SWIM

SWIM Paper

System-Wide Initiative on Water ManagementSWIM

8

INTERNATIONAL WATER MANAGEMENT INSTITUTEP O Box 2075 Colombo, Sri Lanka

Tel (94-1) 867404 • Fax (94-1) 866854 • E-mail iwmi@cgiar.orgInternet Home Page http:/ /www.cgiar.org/iwmi

ISBN 92-9090-380-5

CGIAR

ISSN: 1028-6705

Multiple Uses of Water inIrrigated Areas: A Case Studyfrom Sri Lanka

Editors:Margaretha BakkerRandolph BarkerRuth Meinzen-DickFlemming Konradsen

IFPRI

SWIM Papers

In an environment of growing scarcity and competition for water, increasing theproductivity of water lies at the heart of the CGIAR goals of increasing agriculturalproductivity, protecting the environment, and alleviating poverty.

TAC designated IWMI, the lead CGIAR institute for research on irrigation andwater management, as the convening center for the System-Wide Initiative on WaterManagement (SWIM). Improving water management requires dealing with a range ofpolicy, institutional, and technical issues. For many of these issues to be addressed,no single center has the range of expertise required. IWMI focuses on the managementof water at the system or basin level while the commodity centers are concerned withwater at the farm and field plot levels. IFPRI focuses on policy issues related to water.As the NARS are becoming increasingly involved in water management issues relatedto crop production, there is strong complementarity between their work and many ofthe CGIAR centers that encourages strong collaborative research ties among CGIARcenters, NARS, and NGOs.

The initial publications in this series cover state-of-the-art and methodology papersthat assisted the identification of the research and methodology gaps in the priorityproject areas of SWIM. The later papers will report on results of SWIM studies,including intersectoral water allocation in river basins, productivity of water, improvedwater utilization and on-farm water use efficiency, and multiple uses of water foragriculture. The papers are published and distributed both in hard copy andelectronically. They may be copied freely and cited with due acknowledgment.

Randolph BarkerSWIM Coordinator

ii

SWIM Paper 8

�������������� �����������������������������

��������������������

Editors:

Margaretha BakkerRandolph BarkerRuth Meinzen-DickFlemming Konradsen

International Water Management InstituteP.O. Box 2075, Colombo, Sri Lanka

ii

The editors: Margaretha Bakker is Associate Expert in the Health and EnvironmentProgram at the International Water Management Institute (IWMI); Randolph Barker isSWIM Coordinator and Senior Advisor to the Director General at IWMI; Ruth Meinzen-Dick is Research Fellow in the Environment and Production Technology Division atInternational Food Policy Research Institute; and Flemming Konradsen is EnvironmentalHealth Biologist at IWMI.

The authors: Rekha Mehra is Director of Economic Analysis at the International Centerfor Research on Women (ICRW); Anneke Van Eijk is Hydrologist at the GovernmentService for Land and Water Management, the Netherlands; Fuard Marikkar is Consultant,Yutaka Matsuno is Irrigation Engineer, David Molden is Research Leader of thePerformance and Impact Assessment Program, R. Sakthivadivel is Senior IrrigationSpecialist, Wim Van der Hoek is Research Leader of the Health and Environment Program,and Parakrama Weligama is Consultant/ Agricultural Economist, all at IWMI. The editorsMargaretha Bakker, Ruth Meinzen-Dick, and Flemming Konradsen are the other authors.

This work was undertaken with funds specifically allocated to the System-Wide Initiativefor Water Management (SWIM) by the Ford Foundation, the Rockefeller Foundation, andthe governments of Australia, Denmark, France, Germany, and the Netherlands.

Bakker, M., R. Barker, R. Meinzen-Dick, and F. Konradsen, eds. 1999. Multiple uses ofwater in irrigated areas: A case study from Sri Lanka. SWIM Paper 8. Colombo, Sri Lanka:International Water Management Institute.

/ water management / water allocation / water use efficiency / irrigated farming / waterresources development / water policy / water quality / domestic water / water userassociations / water rights / gender / case studies / households / pricing / water costs /Sri Lanka/ Kirindi Oya /

ISBN 92-9090-380-5ISSN 1028-6705

© IWMI, 1999. All rights reserved.

Responsibility for the contents of this paper rests with the editors.

The International Irrigation Management Institute, one of sixteen centers supported bythe Consultative Group on International Agricultural Research (CGIAR), was incorporatedby an Act of Parliament in Sri Lanka. The Act is currently under amendment to read asInternational Water Management Institute.

iiiiii

Contents

Abstract vii

1. Introduction 1Ruth Meinzen-Dick and Flemming Konradsen

2. Description of the Study Area 3Fuard Marikkar

3. Water Uses in the Kirindi Oya Subbasin 6Anneke van Eijk, David Molden, and R. Sakthivadivel

4. Water Quality Issues 13Yutaka Matsuno

5. The Institutional Environment 16Margaretha Bakker and Ruth Meinzen-Dick

6. Household Uses of Water 21Flemming Kondradsen, Rekha Mehra, and Parakrama Weligamage

7. Valuing the Multiple Uses of Water 26Margaretha Bakker

8. Complementarities, Competition, and Conflicts 31Ruth Meinzen-Dick and Margaretha Bakker

9. Future Directions for Research 35Wim van der Hoek

10. Conclusions: Implications for Policy and Management 38Ruth Meinzen-Dick

Annex 1 Water flow chart 42

Annex 2 Other valuation techniques 43

Literature Cited 45

iv

Figures, Tables, and Boxes

Figure 1 Map of the study area 4

Figure 2 Water accounting for Kirindi Oya subbasin, 1995–96 and 1996–97 11

Table 1 Area cultivated with paddy and OFCs in Kirindi Oya Irrigation System(average from maha 1990 to yala 1997) 7

Table 2 Piped water supply schemes in the study area 9

Table 3 Classification and quantification of the water uses served by the irrigation anddomestic service in the Kirindi Oya subbasin in MCM for 1995–96 and 1996–97 12

Table 4 Water quality measurements and standards in the Kirindi Oya System (September 1997) 14

Table 5 Classification of irrigation water salinity 15

Table 6 Water allocation pattern in Kirindi Oya Irrigation System 17

Table 7 Government agencies and user groups representing different types of water uses 20

Table 8 The importance of irrigation water as first priority water sourcein comparison to other sources of water for a variety of uses 22

Table 9 Results of ranking exercise for domestic water uses and sources 25

Table 10 Value added per m³ of water for different productive water uses (in 1997 rupees) 30

Table 11 Conflicts, competition, and complementarity of water uses in Kirindi Oya 32

Box 1 Water accounting categories 10

Box 2 Formula to calculate the value added of water 28

Box 3 Value added of volume of water consumed, diverted, and total water supply 29

v

Acronyms

BWMC Bundala Wetland Management CommitteeCOFO Cattle Owners’ Farmer Organizationcumecs cubic meter per secondDCO Distributary Channel OrganizationDWLC Department of Wildlife ConservationEC Electrical ConductivityET EvapotranspirationFAO Food and Agriculture OrganizationFCG Field Channel GroupFCS Fisheries Cooperative SocietyFO Farmer OrganizationGVO Gross Value of OutputID Irrigation Departmentha hectareICRW International Center for Research on WomenIMD Irrigation Management DivisionINMAS Integrated Management of Major Irrigation SystemsIWRM Integrated Water Resources Managementkg kilogramkm kilometerKOISP Kirindi Oya Irrigation and Settlement ProjectLB Left BankMCM Million Cubic Metersmg/l milligram/ litermm millimetermS/cm milliSiemens/ centimeterMSL Mean Sea Levelmt metric tonNGO Non-Governmental OrganizationNVO Net Value of OutputNWS&DB National Water Supply & Drainage BoardOFC Other Field CropsO&M Operation and MaintenancePMC Project Management CommitteePPM Parts Per MillionRB Right BankRs Sri Lankan Rupees, at time of the study US$1.00 = Rs 58.80SPC Subproject CommitteeTDS Total Dissolved SolidsWHO World Health Organization

vi

CGIAR Centers

CIAT Centro Internacional de Agricultura TropicalCIFOR Center for International Forestry ResearchCIMMYT Centro Internacional de Mejoramiento de Maize y TrigoCIP Centro Internacional de la PapaICARDA International Center for Agricultural Research in the Dry AreasICLARM International Center for Living Aquatic Resources ManagementICRAF International Council for Research in AgroforestryICRISAT International Crops Research Institute for the Semi-Arid TropicsIFPRI International Food Policy Research InstituteIWMI International Water Management InstituteIITA International Institute of Tropical AgricultureILRI International Livestock Research InstituteIPGRI International Plant Genetic Resources InstituteIRRI International Rice Research InstituteISNAR International Service for National Agricultural ResearchWARDA West Africa Rice Development Association

vii

Abstract

Water is being transferred out of agriculture tomeet the growing demand in other areas, oftenwithout an agreement of or compensation tofarmers with irrigated land and water rights.Furthermore, there is a failure to recognize thatirrigation systems supply water not only for themain fields, but also for domestic uses, homegardens, trees and other permanent vegetation,and livestock. Other productive uses includefishing, harvesting of aquatic plants and animals,and a variety of other enterprises such as brickmaking. In addition, irrigation systems can have apositive or negative effect on wildlife habitats.Thus, the withdrawal of water affects the ruralhousehold, rural economy, and the environment ina number of ways.

This paper argues that to ensure efficient,equitable, and sustainable water use, to reducepoverty and improve the well-being of thecommunity, irrigation and water resources policiesneed to take into account all uses and users ofwater within the irrigation system. The multipleuses of water in the Kirindi Oya irrigation systemare examined in this paper. An interdisciplinarygroup of scientists has investigated a number ofareas including water accounting, water quality,household water use, the valuing of water foralternative uses, and the complementarities,competition, and conflicts among uses and users.

Among the various management issues andthe problem areas identified, the allocation ofirrigation water particularly in periods of scarcity, isperhaps the most critical decision and one that hasprovoked considerable conflicts in the past. Themost appropriate water level to be maintained inthe tanks is another critical decision, with evidenceto suggest that improved management of the tanksystems in the wet season could lead to savingsof water and expansion of irrigated area in the dryseason. Finally, the study highlights the importanceof water quality not only for domestic use but alsofor fishing and wildlife.

Because of the complexity of the issuesinvolved, however, the results of this study mustbe seen as a first step towards the development ofa suitable methodology for studying the range ofuses and interactions among them. While theexact uses and users of water and their relativeimportance will vary from one irrigation system toanother, the issues identified in this pilot study ofKirindi Oya have broader implications for watermanagement policies in Sri Lanka and elsewhere.These relate to the allocation of water betweenirrigation and other sectors; measures of waterquality and efficiency of use; and mechanisms toinvolve all stakeholders in negotiations over waterallocation and use.

1

The 1990s have witnessed a dramatic shift in thepriorities for water resource allocations anddevelopment. Irrigation, which was once seen asthe essential factor for securing sufficient foodproduction, is now seen as a “low-value” use ofwater compared to its municipal, industrial, andeven the environmental uses. Irrigation no longerreceives priority either in the allocation of fundsfor project developments or in the allocation ofwater itself. In a growing number of cases, wateris transferred out of agriculture to meet thegrowing demand in other areas, sometimes withcompensation for the farmers with irrigated landand water rights, sometimes without theiragreement or compensation.

Irrigated agriculture is the biggest consumerof world’s fresh water resources. On a globallevel, irrigation comprises 72 percent of theaverage per capita diversions, with industrial anddomestic sector accounting for 19 percent and 9percent, respectively, of the average per capitawater diversions (Seckler et al. 1998). However,these figures are misleading since irrigationsystems often provide water for many otherpurposes besides irrigated agriculture. Evenwithin the agriculture sector, irrigation systemssupply water not only for the main fields, but alsofor home gardens, trees, and other permanentvegetation through elevated water tables, andprovide water for livestock. Other productive usesinclude fishing, harvesting of aquatic plants andanimals, and a variety of other enterprises suchas brick making. Furthermore, irrigation systemsare generally a major source of domestic water

Multiple Uses of Water in Irrigated Areas: A caseStudy from Sri Lanka

Margaretha Bakker, Randolph Barker, Ruth Meinzen-Dick and Flemming Konradsen

1. Introduction

supply, and can have important environmentalfunctions in supporting plants, birds, and animals.Within some systems, especially in semiaridareas, irrigation water may be the only source ofdomestic water available to the households.Recognizing the importance of this source to thehealth and the social well being of the communityis essential and a more holistic approach toirrigation water management may providesubstantial health benefits. In certain areas,irrigation water may substantially increase the percapita availability of water throughout the year,with an impact of decreasing the prevalence ofskin and eye diseases and the incidence ofhygiene-related diseases. Hence, it is misleadingto equate irrigation only with the agriculturesector, just as it is misleading to equatemunicipal consumption with domestic water use,since much of municipal water goes forproductive enterprises, even gardens and lawns.

Whether or not multiple uses and users havebeen considered in ex ante project assessments,they are generally not included explicitly in theongoing management of irrigation systems. Onereason for this is that most agencies dealing withwater resources have only sectoral responsibilityto deal with either irrigation or drinking water orindustry, or environment. The government as awhole has the responsibility for overall water use,but the implementing agencies have neither themandate nor the incentive to balance the needsof various users (Yoder 1983; IIMI 1995a).

Taking all uses and users into account, watermanagement could contribute to higher total

2

productivity of irrigation systems. Currently, manyirrigation systems are being “modernized” toimprove water use efficiency. Measures such aslining canals or implementing rotational watersupplies may increase the efficiency of irrigationat the farm or system level, but if the availabilityof (ground)water is reduced elsewhere, it mayreduce the overall output, and hence theefficiency at the system or basin level. In somecases, complementarities between uses may befound, but in many other cases there will betrade-offs between different uses, ineither quantity or quality of water, or both(Konradsen et al. 1997).

This paper argues that recognizing themultiple uses of water in irrigation systems iscritical for better water allocation policy. The mainresearch hypothesis is that the value of irrigationwater for non-crop purposes will be of significantmagnitude when compared with the value for usein field crop production.

There are four research issues examined inthis report, drawing on evidence from multipleuses of water in the Kirindi Oya irrigation systemin Sri Lanka. The first issue focuses on a moreaccurate assessment of all the uses and users ofwater in an irrigation system since water inirrigation systems has been undervalued becauseof the failure to recognize the different uses. Thisaccurate assessment will better inform decisionsabout allocating water (and financial resources)between irrigation and other uses. Second, anendeavor is made to value these different usesof water. Third, the interactions and conflictsamong different sectors and users of irrigationwater are identified. For instance, therehabilitation efforts of irrigation systems willaffect the other uses and users of water as well.This raises the question of how to cope with therising competition for water between multiplekinds of users and allocate the water in waysthat are equitable, efficient, and sustainable.Fourth, the fact that the full spectrum of uses ofirrigation water changes the institutional picture ofwater management is recognized. It is no longer

only the Irrigation Department (ID), nor even justthe farmers who play a role in water management,but a much broader range of stakeholders: menand women, different occupations, and differentgovernment departments. Because of thecomplexity of the research issues involved, it is achallenge to develop a suitable methodology tostudy the range of uses and the interactionsamong them. This study has developed a set ofmethodological tools to define and value multipleuses of water. The objective of this paper is not topresent a complete assessment of multiple usesin Kirindi Oya, but rather to lay out the issues andpresent a pilot study to test the methodology. Thismethodology should be refined, adapted, andapplied more completely in a range of sitesbefore conclusive statements can be made aboutthe management of irrigation systems asmultiple-use water resources.

Outline of the Paper

The following chapter gives a description of theirrigation infrastructure and the physical andsocioeconomic characteristics of the study area.The information in this chapter is based on anearlier research done by IIMI as well as on theresearch done during this study. Chapter 3describes the present water use pattern andquantifies the uses of different sectors by using awater accounting procedure. After thequantification of water use, chapter 4 deals withwater quality issues and requirements forirrigation, drinking, and the environmental uses ofwater. Chapter 5 describes the institutionalenvironment in the Kirindi Oya irrigation system,focusing on water-related institutions, land andwater rights, and a variety of governmental andwater user organizations. The description andanalysis of the institutional framework go beyondthe formal, state-defined rights and organizations,and also consider the customary rights and avariety of formal and informal user organizations.In chapter 6, the findings of a household water

3

use survey are presented. A description is givenof the different water sources used for variouspurposes, with an emphasis on the domestic usesof water. Also, an attempt is made to assess theimportance of irrigation water and nonirrigationwater sources for various household water usesby gender of the primary users. Chapter 7discusses aspects of water pricing, economicvalue, and other values of water. The gross valueand the net value added per m³ of water fordifferent productive uses are calculated usingprimary and secondary data. In chapter 8,

complementarities, competition, and conflictsbetween different uses and users are examined.The major types of interactions between uses arealso presented. The penultimate chapterdescribes the future directions for research in aplanned follow-on study and the last chapter,chapter 10, gives the conclusions andimplications for water policy and watermanagement when multiple uses of water arerecognized.

2. Description of the Study Area

Location and Physical Characteristics

The Kirindi Oya Irrigation and Settlement Project(KOISP) is located in the southeastern dry zoneof Sri Lanka, about 260 km from Colombo(figure 1) and is the largest irrigation andsettlement scheme in the south of Sri Lanka. TheKirindi Oya river is 118 km long and is fed by acatchment area of 1,203 km². The irrigationsystem is an expansion of the old Ellagalasystem, which comprised 5 ancient tanks. In the1970s, plans were formulated to rehabilitate andaugment the old Ellagala system. Theimplementation started in 1979, and in 1987, thefirst water supply was issued from the newly builtLunuganwehera reservoir through the new LeftBank (LB) and the Right Bank (RB) canals. Thestudy area is an important tourist destination withthe Bundala National Park situated downstreamof the KOISP and the Yala National Park furtheraway. Also, many pilgrims visit the area on theway to the shrines in Kataragama.

