Sri Lanka - Third Mahaweli Ganga Development Project

Document of The World Bank F ILE CtIY FOR OFFICIAL USE ONLY Report No. 3128-CE STAFF APPRAISAL REPORT SRI LANKA MAHAWELI GANGA DEVELOPMENT PROJECT III May 26, 1981 South Asia Projects Department Agricultural Division A This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENT US$1 = Rs 18.0 Rs 1 US$0.0556 UNITS AND EQUIVALENTS 1 millimeter (mm) = 0.0394 inches (in) 1 meter (m) = 3.2808 feet (ft) 1 kilometer (km) = 0.6214 miles (mi) 1 hectare (ha) = 2.4711 acres (ac) 1 kilogram (kg) = 2.2046 pounds (lb) 1 mstric ton (ton) = 1.0161 long tons (1 ton) 1 m /sec = 35.3357 cubic feet/sec (cusec) 1 bushel of paddy 20.8655 kg PRINCIPAL ABBREVIATIONS USED AO - Agricultural Officer ARTI - Agricultural Research and Training Institute ASC - Agrarian Service Center BC - Bank of Ceylon CB - Central Bank DPM - Deputy Project Manager ERR - Economic Rate of Return FD - Food Department FSS - Food Stamp Scheme GOSL - Government of Sri Lanka GPS - Guaranteed Price Scheme ICB - International Competitive Bidding LCB - Local Competitive Bidding LHG - Low Humic Gley Soils MASL - Mahaweli Authority of Sri Lanka MDB - Mahaweli Development Board MEA - Mahaweli Economic Agency MMD - Ministry of Mahaweli Development MPCS - Multipurpose Cooperative Society NPW - Net Present Worth OCC - Opportunity Cost of Capital OER - Official Exchange Rate PB - People's Bank PMB - Paddy Marketing Board PV - Present Value RBE - Red Brown Earths SCF - Standard Conversion Factor SER - Shadow Exchange Rate SLCC - Sri Lanka Cashew Corporation SMS - Senior Subject Matter Specialist SV - Switching Value T&V - Training and Visit WFP - World Food Programme WMA - Water Management Assistant FISCAL YEAR January 1 - December 31  SRI LANKA FOR OFFICIAL USE ONLY MAHAWELI GANGA DEVELOPMENT PROJECT III Table of Contents Page No. I. SECTORAL BACKGROUND ...... .................. 1 General ..........................1.......................... Agricultural Sector ....... ................. 2 Irrigation Subsector ...... ................ . 3 The Mahaweli Ganga Development Program .... ............ 4 The Bank Group's Role in Mahaweli Ganga Development ........ 5 Review of Implementation Progress of Mahaweli Ganga Development Project II .................................. 6 II. PROJECT AREA ............................................... 6 General .................................................... 6 Climate .................................................... 7 Topography and Drainage .................................... 8 Soils and Land Suitability for Development .... ............. 8 Agricultural Supporting Services. 8 III. THE PROJECT ................................................ 10 General .................................................... 10 Principal Project Features ................................. 10 Description of Project Works, Facilities, and Programs 11 Cashew and Fuelwood Plantations ............................ 19 Water Supply and Demand .................................... 19 Monitoring and Evaluation .................................. 20 Technical Assistance ....................................... 21 Cost Estimates ............................................. 22 Implementation and Settlement Rate ......................... 23 Financing .................................................. 24 Procurement Under the IDA Credit ........................... 25 Disbursement ............................................... 25 Accounts and Audits ........................................ 26 Environmental Impact ....................................... 26 IV. ORGANIZATION AND MANAGEMENT .28 Mahaweli Authority of Sri Lanka ............................ 28 Mahaweli Development Board .28 Mahaweli Economic Agency ................................... 29 Agricultural Supporting Services ........................... 30 Project Design and Construction ............................ 31 Monitoring and Evaluation .................................. 31 This report is based on the findings of an IDA mission that visited Sri Lanka in March-April 1980 comprised of Messrs. T.H. Yoon, S. Baker, A. Seager, W. Nickel, W. Peters, R. Goodland (IDA), R. Tekrony, R. Claringbould, and J. Christopher (Consultants). Representatives of the EEC also participated in project appraisal as a potential co-financier. This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  Table of Contents (cont.) Page No. Operation and Maintenance .................................. 31 Giranduru Kotte Irrigation Training Center and Farm ........ 32 V. PRODUCTION, MARKETING AND PRICES, FARM INCOME, AND COST RECOVERY ............................................ 33 Agricultural Development with the Project .... .............. 33 Future Development without the Project ..................... 34 Yields and Production ...................................... 34 Marketing and Pricing of Project Output .................... 35 Farm Income ................................................ 38 Cost Recovery .............................................. 40 VI. BENEFITS, JUSTIFICATION, AND RISKS ......................... 43 Overview ................................................... 43 Employment and Income Effects .............................. 43 Direct Foreign Exchange Effects ............................ 44 Benefits to the Economy and Economic Analysis .44 Project Risks and Sensitivity Tests ........................ 47 VII. RECOMMENDATIONS ............................................ 49 LIST OF ANNEXES ANNEX 1 Table I : Climatological Factors Related to Estimating Crop Water Requirements Table 2 : Commanded Land Class Inventory in System C Table 3 : Expected Use of Commanded Land in System C Table 4 Irrigation and Drainage System Summary-Zones 3 to 6 of System C Table 5 Social Infrastructure Buildings Table 6 Social Infrastructure Buildings (Zones 3 to 6 only) Table 7 : Management and Operational Infrastructure Table 8 : Facilities to be Provided at the Agricultural Training Center and Farm Table 9 Settlement Assistance and Monitoring Programs Table 10 Estimated Flow of the Mahaweli Ganga at Minipe Table 11 Irrigation Water Requirements for Flooded Rice Table 12 Estimated Total Yearly Water Requ rements for System C at Head of Transbasin Canal (Mm ) Table 13 Estimated Annual Operation & Maintenance Costs Table 14 : Cost Estimate Summary Table 15 : List of Equipment and Vehicles Table 16 : Schedule of Expenditures Table 17 : Proposed Allocation of the Credit Table 18 Estimated Schedule of Disbursements  - iii - Table of Contents (cont.) Chart 1 : Schematic Future Cropping Pattern Chart 2 : Comparison of River Flow at Minipe with Project Irrigation Diversion Chart 3 : Implementation Schedule Chart 4 : Procurement Schedule for Civil Works Procurement Schedule for Equipment and Materials Chart 5 : Organization Structure at Project Level Chart 6 : Organization Structure at Block and Unit Level Chart 7 : Schedule and Cost of Staff and Farmer Training ANNEX 2 : Terms of Reference for Consulting Services ANNEX 3 Table 1 Paddy Area and Production; Rice Availability, Procurement, and Distribution, 1970-79 Table 2 : Project National Rice Balances Table 3 : Summary of System C Cropped Area, Yields, and Production ANNEX 4 Table 1 : Prices Used in Financial and Economic Aanlyses Table 2 : Crop Input Requirements Table 3 : New Settler Farm Budget (Paddy - Paddy Rotation) Table 4 : New Settler Farm Budget (Broadcast Paddy - Upland Annual Crops Rotation) Table 5 : New Settler Farm Budget (Perennial Crops) Table 6 : Farm Budgets of Existing Farmers with Irrigated Holdings in Major Tank Schemes (Paddy - Paddy Rotation) Table 7 : Farm Budget of Existing Farmer With Irrigated Holding in Minor Tank Scheme Table 8 : Economic Cash Flow Table 9 : Relative Importance of Project Cost and Benefit Streams with Switching Values Chart 1(a): Locus of Present Values of Total Project Costs and Benefits Chart l(b): Locus of Total Project Cost and Benefit Switching Values ANNEX 5 : Related Documents and Background Papers in Project File MAPS IBRD 13477R1 - Sri Lanka Mahaweli Ganga Development Project: Existing and Proposed Irrigation Schemes IBRD 15154 - Sri Lanka Mahaweli Ganga Development Project   SRI LANKA MAHAWELI GANGA DEVELOPMENT PROJECT III Staff Appraisal Report I. SECTORAL BACKGROUND General 1.01 Sri Lanka is endowed with relatively favorable resources and diverse agro-climatic conditions that permit wide ranging agricultural pur- suits. It also possesses a highly literate population and has achieved considerable success in curtailing population growth. Sri Lanka comprises a total area of about 65,600 sq km, with an estimated population of about 14.9 M (mid-1981). While population has been increasing at about 2% p.a. in recent years, emigration has reduced this growth to about 1.7%. Per capita GDP is about US$280, with only about 25% of the population estimated to have annual incomes below the poverty level of US$105 per capita. 1/ About 76% of Sri Lanka's inhabitants live in rural areas and are principally dependent directly or indirectly on agriculture. 1.02 In contrast to its considerable social progress, Sri Lanka's eco- nomic performance has until recently been characterized by low output and employment growth. Real per capita GDP grew at 2.1% p.a. during the 1960s, but subsequently declined to 1.3% p.a. during the 1970-76 period; aggravated by deteriorating terms of trade, per capita gross national income rose between 1970-76 by only 0.9% p.a. Slowdown in economic growth in the 1970s was attri- butable to a combination of factors including a series of partial failures of monsoon rains, inadequate investment, ineffective and inappropriate government policies toward growth and investment, and a sharp rise in the cost of imported food and fuel. This disappointing economic performance contributed to increased open unemployment. 