Assessment of Bulk Water Allocation Programme in Mahaweli – H Area

M.M.M. Aheeyar W.H.A. Shantha L.P. Senevirathne

Research Report No: 118 July 2007

Hector Kobbekaduwa Agrarian Research and Training Institute 114, Wijerama Mawatha Colombo 07 Sri Lanka

First Published: 2007

ISBN: 978-955-612-0745

Suggested citation: Aheeyar, M.M.M., Shantha, W.H.A, and Senevirathne, L.P. 2007. Assessment of Bulk Water Allocation Programme in Mahaweli – H Area. HARTI Research Report No: 118, Hector Kobbekaduwa Agrarian Research and Training Institute, Colombo, Sri Lanka. FOREWORD

Competition and conflicts of allocating scarce water resources within and between the sectors are evident in various parts of the country. Increasing water demand for agriculture cannot be met simply by using more water for irrigation. However, the amount of water for agriculture could be used much more efficiently to boost food productivity.

Government of Sri Lanka (GOSL) introduced various projects and programmes to improve the efficiency of irrigation systems and enhance the water management practices through rehabilitation and restorations of irrigation infrastructure and introducing different water management practices with different degrees of successes. Sri Lanka as one of the pioneers of participatory irrigation management (PIM) has been implementing PIM as a national policy for more than two decades. Bulk Water Allocation (BWA) programme is an extended version of PIM which was implemented as a pilot project in Mahaweli System H since 2002. The present study is a timely effort to assess the relevance, effectiveness and efficiency of this programme.

The report covers many aspects related to irrigation and agricultural production which are useful to other researchers, academics and policy makers. I wish to highlight one of the key recommendations made in the report that it is necessary for policy makers and the implementers to focus on the irrigated agricultural development in a broader perspective just beyond the mere cost recovery. It is vital to formulate policies and programmes and create conducive environment that will generate necessary development impulses to increase agricultural productivity, establish market linkages, focus on scale of production and enhance farmers’ income.

According to the findings of the report, irrigation system performances have improved in terms of enhanced performance in water distribution, greater agricultural productivity and better water productivity are some of the remarkable features of BWA as a water management package.

As it has been noted in this report, further studies on technical, institutional and socio- economic features that will determine the feasibility of replication of the BWA concept in other major irrigation schemes will be useful for the irrigated agricultural development of the country.

V.K. Nanayakkara Director

ACKNOWLEDGEMENTS

The research team is very much thankful to the Sri Lanka Council for Agricultural Research Policy (CARP) for providing a research grant to conduct this study. Our sincere thanks are due to Professor H.P.M. Gunasena, Executive Director, CARP and his staff for providing necessary guidance and support for the project and monitoring of the project throughout the study. We are very much grateful to Mr. Jayathissa Bandaragoda, former Researcher, International Water Management Institute for his valuable comments provided in an earlier draft of this report. We also thank the members of the CARP project evaluation committee namely, Dr. Nihal Samarappuly and Dr. Frank Niranjan for their valuable suggestions.

We are grateful to Mr. D.G.P Seneviratne, the former Director of HARTI for his constructive comments and suggestions provided in the early stage of the proposal development. The research team is much thankful to Mr. V.K. Nanayakkara, the present Director of HARTI for his keen interest shown in completing this study on time and the encouragement and the administrative support provided in publishing this report. Thanks are also due to Dr. W.G. Jayasena, Deputy Director (Research), HARTI and Dr. W.M.G.B. Giragama, Head, Irrigation, Water Management and Agrarian Relations Division, HARTI for their support extended throughout the study.

We appreciate the excellent services provided by Mr. P.C.J de Silva, Research Officer, HARTI at the early stage of the study, especially in co-ordinating the field survey and monitoring the data collection. The dedicated service provided by Mr. Nalaka Wijesooriya, Statistical Officer, HARTI was beyond the expected level and it is very much appreciated. Mr. M.D.L. Senarath, Senior Analyst Programmer, HARTI delivered his services in analyzing the data and for timely preparation of the requested data output. Our special thanks also go to Ms. Sharmini Kusum Kumara, Research Associate, HARTI for her assistance in preparing the final report.

We are thankful to Dr. M.U.A. Tennakoon, the former Director General of the Mahaweli Authority of Sri Lanka (MASL) for his approval and other support extended to undertake this study as a collaborative research with the HARTI and the MASL. The support extended by the officials of the MASL attached to the Resident Project Managers’ office and the officials at the block offices in Madatugama and Galnewa is highly appreciated. The service provided by Ms. Shayama Ranasinghe of the HARTI in performing an excellent job of typesetting the report and designing the cover page is also valued.

Thanks are also due to Mr. H.A. Siriwardane, an outside copy editor, for providing expert editorial assistance. The services provided by Mr. S. Rameshwaran, Acting Head, Publication Unit, HARTI in preparing the final version of the report and

arrangements for printing of the report and printing staff of the HARTI for printing the report are highly appreciated. Finally, we wish to pay our tribute to the farming community in the Mahweli H area for their co-operation extended in collecting the necessary information for the study.

Authors

iii

EXECUTIVE SUMMARY

Allocation of funds for sustainable operation and maintenance (O&M) of irrigation systems has been decreasing over the years due to budgetary and fiscal constraints, although irrigation systems have been expanding and improving in Sri Lanka after gaining independence in 1948. Failure to make necessary policy changes to generate and allocate sufficient funds to properly operate and maintain the expanding public irrigation systems has been one of the main concerns of policy makers to prevent the deterioration of irrigation infrastructure and to increase the efficiency of water use in irrigated agriculture which is vital to meet the challenge of competing demand for water within and between different sectors.

Mahaweli H is the oldest system of the Mahaweli development programme where the scarcity of water is acute in comparison with the other systems in the Mahaweli development areas. Therefore, the water management is crucially important in the Mahaweli H to have a successful cultivation especially during the dry seasons (yala seasons). The government of Sri Lanka introduced the participatory irrigation management (PIM) policy in late 1980s, the achievement of which is limited to a few aspects such as soliciting farmer participation for irrigation system maintenance via mobilizing their labour and government cost saving.

Bulk Water Allocation (BWA) is primarily water and irrigation system management package initiated in year 2002 in Mahaweli H areas, for provision of water right to the farmers at a pre-agreed volume for each season, while farmers have to accomplish the responsibility of the O&M of the secondary and tertiary canal system. Farmers also have to pay an O&M fee of Rs 250 per ha per season to distributory canal organization, in addition to providing their labour for maintenance works. The BWA emphasises the active farmer participation and contribution through an established institutional mechanism, while providing volumetric impression of water use to the farmers.

The primary objective of the study is to find out the relevance, effectiveness and efficiency of the BWA concept in Sri Lanka with special focus on assessing the impacts of the programme on performance of water distribution and agricultural production. The study also seeks to estimate the level of resources mobilized for the O&M under the BWA and its implications on the sustainability of irrigation system and assess the pros and cons of the programme.

The findings of the study are based on information and data collected from review of literature, group discussions, key informant interviews and a structured questionnaire survey. Sample survey covered 120 farmers randomly selected from four DCO's from Madatugam and Galnewa blocks. Descriptive and statistical approaches were undertaken to analyze the collected data. A regression model was adopted to examine the relationship with various socio-economic features of farmers on their willingness to pay for improved irrigation system maintenance.

The average family size of the study area varies from 3.6 to 4.4. About a quarter of the population is in the dependency age category. Farming is the main source of income for 66 percent of the total employed labour force. About 96 percent of the farmers are literate. Average monthly income from farming is about Rs 9,500 and that from non farm sources is about Rs 11,500.

The analysis of secondary data clearly indicates that there is a decrease in gross water quota or seasonal tank duty and an increase in the extent of land cultivated per MCM of irrigation water after the BWA. The saved water has been effectively used to cultivate more extent of land especially during the yala seasons as indicated by the increase in extent of cultivation in the dry seasons and the increasing trend of cropping intensity. The impact of the increase of water supply performance is also reflected in agricultural productivity and the farmers’ income. The significant increase shown in the yala season paddy yield and the shift to crop diversification in yala seasons have resulted in an increase in farmers’ income. Gini ratio of head and tail end farmers, estimated for agricultural income distribution is 0.12 and 0.22 respectively suggesting a lesser income inequality compared to that of other major irrigation systems. Land productivity values estimated for the past maha seasons (wet seasons) mark a stagnated condition, while those for water productivity show an increasing trend after the implementation of the BWA.

Farmer perceptions obtained on the impacts of the BWA also justify the findings derived from secondary data. A large proportion of farmers perceived that, crop yield, profitability and income from agriculture have increased with the BWA. About 95 percent of the farmers observed that, they always or usually received their fair share of water quota, which helped them to plan their cropping pattern and other cultural practices in advance without thinking about the risks of an unreliable water supply.

The major advantages of the BWA constitute the reduced number of water conflicts among the farmers, change of farmer attitudes on water use and irrigation management, reduced incidences of crop failure and the increase in extent of cultivation. The main constraints experienced by the farmers with the implementation of the BWA are, inconvenience of receiving night time water issues, longer gap between water rotations at the end of the seasons and the inability to cultivate the crops, the farmers prefer.

The annual estimated value of farmers' contributions to the O&M ranged between Rs 1,000 – 1,775 per ha in kind and Rs 500 per ha, in cash. The value of cash collected is a notable feature in resource mobilization under the BWA, in relation to various failed attempts of cash collection for the O&M in the past. The total irrigation cost for the farmers is Rs 1,450 to 2,275 per ha per annum at 2005 prices.

Majority of the farmers regularly participate and contribute to routine O&M works. The farmers are satisfied with the level of investment made for maintenance and the way the maintenance works are conducted by the DCO. A large number of the farmers are of the opinion that there is a drop in the broken gates, broken pipe outlets and cases of silted canals, after the implementation of the BWA.

The willingness to pay (WTP) on the part of the farmers for improved irrigation management and sustainable infrastructure, in addition to the provision of current labour and the O&M fee was estimated using contingent valuation technique. The estimated amount is Rs 599 – 890 per ha per annum (52 – 77 kg of paddy/ha/annum) at 2005 prices. The WTP value has a significant and positive correlation with the total family income, the size of lowland and the selection of OFC and being at the tail end DCs. The WTP has no significant relationship with level of education, type of farming (fulltime/part time) and contribution of labour hours for maintenance works.

The approach and methodology adopted for the implementation of the BWA programme in Mahaweli H area is a good learning experience for rest of the major irrigation schemes, which are experiencing water allocation problems, despite having suitable infrastructure to practice the BWA. The major lesson of experience is that a comprehensive devolution policy for the irrigation sector and an assured water supply can encourage the farmers to invest for the long term sustainability of their irrigation infrastructure. The devolution policy must pay attention to create local organizations which are self-reliant socially, financially and technically to deal with the local management. The WTP expressed by the beneficiaries should be taken into consideration in future water policy formulation to ensure sustainability of the system, while allowing for future maintenance cost escalations and for averting the increasing trend of system deterioration.

CONTENTS

FOREWORD ...... i ACKNOWLEDGEMENTS ...... ii EXECUTIVE SUMMARY ...... iv CONTENTS ...... vii LIST OF TABLES ...... ix LIST OF FIGURES ...... x LIST OF ABBREVIATIONS ...... xi GLOSSARY ...... xii

CHAPTER ONE ...... 1 Introduction ...... 1 1.1 Background ...... 1 1.2 Statement of the Research Problem ...... 3 1.3 Objectives of the Study ...... 4

CHAPTER TWO ...... 5 Research Methodology ...... 5 2.1 Conceptual Framework ...... 5 2.1.1 Key Concepts ...... 5 2.1.2 Analytical Framework ...... 7 2.2 Selection of Study Sites ...... 10 2.3 Description of Study Areas ...... 10 2.4 Data Collection Instruments ...... 12 2.5 Sampling Design and Procedures ...... 13 2.6 Organization of the Report ...... 14

CHAPTER THREE ...... 15 Socio-Economic and Institutional Features ...... 15 3.1 Demographic Characteristics of the Sample ...... 15 3.2 Labour Force and Income Earning Sources ...... 16 3.2.1 Labour Force ...... 16 3.2.2 Income Earning Sources ...... 16 3.2.3 Household Income...... 16 3.3 Land Ownership and Land Use ...... 18 3.4 Cropping Pattern ...... 18 3.5 Land and Water Issues ...... 22 3.6 Institutional Mechanism Adopted for BWA ...... 23 3.7 Sustainability of the Institutional Mechanism ...... 26

CHAPTER FOUR ...... 31 Irrigation System Performance under ...... 31 Bulk Water Allocation System ...... 31 4.1 Water Supply Performance ...... 31 4.1.1 Gross Water Quota/Seasonal Tank Duty ...... 31

vii

4.1.2 Extent of Land Cultivated in Relation to Water Supply...... 33 4.1.3 Percentage of Land Extent Cultivated in Dry Season (Yala)...... 34 4.1.4 Cropping Intensity ...... 35 4.1.5 Relative Irrigation Supply...... 36 4.2 Performance of Productivity and Farmers Income ...... 37 4.2.1 Changes in Crop Yield ...... 37 4.2.2 Changes in Cropping Pattern ...... 38 4.2.3 Agricultural Income Inequality ...... 40 4.2.4 Land and Water Productivity ...... 41 4.2.5 Farmer Performance Index ...... 42 4.3 Farmers’ Perceptions on Irrigation Systems Performance Realized after BWA ...... 44 4.3.1 Benefits of Bulk Water Allocation System ...... 45 4.3.2 Constraints of Bulk Water Allocation ...... 46

CHAPTER FIVE ...... 47 Role of Bulk Water Allocation in Sustainable Maintenance of Irrigation Infrastructure ...... 47 5.1 Level of Resource Mobilization for Operation and Maintenance ...... 47 5.2 Farmers Investment on Irrigation Infrastructure and the Sustainability ...... 48 5.3 Willingness to Pay for Operation and Maintenance ...... 53 5.3.1 Factors Affecting Willingness to Pay for Operation and Maintenance ...... 54

CHAPTER SIX ...... 57 Findings, Recommendations and Policy Implications ...... 57 6.1 Major Findings ...... 57 6.2 Recommendations and Policy Implications ...... 59

REFERENCES ...... 61

viii

LIST OF TABLES

Table 2.1: Study Sites Selected for the Field Survey ...... 10 Table 3.1: Monthly Household Income Range from Farming ...... 17 Table 3.2: Monthly Household Income from Non-farm Activities ...... 17 Table 3.3: Type of Land Operation (Land Tenure) ...... 18 Table 3.4: Cropping Pattern in Maha Season 2004/2005 ...... 19 Table 3.5: Cropping Pattern in Yala Season 2005 ...... 20 Table 3.6: Crops Newly Introduced after the Initiation of BWA System...... 21 Table 3.7: Increase of Cultivated Land Extent after the BWA ...... 21 Table 3.8: Alternative Irrigation Sources ...... 22 Table 3.9: Reasons for not Practicing Depaweli System ...... 23 Table 3.10: Satisfaction about the FCR’s Performances/Duties ...... 27 Table 3.11: Satisfaction about Jalapalaka’s Activities ...... 27 Table 3.12: Regular Participation for DCO Meetings ...... 28 Table 3.13: Degree of Support from Farmers to the President of the DCO...... 28 Table 3.14: Farmers' Satisfaction on Financial Handling of DCO ...... 29 Table 4.1: Relative Irrigation Supply...... 36 Table 4.2: Adoption and Omission of OFCs after BWA ...... 40 Table 4.3: Summary of Selected Performance Parameters ...... 42 Table 4.4: Impact of BWA on Profitability and Income of Agriculture ...... 44 Table 4.5: Impact of BWA on Crop Yield ...... 44 Table 4.6: Benefits of Bulk Water Allocation System as Perceived by Farmers .... 45 Table 4.7: Constraints of Bulk Water Allocation System as Experienced by Farmers ...... 46 Table 5.1: The Estimated Value of the Mobilized Labour by Farmers for Maintenance – Year 2005 ...... 48 Table 5.2: Estimated Value of Cash Mobilization and Total Value of Mobilized Resources for O&M ...... 48 Table 5.3: Degree of Labour Mobilization for O&M ...... 49 Table 5.4: Degree of Participation for Meetings as Perceived by the Farmers ...... 50 Table 5.5: Supports Given by Officials for O&M and DCO Activities ...... 52 Table 5.6: O&M beyond DC Level by the Line Agency as Perceived by the Farmers ...... 53 Table 5.7: WTP for System O&M (kg of paddy/ha/year) ...... 54 Table 5.8: WTP for System O&M (Rs/ha/year) ** ...... 54 Table 5.9: Estimation of the Regression Model ...... 55