The study area falls within the low-countrydry zone and is situated within 125 meterselevation above sea level (Cooray 1984). It lieswithin the broad, shallow valley of the KirindiOya, with the old Ellagala irrigation system falling

almost entirely within the slightly uplifted flatalluvial plain. The newly developed areas of theirrigation system are situated around the oldareas and have slopes of about 2–6 percent. Thevarious sub-catchment valleys slope into the riverand the tanks. The alluvial plain has an incisedriver, the Kirindi Oya cutting through it, and as aresult, little or no river erosion hazards areencountered. The soils in the new area are moresuited for non-paddy crops while soils in the oldarea are equally suited for paddy and for othercrops.

The climate of the area is tropical and ischaracterized by nearly constant year-roundtemperatures (26–28 oC). Evaporation is uniformthroughout the year, with an annual averageapproximating 2,000 mm. Mean annual rainfall is1,000 mm with the maha (wet) season rainfall(October-March) approximately three times thatof the yala (dry) season (April-September) (IIMI1995a).

Socioeconomic Characteristics

The population of the old area, including theBadagiriya irrigation scheme is estimated as

4

FIGURE 1.Map of the study area.

5

56,222 and that of the new area as 31,528 givinga total population of 87,750 for the entire studyarea. The ethno-religious make up of thepopulation is predominantly Sinhalese (98%). Theaverage household size estimated from fieldsurveys is 4.8 persons per household in the oldarea and 5.1 persons in the new area.

The social composition of the populationreflects the old system/new settlement phases ofirrigation development in the area. Most residentsin the old area are from families that have livedin the area for generations, as would beexpected in an area with a thousand years ofirrigation history. Therefore, 62 percent of theresidents in the old area live in permanenthouses compared with 15 percent of theresidents in the new area. People in the old areahave bigger houses, more consumer durables,and transport (bicycles, two- and four-wheeledvehicles) than those in the new settlement area.Furthermore, homesteads in the old area are welllandscaped with vegetables, fruit trees, and otherpermanent vegetation (IIMI 1995a).

Residents in the new irrigated areas aresettlers who acquired land either after beingdisplaced from old holdings (alternate settlers,55%), or under a government program to allocatenew irrigated land to the landless from otherparts of the country (open settlers, 45%). Manyof the latter reported that political connections intheir place of origin had helped them to acquireland. In addition to the 0.2 hectare highland,each settler was to be allocated 1 hectare ofirrigated land (Stanbury 1989). The poorlydeveloped physical and social infrastructure whenthe project began, combined with the need todevelop their land and homestead sites, madelife difficult for the settlers. For example, forseveral years, the only source of drinking waterwas bowser trucks that delivered water at pointsalong the roadside. Further distress was causedby a lower water availability than was expectedat the time of the project design. Many familieswere forced to migrate seasonally to cities or totheir places of origin and diversify their sources

of income, since livelihoods obtained from theirrigated production alone were insecure.Although it is not formally allowed, some settlersgave the opportunity to someone else to cultivatetheir land.

Under the irrigation development project,there has been a considerable governmentinvestment in the infrastructure and subsidies forhousing. Literacy rates are high in both areas:97.7 percent for men and 94.5 percent forwomen in the old area, and 96.7 percent for menand 93.4 percent for women in the new area.There has been a declining trend in dependenceon agriculture and a rise in salaried employment(including 2% involved in foreign employment).Formal sector employment, especially in thegovernment sector, is higher in the old areabecause of better transport facilities (IIMI 1995a).In 1994, average annual household income in theold area was Rs 66,800 and in the new area,Rs 36,072 (ibid).

Irrigation Infrastructure

The Lunuganwehera reservoir, the construction ofwhich was completed in 1987, provides irrigationwater to 5,400 hectares of the newly developedlands, about 4,200 hectares of the existingirrigated lands under the old Ellagala system, andto 850 hectares under the Badagiriya system.The tanks under the Ellagala system have beenbuilt over a thousand years ago. The old systemwas restored in 1877 by the construction of anew diversion structure (anicut) in the Kirindi Oyariver at Ellagala. The Ellagala anicut diverts waterto 5 ancient and previously independent tanks:Debara Wewa, Tissa Wewa, Weerawila Wewa,Pannegamuwa Wewa, and Yoda Wewa (see alsofigure 1).

The catchment area of the Lunuganweherareservoir is 914 km². The reservoir has a grossstorage capacity of 227 million m³ (MCM) and anactive storage of 200 MCM. The RB main canalof the reservoir is 32 km long ending at the

6

Badagiriya tank, with a discharge capacity of 13cumecs and a command area of 3,500 hectares.The LB main canal is 17 km long with adischarge capacity of 12 cumecs and acommand area of 1,900 hectares.

The Lunuganwehera reservoir is built a shortdistance upstream of the Ellagala Anicut. Afeeder canal from the LB main canal takes waterto the Ellagala diversion. The newly developedareas surround the old irrigated area and draininto the old tanks. About 30 percent of the inflowinto the old system is received directly from theLunuganwehera reservoir and the rest from therainfall runoff from the catchments of the oldtanks and return flow from the water diverted tothe new area.

The areas south of the LB drain into the oldtanks: Debera Wewa, Tissa Wewa, and YodaWewa. Areas east of the RB canal drain into theWeerawila Wewa and Pannegamuwa Wewa andareas south of the RB canal drain into theEmbilikala and Malala lagoons. These lagoonsare part of the Bundala National Park. Embilikala

is an inland lagoon with no direct outfall to thesea, while Malala has a direct outfall to the sea.An incised canal connects the two lagoons.Drainage water from old tanks in the upperreaches (Pannegamuwa Wewa, Debera Wewa,and Tissa Wewa) flows back to the old tanks inthe lower reaches (Weerawila Wewa and YodaWewa) and to the Kirindi Oya river or to otheroutlets to the ocean.

The Badagiriya tank, which is built across theadjoining Malala Ara river basin has its ownindependent catchment of 350 km², and astorage capacity of 11.2 MCM, providing irrigationwater to 850 hectares of land. The Badagiriyatank has been included in the study as it isaugmented by the RB main canal ofLunuganwehera reservoir with a fixed volume ofwater annually. The augmentation had beennecessitated because the water supply to theBadagiriya tank has been reduced considerablyby the construction of many small tanksupstream. The drainage water of Badagiriyaflows into the Malala lagoon.

3. Water Uses in the Kirindi Oya Subbasin

Methodology

Water users in the Kirindi Oya study area are alldependent on the same water from the networkof reservoirs, canals, and rainfall in thecatchment area. They are highly dependent oneach other to obtain adequate water with goodquality. The allocation rules, the means ofdistributing water, and the designs forimprovement require a better understanding ofthe present water use pattern. To understand thewater use pattern in Kirindi Oya, a wateraccounting procedure is used (Molden 1997).This procedure quantifies the different uses ofwater and gives a better understanding of therelative quantities used by different sectors. Italso gives an indication of the performance ofsubbasin water management.

Water accounting classifies water balancecomponents into uses by the different sectors.Water flowing into the subbasin from the KirindiOya river flows to the Lunuganwehera reservoirproviding the main source of water in addition torainfall. The water is put to a beneficial use—drinking water for people, crop consumptive use,and habitat for fish. As a result of the diversionto the use, the water either depletes from thesubbasin, or returns to the subbasin, where thereis a chance that it gets used again. There isconsiderable reuse of return flows in the area,especially within the irrigation sector. Forexample, return flows from the new area arereadily recaptured by the tanks in the old areasand again diverted for agricultural use.

It is useful to study water at three levels:macro or subbasin, mezzo or service, and micro

7

or use level. A macro point of view gives anunderstanding of the overall water resourceavailability. At a finer level of detail, studyingirrigation services gives us an indication of howwell water is being managed. When consideringthe use level, we get a better picture of howfarmers irrigate fields, or how households usewater.

Water Services and Uses

In the Kirindi Oya subbasin, two water servicesare distinguished: irrigation service and domesticservice. The domestic service provides treatedand piped drinking water to some of theinhabitants in the area. The irrigation servicemainly intends to meet the needs of crops, but italso provides domestic water. The paths of waterflow in the study area are shown in annex 1.

Irrigation Service

Irrigated field crop production

The sources of water for the irrigation service arethe water releases from Lunuganwehera that,

together with rainfall provides water for cropgrowth. Within the irrigation service area, two maininterest groups can be distinguished: the newirrigated area, which can be further classified asthe RB and the LB, and the old irrigated area.Most of the supply of the new area is served bythe water remaining after supplying the old area.It is estimated that there are about 300–400agro-wells in the area, providing supplementarywater for cultivating other field crops (OFCs) andto a lesser extent, paddy.

Traditionally, rice is grown in the maha andthe yala seasons if water is available. Newimproved varieties of rice are cultivated inmajority of the area. The rice is broadcast. Themost important OFCs cultivated in the systemare chili, onion, green gram, cow pea, groundnut,and vegetables like brinjal, snakegourd, andtomato. Over the last few years, bananacultivation has been more popular with bananasplanted in paddy fields and areas closer tocanals or other water sources. The average areacultivated with paddy and OFCs and their yieldsare given in table 1.

TABLE 1.Area cultivated with paddy and OFCs in the Kirindi Oya irrigation system (average from maha 1990 to yala 1997).

Irrigable area Maha (average) Yala (average)

Area cultivated with paddy—old area (ha) 4,200 4,092 3,415

Area cultivated with paddy—new area (ha) 5,400 3,703 1,694

Total area cultivated with paddy (ha) 9,600 7,795 5,109

Paddy yield (mt/ha) – 4.22 4.38

Area cultivated with OFCs (ha) – 1,240 881

Total area cultivated in Kirindi Oya Scheme (ha) 9,600 9,035 5,990

Area cultivated in Badagiriya Scheme (ha) 850 665 333

Total area cultivated in KOISP and Badagiriya Scheme (ha) 10,450 9,700 6,323

Sources: Irrigation Department, Department of Agriculture, Irrigation Management Division and Agrarian Services Centers.

8

Domestic use

In dry periods when there is no cultivation, the IDissues water for 48 hours once in every twoweeks to the LB and the RB to provide water forwashing and bathing purposes. In 1995–1996, 1.4MCM was supplied through canals for thispurpose, while in the dry year (1996–1997), 11MCM was supplied. A second source of water fordomestic use, supplied indirectly by the irrigationservice, is well water. Approximately, 2,800homestead wells for domestic purposes aremainly located in the old area. These wells arerecharged by water issued to the irrigated fieldsand rainfall.

Livestock

The main water sources for livestock to drink andbathe are small tanks that obtain their water fromrainfall and the irrigation service. In the dryperiods, these water sources dry up, and theanimals are brought back to the KOISP areamainly to meet their requirements of water fromthe major tanks in the area and for grazing onpaddy stubble. Currently, it is estimated thatthere are over 50,000 cattle and buffaloes, 3,000goats, and 5,000 poultry in the area. Averagemilk yields are low, between 1-2 liters/day/animal.The highest yields are obtained during the wetmonths when food and fodder are available.There are no specific water sources or scheduleddiversions for livestock.

As a consequence of the milk-based curdindustry, there are also a lot of small curd potmaking enterprises. Mud from the tank beds isused to make these pots used for holding curd.Bricks are also made from tank bed mud.

Fisheries

Inland fishing is undertaken in almost all themajor tanks and in the Lunuganwehera reservoir.Depending on the availability, fingerlings areintroduced once a year or once a season.Fingerlings are produced in the fish farms of theDepartment of Fisheries or the NGOs. Fish catchis high when the water levels in the tanks go

down during the dry months (April–September)when fish get concentrated in the shallow waterand are easier to catch. There are no scheduledwater diversions for fisheries. Only the deadstorage or the balance remaining after theirrigation requirements are met can be used ashabitat for fish.

Forest, scrubs and grasses, and chena

Within the study area, there is a significant areacovered with natural vegetation, homesteads, andchena (slash and burn) cultivation. These landuse types benefit from the increase of soilmoisture and groundwater levels due topercolation and seepage from paddy fields andcanals. Much of this use can be consideredbeneficial, because the people use the wood asfuel for cooking, the forest conserves the land,and the grasses and scrubs are used for grazingof cattle.

Domestic Service

In 1997, about 95 percent of the population ofthe new area and 48 percent of the population inthe old area had access to piped water(household survey 1997). This water is treatedand comes from different sources as shown intable 2. In 1996, about 1 MCM of piped waterwas supplied.

Domestic and stand post connections useabout 75 percent of the total supply. Theconstruction and industry sector consumes 0.1percent, tourist hotels 0.05 percent, thecommercial sector 4.6 percent, religiousinstitutions 3.2 percent, and the governmentinstitutions including schools and hospitals 17percent.

The National Water Supply and DrainageBoard (NWS&DB) estimated the total annualdemand for domestic water at 1.7 MCM. Thedomestic water service provides about 1 MCM ofthe demand. The remaining 0.7 MCM is met bythe irrigation service either directly when canalwater is issued, or indirectly when it is pumped

9

from wells. Especially people in the old area relymore on wells for their domestic water uses.About 50 percent of the population in the oldarea use well water as their main source ofdrinking water compared to 5 percent of thepopulation in the new area (household survey1997).

Chena Cultivation

Within the subbasin, the practice of chena orshifting cultivation exists on 3,700 hectares.Chena cultivation relies almost entirely on rain,and does not benefit from the irrigation or thedomestic services.

Environmental Uses

The wetland system of the Bundala NationalPark receives drainage flows from the KOISP.Water outflow from tracts 5, 6, and 7 drains intothe Bundala Park and may affect the ecosystem(see figure 1). A proper analysis of the situationwould require further water accounting studies forthe adjacent river basins. At this moment thereare no regulations about the amount of water tobe drained into the Bundala National Park. Theminimum or the maximum amount of waterrequired to sustain the ecosystem of the park

TABLE 2.Piped water supply schemes in the study area.

Scheme Source Supply area Treatment Amount supplied in1996 (x1000 m3/ year)

Lunuganwehera Lunuganwehera New area, Badagiriya Erosion + sandreservoir scheme filtration + chlorination 570*

Tissamaharama/ Kirindi Oya Tissamaharama/Debera Wewa Debera towns Chlorination 247

Kirinda Groundwater Kirinda area Chlorination 154

Total 971

*Not all of this goes to the piped water supply system. Project bowsers supply 19 percent of the people in the new area. This water comes fromthe Lunuganwehera drinking water supply scheme.

Source: (National Water Supply and Drainage Board).

should be further studied to understand how muchwater should be committed to the area to sustainthe park and how much of drainage outflow can beallowed.

Water Accounting

For water accounting, a water balance isconstructed that considers inflows and outflowsfrom an identified domain of interest, specifyingspatial and temporal boundaries. For this study,the Lunuganwehera reservoir is the Northernboundary (the reservoir itself is not included inthe study) and the other boundaries coincide withthe Kirindi Oya river basin boundaries. Somedefinitions required for the analysis are given inbox 1.

The accounting categories are derived from awater balance for the study period. Any changein the groundwater or the surface water storageis equal to the volume of rainfall plusLunuganwehera reservoir releases minus thesum of evaporation and surface water outflows.As is commonly the case, this amount of datawas not available, and estimates had to bemade. The change in storage over a one-yeartime period was assumed to be negligible.Rainfall and reservoir releases were measured.

10

Water is evaporated from crops, trees andshrubs, bare soils, and free water surfaces.Standard methods (FAO 1992) were used to obtaincrop evapotranspiration. Evaporation of bare soilsbeyond the amount included in evapotranspirationwas assumed negligible. Free water surfaceevaporation was estimated by panmeasurements.

This leaves two major unknowns; evaporationfrom trees and drainage outflows. As a roughestimation, a study from Huruluwewa, also in thedry zone of Sri Lanka (Munasinghe and Somasiri1992) was used as an estimate of evaporationfrom trees and shrubs. The estimates ofevaporation from trees, shrubs, and homesteadsin the dry year were reduced from the wet yearby taking into consideration the amount of rainfallavailable (Sakthivadivel, Molden, and De Fraiture,forthcoming). Finally, drainage outflows wereestimated by closing the water balance. It should

be recognized that there is a potentially largesource of error in the estimation of evaporationfrom trees and shrubs, and thus there is a largedegree of uncertainty around the drainageoutflow estimation. Measurements of drainageoutflow started in October 1997 and will beavailable in the future.

The water balance is made for 12 months,starting in the yala season (April) and ending atthe end of the maha season (March). The yearsof 1995–1996 and 1996–1997 were consideredin this study. Average rainfall in 1995–1996 was1,142 mm and in 1996–1997, only 709 mm(average rainfall over the last 50 years was1,025 mm). In 1995–1996, the total inflow fromrain and the reservoir releases was 607 MCM(figure 2). Process consumption by irrigatedcrops was 98 MCM and for domestic andindustrial uses 2 MCM (table 3). Beneficial non-process depletion includes evaporation from

BOX 1.Water accounting categories.

Water depletion is the use or removal of water from a water basin that renders it unavailable for furtheruse. Two types of depletion are used:

• Process depletion (PD): The amount of water diverted and depleted by a service to produce an in-tended good. Examples of process depletion are evapotranspiration of crops with water delivered bythe irrigation service, and depletion of domestic water provided by the domestic service.

