Document of The World Bank FOR OFFICIAL USE ONLY Report No. 32 39-IN INDIA STAFF APPRAISAL REPORT KARNATAKA TANK IRRIGATION PROJECT February 26, 1981 South Asia Projects Department Agriculture C Division 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 EQUIVALENTS US$1.00 Rupees (Rs) 8.40 1/ WEIGHTS AND MEASURES (METRIC SYSTEM) 1 meter (m) 3.28 feet (ft) 1 kilometer (km) = 0.62 miles (mi) 1 hectare (ha) 3 2.47 acres (ac) 1 million cubic meters (Mm ) = 810 acri-feet (ac-ft) 1 cubic foot per second (cusec) = 0.028 m Is = 28 liter per second (l/s) 1 cubic meter per second (cumec) = 35.3 cusec 1 liter per second = 0.0353 cusec 1 ton = 1,000 kilograms (kg) = 2,205 pounds 1/ The US Dollar/Rupee exchange rate is subject to change. Conversions in this report have been made at US$1.00 to Rs 8.40, which represents the projected exchange rate over the disbursement period. FOR OFFICIAL USE ONLY ABBREVIATIONS AD - Appraisal Division AE - Assistant Engineer ARDC - Agricultural Refinance and Development Corporation CCA - Cultivable Command Area CE - Chief Engineer EE - Executive Engineer GOI - Government of India GOK - Government of Karnataka HYV - High Yielding Variety ID - Irrigation Department IMD - Indian Meteorological Department LCB - Local Competitive Bidding MIC - Minor Irrigation Committee MIW - Minor Irrigation Wing NPV - Net Present Value O&M - Operation and Maintenance OMED - Operational Monitoring and Evaluation Division PFC - Project Formulation Cell SASC - Special Appraisal and Supervision Circle SE - Superintending Engineer TIP - Tank Irrigation Project TMED - Technical Monitoring and Evaluation Division UAS - University of Agricultural Sciences WRDO - Water Resources Development Organization GLOSSARY 1. Deccan - Plateau covering the Southern Indian Peninsula 2. Kharif - Main Agricultural Growing Season (Rainy Season, June to October) \ 3. Rabi - Second Agricultural Growing Season (Dry Season, October to February) 4. Strange's Table - Tables Relating Rainfall and Runoff in Hydrological Catchments for Southern India, developed by A. Strange 5. Switching Value - Value of a Cost or Benefit Parameter that would, Ceteris Paribus, reduce Net Present Value to Zero 6. Taluk - Administrative Subdivision of a District 7. Western Ghats - Mountain Range extending North to South along the West Coast of the Deccan Peninsula FISCAL YEAR GOI, GOK - April 1 to March 31 This document has a restricted distribution and may be used by recipients only in the performance of their offDcial duties. Its contents may not otherwise be disclosed without World Bank authorization. INDIA KARNATAKA TANK IRRIGATION PROJECT STAFF APPRAISAL REPORT TABLE OF CONTENTS Page No. I. INTRODUCTION ............................................ 1 Agriculture and Irrigation in India ..................... 2 The State of Karnataka .................................. 3 - General .............................................. 3 - The Economy .......................................... 3 - Employment ........................................... 4 - Rural Incomes ........................................ 5 Land Reform, Land Tenure and Farm Sizes ................. 5 II. THE AGRICULTURE SECTOR IN KARNATAKA ..................... 6 General ................................................. 6 Topography and Climate .................................. 6 soils ................................................... 7 Land Use ................................................ 8 Agro-Climatic Zones ..................................... 8 Crops and Cropping Patterns ............................. 9 Agricultural Production Trends .......................... 12 Agricultural Supporting Services ........................ 12 Marketing, Processing and Storage ....................... 13 III. THE IRRIGATION SUB-SECTOR ............................... 14 Water Resources ......................................... 14 Irrigation Potential and Utilization .................... 14 General Trends in Irrigation Development .... ............ 15 Regional Pattern of Irrigation Development .... .......... 16 The Role of Irrigation in Karnataka's Agricultural Development .......................................... 17 Bank Group Involvement .................................. 17 IV. PERFORMANCE OF EXISTING TANK IRRIGATION PROJECTS ........ 17 General ................................................. 17 Design and Operation .................................... 18 Project Formulation ..................................... 19 This report was prepared by Ms. M. Nguyen and Messrs. Y. K. Choi, C. Diewald, G. Fauss and P. Ljung, with contributions by Mr. G. Tibor and T. Wickham (consultant). The appraisal mission on whose findings the report is based visited Karnataka in August 1980. - i~i - TABLE OF CONTENTS (Continued) Page No. V. THE PROJECT ............................................ 20 Project Objectives ...................................... 20 Project Components ...................................... 20 Description of a Tank Irrigation System .... ............. 21 Project Scope and Unit Cost ............................. 21 Criteria for Tank Irrigation Projects ................... 22 - General ....... . ..................................... 22 - Planning Criteria .................................... 22 - Design Criteria ...................................... 24 - Implementation Criteria .............................. 25 - Operation Criteria ................................... 25 - Tank Irrigation Committee (TIC) ...................... 25 - Economic Criteria .................................... 26 Dam Safety .............................................. 26 Rainfall and Stream Gauging Network ..................... 27 Development of an Improved Hydrological Forecasting System ................................................ 28 Water Users' Organization Study ......................... 28 Project Costs and Financing ............................. 28 Procurement ............................................. 29 Disbursements ........................................... 30 Accounts and Audits ..................................... 30 VI. ORGANIZATION AND MANAGEMENT ............................. 31 The Irrigation Department ............................... 31 The Minor Irrigation Wing ............................... 31 - Existing Organization ................................ 31 - Proposed Organization ................................ 32 The Minor Irrigation Committee .......................... 32 Procedures for Processing of Tank Irrigation Projects ... 33 - Project Identification ............................... 33 - Project Preparation .................................. 33 - Project Appraisal .................................... 33 - Project Approval ..................................... 33 - Project Supervision .................................. 34 IDA Involvement ......................................... 35 Farmers' Organization ................................... 35 Maintenance of Tank Irrigation Projects .... ............. 35 Cost Recovery ........................................... 35 VII. BENEFITS AND JUSTIFICATION .............................. 37 General ................................................. 37 Production Benefits ..................................... 38 Employment Benefits ..................................... 38 Income and Poverty Impact ............................... 38 Assumptions for Economic Analysis ....................... 39 Economic Rate of Return ................................. 40 Project Risk ............................................ 41 VIII. RECOMMENDATIONS ......................................... 