LIST OF FIGURES

Figure 1.1: Cropping Intensity of Mahaweli System ...... 2 Figure 2.1: Lorenz Curve for Gini Co-efficient ...... 8 Figure 2.2: Mahaweli System H ...... 12 Figure 3.1: Key Elements of the Farmer Based Institutional Setup ...... 25 Figure 4.1: Average Water Duty in Past Yala Seasons and Rainfall Pattern ...... 32 Figure 4.2: Average Water Duty in Past Maha Seasons and Rainfall Pattern ...... 32 Figure 4.3: Average Water Duty in Past Maha Seasons in Main Canals...... 33 Figure 4.4: Average Water Duty and Rainfall Pattern in Past Yala Seasons in Main Canals ...... 33 Figure 4.5: Extent of Land Cultivated per MCM of Water in Mahaweli H (Maha Seasons) ...... 34 Figure 4.6: Extent of Land Cultivated per MCM of Water in Mahaweli H (Yala Seasons) ...... 34 Figure 4.7: Percentage of Land Extent Cultivated in Past Yala Seasons ...... 35 Figure 4.8: Changes in Cropping Intensity in Mahaweli H area ...... 36 Figure 4.9: Changes in Paddy Yield in Mahaweli H – Yala Seasons ...... 38 Figure 4.10: Changes in Paddy Yield in Mahaweli H – Maha Seasons ...... 38 Figure 4.11: Trend of OFC Cultivation – Galnewa block ...... 39 Figure 4.12: Trend of OFC Cultivation – Madatugama block ...... 40 Figure 4.13: Lorenz Curve of Income Inequality ...... 41 Figure 4.14: Changes in Land and Water Productivity ...... 42 Figure 4.15: Farmer Performance Index for Head and Tail Areas (Maha Season-2004/05) ...... 43 Figure 4.16: Farmer Performance Index for Head and Tail Areas (Yala Season-2005) ...... 43 Figure 5.1: Structural Problems of DCs & FCs before and after BWA ...... 50 Figure 5.2: Is the Extent of Resource Mobilization for System O&M by DCOs Sufficient to Ensure Physical Sustainability of Infrastructure? (As a percentage of farmer perceptions) ...... 51 Figure 5.3: Is the Quality of Maintenance Works Conducted by DCOs is Comparatively Higher than the Pre-turnover Maintenance? (As a percentage of farmer perceptions) ...... 51 Figure 5.4: Percentage of DC Canal Properly Maintained after BWA as Perceived by Beneficiary Farmers ...... 52

LIST OF ABBREVIATIONS

BWA Bulk water allocation BLFF Block level federation farmer organization CI Cropping intensity CVM Contingent valuation method DC Distributory canal DCO Distributory canal organizations FC Field canal FCG Field canal group GOSL Government of Sri Lanka IMT Irrigation management turnover JMC Joint management committee LHG Low Humic Gley MASL Mahaweli Authority of Sri Lanka MCM Million cubic meters MDP Mahaweli Development Programme MRRP Mahaweli Restructuring and Rehabilitation Project OFC Other field crops O&M Operation and maintenance PIM Participatory irrigation management RBE Reddish Brown Earth RIS Relative irrigation supply RPM Resident Project Manager WTP Willingness to pay

GLOSSARY

Depaweli Type of an on-farm ditch system Jalapalaka An individual (often himself is a cultivator) responsible of operation of DC and FC gates and of ensuring fair and timely distribution of irrigation schedule under the DC command area Liyadda The smallest bunded unit within a paddy field (Sub field) Maha Cultivation season which normally extends from September/October to February/March and coincide with the North-east monsoon. Paddy Un-husked rice Yala Cultivation season which normally extends from April to August and coincide with the South-west monsoon.

xii

CHAPTER ONE

Introduction

1.1 Background

The multipurpose Mahaweli development programme (MDP) has been the country's biggest investment in the recent past. The project has resulted in the development of 390,000 acres of new land and improved irrigation for double cropping of about 128,000 acres of existing paddy land coming within the development regions. The Mahaweli irrigation development and settlement regions are described as systems A, B, C, D and H.

Each system is divided into administrative blocks, covering about 2,000 ha and having 2,000 families. Each farmer family was provided 1 ha of lowland and ½ acre of upland for homestead. Each block is further divided into units of 200-265 ha each and having 200-250 families. There are about 125,000 settler families in the downstream areas. The Mahaweli areas accounted for about 22 percent of the total national paddy production, 31 percent of big onion and 13 percent of chillie production in the year 2005 (MASL, 2005). The import substitution value of the total crops produced in Mahaweli areas is Rs 6,509 million (MASL, 2000).

Mahaweli H area is the first of the downstream areas benefited by Mahaweli diversion. The system H has the longest history of settlement in the country and had a relatively dense occupation before the Mahaweli water was diverted. After the Mahaweli diversion, additional water was supplied to Mahaweli H area via the Kala Oya basin. This water is diverted into three main storage tanks in the H area - Kandalama, Dambulla Oya and Kalawewa. However, the limited capacity of these reservoirs in comparison with the large extent of irrigable land makes cultivation in the system H vulnerable (Raby & Merry, 1989). In addition to this constraint, Wickramaratne (1986) listed two other important factors limiting the performance of the Mahaweli system H. These are: i) limited capacity of diversion to the system H through Bowatenna tunnel due to mechanical reasons; ii) The system H contains about 60 percent of irrigable land with well-drained Reddish Brown Earth (RBE) soils, which require more water for the cultivation of low land rice.

The total irrigation extent in the system H is around 31,500 ha allocated among equal number of farmers at the rate of one ha per farmer. The cropping pattern in the system H is generally the paddy crop for the entire extent during the maha and paddy and Other Field Crops (OFC) for 50 percent of the total extent during the yala season. Therefore annual cropping intensity in the system H is the lowest in the Mahaweli development area compared to that of system B, K and G. (Figure 1.1) Therefore water management is crucially important in the Mahaweli system H for the successful cultivation especially during the yala season. At the same time, inadequate financial allocation for the operation and maintenance (O&M) due to fiscal constraints lead to the poor performance of irrigation systems and premature rehabilitations, which

aggravate the problem of water scarcity and proper management of the limited available water.

The Mahaweli Authority of Sri Lanka (MASL) implemented various special water management packages in the Mahaweli H area from time to time to meet the challenges of water scarcity and to improve the water use efficiency. Strict management practices had to be followed during drought stricken water crisis seasons. One such management system has been documented by Raby and Merrey (1989).

Figure 1.1: Cropping Intensity of Mahaweli System

230

210

190

170 B 150 C 130 G Cropping intensity H 110

90 95 96 97 98 99 00Year01 02 03 04 05 06

The concept of “Bulk Water Allocation” (BWA) was introduced and pilot-tested by the Mahaweli Restructuring and Rehabilitation Project (MRRP) in the Mahaweli system H to find out a methodology, which can be used as a broad-based solution for water management problems in the major irrigation schemes (Gunaratna, 2004). Under the BWA, the quantity of water to be issued for a particular user and consequently a distributory canal for cultivation practices in a given season is fixed before the commencement of the season. Therefore, a particular user has a legal water right and has to participate and contribute for water management. A maintenance fund has been set up at the DC level with farmer contribution. Under this initiative implemented in the Mahaweli System, it was expected to develop the Mahaweli H as a ‘model demonstration system’ in terms of both irrigation management and commercialized agricultural production.

The overall concept of the BWA was seen as a strategy to achieve the expected objectives of the irrigation management turnover (IMT) project and increase the water use efficiency and water productivity. The project started in the 2001 yala season on a pilot basis in selected locations of system H. Later, it was expanded to cover the entire system H with some improvements made in subsequent seasons. The other main

features of the BWA is that the farmers have to pay an O&M fee of Rs 250 per ha per season to the respective Water User Association for the sustainable O&M of the turnover distibutory system. The farmers have to manage the crop with the agreed quota of water and the MASL has the responsibility of supplying the promised quantity of water. The failure to supply the agreed quota and any consequent crop failure have to be compensated for by the MASL. Any demand for additional quantity of water has to be met at the rate of Rs 1,000 per acre-foot of water.

1.2 Statement of the Research Problem

Historically the Government of Sri Lanka (GOSL) has made various attempts in the past to levy a tax linked either to the use of irrigation services or the benefits received from the irrigation facilities through indirect taxes. The land tax and the implicit product tax are the two main tools used as indirect financing in Sri Lanka.

The user fee of Rs 30 per acre was implemented in major irrigation schemes with over 150 cropping intensity and Rs 20 per acre in the scheme with lesser intensity. The policy was actually implemented for only a short period from 1981-1983 in major schemes, but the collections were minimal (Silva, 1986).

In 1984, the government introduced the direct O&M fee collection at the rate of Rs 250 per ha of paddy land from the farmers promising improved irrigation services. The collected money was credited to the government treasury. The rate of collection had a good beginning with 85 percent of the amount due, but declined sharply during the subsequent years, registering less than 10 percent by 1985.

The GOSL introduced participatory irrigation management (PIM) as a national policy in late 1980s. The PIM adopted an approach of joint management, requiring farmer participation with the irrigation agency in the system management. The farmers are responsible for the O&M of the DCs and the FCs and need to mobilize necessary labour, cash and kind for the sustainable maintenance of turnover responsibilities. The farmers were exempted from the previous O&M fee. The government retains the responsibility of the head works and main systems and for the major and emergency repairs of turnover distributory systems. Nevertheless, a decade of experience in PIM suggests that, Farmer Organizations (FOs) have failed to mobilize adequate resources toward the O&M and some of the maintenance responsibilities have become 'no body's' business (Aheeyar, 1997). The situation has led not only to poor irrigation performances, but also to the quick deterioration of irrigation infrastructure before the expected life period and to the need for pre-mature rehabilitation of the entire scheme.

Considering various unsuccessful cost recovery attempts and the IMT experiences in the past, it is essential to assess the latest BWA concept, which has not only turned over the DCs and FCs and warrants mobilization of sufficient amount of cash, materials and labour for sustainable O&M, but also have strategies to manage water deliveries efficiency. However, there are no documentation which assesses the impacts of BWA and the problems and prospects of the programme. The documentation of the lessons of experiences is a useful input for proposed water

management policy formulation in Sri Lanka. The lessons also can be effectively used for water management in the rest of the irrigation schemes in Sri Lanka helping to avoid wasteful allocation of water resources.

1.3 Objectives of the Study

The major objective of this study is to find out the relevance, effectiveness and efficiency of the BWA concept in Sri Lanka. The specific objectives are:

1. To study the institutional set up established for the implementation of the BWA concept. 2. To assess the impact of the BWA concept on water productivity, farmer income, equality of water distribution, cropping intensity and cropping patterns. 3. To examine the level of resource mobilization for the system O&M under the BWA programme and its implications on sustainability of infrastructure. 4. To find out the benefits/constraints in the adoption of the BWA concept in Sri Lanka. 5. To propose recommendations for policy formulation on the basis of experiences gained through the BWA.

CHAPTER TWO

Research Methodology

2.1 Conceptual Framework

2.1.1 Key Concepts

The conceptual framework of the study is based on some key concepts associated with the implementation of the BWA programme. These concepts are discussed in the following sub sections.

2.1.1.1 Participatory Irrigation Management

Participatory Irrigation Management (PIM) emphasizes the active participation of farmers with the line agency in the irrigation system management. Implementation of the PIM involves three requirements; creation of the Water Users’ Associations (Distributory Canal Organizations), creation of the Joint Management Committees (JMC), which is also called as Project Management Committee (PMC) and the irrigation management turnover (hand over of O&M responsibilities of distributory and field canals to DCOs). Therefore, under the PIM, the O&M of irrigation system, planning of irrigation management, and allocation and distribution of irrigation water are undertaken jointly by the beneficiaries (farmers) and the agency-Mahaweli Authority of Sri Lanka (MASL). Transfer of responsibilities of routine operation and maintenance of secondary and tertiary irrigation systems (Distributory and Field canal systems) to the DCOs is one of the main activities undertaken in the PIM policy.

Prior to the transfer of O&M responsibilities and the turn over of DCs and FCs to DCOs, the rehabilitation of irrigation systems had been completed under the MRRP. Attempts had been made to increase the awareness among the farmers and the office bearers of DCOs on the concept of participatory irrigation management, which was put in place through systematic training programmes. Meanwhile, selected farmers were provided a practical training on the use of water measuring gauges installed at the key points of the irrigation canal systems. Through the awareness building and technical training sessions, attitudes of farmers on the use and management of irrigation systems and water resources have been changed. The programme also aimed at developing a sense of ownership of assets (physical structures of irrigation systems and the water resource) and attitudinal changes towards a “self-responsible” mode of irrigation management and the O&M.

2.1.1.2 Distributory Canal Organizations

Distributory canal organization (DCO) is the key institution at the farmers’ level to undertake the PIM concept. Therefore, to achieve the envisaged objectives of this participatory irrigation management system, strengthening of the pre-existing DCOs and formation of a new institutional setup was undertaken prior to handing over the responsibilities of the O&M of secondary and tertiary irrigation system to the DCOs.

The DCOs are the legally registered organizations at farmers' level consisting of all the legitimate farmers in the DC command area as members. There are field canal level turnout groups (FCG) and Block Level Federated Farmer Organizations (BLFF) above the DCO level, which are not legally registered. Block level water allocation and management plans are developed by respective BLFF taking into consideration the quantity of water demanded by each DCO in relevant block area.

The DCOs and the FCGs are also expected to handle the resolution of conflicts relevant to the O&M and other irrigation management activities at the DC&FC level. On the other hand, turnover of the O&M responsibilities to the DCOs (beneficiary farmers) was considered as a means of reducing public expenditure on the O&M and improving the sustainable maintenance of the irrigation system.

2.1.1.3 Design of Bulk Water Allocation System

The BWA is one of the major activities of the MRRP implemented in Mahaweli H. It was envisaged to improve agricultural productivity through rehabilitation, improvement and converting the farming activities into a more productive commercially oriented venture by the project. It was also anticipated to improve the O&M of the irrigation system in the system-H with more farmer participation, while reducing the dependency on government irrigation subsidies.

Under the BWA system, the DCOs are allocated specific quantity of irrigation water for each season. The specific quantity is referred to as water quota, which is decided on the basis of the total irrigated land within the DCO command area, type of crops to be cultivated and the quantity of water required per rotation to meet the crop water requirement. Each DCO requests the seasonal water requirement and the requirements for each rotation from the BLFF. The BLFF prepares the water requirement for the entire block and submits to the MASL via PMC. The Mahaweli water panel at MASL adjusts the BWA for each system considering the availability of water in the reservoirs. The Project Management Committee (the higher level joint management committee at block level consisting of DCO office bearers and line agency officials) decides the final water allocation for each block as prepared by the Technical Services Division of the Resident Project Manager’s office of the system H. The Block Irrigation Engineer together with the BLFF prepares the quota for each DCOs. The whole procedure is based on the water availability, but there is a guarantee for the allocated water quantity. This strategy has assured the right for water for each individual farmer.