1.03 With the new government elected in 1977 came broad policy changes designed to address economic stagnation. The new Government of Sri Lanka (GOSL) identified its objectives as reviving the economy and increasing employment through: (a) greater capacity utilization in the productive sectors; (b) stimulation of savings and investment; and (c) efforts to boost exports and foodgrain import substitution. Major changes included exchange rate unification at a depreciated rate, import trade liberalization, interest rate increases to encourage savings, abolition of price controls on most commodities, fiscal reforms, and upward revision in agricultural procurement prices. Between 1978-80, GOSL took further steps to complement earlier reforms including: (a) moves to eliminate public sector import monopolies with respect to food and fertilizer; (b) establishment of new autonomous authorities to carry out high priority programs; and (c) initiation of a number of lead investment schemes including the Accelerated Mahaweli Ganga Development Program, new housing and urban development, and establishment of a free trade zone north of Colombo. 1/ Preliminary World Bank estimates, mid-1980.  1.04 Sri Lanka's economy reacted vigorously to these new policies. Real GDP grew by an impressive 8.2% in 1978, by 6.3% in 1979, and by 5.8% in 1980. In the new economic climate, the private sector responded well. Accelerated growth and investment helped ease the pressures of open unemployment. Never- theless, by 1980 it became clear that GOSL's strategy carried certain risks. Overheating of the economy, stemming largely from the ambitious public invest- ment program was reflected in a sharp increase in inflation (implicit 1980 GDP deflation factor of 24%) and the emergence of a severe balance of payments problem necessitating a sizeable drawdown of reserves in 1980. Domestic inflation has been aggravated by large and frequent price adjustments geared to remove subsidies, administered wage increases, heavy demand on the limited capacity of the construction sector, deficit financing, and adverse external price developments. GOSL has begun to address these problems, most notably through a recent substantial reduction in investment allocations in the 1981 budget and on-going scrutiny for further cuts in future expenditures and for increased resource mobilization. The Agricultural Sector 1.05 Agriculture plays a major role in the Sri Lanka's economy, account- ing in 1980 for about 28% of total GDP, over 50% of total employment, about 60% of export earnings, and about 29% of government revenues. In addition, much of the activity in the manufacturing, transport, and service sectors is related either to supply of agricultural inputs or processing of agricultural output. In rural areas, dependence on agriculture is almost total with over 95% of the population either directly or indirectly involved in agriculture. Sri Lanka's climate is suitable for year-round cultivation, and variations in precipitation, topography, temperature, and soils enable production of a wide range of crops. Out of about 2.24 M ha under permanent cultivation (34% of land area), paddy accounts for 0.65 M ha; tree crops (coconut, tea, and rubber), 0.93 M ha; and other perennial crops, 0.04 M ha. The remaining 0.62 M ha comprises mixed rainfed farming, mostly as small gardens around homesteads. An additional 1.0 M ha are under shifting slash and burn cultiva- tion, locally known as chena. 1.06 Policy changes recently introduced by GOSL in the agricultural sector were a response to the disappointing growth recorded in the early and mid-1970s. Average growth, 3.4% p.a. during the 1960s, had declined to only 1.9% p.a. over the 1970-77 period. Initial response to new policy initiatives was encouraging. Recovery of agriculture actually started in 1977 with a 10.4% increase in agricultural production stemming from a 33.2% boost in paddy production. This fine performance was followed in 1978 by a further 5.4% rise in agricultural production--the first sustained increase since the late 1960s. Paddy, increasing by 12.6%, accounted for most of this increment and was undoubtedly helped by a 21.2% adjustment in the publicly guaranteed (GPS 1/) price in November 1977. Growth in agricultural production declined to only 1/ Guaranteed Price Scheme.  - 3 - 2.0% in 1979 mainly due to drought but rose to 3.1% 1/ in 1980 despite con- tinued abnormally low rainfall in the yala season. 2/ 1.07 GOSL had hoped to achieve an ambitious target of 2.21 M tons of paddy in 1980, 15.3% above the record 1979 crop of 1.92 M tons. However, this appears to have been somewhat optimistic given the severe delay in the SW monsoon and its resulting impact on the 1980 yala crop. 2/ Based on preliminary statistics, the total 1980 paddy crop is estimated at about 2.13 M tons. Generally, favorable weather during the maha season, the overall liberalized environment, and the initial impact of the Mahaweli Program were chiefly responsible for such success in paddy output. To maintain producer incentives, the GPS paddy price was raised by 31% during the 1980/81 maha season, partly to offset fertilizer price increases of 55-136% introduced in January 1981. Irrigation Subsector 1.08 Annual mean rainfall in Sri Lanka's wet and intermediate agro- climatic zones, located in the southwest quadrant of the island, ranges from about 2,200 to 3,900 mm and is fairly well distributed throughout the year. Paddy can be grown in these zones under rainfed conditions, not only in the maha season, but in some areas in the yala as well. Irrigation development in the wet zone has been limited primarily to run-of-the-river diversion. In contrast, annual rainfall in the dry zone, averaging about 1,300 mm, is concentrated in the maha season and displays considerable year-to-year vari- ability. Irrigation, therefore, is essential for cultivation in the yala season, and some supplemental irrigation is often necessary for the maha crop. The dry zone, containing about 285,000 ha of irrigated land, accounts for about 78% of the total irrigated area. Traditional irrigation schemes are based on storage tanks designed to provide supplemental irrigation to the maha crop, with any residual water used for limited dry season cropping. There are a total of about 180 such schemes having a storage capacity in excess of 2.5 M m (irrigating nearly 162,000 ha) and a large number of smaller schemes. Traditionally, maintenance of these schemes has been the responsibility of farmers, with only major headworks and main canals under government purview. Water is allowed to flow continuously, running from field to field with only limited development of tertiary channels. Not only have most schemes deterior- ated, but initial designs were also highly wasteful. With proper operation, improved water management practices, and introduction of appropriate cropping systems (in particular to optimize use of rainfall and to diversify into non-paddy crops on more permeable soils), substantial increases in cropped areas could be attained with existing water supplies. Therefore, under the ongoing Tank Irrigation Modernization and Mahaweli II Projects (Crs. 666-CE and 701-CE), Rural Development Projects I (891-CE) and II (1079-CE), and the 1/ Provisional estimate. 2/ Maha is the northeast monsoon season (October-February); yala is the southwest monsoon season (April-September).  -4- proposed Village Irrigation Rehabilitation Project 1/, IDA is providing support for improved water management and a prototype for future development of other dry zone tanks. For the longer-term, GOSL is emphasizing expansion of irrigated agriculture into the dry zone through development of Sri Lanka's major river basins, such as the Mahaweli Ganga Basin. The Mahaweli Ganga Development Program 2/ 1.09 The basic objectives underlying the Mahaweli Ganga Development Program stem from a need to overcome three major problems facing Sri Lanka: (a) unemployment; (b) excessive imports of food and other agricultural pro- ducts; and (c) a shortage of power for industrial development and rural electrification. About 225,000 farm families would be settled on newly developed land, and over a million people would be employed in agricultural and allied pursuits. Agricultural output would reduce the need to import rice and other food items for which the foreign exchange bill was about US$425 M in 1980, or about 40% of export earnings. 1.10 The Program is based on a Master Plan prepared jointly by a UNDP/FAO team and Sri Lankan engineers in 1965-68. The Plan envisaged development of about 365,000 ha of irrigable land in the dry zone, as well as about 500 MW of hydropower. It was divided into three phases, each including several projects for stepwise implementation over a 30-year period. Implementation of the Plan began in 1970 (para 1.13). Late in 1977, however, GOSL realized that at the current rate of construction and population growth, the Program would not have much near-term impact in mitigating the unemployment problem and serious food and power shortages, and that it should be accelerated. An Implementation Strategy Study was carried out with financing from the Netherlands to: (a) review and re-evaluate the UNDP/FAO Master Plan and each of its separate projects; (b) examine the implications of various alternative phasings of the feasible projects; and (c) recommend a specific action plan taking into account technical, financial, economic, institutional, and manpower constraints. While the Study was underway, GOSL selected five major projects comprising five dams and hydropower plants with a total potential capacity of 500-600 MW of power and downstream development of about 112,000 ha to be included under its Accelerated Program. Donors were approached for support for different components of the program and responded promptly. Meanwhile, the Study, which was completed in late 1979, concluded that (a) the base costs of the overall Program had been greatly under-estimated; (b) two of the five projects (Randenigala and Moragahakanda) were not required for irrigation in the Accelerated Program, as the Kotmale, Victoria and the Maduru Oya Dams 1/ Board presentation scheduled for June 1981. 