44 - iii - ANNEX 1 - SUPPORTING TABLES AND CHARTS T-1 - Performance of Selected Major Crops T-2 - Contribution of the Irrigation Sector to Agricultural Growth in Karnataka T-3 - Typical Unit Cost Estimates for Tank Irrigation Projects T-4 - Elements and Features of Six Selected Tank Irrigation Projects T-5 - Cost Estimates for Six Selected Tank Irrigation Projects T-6 - Tentative Distribution of Tank Irrigation Projects by Size and Agroclimatic Zones T-7 - Rainfall and River Gauging Network-Cost Estimate T-8 - Estimated Implementation Schedule T-9 - Schedule of Expenditures T-10 - Proposed Allocation of Credit T-ll - Estimated Schedule of Disbursements T-12 - Estimated Farm Budgets and Project Rent by Agroclimatic Zones T-13 - Standard Assumptions for Yields and Input Requirements T-14 - Cropping Patterns, Crop Budgets and Water Requirements by Agroclimatic Zones T-15 - Value of Water for Agroclimatic Zones T-16 - Estimated Cost and Benefit Streams for One Ha of Command Area CHARTS C-i - Organization Chart Minor Irrigation Wing C-2 - Organization Chart Minor Irrigation Committee C-3 - Project Cycle for Tank Irrigation Projects - iv LIST OF ANNEXES Annex I - Supporting Tables and Charts Annex 2 - Criteria for Tank Irrigation Projects Proposed for IDA Assistance Annex 3 - Terms of Reference for Study on Increased Farmer Participation in Tank Irrigation Projects Annex 4 - Terms of Reference for Panel of Experts on Dam Safety Annex 5 - Reporting Formats for Project Preparation, Monitoring and Evaluation Annex 6 - Empirical Formulae for Estimating Runoff Yields in Karnataka and Proposal for an Improved Forecasting Model Annex 7 - Assumptions for Economic and Risk Analysis Annex 8 - Economic Justification of Canal Lining Annex 9 - Related Documents and Data Available in Project File LIST OF MAPS IBRD 15341 Agroclimatic Zones IBRD 15342 Soils IBRD 15343 Normal Annual Rainfall INDIA KARNATAKA TANK IRRIGATION PROJECT I. INTRODUCTION 1.01 A tank is a small reservoir behind an earthen embankment. Tanks have been the traditional source of irrigation water in the areas of low and erratic rainfall in southern India (Andhra Pradesh, Maharashtra, Tamil Nadu and Karnataka) for many centuries. They are built to store seasonal runoff from streams and minor rivers mainly during the months of the Southwest Monsoon. Water is released either during the rainy season itself to protect the standing crop or, if remaining supply permits, to raise a second irrigated crop in the following months of scant or non-existing rainfall. Priority is given to supplementary kharif irrigation (protective function), but farmers usually try to minimize water use for this purpose in order to make a second crop possible. The distribution of water is through one or two main canals and then through primitive field channels to the crops. 1.02 A tank command area may range in size anywhere between a few hectares and several thousand hectares. In the past, smaller tanks were usually built by village communities and local landlords, larger ones by regional princes. After Independence, State governments took over ownership of all existing tanks with more than 4 ha command area (about 25,000 in Karnataka), construc- tion of new tanks, and their operation and maintenance. However, since the thirties increasing emphasis has been given to large scale surface irri- gation schemes, starting with the Krishnarajasagar Scheme near Mysore, fol- lowed after Independence by several others--most notably the Tungabhadra Scheme commanding about 418,000 ha (within Karnataka). While construction of large schemes continues, e.g., in the IDA-assisted Upper Krishna Scheme, the Government of Karnataka (GOK) has realized the limitations of large projects, which provide irrigation only to certain areas of the State while leaving other dry and drought prone areas without irrigation development. The merits of minor irrigation (tanks, lift schemes, pick-up weirs, wells) once again became recognized, for it allows greater geographical flexibility and provides water to many areas poorly endowed with water resources. 1.03 As part of the current Five Year Plan (1980/81 to 1984/85), GOK has scheduled a Minor Irrigation Program of about US$126 M, of which US$108 M will be used for construction of tanks and lift irrigation schemes and for repair of existing schemes. Government of India (GOI) has requested IDA's assistance to finance a four year time slice (1981/82-1984/85) of GOK's tank construction program, covering 120-160 new tank irrigation projects (TIPs) spread throughout the State of Karnataka. 1/ This would be the first time that IDA would assist directly in the development of minor surface water schemes in India. (Minor irrigation has been receiving support by IDA through agricultural credit operations.) 1/ A tentative list of TIPs has been prepared by GOK, including 160 individual projects. -2- 1.04 Tanks to be financed under the proposed project would be located mostly in drought prone areas and where large scale irrigation development is not feasible. They have short gestation periods and thus would yield quick returns. The proposed investment would promote a regionally more balanced development and reduce rural poverty and unemployment. The tanks would also provide opportunities to experiment with more modern irrigation practices, for crop diversification and the setting up of water management organizations at the grass root level. Lessons learned from the tank project could be used in the design and subsequent management of major irrigation projects where small management units (irrigation blocks) could be disaggregated from the large conveyance system and be operated somewhat like individual tank projects. 1/ Agriculture and Irrigation in India 1.05 India has a population of about 660 M which is growing at an annual rate of 1.9%. Although since 1960 per capita income grew at an annual rate of 1.4% and reached US$180 in 1979/80, and although access to public services has improved, there has been little change in the incomes of the vast masses of urban and rural poor, who comprise about 50% of the total population. Therefore, GOI's development plans give priority to alleviating poverty and creating employment, especially in rural areas. 1.06 Agriculture is the dominant sector of the Indian economy and contri- butes about 45% of GNP. It engages about 70% of the labor force and provides the base for about 55% of India's exports. During the last decade, GOI devel- opment plans have emphasized agriculture and sought to raise food production by increasing the use of fertilizers, plant protection chemicals and improved seed varieties. In suport of this, GOI has modernized and expanded its agri- cultural credit institutions and accelerated the development of irrigation. 1.07 Some 50 M ha are presentlv irrigated, about three-fifths from surface water sources and two-fifths from groundwater. Irrigated areas are almost four times as productive as rainfed areas and account for about 60% of agricultural output. The expansion of irrigation facilities and the productivity gains on irrigated lands have accounted for at least three- quarters of the agricultural growth since 1960. 1.08 Actual utilization in many completed irrigation projects remains low. In an attempt to remedy the situation, Command Area Development programs-- focusing on development below the Government outlet, i.e., on the construction of field channels and drains and on landshaping---have been initiated in most major irrigation projects. Experience from such programs indicates, however, that a lack of on-farm development is only a partial explanation for the 1/ Instead of the traditional irrigation outlets in the major projects (which serve areas of about 40 ha) a proposal under consideration (not part of this project) is to build small tanks near these outlets where water can be stored overnight or possibly longer and supplied to the consumers below the tank on a more flexible basis than the traditional rotational system. - 3 - underutilization of the irrigation potential. It has become increasingly clear that one of the major constraints to greater utilization lies in the deficiencies of the distribution system between the water source (dam or diversion) and the Government outlet. 1.09 Project design, water allocation and canal operation procedures have changed little since the last century when the prime objective of irrigation was to provide "insurance" against drought and famine. Existing designs and management procedures make it almost impossible to provide a timely and reliable water supply at the farmgate--a prerequisite for modern agriculture based on high yielding, high value crops. Furthermore, actual water conveyance losses are much higher than assumed for the design of the canal systems, which reduces the area that can actually be irrigated. As a consequence, on-farm development is resisted by the farmers who have to pay for its cost and may not see the benefits. Improvements of both irrigation structures and on-farm development should be complementary and reinforce each other to meet the demand of modern agriculture. These objectives have been pursued by many recently Bank-financed irrigation projects in India. The State of Karnataka 1.10 General. The State of Karnataka has an area of 19.1 M ha and a population of about 35.5 M (1979). Population growth during the last two decades is estimated at about 2.2% per annum. About 70% of the labor force is employed in agriculture. The ability of urban centers to accommodate migrants from rural areas is limited and the welfare of the majority of the State's population is critically dependent on growth in agriculture. 