• Non-process depletion (NPD): Water is depleted, but not by the process it was intended for. This canbe further classified as beneficial non-process depletion or non-beneficial non-process depletion. Ex-amples of non-process depletion are evaporation from trees and shrubs, evaporation from fallowlands, and evaporation of water that was delivered by irrigation services for domestic uses.

Committed water (C): That part of the inflow that is committed to other uses. An example is water com-mitted to other users outside the river basin, or water committed to downstream environmental uses suchas lagoons.

Uncommitted outflow (UC): Water that is neither depleted nor committed, and thus available for use withina basin or for export to other basins, but flows out due to lack of storage or operational measures. Theuncommitted outflow at the subbasin level is the water that flows out of the Kirindi Oya into the ocean.

Nondepletive use (ND): A use of water that does not cause any depletion. Fisheries is a nondepletive useof water unless the reservoirs or channels are kept at a minimum level to keep the fish alive.

11

trees, shrubs, and homesteads (149 MCM), anddomestic consumption from well water (1 MCM).Well water is considered a non-process use ofthe irrigation service because the water wasunintentionally derived from the irrigation system.Non-beneficial, non-process depletion is ET fromfallow land and water surface (23 MCM). Theoutflow to RB tracts 5, 6, and 7 and theBadagiriya tank is considered as committedwater (78 MCM). Finally, the outflow of water tothe ocean, which during this year was estimatedat 219 MCM is considered as uncommitted waterand thus available for a use within the basin orfor export to other basins.

For 1995–1996, the depleted fraction (DF),i.e., the amount depleted divided by the totalinflow, was 0.51. This indicates that 51 percent ofthe water entering the area is depleted. Theprocess fraction of depleted water, defined as theprocess depletion divided by total depletion, is0.45. This indicates that only 45 percent of theamount of water that was depleted went tointended processes and this shows scope forconsiderable water savings. Most of thesesavings could be obtained by decreasingdrainage outflow to the ocean.

A water accounting classification andquantification for the Kirindi Oya subbasin isshown in table 3. Each water use is classified asprocess depletion, beneficial non-processdepletion and non-beneficial non-processdepletion, non-depletive uses, committed, oruncommitted water.

Discussion

The results of the water balance should only beconsidered as an indication of the order ofmagnitude of the different water accountingprocesses in the Kirindi Oya subbasin as manywater balance data are based on roughestimates or assumptions. In particular, data aremissing for drainage outflow, ET from the forest,and scrubs and chena, and could only be roughlyapproximated. However, often first orderestimates provide the basis for more in-depthanalysis that provides important clues onincreasing water productivity (Molden 1997). Atthe end of 1997, IIMI started measuring drainageoutflows and at this moment a study is going onto calculate the ET of different types of land use

FIGURE 2.Water accounting for Kirindi Oya subbasin, 1995–1996 and 1996–1997.

12

in the area using remotely sensed imagery. In thefuture, the water balance can be updated andadjusted with more accurate data.

It is important to notice the order ofmagnitude of the drainage outflow. The KirindiOya system is known as a water-short system,but in a relatively wet year like 1995–1996, a lotof water is flowing to the ocean without beingreused. In contrast, 1996–1997 was a relativelydry year. There was no cultivation during theyala season and only little cultivation during themaha season in the new area. During this dryyear, it appeared that only a small amount ofwater was drained. The variability of water supply

should be a consideration in the management ofthe area.

By far, the major user and consumer of wateris irrigated agriculture. Except for a relatively dryyear as 1996–1997, water diverted and depletedby domestic and industrial uses is very small incomparison to the quantities used by crops.Certain important uses such as fisheries andlivestock deplete only a minimum amount ofwater, but add an important economic value. Itwould appear that at least on an annual basis,there is no shortage of water supplies, andhence there is a large scope to improve irrigationservices to obtain more crop consumptive use.

TABLE 3.Classification and quantification of the water uses served by the irrigation and domestic service in the Kirindi Oyasubbasin in MCM for 1995–1996 and 1996–1997.

Year 1995–1996 1996–1997

Unit MCM MCM

Inflow Precipitation 302 188Reservoir water issues 305 168Gross inflow 607 356

Process depletionIrrigation service ET paddy (irrigated) 96 45Irrigation service ET OFC (irrigated) 2 7Irrigation service Domestic use (canal water) 1 1Domestic service Domestic use (piped water) 1 1Rainfall ET chena 38 24

Total 138 78

Beneficial non-process depletionMet indirectly through irrigationservice and rainfall Domestic use (wells) 1 1

ET forest 28 19ET scrub/grassland 96 60ET homestead 24 15Total 149 95

Non-beneficial, non- process ET fallow land 4 9depletion ET water surface 19 12

Total 23 21

Non-depletive uses LivestockFishing

Committed Badagiriya tank 1 1RB tracts 5,6, and 7 77 35Total 78 36

Uncommitted Drainage outflow 219 126

13

Although the quantity of water for domestic use isa small proportion of the total water depleted in anirrigated area, the quality requirements fordomestic water use are considered to be higherthan the irrigation water quality requirements.Satisfying the irrigation water quality requirementsdoes not mean satisfying the domestic and theenvironmental water quality requirements. Forexample, nutrient-rich water benefits irrigatedagriculture, but it is a concern for domestic usebecause of the possible negative health effects ofhigh nitrogen, phosphorus, and bacterial content.

The availability and quality of shallowgroundwater in Kirindi Oya is influenced byirrigation water supply. The groundwater isrecharged with water from the fields, tanks, andcanals as seepage water in addition to the directpercolation of precipitation. During this process,soil minerals and other chemicals get dissolved inthe water. Many shallow wells are constructedadjacent to the field canals and rivers becausewater tables in those locations are expected to behigh. The water table depth is often less than 1m,but fluctuates greatly depending on the time of theyear.

In September 1997, the water quality of theLunuganwehera reservoir, 6 major tanks, theEmbilikala lagoon, 2 sites in the Kirindi Oya river,2 main drainage canals, and 8 dug wells in thestudy area was measured for 10 selectedparameters. The water quality analysis followedthe sampling procedure and analysis methodsrecommended by the Laboratory Service of theNWS&DB. Table 4 shows the results of themeasurements. Additionally, the electricalconductivity (EC) of 85 shallow wells wasmeasured during late December 1997 to earlyJanuary 1998 using a portable conductivity meter.The EC is commonly used to estimate the solublesalt content in solution, in this case water. The

4. Water Quality Issues

average EC of the wells was 2.1 milliSiemens percentimeter (mS/cm) with a standard deviation of1.8 mS/cm.1 Further, household interviews onwater quality perceptions were carried out amongthe households using the 85 wells for domesticpurposes. Of the 10 parameters in table 4, onlythe first four are discussed in the followingsections.

Irrigation

During most of the year, evaporation is higher thanprecipitation, which may cause salinity problems ifexcess salts are not leached out of the plant rootzone. In fact, a salinity problem was reported inthe RB command area after the construction ofthe Lunuganwehera reservoir, which resulted in areduction of rice production (IIMI 1995a). Asalinity-affected area of approximately 200hectares is also recorded in the lower parts of thecommand area, caused by improper drainage,which resulted in the accumulation of salts.

The quality of water in the reservoir, tanks,and the main canals when used for irrigationpurposes falls into the Class 2 category (tables 4and 5). This means that there is a moderatesalinity hazard and the water can be used forirrigation with moderate leaching for most crops.The lowest EC value 0.33 mS/cm was observedin the Lunuganwehera reservoir. The EC of thedownstream tanks was higher than that in theupstream tanks, indicating an entry of elementsalong the watercourses.

Domestic Use

Besides the water quality measurements forirrigation, table 4 also shows the drinking water

1Minimum EC measured was 0.2 mS/cm and the maximum, 10.4 mS/cm; skewness of the measurements was 2.4 mS/cm, and themedian was 1.5 mS/cm.

14

quality standards for Sri Lanka. For all sources,color exceeded the maximum desired levels,especially for surface water. Color is known to bean important criterion on which people base thechoice of their drinking water source (WHO 1984).

The shallow ground water, Kirindi Oya river,and drainage water had high levels of hardnessand alkalinity. Like the Kirindi Oya river, drainagewater exceeded the maximum desirable levels ofalkalinity and hardness (see table 4). The highlevel of hardness can result in scale deposition,particularly when heating the water, and can leadto an increased incidence of urolithiasis (kidneystones) (WHO 1984 and 1993). It was reportedthat the people have sticky hair when well water isused for bathing. Prevalence of urolithiasis ingoats in the area has been reported.Water is not suitable for drinking when it has totaldissolved solids (TDS) of more than 1,000 mg/l(WHO 1993). A TDS below 600 mg/l is consideredto be good drinking water while water becomesincreasingly unpalatable when the concentration isgreater than 1,200 mg/l. Based on the EC

measurements in table 4, TDS can be calculatedapplying the formula:TDS (mg/l) = 640 x EC (mS/cm) (Tanji 1996). Thisresults in a TDS of 1,459 mg/l for shallow wells,448 mg/l for irrigation water, and 570 mg/l forKirindi Oya water. Therefore, as far as TDS isconcerned, well water is unsuitable for drinking,while the irrigation water and river water aresuitable. The EC measurements and householdinterviews showed that 22 percent of the 85shallow wells were not preferred for drinking, and40 percent of the wells used for drinking had asaline taste.

Even though the results of the measurementsindicate that surface water is better for drinkingthan groundwater in terms of chemical quality,surface water is more likely to be contaminatedwith fecal material. There is a concern ofincreasing public health risks due to the usage ofuntreated tank water (CEA 1994). People use thetanks and canals for bathing and washing clothes,while livestock also use them for drinking andbathing.

TABLE 4.Water quality measurements and standards in the Kirindi Oya System (September 1997).

Water quality in the Kirindi Oya system Drinking water qualitystandards of Sri Lanka

Measurement Units Irrigation Kirindi Drainage Shallow Embilikala Maximum Maximumwater Oya well lagoon desirable permissible

water level level

Electrical mS/cm 0.70 0.89 1.67 2.28 1.80 0.75 3.5Conductivity

Color TCU* 7.7 7.5 7.6 7.5 8.0 5 30

Alkalinity (as CaCo3) mg/l 211.3 338.0 265.0 497.1 190.0 200 400

Hardness (as CaCo3) mg/l 192.3 240.0 400.0 650.3 450.0 250 600

Temperature C 29.6 30.0 33.1 29.0 33.0 – –

Nitrate (as N) mg/l 0.56 0.70 0.75 0.43 0.50 – 10

Nitrite (as N) mg/l 0.13 0.01 0.12 0.01 0.06 – 0.1

Phosphate (as PO4) mg/l 0.18 0.23 0.15 0.38 0.17 – 2.0

Sulphate (as SO4) mg/l 36.3 50.0 45.0 72.1 NA 200 400

Chloride (as Cl) mg/l 93.5 78.0 320.0 416.9 380.0 200 1200

*True Color Units (WHO 1984).

15

Environment

The Kirindi Oya irrigation system includes theWeerawila-Tissa Wildlife Sanctuary which contains4 tanks with a total surface area of 1,590hectares. The tanks are defined as wetland andrecognized as a refuge for ducks and waterbirds.The CEA (1994) reported possible effects ofchanges in water levels in the tanks and waterquality on the ecology and wildlife of thesanctuary as follows:

• Negative impact on fish population anddiversity caused by reduced water levels,especially in the yala season.

• Negative impact of extreme water levels onresident and migratory birds, when tanks aredry or when the water level is too high forwaders, dabbling birds, etc.

• Risk of eutrophication of tanks resulting fromdung and urine from livestock, discharge ofdomestic effluents, and fertilizer runoff.

There is no doubt that the quantity and qualityof water in the sanctuary are greatly influenced byirrigation and human activities within the KirindiOya scheme as the hydrology of the sanctuary isdirectly related to the irrigation management. Thetanks receive water through the LB main canal of

the Lunuganwehera reservoir, in addition to thedrainage water from the RB tracts.

The Bundala National Park is locateddownstream of the Kirindi Oya scheme. It covers6,216 hectares and encloses 5 brackish lagoons of2,250 hectares (CEA 1993). The drainage waterfrom the RB tracts 5, 6, and 7 and the BadagiriyaIrrigation Scheme flows into Embilikala and Malalalagoons, respectively. Over the past years, it hasbecome clear that the ecosystem of those twolagoons has been severely affected by the drainagewater coming from the Kirindi Oya scheme. Naturalfluctuations in salinity levels have disappeared andthe two lagoons have become freshwater lakes.The change in salinity levels influences thepopulation of water birds as it affects the qualityand quantity of their food supply (CEA 1993). Prawnfarming requires brackish water conditions andpreviously it sustained several hundred families,but it has now almost disappeared from the area.Eutrophication is also a problem in the lagoons.Water has a greenish color due to the accumulationof nutrients and increase in green algae. This mightbe caused by overgrazing with direct deposit ofanimal feces in surface water as well as highfertilizer runoff from the irrigated area. On the otherhand, an increase of drainage flow may benefit theother wildlife and livestock in the national park asmore fresh water becomes available.

TABLE 5.Classification of irrigation water salinity.

Class EC (mS/cm) TDS in PPM Characteristics

C1 0.1–0.25 60–160 Low salinity hazard; water can be used for most crops on most soils.

C2 0.25–0.75 160–480 Moderate salinity hazard; water can be used with moderate leachingfor most crops.

C3 0.75–2.25 480–1,440 Medium-high salinity hazard; water for use on soils with moderategood salt tolerance, leaching is required.

C4 2.25–4.00 1,440–2,560 High salinity hazard; water for use on well-permeable soils with salttolerant crops; special leaching requirements must be met.

C5 4.00–6.00 2,560–3,840 Very high salinity hazard; water generally undesirable for irrigation; tobe used only on highly permeable soils with frequent leaching andwith highly soil-tolerant crops.

C6 Above 6 Above 3,840 Excessive salinity hazard; water unsuitable for irrigation unless undervery special conditions.

Source: International Land Development Consultants (ILACO) 1981.

16

5. The Institutional Environment

The Government of Sri Lanka claims legalownership of all surface water and does notrecognize any system of individual or group waterrights. Water use rights are allocated to land indesignated irrigated areas through a process ofseasonal planning or allocation (Brewer,forthcoming). The Irrigation Ordinance definesmechanisms for seasonal water allocations butadds that all decisions are subject to review andchange by the government. The final authority isthe Minister in charge of irrigation (IIMI 1995b).The negotiation processes for water are influencedby the structure of the government and the userorganizations. Even within the government,different agencies’ regulations (or the interpretationof different officials within an agency) may vary.Another complication with the definition ofwater rights is the overlap of various hydrologicunits and administrative boundaries (Hoogendam1995).

There is no comprehensive water resourcespolicy, but several institutions at different levelsformulate water policies relating to their respectivesectors (e.g., Irrigation Department andNWS&DB). Until April 1996, there was noinstitutional mechanism to coordinate the activitiesof these institutions, so that there wereconsiderable duplications of efforts. In April 1996,the establishment of a Water Resources Councilwas officially approved. This council iscoordinating the implementation of the action planfor comprehensive water resources management.The Water Resources Council is only an advisorybody and is part of a temporary arrangementaimed at coordinating activities in the watersector, including the formulation of waterresources policies and legislation.

Water Rights and Institutions in KirindiOya

The KOISP falls within the Southern Province, butthe main reservoir and the catchment area arewithin the Uva Province. Therefore, the CentralGovernment has legislative and executive powersover the entire system. The Kirindi Oya river basinfalls within 3 districts and 7 divisions while theKOISP falls within 2 districts and 4 divisions.In addition to these administrative jurisdictions,there is a Project Management Committee (PMC)that coordinates the activities of variousgovernment agencies related to irrigatedagriculture and makes decisions for seasonalwater allocation.

Field Crop Production

Water rights tied to irrigated land constitute one ofthe most widely recognized forms of water rightsin Sri Lanka, particularly in Kirindi Oya. Generally,farmers in the old area own their homesteads andirrigated land, while settlers in the new area havebeen allotted management and use rights2 by thegovernment. Alienation rights to land for settlersare limited: they cannot legally sell or lease it, andwhile it can be inherited by a successornominated by the original allottee, the land cannotbe subdivided between heirs (Stanbury 1989).

In developing the KOISP, the governmentrecognized the seniority of the existing waterrights of farmers in the old area. Those farmerswere assured that their water use would not bereduced by the project, and in fact, they were ableto increase their cropping intensity due to morereliable water supplies from the new reservoir.

2The following analysis of rights uses a hierarchy of bundles of rights identified by Schlager and Ostrom (1992) : access and withdrawal(use rights), management, exclusion, and alienation (control rights).

17

TABLE 6.Water allocation pattern in the Kirindi Oya Irrigation System.

Area Maha season (wet season) Yala season (dry season)

Old area Enough water to cultivate 100% command Enough water to cultivate 70% command area witharea with paddy. paddy.

New area 2/3 command area with paddy; others are Areas that did not get water for paddy during Mahaencouraged to plant OFC. If there is have priority.sufficient rainfall they can cultivate a latepaddy crop.

Source: Brewer, forthcoming.

Based on this guarantee, the lands in the old areawere given priority for water, even if it meant thatthe new area did not get any water. The generalbasis for water allocation to the old and the newareas is given in table 6.