The farmers in the relevant DCOs have to bear the cost of the O&M of turned over irrigation systems through a maintenance fund and they have to be involved in management and the O&M of their own irrigation system through mobilization of cash, labour and materials. In order to ensure the delivering and distribution of water among the farmers in a timely manner, a Water Master (Jalapalaka) has been appointed by the farmers of each DCO on payment of a salary for his services, from the O&M fund maintained by the DCO.

The efficient use of available water resource and equitable distribution has been one of the expectations of the BWA mechanism, which also envisages result better discipline in water management through active farmer participation, creating a sense of ownership and an attitudinal change in the farmer through fabricating volumetric impression of water use. In this process, an assured irrigation water supply was targeted resulting in the reduction of risks in crop cultivation and motivating the farmers to cultivate high value cash crops in the dry season. The target of this PIM is to develop a well functioning irrigation management mechanism via generating tangible benefits for the farmers to manage their own resources in a sustainable manner.

2.1.2 Analytical Framework

The data and information gathered from the structured sample survey and other primary and secondary data collecting methods were subjected to different techniques of analysis, ranging from appropriate quantitative and qualitative data analysis techniques.

The assessment of the impacts of the BWA in relation to the irrigation system refers to the measurement and analysis of certain indicators of the performance of the irrigation and agricultural production under the programme. The performance assessment was measured using both qualitative and quantitative methods. Qualitative assessments were used to appraise the farmer perception of the changes in selected indicators before and after the implementation of the BWA such as quality of maintenance, delivery of fair share of water, canal condition, changes in agricultural production and income, organizational management performance and the pros and cons of the BWA system. The quantitative analysis is based on the performance indicators described by Molden et-al (1998) linking the performance of irrigation system to the financial, agricultural, hydrological and economic aspects. The main quantitative parameters used for the assessment are: i) Cropping intensity = Areacultivated in thedry season + Area cultivated in the wet season *100 Cultivable area

ii) Gross water quota for the season per unit area (m) = Gross quantity of water for the season(m3 )

Actual irrigation area(ha)

Actual irrigationsupply iii) Relative irrigation supply = Planned irrigationsupply

The relative irrigation supply provides some insight of the water scarcity or abundance and an indication on how accurately water was delivered with reference to the planned water schedule.

iv) Yield of the major crop

It was expected to obtain increased yield with improved irrigation services via adequate supply of water on time and reliable manner.

v) Percentage of the land area cultivated in the dry season

Cultivation of full extent during the yala seasons was one of the expectations of the BWA. Therefore, examining the changes in the cultivated extent in the yala season is a good indicator to assess the performance of the BWA system.

vi) The changes in the cropping pattern (extent under different crops)

The BWA encourages the farmers to shift toward less water consuming crops in order to save water and increase the extent of cultivation. The changes in the extent under different OFCs provide the pattern of diversification over the year.

vii) The cost of irrigation to the farmers (cash cost +labour)

viii) The income inequality using Gini co-efficient

Gini Co-efficient was used to assess the distribution of agricultural income pattern within the DCO and between the head and tail areas of the Mahaweli H system. Gini co-efficient (G) is a summary statistic of the Lorenz curve and a measure of inequality in a population. G is defined as a ratio of the areas on the Lorenz curve diagram (figure 2.1). If the area between the line of perfect equality and Lorenz curve is A, and the area under the Lorenz curve is B, then the G is A/(A+B) (Morgan, 1962).

Figure 2.1: Lorenz Curve for Gini Co-efficient

ix) Land and water productivity

Land productivity is defined as value of output obtained from a unit of cultivated area, while water productivity is the value of output received from a unit of irrigation water supplied.

Total valueof production($) Land Productivity ($/ha) = Total cultivated area(ha) Total valueof production($) Water Productivity ($/m3) = Diverted irrigationsupply(m3)

Standard gross value of production (SVGP) is used to measure value of production. SVGP is standardized estimates of production calculated using international price of rice (major crop cultivated under irrigated condition).

Pi SVGP = ( ∑ AiYi )Pw Pb Ai - Area of crop i Yi - Yield of crop i Pi - Price of crop i Pb - Price of base crop (rice) Pw - Price of base crop traded at world price

x) Willingness to Pay for the O&M

Contingent valuation technique was adopted to estimate the farmers’ willingness to pay for improved irrigation services. The contingent valuation method (CVM) is a survey technique that elicits values people provide on the increase or decrease of benefits from natural resources at hypothetical market condition. The method basically depends on setting up a contingent market which is close to those that would be revealed if an actual market did exist. There is substantial literature to explain the theoretical and empirical application of CVM (Hufschmidt et-at, 1990; Whittington, D. et-al, 1990, Randall, 1981; Upawansa & Abeygunawardana, 1993).

The questionnaire was designed as a measure to elicit willingness to pay (WTP) for sustainable operation and maintenance (O&M) of the system in order to provide improved irrigation services and also to prevent deterioration of infrastructure.

The factors affecting the amount of WTP and relationship between different variables were established developing a multiple regression model. The following ten independent variables were identified and information was collected from sample farmers.

1. Total farming income (Inc) – Rs/year 2. Total low land extent (low land) – ha 3. Major yala crop (crop) – OFC = 0, Paddy = 1 4. Age (age) – Years

5. Education (Edu) – No. of years 6. DC location (DCloc) – Head = 0, Tail= 1 7. Type of farming (Farm) – Full time = 0, Part time = 1 8. Labour contribution for meetings (Labmeet) – No. of days/season 9. Labour contribution for maintenance (Labmain) – No. of days/season

The willingness to pay (WTP) estimate was regressed as a function of listed nine variables.

WTP = β0 + (β1 Inc + β2 Lowland + β3 Crop + β4 Age + β5 Edu + β6 DCIoc + β7 Farm + β8 Labmeet + β9 + Labmaint)

2.2 Selection of Study Sites

The study site of this research was the Mahaweli System H where the BWA mechanism based on participatory irrigation management system is practiced. Further, the Mahaweli System H is the first and the largest area developed under the MDP introducing and implementing various programmes in terms of irrigation, resettlement and agricultural development. Usually the lessons from the programme initiated in the System H have been used to formulate development plans for the rest of the Mahaweli areas. The Mahaweli system H has been considered as a ‘Model Demonstration System’ in implementing MRRP.

Two Block areas called Madatugama and Galnewa representing the Kandalama tank and the Kalawewa tanks were selected respectively for the study. Four DCOs (two DCOs from each block) were randomly selected to undertake the in-depth field research activities. One DCO from a selected block area represents the head end of the branch canal, while the other DCO from the same block represent the tail end of the particular branch canal. Thus, 101/D4 (head end) and 104D7 (tail end) DCOs from Madatugama block and 302/D1 (head end) and 306/D4 (tail end) from Galnewa block were randomly selected to administer the field survey and other research activities. Table 2.1 gives the relevant information of study sites.

Table 2.1: Study Sites Selected for the Field Survey

Block Area Irrigation Water Total No. of DCOs Selected Source DCOs in the Block Area Madatugama Kandalama Tank 51 101 /D4 (Head end) 104 /D7 (Tail end) Galnewa Kalawewa Tank 40 302 /D1 (Head end) 306 /D4 (Tail end)

2.3 Description of Study Areas

Mahaweli System H covers the Kala-oya river basin from Dambulla (Kandalama) to Nochchiyamaga. The area belongs to three administrative districts and the largest

portion lies in the Anuradhapura district, while only 3 block areas are in the Matale and Kurunegala districts.

The System H is headed by a Resident Project Manager (RPM) and he is assisted by nine block level managers. A block comprises 3-4 units and each unit is under the administration of a Unit Manager. In this study two DCOs selected in Galnewa block, 302/D1 and 306/D4 representing two unit areas called Aukana and Kalankuttiya.

The System H contains a total land area of 50,994 hectares and the total asweddumized land area was 31,956 hectares in the 1999/2000 maha season. The System H predominantly lies in the dry zone although a part of it is located in the Matale district, which represents the intermediate zone. Thus, DL1 (Dry Zone Low Country One) is the agro-ecological region where all the study locations are sited. The envisaged average annual rainfall in this area is over 775 mm (mostly 775-900 mm) and it mainly occurs in the North-east monsoon, during the period from September to February. According to the data available at Mahaillupallama meteorological station, the average annual rainfall during the last 10-year period up to 2003 was 1,450 mm and it varied from 1,018 mm to 1,897 mm (CBSL, 2004). The lowest occurrence of rainfall is during the period from June to August while the highest figures are recorded from October to December in an average year. More than two third of the rainfall is received in the period from October to December while the yala season receives about 445 mm on an average. The Mahailluppallama pan evaporation values give an annual average of 1,706 mm over the period from 1974 to 1996 (Mininstry of Irrigation and Water Management, 2003).

Reddish Brown Earths (RBE) and Low Humic Gley (LHG) are the main soil groups that are much favourable for tropical food crop cultivation. As envisaged in the Mahaweli Development Programme, the paddy cultivation is the main agricultural activity in the Mahaweli H area, especially in the maha season and cash crops and other field crops are prominent in the yala season, following the problems relating to irrigation water scarcity. As a cash crop, big onion is cultivated in the yala season especially in Madatugama and Galnewa block areas. In the maha seasons, the farmers tend to pursue paddy farming in their whole irrigated lowland areas. In the maha 1999/2000, paddy cultivation had been undertaken in over 31,254 ha accounting for 97.8 percent of total asweddumized land area of the System H (MASL, 2000). During the last five maha seasons since 2000/01 upto 2004/05 the average land area under paddy cultivation in the Mahaweli H area was 24,613 ha accounting for 77 percent of the total asweddumized land area.

Galnewa and Madatugama block areas were the main paddy cultivation areas in the Mahaweli H System in the 2004/05 maha season. Paddy cultivation had been undertaken in over 3,959 ha and 3,663 ha in Galnewa and Madatugama block areas respectively. The Mahaweli H System recorded the highest average paddy yield (5.75 t/ha) in the same season among all the Mahaweli systems, which was the highest national average yield as well. Cash crop cultivation under the contract farming system or forward agreements is also practiced in the study areas. Soybean, maize, gherkin and vegetables are the main crops cultivated under these agreements with the

outside private companies. In addition, in the year 2000 the Mahaweli System H recorded the highest paddy production compared with that in the other Mahaweli Systems. The Mahaweli H area has the largest extent of land under soybean in both the yala and the maha seasons in Sri Lanka. It is also the third highest cultivation area of big onions in the yala season and second only to the Matale and the Puttalam districts. Tobacco is widely cultivated in the yala seasons as a cash crop in the H area and only the Matale district has a larger extent under tobacco than the Mahaweli H.

The total number of families settled in the Mahaweli H system as at the end of year 2000 was 31,866 and it comprises 24,088 farm families and 7,778 non-farm families. The estimated population of the system during this period was 170,444 (MASL, 2000).

Figure 2.2 depicts the locations of this research study.

Figure 2.2: Mahaweli System H

2.4 Data Collection Instruments

The study was mainly based on primary and secondary data representing both qualitative and quantitative aspects, collected applying several data collection methods and approaches. Literature survey and data extracted from secondary sources especially from reports and records available at the RPM office and block

level offices of the MASL as well as the DCO offices were the main sources for the secondary data.

Available documented data and information pertaining to the BWA mechanism in terms of institutional strengthening, cropping intensity and change of cropping pattern, land productivity, water productivity, O&M of tertiary irrigation systems, crop yield, water duty etc. was derived from secondary sources.

Different types of data collection tools were used to gather primary data required for this study. Field visits of the research team to the study areas were made prior to formulate the formal data collection schedules. During these field visits, information gathered from informal discussions with the farmers, DCO office bearers, officials at management level and field level and direct field observations were much useful to have a proper perception and a through awareness about the BWA and the participatory irrigation management system in the Mahaweli System H. The main primary data collection instrument used was the field level sample survey. The sample survey was administered using a pre-tested comprehensive structured questionnaire with the direct participation and close supervision of the research team.

Key informant discussions with the Mahaweli Officers [Block Managers, Irrigation Engineers, Agricultural Officers, Deputy Resident Project Manager (technical services)] and DCO office bearers, office bearers of federations, Water Controllers (Jalapalaka) as well as selected prominent farmers were of immense value to obtain more in-depth information about the BWA mechanism and constraints and bottlenecks faced by both the farming community and the authorities. Focus group discussions conducted with selected farmers also provided some insight of the BWA programme.

The sample survey was conducted during the yala 2005, but information on both the yala 2005 and the maha 2004/05 was also collected.

2.5 Sampling Design and Procedures

Multiple stage stratified random sampling technique/ design was used to select a total of 120 farmers. The membership lists maintained by the selected DCOs were used as the sample frame.

At the initial stage, DCOs were selected to represent the head and tail areas of the two selected blocks. Each DCO consists of a number of field canals and the number varies among DCOs’. Generally, each and every DCO comprised over 30 FCs. The number of farmers in a field canal system also differs among FCs and the range is 8-20 farmers. As the second step of the randomization/ sampling procedure, each and every FC in the selected DCO was given a weight depending on the number of farmers attached to the particular FC. Therefore each FC in the respective DCO was given equal opportunity in selecting sample farmers to the ultimate sample, 30 farmers from a DCO. The weight given for a specific FC reflected the fact that the number of farmers had to be randomly selected from that FC.

In the last stage the relevant numbers of farmers (compatible with the weight) from each FC were randomly selected.

2.6 Organization of the Report

The report is presented in six chapters. Chapter one, as the Introduction, mainly consists of the background of the BWA-based participatory irrigation management system practiced in the Mahaweli System H, the problem statement and the objectives of the study. Chapter two deals with the research methodology, including the conceptual framework, selection of study areas, sampling design and analytical framework. Socio-economic and demographic characteristics of the study areas and the institutional features of the BWA mechanism are discussed in chapter three, whereas chapter four presents the most important findings and derivatives of the data and information on the impacts of the BWA. The role of the BWA in sustainable maintenance of infrastructure and the degree of strength of the current level of resource mobilization for irrigation system maintenance are presented in chapter five. Conclusions and recommendations for future policy directions are represented in the final chapter six.

CHAPTER THREE

Socio-Economic and Institutional Features

3.1 Demographic Characteristics of the Sample

Over 75 percent of the total households (120 households) surveyed are headed by males and the corresponding figure for female headed households is 24 percent (29 households). The total strength of the 120 families is 477 members representing 228 from the Madatugama block area and 249 from the Galnewa block area. Average family size of the sample households is 3.9 and it varies from 3.6 – 3.9 in Madatugama area and 3.9 to 4.4 in Galnewa area. Depending on the number of members in the families, sample families were categorized into three groups, namely families with 1-3 members, 4-5 members and more than 6 members. About 52 percent of the total families have 4-5 members, while families with 1-3 members and 6 or more members are 35 percent and 13 percent of the total families respectively. A well balanced gender distribution could be observed among the population with the composition of 50.1 percent of males (239) and 49.9 percent of females (238). Therefore, the masculinity ratio (number of males per 100 females) of the study area is almost 100.

Among the farmers, majority of them (about 60 percent) are between 40-60 years of age indicating the less involvement of youth in farming activities. Only 23 percent of the total farmers represent the youths between 15-40 years of age. However, 40 percent of the total population is youths in the age category of 10-30 years and it is the second largest age category in the population. These findings reveal that a large segment of the population is in the employment seeking category. Population below 14 years of age and above 65 years of age account for about 19 percent. Of the total population, 15 percent are over 55 years. The age category between 40-55 years represents 26 percent to the total population. Thus the age dependency ratio of the study area is nearly 24 percent

age group (10 −14) + age group 65 & more Age dependency ratio= age aroup (15 − 64 years)

According to the survey data, majority of the total population (58 percent) have received formal education up to G.C.E. Ordinary Level, but only 29 percent have been successful at the G.C.E. (O/L). The percentage with G.C.E. (A/L) qualifications is limited to 15 percent, while only 2 percent of the total population is graduates. 19 percent of the population has received only primary education and nearly 4 percent are recorded as illiterate.