1.11 The Economy. Between 1960/61 and 1978/79, State income grew at an annual rate of 3.5%, slightly higher than the national average. Sectoral growth in Karnataka differed from that of the rest of India as follows: Share in Trend Growth Net Domestic Product 1960/61 1978/79 Karnataka India Karnataka India (average 1974-79, %) (% per annum) Agricultural & Allied Activities 48 40 2.9 2.2 Mining, Manufacturing, Construction, etc. 23 25 4.4 4.3 Services 29 35 4.1 4.5 Total 100 100 3.5 3.3 Karnataka's agriculture grew faster than agriculture in the rest of the country, yet it showed the lowest growth among all sectors in Karnataka. Since it is the main activity in the State and has close links to transport, storage, trade and other services, the comparatively poor performance of the service sector can partly be explained by the slow growth of agricultural - 4 - output. The fastest growing sector has been manufacturing. However, with its small and highly concentrated base, it contributed only marginally to the welfare of the majority of population in the rural areas (75%). 1.12 Per capita income grew at 1.3% p.a., close to the national average. In 1977/78, average per capita income was estimated at about Rs 1,130 (US$135). Variations in the regional income levels are largely a result of ecological factors: Average Per Capita Rural per Income in Zone /a Rainfall Capita Income /b Karnataka (mm) (Rs) (Rs) Dry 450- 900 680 1,180 Transitional 660-1,300 700 1,040 Hilly 1,000-6,400 1,380 1,690 Coastal 3,000-3,700 830 1,350 State 750 1,130 /a The State is divided into ten agro-climatic zones, based on soils and rainfall patterns (see para 2.06). In this table they have been further aggregated into four groups, based primarily on the amount and relia- bility of rainfall. /b Value-added in the agricultural sector per head of rural population. The table indicates that the high rainfall areas have a higher agricultural output per head of rural population and higher average per capita income. The removal of regional inequities depends largely on the increase of agricultural output. Since unpredictable and low rainfall tends to restrict growth in agriculture, an expansion of reliable irrigation facilities would reduce this constraint, particularly in the dry and transitional zones. 1.13 Employment. The labor force constitutes about 35% of Karnataka's population. The participation rate is higher in rural (36%) than in urban areas (27%). In 1971, cultivators accounted for 40% and agricultural laborers for 27% of the workforce. Productive employment has failed to keep pace with population growth. Open unemployment in 1977/78 was 9.3% of the rural and 11.4% of urban labor force (9.8% of total labor force). 1/ No estimates of underemployment are available, although it is considered more significant. Removal of unemployment and underemployment is one of the main development objectives at present. Highest priority is accorded to irrigation, agricul- ture and allied activities which generate high employment. In addition, in 1/ Based on National Sample Survey (1977/78) of the distribution of persons in the age group 15-59 by daily activity in Karnataka. - 5 - March 1979, GOK introduced the "Employment Affirmation Scheme" which aims at providing work of about 100 days in a year to all able-bodied adults who cannot find work during the slack agricultural season. 1.14 Rural Incomes. Average rural per capita income in Karnataka is estimated at about Rs 750 (US$89) per year. 1/ About 70% of the rural population is considered as living at or below the poverty threshold. 2/ Because of sluggish growth in per capita rural income, the incidence of poverty has basically remained unchanged. GOK has attempted to implement a series of programs to raise standards of living of the poor, including land reform, minimum wages, assistance to small and marginal farmers, subsidies to artisans, special programs for scheduled Castes and Tribes and other backward minorities. Experience indicates that the Employment Guarantee Scheme in Maharashtra had a significant effect on incomes of a large group of landless laborers and small farmers. The Employment Affirmation Scheme in Karnataka is expected to have similar effects. Land Reform, Land Tenure and Farm Sizes 1.15 Land Reform has been in effect since 1965. It provides land owner- ship to tenants, fixes land ceilings, and redistributes surplus land among the landless. Land ceilings are fixed as 20 ha for rainfed and 8 ha for irrigated land per family of five, but are increased for each family member in excess of five, up to double the basic ceiling. So far, about 0.22 M tenants have been declared owners of about 0.44 M ha land. In addition, 0.13 M ha have been declared surplus to be distributed to landless farmers. At present, about 90% of all holdings are wholly owned and self operated; 5% are partly owned, partly rented and the remainder 5% are wholly rented. 1.16 The average farm size in Karnataka decreased from 3.20 ha in 1970/71 to 2.98 ha in 1976/77, due to a reduction in both the number and the area of large holdings. Marginal and small farmers continue to be predominant. The change in holdings of various categories of farms occurred as follows: % of Numbers % of Operational Category of Farms of Holdings Area Average Size (ha) 1970-71 1976-77 1970-71 1976-77 1970-71 1976-77 Marginal (below 1 ha) 30 33 5 6 0.51 0.50 Small (1-2 ha) 24 24 11 12 1.46 1.49 Medium (2-10 ha) 40 38 53 54 4.25 4.24 Large (10 ha and above) 6 5 31 28 16.43 16.30 Total 100 100 100 100 3.20 2.98 1/ The value added of agriculture and allied activities per head of rural population. It is about two-thirds of the state per capita income of US$135. 2/ Estimated as about Rs 780 per capita per year for rural India (1980). -6- 1.17 Regional variations in farm size distribution are determined by agroclimatic conditions; small farms are much more frequent in high rainfall areas than in low rainfall areas: % of Farms in Zone Annual Rainfall Size Group (ha) Average Size (mm) 0-1 1-2 2-10 10+ (ha) Dry 450- 900 33 23 38 6 3.14 Transitional 600-1,300 27 24 44 5 3.07 Hilly 1,000-6,400 47 22 29 2 2.16 Coastal 3,000-3,700 57 25 17 1 1.33 Karnataka 33 24 38 5 2.98 However, small farms in high rainfall areas support considerably more people than similar size farms in other areas. II. THE AGRICULTURAL SECTOR IN KARNATAKA General 2.01 Agriculture contributes about 45-50% of Karnataka's State income and employs about 70% of its labor force. From 1955/56 to 1977/78, production of most major crops increased at a moderate rate. Topography and Climate 2.02 The State forms the southwestern part of the Deccan Peninsula. Most of the State is on a plateau, bordered in the west by a coastal strip and the Western Ghats (up to 1,500 m in altitude) and tapering off in altitude from the Ghats towards the east. All rivers originate in the Ghats, flowing west or east. The State includes four distinct physiographic zones (see Maps IBRD 15341 and 15343): (a) The coastal plain between the Western Ghats and the Arabian Sea. It averages about 30 km in width, and has an annual rainfall ranging from 3,000 to 3,700 mm, virtually all of it during the Southwest Monsoon. (b) The Western Ghats, a mountain range averaging 50 km in width and extending for 650 km from the North to South. It includes the areas with the highest rainfall in the State, ranging from 1,000 to 6,400 mm and averaging about 2,500 mm. (c) The Northern Plateau, lying to the East of the Ghats, with mean annual rainfall varying from 900 mm at the western margin to 500 mm at the eastern. However, rainfall increases again towards the Northeast to about 900 mm. Elevation averages about 600 m. - 7 - (d) The Southern Plateau is similar to the Northern, but with more rolling topography and annual rainfall in the range of 750 mm to 500 mm. 2.03 Karnataka has a tropical monsoon climate. It receives the major portion of its rainfall from the Southwest monsoon which usually starts in early June and continues with some intervals till the end of September. The Northeast monsoon commences in October and usually ceases by the end of December. There are considerable variations in both amount and distribution of monsoon rainfall. The times of onset and end of the monsoon are uncertain and long dry spells frequently occur within the rainy season. In fact, as many as 135 out of 175 taluks 1/ in the State are often exposed to drought (classified as "drought prone"). These conditions make irrigation important for successful crop production even during the monsoon season. Soils 2.04 A wide variety of soils occurs in different parts of the State (Map IBRD 15342). Deep black soils are mostly confined to the northern parts of Karnataka. They are over 90 cm deep and usually dark brown in color, silty clay and clay, ranging from 40-65%. They have high water holding capacities. These soils have low permeability (0.1-10 mm per hour) and a tendency to swell and shrink as the soil moisture fluctuates. They are fertile and pro- duce good yields when moisture is adequate. Medium deep black soils occur mostly in the north central part of the State. These soils are moderately deep (30-90 cm) and dark in color. They contain a high percentage of clay (30-40%), have a moderate water holding capacity, ranging from 30-40%, and