Since 1978, various experiments givinggreater responsibilities to farmers in irrigationmanagement were carried out in Sri Lanka (Brewer1994). Kirindi Oya was brought under theIntegrated Management of Major IrrigationSystems (INMAS) program in 1986. In the KOISP,this program included the creation of 690 FieldChannel Groups, 59 Distributary ChannelOrganizations (FCGs and DCOs), 4 Sub-ProjectCommittees (SPCs), and 1 PMC. The FarmerOrganizations (FOs) were created to assist themanagement of the irrigation system, while theProject Committees (SPCs and PMC) givefarmers a voice in allocation decisions.

The PMC is the main organization involved inwater allocation in Kirindi Oya. This is a jointgovernment-user group entity composed of farmerrepresentatives from various parts of the systemalong with representatives from a range ofgovernment agencies (e.g., Irrigation Department,Department of Agriculture, and Department ofAgrarian Services). In addition to water allocationdecisions for irrigation, the PMC attempts toresolve other problems brought to them,particularly problems that require the assistance ofone of the government agencies.

The PMC allocates water for agriculturalpurposes by negotiated seasonal planning

meetings. In developing the plans it is assisted bythe SPCs for Ellagala, new area LB, new area RB,and Badagiriya. The seasonal planning is flexiblein the sense that it adjusts water allocation towater availability. Until 1991, allocation decisionswere made by the officials without direct inputfrom the farmers. The seasonal decisions madeby the PMC were more acceptable since thefarmers had some input. The PMC authority wasalso accepted by the government officialsbecause of the government’s participatorymanagement policy.

Water distribution—the delivery of water toexecute the water allocation plan—is theresponsibility of the ID at the reservoir and themain canal levels, and the responsibility of FOsbelow the distributary level. In addition to waterdistribution, the FOs are responsible formaintenance of the distributary and field channels.If they wish to, the FOs can take on otherfunctions. Their performance in maintenanceseems to be satisfactory and, clearly, they havehelped improve water distribution, at least at thefield channel level (IIMI 1995a). Above that level,most FOs are weak and do not play a significantrole in system operation. The FOs in the new areaare weak because many farmers do not reside thewhole year round or lease their lands to others.Another problem in the new area is that one FOcan include farmers from different hamlets. Thissocial division hampers the functioning of the FOs(ibid).

18

As individual water users, farmers with land inirrigated areas have use and alienation rights towater on their fields. Through participation in theFOs and the PMC, farmers have also acquiredsome management rights. The strength of theserights depends on the degree of participation ofthe individual farmers in the FOs and on thestrength of farmer representative’s voices on thePMC. Nevertheless, farmers’ interests in water forfield (especially paddy) crop production are betterrepresented than any other type of use in waterallocation decisions.

Garden Crop Production

There is no recognized water right for homesteadgardens. On the contrary, taking water from eitherthe irrigation canals or the piped water supply forgardens is prohibited (NWS&DB 1997a). Takinggroundwater from private wells, however, is notregulated and the development of agro-wells iseven promoted by the Agricultural DevelopmentAuthority. Informally, a certain amount of wateringgardens from canals or domestic supply systemsmay be tolerated, and runoff or wastewater fromdomestic use is certainly applied to gardens.However, garden production is treated as anindividual use, and there is no user group torepresent these interests.

Livestock

Water is not especially issued for livestock useslike drinking and bathing. The water use rights oflivestock are informal and not clearly defined. Thefact that customary lands for cattle grazing andwatering places were not recognized in thedevelopment of the Kirindi Oya system indicatesthe relatively weak water rights for livestock. Thisincreased the contact between herds and fields,which causes crop damage and conflict betweenlivestock and crop production. To solve thisproblem, three Cattle Owners’ FarmerOrganizations (COFOs) were formed in 1991.

These organizations are working together withthe government agencies to find alternativegrazing for the herds. The leaders try to work withthe FO leaders to resolve disputes about cropdamage. Even though livestock owners arerepresented on the PMC, their participation in thatforum is primarily related to crop damage anddoes not involve water management decisions,and therefore they are not granted managementrights.

Fisheries

There are a variety of government, NGO, and userorganizations involved in fisheries, but there is nocoherent policy towards water use for fishing.There are Fisheries Cooperative Societies (FCS)for each tank and reservoir, and tank fishing rightsare legally restricted to these cooperatives(Steele, Konradsen, and Imbulana 1997).Government assistance to fishermen is channeledthrough the cooperatives, that are responsible forchecking whether the fishermen stick to the rules(e.g., size of holes in the nets). Access rights towater and the right to withdraw fish are regulatedthrough the FCSs to their members. Becausefishermen do not have a voice on the PMC forregulating water levels, they do not havemanagement rights over water. However, thefishermen do not seem to make an issue of thisbecause most have agricultural land. Theyconsider fishing as a secondary activity, asubsidiary use of water, while the first and mostimportant activity is agriculture.

Domestic

Although the ID was nominally responsible forconstruction of the domestic water supplyscheme, responsibilities were surrendered to theNWS&DB, and it continues to be responsible forthe piped water supply system (IIMI 1995a). Whenthere is no irrigation, the ID supplies water in thecanals once in 14 days for domestic purposes.

19

A certain amount of the Lunuganweherareservoir water is set aside for the NWS&DB toprovide water for the piped water supply system.When the water is above the dead storage level,i.e., 45.5 meters above the mean sea level (MSL),the NWS&DB has a water right to extract 5,000m3

of water per day. When the water level is at orbelow 47.5 meter above MSL, only the NWS&DBhas the right to pump water from theLunuganwehera reservoir. Further, a water right toabstract 600m3 from two tube wells at Kirinda isgranted to the NWS&DB to supply the old area.The NWS&DB also has a right to abstract waterfrom Tissa Wewa to provide water for domesticpurposes. During the 1992 yala season, waterissues from the reservoir for irrigation were evenstopped in early July to protect the domesticwater supply (Brewer, forthcoming). This is anindication of the priority given to the domesticwater supply. It is noteworthy that the NWS&DB isnot represented on the PMC.

On the users’ side, standpipe committees ofapproximately 15–20 households are establishedunder the supervision of the NWS&DB to managestand posts for piped water supply. Theseassociations are informal, i.e., no authority isvested in them under the existing legislation,although they are responsible for collecting usercharges from the households who make use ofthe standposts. User charge for one household isRs 11 per month. It is also the responsibility ofthe members of the standpipe committee tosafeguard the water stand, and the committee isliable for the misuse of water by the standpipeusers. If the users do not stick to the rules andregulations set by the NWS&DB, the water isdisconnected and a reconnection fee of Rs 250has to be paid by the committee (NWS&DB1997a). Although the NWS&DB does not allow foruses other uses than drinking (ibid), the waterfrom standpipes serves a variety of uses e.g.,bathing, laundering, and brick making. The usersallow each other to use standpipe water for thesekinds of purposes because there is no other watersource nearby.

The formal rules grant members of thestandpipe committees limited use rights, andrights (which they may or may not exercise) toexclude other users, or users who do not pay.However, the rules are specified by the NWS&DB,so that users have no formal management rights.Informally, however, each group decides whatuses will be tolerated or even consideredlegitimate, so there are some de factomanagement rights. Those who draw theirdomestic supplies from sources other than theNWS&DB system have acknowledged rights towater through reservoir releases even when thereis no irrigation. Beyond this, there is much lessregulation of use.

Other

No special water rights and allocation are recordedfor recreation, wildlife, and the environment. TheAir Force has a water right of 2,000m³ per monthfrom the NWS&DB. As far as industrial water useis concerned, there is one garment factory inKirindi Oya, which has a water right for 1,300m³per month from the NWS&DB. According to theID, a number of hotels have requested water, buthave been denied permission to take water fromtanks and other surface sources. They thereforeturn to groundwater abstraction, which is regulatedless (although the ID notes that this waterultimately comes from the irrigation system).Hotels have also turned to groundwater becausethe NWS&DB charges for hotels (Rs 27 /m³) havetripled since 1984, and are considerably above thecharges for household use, schools, and religiousinstitutions (NWS&DB 1997b). Water is notallocated to small-scale enterprises like curd potmaking and brick making. People make use of theavailable water, which has been allocated for otherpurposes like irrigation and drinking. No usergroups were encountered representing the waterinterests of industrial or micro-enterprise waterusers.

20

TABLE 7.Government agencies and user groups representing different types of water uses.

Type of water use Government agency User group

Field crops Irrigation Department* Farmer Organizations*Irrigation Management Division* (distributary as well as field-channel level)Land Commissioner Department*Agrarian Services Department*Department of Agriculture*Divisional Secretaries*

Garden crops Agricultural Development Authority Not represented

Fisheries Aquaculture Development Division of the Fisheries Cooperative SocietiesMinistry of Fisheries

Livestock National Livestock Development Board Cattle Owners’ Farmer Organizations*Department of Animal Production and Health

Domestic National Water Supply and Drainage Board Local standpipe committeesLocal Government Authorities

Industry/ small-scale National Water Supply and Drainage Board Not representedenterprises Local Government Authorities

Wildlife/ environment Department of Wild Life Conservation National/International NGOsCentral Environmental Authority

* Represented on the Project Management Committee.

Institutional Structure for Multiple WaterUse

Many of the use and management rights ofdifferent categories of water users are negotiatedand mediated by a range of formal and informalorganizations. Table 7 gives an overview of therange of organizations found in Kirindi Oya thatrelate to a type of water use.

In many cases, there are parallel governmentagencies and user groups for each type of wateruse. In some cases, there are even multiplegovernment departments related to a type of wateruse. However, effective coordination amongdepartments is very difficult. In Sri Lanka,government organizations are strongly hierarchicalwith clear lines of authority. Officers are generallynot rewarded for the effort put to coordinate withother departments and are sometimes punishedfor it (IIMI 1995a).

We were not able to identify any user grouprepresenting homestead gardens, perhaps

reflecting the autonomous (and sometimes evenatomized) production. While wildlife does not havea user organization, the interests of wildlife arerepresented, to some extent, by the environmentalNGOs, both national and international. However,these do not make their presence felt at the locallevel. There is a district-level Bundala WetlandManagement Committee (BWMC) that includesrepresentatives from government agencies (e.g.,Department of Wildlife Conservation, IrrigationDepartment) as well as from user groups (e.g.,salt farm). Although user groups’ representativesare included on this committee, it does notprovide a forum to deal with problems related tomultiple uses of water because it is not on theirrigation system level. There is no linkagebetween the BWMC and the PMC, nor is theDepartment of Wild Life Conservation representedon the PMC.

The negotiations over water allocation betweendifferent uses take place not only among usergroups, or between user groups and the

21

government, but also among governmentagencies. Of these, the ID is seen as thestrongest, with the greatest control over waterreleases. For example, the Department of WildlifeConservation has requested changes in waterflowing to the Bundala sanctuary to preserve thesalinity balance in the wetlands, but it does notfeel it can direct the ID. The ID feels it isresponsive in allocating domestic water from thereservoir, but the NWS&DB does not always thinkthis is the case.

Traditionally, water allocation has been carriedout by the ID under the Irrigation Ordinance. Therecent development with many subsectorscompeting for water stresses the need to create anew institutional setup to handle water sectorcoordination problems (Rasmussen 1994). Thiskind of organizational framework thatencompasses all water uses and users is lacking,even within the government. The PMC and itsseasonal planning meetings before every season

provides a first step in this direction, by includingvarious government departments as well as farmerrepresentatives from each command area. Toresolve disputes due to damage of crops bycattle, representatives of COFOs also attendthese meetings. Domestic water allocation issuescan also be raised and considered at thesemeetings, but they do not cover other types ofwater use. Seasonal planning meetings might givescope for dealing with other types of water usesand issues when representatives of relevant usergroups and agencies can participate. For this tobe effective, representatives of different usergroups must not only be included on the PMC, butalso have a strong enough voice to raise their ownwater-related issues. Further coordination, betweengovernment agencies and user representatives isneeded if the Kirindi Oya system is toaccommodate the needs of all water users, anddeal with the potential complementarities as wellas trade-offs involved in multiple water use.

6. Household Uses of Water

The information presented in this chapter relatesto the findings from the household water usesurvey among 156 families in the study area. Theobjective of this household survey was to identifythe variety of agricultural and nonagricultural usesof irrigation water. At the same time, an attemptwas made to assess the importance of irrigationwater when compared with nonirrigation sources.

Methodology

The surveyed households were located in 10clusters with 5 each in the new and the old areasof the Kirindi Oya system. The households wereselected at random from the total number in theclusters from the old and the new areas at a fixedpercentage of 0.70 and 1.30, respectively. The

random selection was based on the “voter’s list”generated for the local elections in Sri Lanka in1996. The information collected is believed to berepresentative of the whole Kirindi Oya system butthe data presented below do not considerseasonal differences. The survey was carried outover a 7-month period from May to November1997. Quantitative information collected through aquestionnaire presented to each household wassupplemented with qualitative information collectedfrom repeated visits to the same households andvia informal or formal group discussions in the 10selected clusters. With regard to questions relatingto the water used in agriculture, the informationcollected represents the last season when thefields were cultivated. The qualitative informationwas collected to provide more in-depth informationto explain why a certain source was used and who

22

was the person responsible within the householdfor providing labor in relation to a certain wateruse activity. In the description given below, thedifferent water sources used by the households forvarious purposes are given with an emphasis onthe domestic uses of water.

Household Uses of Water by Sector

Irrigation water is defined as water from tanks andirrigation or drainage canals. Nonirrigation watercomes from the piped water supply system,homestead wells, the Kirindi Oya River, or rainfall.It should, of course be acknowledged asmentioned before, that a large proportion of thegroundwater used from the homestead wells is infact, seepage from the irrigated areas. Also, partof the piped water system has its origin at theLunuganwehera reservoir. However, the water nolonger flows via irrigation structures and is in oneway or other treated and possibly pumped.

The first priority water source for a range ofactivities is given in table 8. Rice cultivationobtains water exclusively through the irrigationcanal system and all inland fisheries take place in

the reservoir and irrigation tanks. The smallnumber of families involved in home industries isalmost equally divided as families dependent onirrigation and nonirrigation sources. Of the mostimportant income-generating activities for thehouseholds, only raising livestock and shiftingcultivation are dependent on nonirrigation sourcesto any significant extent, mainly rainfall. Close tohalf the households prefer or are dependent onirrigation water as a source for laundering, bathing, orrecreational uses. But hardly any household makesuse of irrigation water for drinking and cooking.

Water-Related Labor Input forFisheries, Agriculture, and Livestock

In the vast majority of households, a male wasthe main person responsible for water-relatedactivities in rice farming (86%). In the remaininghouseholds it was a shared activity between menand women. For overall labor input into riceproduction, the male members were principallyresponsible in 67 percent of the households, withthe responsibility being shared in the remaininghouseholds. In homegarden production, a female

TABLE 8.The importance of irrigation water as first priority water source in comparison to other sources of water for a variety ofuses.

Uses Number of Use of different sources (%)

respondents Nonirrigation water Irrigation water

Irrigated agricultureRice 93 0 100Other field crops (e.g., onion) 17 0 100

Shifting cultivation 30 100* 0

Homegarden 54 87* 13

Cattle, buffaloes, and goat 20 45 55

Inland fisheries 9 0 100

DomesticLaundering, bathing, and recreational 156 96 4Drinking, cooking, sanitation, andwashing utensils 156 53 47

Home industries 17 41 59

*Including water from rainfall.

23

member of the household was the main personresponsible for providing irrigation water in 52percent of the households and it was in only 37percent that a male member was the main personresponsible. However, as indicated in table 8,most of the homegarden production wasdependent on rainfall and the additional water fromirrigation was limited. In 44 percent of thehouseholds, the women provided most of theoverall labor into homegarden cultivation while only32 percent of the males provided most of thelabor, with the responsibilities being shared in therest of the households. Precipitation was the onlysource of water for shifting cultivation andtherefore, there is no gender differentiation inwater-related labor input. In shifting cultivation, thework is shared almost equally among thehousehold members but some gender differencesare associated with the different tasks. The typeof inland fisheries practiced depended on theavailability of water in the irrigation tanks and thehouseholds provided no water management orwater-related labor input. In the case of homeindustries, the types of production were so diversethat no general conclusions could be drawn.

Domestic Uses of Water

A very heavy burden was placed on thehouseholds, where the source of domestic waterwas distant from the homestead. Clearly, a reliablewater supply close to the settlement is a veryhigh priority for the community. In some of thehouseholds, the time spent fetching domesticwater could amount to ‘several’ hours a day andan individual from the household sometimes hadto go to the standpipe as early as 4 or 5 o’clockin the morning to wait in a queue for water. Attimes of the year when the homestead well runsdry and water is not available, more time is spentfetching water in the irrigation canals. Fetchingwater for domestic use was mainly done bywomen or children (male and female). However,where the household is dependent on a water

source some distance away from the house, themales were often involved in fetching water inlarge buckets that they transport on bicycles ortractors. Only two households included in thestudy mentioned that they purchased water fromwater vendors.

Drinking and Cooking

Respondents stated that they obtained water fordrinking and cooking from either a tap (70%) or awell (30%) because these sources were perceivedas clean. Convenience was another importantfactor as the taps and the wells were generallylocated close to the homes. Some peoplepreferred using well water because, unlike pipedwater, there was no fee for its use. Some families,especially those living close to the sea, preferredtap water because the groundwater is saline.Water for cooking was mainly used by females ofthe household (69%). Virtually all members of thehousehold were involved in fetching water. But in44 percent of the households, women providedmost of the labor to fetch water for drinking andcooking, whereas, men were mainly responsiblefor this task in only 18 percent of the households.Only a few families mentioned that they boiledwater before drinking, often only for children andsick people.