The statistics on the education of the respondent farmers show that the majority of them (32 percent) have received formal education up to G.C.E (O/L). Of them, 27 percent have received primary education, while 7 percent have not received any kind

of formal education. Among the respondent farmers, university education is negligible (0.8 percent) and even the G.C.E. (A/L) qualified represent only 11 percent.

3.2 Labour Force and Income Earning Sources

3.2.1 Labour Force

In this study, males and females in the 15-64 year age group have been considered for the labour force. The total labour force of the study areas accounts for 83.4 percent of the total population- 52 percent of the males and the remaining 48 percent females (190 persons). It could be noticed that the gender composition of the labour force is also compatible to with that of the total population.

Over two third (68 percent) of the total labour force are employed. Employment seekers, school children, unpaid family workers, disabled, and those who follow various courses of studies are included in the unemployed category. Economic dependency ratio is somewhat higher in the study areas. The economic dependency ratio is around 66 percent in the area irrespective of study locations

[Economic dependency ratio= unemployed + housewives + students + discouraged * +others ** ] total employed * Those not looking for employment ** Retired, disabled, too young and too old

3.2.2 Income Earning Sources

15 percent of the total employed labour force are engaged in both on-farm and off- farm activities for their income, while the rest have only one source of income. Farming is pursued by 66 percent of the total employed labour force. About 61.5 percent of the total employed labour force is involved in farming as the main income earning source, while 4.5 percent is engaged in farming as a secondary source of income. The government sector is the second highest employment provider (14 percent) to the members of the sample families. Private sector employment (9 percent), self-employment (9 percent) and migration to foreign countries (3 percent) are the other means of income.

3.2.3 Household Income

Monthly average household income is calculated to include both on-farm and off- farm aspects. Income from farming as the main income source is in the range of Rs 1,000 to over Rs 20,000 per month. Household income was distinguished between head and tail ends of the branch canals in the selected block areas. Farm income of the farmers in head DCs (101/D4 and 302/D1) were compared with tail end DCs (104/D7 and 306 D4). Table 3.1 shows the monthly household income range from farming in both head end and tail end areas.

Table 3.1: Monthly Household Income Range from Farming

Income Range Head End Tail End Total (Rs.) No. % No. % No. % (N=60) (N=60) (N=120) < 1000 - - 1 1.7 1 0.8 1000 – 4999 9 15.0 9 15.0 18 15.0 5000 – 9999 21 35.0 22 36.7 43 35.8 10000 – 14999 20 33.3 12 20.0 32 26.7 15000 – 19999 7 16.7 7 11.6 14 11.7 > 20000 3 5.0 9 15.0 12 10.0 Total 60 100.0 60 100.0 120 100.0 Source: Authors’ Survey Data, 2005.

As shown in the table 3.1, there is hardly a difference in income ranges between two canal areas (head end and tail end). Monthly income between Rs 5,000 and Rs 10,000 is the conspicuously high range for both canal areas. In the second highest group, 33.3 percent of the farmers in the head end areas and 20 percent of the farmers in the tail end areas receive a monthly income of Rs 10,000-15,000. An income of over Rs 20,000 per month from farming is higher in the tail end areas (15 percent of households) than in the head end areas (5 percent of households). Average monthly household incomes from farming are Rs 9,409 and Rs 9,713 per month in the head end and the tail end areas respectively. In the tail end areas, the slightly higher average monthly income is due to the tendency to cultivate high value cash crops in comparatively larger land extents other than cultivating high water consumption traditional crops (specially paddy), in contrast to the head end areas.

Total number of households which depend on non-farm employment activities is 79 (66 percent) among the sample households. Average monthly household income from such activities is in the range of Rs 11,517 and Rs 11,410 in the head end and the tail end areas respectively. Income ranges from non-farm activities are given in the table 3.2.

Table 3.2: Monthly Household Income from Non-farm Activities

Income Range Head End Tail End Total (Rs.) No. % No. % No. % (N=43) (N=36) (N=79) < 1,000 0 - 1 2.8 1 1.3 1,000 – 4,999 8 18.5 15 41.8 23 29.2 5,000 – 9,999 18 43.0 10 27.8 28 35.5 10,000 – 14,999 6 13.9 5 13.8 11 13.9 15,000 – 19,999 4 9.3 3 8.2 7 8.8 20,000 – 24,999 3 7.0 1 2.8 4 5.0 25,000 < 4 9.3 1 2.8 5 6.3 Total 43 100.0 36 100.0 79 100.0 Source: Authors’ Survey Data, 2005.

Monthly income from non-farm activities between Rs 5,000 and Rs 10,000 is the prominent income range in the head end area, while the range in the tail end area is from Rs 1,000-5,000 per month which indicates that the income earning opportunities from non farm sector is higher in the head end areas. In all the areas the number involved in non-farm activities is only 66 percent of the farmer households.

3.3 Land Ownership and Land Use

Land is being an indispensable resource for any agricultural production system. Land ownership and availability have become the most crucial factors in the Mahaweli System H. At the initial stages of the Mahaweli development, settlers were given 2.5 acres of irrigated lowland and 0.5 acre of highland for homestead. However, with the population increase the pressure on limited land resources in the Mahaweli areas has increased. The pressure on land from the second and the third generation settlers for both homestead development and cultivation purposes intensified and resulted in encroachments on state lands and reservations and joint farming systems in limited lowland plots. In addition hidden transaction of land via leasing out and renting out also contributed in changing land size cultivated by individual farmers.

Table 3.3 compares the pattern of operation of lowlands in the areas concerned. Almost a similar pattern of land operation (land tenure) system is evidenced for both areas. Table 3.3 shows that in the Mahaweli system H area, a large proportion of land owners, especially in head end areas continue to cultivate their lands.

Table 3.3: Type of Land Operation (Land Tenure)

Type of land Head End Tail End operation No. % Total Avg. No. % Total Avg. Ext. (ac) Ext. (ac) (ac) (ac) Owner 54 90.00 120.00 2.22 48 80.00 108.00 2.25 operators Share tenancy 4 6.67 10.25 2.56 6 10.00 18.50 3.10 Share 2 3.33 5.00 2.50 4 6.67 9.00 2.25 cropping Lease out 1 1.67 2.50 2.50 2 3.33 5.00 2.5 Lease in 2 3.33 3.25 1.63 3 5.00 4.50 1.5 Encroachment 1 1.67 2.50 2.50 - - - - Total no. of 60 143.50 60 145.00 farmers Source: Authors’ Survey Data, 2005.

3.4 Cropping Pattern

The MASL provides water for cultivation in the lowlands for both yala and maha seasons, but not for an intermediate season. Water supply is restricted in most of the yala seasons due to the scarcity of water in feeder tanks. In the past, the farmers

adopted ‘Bethma’ System1 in water scarce seasons, which has ceased after the implementation of the BWA system. The farmers are encouraged to cultivate low water consuming crops especially in the yala seasons. Creating awareness among the farmers on the importance of cultivating the low water consuming, high value cash crops and their profitability was one of the main activities undertaken by the MASL.

Thus, the farmers have to consider the water availability as well as the risk component of the crops to be cultivated during the yala season. Therefore, the cropping pattern of the farmers in the Mahaweli areas is dependent on several factors. However, the farmers who have alternative irrigation resources tend to practice perennial crop cultivation and to undertake short duration crop cultivation in intermediate seasons as a third crop of the year.

With the adequate irrigation water supply after the onset of the North-East monsoon rain, most of the farmers tend to cultivate paddy during the maha. However, OFCs such as cereals, pulses, vegetables and oil crops are also cultivated in the maha season in a limited extent of land. Table 3.4 indicates the cropping pattern in the maha season (2004/2005) in the study areas.

Table 3.4: Cropping Pattern in Maha Season 2004/2005

Crop Head End Tail End Total Extent % Extent % Extent % (acres) (acres) (acres) Paddy 138.83 97.1 152.58 92.2 291.41 94.5 Chillies 1.00 0.8 0.75 0.5 1.75 0.5 Vegetables 0.99 0.7 7.50 4.5 8.49 2.7 Other crops 2.25 1.5 4.70 2.8 6.95 2.3 Total 143.07 100.0 165.53 100.0 308.60 100.0 Source: Authors’ Survey Data, 2005.

Paddy is the dominant crop cultivated in both areas during the maha season. Cultivation of low water consuming vegetables and other field crop varieties is relatively higher in the tail end areas (7.3 percent). Though, about 10 years ago the Mahaweli H was the major chillie producing area of the island, present chillie cultivation has become a much less important activity after the bad experiences of severe Thrips (pest) attacks and the marketing problems occurred in late nineties. Thus, the total land area under chillie cultivation at the moment in both the head end and the tail end areas in maha 2004/2005 are rather low. Cropping pattern in the yala season is described in table 3.5. The table shows that although the farmers have given high priority for paddy cultivation in the yala season accounting for nearly 50 percent of the total cultivated land extent, crop diversification is much more intense compared with that of the maha season. However as described in the previous section, total extent cultivated has dropped in the yala season, by 31 percent and 45 percent in the

1 Bethma is an instuition in which cultivation is practiced only in the upper reach of the system sharing the land with lower reach farmers when available water is insufficient to cultivate entire command area.

head end and the tail end areas respectively compared to that of the immediate preceding maha season.

Soybean (20 percent) and big onion (34 percent) are the two major OFCs cultivated in the head end and the tail end areas respectively. Soybean is one of the low water consuming crops mostly cultivated under forward purchase agreements with private sector companies. This kind of private sector intervention is observable in 101/D4 DCO area (head end) making soybean cultivation, rather conspicuous in that area. Big onion is also one of the main crops cultivated in the yala season in both Madatugama and Galnewa block areas (Galnewa block has the largest big onion cultivation under the Kalawewa tank). Though the water requirement for Big onion is comparatively higher (but less than that for paddy), the farmers resort to this crop for its high profitability. Vegetables and OFCs are also grown in larger extents in the tail end areas during the yala seasons than in the maha season. Cultivation of low water consuming crops like maize (5 percent) and other field crops (11 percent) is also featured in the yala season.

Table 3.5: Cropping Pattern in Yala Season 2005

Crop Head End Tail End Total Extent % Extent % Extent % (acres) (acres) (acres) Paddy 54.25 54.7 41.25 45.0 95.50 50.0 Soybean 19.33 19.5 0.25 0.3 19.58 10.2 Big onion 7.08 7.1 26.87 29.3 33.95 17.8 Vegetables 8.25 8.3 6.49 7.0 14.74 7.7 Chillies 5.05 5.0 0.87 0.9 5.92 3.1 Maize 0.75 0.8 4.87 5.3 5.62 3.0 Others 4.50 4.6 11.18 12.2 15.68 8.2 Total 99.21 100.0 91.78 100.0 190.99 100.0 Source: Authors’ Survey Data, 2005.

The drastic drop of cultivated land extent of paddy could be seen in both head end and tail end areas in yala season compared to the maha season. The percentage difference of cultivated paddy land extent between two seasons (maha 2004/05 and yala 2005) is around 51 percent in tail end areas, while it is around 42 percent in head end areas. Since the water supply is assured by the BWA system, the farmers tend to grow low water consumption high value cash crops especially in water scarce tail end areas. Even in head end areas the farmers have shown greater interest in cultivating cash crops in yala season.

The initial investment needed to cultivate high value cash crops is much higher compared to the paddy cultivation. Nevertheless farmers have shouldered the risk of the investment with assured water supply under BWA programme.

Table 3.6: Crops Newly Introduced after the Initiation of BWA System

Head End Tail End Total Crop No. of % No. of % No. of % farmers (N=60) farmers (N=60) farmers (N=120) Soybean 9 15 7 12 16 13.0 Black gram/ - - 9 15 9 7.5 Green gram/ Cowpea Big onion 3 5 2 3 5 4.0 Banana/Papaw 1 2 2 3 3 2.5 Vegetables 5 8 2 3 7 9.0 Gingerly (sesame) - - 3 5 3 2.5 Maize 1 2 2 3 3 2.5 Chillies 2 3 3 5 5 4.0 Gherkin/ Tobacco 1 2 2 3 3 2.5 Total 22 37 32 53 54 45.0 Source: Authors’ Survey Data, 2005.

After the initiation of the BWA, 45 percent of the farmers have changed their cropping pattern introducing new low water consumptive and high value cash crops. The majority of such farmers are from the tail end areas where water scarcity is severe. Those farmers accounted for nearly 60 percent. On the other hand the farmers have shifted to low water consuming cash crops with the intention of maximum utilization of available land and irrigation water. Soybean (13 percent), pulses (7.5 percent), maize (2.5 percent), sesame (2.5 percent) are the major crops preferred.

High value cash crops like big onion (4 percent), gherkin/ tobacco (2.5 percent), perennial crops (papaya and banana- 2.5 percent), vegetable crops (9 percent), chillies (4 percent) have also been adopted after the introduction of the BWA.

The introduction of the BWA marked a significant increase in the extent cultivated in the head end and the tail end areas in both seasons. Table 3.7, shows these changes. Apparently the total number of farmers who have increased the cultivated land extent after the BWA is higher in the tail end areas (32.5 percent) than in the head end (22 percent). Those increases resulted in changes in the cropping patterns of the respective canal systems.

Table 3.7: Increase of Cultivated Land Extent after the BWA

Yala Season Maha Season Total No. of % No. of % No. of % Canal System farmers farmers farmers N=60 N=60 N=120 Head end 17 28.3 9 15.0 26 21.7 Tail end 28 46.7 11 18.3 39 32.5 Source: Authors’ Survey Data, 2005.

3.5 Land and Water Issues

As discussed in the Section 3.3, the majority of the farmers (85 percent) operate their own land lots (lowland), but share tenancy is the second highest land operating pattern, 7 percent and 10 percent of the farmers in both the head and the tail end areas. Share cropping system is the third highest land operating type in the study areas. Such land operation systems have resulted in a drop in the income level from agriculture. Lack of profitability and income from agriculture, the non availability of land have weaned the youth away from agriculture, rendering it a main issue of the second and the third generation settlers, in the context of land and irrigation water.

In addition to the land scarcity, ill-drained lands and improper soil condition are other problems affecting the crop cultivation. Lands with no proper draining restrict the farmers in adopting high value cash crops (soybean, big onion, etc). About 12 percent of the farmers in the head end areas and 6 percent of the farmers in the tail end tend to practice paddy cultivation in the yala seasons due to the ill drained soil condition. Inadequate irrigation water supply to cultivate the full extent in the yala season is the main concern of the farmers in both ends although an increase in extent is seen in contrast to the past yala seasons.

47.5 percent of the respondent farmers have no access to alternative irrigation sources. However, the farmers who have such facilities tend to cultivate crops in the intermediate season along with the perennial crop cultivation. About 52 percent of the respondents have access to some kind of alternative irrigation sources (table 3.8).

Table 3.8: Alternative Irrigation Sources

Source Head End Tail End Total No. of % No. of % No. of % farmers (N=60) farmers (N=60) farmers (N=120) Own Agro-well 24 40 26 43 50 42 River 3 5 1 1 4 3 Drainage water 4 7 3 5 7 6 Neighbour’s agro- 1 1 1 1 2 2 well 32 53 31 52 63 52.5 Source: Authors’ Survey Data, 2005.

Agro-well is the major source of alternative irrigation for the farmers in both ends. 50 agro-wells in the study area are used by 57 farm families. Nearly 3 percent of the farmers have benefited from the neighbours’ agro-wells during spells of severe drought. Those farmers have cultivated 45.25 acres of the total land extent using agro- wells as a supplementary source of water in the study area. Some farmers in Galnewa block area, who have encroached on land near the river side of the Kala oya, pump water from the Kala oya illegally using lift irrigation methods. Use of drainage water flows from the fields in the upper reaches to the respective farmers’ land lot is another alternative for nearly 6 percent of the farmers.