2/ 'The Mahaweli Ganga Development Program or Scheme, as shown in the UNDP/ FAO Mahaweli Ganga Master Plan of 1968, includes about 365,000 ha of net irrigable land situated in the Mahaweli Ganga and Maduru Oya Basins (192,000 ha) and the North-Central part of Sri Lanka (173,000 ha). The Accelerated Mahaweli Development Program encompasses about one-third of the Scheme's net irrigable area, representing Systems A, B, C, D-1, D-2, E, F, and G.  - 5 - would provide sufficient regulation to irrigate the entire Accelerated area; (c) power demand studies confirmed that substantial additions to existing capacity were needed (later feasibility studies justified proceeding with Randenigala on the basis of power alone). 1.11 GOSL accepted these conclusions and revised the Accelerated Program accordingly. Initial works of the Program (1980-85) were confined to four headworks projects having a total installed capacity of about 500 MW (Kotmale, financed partly by Swedish import support funds; Victoria, by a L 100 M grant from the UK; Maduru Oya, by a Canadian loan of C$76 M; and Randenigala, by a DM 400 M loan from the Federal Republic of Germany). Downstream development was to be limited to about 60,000 ha in Systems B and C together with the Minipe Anicut, the Right Bank Transbasin Canal, and the Ulhitya and Ratkinda Reservoirs. The US is providing finance for downstream development of the Maduru Oya Project (System B) (US$90 M) and has financed an in-depth study of the environmental impact of the Accelerated Program (paras 3.48-3.51). Canada and the UK financed feasibility studies of Systems B and C, respectively. The EEC is supporting development of Zone 2 in System C, while the Asian Develop- ment Bank is providing assistance to plan and construct roads in the Accele- rated Program area. Under a technical assistance credit, IDA and the EEC are financing consultancy services to review designs and tender documents for the Transbasin Canal and for the detailed designs and engineering of irrigation infrastructure in System C (para 3.38). Even though the Bank Group has assisted in Mahaweli development over the last decade (para 1.13), the pro- posed project in System C would mark its first area development effort under the Accelerated Program. 1.12 The overall costs of the Program have risen substantially during the past years, due to rapid domestic and international inflation and to the increased costs associated with more difficult site conditions for the Kotmale project than originally estimated. GOSL responded to this by making various adjustments (such as reducing the height of Kotmale Dam and deferring downstream development on the right bank of System B). Nevertheless, expen- ditures on the program are projected at US$1.6 billion in the 1981-85 period, or about 44% of projected resources for the entire public investment program. Such an ambitious program will strain available financial and real resources, particularly in 1982 and 1983. Financial constraints can be minimized to the extent that donors step up assistance levels for the Program and GOSL increases its own domestic resource mobilization efforts. Some donors have already moved in this direction, and GOSL proposes to take measures to raise domestic resources. The Bank Group's Role in Mahaweli Ganga Development 1.13 With Bank Group assistance, initial construction in the Program began in 1970 (Cr. 653-CE and Ln. 174-CE) with diversion headworks at Polgolla on the Mahaweli Ganga and at Bowatenna on the Amban Ganga, including a 40 MW hydro-power plant at Polgolla. The project was completed in 1978. This project provided an improved water supply to about 52,650 ha of existing irrigated land and new irrigation to about 29,000 ha in System H. A project completion report for Mahaweli I indicates that the project was well designed and is fulfilling its objectives. Under the second IDA credit (Cr. 701-CE),  -6- with co-financing from Canada, the Netherlands, UK, US, and EEC, construction work is in progress on irrigation infrastructure for about 60% of this 29,000 ha newly irrigated area. 1/ The project also includes settlement assistance and agricultural extension and support services. Under a technical assistance credit (Cr. 979-CE), IDA, with EEC co-financing, is providing technical assis- tance for a study to determine optimal use of surplus water from the Mahaweli Ganga. In addition to financing earlier Mahaweli projects, the Bank has also assisted GOSL in its efforts to accelerate the Program, principally through external assistance coordination for project preparation and implementation. Review of Implementation Progress of Mahaweli Ganga Development Project II 1.14 About 85% of irrigation system works and 40% of social infrastruc- ture were completed by April 1981. The remaining works are planned to be completed by end 1982. GOSL has agreed to provide sufficient funds from within its Mahaweli Program allocation to ensure successful completion of the Project by end 1982. Actual disbursements from the IDA Credit were about US$5.1 M at the end of April 1981, compared with an appraisal estimate of about US$16.0 M. The rate of disbursements achieved understates actual implementation progress principally because of the devaluation of the rupee since appraisal, resulting in lower foreign exchange values for civil works for which the IDA Credit and Netherlands and EEC loans are mainly disbursed. Since no concomitant adjustments in disbursement percentages were made, disbursements against foreign exchange equivalents of civil works costs have been much lower than anticipated. IDA has agreed to a recent GOSL request to increase disbursement percentages for civil works. Also, GOSL implementing agencies have been successful in effecting economies in construction costs and by using previously acquired equipment, which resulted in lower disbursements than anticipated. In view of the above, some funds from the IDA Credit may not be utilized and may be cancelled. The recovery of land improvement and O&M costs of the irrigation system in the Project was to be carried out under the Sale of State Lands Law No. 43 of 1973 and the Land Betterment Law of 1977. However, neither of these laws could be effectively implemented as a series of technical and procedural problems arose which defied resolution. GOSL has again provided for the recovery of costs of O&M through water charges under the Irrigation Ordinance No. 32 of 1946. It has also provided for the transfer of land to settlers upon payment for costs of land improvements under the Land Development Ordinance of 1936 as amended by an act in April 1981. The technical and procedural arrangements for the implementation of these laws are now being made and are expected to be finalized by September 30, 1981. These laws will be implemented nationwide and will cover all settlers within the Mahaweli Ganga Development Project II area. II. PROJECT AREA General 2.01 The project area comprises the northern half of System C (Map 13477R1), totalling about 66,700 ha situated in the east-central sector of 1/ GOSL independently undertook development of the remaining 40%.  - 7 - Sri Lanka. System C comprises about 24% of the area to be irrigated under the Accelerated Mahaweli Program and represents about one half of the down- stream development area to be served by the Victoria Dam Project, a principal component of the Accelerated Program (para 1.10). The proposed project would cover specifically Zones 3-6 of System C and include a net irrigated area of 18,500 ha out of the 66,700 ha gross System C area. The project would also include activities to facilitate the development of System C as a whole (para 3.01). Located along the right bank of Mahaweli Ganga just downstream from the Minipe Anicut, System C extends about 70 km north and, at its widest part, 20 km east from the Mahaweli River (Mahaweli Ganga). The Right Bank Transbasin Canal forms the eastern boundary of the project area to its mid- point at Ratkinda Oya Reservoir (Map 15154). Situated in Badulla, Polonnaruwa, and Amparai Districts, System C traverses both the intermediate and dry agro- climatic zones. At present, permanent settlements have been established in only the southern part of the System, principally around three rehabilitated major tanks, minor village tanks, and small lift irrigation schemes. 2.02 Gradually over the last 25 years, unsanctioned encroachment has destroyed a large portion of the forest cover north of these settlements. As a result, by 1979 only the northwest third of the area remained under relatively dense mixed evergreen forest, all located within the project area. However, traces and roads have since been cut into the forest to enable soil and topographic surveys and timber extraction to be undertaken, thereby making the area increasingly accessible to encroachers and illicit timber removal. Much of the 29,000 ha of forested area in the northern part of System C is currently being cleared of marketable timber in preparation for settlement and agricultural development. 