Washing of Utensils

The tap was the preferred source of water forwashing of utensils, (64%) followed by well water(30%), and the canal (3%). As in the case ofwater for cooking and drinking, these were thepreferred sources for reasons of hygiene andconvenience. Easy accessibility was given as animportant reason for the use of well water (45%)and tap water (30%). For about 20 percent of thehouseholds, the tap was the only source availableto wash utensils. Women were primarilyresponsible for washing utensils in 70 percent of

24

households and in the rest, the task wasperformed by children, men, and in some cases,by “all members” of the household.

Laundering

Canal water was used by 38 percent of thehouseholds for laundering, followed by tap water(30%), wells (15%), tanks (10%), and the river(7%). Fifty-seven percent of the respondentsclaimed they used canal water for laundering atone time or another (as compared to the 38% whoused it as their main source). The free availabilityof canal water appeared to be the mainconsideration in the preference displayed in itsuse (60% of respondents). It also appeared to bethe reason for the use of tanks (32%) and theriver (33%) for laundering. Convenience was alsoa factor, as indicated by one-fourth of therespondents who used the river and those whoused canals (22%), and tanks (19%).Respondents stated that these sources were“easy to use” probably because there were nowaiting times, nor did users have to fetch thewater in a bucket, and the availability of largeamounts of water made washing easier.

Thirty-one percent of those using tap water forlaundering said it was either because no othersource was available or “it is easy to use” (25%)(convenience), and finally, because it was clean(16%). Similar reasons were cited for the use ofwell water.

In 65 percent of the households all membersof the household claimed they participated inwashing clothes and fetching water. It was notclear from the survey as to who was primarilyresponsible for laundering, but the qualitative dataindicated that the senior woman in the householdhas much of the responsibility for this chore.

Bathing and Recreational Use of Water

The main sources of water used by thehouseholds for bathing (personal hygiene) and

recreational purposes were the canals (35%), taps(31%), wells (14%), and the river (8%). Fifty-sixpercent of respondents claimed they used canalwater for bathing and recreational purposes atsome time and almost a fourth made use of thetanks at some point in time. In 80 percent of thehouseholds, all members were said to be involvedin fetching water for this purpose. In approximately10 percent of the households, a female was saidto be the main provider of the labor input intothese activities. The reason to prefer the canal asthe main source of water for bathing andrecreational use was because of the largeamounts of water easily available at no cost (“adlib use” 55%, “source is easy to use” 24%, and“close to household” 13%). From the qualitativeinformation collected from the households, anadditional reason for preference was given as“water quality.” Canal water was perceived as of aquality ideal for bathing and recreation purposes.Similar reasons of cost and accessibility werecited as the reasons for tank water use although13 percent of the respondents indicated that theyused tanks because they were the only availablesources. Tap water was used because it was theonly source available (34%), “easy to use” (28%),and clean (19%); among its main users, well waterwas also regarded as “easy to use” (36%) and “aclean source” (27%). Convenience was cited asthe most important determinant of the use of riverwater (77%), probably because these respondentslived close by the Kirindi Oya river.

Household Cleaning

The main source of water used to clean the houseis tap water in 57 percent of the households andwell water in 12 percent. Water from canals,tanks, and the river only plays a minor role forthis activity, with around 3 percent of thehouseholds indicating one of these sources as themost important. The question of water being usedfor cleaning of the house was answered by 78percent of the households and only 30 percentgave any additional information on labor input or

25

reasons for preference. However, in thehouseholds that did respond, this activity seemsto be mainly a female responsibility. Sixty percentof the responding households claimed that thisactivity is exclusively done by women followed by26 percent of the households stating that this wasan activity carried out by both the male head ofthe household and his spouse. Tap water and wellwater are used mainly because it is seen as aclean source and ‘easy to use.’

Sanitation

Tap water was the main source for sanitary uses(59%); followed by the wells (28%) and canals,tanks, and the river (13%). The majority of thehousehold members are involved in using andfetching water for sanitary purposes. Twentypercent of the households reported that the seniorwoman in the house was mainly responsible forthis activity, probably because she, in addition toher own sanitation, takes care of the children’sneeds. Convenience was the main reason cited forthe reliance on tap water and well water. Tap waterand well water were not used out of qualityconsiderations but mainly since this was the waterthat was brought to the house anyway.

Preferred Water Sources in the Newand the Old Areas

The sources of water available to the householdsdiffered substantially between the old and the newareas. This affected the selection and prioritizationof water sources for different purposes. In the oldarea, wells were more prevalent than taps, whilethe reverse was true for the new area. Also, in theold area, tanks and the river were an importantsource of water for some purposes. A Wilcoxonpaired sign rank test was conducted for allsources and not just for the first priority source.Water sources, which contributed less than 5percent of the total water used by the household,were eliminated. The results are summarized intable 9.

Interestingly, in the new area, tap water is thepreferred source for all purposes. Wells are alsoused for all purposes and there is an equalpreference between wells and canals forlaundering, bathing, recreation, and sanitation. Inthe old area, the sources and preferences aremore varied with wells being either preferred orequal to all other sources.

TABLE 9.Results of ranking exercise for domestic water uses and sources.

Purpose Order of ranking

Old area (n=82) New area (n=74)

Drinking and cooking Well = tap Tap > well

Washing of utensils Well > tap Tap > well

Laundry Well = canal > tap = tank Tap > well = canal

Bathing and recreation Well = canal > tap = tank Tap > well = canal

Sanitation Well > tap > canal = tank = river Tap > well = canal

26

Irrigation water is used for many purposes otherthan irrigating field crops, as is shown in theprevious chapters. The importance and value ofmultiple uses of irrigation water are oftenunderestimated. According to Bhatia (1997), directeconomic benefits to the farmer (from crop output)reflect only a small proportion of the total benefitsto the community of using water in irrigatedagriculture. An irrigation infrastructure providesnonirrigation benefits to other user sectors andignoring these benefits will result in a seriousunderestimation of the benefits available from thevolume of water that is diverted for irrigation. Thevaluing of water for multiple uses should ensurethat the full range of values placed on water incompeting uses is observed (Pigram 1997) andtaken into account when water allocation decisionsare being made.

The Value of Water

While talking about valuing the water, it is good todistinguish different concepts: water pricing,economic value of water, and other values ofwater.

Water Pricing

Water pricing (water use fee) is meant to collectmoney from the users in such a way that all or aportion of the construction, operation, andmaintenance costs of the system are recovered.Hence, users pay a price to use a certain service,either the irrigation or the domestic service. Thereis, in fact, a considerable debate amongprofessionals regarding the amount to charge forthe use of irrigation. Further, in a multiple usercontext this leads to questions of who should payfor the water service: the irrigators, livestockowners, fishermen, domestic users, or brickmakers? Even when an irrigation system isconsidered as a single purpose unit, it is

7. Valuing the Multiple Uses of Water

questionable who should pay for the water. Thegeneral consensus is that farmers should pay theoperation and maintenance (O&M) costs. However,the construction of government irrigation schemesand expansion of irrigation has resulted in lowercereal grain prices. Consumers have been the majorbeneficiaries. Thus it seems unreasonable to chargefarmers the cost for construction or rehabilitation.

In Sri Lanka, farmers enjoy irrigation facilitiesthat are provided by the government free ofcharge. The government has invested money inconstruction and maintenance of the irrigationschemes and it offered free O&M of the schemesin addition to free land to settlers (Upasena andAbeygunawardena 1993). In 1984, the governmentimposed an irrigation fee in major irrigationschemes. At first, farmers were to be charged halfof the O&M cost (then estimated Rs 495 perhectare) but the charge would eventually rise tothe full cost (Brewer 1994). In 1988, the feecollection scheme collapsed. In turn, fullresponsibility for resource mobilization and theO&M of the field and distributary channels of themajor irrigation systems was turned over to theFOs.

The NWS&DB which supplies part of thedomestic water in Kirindi Oya, makes use ofquantity-based prices which vary according to theuser. For example, religious institutions pay Rs 2/m³ and industries pay Rs 25/m³ for water (duringthe study period US$1.00 = Rs 58.80) (NWS&DB1997b). To discourage the waste of water bydomestic consumers, the tariff increasesprogressively by stages, depending on the amountof water consumed. For instance, for the first 10m³ Rs 0.60/m³ and for all cubic meters above 50,Rs 32.50/m³. The NWS&DB (1985) estimated thecost of supplying 1 m³ of water in 1995 at Rs 0.95.

Economic Value of Water

Briscoe (1996) argues that there is an emergingconsensus that effective water resources

27

management includes the management of wateras an economic resource. A way to do this is toestablish water markets. Several conditions needto be met for establishing a water market. Amongothers, Meinzen-Dick and Rosegrant (1997), Perry,Rock, and Seckler (1997), and Reidinger (1994) allgive a set of preconditions for the beneficialintroduction of water markets. It is argued thatmarkets increase economic efficiency byallocating resources to their more valuable uses(Bauer 1997). However, there are strong socialnorms that argue against water being treated as asimple marketable commodity because it is abasic need. Pursuing efficiency through marketallocation may not be politically or sociallyacceptable if equity considerations are not met(Perry, Rock, and Seckler 1997; Meinzen-Dick andRosegrant 1997). Besides the fact that water is anessential, life-supporting commodity with nosubstitute, water has some other complicatingfeatures, which make it more difficult to establisha market for it.

Water is• a ‘fugitive,’ reusable good

• a common property or object of shared rights

• subject to economies of scale in provision

• associated with many non-market,environmental qualities (Morris et al. 1997;Caldas, Sousa, and Pereira 1997)

Since there is no developed water market inSri Lanka, the value of water cannot be derivedfrom the marginal value reflected on that particularmarket.

Water has an economic value because it hastwo qualities: desirability and scarcity(Abeygunawardena n.d.). In economics, the term“value” refers to monetary measures of changes ineconomic welfare (Young 1996). Economicvaluation can be defined as the attempt toquantify goods and services provided by natural

resources, whether or not market prices areavailable. The economic value of any good orservice is generally measured in terms of thewillingness to pay for the commodity, minus whatit costs to supply it. Where a resource simplyexists and provides us with products and servicesat no cost, it is our willingness to pay alone whichreflects the value of the resource in providingsuch commodities, whether or not we actuallymake the payment (Barbier, Acreman, andKnowler 1997). The value of water, i.e., thedesirability and scarcity, can vary considerablyacross seasons and regions. In Kirindi Oya, waterwill have a different value for farmers in the newand the old areas, during the maha and the yalaseasons, and in wet and dry years. On a historicbasis, the old area farmers claim that they havemore rights on irrigation water than the new areafarmers. The ID recognizes these rights and theold area farmers receive more water than thefarmers in the new area. In this case, the value ofwater is influenced by the historical users’ rights,which determine the access and control overwater, and this, in turn, influences the scarcity ofwater for certain users and uses. Hence,maintaining a constant value of water does notreflect the reality of changing water supply anddemand conditions through time and region.Therefore, information on economic values ofwater must always be indicative rather thanabsolute (Pigram 1997).

Other Values of Water

Besides the economic value of water, it is alsoimportant to take other values into account whenwater allocation decisions are made. If onlyeconomic considerations and values woulddetermine water allocation, the poor of the worldcould be very much worse off. For instance,willingness to pay depends largely on the ability topay. Thus even with the same basic need or valueof water, the rich will get more than the poor(Perry, Rock, and Seckler 1997).

28

Irrigation water has a social value in thesense that it creates opportunities fordevelopment. Without irrigation these opportunitiesdo not exist. Consider, for instance, theemployment generated by irrigated agriculture. Itcreates direct job opportunities in the field as wellas indirect jobs that are linked with agriculturalbusinesses and services. In Kirindi Oya, between1986 and 1994 there was a decrease inagriculture-related employment opportunities due tothe persisting drought and the consequentinadequate water supply (IIMI 1995a). Further, adecline of irrigation could also increase the trendof migration from rural to urban areas, which couldenhance social conflicts. Besides employmentgeneration and social consistency, irrigatedagriculture contributes to food security. Accordingto Fereres and Cena (1997) the risk of notmaintaining a productive agriculture is a strategicmistake. A substantial decline in irrigatedagriculture would make dry areas very vulnerablein the long run, regardless of the level ofeconomic development.

The availability and the accessibility ofirrigation water for domestic uses generate health

benefits. Respondents of the household survey inKirindi Oya mentioned that there is a higherincidence of eye diseases and skin problems inthe dry season due to shortage of water. Unfortu-nately, there were no data available to confirmthese statements and the relationship between theavailability and the accessibility of irrigation waterand the incidence of these illnesses.

Besides this, irrigation water has anenvironmental value. Examples include therecharge of groundwater table and preservation ofthe ecology of wetlands. This environmental useof water is especially hard to value because onehas to deal with intangible aspects and somebenefits only show in the long run.

Valuing the Productive Uses of Water

To value the productive uses of water like cropproduction, livestock, fisheries, and industries, thevalue added of water can be calculated. This isthe so-called factor productivity of water and isdefined in box 2.

This is one way to measure the productivity ofwater for its different productive uses. Thenumerator shows the net value of output (NVO).Although it is harder to get data on the net valuethan on the gross value of output (GVO), it isworth the effort because if inputs are notdeducted, all the value added is attributed towater, which doesn’t reflect reality.

The value added for water can be calculatedfor three different levels:

1. Private farmers’ viewpoint: shows the impactof water uses on a farm level and usesfinancial prices or those paid and received byfarmers.

BOX 2.Formula to calculate the value added of water.

The value added of water is defined as: � (PjQj - CjIj)W

Pj = price of output j; Qj = quantity output j; Cj = cost of inputs necessary to produce output j and

Ij = quantity of inputs necessary to produce output j; W = volume of water.

29

2. National viewpoint: shows the impact ofdifferent water uses from the national point ofview, uses economic prices. Shadow prices oropportunity costs should be used when anenvironmental externality like damage towetlands is included. It is possible to make adivision into sectors like irrigation, municipaland domestic, industries, and theenvironment.

3. Global viewpoint: shows impact from theinternational point of view. For instance,impacts of water uses on maintenance ofbiodiversity, migratory birds, etc. It is hard tovalue those impacts. One way to value thoseimpacts is the Contingent Valuation method(see annex 2).

Going from level 1 to 3, it is necessary tomake more assumptions related to prices andimpacts. For level 1, net values are essential, butfor levels 2 and 3, societal perspective, grossvalues are suitable.

It can be hard to get an accurate measure ofthe volume of water used (denominator) bydifferent uses especially when these uses include,for instance, reuse and non-consumptive uses ofwater. Another difficulty is the term “water use”

itself. This could refer to irrigation diversions, cropevaporation and application, all having differentmeanings and implications. If the denominatorreflects the water diverted, all the value added isattributed to irrigation water while rainwater alsocontributes to production. Furthermore, the amountdiverted may be reused elsewhere. By taking thecrop evaporation as the denominator, rainwater isincluded, so the value added is attributed to rainas well as irrigation water while the problem ofreuse is eliminated. In box 3, different formulasare shown to calculate the value added pervolume of water consumed, diverted, and the totalwater supply.

Unfortunately, there enough data were notavailable on small-scale industrial activities, likecurd pot making and brick making, and thereforecould not be included in the third formula in box 3.

Results of value added calculations forproductive uses of water in Kirindi Oya are given intable 10. The gross value and the net value addedof water are calculated from the farmers’ point ofview. Therefore, no cost is imputed for the use ofland and family labor. The data used for thecalculations come from different sources. Thevalues of the denominator come from the chapteron water uses in the Kirindi Oya subbasin(chapter 3). With regard to the numerator, primary

BOX 3.Value added of volume of water consumed, diverted, and total water supply.

1. Value added per m³ of water consumed refers to evapotranspiration (ET) and can be calculated for paddy and forOFC separately:

(a) NVO paddy • ha paddy (b) NVO OFC • ha OFCvolume of ET volume of ET

2. Irrigation water is diverted to paddy and OFC cultivation, value added of volume of water diverted is:

(NVO paddy • ha) + (NVO OFC • ha)

volume of water diverted to irrigation

3. Total water supply includes irrigation water diverted and precipitation, and the different water uses are paddy, OFC,homestead, chena cultivation, livestock, and fisheries. Value added per m³ of total water supply:

(NVO paddy•ha)+(NVO OFC•ha)+(NVO homestead•ha)+(NVO chena•ha)+NVO livestock+NVO fish

volume of water diverted and precipitation

To calculate Gross Value of Output per volume of water, replace NVO with GVO in the three formulas.

30

and secondary sources were used. For paddy, itwas possible to use the data from our householdwater use survey, although the survey was notdesigned for this purpose. Secondary data sourcesare used for OFC, homestead, chena cultivation,and livestock calculations (IIMI 1990 and 1995a;EA1P 1997). The key informant interviews providedsupplementary data for livestock and fisheries. It isstriking that data on homestead and chenacultivation, especially related to water use, are veryscarce. To find the differences between wet and dryyears, the calculations are made for 1995–1996(wet) and 1996–1997 (dry). The values are allexpressed in constant 1997 Sri Lankan Rupees(US$1.00 = Rs 58.80 in 1997) and calculated perm³ of water.

As shown in table 10, for ET, a distinctioncould be made for paddy and OFC. Since there areno separate scheduled water diversions for paddyor OFC and other uses like fisheries, it was notpossible to distinguish the m³ diverted water for thespecific uses. Consequently, also the total watersupply has to be taken as one figure. In this case,OFC refers to a mixed cropping system of chili,groundnut, big onion, bananas, and vegetables.Paddy, as a high water-consuming crop, has muchlower gross value and net value added per m³ ETthan OFC. The OFCs cultivated in Kirindi Oya,especially onion and chili are high-valued crops.Hence, this results in higher gross value and netvalue added per m³ ET for OFC than for paddy. Thedifferences between the values of a dry and a wetyear are negligible for paddy but not for OFC. Apossible explanation is that the composition of theOFCs is different for the dry and the wet years.