Land productivity with maximum water is one of the main objectives of the BWA in the Mahaweli H area. It was found that the majority of the farmers strive to save the water. On-farm ditch system (Depaweli) is one of the strategies adopted by the MASL for water conservation and to optimize water utilization. The method cuts down the water delivery time to the farmer’s land lot in a water rotation and also reduces the conveyance losses. The conventional method takes more time to feed the whole land and water distribution from one field to another leads to more water loss. However, about 40 percent of the farmers experience difficulties in adopting the ‘Depaweli’ system due to the nature of the geographical location of their land. In other words the sub fields (liyadda) which are not located in proper gradient or slope cannot receive water via on farm ditch system made across the field. Out of the remaining farmers, only 58 percent have adopted the on farm ditch system without any difficulties. Reasons for this non-adoption are presented in table 3.9. The major reasons for refraining from this practice as perceived by the farmers are availability of sufficient water (7.5 percent), non-awareness about the ‘Depaweli’ system and the need of extra land for the construction of the farm ditch.

Table 3.9: Reasons for not Practicing Depaweli System

Reason Head End Tail End Total No. of % No. of % No. of % farmers (N=60) farmers (N=60) farmers (N=120) Location of the land 26 43 22 37 48 40.0 Wasting much land area 3 5 7 12 10 8.0 No need of such system 6 10 3 5 9 7.5 Ignorance about the 4 7 6 10 10 8.0 system Additional labour - - 1 2 1 1.0 requirement Total 39 65 39 65 78 65.0 Source: Authors’ Survey Data, 2005.

3.6 Institutional Mechanism Adopted for BWA

The development of the Mahaweli H area under the Mahaweli Restructuring and Rehabilitation Project (MRRP) has centred on five main activities, namely; i) irrigation rehabilitation, ii) water management, iii) institutional development, iv) agricultural development and agricultural produce marketing and v) self-employment and income generating activities. The BWA system was one of the components of the water management programme implemented by the MRRP that also has a direct relationship to the institutional development.

The BWA is a type of joint management system with the specified tasks allocated for different stakeholders including the MASL, the farmers, and the line agencies, which also require active participation of all stakeholders. The project aimed to establish an institutional setup considering the interests of all the concerned stakeholders.

The MASL has a well established administrative structure to manage the Mahaweli systems. The Mahaweli system H is headed by a Resident Project Manager (RPM). The system H has been divided into 9 administrative units called blocks, each of which is headed by a Block Manager. A block comprises of 3-4 units and units are administered by the Unit Managers (the total number of units in system H is 28). The unit Manager is the lowest level administrative officer who deals with the farmers and their organizations.

Farmer participation was solicited as an essential and a pre requisite of all stages of the project including project planning, designing and implementation and monitoring of development activities. Therefore, strengthening or establishing appropriate farmer based organizations for institutionalizing the BWA concept at different levels of the system was a vital requirement of the development programme.

Figure 3.1 gives a structural description of the farmer based institutional setup. Field channel farmer groups (FCG) are the lowest level of the institutional hierarchy, and it is an informal group. Field Channel Representative (FCR) is the leader of the FCG. The FCGs undertake the water distribution and O&M at FC level. The Distributory Channel Organization (DCO) is formulated with the participation of all the legitimate farmers in the DC command area (there are 256 DCOs in System H). The DCOs are legally registered organizations. The main tasks of the DCOs are proper distribution of water within the DCO command area, collection of O&M contribution from the farmers, mobilization of the farmers for sustainable O&M and preparation of the seasonal agricultural plan and the implementation of the agricultural plan with the collaboration of other organizations in the block area. In terms of Irrigation Ordinance (revised) Act, No. 13 of 1994, the farmer organizations have the legal power to charge an O&M fee from the member farmers. The Project Management Committee has decided to collect Rs 250 per hectare of land per season (Rs 500 per year). The unit level Support Committees are at the next level of the institutional hierarchy of farmer based institutional setup and it comprises the representatives from the DCOs in the relevant unit area and the MASL officials. The sub committee decides on the water delivery system based on the water quota issued to each DC.

At block level, a Sub Project Management Committee presided over by a farmer representative has been formed. The DCOs present all their issues related to water management and the seasonal agricultural plan to this committee. Apart from those institutions/organizations/committees, block level farmer federations (BLFF) were also formed under the MRRP. The sub project committee is basically a decision making body and its decisions are implemented by the BLFF. The BLFF has only farmer membership and office bearers of farmer based organizations (no MASL or other officials) and therefore it does not fall into the federation category mentioned in the Agrarian Development Act. Thus, the federations have no legal basis as an organization. The main entrusted tasks of the BLFF are, bulk water distribution from the main canal to all the DCOs in the block area, drawing up a water schedule for the season, undertake the main canal maintenance contracts from the MASL (15 percent commission from such contracts to the BLFF) and supply of seeds and other inputs to the farmers at reasonable price. The system level Project Management Committee

(PMC) is the highest organizational body consisting of the farmer representatives from each block (5 representatives from each BLFF) and the representatives from line agencies including irrigation, agriculture, banks, Crop Insurance Board and other relevant organizations. The main feature of the PMC is that, the number of farmers always outnumbers the officials. The PMC makes policy decisions on irrigation and agricultural activities.

Figure 3.1: Key Elements of the Farmer Based Institutional Setup Agriculture Irrigation System level PMC Credit

Others

Sub project Sub project Sub project

committee committee committee

Block level

BLFF BLFF BLFF

Unit level Unit level support committee

DC level DCOs DCOs DCOs

FC level FCG FCG FCG

Farmers Farmers Farmers

Direct involvement of the farmers in all the stages of the rehabilitation process including planning, decision making and project implementation and the contribution of resources in kind (not less than 10 percent of the total project cost) for the irrigation rehabilitation leads to create a sense of ownership among the farmers. The MRRP

also offered small rehabilitation contracts to the DCOs in order to develop the technical capabilities of the DCOs and to build up their financial capacities for future self operation of the canal system.

Before the turnover of the irrigation systems to the DCOs, a series of awareness creation programmes on PIM, seasonal agricultural planning, calculation of water requirements, water allocation, distribution, and measurements and monitoring of water management programme were provided to the farmers and the farmer leaders by the MASL. However, the MASL has not totally withdrawn their involvement in O&M of tertiary irrigation systems (DC and FC systems). The MASL has been providing technical guidance and specifications to the DCOs to undertake complicated maintenance works in a sustainable manner. In addition, the DCOs have to obtain the approval from the MASL for any rehabilitation work to be undertaken with their own cash investment.

The responsibility of distributing the allocated water quantity for the DC among all the FCs, considering the number of land lots for the given water quota devolves on the Water Master (Jalapalaka). He is trained in these activities and is an employee of the DCO. The Water Controller has been given adequate training in water measuring, allocation, distribution etc. The farmers are pre informed about the time schedule of water delivery and the quantity of water allocated for them is published through the public notice board kept for the purpose at each DCs. Therefore, the farmers have the volumetric impression about the water quantity they have already received and the quantity they are to receive in future water rotations.

3.7 Sustainability of the Institutional Mechanism

The sustainability of the institutional mechanisms is basically dependent on the strength of the DCOs and the performance of the DCOs in organizational management and water distribution and the degree of support provided by the line agency for the sustainable operation and maintenance. Organization leadership, members’ participation at the DCO meetings and other community activities and financial transparency and viability are considered as the indicators which determine the strength of the organization. Performance of the DCOs is needed in water distribution and system operation and maintenance.

3.7.1. Strength of DCOs

The Field Canal Representatives (FCRs) are the lowest level volunteer organizational cadre involved in the O&M of the irrigation systems and irrigation water management. Table 3.10 shows the farmers perceptions about the FCRs performances in both the head and the tail end areas. It is evident that the majority of the farmers (87.5 percent) in both areas are satisfied with the performances of the FCR. 97 percent of the farmers in the tail ends where water scarcity is more acute have appreciated the volunteer services provided by the FCR in the relevant field canal.

Table 3.10: Satisfaction about the FCR’s Performances/Duties

Head End Tail End Total Satisfaction No. of % No. of % No. of % farmers farmers farmers Satisfied 47 78 58 97 105 87.5 Not satisfied 13 22 2 3 15 12.5 Total 60 100 60 100 120 100.0 Source: Authors’ Survey Data, 2005.

The research findings also reveal that almost all the farmers are always or most of the time aware of the information on water delivery (time schedule and quantity of water). Further, 95 percent of the farmers opined that they get their fair share of irrigation water (62 percent of farmers always and 33 percent of farmers usually). These responses reflect that the institutional mechanism adopted at FC level is satisfactory.

‘Jalapalaka’ is one of the key persons operating under the BWA system who is responsible for issuing, distributing, delivering, measuring and monitoring of irrigation water among the farmers at FC level within a DC area with the assistance of the FCRs, the office bearers of the DCO and MASL officials. Therefore, the role of ‘Jalapalaka’ is very crucial in the BWA system. Table 3.11 presents the farmers’ perception about his activities in terms of irrigation water management. Over 85 percent of the farmers in the sample are happy with their performance in the relevant DC areas. It is noteworthy to observe that, 97 percent of the tail end farmers also hold the same view.

Table 3.11: Satisfaction about Jalapalaka’s Activities

Satisfaction Head End Tail End Total No. of % No. of % No. of % farmers farmers farmers Satisfied 45 75 58 97 103 86 Not satisfied 15 25 2 3 17 14 Total 60 100 60 100 120 100 Source: Authors’ Survey Data, 2005.

Participation of farmers for the organizational and O&M activities is another feature that determines the strength of the DCOs. About 95 percent of the farmers are of the opinion that most of the members of the DCO are providing their labour voluntarily for the maintenance of the FCs, the DCs and the branch canal systems. In addition to the beneficiaries’ contribution in kind for the O&M, the farmers mobilize their labour time for the DCO management activities via participating at various DCO meetings. The DCOs hold the members’ meetings and the office bearers’ committee meetings regularly. The minutes show that the past meetings were held with an agenda and the decisions taken have been implemented satisfactorily. The details of the beneficiary participation at the DCO meetings are tabulated in table 3.12. According to the table a

great majority of the members are regular participants at the DCO meetings indicating the higher strength of the DCOs.

Table 3.12: Regular Participation for DCO Meetings

Head end Tail end Total No. of % No. of % No. of % farmers farmers farmers Yes 57 95 55 91.7 112 93.3 No 3 5 5 8.3 8 6.7 60 100 60 100 120 100 Source: Authors’ Survey Data, 2005

In addition, almost all the farmers (119 out of 120 respondent farmers) extend their fullest support to the President of the DCO and other office bearers and co-operate in the organizational activities in water allocation and distribution and the O&M of irrigation systems and other agricultural development pursuits reflecting the positive aspects of the DCOs.

Table 3.13: Degree of Support from Farmers to the President of the DCO

Head end Tail end Total No. of % No. of % No. of % farmers farmers farmers Yes 60 100 59 99 119 99 No 0 0 1 1 1 1 60 100 60 100 120 100 Source: Authors’ Survey Data, 2005

Rise and fall of any organization is mainly dependent on its financial management. This aspect was assessed by investigating how the DCOs maintain their financial records, the bank account of the DCOs, and the procedures of fund disbursement for various activities. The research findings show that all the sample DCOs maintain accounts following accepted book keeping methods and audited by an external auditor. The joint bank accounts are maintained in the names of the treasurer, the president or the secretary of the DCO and authorized MASL officials and fund disbursement is possible only with the approval of the authorized officials of the MASL. The accounts are presented to the membership at the annual general meetings of the DCO. The farmer perception of the DCOs financial handlings was obtained and arrayed in table 3.14, the findings shows that over 85 percent of farmers are satisfied about the DCO financial handlings.

As O&M activities of the irrigation canal system below the DCs have been turned over to the DCOs, it is necessary to develop the financial viability of the DCOs. The DCOs maintain the O&M fund, raised through the undertaking of maintenance contracts at the initial stage of the MRRP and collecting O&M fees from the member farmers and various other business activities. 96 percent of the farmers regularly

make the O&M payments in the sample areas. The officials of the DCOs have been given the authority to take actions against the defaulters of the O&M fee. Imposing a fine for late payments or taking the defaulters to the ‘Samatha Mandalaya’ (a local arbitration board) are the most common actions taken. However, such instances are almost non-existent.

Table 3.14: Farmers' Satisfaction on Financial Handling of DCO

Head end Tail end Total No of % No of % No of % farmers farmers farmers Satisfied 44 73.3 59 98.3 103 85.8 Not satisfied 16 26.7 1 1.7 17 14.2 60 100 60 100 120 100 Source: Authors’ Survey Data, 2005

3.7.2. Performances of DCOs

Research findings have shed light on the fact that over 93 percent of the farmers are of the opinion that there is an improvement of the services provided by the DCOs after the initiation of the BWA. The findings on the farmers’ perception on the DCO’s performances in sustainable O&M show that, over 93 percent of the farmers are contented with the performances of the DCOs in efficient management of the irrigation infrastructure. Only a very negligible number (4 percent) opined otherwise. About 97 percent of the farmers are in agreement with the way resources are allocated (money, labour, time etc) for the sustainable O&M of the irrigation systems. The largest majority of the respondent farmers agreed that the DCOs regularly take necessary measures to maintain the canal system and to prevent it from damages.

However, the concept of developing block level farmer federations (BLFF) has failed at the time of the investigation, and their entrusted tasks have been undertaken by the MASL.

3.7.3 Agency Performance in System Maintenance

As the BWA programme is a joined management effort, the MASL has to undertake the allocated tasks routinely with sufficient resource mobilization to ensure the system sustainability beyond the DC level. However, the farmers’ perception on the support and assistance given by the MASL officials remain positive. Over 80 percent of the respondent farmers are of the view that they are always or usually receive the maximum support from the MASL officials in successfully implementing the BWA programme. Further, 93 percent of farmers appreciated the performance of the tasks entrusted to the MASL, especially doing a good job in the maintenance of the main canal systems. However, the level of services provided by the agency is basically dependent on the amount of fund allocated by the central financial agency.

As perceived by all the DCO leaders, agency participation for the PMC and the sub PMC is very impressive and they contribute very well in making proper decisions and being responsive to the farmers’ views. The officials have played a crucial role at the initial stages of the project in developing the farmers’ skills and capabilities to undertake turned over responsibilities.

CHAPTER FOUR

Irrigation System Performance under Bulk Water Allocation System

4.1 Water Supply Performance

The evaluation of the MRRP by the TEAMS (2003), highlights the irrigation system performance in Madatugama block during the early stages of the BWA, according to which the water delivery performance ratio for 2002 yala and 2002/03 maha signifies the water saving and higher water productivity values than expected. Mathmaluwa (2003) found that, water quota per ha for transferred canals was lesser than that for non-transferred canals in the year 2002. She also shows the increase in output per irrigation supply. The present study assesses the water supply performance through the following indicators:

4.1.1 Gross Water Quota/Seasonal Tank Duty

Gross water quota or seasonal tank duty is an indicator used to asses the performance of irrigation supply. Data was analyzed to find out the performance of tank duty before vs after the BWA in Mahaweli H area. The tank duty gives an overall idea of the efficiency of the whole system in providing water to save crop needs.

Tank duty (m) = Actual quantity of irrigation water used (m3) Actual extent cultivated (m2)

Figures 4.1 and 4.2 illustrate the trend of the tank water duty over the years, indicating the improvement in tank water duty after the year 2002. As Mathmaluwa (2003) pointed out, lowering of tank water duty is a combined effect of restricted water quota delivered under the BWA, cultivation of low water requiring crops, effective water management under the IMT and rehabilitation of the irrigation system (MRRP).