2.03 Most of the northern half of System C has no infrastructure or resident population. The southern half is more developed and contains about 95% of the estimated 30,000 inhabitants (1979) within the System. About 80% of the population is engaged directly in agriculture. Such activities are basically confined to Zone 1 where about 1,770 ha of irrigated paddyland is served mainly from several major tanks and 1,800 ha is served from small scattered tanks. In Zones 2-6, another 500 ha of rainfed and partially irrigated paddy and about 3,000 ha of chena are cultivated each year. Recent pressure on land has been such that areas are being brought under chena again before forest growth regenerates from the previous slashing cycle. Communica- tions, health, education, and technical services are concentrated around the town of Mahiyangana at the southern end of the System. Most farm families outside the areas serviced by tanks earn less than Rs 3,600 p.a. (US$200), the eligibility level for the national Food Stamp Scheme (FSS). Climate 2.04 System C lies partly in the intermediate climatic zone and partly in the dry zone. The project area (Zones 3-6) is located principally in the dry zone. Annual mean rainfall varies from about 2,250 mm in the southern end of System C (Zones 1-2) to about 1,750 mm in the north. About 60% of this rainfall occurs during the NE monsoon period of November-February (maha), and the balance, during the SW monsoon period of May-September (yala). The unpredictability of rainfall in the maha makes irrigation necessary to ensure high and stable yields. Crop cultivation during the yala is almost entirely  - 8 - I dependent on irrigation. The climate is tropical with maximum temperatures ranging from 30 to 36 0C and minimum temperatures, from 20 to 24 0C, making year-round cropping feasible (Annex 1, Table 1). Winds are gusty in the yala season. Topography and Drainage 2.05 Potentially irrigable land in the northern half of System C is scattered in irregular blocks. Area lands are undulating and incised with numerous gullies. North-south inclined ridges, some rather steep and high, are prevalent and result in a sharp distinction between irrigable soils in valleys and shallow, non-irrigable, sloping land on sides of ridges. Land slopes in this area range from the almost flat bottom to slopes of 6-10% on sidehill areas. Slopes in major blocks to be developed range from 0-6%, with an average of 1-3% over most of the area. Because of the diverse erosional pattern, shallow natural drainageways have developed in more steeply sloping areas. The coarser, freely draining soils located on higher lands are mostly underlain by bedrock at relatively shallow depths. The southern portion of System C (Zones 1-2), in comparison, is flatter and exhibits an average slope of 1-2%. Soils and Land Suitability for Development 2.06 System C soils consist mainly of two types: poorly drained Low Humic Gleys (LHGs) in flat valley bottoms, and relatively better draining Red Brown Earths (RBEs) in higher, more steeply sloping areas. Between these two types are intermediate, imperfectly drained soils. RBEs are subject to high erodability due to poor structure, and the majority of these are underlain with bedrock at depths ranging from 2-3 m, as are many of the imperfectly drained soils. Soil salinity, alkalinity, and other possible adverse proper- ties such as iron toxicity will not be problems for agricultural development considering (a) area rainfall levels and quality of water to be used for irrigation, and (b) drainage proposed. Available data indicate that soil acidity will not be significant. 2.07 Project area lands are predominantly more suitable for paddy than other crops because of the soil-water regimen expected to develop with forest removal and irrigation. Prevailing characteristics of shallow soil to drain- age or construction barrier (bedrock) preclude the irrigation of crops that must be "irrigated dry" because of the high cost of subsurface drainage required. Relatively small non-contiguous parcels exist, however, that would be suitable for such diversified cropping. These areas were not identified during feasibility studies, but would be determined during detailed investi- gations carried out during pre-construction activities. A land use inventory, based on earlier semi-detailed soil surveys and preliminary land use studies, is shown in Annex 1, Table 2. Agricultural Supporting Services 2.08 At present, no agricultural supporting services are available in Zones 3-6 or in the northern part of Zone 2. Zones 1-2 contain normal trad- ing and infrastructure facilities. Mahiyangana and, to a lesser extent,  - 9 - Alluterama are the only trading centers of significance. In these latter zones, fertilizers, pesticides, and seeds are readily available through Agrarian Service Centers (ASC) as well as from private traders. Banking facilities for short- and medium-term credit also exist; however, agricultural extension services are understaffed. The latter are scheduled to improve as implementation of with the IDA-supported nationwide Agricultural Extension and Adaptive Research Project (Cr. 931-CE) gains momentum. 2.09 Rural Credit. Institutional rural credit in Sri Lanka is disbursed principally by the two nationalized commercial banks, the People's Bank (PB) and the Bank of Ceylon (BC). 1/ PB operates through Multipurpose Cooperative Societies (MPCS) and their banking divisions, Cooperative Rural Banks. BC provides credit services through its main branches with representation main- tained at ASCs. National rural credit policy is established by a Rural Advisory Board chaired by the Governor of the Central Bank (CB). CB's Department of Rural Credit functions as the Board's secretariat and executive. PB maintains a branch office in System C only at Mahiyangana, and BC has representation there through the local ASC. 2.10 A long-standing problem of rural credit in general, and agricultural credit in particular, has been a high default rate, which was aggravated during the 1977/78 maha season (70%) when GOSL relaxed lending criteria. However, to remedy past trends, first, the CB guarantee for agricultural credit disbursed by PB and BC was withdrawn in 1978 forcing the banks to be more cautious about borrower creditworthiness and loan recovery. Second, GOSL launched an aggres- sive recovery drive nationwide, holding defaulters accountable for arrears and making repayment or rescheduling conditions for new credit. 2.11 With this recent crackdown came improved nationwide recoveries in 1979 (58% 2/), but also a drastic curtailment of rural lending. Total rural credit contracted from Rs 326.0 M in 1978 to Rs 106.1 M in 1979, within which crop cultivation loans fell from Rs 259.5 M in 1978 to Rs 55.9 M in 1979. The transition, while justified on budgetary grounds, was abrupt in the sense that the institutional credit system previously had been void of responsible credit supervision conducive to repayment discipline. However, a national rescheduling effort, part of the new recovery program, should allow future expansion of a sounder agricultural credit program, hopefully predicated on an improved credit delivery system and healthier farmer-bank relationships. 2.12 Cultivation loans for annual crops are generally provided for 6-10 months at 9.0% interest p.a. with a penalty of 5% charged on overdues. Such loans are granted according to established scales of finance for different crops and are provided partly in kind (inputs) with direct payment to suppliers and partly in cash. Longer-term credit is available to farmers for investments such as tractor purchase involving 80% bank financing, 14% interest, p.a., and three years repayment. In view of the sharp rise in inflation particularly 1/ The Hatton National Bank Ltd. supplies a small number of crop cultivation loans under the Comprehensive Rural Credit Scheme. 2/ As of December 1979. Recovery estimate based on cultivation loans (6-10 months terms) only.  - 10 - since 1978, agricultural on-lending rates are currently negative in real terms. GOSL, however, is taking steps to reduce inflation and thereby curb the impli- cit subsidy on such interest rates. Other institutional credit schemes of interest to prospective settlers under the project would include small-scale industrial credit (e.g., rice hullers, sewing machines, bicycle repair shops), house establishment credit (under consideration), and animal husbandry financing. III. THE PROJECT General 3.01 The proposed project would be the third irrigated agricultural development project assisted by the Bank Group under the Mahaweli Ganga Development Program and the first downstream development to be undertaken under GOSL's Accelerated Program. The project would provide new irrigation capability for about 24,100 ha and enhanced capability for about 3,620 ha in System C, comprising in total about 42% of the gross area of the System. Within Zones 3-6 (18,500 ha net of newly irrigated area), the project would provide: irrigation and drainage infrastructure; construction camps; land clearing and on-farm development; social infrastructure; market, village and hamlet roads; fuelwood and cashew plantations; management offices, quar- ters and vehicles; and settlement assistance. 