Compared to the value added per m³ ET, thegross value and the net value added per m³diverted water and total water supply are quitesmall. As shown in box 3, the numerator of theformulas include more uses and are thus higherthan the numerator of the first formula in box 3.However, the denominators of formulas 2 and 3have much higher values. In the dry year(1996–1997), when there were less water inflowand rainfall, people were able to obtain highervalue added per m³ of diverted water and m³ oftotal water supply than during the wet year(1995–1996). This shows that people have thetendency to use water more efficiently when thereis less water available. Since the calculations arebased on first order estimates and only for a2-year period, we have to be careful with ourconclusions.

Valuing the Nonproductive Uses ofWater

A main hypothesis of this study is that the valueof water for non-crop purposes will be of asignificant magnitude when compared with thevalue for use in crop production, and it thereforefollows that non-crop uses must be taken intoaccount in the management of irrigation waterresources. The limited literature and dataavailability on the subject of valuing water forother productive uses than crop uses give us theimpression that these other productive uses areoften overlooked and, therefore, hardly taken intoaccount in water management decisions.

TABLE 10.Value added per m³ of water for different productive water uses (in 1997 rupees).

1995–1996 1996–1997

Gross value (Rs/ m³) Net value (Rs/ m³) Gross value (Rs/ m³) Net value (Rs/ m³)

ET Paddy 7.2 3.7 7.1 3.6

ET OFC 38.9 30.7 46.9 37.1

Diverted water 3.4 1.8 4.9 3.2

Total water supply 1.7 1.0 2.4 1.6

31

Moreover, we are not able to give a value forthe domestic and the environmental uses of water,though for classification purposes we considerthem as nonproductive uses. Hence, we cannotdetermine if the research hypothesis is acceptedor rejected. These uses of water are not traded inmarkets and are therefore difficult to value. Asshown before, the value of water for productiveuses is assessed through the productivity factorof water for different goods. To value, and thusquantify the domestic and the environmentalservices of water, other methodologies have to beused (see annex 2). However, there are stillproblems and concerns related to thequantification of these services. One of theproblems is the risk of missing the qualitativeessence of uses (e.g., recreation) if intangiblevalues and benefits of water are to be accountedfor quantitively (Seckler 1966). Second, thewillingness to pay is a function of the ability topay. The value reflected by the willingness to payis higher for people with a higher income. So itappears that people with a lower income attach alower value to a certain service. This doesn’tnecessarily have to be the case. Third, accordingto conventional economic theory, monetary termsare used to analyze the efficiency of resourceuse. Efficient allocation of the resource, in thiscase water, does not have to be compatible withsustainable use and equitable distribution(Bingham et al. 1995). Therefore, it is importantthat water allocation decision makers do also take

into account other information than only thequantitative values of water.

Conclusions

Values presented in this chapter are based on firstorder estimates of water accounting. Because ofthis, together with the lack of data available onother productive uses of water than cropproduction and data on domestic and theenvironmental values of water, the values in thischapter should be seen as a first indication andnot as absolute values of water. Further,substantial productivity gains should be expectedif ways could be found to save and utilize thewater lost to the ocean (see chapter 3).

With the value-added method, we are onlyable to calculate the value of water for productiveuses, where water is consumed. To value thedomestic and environmental uses of water, othermethodologies should be used. The next step isto look for suitable methodologies and tooperationalize and apply these for the Kirindi Oyacase study. Supplementary data should becollected, for example, data on positive andnegative impacts of irrigation water on thewetlands and the time spent in fetching water fromdifferent sources for domestic purposes. We willbe able to test our main hypothesis, only if we areable to calculate the value for the domestic andthe environmental uses of water.

8. Complementarities, Competition, and Conflicts

Agriculture remains the largest consumer of waterin the Kirindi Oya system, particularly paddy.Many of the other uses such as fishing or bathingare nonconsumptive, while others such asdrinking, watering livestock, collecting lotuses orreeds, and brick making consume relatively smallamounts of water compared to field irrigation.

Because they draw their water directly from theirrigation system (canals and tanks) or indirectly(from wells through groundwater recharge), there isa complementarity between these uses and fieldirrigation. When water is available in the tanks andcanals for paddy fields, it is also available forgardens, fishing, lotus production, bathing,

32

livestock production, curd pot making and brickmaking. Moreover, when water is abundant, waterquality problems are also reduced. When there isno water for irrigation, agro-wells dry up, fishstocks are depleted, milk production decreases,lotus stems must be removed, domestic water isunavailable from the canals, wells fall dry, and thepollution and salinity concentrations of theremaining water increase.

When water becomes scarce, there is morecompetition, and even conflicts over water. From ananalytical perspective, the different types of usescompete. However, from the households’ perspec-tive, this competition is not so much betweensectorally defined uses, because all householdsengage in multiple activities involving water. Withinthe household, some members may be moreaffected by the shortage (or benefit from abundance)of water for certain uses than others. The overallperception of competition is between users—particularly between the old and the new areas.

Competition, Complementarity, andConflict between Uses

The major types of interactions between differenttypes of uses are summarized in table 11.Because irrigation of field crops is the largest

water user, and holds the strongest rights to water,the relationship between each use and irrigatedcrop production is the most important form ofinteraction. However, the potential positive andnegative interactions between other uses may alsobe significant, as indicated below.

Irrigated Crop Production

The chief competition and conflict in Kirindi Oyaare not between the different types of uses, butregarding irrigating fields in the old and the newareas. There is an ongoing tension between thedemands of the Ellagala area for full paddyirrigation in two seasons, based on their historicalclaim to water, and those of the different parts ofthe new area to receive water for paddy in at leastone season. Although broad priorities have beenset, fluctuating weather and hence wateravailability require renegotiation of the areas thatcan receive water in each season. Despite theseasonal allocation process through the PMC,there have been tension and conflict, particularlyin years of low rainfall and reservoir inflow. Evenwithin the old or the new area, there may becompetition between the fields that get irrigated,especially when water supplies are short.

In terms of water quality, there are alsosignificant negative interactions between irrigated

TABLE 11.Conflicts, competition, and complementarity of water uses in Kirindi Oya.

Irrigated Live- Fishing Laundering Drinking Home Home Environ-crops stock and bathing industry gardens ment

Irrigated crops • – – – – – – –

Livestock •/+ x – – – – – –

Fishing •/+ • x – – – – –

Drinking •/+ x x • • – – –

Laundry bathing + • x x – – – –

Home industry •/+ x x x •/+ x – –

Home gardens •/+ x x x • x x –

Environment •/+ • • x x x x •

• conflicts and competitionx no conflicts and competition+ complementarity

33

fields at the local level. The salinity problem dueto poor drainage is becoming more evident inKirindi Oya, as discussed in the chapter on waterquality issues (chapter 4).

Livestock

When the Kirindi Oya system was expanded, itdisplaced a considerable number of livestock thathad been using the area for grazing and watering.Much of the ongoing competition between irrigatedcrops and animals relates primarily to grazing,rather than to water per se. Crop damage fromcattle is a recurring tension. While the volumes ofwater consumed by the animals are not asignificant source of competition, conflicts overwatering animals arise because cattle damage theirrigation infrastructure.

At the household level there is somecomplementarity between crops and livestock,because water in the irrigation system makeswater more available for animals as well. Thefarming system includes both crops and animals,with crops providing fodder, and animals providingdraft power and manure. Use of irrigation facilitiesby livestock is tolerated, as long as they do notdamage crops.

Fisheries

At one level there is considerable complementaritybetween irrigation and fishing. The reservoir,tanks, and canals of the irrigation system providethe environment for fish production, and fishingprovides a fallback occupation for some agricul-tural households during difficult times. However,the pesticides used for paddy and other irrigatedcrops flow into the water. Fish and other aquaticanimals are very susceptible to changes in waterquality. Furthermore, low tank water levels reducethe number and diversity of fish species. “Optimiz-ing” the use of tank water for irrigation purposesby reducing the level of dead storage or theamount of water stored in lower tanks during theseason (to take advantage of return flows fromupper irrigated fields) may reduce the potential forfishing.

Drinking

In terms of water quantities, there is acomplementarity between irrigation and drinkingwater, especially in the old area, where wells andseepage from surface irrigation sources providethe main source of drinking water. However, in dryyears, irrigation and drinking compete for water.The biggest conflict is generally over the right ofthe NWS&DB to keep the water in the reservoir ata certain level to guarantee domestic water needsthrough the piped water supply system. Irrigationsupply was stopped to safeguard domestic needsin 1992 which led to serious conflicts. Farmersdemanded for more water releases for irrigation,and politicians got involved in trying to settle thedisputes (Brewer, forthcoming). These issuesresurfaced in 1995 and 1997.

There is more conflict between irrigation anddrinking uses when it comes to water quality.Agrochemical runoff and leaching of minerals aswater seeps and percolates from paddy fieldshave contaminated both surface sources andgroundwater within the Kirindi Oya system, makingwell water unsuitable for drinking.

Laundering and bathing

Irrigation and bathing and laundering are largelycomplementary, because canals and tanks arevery important sources and locations for thesedomestic water uses. When irrigated cropproduction takes place, there is more wateravailable for bathing and laundering, and since thelatter are in-stream uses, they do not take wateraway from field crops. There can be, however,conflicts over water quality, as high salinity levelsmake the water less suitable for laundering andbathing.

Home industry and home gardens

The interaction between field irrigation and homegardens (household industries) is largelycomplementary. The latter use relatively smallamounts of water, which usually come from rainfallor the irrigation system (either through pumping ofwater from canals or recharge of wells). In some

34

instances, there can be a conflict if people at thehead of the canal use the water for home gardensor home industries and thereby reduce the wateravailability for the tail enders’ fields.

Environment

The availability of water in the irrigation systemhas provided a habitat for wildlife, especially birds.The Weerawila tank was designated as a birdsanctuary before the expansion of the irrigationsystem, and despite predictions that birds wouldbe displaced, their population has actuallyincreased. However, eutrophication of the tank isbecoming a problem for wildlife—caused in part byfertilizers, and also by livestock. While wadingbirds do not cause much conflict with cropproduction, intrusions by elephants from nearbywildlife areas looking for water cause conflictsresulting in crop damage.

Water quality issues are even more importantin the neighboring Bundala National Park. Inaddition to eutrophication causing excessivegrowth of algae, there are problems maintainingthe salinity balance in the lagoons if there is toomuch discharge from the irrigation system. Twobrackish lagoons have been converted intofreshwater lakes. This has had a negative impactnot only on water birds, but also on the hundredsof families that were formerly engaged in prawnfishing in the lagoons.

Livestock

Fisheries

Since neither livestock nor fisheries consumemuch water, the interactions with respect toquantity are negligible. However, livestock pollutesthe water with dung and urine. This can causeeutrophication, which can have a negativeinfluence on the fish (although some speciesthrive under such conditions).

Laundering and Bathing

As in the case of interactions with fisheries,livestock pollutes water. Since people enter the

tanks and canals to bathe and launder, the dungand urine from livestock can cause a public healthrisk. The negative effect on bathing and launderingis especially pronounced when the water level is low.

Environment

Conflicts between livestock and the environmentalinterests over access to tank areas and land haveled to a number of disputes between the COFO andthe Wildlife Department. In terms of water quality,there are further conflicts because overgrazing anddeposits of dung and urine in the water can causeeutrophication, especially in the Bundala lagoons.

Fisheries

Environment

In some of the tanks and the Lunuganweherareservoir, conflicts arise between the Departmentof Wildlife Conservation and the FishermenCooperative Societies. These conflicts relate toaccess to the tank areas in the evening (and thusdisturbing wildlife) which is the preferred time to fish.

Drinking

Although the total amount of water for drinking andcooking is relatively small, this water is veryimportant to satisfy basic human needs, and there isnot always enough available at the standposts. Thus,conflicts between different users arise at the tap.

Laundering and bathing

When there is no water available in the canals,people use the tap water also for laundering andbathing. This is prohibited by the NWS&DB(NWS&DB 1997a), and causes conflicts becauseif some take water for bathing, there is lessavailable for others at the standpipe, and theadditional costs are borne by all.

Home industry and home gardens

As in the case of interaction between drinking andbathing, when the tap water is used for gardens orhome-based industries, conflicts arise because

35

people use more than they are supposed to, whichleads to longer queues and higher costs for othersat the standpost. However, there is also a strongcomplementarity between drinking (and especiallycooking) water and home gardens, becausewastewater is used as a source for gardens.Surplus water spilled at the standpipes is alsooften channeled into nearby gardens.

Environment

Salinity in paddy fields has become a significantenvironmental problem in certain parts of thecommand area. But mitigating this problemthrough leaching and drainage displaces theminerals and solids, which can cause problems inother areas, including domestic wells and thewildlife sanctuary.

Interaction between Users: TheHousehold Perspective

Paddy production is the primary source oflivelihood for many people. Thus, most people(especially those in households with at least somepaddy fields) are more concerned with theallocation of water between different areas(primarily for irrigation) than between different

uses. However, the potential complementaritybetween irrigation and other uses should not beoverlooked. When water is released in the canalsand tanks in an area to supply crops, it is alsoavailable for livestock, fishing, gardens, bathing,and other enterprises. Not all members of ahousehold have an equal interest in all types ofwater use. Men, women, and children havedifferent responsibilities related to each type ofwater use, and derive different benefits.

While water quantity issues continue to be themost prominent, there is considerable awarenessof water quality problems, particularly for drinkingand bathing. People complain not only of thepalatability of water, but also of its hardness. Theyare also aware of the role of agricultural chemicalsand seepage in contributing to water qualityproblems. However, this does not prevent themfrom using chemicals on their own fields. Sincerunoff upstream is what affects each area’s waterquality, a family would not benefit, in terms ofwater quality, from refraining from chemical use.Because the problem is not localized, localcollective action is also unlikely to have an effect.Addressing the water quality problems—for thebenefit of both household use and theenvironment—would require more widespreadregulation.

Introduction

In the classical sectoral approach to irrigatedagriculture, the irrigation infrastructure,management, and institutional arrangements servethe objective of efficient use of water for foodproduction. In this report we have argued thatsuch a sectoral view that does not take intoaccount water uses pertaining to other sectors istoo limited in scope. The sectoral approach is

9. Future Directions for Research

more and more replaced by the concept ofIntegrated Water Resources Management (IWRM).While the theoretical concept is still developing, itseems that IWRM takes water as a naturalresource as the starting point and then analyzeshow this resource can be managed in anintegrated and sustainable way by buildinginstitutional capacities to satisfy human needs,promote food security, and protect theenvironment. Our approach of looking at the

36

multiple uses of water differs from IWRM in thesense that it is more anthropocentric, i.e., it startswith the question: who are the users of water andwhat are their uses? It puts the people, themultiple users of water at the fore whileappreciating that people operate in a certainphysical and institutional environment. Itdocuments why certain sources are preferredabove others, how people cope with periods ofdrought, and analyzes how far the tasks in relationto water are gender-specific. Finally, it tries toestimate the value of different uses of water. Wethink that such an analysis of the multiple usersand uses of water should precede any attempt forimplementing IWRM.

Research Methodology

A combination of qualitative and quantitativemethods as used in the present case study willremain the best option to describe the differentwater uses at the sectoral and the householdlevels. While certain patterns are likely to berepresentative for other areas of Sri Lanka,variations between countries will be considerable.The methodology for the systematicdocumentation of multiple uses of water has to berefined further and applied elsewhere. While mostof the information required to describe multipleuses and users of water could be obtained withthe methodologies used in this case study, theprocedures were rather labor-intensive and time-consuming, especially the household interviewsurvey. More efficient research methods should beconsidered. Ideally, these would be rapidassessment and even participatory appraisalprocedures that can validly describe sectoral andhousehold level water uses and users (Gosselinkand Thompson 1997; Chambers 1994; Gosselinkand Strosser 1995; Pretty et al. 1995).

In reviewing the output of the present study, alot of descriptive information was available throughkey informant interviews and direct observations inthe field. Another point of attention in further

studies should be the validity of interview-baseddata. To quantify use and users at the householdlevel, direct observations could provide morereliable data than household interviews, but at agreater cost. Methodologies for direct observationscould be obtained from the extensive literature thatis available on human water contact studies inrelation to schistosomiasis (see for example Klooset al. 1983). Provided a valid sampling frame isused, such studies can also give a better idea ofgender-based water use than the householdinterviews. Group discussions, when used at thebeginning of a case study, can be a usefulexploratory process when little is known about thestudy population. They make it easier to compile aquestionnaire for collecting quantitative data orrecording sheets for water contact observations.This would prevent the gathering of exhaustive,irrelevant information. However, the quantitativedata collected through household interviews canstill be difficult to interpret, owing to the complexityor ambiguities they contain. A second round ofgroup discussions at the end of the case studiescan then provide a fuller understanding of thenumerical results, and provide relevant feedback topeople in the study community.

Research Priorities

Measuring Water Use and Economic Values

In the discussion on competition for scarce waterresources in water basins with multiple uses, it isvery important to be able to assign economicvalues to the different uses. While techniques(see annex 2) exist for valuing water for economicuses, as seen in this study, they are oftencumbersome and expensive to apply. Simpler, butrobust techniques are needed, not only forresearch purposes, but to ensure that thevaluation methods are understood by systemmanagers and policy makers. Further, there is alack of data regarding the inputs and outputs ofproduction for water uses other than the main field

37

crop (e.g., chena cultivation, fisheries, etc.) and itis recommended that more time and effort are putin gathering this kind of data.