are moderately permeable with an infiltration capacity of 0.5-2.5 cm per hour. Shallow black soils occur mainly in the northern parts of the State. These soils are less than 30 cm deep, gravelly clay loam to clay loam and clay in texture with low water holding capacity of 20-35%. They are moderately to highly permeable with moderately high infiltration. The soils are well drained. Red loamy soils are found in a long strip east of the Western Ghats. The soils are sandy loam, gravelly sandy loam to clay loam, with low fertility and shallow rooting depth and low water holding capacity, ranging from 20-25%. They are well drained with moderate permeability. The infiltration capacity ranges from 1.5-2.5 cm per hour. Red sandy soils occur in the southern and central part of the State. They are 50-80 cm deep with a high proportion of quartz gravel. The surface is a thin layer of gravelly to cobly sandy loam to loamy sand; the sub-soils are gravelly to strong sandy clay with gravel ranging from 30-50%. These soils also have low fertility and shallow rooting depth. The soils are moderately rapid in permeability with an 2-3 cm per hour infiltration rate. Adequate doses of organic manures to build up soil struc- ture, fertility and water holding capacity is necessary to obtain satisfactory crop yields in these soils. Laterite soils mainly occur on gently undulating rolling plains, in hilly topography, and along the coastal high rainfall areas. They are deep, dark red in color, and clay loam to gravelly sandy loan on the surface and clay loam to gravelly sandy clay or clay in the subsurface 1/ Taluk is an administrative subdivision of a district, typically with a population of about 150,000. - 8 - horizon. They are acidic and low in water holding capacity (about 20-30%) with moderately high permeability and infiltration. Alluvial soils occur on the coastal plains. They are deposited soils consisting of washed-down material from the Western Ghats. The soils are loamy sandy to sandy loam on the surface, with loamy sand, gravelly sandy loam to clay loam in the sub- surface horizon. They have low water holding capacities, with moderately slow permeability. The water table in the low-lying areas is usually at 1 to 1.5 meters for most parts of the years. Land Use 2.05 About 11.9 M ha (62% of the State's area of 19.1 M ha) is under cultivation. The forest area is 2.9 M ha (15%), less than what is required to maintain a healthy ecological balance. About 1.9 M ha (10%) are either barren, uncultivable or put to non-agricultural uses. Cultivable waste land accounts for 0.6 M ha (3%). The remaining 1.8 M ha (10%) are used as permanent pastures or as grazing land. In practice there is little scope for increasing the cultivable area. Agro-Climatic Zones 2.06 Karnataka has been divided into ten agro-climatic zones based on climate, topography, soils and vegetation (see Map IBRD 15341). These zones are: (a) The Northeastern Transition Zone (5%) 1/ receives between 800-900 mm of rainfall annually. The soils are black clay with poor drainage. Main crops grown are sorghum, maize, cotton and chillies in the wet season; pulses, maize, and oilseeds in the dry season. (b) The Northeastern Dry Zone (13%) receives from 600 to 800 mm of rainfall. The highest rainfall occurs in September and October. The soils are deep black clay in most areas. Sorghum, millet, oilseeds, pulses and cotton are the main crops grown. (c) The Northern Dry Zone (34%) has a semi-arid monsoon climate. Rainfall varies widely over the large area, being somewhat higher in the western and northern parts. Much of the zone receives less than 600 mm of rainfall per year. Two types of soils, black clay and red sandy loam, exist in the area. The principal crops are sorghum, millet, pulses and oil- seeds. (d) The Central Dry Zone (10%) also has a semi-arid monsoon climate and receives between 400-700 mm of annual rainfall, mostly during the pre-monsoon (April to May) and monsoon 1/ Figures in brackets indicate a zone's share in total cultivable area of the State. - 9 - (July to October) seasons. The soils are red sandy loams with some mixes of deep black clays. The important crops are sorghum, millet, pulses, chillies, oilseeds and some cotton. (e) The Eastern Dry Zone (8%) also receives pre-monsoon and mon- soon rains which range between 800-900 mm annually. The soils are mostly red sandy loam with some clay laterites. The main crops are millet, paddy, pulses, maize and oil- seeds. As a cash crop, mulberry is extensively grown throughout the year. (f) The Southern Dry Zone (7%) has natural features similar to the Eastern Dry Zone, except that pockets of black clay soils are intermixed with red sandy loam in some areas. The prin- cipal crops grown are paddy, sorghum, millet, pulses and oilseeds. Sugarcane and mulberry are important cash crops in this zone. (g) The Southern Transition Zone (7%) forms a narrow strip wedged between the dry zones (Southern and Central) and the southern part of the Hilly zone. Annual rainfall varies widely from place to place and ranges between 600-1,000 mm. Red sandy loams and loamy soils dominate in the area. The main crops are paddy, sorghum, millet, pulses and sugarcane. (h) The Northern Transition Zone (8%), between the Northern Dry Zone to the east and the Hilly Zone to the west, forms a strip extended from the Southern Transition Zone, where annual rainfall ranges between 600 to 1,300 mm. The soils are black clay and red sandy loam in equal proportion. The principal crops are paddy, sorghum, pulses, oilseeds, cotton and sugarcane. (i) The Hilly Zone (6%) occupies a long narrow mountainous area to the east of and parallel to the coastal belt. Rainfall is very heavy and reliable. The soils are red clay loams and laterites in major areas. The main crops are paddy, condi- ments, spices and plantation crops such as coffee. (j) The Coastal Zone (2%), a long narrow coastal belt lying between the Western Ghats and the Arabian Sea, receives a high and assured rainfall during the monsoon season. Paddy is the main crop and coconut, arecanut, cashews, fruits and other perennials are extensively grown. Crops and Cropping Patterns 2.07 Karnataka has three cropping seasons: Kharif (usually June-October); Rabi (November-February); and hot weather (February-May). Most of the crop- ping takes place during the kharif season, utilizing the monsoon rainfall. - 10 - In the northeastern and northern dry zones where over 50% of annual rainfall occurs between September and December, the main growing season is later than in other parts of the State. Rabi crops are grown on residual soil moisture, occasional rainfall or with irrigation. No cultivation is practiced during the hot season, except in areas where perennial irrigation is available. The variable agro-climatic conditions in the State have caused a diversity of cropping patterns. Major crops are described below: 2.08 Paddy is the most important cereal crop of the State, accounting for about 10% of the total cropped area. Almost 65% of the crop is irrigated, and paddy is the preferred crop for irrigated land, taking up about 35% of the gross irrigated area. Rainfed paddy is mostly grown in areas of assured rain- fall, i.e., the coast, the hills and in the transitional zones. The average yield in the State is estimated at about 3.2 t/ha; rainfed yields average about 1.8 t/ha. 50 to 60% of paddy consists of high yielding varieties (HYV), with yields of 4.0 to 4.5 t/ha under average irrigated conditions, and about 2.0 to 2.5 t/ha when rainfed. Highest yields are achieved in the hot season (February to April) under irrigation. HYVs are usually transplanted. Local varieties, mostly drilled and grown under rainfed conditions, give average yields varying with rainfall from 1.0 t/ha to 2.0 t/ha. 2.09 Sorghum is the major dryland crop in the State, grown rainfed during the monsoon season and on residual soil moisture in the dry season. Sorghum is the staple food crop especially in the northern dry zones and occupies about 21% of the cropped area. Several hybrid varieties are gradually replac- ing local varieties. Average yields are low at 0.6 t/ha for local varieties and 2.0 t/ha for the hybrids. Hybrids under irrigation can be expected to produce about 3.0-3.5 t/ha. 2.10 Millets are an important crop group in the low rainfall areas of the State, accounting for about 20% of the cropped area. They are mainly grown in the kharif season and to a small extent under irrigation in the rabi season. Ragi or fingermillet 1/ is the staple food crop in the southern parts of the State. About 90% of it consists of HYV. Hybrid bajra or pearl millet 2/ and minor millets occupy the remaining area. The area under minor millets is gradually decreasing as the hybrids and HYVs are more widely adopted by the farmers. Crop yields are low at 0.4 t/ha for local varieties, and 0.6 t/ha for HYV, both under rainfed condition. When irrigated, HYV ragi and hybrid bajra average yields are about 1.6 t and 2.4 t per ha, respectively. 