1/ Within the total System C area, the project would provide: assistance for completing the Right Bank Transbasin Canal; technical assistance; monitoring programs; maintenance equipment and workshops; equipment and facilities for a farm machinery hire service; and equipment for a training center and farm. Principal Project Features 3.02 Works, facilities, and programs to be implemented over a six-year period would include: (a) Minipe Right Bank Transbasin Canal and allied structures from km 4.7 to its end near the Ulhitiya Reservoir (about 26 km); (b) Within Zones 3-6: (i) Construction of construction camp and workshops; (ii) Construction of main and branch canals, including the Main Canal No. 2; 1/ GOSL is currently undertaking similar investments in Zone 2 (5,300 ha net of newly irrigated area) with EEC assistance. GOSL will implement a similar program in Zone 1 (300 ha net of newly irrigated area) after completion of Zones 2-6. About 3,620 ha of currently irrigated area in Zones 1-2 would receive additional water and would benefit from technical infrastructure supplied under the project.  - 11 - (iii) Construction of distributor and field canals; (iv) Construction of main drains; (v) Construction of secondary and field drains; (vi) Land clearing and on-farm development; (vii) Construction of social infrastructure including schools and teachers' quarters, medical and public health facilities, community development centers, administra- tive service buildings, and agro-chemical, seed and food stores; (viii) Construction of market, village, and hamlet roads; (ix) Establishment of fuelwood and cashew plantations; (x) Assistance to settlers through provision of core house grants, seeds, and tools; and (xi) Construction of offices and staff quarters and provision of vehicles for the Mahaweli Economic Agency. (c) Within Total System C: (i) Provision of maintenance equipment and workshops; (ii) Construction of irrigation infrastructure and provision of equipment for the Giranduru Kotte agricultural development training center and farm; (iii) Implementation of a program to monitor and evaluate the project's agro-economic impact as well as irri- gation deliveries and system losses; (iv) Engineering, supervision, and management of project works; (v) Introduction of a sound water management system; (vi) Assistance to settlers through the establishment of a farm machinery hire service and comprehensive agri- cultural support services; and (vii) Provision of technical assistance and training. Description of Project Works, Facilities, and Programs 3.03 Minipe Right Bank Transbasin Canal. The Transbasin Canal would convey water from the Mahaweli Ganga at Minipe to the Ulhitiya Oya/Ratkinda  - 12 - Reservoirs, where it would be regulated for release to Systems B and C. Con- struction of the new Minipe Anicut, 1/ the first 4.7 km of the Transbasin Canal (including allied structures), and the embankments and outlet works of the Ulhitiya Oya and Ratkinda Reservoirs were contracted to GOSL construction agencies and private contractors before project appraisal. The remaining portion of the Transbasin Canal (about 26 km) would be included in the project. A contract for this latter work (excluding regulatory gates estimated to cost about Rs 8 M) costing Rs 537.8 M was awarded in August 1980, following ICB in accordance with JDA guidelines. 2/ The diversion capacity of the canal would be about 64 m /sec; it would be concrete-lined to reduce seepage losses, maintenance problems, and unauthorized diversions from the canal. Final designs for the Transbasin Canal have been completed and the preparation of working drawings is keeping pace with construction development. The following major auxilliary structures would be constructed, the first three under GOSL financing: Structures Status Minipe Head Sluice Contracted to GOSL agency Badula Oya Crossing (aqueduct) Contracted to GOSL agency Badula Oya Tunnel Awarded to private contractor Loggal Oya Level Crossing, spillway, and inlet and outlet structures Hepola Oya Level Crossing, spillway, Included in contract awarded and inlet and outlet structures August 1980. Diyabana Oya Level Crossing, spillway, and inlet and outlet structures Ulhitiya Oya Inlet Structure 3.04 Irrigation and Drainage Systems. Irrigation deliveries to Zones 3-6 would be conveyed from the Ratkinda regulating reservoir by Main Canal No. 2. 3/ Branch and distributor canals (D-canals), located on tops of ridges where practical to command the largest area possible, would distribute water throughout the project area. Balancing reservoirs would be located along the canals at appropriate locations to provide operational flexibility, reduce 1/ Diversion structure. 2/ It is estimated that US$10.0 M would be disbursed retroactively from the IDA credit for mobilization and initial construction costs of the Right Bank Transbasin Canal incurred subsequent to August 8, 1980 (para 3.46). 3/ Main Canal No. 1, starting at the lJlhitiya Oya regulating reservoir, would provide irrigation water to a separate irrigation system serving Zone 2, which is not included in the project.  - 13 - operational losses, and limit peak canal design capacity requirements. Reflec- ting GOSL's objective to optimize the use of available water, an integral part of the system plan would be to design some of the lower lying canals to inter- cept drainage flows and thereby make full use of return flows within the project area. Main and branch canals would be lined only in areas of rock excavation or where canal alignment would traverse extremely porous soils. 3.05 The basic pattern of development would be to group proposed 1.0 ha farms in blocks of about 10-15 farms, supplied by a single farm canal from one distributor turnout. The location and design of distributor and farm canals is being guided by ongoing irrigation system trials in Block 404 of System H. 1/ pipe irrigation pilot design to determine its applicability to Sri Lankan conditions and whether it would result in more efficient and equitable distri- bution of water with increased production. The development plan to be adopted for the project would be determined by the final design of a 1,000-ha repres- entative area undertaken by the Consultant (paras 3.07 and 3.38). It is planned to distribute water at the farm canal level by simple rotation among farmers in the turnout group. To improve farmer irrigation and water manage- ment capability, cooperation among farmers, and efficiency of water use, an irrigation training program for both project staff and farmers would be provided at the training center to be established in Zone 2 (para 3.18). 3.06 Because bedrock underlies the major part of the more permeable RBE soils at relatively shallow depths, most of the area would be developed for double-cropped paddy, which would require a relatively simple surface drainage system to provide drainage at harvest. Where return flows are to be used within the project area, the system would be designed accordingly. Small areas of perennial and diversified annual crops would be grown in free drain- ing soils where drainage problems are not expected to develop. 3.07 A summary of the proposed irrigation and drainage system for Zones 3-6, based on designs employed in System H, is given in Annex 1, Table 4. Field location surveys of main and branch canal alignments are in progress. Final designs of the irrigation and drainage systems, as well as of required social infrastructure, are currently being prepared (para 3.38), and final design specifications and tender documents are scheduled for completion in 1981. Detailed designs of the 1,000 ha representative area (down to farm level) include distribution and farm canal layouts, structure designs, and on-farm development plans. 3.08 Land Clearing. Present land use in Zones 3-6 consists of about 29,000 ha of forest, 2,000 ha of grassland, 6,500 ha of chena, and 100 ha of rainfed paddy. About 82% of this area would be cleared or modified in devel- oping the estimated 31,000 gross ha required for irrigated farms, project infrastructure, residential areas, and plantations. However, a 1,000 ha block of Sri Lankan Dry Mixed Evergreen Forest would be preserved as a conservation area. 1/ These trials were undertaken as part of feasibility studies for System C. They investigated four different field irrigation systems and various settlement patterns, modes of on-farm development, and lining materials.  - 14 - 3.09 Land clearing techniques and procedures used for new settlement in recent years have not proved entirely satisfactory, although some improve- ments have been introduced in System H. A better standard of land clearing and on-farm development is proposed for System C to help settlers attain full production levels rapidly. Before clearing commences, all usable trees would be removed by the State Timber Corporation and its contractors. This work is now in progress. Medium density forest would be cleared by dozers with straight or stinger blades. Hand-felling of isolated trees and removal of stumps and roots by dozers with rippers and front-end rakes would be carried out in light forest/chena areas. Settlers would be encouraged to use felled trees and underbrush for fuelwood. All clearing work would be contracted privately and would be scheduled for completion just prior to start of on- farm development work. Medium density forest clearing is estimated to cost Rs 6,000/ha, and light density forest/chena clearing, Rs 2,000/ha, averaging about Rs 5,200/ha. During clearing, dozers would also do rough levelling and filling. 