One common feature of all types of economicvaluation is the denominator: water use. Gettingan accurate measure of water use by differentsectors can be difficult (if not impossible),especially when the uses include return flows,reuse, and nonconsumptive use. Adding to thiscomplexity, the term “water use” has differentmeanings to different people and could refer to(irrigation) diversions, application, or evaporativeuse, all having different implications. This studyhas suggested a means to account for wateruse by different users in a consistent manner. Itwould be useful to refine and adapt this method toother situations so that it can have broaderapplicability.

Valuing the Nonproductive Uses of IrrigationWater

As complicated as it is to estimate the value ofwater used for productive purposes, it is muchmore difficult for the nonproductive uses; domesticand the environmental uses. In fact,methodologies for valuing nonproductive wateruses are existing but they are applied only inrelation to a single sector or use of water, eitherthe domestic use (Altaf, Jamal, and Whittington1992; Whittington and Swarna 1994) or theenvironmental use (Barbier, Acreman, and Knowler1997). Further development and testing of suchmethodologies and developing a framework forvaluing water in an integrated water resourcesystem are priorities for further research. Like withall aspects of multiple use, this should be aninterdisciplinary effort. A lot of useful informationon the value of uses in a particular sector mightbe available in the specialized literature. Anexample is the cost-of-illness methodologiesdeveloped for water-related diseases.An extensive interdisciplinary literature review istherefore needed, including less traditionalliterature sources, before designing newstudies.

Water to Sustain Aquatic Ecosystems

The environmental functions of irrigation water willbe addressed in a follow-up to the present KirindiOya case study. Irrigation drainage water affectsthe important wetland ecosystem of the BundalaNational Park. The aim is to develop amethodology that could be applied elsewhere andthat could address the following questions:

• How does irrigation water management affectthe ecology of downstream wetlands?

• What are the water management options thatcould conserve wetlands?

• What are the water use options that will bestserve the interests of different users,especially those of poor rural communities?

Water as a Basic Human Need

A case study similar to the one in Kirindi Oya wascompleted in Pakistan (Jehangir et al. 1998). Themain objective of that study was to get an accurateassessment of all the uses and users of water inthe irrigation system. In the study area, peopledepend on irrigation water for all their domesticrequirements, even for drinking. Therefore, inPakistan, health impacts are more important issuesthan in Sri Lanka. To bridge the gap betweenirrigation and the domestic water supply sector, adetailed epidemiological and water quality studyhas now been started. In Pakistan and elsewhere,we want to address the following questions withrespect to water as a basic human need:

• To what extent is irrigation water used fordomestic purposes?

• What is the health impact of these domesticuses of irrigation water?

• What adaptations in irrigation system designand operation are needed to make domesticuse of irrigation water a safe option?

• How will more efficient irrigation and“improved” irrigation water managementpractices affect the quantity and quality ofwater available for domestic purposes?

38

Introduction

Disciplinary and subsectoral emphases have toolong focused the attention of researchers, policymakers, and agency staff involved with waterresources on only one of the following wateruses—irrigation, domestic use, fishing, orlivestock—when in fact, people have been usingwater in irrigation systems for many purposes.Going beyond the disciplinary and sectoralblinders changes our picture of irrigation systemsand allows us to see the full spectrum of wateruses. It also expands the view of water usersbeyond those (primarily male) farmers in the fieldsor those (primarily women and children) drawingwater at the standpipe. In this study, we haveseen how fishermen, livestock herders, and curdpot makers, and even the birds and animals arewater users who depend on the irrigation systemfor their livelihoods.

Recognizing the full spectrum of uses is farmore than an academic exercise. It has importantimplications for the management of water withinthe irrigation system, and also for a broader waterresource policy. The present study has been onlya pilot activity, and could not explore and quantifyall the water uses and their values. Nevertheless,it points to critical issues to be addressed. Thisconcluding chapter identifies a number of theseissues, within Kirindi Oya, and then addressespolicies such as intersectoral allocation andinfrastructure development.

Management Issues within Kirindi Oya

The allocation of irrigation water between differentparts of the Kirindi Oya system continues to bethe single most important management issue,even if multiple uses are taken into account. It isnot so much that a rising tide lifts all ships, butthat an irrigation release fills all pots, meets mostother water needs, and dilutes the contaminants.

10. Conclusions: Implications for Policy and Management

When water is scarce, cutting back on irrigation toreserve water in the reservoir and providingspecial water releases in the canal and river fordomestic supply become critical decisions thathave provoked considerable conflicts in the past.

The issue of the water level that should bemaintained in the tanks is another managementdecision, which illustrates some of the potentialtrade-offs between different uses. Keeping thetank levels high during the maha season causesconsiderable spillage, which is not recaptured, andwater flows out of the system to the ocean. If thetank levels were kept lower, more of the rainfalland drainage from the new area could berecaptured. This would be more efficient from thestandpoint of irrigation, because the same areacould be irrigated with less reservoir releases,saving water for other areas or for the nextseason. However, the lower tank levels wouldreduce the availability of water for bathing,washing, livestock, and fishing in the tanks.Moreover, reducing the water levels in the tanksand relying more on recycling would increase theconcentration of various contaminants (includingagrochemicals and fecal coliform bacteria). On theother hand, conserving water would reduce theproblems in getting water for domestic and otheruses in the yala season, when water is usuallyvery scarce, and the reduction in drainage wouldallow the Bundala lagoons to remain brackish forprawns and other fauna that depend on the originalbrackish conditions. Considering all the uses ofwater complicates the decision-making processbecause it shows that maximizing the efficiencyfor irrigation may not be the same as “optimizing”for all uses.

From a household perspective, it is unlikelythat a reduced quantity of irrigation water allocatedto the system has a major impact on thehousehold supplies of water for drinking, cooking,sanitation, and the washing of utensils. However,in the long term, reduced seepage of irrigationwater may have an impact on the availability of

39

water in the homestead wells and thereby affectwater availability, for instance, for sanitarypurposes. A reduced availability of water in thecanals and tanks will have a significant impact onthe availability of water for personal hygiene,laundering, and recreational use. The impact willdiffer throughout the system, where some familieswill be able to shift their priority to other sources,while many families living in areas with poor pipedwater supply will definitely feel an impact on theirhouseholds. The impact of a reduced irrigationwater supply will increase the pressure on thepiped water supply throughout the system,especially in the areas where the groundwater isbrackish. This is likely to lead to increasedconflicts between the users of the standpipes.Another possible general impact is a reducedoverall per capita use of water for hygienepractices.

Water rights in Kirindi Oya are not clearlydefined, especially for uses other than irrigationand domestic supply. The PMC has theresponsibility for water allocation, but despite therange of government and user organizationsrepresented, it has not recognized the range ofwater uses, or the challenge of managing water tomeet all needs.

Because many of the water uses arenonconsumptive (e.g., fishing), or require relativelysmall amounts of water (e.g., curd pot making),they do not compete with other uses in terms ofthe volume of water, and as long as water isrelatively abundant, it is not worthwhile to define aquantitative right. Although many of these usesmay not directly conflict with one another, whenwater demand increases, or when water supplydecreases, competition for water resourcesfollows. For many uses, quality issues are oftenmore important than quantity (e.g., domesticwater, fishing, or wildlife). Hence, the critical rightsare not for withdrawal, but for management of theresource (and potentially for exclusion of otherusers that pollute). For example, fish is highlysensitive to salinity and agrochemical pollution,and its production is reduced when water levels in

the tank are too low or too high. Thus, althoughfishing is a nonconsumptive water use, fishermenhave a strong interest in the management of tankwater levels, and in the interactions with otheruses. However, the rights and coordinationmechanisms to deal with such issues are notdefined at present.

Finally, considering the interactions among themultiple water uses highlights the issues of waterquality. These are critical not only for domesticuse, but also for fishing and the wildlife. Atpresent, there is virtually no attention given towater quality. Measures to handle sewage andlivestock wastes are ineffective, and there are nomeasures to limit contamination fromagrochemicals. Experience in industrializedcountries has shown that it is difficult to handlesuch nonpoint source pollution. However, raisingawareness and discussing the issues arenecessary if the water of Kirindi Oya is tocontinue to support multiple uses.

Implications for Water ManagementPolicies

While the exact uses and users of water and theirrelative importance vary from one irrigation systemto another, the issues identified in this pilot studyin Kirindi Oya have broader implications for watermanagement policies in Sri Lanka and elsewhere.These relate to the allocation of water andfinancial resources between irrigation and othersectors; measures of water quality and efficiencyof use; and mechanisms to involve allstakeholders in negotiations over water use.

With the growth of cities and industries andthe relative decline of agriculture in economiesaround the world, inter-sectoral competition forwater has become a major issue, and irrigationsystems often lose out relative to municipal andindustrial uses. Inter-sectoral water allocation isgenerally viewed as a process of determining howmuch water goes into a municipal system, afactory intake, an irrigation system, or natural

40

reserve—implicitly or explicitly equating domesticuse with municipal systems, industrial use with(licensed) factories, agriculture with irrigationsystems, and biodiversity or environment withwetlands and reserves. This study hasdemonstrated that water in the Kirindi Oyairrigation system is used for much more thansimply producing the field crops, which is oftenattributed to irrigation. Although an exactquantification is extremely difficult, the social andeconomic value of water within the irrigationsystem is much higher when we account for alluses, than is recognized in many conventionalanalyses of inter-sectoral water uses.

Because of Kirindi Oya’s distance from majorurban centers, municipal and industrial uses havenot been major competitors with irrigation for thewater resources available in the system. There areplans to build an oil refinery on the coast, whichwould require substantial allotments of water everyday from the Kirindi Oya system. Recognizing themultiple uses of water changes the analysis ofinter-sectoral water allocation, especially forreallocation out of irrigation. A simplistic analysismight suggest that the water needed for therefinery could be met by improving the “efficiency”of irrigation deliveries, taking a certain area out ofpaddy production, or a combination of theseapproaches. A more comprehensive view wouldrecognize that changing the quantity and timing ofwater deliveries to supply such a major industrywould not only affect system managers or thefarmers who have to switch crops but also theother users of water. It would have ripple effectsthroughout the system, affecting groundwaterlevels and hence water availability for drinking andhomegardens; tank levels and hence fishproduction; runoff and hence concentration ofchemicals; and salinity entering the wetlands,among other factors.

Water for hotels to meet a growing touristindustry is also a serious current issue of waterallocation within Kirindi Oya, and illustratesanother aspect of inter-sectoral water use: theimpact on water quality. Although hotels, like other

domestic uses, may not deplete water supplies asmuch as agriculture, they create wastes that cancontaminate downstream water. Currently, the IDhas denied requests for surface water allocationsto hotels, but there has been no effectiverestriction on groundwater abstractions.Mechanisms for monitoring such abstractions andthe impact of their uses on water quality arecurrently weak. Dealing with these aspects ofwater resource management requires going farbeyond the existing sectoral approaches todeveloping irrigation or domestic water supplies.

As water resources become more fullydeveloped, countries often turn from investmentsin new systems for water capture and storage toimprove the efficiency of existing water supplysystems (particularly irrigation systems). Canallining, sprinkler or drip application systems, androtational irrigation schedules are common meansof increasing the proportion of water in the systemthat is used for crop evapotranspiration. The water“saved” through these measures is then seen asavailable for reallocation to other uses or users(e.g., expanding the area irrigated or supplying tomunicipal systems). But reducing the seepage andpercolation through canal lining or sprinkler anddrip systems often lowers water tables, affectingthe wells for drinking and gardens. In places likeKirindi Oya, where permanent vegetation relies onhigh water tables recharged by irrigation seepage,such measures could also threaten quite a bit ofhigh-value horticultural production (e.g., coconut,mango). Rotational water deliveries could alsocreate problems for livestock and bathing; thecrops may be able to go on for days or evenweeks between waterings, but people and animalswho depend on the canals as sources of bathingor drinking water need the water on a daily basis.This is to conclude that when measures are takento improve irrigation efficiency, the impact it mayhave on other water users also needs to be takeninto account.

Recognising the various uses and users ofwater is an important step toward managing thesystem to accommodate all needs. However,

41

trade-offs between different uses are inevitable. Insuch instances, decisions can be externally made(e.g., by a government agency), or can benegotiated among various stakeholders. The latterhas the potential to reach decisions that are moreacceptable to a range of parties, but it requiressome form of platform for negotiation (Röling1994; Steins and Edwards 1998). The governmenthas been striving to devolve management andincrease user involvement, and the establishmentof a range of formally recognized userorganizations (e.g., the hierarchy of FOs, COFOs,and FCSs) can be seen as a step towardestablishing a platform for negotiation within asingle use sector. Institutions such as the PMC,which bring together farmer representatives fromvarious parts of the system along withrepresentatives from a range of governmentagencies, can provide a platform for negotiationover the use of water for irrigating field crops. Itmay be possible to expand the PMC, whichcurrently includes government representativesfrom a number of agencies, as well as farmersand livestock owners’ user groups, to representother interests such as fisheries, domestic water,and the environment. Bringing in representativesof other categories of users can be done throughpublic meetings to discuss water managementplans, either on a seasonal basis, or especiallywhenever management plans are suggested.

To the extent that the members of the FOsalso use water for domestic purposes, gardens,and even fishing, it might appear that these othertypes of uses could be represented by the farmerrepresentatives. However, experiences from KirindiOya in 1992, 1995, and 1997 show that at criticaltimes, the farmers give priority to water forirrigating the field crop instead of giving priority towater for domestic purposes. While farmhouseholds may be involved in all of these uses,there are important intra-household differences inresponsibility for these various types of uses.Membership in the FOs at all levels is heavily

dominated by males (Meinzen-Dick andZwarteveen 1998). Even though the interests ofmen and women are often complementary, thereare important differences in priorities for water use(Zwarteveen 1994). Thus, the significant barriersto gender equity in participation should beaddressed for effective overall management of allwater uses.

The twelfth century Sinhala King Parakrama-bahu I is quoted as saying:

Let not even a small quantity of waterthat comes from the rain flow into the oceanwithout being made useful to man.

If we recognize the multiple uses of water, wesee that, while some water still runs to the sea,much of it is used by men and women, severaltimes over. This undoubtedly increases the valueof water use, and it needs to be taken intoaccount when evaluating irrigation systemperformance. But in the process of being used forso many purposes, the water picks up a variety ofcontaminants, such as fertilizers, pesticides,salinity from the fields, soap residues, bacteriafrom domestic and livestock uses, and othertypes of chemicals. Recognizing the interactionsbetween uses and users may also provide scopeto better accommodate the various uses, therebyto increase the efficiency of water use. But toincrease the total value of productive and non-productive uses within irrigation systems, moreattention should be focussed on water qualityissues, as well. Unfortunately, as difficult as it isto develop accurate empirical measures ofquantitative efficiency or system performance, it iseven more difficult to measure and incorporatemeasures of water quality. On the output side, theuses that are particularly susceptible to waterquality (especially drinking water and wildlife uses)are very difficult to place quantitative values on.This remains a critical area for further research aswell as for action in water management.

42

ANNEX 1

Water flow chart

43

Alternative Costs/ Opportunity Costs

If a given water allocation with specified outputcosts less than the next water allocation whichcan achieve the same output, then the cost of thenext-best option can be considered as the benefitto the water allocation under consideration. Thismethod may be a solution when estimations of adirect demand schedule prove to be difficultbecause of lack of data or other reasons. This issimilar to the opportunity costs calculation, whichcalculates the benefit foregone by using a scarceresource for one purpose instead of its next-bestalternative use. The benefits foregone can beused to value the water for that purpose. Anaspect which can be taken into account whenusing this methodology is the quality of waterbecause this influences the suitability of the waterfor other purposes and thus affects the number of(next-best) alternative uses. This is important tonote because water quality issues are oftenignored.

This method can be used to value thedomestic water use from different sources. In thevillages in Kirindi Oya, there are several watersources that are used for domestic uses. Eachwater source has its own characteristics (distancefrom the house, quality perceived, reliability, etc.)and is selected for different uses (see alsochapter 6). According to Whittington and Swarna(1994), water from different sources is a differentgood in terms of quality and servicecharacteristics. Water from different sources is aclose but not a perfect substitute. The costsconsist of resource costs plus a money price ofobtaining water from that source. For a public tap,for instance, this includes the opportunity cost oftime spent walking to the tap, waiting in the

queue, walking home, plus the price paid for tapfacilities. The total costs will vary amonghouseholds, as the opportunity cost of the timespent for collecting water will be differentdepending on the distance from the water source.This will result in different water values for waterfrom different sources.

In addition, this method can be useful to valuethe water for agricultural crop and non-cropproduction like fisheries, industrial use, and theenvironmental use (recharging groundwater table).

Contingent Valuation Method

Another way to value water, often applied whenservices are improved is the Contingent ValuationMethod. This method determines the market valueby trying to get people to reveal what they wouldbe willing to pay for water and services inhypothetical markets. Individuals are surveyed todetermine their willingness to pay for a specifiedchange in the quality or quantity of water. Themean value of the willingness to pay across allbids (including valid zero bids) is then used toprovide an indication of the economic value of thespecified change. The quality of the results of thismethod depends on how well-informed people are;it does not adequately incorporate long-term goalssince it excludes future generations from biddingin the markets. It is also difficult to induceindividuals to reveal their true willingness to payfor natural resources when the question is putdirectly (Erskine 1997). Potentially, serious biasescould arise from the use of this method and theestimates derived should be viewed as broadlyindicative, rather than knowledge-based (Pigram1997).