2.11 Maize has been a minor crop in the past. However, since the intro- duction of hybrid seeds in the early 1970s, the area planted to maize has rapidly increased. Still the total cropped area in maize is only about 1%. Further growth in maize production is expected from its increased acceptance as a foodgrain, as livestock feed and for industrial processing. Hybrids occupy about 90% of the maize area and are grown under irrigation. In the Southern and Northern Dry Zones, hybrid maize is also being successfully grown as a rainfed crop. Average yield of hybrid maize in the State is about 2.9 t/ha as against a national average of about 1.0 t/ha (all varieties). 1/ Eleusine coracona. 2/ Pennisetum typhoideum. - 11 - 2.12 Wheat is a popular rabi crop cultivated in the northern parts of the State. It accounts for 3% of the cropped area in the State. About 40% of the cropped area is covered with HYVs of the semi-dwarf Mexican types and local improved varieties. The present yield of HYV wheat is low at 0.8 t/ha as against the average of about 1.3 t/ha in India. Locally improved rainfed varieties give about 0.5 t/ha. These low yields are attributed to the southern location and high winter temperatures. 2.13 Pulses account for about 13% of the cropped area in the State. Pulses comprise cow pea, pigeon pea, bengalgram, blackgram, greengram, and horsegram. They are grown in all seasons, virtually all under rainfed conditions. Pigeon pea is the most important pulse in the kharif and bengal- gram in the rabi season. Yield of pulses under rainfed is about 0.4 t/ha and irrigated about 0.5 t/ha. GOK is encouraging farmers to grow more pulses as a cheap source of protein supply. Higher yielding varieties have recently been introduced for some species. 2.14 Oilseed crops are groundnut, safflower, sunflower, sesamum, linseed, niger and coconut. They account for about 9% of the total cropped area in the State. The most widely grown oilseed crop is groundnut. Presently, most oilseeds are grown rainfed during kharif, but farmers start growing them also during the rabi season. The average yield of unshelled groundnut (the most important oilseed) is low at 0.5 t/ha rainfed and about 0.8 t/ha under irriga- tion. GOK encourages expansion of oilseeds in line with a national policy. 2.15 Cotton is an important fiber crop, grown on 9% of the cropped area, primarily on the black soils in the northern parts of the State. Karnataka contributes about 10% of the total cotton production of the country. Pre- sently, long staple hybrid and HYV cottons are grown mainly on irrigated areas (about 7% of the cotton area). Most of the rainfed varieties are Asiatic and American upland cottons. They have usually short to medium staple length. A poor average yield of about 0.3 t/ha for rainfed cottons is attributed to: (a) the use of mixed seed of low viability; (b) poor pest control; (c) the vagaries of rainfall, particularly the timing of early rains; and (d) the unwillingness of farmers to invest in inputs under those conditions. Since 1972, two hybrid varieties have been successfully introduced in Karnataka. They are especially suited for the area and for irrigation, and are of the long staple type. Yields under average irrigated conditions are 1.5-2.0 t/ha (seed cotton). 2.16 Sugarcane is an important cash crop. All sugarcane is grown under irrigation. While only 1% of the cropped area is currently devoted to sugarcane, there is a tendency to expand the area as more land comes under irrigation. Sugarcane is grown throughout the State. The average yield is estimated at 75 t/ha, which is among the highest in India. 2.17 The mulberry area in the State is estimated at 110,000 ha, of which about 27% is irrigated. About 4% of the cropped area is planted to an improved variety. Although mulberry grows well in a wide range of soils in the State, at present it is largely confined to the Eastern and Southern Dry Zones. Recently, evidence has accumulated that sericulture is feasible in many other parts of the State. A rapid increase in mulberry planting is - 12 - therefore anticipated. 1/ Present average leaf yield is estimated at 3.0 t/ha (rainfed) and 8.0 t/ha (irrigated), far below its potential (about 25-30 t/ha). Due to high net returns, farmers are increasingly interested in this crop. Agricultural Production Trends 2/ 2.18 Between 1955/56 and 1977/78, foodgrain production grew only about 2.7% per annum, slightly above the population growth of 2.2%. Areas under foodgrains declined while yields showed an increasing trend. The performance of cash crops remained stagnant except for the production of sugarcane and cotton, which increased at 5.7% and 2.9%, respectively. However, since these crops only occupy a small portion of the total cropped areas, they did not significantly affect the overall performance of the agricultural sector. 2.19 From the mid-50s to mid-60s, agriculture was characterized by both area expansion and yield increases for most crops. Rice and jowar accounted for most of the gains in foodgrains production which grew at a rate of 1.4% per annum. Production of most cash crops, except for sugarcane, declined by about 1-2% p.a. The trend changed significantly after 1967/68, with a gradual introduction of HYV and rapid increase in the use of fertilizers (11% p.a.). Increases in production of most crops during this period were largely the result of impressive yield improvements particularly for jowar, ragi, maize and wheat. This was, however, not the case for rice whose yield only grew at a rate of 2% p.a. during this period. The performance of cash crops has been mixed. Production of oilseeds remained relatively unchanged from the previous period; sugarcane grew at a much slower rate than before and cotton production increased spectacularly at 8.2% p.a. mainly due to changes in varieties and corresponding yields. Because of significant increases in foodgrain production since the mid-60s (about 3.4% p.a.) Karnataka has become a net exporter of cereals. It is marginally self-sufficient in pulses and edible oilseeds. GOK plans to stimulate the diversification of the cropping pattern in the State towards more pulses, oilseeds, horticultural crops and sericulture. Agricultural Supporting Services 2.20 General. Agriculture in Karnataka experienced a faster growth in the past than did India's agriculture as a whole (para 1.11). This may be explained in part by a more advanced structure and performance of supporting services such as extension and research, credit and input supply and marketing. The proposed project, covering virtually the entire State, would build on the satisfactory state of this supporting system; deficiencies still encountered are usually common to all states and to be addressed at a more general level. 1/ IDA is assisting Karnataka's sericulture development under Cr. 1034-IN, US$54 M, 1979. 2/ For detailed information on performance of major crops refer to Annex l, T-1. - 13 - 2.21 Agricultural Research. The University of Agricultural Sciences (UAS), established in 1965, is responsible for agricultural research. It has two main campuses at Bangalore and Dharwar and five regional stations repre- senting the main agro-climatic zones in the State. A network of 35 substa- tions is dispersed throughout the State to provide location-specific testing of research results. UAS owns some 600 ha of farmland and is staffed with over 200 scientists. It also has about 80 Extension Units involved in testing research findings on farmers' fields. Existing research facilities are adequate for project support. 2.22 Agricultural Extension Services. The "Training and Visit System" has been introduced in Karnataka since 1978 assisted by IDA under the Composite Extension Project (Credit 862-IN). The extension program under this project is progressing well and now covers virtually all districts of the State. There is a close linkage between the extension service and research provided by UAS. The proposed project is expected to greatly benefit from the inten- sity and coverage of the extension system. 2.23 Agricultural Inputs. There is a well organized system for distri- bution and sale of seeds, fertilizer and pesticides throughout the State. Procurement and distribution is organized by the Karnataka State Cooperative Marketing Federation, through Cooperative Marketing Societies. These socie- ties are usually located in regulated markets at Taluk level where they have offices and storage facilities. Primary agricultural credit cooperative societies, functioning at village level, are members of marketing societies from which they obtain fertilizers, pesticides and seeds for sale and distri- bution to farmer members. Other outlets are the Department of Agriculture and private dealers. Input prices are controlled by GOK. Many farmers retain their own seed or buy or exchange seed with their neighbors. 