3.10 On-Farm Development. To achieve efficient water use, farms must be laid out well, fields levelled properly, and bunds constructed and maintained satisfactorily. Considerable improvement in on-farm development is being achieved in System H, compared to earlier attempts, by combining mechanized and manual operations, the latter performed under contract with settlers. To effect further improvements in the project area, mechanized work would be contracted privately, specifying use of heavy duty disc harrows pulled by light to medium crawler tractors, enabling land preparation to proceed year-round. Such tractors could also be used in levelling/grading work. The following sequence of operations is proposed: (a) rough levelling; (b) preliminary land shaping; (c) heavy disc harrowing to chop roots and loosen soil; (d) surveying grades and locating bunds; (e) land grading, forming, and cross levelling; (f) establishing farm bunds, soil conservation contour bunds, field channels, and drains; and (g) final levelling. Machines would be used for activities (a) to (c); GOSL technical staff would perform (d); either machines or manual labor would accomplish (e); and (f) and (g) would be contracted to settlers. In the event that the settlers cannot accomplish (f) and (g) satisfactorily, private contractors would be employed. Costs of on-farm development work are expected to average about Rs 4,500/ha for mechanized development plus Rs 4,000/ha for remaining manual work. Proper phasing of clearing and on-farm development would be required to avoid costly clearing of secondary growth and reconstruc- tion of works previously completed. 3.11 Social Infrastructure and Community Development Facilities. Most existing infrastructure in System C is located in Zones 1-2. The only facility in Zones 3-6 is a primary school in Yakkure. The project would provide settlers with adequate health, educational, and other facilities to attain living stand- ards comparable to other developed areas in Sri Lanka. A unit, comprising about 250 farmer and 75 non-farmer settlers, would be the basic project develop- ment unit, and its service center would be the first element of social infra- structure constructed after completion of main roads. A village center would be developed for every four unit centers, and an area center would be developed for every eight unit centers. A town center would be established to serve every ten village centers. Two town centers would also be established, Giranduru Kotte (Zone 2) and Dehiatte Kandiya (Zone 4). A total of about 75 unit centers and 18 village centers would be established in Zones 3-6.  - 15 - 3.12 About 31,300 new families (about 180,000 persons) would settle in System C after development, including 24,100 farmer-settlers and 7,200 trades- men, artisans, and service personnel, including GOSL staff. The project would provide social infrastructure required in Zones 3-6, including quarters for teachers, medical staff, and postal employees. Community development centers and principal public service buildings would also be provided. A sites-and- services approach to town development would be carried out under the project. A summary of the infrastructure to be provided under the project is given in Annex 1, Tables 5 and 6. Settlement assistance proposed is described in paras 3.22-3.23. 3.13 Since most of the expected settler population is likely to be young, immediate educational requirements would be mainly at the primary and junior secondary level, with senior secondary schools to follow later. To minimize walking distances, a primary school would be provided for each unit. Thus, about 96 new primary schools would be required in System C, 75 to be provided under the project in Zones 3-6. Junior secondary schools would be situated in each of the 24 village centers (18 in Zones 3-6) and have about 250 pupils per school. One senior secondary school would be established at Giranduru Kotte (Zone 2) and one in Dehiatte Kandiya (Zone 4); the latter would be financed under the project. Limited boarding facilities would be provided for pupils coming from distant parts of the project area. School buildings and quarters for principals and head teachers would be provided under the project and would be of standard design. Since junior teaching staff would probably be drawn from the resident population, housing would be provided for only about one half of the primary school teachers and all of the junior secondary school teachers. All senior secondary school staff would be provided housing. The Ministry of Education would equip the schools and defray recurrent costs of the buildings, teachers' salaries, and other expenses. 3.14 The Ministry of Health currently faces major staffing problems for all posts in its service and finds particular difficulty in filling vacancies in remote rural areas, principally because of higher salaries available abroad and in private practice. To circumvent this constraint, the project would rely on strategically located medical facilities with effective ambulance and mobile unit service. The Mahiyangana district hospital would continue as the focal point of medical and public health services. A touring public health team for malaria control and a touring dispensary team holding clinics in village public dispensaries are based at this hospital. Two peripheral health units, each with a 48-bed hospital, one in Giranduru Kotte and one in Dehiatte Kandiya (the latter to be financed under the project), would also serve the population of System C. The project would provide nine public health dispen- saries and clinics at area centers for 37 public health midwives in Zones 3-6. Each midwife would serve two hamlets. 3.15 Meeting halls would be provided in each of the village centers. Other facilities provided would include town council offices, police stations and police posts, telephone and postal facilities, rural banks, and markets. Storage for fertilizers, seeds, and food provided by the World Food Programme (WFP) also would be provided at the unit center level.  - 16 - 3.16 Roads. The Asian Development Bank is assisting Sri Lanka in plan- ning and developing the major road network in the downstream development area under the Accelerated Program. Necessary market roads and other roads in the hamlets, towns, and village centers would be provided in Zones 3-6 under the project. About 90 km of market roads (9 m in width and unpaved), 70 km of 11 m wide urban roads (5 km metalled and tarred), and 240 km of graded laterite-surfaced hamlet roads (4.3 m wide) would be provided. 3.17 Management and Operational Infrastructure. The project would be managed by the Mahaweli Economic Agency (MEA). Offices and quarters for MEA staff would be provided under the project. During project implementation, MEA's recurrent costs (e.g. salaries, allowances, vehicle operation) would be covered as part of the O&M costs of the project. Housing would be provided for all MEA management staff at the unit and block levels and at project head- quarters as well as for half of the vehicle and tractor drivers. The project would provide office buildings at all levels, a central workshop and store, shed and minor repair facilities for the machinery hire unit, and facilities for fuelwood and cashew nurseries. A summary of management and operational infrastructure required for the project is shown in Annex 1, Table 7. 3.18 Giranduru Kotte Training Center and Farm. Farmer-settlers, some with no practical irrigation experience, would require training to gain proficiency in water management techniques. They would need to learn the fundamentals of, and accept responsibility for, operating and maintaining the irrigation system downstream of D-canal turnouts. The keys to the success of the Mahaweli Program and its individual projects are knowledgeable, effi- cient farmers, capable of managing water resources effectively in cooperation with other farmers in the same turnout area. With water use in similar irrigation projects currently running 2-3 times higher than actually needed, it is evident that a major effort must be made to (a) construct proper irri- gation systems to control water issues efficiently and to distribute them equitably, and (b) train farmers to be good water managers. The proposed irrigation system for System C would provide the former. Staff and farmer training programs at the proposed Center in Giranduru Kotte are expected to accomplish the latter. 3.19 The Farm at the Center would encompass the commanded area of one distributor canal (about 250 ha) and would demonstrate diverse cropping and irrigation systems. The Center, to be financed by the EEC, would have necessary classrooms, research buildings, and laboratories for demonstrating correct agricultural practices and irrigation techniques. Hostel accommoda- tions would be provided for 100 participants attending basic courses. Quarters for resident staff, offices, workshops and ancillary buildings would also be provided. Equipment and irrigation infrastructure for the Center and Farm would be provided under the project. 3.20 The Center would organize (a) pre-service and in-service training for management operations staff; (b) training for contact farmers and com- munity leaders; and (c) as space permits, training for other farmers within System C. The Center would also serve as a venue for demonstrations and field days for the settler community. Small demonstration plantations of fuelwood and cashews would be developed. A list of the facilities to be provided by the EEC is shown in Annex 1, Table 8.  - 17 - 3.21 Settlement. Farmer-settlers would each be allotted a 1.0 ha irri- gated plot and 0.2 ha for an unirrigated houseplot (para 5.03). Non-farmer settlers and GOSL employees in the area would be allotted 0.1 ha unirrigated houseplots. 