Performance achieved in tank water duty is a result of the efficiency in water distribution and utilization of water at secondary and tertiary level by the DCOs and water supply performance in the primary canal system. Therefore, the study analyzes the data of main canal water duties of the Mahaweli H to find out the trend of water duty in the main canals, where canal maintenance and water distribution are mainly handled by the MASL. The average main canal water duty during the yala seasons in the past 5 years prior to the implementation of the BWA is 4.85 ft, while the average value after the BWA is 3.43 ft. The main reasons for this reduction are the lesser conveyance losses due to the rehabilitation of the canal system and the strict management practices adopted in water supply after the BWA as perceived by the irrigation officials. The values of main canal duty showed a significantly increasing trend during the last 3 years in both seasons indicating a lowering of the performances. According to the officials of the MASL, the reduced government financial allocation for the O&M is a reason for the above situation.

Figure 4.1: Average Water Duty in Past Yala Seasons and Rainfall Pattern

7 600

6 500

5 400 4 300 3 Rainfall (mm) Water DutyWater (ft) 200 2

1 100

0 0 Duty 90 92 94 96 98 00 02 04 R/F Year

Figure 4.2: Average Water Duty in Past Maha Seasons and Rainfall Pattern

6 1800 1600 5 1400 4 1200 1000 3 800 Rainfall (mm) Water DutyWater (ft) 2 600 400 1 200 0 0 Duty 89 91 93 95 Year97 99 01 03 R/F

Figure 4.3: Average Water Duty in Past Maha Seasons in Main Canals

Duty Yeild 6.00 120.00

5.00 100.00

4.00 80.00

3.00 60.00 Duty (ft) Yeild (bu/ac) 2.00 40.00

1.00 20.00

0.00 0.00 90/91 92/93 94/95 96/97 98/99 00/01 02/03 Year 04/05

Figure 4.4: Average Water Duty and Rainfall Pattern in Past Yala Seasons in Main Canals

7.00 100.0 90.0 6.00 80.0 5.00 70.0 4.00 60.0 50.0 3.00 40.0 Rainfall(mm) Water DutyWater (ft) 2.00 30.0 20.0 1.00 10.0 0.00 0.0 90 92 94 96 98 00 02 04 Duty Year Yield

4.1.2 Extent of Land Cultivated in Relation to Water Supply

Extent of land cultivated per MCM of water was estimated in both maha and yala seasons from 1990 to 2005. The figures show that, the number of hectares cultivated per MCM water issued, had registered an increasing trend after the adoption of the BWA in 2001. Extent so cultivated for both seasons before the BWA was generally lesser than 90 ha and it has increased up to 130 and 160 ha after the implementation of the BWA during the maha and the yala seasons respectively. The figures No. 4.5 and 4.6 illustrate the pre and post situations. The lower value shown in 2004 yala is due to a severe drought that prevailed in the latter parts of 2003.

Figure 4.5: Extent of Land Cultivated per MCM of Water in Mahaweli H (Maha Seasons)

ha/mcm 140

120

100

60 Extent (ha) Extent 40

0 90 91 92 93 94 95 96 97 98 99 2000 2001 2002 2003 2004 2005 Year

Figure 4.6: Extent of Land Cultivated per MCM of Water in Mahaweli H (Yala Seasons)

ha/mcm 180 160 140 120 100 80

Extent (ha) Extent 60 40 20 0 90 91 92 93 94 95 96 97 98 99 2000 2001 2002 2003 2004 2005 Year

4.1.3 Percentage of Land Extent Cultivated in Dry Season (Yala)

Area cultivated during the past yala seasons in the Mahaweli H area before the BWA was far below the expectations of the original plan of the Mahaweli development programme and in many years it had been below 50 percent of total cultivable land. However, the extent cultivated after the 2002 yala was over 60 percent of the total land extent except in the year 2004 which experienced a severe drought all over the country. Figure 4.7 illustrates the percentage of land extent cultivated in the past yala seasons. This is basically an outcome of the water saving of the maha season following the strict water management policies adopted with the implementation of the BWA and more farmers shifting to less water consuming crops.

The aim of the BWA programme is to reach 100 percent extent of cultivation, but it is yet to be realized. The extent cultivated in 2005 yala has reached about 93 percent of the total extent, a remarkable achievement compared with the past seasons.

Figure 4.7: Percentage of Land Extent Cultivated in Past Yala Seasons

100 90 80 70 60 50 40 Extent (ha) Extent 30 20 10 0 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 Year

4.1.4 Cropping Intensity

Annual cropping intensity (CI) for the past 10 years of the Mahaweli H area was examined to understand the change over the years. The findings are shown in figure 4.8, which indicate a gradual increase in the cropping intensity after the implementation of the MRRP and the BWA in 2002. One of the reasons for the lower CI before 2002 might be that the infrastructure was in a state of disrepair prior to rehabilitation. According to the farmers and the officials a higher CI was achieved not only due to the MRRP, but also with resorting to less water consuming crops and the efficient use of water in both seasons.

Figure 4.8: Changes in Cropping Intensity in Mahaweli H area

250

y = 4.7273x + 127.4 200 R2 = 0.4851

150

100 Cropping Intensity Cropping

0 96 97 98 99 00 01 02 03 04 05 Year

4.1.5 Relative Irrigation Supply

Relative irrigation supply (RIS) = Actual irrigation supply Planned irrigation supply

The RIS provides some insight of the water scarcity or abundance and is an indication on how accurately water was delivered with reference to the planned water schedule. Therefore, the RIS is an indication of the efficiency of the management of the system. Under the BWA, the farmers are involved in preparing seasonal water schedules from the beginning, and this is a good indicator to assess the success of the farmer participation.

In this study, the RIS value was calculated for the maha and the yala separately from the year 2001/02 to 2004/05 for selected canals. The selected canals are 104 D7 and 101 D4 in Madatugama block and 302 D1 and 306 D5 in Galnewa block. The findings are given in table 4.1.

Table 4.1: Relative Irrigation Supply

Maha Seasons Relative irrigation supply 104 D7 101 D4 302 D1 306 D5 2001/02 0.96 0.91 1.00 1.00 2002/03 0.97 0.93 1.00 1.00 2003/04 0.98 0.98 1.00 1.00 2004/05 0.96 0.91 1.00 1.00 Yala Seasons 2002 1.06 0.98 N.A N.A 2003 - - 1.12 1.12 2004 0.69 0.79 1.00 1.00 2005 0.91 0.95 1.00 1.00 Source: DCO Records

The RIS values were calculated using the data gathered from the BLFFs of Madatugama and Galnewa. The RIS equal or closer to 1.0 indicates effective irrigation planning at the DCO. Lesser than 1.0 means there is water stress in the particular seasons or there is an over estimation of water requirement at the planning stage. Despite a RIS value of lesser than one in some season, certain canals have achieved the planned cultivation extent without any significant crop damages. Therefore, one can conclude that these canals are not under water stress condition.

The method of approximation used in calculating the water requirement at the planning stage may also cause deviation of values in the RIS (Mathmaluwa, 2003).

4.2 Performance of Productivity and Farmers Income

4.2.1 Changes in Crop Yield

The major crop cultivated in the Mahaweli H system is paddy in both seasons. Therefore, the change in the paddy yield was observed to understand the trend of yield over the years. The findings are illustrated in figures 4.9 and 4.10, which mark an increasing trend of yield over the years. The increase of yield in the yala seasons is very prominent after the year 2002. The outcome observed in yield may be due to multiple factors such as improved variety, access to extension, use of straw/organic fertilizers and timely cultural practices. However, the DCO leaders perceived that systematic water management and reliable supply is one of the main reasons for this increase. According to TEAMS (2003), the yield level was low before the BWA programme and some tail end farmers received a yield of only 40-50 bu/ac2 in that period.

2 1bu(bushel) = 20.87 kg

Figure 4.9: Changes in Paddy Yield in Mahaweli H – Yala Seasons

y = 1.4213x + 50.788 R2 = 0.581 100 90 80 70 60 50 40 Yeld (bu/ac) 30 20 10 0 81 83 85 87 89 91 93 95 97 99 01 03 Year

Figure 4.10: Changes in Paddy Yield in Mahaweli H – Maha Seasons

y = 0.5998x + 88.787 120 R2 = 0.1956

100

Yield (bu/ac) 40

0 81 83 85 87 89 91 93 95 97 99 02 04 Year

4.2.2 Changes in Cropping Pattern

Types of crops/crop combinations grown in both the maha and the yala seasons before and after the BWA programme were studied. There is no much difference in the cropping pattern practiced in the maha season and the extent under the OFC is generally between 650 ha to 1700 ha. However, the extent under the OFC in the past yala seasons has increased as a result of multiple factors especially, the reliable water supply after the BWA, establishment of forward purchase agreements with the private sector and the BLFF and priority given by the BLFF to cultivate low water consuming crops. The cultivation of the OFC requires more capital investment involving high

risk in the entire cultivation and marketing of the output. However, assurance of specified quotas of water under the BWA programme and the establishment of forward purchase agreements for selected OFCs’ with the private sector reduced the risk factor motivating the farmers towards the OFC cultivation.

There are arguments that, motivation of farmers to cultivate low water consuming crops has lead to a reduction in extent under high value cash crops (onion, chillies, vegetables) and an increase in low return field crops (green gram, soy bean, cowpea) (Mathmaluwa, 2003). The latest situation in this regard was assessed in Madatugama and Galnewa blocks using the extent of the OFC cultivated in the past yala seasons. The results are shown in figures 4.11 and 4.12.

These figures reveal that, the OFCs’, irrespective of high or low value, have been cultivated increasingly after the BWA except for the chillie crop. The reason for the drop in the extent of chillie cultivation is a pest attack in the area causing heavy losses to the farmers. It should be noted that, the selection of the OFC also depends on the price factor in the local market in the cultivation year and the previous year.

The cultivation of the OFCs was in practice from the beginning of the Mahaweli development programme in the H area especially during the yala seasons due to scarcity of water. However, as discussed earlier, encouragement given to cultivate the OFCs was higher after the BWA. The present study attempted to assess the percentage of farmers who took to new OFCs’ after the implementation of the BWA. The results are given in table 4.2 which again prove that, the farmers have got involved in crop diversification incorporating both high value and low value crops.

Figure 4.11: Trend of OFC Cultivation – Galnewa block

700 650 600 550 500 Chillies 450 Big onion 400 Vegetables 350 Soybean 300 Extent (ha) Extent 250 200 150 100 50 0 97 98 99 00 01 02 03 04 05 Year

Figure 4.12: Trend of OFC Cultivation – Madatugama block

800 750 Chillies 700 Big onion 650 Vegetables 600 Soybean 550 500 450 400 350 Extent (ha) Extent 300 250 200 150 100 50 0 97 98 99 00 01 02 04 05 Year

Table 4.2: Adoption and Omission of OFCs after BWA

% of new farmers adopted % of farmers given up Crop in Yala (N=120) the crop in Yala (N=120) Chillies 4 10 Big onion 4 4 Vegetable 5 2 Papaw 1 0 Banana 2 0 Soybean 13 0 Ground nut 3 1 Tobacco 1 0 Maize 2 0 Red onion 1 0 Gherkin 1 0 Black gram 4 1 Green gram 3 0 Source: Author's Survey Data, 2005

4.2.3 Agricultural Income Inequality

The estimated average Gini ratio (G) for agricultural income of the head end areas in the study location of Madatugama and Galnewa areas is 0.12 and that for the tail end area is 0.22 (figure 4.13). The average national Gini ratio in 2003 was 0.42,

compared with which the Mahaweli system H under the BWA shows more equality in terms of income distribution in both the head and the tail ends. Very interestingly, the G value estimated in the Mahaweli H is much lower than that of many other irrigation schemes estimated in earlier studies (Hemaratna et-al 1991, Ratnayake, 1986). The major reason for the achievement can be attributed to the outcome of the BWA and the fair allocation of irrigation water adequately and timely on reliable rotations.

Figure 4.13: Lorenz Curve of Income Inequality

120

100

40 Tail end

Cumulative % of Income of % Cumulative Head end 20

0 0 20 40 60 80 100 Cumulative % of Population

4.2.4 Land and Water Productivity

Land and water productivity values are calculated in terms of gross value of output obtained per hectare of land and per cubic meter of irrigation water respectively. Rice is the major crop in the Maheweli system H in the maha seasons. Although there is a small extent of land under the OFCs in this season, non-availability of reliable data on the yield and the prices of the OFCs restricted the valuation to rice cultivation. The international price of rice was calculated at 2005 constant US dollar value. Productivity values were estimated from 1999/2000 to 2004/2005 maha season. To avoid the effect of price fluctuations of rice in the local and the world market, the average prices are used for the reference period (1999-2005).

The findings on land and water productivity are illustrated in figure 4.14, which reveal that land productivity has not increased over the time compared with the pre or post BWA. However, water productivity values mark an increasing trend after the BWA indicating the performances in water distribution. Although there is an increase in water productivity values comparatively after the BWA, the increased value has dropped during the last three years, which gives an alarming signal of a decrease in water use efficiency as witnessed by the increasing main canal water duty during the last three years (figures 4.3 and 4.4).

Figure 4.14: Changes in Land and Water Productivity

Land Productivity Water Productivity 1400 0.16

1200 0.14 0.12 1000 0.1 800 0.08 600

SVGP $/ha 0.06 SVGP Rs/m3 400 0.04

200 0.02

0 0 99/00 00/01 01/02 02/03 03/04 04/05 Year

Table 4.3: Summary of Selected Performance Parameters

Performance Performance before BWA after BWA Percent change Performance parameters (1996-2000) (2001-2005) Maha Yala Maha Yala Maha Yala 1. Tank water duty (m) 1.18 1.50 0.889 0.94 -24.6 -37.3 2. Main canal water duty 4.29 5.22 3.12 3.41 -27.1 -34.53 (m) 3. Area irrigated per unit 88.9 81.39 119.36 119.96 +34.26 +47.38 of water (ha/MCM) 4. Water productivity 0.1 - 0.13 - +30 - (maha) (US$/m3) 5. Land productivity 1087.5 - 1148.9 - +5.6 - (maha) (US$/ha) 6. Paddy yield (t/ha) 5.03 3.98 5.21 4.49 +3.6 +12.7 7. Extent under OFC (ha) 865.75 5854 1041.75 8971 +20.3 +53.2 8. % of land extent - 42.45 - 64.63 - +52.2 cultivated in yala season 9. Annual cropping 148.25 164.11 +10.7 intensity

4.2.5 Farmer Performance Index

Pingali et-al (1990) has developed an index called farmer performance index, which was defined as the ratio of farmer yield to the location-specific yield potential. The indictor provides an idea of farmers' ability to exploit the yield potential in the given circumstances. According to the Deputy Resident Project Manager (Agriculture) of the Mahaweli H, the potential paddy yield of the Mahaweli H area is 6 metric tones

per hectare. The performance index has been calculated separately for the paddy farmers in the head end and the tail end of branch canals in both seasons. Figures 4.1.5 and 4.1.6 illustrate the findings. On an average, the farmers have achieved 93 percent (5,580 kg/ha) and 80 percent (4,796 kg/ha) of the technical efficiency in paddy cultivation during the maha and the yala respectively in the year 2004/2005. The findings also suggest that about 31 percent of the farmers during the maha in both ends and 15 percent of the farmers during the yala were able to exceed the yield potential. About 55 percent and 38 percent of the farmers achieved the technical efficiency of 85-110 during the maha and the yala respectively.

Figure 4.15: Farmer Performance Index for Head and Tail Areas (Maha Season- 2004/05)

220 Tail 200 Average Head 180

160

140

120

100

80 Performance Index 60

0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 Farmer No.

Figure 4.16: Farmer Performance Index for Head and Tail Areas (Yala Season- 2005)

140 Head Tail 120 Average

100

60 Performance Index Performance 40

0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 Farmer No.