2.24 Agricultural Credit. Institutional credit organization follows the normal pattern for India. The cooperative banks, which provide short and medium term loans mainly for agricultural inputs at an annual rate of 11-12% to the farmer are organized on a three tier basis with the State Cooperative Bank at the State level, 20 central banks at the district level and 5,000 primary cooperatives at the village level. The State Land Develop- ment Bank (LDB) is handling long term credit (up to 15 years) through 140 branches. Commercial Banks (CB) with 1,600 branches in the State, have just started entering the field of agricultural lending. Most banks, particularly LDBs, have had considerable experience of Bank Group supported lending, notably through the Mysore Agricultural Credit Project (Cr. 278-IN) and three Agricul- tural Refinance and Development Corporation (ARDC) Credit Projects (Credits 540-IN, 715-IN and 947-IN). Marketing, Processing and Storage 2.25 Marketing in Karnataka is regulated under the Agriculture Produce Marketing Act, 1966. Trade is controlled by the State Marketing Department through local Agricultural Produce Marketing Committees. At present, there are 115 main markets and 202 sub-markets under the control of the State - 14 - Marketing Department. 44 out of the 115 main markets are being further developed with the assistance of an IDA credit (Cr. 378-IN, 8.0 M, 1973). 2.26 Private processing facilities exist in most major producing areas. GOK is also encouraging the development of cooperative processing plants such as sugar factories and textile mills; or in some cases, silk factories and oil crushing units. The State Warehousing Corporation is developing storage facilities in all major market yards and the Central Warehousing Corporation has larger depots in the major centers. III. THE IRRIGATION SUB-SECTOR Water Resources 3.01 Three major river systems traverse the State. The Krishna, Godavari, and Cauvery--three of India's four largest rivers--originate in the Western Ghats and flow across the Deccan Plateau before discharging into the Bay of Bengal. They drain about 80% of the State's area. The remaining 20% of the land area lie in the basins of smaller west-flowing rivers, the North Pennar, South Pennar and Palar which rise on the western side of the Ghats and drain into the AraVian Sea. The total discharge of the State's rivers is estimated at 97,300 Mm , about half of which could be utilized. The potential for development of these large surface water resources, however, is limited, inter alia, because all the tributaries of the major rivers are seasonal, and even in the case of the perennial rivers more than 90% of the total flow occurs during the June-September monsoon when the need for irrigation is lowest. Also, the Krishna, Godavari and Cauvery Rivers all have a major part of their catchment in other States. The availability of water for Karnataka depends, therefore, largely on interstate agreements. Because plateau areas of the State are underlain by crystalline rocks, groundwater aquifers are low yielding and costly to develop. The State Groundwater Cell 3 estimates that the total groundwater potential of the State is about 11,700 Mm Irrigation Potential and Utilization 3.02 At present, the irrigation potential and its utilization in the State is roughly estimated as follows: Potential Water Presently % of Resource Developed Potential Mm3 M h /a Mm3 M ha/ Water Area Surface Water 47,000 4.00 19,000 1.50 40 37 Groundwater 12,000 1.80 1,800 0.40 15 21 59,000 5.80 20,800 1.90 35 33 /a In terms of gross irrigated area (i.e., including double cropped area). - 15 - The figures indicate the ample scope for further irrigation development from both surface and groundwater resources. Development of groundwater resources is undertaken mainly through private investments with financial support from the institutional credit system. Through the Mysore Agricultural Credit Project (Cr. 278-IN, 1971, US$40.0 M) and the line of credit to the Agri- cultural Refinance and Development Corporation (ARDC), the Bank Group has supported and still is supporting groundwater development in Karnataka. General Trends in Irrigation Development 3.03 Before Independence, the area under irrigation was less than 5% of the total area under cultivation. Large scale irrigation development was negligible and most of the irrigation was from tanks, small diversion weirs and dugwells. A large number of major and medium projects were constructed during the First and Second Five Year Plans (1950/51 to 1960/61). As a result, by 1960/61, about 27% of the total net irrigated area were under major and medium projects, 41% under tanks, 15% was irrigated from wells and 17% by other sources. 3.04 Between 1960/61 and 1977/78, the total net irrigated area grew at a rate of some 2.5% per annum. The expansion of irrigated areas under major and medium projects and wells accounted for most of this growth. Irrigated areas under tanks declined at annual rates of 0.6%. This was probably due to the silting of tanks and, in some areas, due to switching from tank to canal and well irrigation. Progress of irrigation development in terms of net irrigated area in the State since 1960/61 is shown below: 1960/61 1965/66 1970/71 1977/78 -- 000 ha------------- Major & Mediums 235.8 360.4 448.7 561.9 Tanks 343.7 329.3 364.8 347.5 Wells 132.8 163.1 459.5 356.7 Others 145.9 127.5 92.4 121.4 Total 858.2 975.3 1365.4 1387.7 3.05 By 1977/78, the share of irrigated areas under major and medium projects had increased to 40%, while those under tanks had diminished to 25%. Irrigated areas under wells accounted for about 26% in 1977/78 as compared to 15% in 1960/61, and other sources provided water for only 9% of the total irrigated areas. In spite of this growth, only about 13% of the net area sown in the State is irrigated. This proportion is one of the lowest in India and far below the national average of 30%. The gross irrigated area (i.e., including multiple croppi;ig) accounted only for about 15% of the gross cropped area. This is also less than half the national average. - 16 - Regional Pattern of Irrigation Development 3.06 There is a wide variation in the level of irrigation development in the State. The high rainfall areas (hilly and coastal zones) have a higher proportion of land irrigated than the lower rainfall areas (dry and transitional zones). In 1977/78, the level of irrigation development in different parts of the State was as follows: Net Irrigated Area Zones Rainfall (as % of Net Area Sown) (mm) Dry 450-900 12.9 Transitional 600-1,300 11.5 Hilly 1,000-6,400 20.7 Coastal 3,000-3,700 36.4 State 13.5 3.07 Between 1960/61 to 1977/78, virtually all of the increase in irrigated area occurred in the dry and transitional zones. This indicates that GOK's investment policy for irrigation has largely been oriented towards an equalization of the level of irrigation development in different parts of the State. Special emphasis has been given to projects in drought prone and economically backward areas. 3.08 An examination of the sources of irrigation indicates that the high level of irrigation development in the high rainfall areas is mainly due to private investments in groundwater, river lift schemes and both public and private investments in tanks. On the other hand, in dry and transitional zones, a large part of the irrigated areas developed were under major and medium schemes (canals). In 1977/78, the following pattern emerged: Zones Canals Tanks Wells Other Total --------- % of total ---------- % Dry 49 19 28 4 100 Transitional 35 30 26 9 100 Hilly 9 74 2 15 100 Coastal - 7 25 67 100 The Role of Irrigation in Karnataka's Agricultural Development 3.09 Between 1960/61 and 1975/76, the gross value of crop production 1/ increased by about 38%, or 2.2% p.a. The growth in production from irrigated areas existing in 1960/61 contributed about 18% to the overall increase, while 1/ At constant 1975/76 prices. - 17 - the balance of increases of production on new irrigated areas and correspond- ing losses of rainfed production on the same area contributed about 52%, and the growth of production on rainfed areas (as existing in 1960/61) contributed about 30%. The shift from rainfed to irrigated cultivation alone, even if yields had not increased during this period, contributed about 37% of overall growth. Yields on irrigated areas improved at a moderate rate of 1.9% p.a. whereas those on rainfed land grew only at about 1.3% p.a. (A detailed presentation of the structure of growth in both sectors is to be found in Annex 1, T-2). This indicates that the expansion of irrigation facilities and the associated yield improvements have been the major contribution to the