3.22 Displaced persons (e.g., from reservoir areas) with agricultural background would receive priority in plot allotments. Other farmer-settlers would be expected to meet the following criteria established by GOSL: (a) currently owning no more than 0.4 ha of agricultural land; (b) proven experience in agriculture either as a sharecropper or agricultural laborer--priority would be given to appli- cants who participated in the Department of Agriculture's Practical Farm School Certificate course; and (c) availability of family labor: applicant must be married w`Lth children or be the breadwinner of an extended family. Farmer-settlers would receive their houseplots and farms about six months before irrigation water becomes available. Therefore, their first harvest would be obtained about one year after settlement. 3.23 A total of 24,100 new farmers are expected to settle on irrigated land developed in System C, 18,500 of whom would be located in Zones 3-6. These would be given the opportunity to work for competitive wages in develop- ing their lands, including preparation of farm ditches and drains. Each farmer-settler family would also be provided under the project with a core house grant (Rs 1,500) and Rs 500 to purchase field tools and seed (Annex 1, Table 9). Other settlers would be expected to construct houses and business establishments from their own resources, institutional credit, or other development programs provided by GOSL independently from the project. Settlers would be eligible to receive food aid for a period of 12 months after arrival under an agreement between GOSL and WFP. No difficulty is foreseen in attracting sufficient settlers for the project. 3.24 Land Use. Most irrigable land in the project area would be devel- oped for paddy production rather than for "irrigated dry" crops due to the soil-water regimen expected to result with the project. This would be the most economically viable alternative and is in consonance with GOSL's goal of self-sufficiency in basic foodgrains. Expected irrigable land use would be as follows: Crop Rotation Area (ha) Area (%) Paddy-paddy 16,000 86 Paddy-upland crops 1,500 8 Perennial crops 1,000 6 Total 18,500 100 The proposed cropping calendar for these crops at full development is shown in Annex 1, Chart 1.  - 18 - 3.25 Agricultural Extension. A strong agricultural extension program would be provided to ensure that expected yields and intensities are achieved as quickly as possible. Extension staff would receive thorough training be- fore and after arrival in the project area. Periodic courses at the Giranduru Kotte Center (para 3.18) would complement proposed pre-service training of field staff. The extension program provided would be consistent with the Training and Visit (T&V) 1/ system currently being introduced throughout Sri Lanka. Assurances were received from GOSL that implementation of agricultural extension services established in the project area under MASL would be in accordance with the systematic approach agreed to under the nationwide Agri- cultural Extension and Adaptive Research Project assisted by IDA. 3.26 Farm Machinery Hire Service. The proposed development of the project area is principally for double-cropped paddy. The maha crop should be planted by end-November to make most effective use of maha rains. There- fore, farmers must have sufficient farm power to complete land preparation within a period of about 6-8 weeks before the maha and 4-6 weeks before the yala, as well as assistance to thresh the maha crop. Since necessary farm power would not be available from farmers' own resources during the early stages of the project, MEA would operate a tractor hire service consisting of 150 four-wheel 40-50 HP tractors with suitable tillage implements provided under the project. It would not be necessary to replace these MEA tractors because the farming community eventually is expected to acquire sufficient two-wheel tractors and draft animals to make a publicly operated tractor hire service no longer necessary. Provision would be made for both individual livestock holdings and communal grazing areas. 3.27 Traditional threshing practices using animal power would not be adequate to handle the paddy crop expected with the project. Therefore, the project would provide axial-flow and drum-type (tractor or engine powered) threshers for MEA to hire out to farmers during the first 6-10 years of the project. After that, it is expected that the farmers would obtain threshing equipment with their own resources. 3.28 Agricultural Supporting Services. The marketing and distribution of seeds, fertilizers, and pesticides would be supported by the construction of a central store at Dehiatte Kandiya and by smaller stores to serve as retail outlets at unit-level. Existing storage at Giranduru Kotte is suffi- cient for such use. Since fertilizers would be blended and stored in large quantities at the Ceylon Fertilizer Corporation regional depot at nearby Badulla, the project would provide only minimum on-site storage facilities, but would provide sufficient transport vehicles to replenish stocks on short notice. Supplies over the long-term are expected to come from bulk depots at Polonnaruwa. 3.29 Rural Credit. Most farmer-settlers within the project area would be selected from the landless, unemployed, and underemployed segments of the population and would be dependent upon credit to obtain basic cultivation and domestic requirements. Furthermore, about 10-20% of the new occupants 1/ Daniel Benor and James Q. Harrison, Agricultural Extension: The Training and Visit System, World Bank, Washington, 1977.  - 19 - of System C would belong to the trading and service sectors and would require development capital to finance investments in support activities. Institu- tional credit, therefore, would be an integral part of the settlement require- ments. Recent efforts to improve credit recoveries, including rescheduling (para 2.11), are not expected to affect adversely credit eligibility for those settlers selected under the project. 3.30 The project would support overall banking operations in the project area by providing ten simply constructed rural bank facilities (para 2.09) in village centers. In addition, settlers and current residents alike are expected to benefit directly from complementary programs currently being carried out by major banking institutions to expand rural lending based on repayment and supervisory discipline, to reinforce their field organizations and facilities, and to broaden lending schemes as new rural financing needs arise. Their lending schemes and scales of finance, both existing and envis- aged, should, however, be reviewed periodically to ensure that financial assistance is not eroded by inflation. To permit realization of agricultural production potential from project investments and to ensure growth of a viable service sector, assurances were received from GOSL that it would make arrange- ments to provide adequate credit to project area residents, that such institu- tional credit would be supported by adequate supervisory staff and facilities, and that credit arrangements would reviewed periodically and adjusted as required. Cashew and Fuelwood Plantations 3.31 Cashews. For optimal utilization of out-of-command areas and in support of the country's policy of increasing exports of high-value commodi- ties, about 2,000 ha of unirrigated land would be planted to cashew trees. Land clearing would be carried out under the project. About 500 ha would be planted to cashew each year over a four-year period beginning in 1983 either by private entrepreneurs or by the Sri Lanka Cashew Corporation. Establish- ment costs have been estimated at about Rs 2,400/ha with maintenance costs averaging about Rs 920/ha each year thereafter, all to be borne by the planters. 3.32 Fuelwood. To reduce the country's reliance on imported fuel and provide a convenient source of fuel for settlers, the project would establish fuelwood plantations after clearing and development of the project area. Plan- tations would be systematically harvested and replanted. About 3,000 ha of land would be required to satisfy local needs. Plantations would be concen- trated in marginal, out-of-command areas. The fuelwood component of the project would be in line with the well-conceived afforestation program being carried out nationwide by the Forestry Department. About 300 ha would be planted to fuelwood trees each year over a ten-year period beginning in 1983 to ensure a continuous supply for area settlers. The plantation would be established either by the Forestry Department or by MEA with technical guidance from the Forestry Department. Water Supply and Demand 3.33 Based on a 28-year water requirements study carried out during appraisal, the total aveSage annual irrigation diversion requirement for System C would be about 500 M m , most of which would be supplied from Mahaweli Ganga  - 20 - diversions. Return flow and local inflow from irrigation within System C would be used to the extent possible to satisfy remaining irrigation require- ments. Comparisons of estimated Mahaweli Ganga flow at Minipe and estimated diversion requirements of System C (Annex 1, Chart 2) show that adequate water would be available for diversion. An assurance was received from GOSL that it would allocate to System C the full amount of water required fr3m the Mahaweli Ganga for project use, estimated to average about 500 M m annually. 