4.3 Farmers’ Perceptions on Irrigation Systems Performance Realized after BWA

Almost all the farmers were in favour of the performances realized from the BWA compared with water management practices implemented in the past. The main benefit of the BWA is that the farmers are aware of the water quota they are expected to get in the season in advance, facilitating them to plan for the cropping system and the cultivation practices accordingly. About 95 percent of the farmers perceived that they always or usually get their fair share of water, while only 5 percent viewed that they get the specified quota sometimes.

The perceptions on the impact of the BWA on crop yield, income from agriculture and profitability of agriculture are positive for a large proportion of farmers (table 4.4). About half of the respondent farmers perceived that, income and profitability of agriculture has increased after the implementation of the BWA. The DCO leaders of all selected locations perceived that, the requested amount of water supply has led to a significant yield increase in paddy cultivation.

Table 4.4: Impact of BWA on Profitability and Income of Agriculture

% of responses Impact Head end farmers Tail end farmers (N=60) (N=60) Increased income and profit 52 45 No change in income and profit 40 48 Reduced income and profit 8 7 Source: Authors’ Survey Data, 2005

The farmers have realized an enhanced income during the yala seasons following the increase in the extent of cultivation and the cultivation of high value crops. The farmers' perceptions on advantages and disadvantages of the BWA system are discussed in the following sections.

Table 4.5: Impact of BWA on Crop Yield

Impact % of responses Paddy Paddy OFC OFC (Maha) (Yala) (Maha) (Yala) Head end (N=59) (N=49) (N=19) (N=42) Increased yield 33 35 27 43 No difference in yield 47 47 68 47 Decreased yield 20 18 5 10 Tail end (N=56) (N=48) (N=20) (N=37) Increased yield 29 30 10 20 No difference in yield 55 58 85 70 Decreased yield 16 12 5 10 Source: Authors’ Survey Data, 2005

4.3.1 Benefits of Bulk Water Allocation System

The implementation of the BWA system has witnessed a number of advantages as responded by the farmers. The major advantage for about 83 percent of the beneficiaries is the reduction in the number of water conflicts among the farmers in sharing water, due to the systematic water allocation of the BWA. This was a plus point irrespective of the location of the irrigation channel (head or tail). TEAMS (2003) recorded similar experiences and referred to many conflicts and disputes among the farmers and between the farmers and the officials on the issue of sharing and allocating water for their fields prior to the MRRP project. The major reasons for this adverse situation were inadequate water deliveries, damaged structures, lack of co-operation among the farmers and their ignorance (ibid).

The other very important achievement of the BWA as perceived by the farmers is the change of attitudes of the farmers in water use and irrigation management. The farmers who receive irrigation water to their fields make efforts to save water and reduce the off takes of water as soon as they have an adequate supply. The farmers are conscious that, excess water is not only bad for their crop, but also a waste that will cost them in the next season. According to table 4.6, it is interesting to note that, more farmers in the tail end areas have experienced reduced incidence of crop failure, increased extent of cultivation especially in the yala seasons and increased income compared to that of the head end farmers.

Table 4.6: Benefits of Bulk Water Allocation System as Perceived by Farmers

Head end farmers Tail end farmers Benefits (% of responses) (% of responses) N=60 N=60 1 Decreased number of water 83 83 conflicts among farmers 2 Change of farmers attitudes on 60 63 water use and management 3 Reduced incidences of crop 45 55 failures 4 Increase in farmers income 32 40 5 Increase in extent of cultivation in 28 50 yala 6 Change in agency attitudes on 27 22 water distribution 7 Increased cultivated extent in 15 18 maha season 8 Reduced cost of production 2 5 Source: Authors’ Survey Data, 2005

4.3.2 Constraints of Bulk Water Allocation

Under the BWA, the DCO has to allocate water among the farmers once they receive the water quota for each rotation irrespective of day or night. Some farmers get their water in the night time. One of the major constraints for about 33 percent of the farmers in the head end and 17 percent of the farmers in the tail end is their experiences of difficulties in receiving the water issues by night after the BWA (table 4.7).

The farmers expected to cultivate 100 percent of the land extent in both seasons after the BWA system. In the yala season the extent of cultivation has increased, but not by 100 percent. This was one of the concerns for about 25-30 percent of the farmers. Another problem was the time gap that exists between two water rotations, especially when different farmers cultivate different OFC in the same FC. For instance, the gap between two-irrigation rotations set for soybean is not suitable for onion cultivation, especially during the bulb development stage. Onion needs more water than soybean, maize and chillies. A longer gap between two rotations causes drying of crops leading to a drop in the yield.

Inability to cultivate the desired crop is another major drawback expressed by the beneficiaries. For instance if a farmer wants to cultivate the rice crop for economic or social reasons, he is encouraged to cultivate other field crops depending on the water allocated.

Table 4.7: Constraints of Bulk Water Allocation System as Experienced by Farmers

Farmers in Head Farmers in Tail Problem end areas (% of end areas (% of responses) (N=50) responses) (N=60) 1 Logistic difficulties in receiving 33 17 night time water issue 2 Inability of cultivate full extent 25 30 3 Inability to cultivate desired crops 22 25 4 Non provision of promised water 18 22 quantity 5 Larger gaps between the water 17 22 issues in later stages of cultivation 6 Difficulties in marketing of OFC 12 28 7 No problems 13 10 8 Need to pay O&M fee 8 3 9 Unsuitable irrigation rotations for 7 7 different OFCs Source: Authors' Survey Data, 2005

CHAPTER FIVE

Role of Bulk Water Allocation in Sustainable Maintenance of Irrigation Infrastructure

5.1 Level of Resource Mobilization for Operation and Maintenance

The complete turnover of the O&M responsibilities from the FC to the DC under the BWA necessitates the farmers’/DCO’s responsibility of mobilizing sufficient resources for sustainable O&M. Sustainability of infrastructure largely depends on proper maintenance of the system from the primary level (head system) to the tertiary level (farm gate). The task needs mobilization of beneficiary labour time and cash. There are broadly four categories of resource mobilization in practice in irrigation system management at the farmer level (Aheeyar, 1997).

(a) Mobilization of labour for group activities – eg. distributory canal maintenance (de-silting and jungle clearing) (b) Mobilization of labour for individually allocated tasks. eg. maintenance of field canal (c) Mobilization of beneficiary time for various meetings. eg. planning, decision making and various DCO meetings. (d) Mobilization of cash and materials. eg: payment of the O&M fee at the rate of Rs 250 per ha per season (for masonry works, structural repairs, and painting and greasing of the structures and for the salary of operation, labour and expenses of DCO administrative work)

The level of resource mobilization by the farmers in kind and cash for the system O&M is indicated in tables 5.1 and 5.2. Value of mobilized labour ranges from Rs 1,000 to Rs 1,775 per ha per annum. The amount of cash mobilized by the farmers to the DCOs is about Rs 500 per ha per annum. Samad and Vermillion (1999) estimate that the annual irrigation cost to the farmers in Nachchahduwa and Hakwatuna-Oya major irrigation schemes is about US$ 15 per ha. However, the cost for the farmers for irrigation water under the BWA was over US$ 20 per ha at 2005 US$ constant prices except in the DCO called as 302D1. The US$ 5 difference shown in the two estimates is basically due to the current O&M fee of US$ 5 per ha per year. The amount of cash mobilized for the system O&M is a significant feature in resources mobilization under the BWA compared to various past attempts of cash collection for the O&M by the DCO’s and the government agencies for turned over irrigation channels in the major irrigation schemes. Aheeyar (1997) found in selected irrigation schemes that, the value of cash and material mobilization by the farmers for the DCO’s for O&M was desperately low and sometimes it was recorded as zero.

Table 5.1: The Estimated Value of the Mobilized Labour by Farmers for Maintenance – Year 2005

Name of Average Average labour % of farmer % of Estimated the DCO labour mobilization of participation farmers value of mobilization meetings per for group participation mobilized for group farmer (in works for meetings labour works per days) (c) (Rs/ha) farmer ( in (b) (d) (e) days) (a) 101D4 2.94 2.12 95 75 1424.50 104D7 3.03 2.99 98 75 1772.50 302D1 2.08 2.6 94 70 1032.00 306D4 2.51 2.97 97 75 1445.00 Note: e= [No. of farmers in the DCO* (a) *opportunity cost of labour] + [No. of farmers in the DCO * (b) * (d) * opportunity cost of labour] / DCO command area Source: Authors’ Survey Data

Table 5.2: Estimated Value of Cash Mobilization and Total Value of Mobilized Resources for O&M

Number of DCO No. of Value of cash Total value of the DCO command farmers mobilization mobilized area (ha) regularly Rs/ha/year resource (Value paying O&M of labour + Cash) fee (Rs/ha/year) 101D4 40 39 487.50 1912.00 104D7 130 130 500 2272.00 302D1 153 143 467.32 1447.32 306D4 49 49 500.00 1945.00 Source: Authors’ Survey Data, 2005

5.2 Farmers Investment on Irrigation Infrastructure and the Sustainability

The sustainability of infrastructure is basically dependent on the level of farmer participation on O&M, regular payment of the O&M fee, investment made for the infrastructure maintenance by the DCO and O&M and beyond the DC by the line agency. The value of the resource requirements for adequate O&M activities as estimated by TEAMS (1991) in major irrigation schemes is Rs 914 per ha/year at 1990 prices. This is equivalent to Rs 2,726 at 2003 prices. However, the estimated total value of mobilized resources towards the system O&M in the Mahaweli system H under the BWA programme ranges from Rs 1,450 to Rs 2,275 at 2005 prices. The finding reveals that the mobilization of resources is little lower than the TEAMS estimate, but the present contribution may be sufficient due to the following reasons:

a) The infrastructure was recently rehabilitated and therefore maintenance requirement is lower

b) The actual value of mobilized labour by the farmers can be higher than the estimated value because of the farmers’ commitment and completion of allocated maintenance tasks as early as possible to reduce voluntary labour time, since labour has an opportunity cost for the farmers.

Farmer perceptions were obtained to understand the various aspects of resource contribution and investment for the O&M and its implications. Farmer participation in the O&M of the DCs and the FCs is very important and virtually all the farmers actively participate in the routine maintenance of the DCs and the FCs (tables 5.3 and 5.4) as perceived by a majority of the farmers. The sample farmers were inquired about the structural problems of the DCs and the FCs after the BWA. About 90 percent of the farmers perceived that unlike in the pre-BWA period, there is a decrease in the number of broken gates, broken structures, damaged pipe outlets, breached bunds and cases of sifted canals with the implementation of the BWA programme.

The farmers are routinely involved in jungle clearing and de-silting of the DCs and the FCs twice a season. The DCOs invest money on structural repairs and the greasing and the painting of structures at the beginning of each season. The farmers were asked about their perception on the following statement "the level of resource mobilization for system O&M by the DCO is sufficient to ensure physical sustainability of infrastructure". About 26 percent of the farmers 'strongly agree' with the statement, while about 71 percent 'agree'. Only 3 percent disagreed (figure 5.2). The above statement was further confirmed when the farmers were asked to comment on the following statement. The "DCO maintains the infrastructure very efficiently and the quality of maintenance work is well above the level of pre-BWA maintenance". The above statement was rejected only by 4 percent of the farmers and 35 percent of the farmers strongly agreed with the statement (figure 5.3).

Table 5.3: Degree of Labour Mobilization for O&M

% of respondents Degree of mobilization Madatugama Galnewa (N=60) (N=60) 1. All members in the FC participate 42 29 2. Most members in the FC participate 53 68 3. Some members in the FC participate 5 3 4. No members in the FC participate - - Source: Authors’ Survey Data , 2005

Table 5.4: Degree of Participation for Meetings as Perceived by the Farmers

% of respondents Degree of participation Madatugama Galnewa (N=60) (N=60) 1. All members actively participate in 5 8 DCO meeting 2. Most members actively participate in 80 74 DCO meeting 3. Some members actively participate in 13 15 DCO meeting 4. No members actively participate in 2 3 DCO meeting Source: Authors Survey data 2005

The farmer perceptions were obtained about the proper maintenance of the DC canal by the DCOs. Over 50-60 percent of the farmers perceived that over 75 percent of the total canal length is properly maintained, while about 90 percent of the beneficiaries mentioned the corresponding figure as above 50 percent of the canal length by DCO (figure 5.4).

Figure 5.1: Structural Problems of DCs & FCs before and after BWA

90 80 70 60 Decreased 50 No change 40 30 20 10 - - - - -

Broken gates Broken Broken pipe Broken bund Silted canal structures outlet problem

Figure 5.2: Is the Extent of Resource Mobilization for System O&M by DCOs Sufficient to Ensure Physical Sustainability of Infrastructure? (As a percentage of farmer perceptions)

Dis Dis agree agree Strongly 5% Strongly 2% agree agree 20% 32%

Agree Agree 78% 63%

Pre Project Post Project

Figure 5.3: Is the Quality of Maintenance Works Conducted by DCOs is Comparatively Higher than the Pre-turnover Maintenance? (As a percentage of farmer perceptions)

Dis agree 2% Dis agree Strongly 7% Strongly agree agree 35% 35%

Agree 63% Agree 58% Pre Project Post Project

Figure 5.4: Percentage of DC Canal Properly Maintained after BWA as Perceived by Beneficiary Farmers

Galnewa Madatugama <50 2 5

50-75 40 43 % of Canal

>75-100 58 52

0 20 40 60 80 100 120 % of Farmers

The O&M fee is regularly paid by 92 percent of the farmers in the head end areas and 100 percent in the tail end areas, showing the enormous concern of the farmers in sustainable O&M of the irrigation system. The major reasons for the non-payment or irregular payment of the O&M fee by about 8 percent of the head end farmers are:

(a) Crop failure (60 percent of cases) (b) Lack of confidence over the DCO leadership (40 percent of cases)

About 80 percent of the sample beneficiaries perceived that the line agency officials were easily accessible always or usually, that they provide their complete assistance for irrigation system management (Table 5.5). About 95 percent of the farmers revealed that the line agency always or often undertakes the routine maintenance above the DC level (main canal and head system) (Table 5.6).

Table 5.5: Supports Given by Officials for O&M and DCO Activities

Degree of support % of responses Madatugama Galnewa 1. Officials are always easily approachable and 55 40 provides full support 2. Officials are usually easily approachable and 22 43 provides full support 3. Officials are sometimes approachable and 22 17 provides full support 4. Relevant officials are never approachable 1 - Source: Authors’ Survey Data, 2005

Table 5.6: O&M beyond DC Level by the Line Agency as Perceived by the Farmers

% of respondents Degree of work Madatugama Galnewa 1. Always do the routine maintenance in mainland 48 35 2. Often do the routine maintenance in mainland 48 58 3. Rarely do the routine maintenance in mainland 4 7 4. Never do the routine maintenance in mainland - - Source: Authors’ Survey Data, 2005

5.3 Willingness to Pay for Operation and Maintenance

The Mahaweli H system was properly rehabilitated under the MRRP before the commencement of the BWA programme in 2001. Therefore, the expenditure requirement for maintenance activities was lower in the initial years. However, the need for maintenance will increase over time and the current level of resource mobilization may be insufficient. Therefore, the study made an attempt to assess the farmers’ willingness to pay (WTP) for sustainable O&M of turnover irrigation system for the DCO in addition to their current level of cash and labour mobilization.

A direct inquiry on an irrigation fee is a very sensitive issue with the farmers and it may be hard to elicit an answer. However, a traditional custom, which exists in Sri Lanka, is the donation of a certain proportion of paddy to the Irrigation Headmen (or Jalapalaka) after each harvest for his services. But, it is not in practice in the new irrigation schemes. Therefore, the farmers chosen for the survey were asked about their WTP for improved irrigation services and sustainable O&M of irrigation infrastructure. Before posing the question, the beneficiaries were clearly explained about the existing status of irrigation infrastructure, the expected future cost escalation, and the institutional context in which the water is to be provided. They were also made aware of the funding procedure and responsibilities under the agreement signed with the BWA programme. Then the WTP question was stated as follows:

“How many kg (or bushels) of paddy are you willing to give to your DCO per hectare of cultivated land per season in addition to the payment of your current O&M fee and voluntary labor mobilization for maintenance and DCO activities, in order to maintain the turn over irrigation infrastructure in a better condition to provide improved irrigation services".