growth of agricultural production. Bank Group Involvement 3.10 The Bank Group is directly involved in surface irrigation development in Karnataka through the Karnataka Irrigation Project (Cr. 788-IN, US$126.0 M, 1978). The project finances a five year time slice of the development of the Upper Krishna project, estimated to require 15 years for completion.' The project made a slow start, mainly due to procurement problems. -Tt has now picked up momentum and progress in construction of dams and main canals is satisfactory. Planning and implementation of the distribution system and command area development suffer from lack of experienced staff and inadequate cooperation between the Command Area Development Authority and the Irrigation Department. The road component suffers from inadequate funding as well as organization and staffing deficiencies. These issues have been raised with GOK and are being followed up. 3.11 The Bank Group also supported groundwater development through the Mysore Agricultural Credit Project (Cr. 278-IN, US$40.0 M, 1972) which was completed in 1977. The Performance Audit Report (No. 2892, 1980) on this project concluded that the project exceeded its target in terms of numbers of minor irrigation investments financed (51,000 wells and 27,000 pumpsets vs. 16,000 wells estimated at appraisal). Also against appraisal expectations, farmers preferred by far the more economical and less risky dugwells against borewells. Only half of the wells and about two thirds of the pumpset invest- ments financed under the project were completed at the end of the project period; most of the balance were to be completed under IDA's first and second line of credit to ARDC. In spite of a higher number of investments, total irrigated area increased only modestly due to a smaller benefiting area per well than was estimated at appraisal. Moreover, cost per well turned out to be higher than anticipated. Nevertheless, groundwater investments, where feasible, proved to be economically well justified and financially profitable to farmers. Through a third line of credit to the ARDC, funds from IDA are made available to refinance credits for private groundwater investments as well as land levelling and land shaping, through Land Development Bank and commercial banks. - 18 - IV. PERFORMANCE OF EXISTING TANK IRRIGATION PROJECTS General 4.01 Tanks were constructed as early as the 12th century. Many old tanks still exist and are operational. At present, there are about 41,000 tanks, widely dispersed throughout the State. Of these, about 25,000 tanks with a command area above four ha each are owned and operated since 1949 (after Independence) by the State's Irrigation Department (ID). Another 16,000 tanks, with less than four ha command area each, are under control of Taluk Development Boards. Together, the 41,000 tanks cover a nominal command area of about 850,000 ha. However, the actual irrigated area is probably not more than 350,000 ha. Design and Operation 4.02 As indicated in the introduction to this report, tanks are designed to catch and store runoff from streams and minor rivers during the rainy season. They serve the double purpose of protecting the standing crop from dry spells during the kharif season which in dry areas would destroy the crop, and to facilitate the growing of a second crop. Since catchment and command area are usually subject to similar climatic conditions, high irrigation demand occurs typically in seasons with low runoff from the catch- ment and therefore low water supply. Therefore, and since the protective use has absolute priority, a second crop is usually only possible in good rainfall years. Due to the drought protection objective, command areas have usually been defined as large as possible, so as to reach a maximum of beneficiaries. 4.03 The design of the reservoir (dam and overflow section) is usually based on empirical formulae, i.e., on standard correlations of annual rainfall and runoff, given the characteristics of the catchment area, rather than on actual rainfall/runoff relationships observed at the site (see Annex 6). This, of course, is almost inevitable given the number of tanks developed and the paucity in stream gauging stations. Only recently have actual measure- ments at a few tank sites been started. While nothing is known about the hydrological criteria used for very old tanks, actual performance indicates that the design of overflow structures seems to be satisfactory. 1/ Only very few tanks breach every year; in such a case, the damage is mostly confined to agricultural areas on the banks of the stream. A major problem for existing tanks is the silting of the reservoir, caused by unstable, over-cultivated catchment areas or by excessive forest clearing. Silting reduces the potential of the tanks to store water and is one of the reasons for the gap observed between areas commanded by the tank canal system and the actual irrigated area. 2/ 1/ More recent engineering practice makes use of an empirical formula to estimate peak runoff (Ryve's Formula), or unit hydrographs for tank catchments. 2/ The criteria for implementation of the proposed project stress the need for adequate catchment protection works in unstable catchment areas (para 5.08). However, since siltation affects "dead storage" first, no adverse effects are expected from such process before about 50 years of project life. - 19 - 4.04 The distribution system in most cases consists of unlined main canals (usually two, one on each side of the stream) with outlets leading either directly to the fields or to long, poorly shaped and uncontrolled field channels. Main canals are usually rather long (up to 15 km), in order to serve a generously defined command area, which results in high conveyance losses. Where there are only few cross regulators and gated outlets, opera- tional staff is unable to regulate the water supply adequately. Farmers at the head of the system or of a particular outlet command are free to take more than their share of water. Water often flows from field to field towards the main drain, even in the case of crops other than paddy, not permitting control of amount and timing of water application. These and other "operational" losses are estimated to account for about 20% of overall losses. 4.05 Unlined channels cause excessive seepage losses. A recent study in three tanks in gravelly soils shows that losses are in the range of 63 to 78 cusecs per million square feet, a range of values many times larger than the normally assumed losses (about 8-10 cusecs per million square feet). Along the entire length of a channel, 40-80% of the head discharge are lost. As a result, no water reaches the middle and tail-end fields in the command areas. 4.06 Operation of tanks is presently under the instruction of a Govern- ment irrigation officer, who consults with the Taluk Development Committee (a committee consisting of concerned government staff and local farmers' repre- sentatives). The officer has the authority under the Karnataka Irrigation Act (1965) to specify time, period and quantity of water releases as well as the area to be served. Such decisions are made public for each tank before the irrigation season and are known to the farmers. A canal inspector, or Nalamanegar, from the staff of the Irrigation Department, regulates the supply at the gated reservoir outlet, while further distribution among outlets to field channels is carried out by auxiliary staff, called Sowdies (also employed by the Government). In larger tanks, each Sowdi is responsible for a section of the canal serving about 200 ha. Night irrigation occurs in time of heavy demand (paddy pre-planting and transplanting periods). Farmers generally know the day and time of water supply to their outlet, but are not aware of the exact time and quantity to be supplied to their individual holdinigs. Although there are no formal organizations within a tank command, informal arrangements for water-sharing exist inside the outlet command. However, these arrangements are precarious and often result in inequitable distribution of water. 4.07 Water supply is rotated in many tanks between groups of outlets for periods at one to two week intervals in order to maximize flows to the lower reaches of the system. However, with the prevailing features of the canal system, sufficient flow to tailenders is the exception rather than the rule. Where double cropping is only possible for part of the command area, supply of water is rotated betwe2n parts of the system (usually along one of the main canals) on a yearly basic,. 