness 5institutional models 46–48interactional analysis 11matrix 11, 12mechanisms 218–227organizational configurations

14–16planning council for Dong Nai River

basin (Vietnam) 191political attributes 13politics 228poverty 240–242representation 242–244resources 13, 14, 226–227services 227–228stakeholders 242, 243water account 9

river basin organizations 4charge assessment methods 83–84charging impacts 86–87, 89costs components 78–80council for Lerma–Chapala basin

(Mexico) 134–135, 141equity 87, 88expenditure control 85–86financial flows 89, 90, 91financing systems 88–91

254 Index



funding 79sources 81–82

gradual integration 90infrastructure managers 79regulation 78–79regulatory functions 79, 80–81revenue sources 82service delivery 80–81service providers 79

riverine ports, Dong Nai River basin(Vietnam) 189

riverscumulative uses of resources 1–2hydrologic regime change 25return flow 83

runoff, mean annual 149

Sacramento–San Joaquin Valley(California, USA) 112

Sai Gon River (Vietnam) 177salinity credits 33salinization 25semi-arid areas

irrigation 231precipitation 36

service delivery 80–81agencies 89function 88

service providers 79, 81, 228Central Valley (California) 115privatization 33, 228

Shasta dam (California) 112smallholders, irrigation 157small-scale water user forums (SWUFs)

157social analysis levels 46–47social democracy 243socioeconomic development level 43–45South Africa

equity in water management 242government agencies 153–155water rights 237see also Olifants River basin (South

Africa)South African Development Trust 154stakeholders 7, 10

data ownership 58Dong Nai River basin (Vietnam)

179–184France 94–96Gediz River basin (Turkey) 207

Lerma–Chapala basin (Mexico)132–135

Murray–Darling River system 33Olifants Basin (South Africa) 221river basin management 242, 243

standard setting 228State Planning Organization (Turkey) 202State Water Project (California, USA) 112State Water Resources Control Board

(California) 114–115, 116surface water 11

allocation 119, 137–140Central Valley (California) 219constrained availability 227developed resources 110Dong Nai River basin (Vietnam) 189Gediz River basin (Turkey) 195, 196,

200–201, 208–209, 222–223Lerma–Chapala basin (Mexico)

137–140markets 238Olifants Basin (South Africa) 149

sustainabilitycosts 102–103, 104, 107locally managed irrigation 234–235,

240water management 5–6

tankswater collection 36see also water storage

Tennessee Valley Authority (TVA) 3–4, 15model 34

timelines 66trans-basin diversion 8tribal leaders, Olifants River basin (South

Africa) 154–155tubewell owners 42Turkey 193, 194, 226

see also Gediz River basin (Turkey)

urban watersupply in Lerma–Chapala basin

(Mexico) 130–131use in Central Valley (California)

112–113urbanization 44–45

charging impacts 86, 87inequity 87Turkey 193

Index 255



user pays principle 33–34

Vietnamcivil society 183enabling conditions 224management structure 179–182provincial actors 182see also Dong Nai River basin

(Vietnam)villages

Gediz River basin (Turkey) 202groundwater cooperatives 197

wastewater 237Lerma–Chapala basin (Mexico) 221Mexico 141Turkey 209

wateraccess in Olifants River basin (South

Africa) 152capital investment 82consumption 75

Lerma–Chapala basin (Mexico)127

depletion 25distribution 25exports from Olifants River basin

(South Africa) 149–150financial productivity 77–78flowing 19–20human demands 1human needs 76, 77imports to Olifants River basin (South

Africa) 150impounding of large bodies 25loss 39–40ownership 36price rises 100, 102reallocation 8, 22, 25, 26, 27regulation in Central Valley

(California) 115–116retail delivery services 119shortages in Dong Nai River basin

(Vietnam) 189strategic value 103–105value 26withdrawal 98

water account 9water accounting framework 19, 20, 21

water productivity 38

water allocation 25Dong Nai River basin (Vietnam)