3.34 Water Quality. The chemical quality of water in the Mahaweli Ganga and in small catchments located in the project area is excellent for irriga- tion. Water quality would deteriorate somewhat with the expected use of return flows but would still remain satisfactory. Also, analyses from 16 water samples collected from wells and reservoirs within System C indicate surface and groundwater supplies would be of good quality for domestic use. Monitoring and Evaluation 3.35 A monitoring and evaluation system would be developed under the project to review progress in project implementation and success in meeting its objectives. This system would focus on two basic areas: (a) input delivery and use covering civil works construction, on-farm development, equipment procurement, settlement, project costs, O&M, agricultural input provision, and credit; and (b) output achievement in terms of increased production, improved services and living standards, and water management (including irrigation deliveries and system losses). 3.36 Two specific activities would be carried out under the project-- agro-economic surveys and a water management evaluation. Other relevant evaluation activities would be specified during the first year of project implementation. The two monitoring exercises mentioned above are briefly described as follows: (a) Agro-economic Surveys. Separate surveys would be made in the currently developed area (Zone 1) and in areas to be developed to ascertain the project's impact on production levels, farm incomes, income distribution, settlement pro- gress and problems. These would start in 1982 with benchmark measurements and would continue on a periodic basis to 1990. Management effectiveness, farmer training, and adaptability and performance of farmer turnout groups also would be evaluated. Effectiveness of the project in achieving planned development and production goals would be studied, and causes of deviation from the planned project's targets would be identified. Results from these studies would be used to make adjustments necessary to improve implementation and to guide future settlement efforts in similar projects (para 4.10). (b) Irrigation Deliveries and System Losses. A water management program would be set up to determine irrigation deliveries accurately throughout the system and areas of unacceptable seepage and/or operational loss. Where necessary to obtain  - 21 - continuous measurements, gauging stations with automatic recorders would be established. Data obtained would be used in determining system operation criteria and possible system design modifications, including identification of canal sections in need of lining. Farm irrigation efficiency studies for paddy cultivation on both LHG and RBE soils would be made to determine percolation losses and water application losses under flooded conditions. Data obtained would be used to guide development of future systems with similar conditions (para 4.10). 3.37 Assurances were received from GOSL that a special monitoring and evaluation program would be established to assess project performance over the 1982-90 period, that details of this program would be submitted for review by IDA by March 31, 1982 and that it would implement a monitoring and evaluation program satisfactory to IDA by September 30, 1982. Estimated costs of monitoring activities are given in Annex 1, Table 9. Technical Assistance 3.38 A consortium of consultants (Consultant), 1/ financed under an EEC Special Action Credit, has been retained by GOSL to review final designs and prepare tender documents for the Transbasin Canal and to proceed with pre- construction investigations and final designs for System C development (paras 3.03 and 3.07). The Consultant, financed earlier by the UK, previously prepared the Feasibility Study for System C. Designs and tender documents for the Transbasin Canal have been completed. 3.39 The project would provide additional technical assistance for planning and scheduling project implementation work, contract administration and construction supervision of irrigation and drainage works and social infrastructure in Zones 3-6 and for general agricultural and water management aspects of the project. The following specialists would be provided: Proposed Period Speciality of Assignment (man-years) Senior Construction Engineer 4 Planning and Scheduling Advisor 3 Contract Administration Engineer 3 Senior Construction Inspector 4 Construction Management Engineer (records) 3 Agricultural Advisor 3 Water Management Specialist 3 On-Farm Development Advisor 4 Total 27 1/ Hunting Technical Services Ltd. in association with Sir Alexander Gibb and Partners, and Huszar Brammah and Associated.  - 22 - 3.40 Terms of reference for consultancy services required are presented in Annex 2. Total costs of these services are estimated at about US$3.9 M, inclusive of travel, shipping, and other expenses. Average man-month cost for consultancy services would be about US$12,000. An assurance was received from GOSL that the consultants required to assist in project impletation would be appointed on terms and conditions acceptable to IDA. The project would include training for GOSL staff in countries where specific programs are available in planning, design, and execution of irrigation development and settlement projects. Cost of such training is estimated at about US$0.2 M. Cost Estimates 3.41 Total costs of the project are estimated at about US$201.8 M, including US$68.9 M for price escalation. No import duties and taxes would be charged under the project. The foreign exchange cost would be about US$87.4 M, or 43%. Cost estimates are based on prices prevailing in mid-1980. Unit prices used for main and branch canals are based on recent bid prices for similar work. The cost of the Transbasin Canal is the contractor's bid price for constructing the canal, awarded after ICB conducted in accordance with IDA guidelines. Cost estimates for land clearing and onfarm development are based on recent MDB experience in System H, upgraded to take account of standards proposed in the project area. Costs of distributor and field canals are based on typical layouts constructed for irrigation research trials in Block 404 of System H. MDB's costs for administration and engineering are estimated to be about 10% of construction costs for civil works and 8% of such costs for social infrastructure. Physical contingency factors of 15% and 10% are applied to costs of civil works and social infrastructure, respectively. Price contingencies are estimated to be about 52% of base costs, including physical contingencies and O&M, in accordance with inflation rates specified in Table 3.1. O&M costs at full development are estimated to average Rs 690/ha (US$16/ac) as shown in Annex 1, Table 13. Cost estimates are detailed in Annex 1, Table 14 and are summarized below in Table 3.1. Cost estimates of equipment to be provided under the project are shown in Annex 1, Table 15.  - 23 - Table 3.1: SUMMARY COST ESTIMATE Foreign Local Foreign Total Exchange _-------US$ M------ () Civil Works Construction Camp 1.5 1.0 2.5 40 Right Bank Transbasin Canal 12.2 18.3 30.5 60 Irrigation System & Land Development in Zones 3-6 19.4 22.5 41.9 54 Roads 1.6 1.3 2.9 45 Social & Administration Infrastructure Social Infrastructure 4.1 1.8 5.9 30 Administration Offices & Quarters 4.5 1.9 6.4 30 Engineering, Supervision & Administration 10.0 - 10.0 - Equipment & Vehicles 0.5 9.1 9.6 95 Settlement Assistance 1.8 0.3 2.1 14 Monitoring Program 0.7 - 0.7 - Technical Assistance & Training 0.5 3.6 4.1 88 O&M During Project Construction 3.0 0.5 3.5 14 Base Cost 59.8 60.3 120.1 50 Physical Contingencies 6.0 6.8 12.8 53 Price Escalation /a 48.6 20.3 68.9 30 Total Project Cost 114.4 87.4 201.8 43 /a 1980 1981 1982 1983 1984 1985 1986 (percent p.a.) Foreign Costs 12.5 9.0 8.5 7.5 7.5 7.5 6.0 Local Costs 25.0 25.0 15.0 12.0 10.0 8.0 8.0 Implementation Schedule and Settlement Rate 3.42 The project would be completed in six years, or by December 1986. A detailed schedule of project implementation is provided in Annex 1, Chart 3. Settlement in System C is expected to proceed simultaneously with that in System B and therefore, could be subject to various constraints. Table 3.2, however, represents a projected settlement rate that is considered achievable and realistic: Table 3.2: ESTIMATED SETTLEMENT RATE FOR SYSTEM C /a (families) 1982 1983 1984 1985 1986 Total Zone 1 - - - - 300 300 Zone 2 4,300 1,000 - - - 5,300 Zones 3-6 - 3,500 5,000 5,000 5,000 18,500 Total 4,300 4,500 5,000 5,000 5,300 24,100 /a Assumes simultaneous settlement in System B.  - 24 - Financing 3.43 Co-financing arrangements for the project are being finalized by GOSL. The Kuwait Fund has pledged US$45.0 M equivalent to finance about 22% of total project costs, including approximately 17% of foreign costs and 27% of local costs. Japan is seriously considering an official request from GOSL for a concessionary loan of US$45.0 M, which would also finance about 22% of total project costs, including 36% of foreign costs and 12% of local costs. The above bilateral assistance combined with the proposed IDA contribution of US$90.0 M would finance about 89% of total project costs, including approxi- mately 81% of local costs. The remaining funds required (US$21.8 M equivalent) would be provided by GOSL. A condition of effectiveness for the IDA credit would be the satisfactory arrangements for commitments from the co-financiers participating in project financing. An assurance was received from GOSL that adequate and timely funds would be provided for project implementation. An estimated schedule of expenditures and an IDA credit allocation schedule are shown in Annex 1, Tables 16 and 17, respectively. Financing of discrete proj- ect components has been tentatively agreed with the prospective cofinanciers as shown in the following table: Category GOSL IDA Kuwait Japan --%