The approach adopted in this study to find out the WTP in terms of kind (paddy) rather than in cash was very successful and all the farmers responded well. The WTP in terms of paddy was converted into monetary value using 2005 paddy prices prevalent in the study areas.

The average annual WTP ranges from 52 kg/ha to 77 kg/ha. This is equivalent to the money value of Rs 599/ha to 890/ha/annum at 2005 paddy prices. Only 8 farmers

were unwilling to provide any thing other than the current level of resource mobilization. It is noteworthy to mention that WTP values estimated is in addition to the current level of labour mobilization and O&M fee collection.

Table 5.3 shows the WTP values obtained in selected DCOs. The higher values of standard deviations of WTP indicate the high variation in amount that various farmers are willing to pay.

Table 5.7: WTP for System O&M (kg of paddy/ha/year)

Name of the 101 D4 104 D7 302 D1 306 D4 DCO Mean 77.37 71.87 76.96 52.12 Maximum 217.4 217.4 326.1 217.4 Minimum 21.75 0 0 0 Std division 58.99 65.14 79.82 47.92 Source: Authors' survey Data, 2005

Table 5.8: WTP for System O&M (Rs/ha/year) **

Name of the 101 D4 104 D7 302 D1 306 D4 DCO Mean 889.73 826.55 885.09 599.30 Maximum 2500.70 2500.70 3751.06 2500.7 Minimum 250.07 0 0 0 Std division 678.48 749.12 918.03 551.08 **Calculated using average farm gate paddy (Nadu rice) price at year 2005 Source: Authors' Survey Data, 2005

The amount that farmers are WTP to the DCO's towards system O&M is an impressive point compared with attempts made in the past to collect O&M fee through the centralized financial agency, which had an unsuccessful short life of 4 years. The existing WTP should receive the attention of the relevant authorities to cope with future cost escalations and increasing maintenance requirements of ageing rehabilitated infrastructure.

5.3.1 Factors Affecting Willingness to Pay for Operation and Maintenance

Multiple regression model specified in section 2.12 was run using the SPSS computer software package. Ten independent variables [total family income, major yala crop, total low land extent, DC location, age, education, type of farming (full time/part time), contribution of labour for DCO meetings and contribution of labour for maintenance work] were identified as the factors, which may influence the amount of WTP.

The results of the analysis are presented in table 5.9. Out of nine variables, four variables are statistically non significant (age, type of farming, labour contribution for maintenance and level of education). Four independent variables are statistically

significant at 1 percent level (total family income, total low land extent, amount of labour contribution for meetings and gender) and the type of crop cultivated in the yala season is statistically significant at 10 percent level and the location of the DC (head or tail) is significant at 5 percent level. The positive and significant co-efficient on total family income indicates that beneficiaries having a higher income are willing to pay more for improved irrigation services and are more concerned about the sustainability of irrigation infrastructure. Similar results obtained for the total low land extent cultivated, show that the farmers who cultivate larger low land extents have expressed more willingness to pay. Positive and significant co-efficient for the main crop cultivated in the dry season suggests that respondents who are engaged in high return cash crop cultivation (OFC) are willing to pay more for improved irrigation services. The positive relationship with total family income, farm size and cultivation of OFC crops indicates the need for strategies to commercialize the small farm agriculture which will increase the capacity and willingness of the farmers to mobilize more resources toward irrigation system maintenance. Therefore, intensification of land use through promotion of appropriate cropping systems and integration of different components such as livestock, bee keeping and agro- processing into the farming system is more suitable.

Negative co-efficient for the DC location (head end or tail end) indicates that the tail end farmers want to provide more for improved services. The higher value of willingness to pay by the tail end farmers is due to their comparatively higher agricultural income and the greater assurance of water after BWA compared to the past.

Table 5.9: Estimation of the Regression Model

Variable Co-efficient Statistics Constant 23.534 4.65 Income 0.221 3.34*** Yala crop 3.760 1.80* Low land extent 3.924 3.43*** Age 0.005 0.057 Education 0.213 0.783 DC location -4.003 -1.924** Type of farming -3.467 -0.971 Labour contributions -0.474 -2.388*** (meetings) Labour contribution 0.278 0.392 (maintenance) ***significant at 1 percent level, **significant at 5 percent level, *significant at 10 percent level

CHAPTER SIX

Findings, Recommendations and Policy Implications

6.1 Major Findings

1. Demographic and Socio-economic Features

Average family size varies from 3.6-4.4 with an average of 3.9. About 52 percent of the families have 4-5 members, while 35 percent have 1-3 members. The masculinity ratio is almost 100. The majority of the farmers (over 60 percent) are in the age group of 40-60 years, which reflects the low level involvement of youth in Agriculture. About a quarter of the population is in the dependency age category.

About 96 percent of the population had some degree of education, with 58 percent of the population having an education up to GCE ordinary level. Education for about 27 percent of the farmers was limited to primary level education and 4 percent were illiterate. Over two third of the total labour force is employed. However, the economic dependency ratio of the study area is 77 percent, which is relatively a high figure.

Farming is the main source of income for 66 percent of the total employed labour force. Average monthly income from farming is about Rs 9,400 in the head end and Rs 9,700 in the tail end areas. However, the average monthly income from non- farming activities is Rs 11,500 and 11,400 in head end and tail end areas respectively.

2. Institutional Mechanisms Adopted

A five-tier institutional mechanism has been adopted for the implementation of the BWA programme. The main organizations functioning in this five tier institutional mechanism are the FCG at the FC level, the DCO at the DC level, unit level support committee at the unit level, the BLFF and sub project committee at the block level and the PMC at system level.

The strength and performance of the DCOs in control structure management, organizational management and water management demonstrate the sustainability of the DCOs, which is the key organization working at the farmer’ level. The extent of support given by the line agency ((MASL) in the joint management of irrigation system and co-operation it extends to the farmer groups indicate the smooth functioning of the institutional mechanism at the apex level. However, the performance of the line agency is highly dependent on the financial allocation made to them by the central financial agency and the equality of leadership available at the agency.

3. Irrigation System Performance

Irrigation system performance was assessed in terms of water supply performance, the performance in productivity and improvement in the farmer’ income. According to

the findings, the performance of water supply after the BWA has improved considerably in terms of gross water quota allocated during both maha and yala at the block levels and the main canal level. The block level water duty has dropped at an average of 24.6 percent and 37.3 percent during the maha and the yala respectively. Similarly the canal water duty has declined by an average of 27 percent and 35 percent respectively. The extent cultivated during the dry seasons after the introduction of the BWA, has increased at an average of 52 percent, with an increase of annual cropping intensity by 10.7 percent.

Area irrigated per unit of water too has marked an increase by 34 percent and the water productivity values have risen by 30 percent following the BWA programme. The extent under OFC has shown a 52 percent upward trend with the implementation of the BWA. The estimated Gini ratio for farm income in the head and the tail ends is 0.12 and 0.22 respectively indicating more equality in the agricultural income distribution under the BWA. On an average, the farmers have achieved 93 percent and 80 percent of the technical efficiency in paddy cultivation during the maha and the yala respectively in the year 2004/2005.

4. Pros and Cons of BWA

The major benefit realized by the BWA as perceived by 83 percent of the beneficiaries is the drop in water conflicts among the farmers. In addition, the farmer’ attitudes have shown a positive change so far as water use and irrigation system management are concerned. Majority of the farmers are making efforts to save water and they have gained awareness that, the excess water is bad for their crop and wasting of water would cost them in the next season. The findings also highlight the increase in the extent of cultivation and income and a drop in crop failures.

The major constraint of the BWA as reported by 25 percent of the farmers is the difficulties of receiving night time water rotations. Inability to cultivate the desired crops, and the full extent in the yala seasons, and longer gaps between some water rotations especially in the latter parts of the seasons are some other snags experienced by the farmers.

5. Sustainability of Turnover Irrigation Infrastructure

The farmers are mobilizing resources in kind and cash for system O&M. The value of mobilized labour ranges from Rs 1,000-1,775 per ha per year and the value of cash mobilization to DCOs is about Rs 500 per ha per year. This is a significant achievement compared with the past failed attempts of mobilizing resources from the farmers in terms of cash. However, the level of resources mobilized for system O&M is a little lower than the estimated requirement for sustainable maintenance. Nevertheless, over 90 percent of the farmers agreed that, the number of broken structures of bund, of broken pipe outlets, of gates and incidences of silted canals have decreased after the BWA and the amount of resource mobilization for system O&M by the DCOs is sufficient to ensure physical sustainability of infrastructure. This is understandable on account of the present good condition of the infrastructure

due to recent rehabilitation under the MRRP. About 95 percent of the farmers are also of the view that, line agency officials always or often under take routine maintenance at the DC level.

In addition to the current level of labour and cash mobilization by farmers for system O&M, the farmers are willing to pay Rs 599-890/ ha/annum for improved irrigation services considering the possibility of future system deterioration and maintenance cost escalations. The WTP value has positive and significant relationship with total family income, size of low land, amount of labour contributed by the farmers for DCO meetings, location of farm in DC canal (head or tail) and cultivation of cash crops over paddy.

6.2 Recommendations and Policy Implications

1. The approach and methodology adopted for the implementation of the BWA programme in the Mahaweli H area is an encouraging experience for the rest of the major irrigation schemes, which are having suitable infrastructure and experiencing water allocation problems. In addition to the hardware aspects of development by the MRRP, the software parts of development activities including the institutional development, the private sector involvement in forward purchase arrangements and collaborative supports from other government agencies are noteworthy features of this water management programme.

2. A comprehensive devolution policy for the irrigation sector with clear arrangements for the roles and responsibilities and an assured water supply would provide tangible benefits which can encourage the farmers to invest in the long term sustainability of their irrigation infrastructure. Allocation and sharing of responsibilities should be with clear policy demarcations on who is responsible for the specified activities and who can be made accountable in case of a breakdown of these activities.

3. In order to realize the benefits of the devolution policy, the implementers should provide attention to create self-reliance local organizations in terms of social, financial and technical perspectives for better local management.

4. The WTP expressed by beneficiaries should be taken into consideration in future water policy formulation to ensure sustainability of the system allowing for future maintenance cost escalations and to avert increasing trend of system deterioration.

5. The positive relationship of WTP with the total family income, the irrigated land extent, the type of crop cultivated and the location of irrigation canal indicate the direct linkages with commercialization of small farm agriculture.

6. The irrigation management transfer should widen up the focus beyond mere cost recovery, but in helping and creating an environment to generate the necessary development impulses for increasing agricultural productivity and

farmers’ income, establishing market linkages for agricultural inputs and outputs and considering the scale of production.

7. The Government should take note of the achievement of the BWA in improving water use efficiency and increasing the extent cultivated in the dry season and study the feasibility of duplicating this concept in other irrigation systems where water scarcity and water wastage is limiting the system performance, provided the necessary infrastructure is available for the implementation of the BWA.

REFERENCES

Aheeyar, M.M.M. 1997, “Participatory Irrigation Management and Sustainability of Irrigation Infrastructure: A Case of Irrigation Management Turnover in Sri Lanka”, Proceedings of the International Conference on Large Scale Water Resources Development in Developing Countries: New Dimensions of Problems and Prospects, October 20th-23rd, Kathmandu, Nepal. Central Bank of Sri Lanka, 2004, Economic and Social Statistics of Sri Lank- 2004. Central Bank of Sri Lanka, Colombo, Sri Lanka. Department of Census and Statistics, 2005, Paddy Statistics – 2004/2005 Maha Season, Department of Census and Statistics, Colombo, Sri Lanka. Gunaratne, B. 2003, Bulk Water Allocation Concept for Improving Water Productivity- A Pilot Study in Mahaweli System H. In R.P. De Silva (ed), Water Resources Research in Sri Lanka, Postgraduate Institute of Agriculture, University of Peradeniya, Peradeniya. Hemaratne, H.P., Abeygunawardane, P. and M.W. Thilakaratne, 1991, Income Inequalities and Different Levels of Water Availabilities in Irrigation Schemes: The Case of Minipe, Tropical Agricultural Research, Vol.3: 290- 301. Hufschmide, M.M. et-al, Environment, Natural Systems and Development: An Economic Valuation Guide, Johns Hopkins University Press, Baltimore. Mahaweli Authority of Sri Lanka, 2005, Statistical Bulletin-2005, available at http://www.mahaweli.gov.lk/other/2005 Mahaweli Authority of Sri Lanka, 2000, Mahaweli Statistical Hand Book-2000. Planning and Monitoring Unit, Mahaweli Authority of Sri Lanka, Colombo, Sri Lanka. Mahaweli Authority of Sri Lanka, 1999. Mahaweli Statistical Hand Book -1999, Planning and Monitoring Unit, Mahaweli Authority of Sri Lanka, Colombo, Sri Lanka Mathmaluwa. S, 2003, The Effect of Irrigation Management Transfer on the Productivity of an Irrigation System, A study on the Madatugama block of Mahaweli system H. Thesis submitted in partial fulfillment of the degree of Master of Science in Soil and Water, Waganingen University, The Netherlands. Ministry of Irrigation and Water Management, 2003, Final Interim Report on Comprehensive Plan for Kala Oya Basin. Mahaweli Authority of Sri Lanka, Ministry of Irrigation and Water Management, Colombo, Sri Lanka. Molden, D.J. et-al. 1998, Indicators for Comparing the Performance of Irrigated Agricultural Systems, Research Report No, 20, International Water Management Institute, Colombo.

Morgan, J. 1962. On the Measurement of Income Inequality, The Review of Economics and Statistics, Vol. XLIV, No.3. Raby,N. and D.J.Merrey, 1989, Professional Management of Irrigation Systems: A Case Study of Performance Control in Mahaweli System H, Sri Lanka. IIMI country paper, Sri Lanka No. 1, International Irrigation Management Institute, Colombo. Randall, A. 1981, Resource Economics - An Economic Approach to Natural Research and Environmental Policy, Grid Publishing, Inc., Columbus, Ohio. Ratnayake, R.M.G.K.B. 1986. An Empirical Assessment on the Relation of the Scale of Irrigation Systems on Quantity and Equity of Farm Production, Thesis submitted for the partial fulfillment of Bachelor of Science in Agriculture, Department of Agricultural Economics, Faculty of Agriculture, University of Peradeniya, Peradeniya Samad, M. and D. Vermillion, 1999, Assessment of Participatory Management of Irrigation Schemes in Sri Lanka: Partial Reforms, Partial Benefits, Research Report No. 34, International Water Management Institute, Colombo. Silva, M. 1986. The Evolution of Policies in Sri Lanka Relating to the Covering of Water Charges from Farmers for Irrigation Service, International Irrigation Management Institute, Colombo TEAMS, 2003, An Independent Audit to Evaluate the Performance of Madatugama block in System H under MRRP Programme, Report submitted to Mahaweli Authority of Sri Lanka, T.B. Jayah Mawatha, Colombo 10 (unpublished). Upawansa, W.J.I. and P. Abeygunawardana, 1993, Economic Value of Irrigation Water in Dewahuwa Irrigation Scheme, Tropical Agricultural Research, Vol. 5, pp 181-193 Wickramaratne, H.A, 1986, Water Management in System H of the Mahaweli Project, Diyawara 3 (Aug):1-6, Agrarian Research and Training Institute, Colombo. Whittingten, D. et-al, 1983; Cost Recovery Strategy for Rural Water Delivery in Nigeria, Urban Development Working Paper, World Bank, Washington D.C.