188–190environmental needs 33Gediz River basin (Turkey) 204, 205,

206, 209–210, 223mechanisms 235–237Olifants River basin (South Africa)

167pricing in control 105–107pro-poor policies 241

water balanceDong Nai River basin (Vietnam)

173–174Olifants River basin (South Africa)

150–151water boards (South Africa) 156water budgets 62water charges, collecting 42, 84–85water control agency 116water courts

Central Valley (California) 117Olifants River basin (South Africa)

155water demand 75–77

Central Valley (California, USA)112–113

Dong Nai River basin (Vietnam)172–173

Gediz River basin (Turkey) 200Olifants River basin (South Africa)

150, 151water development in France 93water diversions

allocation 38cap 33permits 33

Water Law (1992, France) 96Water Law (1999, Vietnam) 177–179, 183water law reserves, South Africa 152water licensing

compulsory 242see also abstraction, licences

water managementequity 242institutional effectiveness 5intractable/wicked problems 5organizations in Lerma–Chapala basin

(Mexico) 132–135policies 6–7sustainable 5–6

water markets, inter-sectoral 238–239

256 Index



water policyOlifants River basin (South Africa)

151–158South Africa 146

water productivity 38increase 239–240

water quality 216, 217assurance 119Central Valley (California) 219Dong Nai River basin (Vietnam) 190,

191enforcement of standards 25, 72, 119Gediz River basin (Turkey) 200, 205,

206, 208–209, 211, 223irrigation 16monitoring 25Olifants River basin (South Africa)

151, 167–168water resources

capture 9Central Valley (California) 121, 219depletion 23development 19–20

best practice 106–107Dong Nai River basin (Vietnam) 188enforcement of standards 72events 66Gediz River basin (Turkey) 195–200,

211, 212Lerma–Chapala basin (Mexico)

220–221management

developing countries 44Olifants River basin (South Africa)

163need changes 72Neste system (France) 98open-access 236overexploited 23, 232pressures 67–68reallocation 8, 22, 25, 26, 27South Africa 166Vietnam 182

water rights 33, 36, 72access to 232appropriative 114, 236basin closure 236Central Valley (California) 113–114,

121Dong Nai River basin (Vietnam) 191Gediz River basin (Turkey) 200–201,

210, 212, 223

Lerma–Chapala basin (Mexico)131–132, 221

Olifants River basin (South Africa)221

riparian 37, 114South Africa 237systems 235–237transfers 239

water scarcity 7, 8, 26, 241irrigation service provision 231–244

water sector organization 40–43water security 241water service providers (South Africa)

156, 163–164Water Services Act (1996, South Africa)

152water storage

Dong Nai River basin (Vietnam) 172,191

infrastructure 227local 37Olifants River basin (South Africa)

149–150see also reservoirs

water supply 218Central Valley (California) 110–112Dong Nai River basin (Vietnam) 170,

172, 177, 189Gediz River basin (Turkey) 197–198,

199–200Neste system (France) 98

water systems, closed 110water transfers 237–239

inter-basin 23inter-sectoral 232trans-basin diversion 8

water use 25, 75–77basin-wide impacts 1–2changing patterns 19Dong Nai River basin (Vietnam)

174–177efficient 16environmental 23

Dong Nai River basin (Vietnam)177

Gediz River basin (Turkey)198–199

evaporative 23Gediz River basin (Turkey) 196–198monitoring 42outside-basin 198over-withdrawal 25

Index 257



water use continuedregional impacts 1–2within-basin 197–198

water usersCentral Valley (California) 115, 219charge impacts 89fees 81–82Gediz River basin (Turkey) 209–210Neste system (France) 97–98Olifants River basin (South Africa)

224

small 42water users associations (WUAs) 13, 41

developing countries 231–232Mexico 130, 133, 139–140Olifants River basin (South Africa)

153, 164performance benchmarking indicators

69–70South Africa 153, 156, 164

water-harvesting devices 20, 21wetlands, drying up 25

258 Index