ANNEX 2

Other Valuation Techniques

44

Suitability of valuation techniques in the Kirindi Oya case study.

Agriculture Fisheries Domestic Industry Recrea- Environ-

Crop Non-crop tion ment

Alternative cost/ opportunity cost ✔ ✔ ✔ ✔ ✔ ✔ ✔

Contingent valuation ✔ ✔ ✔ ✔ ✔ ✔ ✔

Hedonic pricing – – – – – – –

Travel cost – – – – – ✔ ✔

This method can be used to reveal thewillingness to pay for improved water quality fromthe wells or reliable water supply from thestandpipes in Kirindi Oya. Altaf, Jamal, andWhittington (1992) applied the Contingent ValuationMethod in the Punjab, Pakistan. They used themethodology to determine the willingness of thehouseholds to pay for improved service levels. Itis also possible to apply the Contingent ValuationMethod for agricultural crop production. Farmers,especially in the new area can be asked whatthey are willing to pay for a more reliable irrigationwater supply. Further, this methodology isapplicable to agricultural non-crop production,industrial, and recreation/environmental water use.For instance, people can be asked what they arewilling to pay for preserving the wetlands inBundala or for having a bath in one of the tanks.

Hedonic Pricing Method

This method is based on the concept that theprice paid for a complementary good (e.g., aresidential property) reflects the buyer’swillingness to pay for a particular environmentalgood (e.g., adjoining a river). Application of thismethod requires the use of regression analysis todetermine the relationship between the marketprice of the property and its attributes, of whichone (set) relates to associated environmentalcharacteristics. From this, the implicit price for theassociated environmental characteristics isderived. This method rests on the assumption that

the price of some marketed good is a function ofits different characteristics, and an implicit priceexists for each of the characteristics (Young1996). So, it is necessary that active andcompetitive agricultural land and real estatemarkets are in place to use the Hedonic PricingMethod. In Kirindi Oya, agriculture and real estatemarkets are not active and competitive. If thesepreconditions are there, land values can bederived from samples of land sales representingirrigated and nonirrigated land. A comparison ofthe irrigated and the nonirrigated land values canprovide a useful and relatively convincinginformation about the revealed preference forirrigated land. The difference between the value ofirrigated and nonirrigated land represents the valueof irrigation water.

Travel Cost Method

This method is based on the concept that peoplespend time and money traveling to recreationalsites and that these expenditures, or costs, canbe treated as revealing the demand for the site.Surveys of site visitors are undertaken todetermine the demand for a site. Visit rates are afunction of travel expenditure, income, entry fees,environmental characteristics and the availabilityof substitute sites (Postle, Berry, and Westscott1997).

The Travel Cost Method might be useful tovalue the Bundala National Park which is part ofthe research area. However, the money spent on

45

traveling and entrance fees reflects the value ofthe recreational site as a whole and not just thevalue of the water that it contains. So, when usingthis methodology to value the water in BundalaNational Park, the water will be overvalued (Burrill1997).

The table below gives an overview of themethodologies described in this annex, and thesuitability to apply those on the different wateruses in Kirindi Oya.

Literature Cited

Abeygunawardena, P. n.d. Total economic value of forests: The case of Sinharaja Reserve in Sri Lanka. Sri Lanka:Faculty of Agriculture, University of Peradeniya.

Altaf, A., H. Jamal, and D. Whittington. 1992. Willingness to pay for water in rural Punjab, Pakistan. UNDP-World BankWater and Sanitation Program. Water and Sanitation Report No.4. Washington, D.C., USA: The World Bank.

Barbier, E. B., M. Acreman, and D. Knowler. 1997. Economic valuation of wetlands. A guide for policy makers andplanners. Gland, Switzerland: Ramsar Convention Bureau.

Bauer, C. J. 1997. Bringing water markets down to earth: The political economy of water rights in Chile, 1976-95. WorldDevelopment 25(5):639-656.

Bhatia, R. 1997. Food security implications of raising irrigation charges in developing countries. In Water: Econom-ics, management and demand, eds. M. Kay, T. Franks, and L. Smith. London: E & FN Spon.

Bingham, G., R. Bishop, M. Brody, D. Bromley, E. Clark, W. Cooper, R. Constanza, T. Hale, G. Hayden, S. Kellert, R.Norgaard, B. Norton, J. Payne, C. Russell, and G. Suter. 1995. Issues in ecosystem valuation: Improving informa-tion for decision making. Ecological Economics 14(1995):73-90.

Brewer, J. D. 1994. The participatory irrigation system management policy in Sri Lanka. Economic Review 20 (6):4-6, 8-9.

Brewer, J. D. Forthcoming. Negotiating water allocation rules in a government managed irrigation system: Conflictsin Kirindi Oya. In Negotiating water rights, eds. B. Bruns, and R. S. Meinzen-Dick. New Dehli: Sage.

Briscoe, J. 1996. Water as an economic good: The idea and what it means in practice. Special Session R 11, Six-teenth Session. Cairo, Egypt: International Commission on Irrigation and Drainage.

Burrill, A. 1997. Assessing the societal value of water in its uses. Sevilla, Spain: Institute for Prospective Technologi-cal Studies, Joint Research Centre of the European Commission.

Caldas, J. C., P. L. Sousa, and L. S. Pereira. 1997. Need for new approaches in valuing water: Learning from a lesssuccessful case. In Water: Economics, management and demand, eds. M. Kay, T. Franks, and L. Smith. London:E & FN Spon.

CEA (Central Environmental Authority). 1994. Wirawila-Tissa sanctuary & Yodawewa. Wetland site report & conser-vation management plan. Sri Lanka: Central Environmental Authority/Euroconsult, Ministry of Transport, Highways,Environment and Women’s Affairs.

CEA. 1993. Bundala national park. Wetland site report & conservation management plan. Sri Lanka: Central Envi-ronmental Authority/Euroconsult, Ministry of Environment and Parliamentary Affairs.

Chambers, R. 1994. Participatory rural appraisal (PRA): Analysis of experience. World Development 22(9):1253-1268.

46

Cooray, P. G. 1984. The geology of Sri Lanka (Ceylon). Colombo: National Museum Publication, Department of Gov-ernment Printing.

EA1P (Environment Action 1 Project). 1997. Economic and environment linkages in land use for irrigated versusrainfed farming in the dry zone. Colombo: Ministry of Forestry and Environment and The World Bank.

Erskine, J. M. 1997. Sustainability measures for natural resources. In People and participation in sustainable devel-opment. Understanding the dynamics of natural resource systems, eds. G. Shivakoti, G. Varughese, E. Ostrom, A.Shukla, and G. Thapa. Proceedings of an International Conference held at the Institute of Agriculture and AnimalScience, Rampur, Chitwan, Nepal, 17-21 March 1996. Bloomington, Indiana, USA: Workshop in Political Theoryand Policy Analysis, Indiana University.

FAO (Food and Agriculture Organization). 1992. CROPWAT, a computer program for irrigation planning and man-agement. Irrigation and Drainage Paper 46. Rome, Italy: FAO.

Fereres, E., and F. Cena. 1997. Social benefits and environmental constraints of irrigation in an era of water scar-city. In Water: Economics, management and demand, eds. M. Kay, T. Franks, and L. Smith. London: E & FN Spon.

Gosselink, P., and P. Strosser. 1995. Participatory rural appraisal for irrigation management research: Lessons fromIIMI’s experience. Working Paper No. 38. Colombo, Sri Lanka: International Irrigation Management Institute.

Gosselink, P., and J. Thompson. 1997. Application of participatory rural appraisal methods of action research on watermanagement. Short report series on locally managed irrigation No. 18. Colombo, Sri Lanka: International Irriga-tion Management Institute.

Hoogendam, P. 1995. Water rights in irrigation management: The property issue. Unpublished lecture notes,Wageningen Agriculture University.

IIMI (International Irrigation Management Institute). 1990. Irrigation management and crop diversification. Volume II,final report on the Technical Assistance Study. Colombo, Sri Lanka: International Irrigation Management Institute.

IIMI. 1995a. Kirindi Oya Irrigation and Settlement Project. Project Impact Evaluation Study. Volume II: Annexes (finalreport). Colombo, Sri Lanka: International Irrigation Management Institute.

IIMI. 1995b. Irrigation management and crop diversification. Volume II, final report on the Technical Assistance Study,phase II. Colombo, Sri Lanka: International Irrigation Management Institute.

ILACO (International Land Development Consultants). 1981. Agricultural compendium for rural development in thetropics and the subtropics. International Land Development Consultants, Arnhem, The Netherlands: Elsevier Sci-entific Publishing Company.

Jehangir, W. A., M. Mudasser, Mahmood-ul-Hassan, and Z. Ali. 1998. Multiple uses of irrigation water in the Hakra6-R distibutary command area, Punjab, Pakistan. Report No. R-61. Lahore, Pakistan: International Irrigation Man-agement Institute.

Kloos, H., G. I. Higashi, J. A. Cattani, V. D. Schlinski, N. S. Mansour, and K. D. Murrell. 1983. Water contact behaviorand schistosomiasis in an upper Egyptian village. Social Science and Medicine 17:545-562.

Konradsen, F., W. van der Hoek, C. Perry, and D. Renault. 1997. Water: Where from, and for whom? World HealthForum 18(1):41-43.

Meinzen-Dick, R., and M. W. Rosegrant. 1997. Water as an economic good: Incentives, institutions, and infrastruc-ture. In Water: Economics, management and demand, eds. M. Kay, T. Franks, and L. Smith. London: E & FN Spon.

Meinzen-Dick, R., and M. Zwarteveen. 1998. Gendered participation in water management: Issues and illustrationsfrom water users’ associations in South Asia. In Gender analysis and reform of irrigation management: Concepts,cases and gaps in knowledge. Proceedings of the international Workshop on Gender and Water. September 1997,eds. D. J. Merrey, and S. Baviskar. Colombo: International Water Management Institute.

Molden, D. 1997. Accounting for water use and productivity. SWIM Paper 1. Colombo, Sri Lanka: International Irri-gation Management Institute.

47

Morris, J., E. K. Weatherhead, J. A. L. Dunderdale, C. Green, and S. Tunstall. 1997. The feasibility of tradeable per-mits for water abstraction in England and Wales. In Water: Economics, management and demand, eds. M. Kay, T.Franks, and L. Smith. London: E & FN Spon.

Munasinghe, M. A. K., and S. Somasiri. 1992. Moisture consumption by dominant vegetation types in the dry zoneof Sri Lanka and its influence on the hydrological cycle. Tropical Agriculturist 148. Peradeniya, Sri Lanka: Landand Water Management Research Centre.

National Water Supply and Drainage Board (NWS&DB). 1985. Feasibility study on water supply to new towns & settle-ment areas under Kirindi Oya project. Sri Lanka: National Water Supply and Drainage Board.

NWS&DB. 1997a. National Water Supply and Drainage Board, Notice to all consumers of water. Sri Lanka: NationalWater Supply and Drainage Board.

NWS&DB. 1997b. Water tariff structures. Colombo, Sri Lanka: National Water Supply and Drainage Board, Ministryof Housing and Construction.

Perry, C. J., M. Rock, and D. Seckler. 1997. Water as an economic good: A solution, or a problem? Research Report14. Colombo, Sri Lanka: International Irrigation Management Institute.

Pigram, J. J. 1997. The value of water in competing uses. In Water: Economics, management and demand, eds. M.Kay, T. Franks, and L. Smith. London: E & FN Spon.

Postle, M. G., N. Berry, and R. Westcott. 1997. The economics of low flow rivers. In Water: Economics, managementand demand, eds. M. Kay, T. Franks, and L. Smith. London: E & FN Spon.

Pretty, J. N., I. Guijt, I. Scoones, and J. Thompson. 1995. A trainer’s guide for participatory learning and action. Lon-don: International Institute for Environment and Development.

Rasmussen, L. 1994. Comprehensive water resources management. In Water resources management in Sri Lanka.Draft strategic framework and action plan, Volume 2: Annexes. Working paper prepared under Technical Assis-tance Project. TA 1918-SK1, with the Asian Development Bank and USAID.

Reidinger, R. 1994. Observations on water markets for irrigation systems. In Water policy and water markets: Selectedpapers and proceedings for the World Bank’s ninth annual irrigation and drainage seminar, Annapolis, Maryland:December 8-10, 1992, eds. G. Le Moigne, K.W. Easter, W.J. Ochs, and S. Giltner. Washington D.C.: The World Bank.

Röling, N. G. 1994. Communication support for sustainable natural resource management. IDS Bulletin 25:125-133.

Sakthivadivel, R., D. J. Molden, and C. De Fraiture. Forthcoming. Water accounting for use and productivity of basinwater: Some examples. Research Report. Colombo, Sri Lanka: International Water Management Institute.

Schlager, E., and E. Ostrom. 1992. Property-rights regimes and natural resources: A conceptual analysis. Land Eco-nomics 68(3):249-262.

Seckler, D.W. 1966. On the uses and abuses of economic science in evaluating public outdoor recreation. Land Eco-nomics X L11(4):485-494.

Seckler, D., U. Amarasinghe, D. Molden, R. de Silva, and R. Barker. 1998. World water demand and supply, 1990-2025: Scenarios and issues. Research Report 19. Colombo, Sri Lanka: International Water Management Institute.

Stanbury, P. 1989. Land settlement planning for improved irrigation management: A case study of the Kirindi Oya Ir-rigation and Settlement Project. IIMI Country Paper Sri Lanka 4. Colombo, Sri Lanka: International Irrigation Man-agement Institute.

Steele, P., F. Konradsen, and K. A. U. S. Imbulana. 1997. Irrigation, health, and the environment: A literature reviewwith examples from Sri Lanka. Discussion Paper No. 42. Colombo, Sri Lanka: International Irrigation ManagementInstitute.

Steins, N. A., and V. M. Edwards. 1998. Platforms for collective action in multiple-use CPRs. Paper presented at the7th Conference of the International Association for the Study of Common Property, June 10-14 1998, Vancouver.

48

Tanji, K. K. (ed). 1996. Agricultural salinity assessment and management. ASCE manuals and reports on engineer-ing practice No.71. New York: American Society of Civil Engineers.

Upasena, J., and P. Abeygunawardena. 1993. Economic value of irrigation water in Dewahuwa irrigation scheme.In Tropical agricultural research. Vol. 5, 1993: Proceedings of the 5th Annual Congress of the Post Graduate Insti-tute of Agriculture, 4-5 November 1993, ed, A. A. Jayasekera. Peradeniya, Sri Lanka: Post Graduate Institute ofAgriculture.

Whittington, D., and V. Swarna. 1994. The economic benefits of potable water supply projects to households in de-veloping countries. Economic Staff Paper No. 53, Manila: The Asian Development Bank.

WHO (World Health Organization). 1993. Guidelines for drinking water quality. Volume 1: Recommendations. Sec-ond edition. Geneva, Switzerland: WHO.

WHO. 1984. Guidelines for drinking water quality. Volume 2: Health criteria and other supporting information. Geneva,Switzerland: WHO.

Yoder, R. 1983. Non-agricultural uses of irrigation systems: Past experiences and implications for planning and de-sign. London: Overseas Development Institute.

Young, R. A. 1996. Measuring economic benefits for water investments and policies. World Bank Technical PaperNo.338. Washington, D.C.: The World Bank.

Zwarteveen, M. 1994. Gender issues, water issues. Working Paper No. 32. Colombo, Sri Lanka: International Irriga-tion Management Institute.

SWIM Papers

1. Accounting for Water Use and Productivity. David Molden, 1997.

2. How to Manage Salinity in Irrigated Lands: A Selective Review with ParticularReference to Irrigation in Developing Countries. Jacob W. Kijne, S. A. Prathapar, M.C. S. Wopereis, and K. L. Sahrawat, 1998.

3. Water-Resource and Land-Use Issues. I. R. Calder, 1998.

4. Improving Water Utilization from a Catchment Perspective. Charles Batchelor, JeremyCain, Frank Farquharson, and John Roberts, 1998.

5. Producing More Rice with Less Water from Irrigated Systems. L. C. Guerra,S. I. Bhuiyan, T. P. Tuong, and R. Barker, 1998.

6. Modeling Water Resources Management at the Basin Level: Review and FutureDirections. Daene C. McKinney, Ximing Cai, Mark W. Rosegrant, Claudia Ringler,and Christopher A. Scott, 1999.

7. Water Harvesting and Supplemental Irrigation for Improved Water Use Efficiency inDry Areas. Theib Oweis, Ahmed Hachum, and Jacob Kijne, 1999.

8. Multiple Uses of Water in Irrigated Areas: A Case Study from Sri Lanka. MargarethaBakker, Randolph Barker, Ruth Meinzen-Dick, Flemming Konradsen, 1999.

SWIM

SWIM Paper

System-Wide Initiative on Water ManagementSWIM

8

INTERNATIONAL WATER MANAGEMENT INSTITUTEP O Box 2075 Colombo, Sri Lanka

Tel (94-1) 867404 • Fax (94-1) 866854 • E-mail iwmi@cgiar.orgInternet Home Page http:/ /www.cgiar.org/iwmi

ISBN 92-9090-380-5

CGIAR

ISSN: 1028-6705

Multiple Uses of Water inIrrigated Areas: A Case Studyfrom Sri Lanka

Editors:Margaretha BakkerRandolph BarkerRuth Meinzen-DickFlemming Konradsen

IFPRI