4.08 Maintenance of reservoir structures and canals is the responsibility of the Irrigation Department. The farmers do not contribute to these works directly; however, they have to maintain their own field channels. Due to a low density of supervisory staffing, and due to the unlined canals and - 20 - channels, maintenance of the distribution network is often less than satis- factory, leading to higher operational losses. Higher standards of main- tenance can be expected for lined systems and with more participation and interest in a properly functioning system on the part of benefiting farmers. Project Formulation 4.09 About 54% of the State's geographical area is drought prone compared to 16% of the total area in India. While, on average, only about 10% of the population in India live in drought prone areas, at least 50% of the popula- tion of Karnataka live in such areas. Furthermore, only about 13% of the net area sown in the State is irrigated, which is one of the lowest ratios in India and far below the national average of 30%. Therefore, expansion of irrigation facilities is one of the highest priorities in GOK's agricultural development plan. Given the limited scope for further groundwater develop- ment, 1/ GOK plans to accelerate development of surface irrigation projects. Within the latter, minor surface irrigation projects have a high priority because of their effect on regional equity and rural poverty. 4.09 During the current Five-Year Plan (1980/81 to 1984/85), GOK will allocate about US$126 M of its development budget to minor irrigation programs. Out of this, about US$18 M will be spent during 1980/81 for completing on- going minor schemes. The remaining US$108 M will be used for constructing new tanks, lift irrigation schemes and for repair of existing schemes. GOK has proposed to IDA a project consisting of the construction of about 120-160 new tanks covering an aggregate area of some 25,000 ha at a cost of US$77 M during four years (1981/82 - 1984/85). The reason for including only new tank irri- gation schemes under the proposed project is to experiment with new planning and design criteria, the effect of which would be carefully monitored and evaluated. Lessons learned would subsequently be applied to the modernization of existing tanks. V. THE PROJECT Project Objectives 5.01 The proposed project is formulated with the following objectives: (a) spread irrigation to areas where large scale irrigation development is not feasible; (b) achieve more economic use of scarce surface water resources; (c) ensure a more reliable irrigation supply to consumers; (d) build up GOK's capability to prepare, appraise, supervise and evaluate irrigation projects; 1/ Groundwater development is also mainly a private activity, albeit supported by subsidized institutional credit. - 21 - (e) increase farmers' participation in the operation of tank irrigation projects (TIPs); and (f) improve the hydrological data base for minor surface irrigation projects. Project Components 5.02 The proposed project would finance: (a) Construction, over a four-year period, of 120-160 TIPs throughout the State, estimated to irrigate some 25,000 ha; (b) a rainfall and stream gauging network to improve the hydro- logical data base for the planning of future minor irrigation projects; (c) development of an improved model for catchment runoff estimation; and (d) a study of possible approaches to the formation of water users' organizations. Description of a Tank Irrigation System 5.03 Under the proposed credit, individual TIPs would cover cultivable command areas of between 20 and 2,000 ha. Each TIP would generally consist of: (a) an earthfill embankment to create a reservoir, with an ungated spillway and head regulator(s); (b) a fully lined canal network which would deliver water through outlets serving approximately 8 ha blocks; (c) a system of normally unlined field channels within the 8 ha blocks; and (d) a drainage network connected to natural drains. Project Scope and Unit Cost 5.04 Project scope, i.e. the total area to come under irrigation through the proposed project, depends on the average cost per ha. Cost per ha is subject to some uncertainty, since only a fraction of the planned TIPs have been fully designed and costed. 1/ Analysis of available cost estimates shows 1/ GOK has prepared nine model projects, one for each agroclimatic zone of the State (except Northern Transition Zone). Out of the nine, six proj- ects have been selected as typical. Their features, design criteria and cost estimates are shown in Annex 1, T-4 and T-5. Some modifications to cost estimates were made by the mission to reflect improved design standards for the distribution and drainage systems. - 22 - considerable variation between TIPs, even between TIPs of similar size, in terms of cost per ha as well as cost per unit of water deliverable at the tank outlet. On the basis of available information, and of separate cost estimates for the reservoir (embankments and overflow) and for the distri- bution system for various sizes of TIPs, a typical cost estimate for five different size groups of TIPs has been developed (Annex 1, T-3). More extreme deviations from the assumed norm due to particular topographical conditions, have been disregarded to arrive at this estimate. The cost of the tank itself accounts typically for about 75-80% of the total cost of a TIP, the cost of the distribution and drainage system accounting for the rest. However, cost per ha of the reservoir varies considerably with size, decreasing for larger units to about 60% of project cost. This economy of scale outweighs by far a higher per ha cost of the canal system in larger projects, resulting in a lower per ha total cost for larger TIPs. With the social objective of spread- ing the benefits of the project more widely in a geographical sense, especially in drought prone areas, GOK intends to build about one third of the TIPs in the size group between 20-100 ha, another third between 100 to 200 ha, and the balance larger than 200 ha. The average size of the TIPs financed would be about 230 ha. Depending on the actual cost of the projects built, the number of tanks financed might be reduced, with possible changes also in the size distribution. The weighted 1/ average cost per ha of CCA (in 1980 prices) is estimated at about Rs 21,300. This includes physical contingencies and overhead for design and supervision, as well as the cost of land acquisition. The estimated unit cost is higher than for medium and large surface irrigation projects (Rs 11,000-13,000/ha), firstly because tanks per se are somewhat more costly than larger schemes, and secondly because of the deliberate policy of including a larger number of relatively small TIPs in the construction program. However, smaller schemes have considerably shorter gestation periods, about one to one and a half years for the smallest (20-100 ha). (The average construction period is assumed to be about 3 years.) Also, TIPs have generally higher conveyance efficiencies than medium and large schemes, and are more manageable in terms of efficient operation.3 Assuming an annual water avail- ability at the reservoir of about 10,800 m /ha of CCA, the average cost per cubic meter of water is estimated at about Rs 1.97. Given the funds budgeted for the Tank Irrigation3Program, 2/ the project would provide water to some 25,000 ha (about 270 Mm in an average year). 5.05 The proposed regional distribution of TIPs would favor the dry zones of the State with 52% of the proposed CCA and 65% of the projects, and the transitional zones with 36% of proposed CCA and 29% of the projets. The balance would be in the coastal and hilly zones. The tentative regional and size-wise distribution of TIPs is shown in Annex 1, T-6. 1/ Weights being derived from the size distribution intended by GOK. 2/ About Rs 524 M after allowing for price contingencies (23%). This does not include expenditures for other project related works or operations, such as a catchment protection program or resettlement of displaced families, to be financed from other budget funds. - 23 - Criteria for Tank Irrigation Projects 5.06 General. In view of the proposed dispersion of the tank project throughout the State, detailed appraisal by IDA of each tank would be quite impractical. Therefore, criteria for planning, design, implementation, operation and economic viability have been developed and discussed between GOI, GOK and IDA for projects to be financed from the credit. An assurance has been obtained from GOK that TIPs financed under the credit would follow the agreed criteria. The criteria are described in detail in Annex 2 and are summarized below: Planning Criteria 5.07 Basic Data Requirements for a Tank Project: (i) topographical map; (ii) reconnaissance type soil survey and land classification dividing the area commanded into three categories: areas suited for paddy, areas suited for both paddy and upland crops, and areas suited for upland crops only; (iii) data for three years (minimum) of rainfall in the catchment and its correlation with actual streamflow data for the tank site (exceptions see under para 5.08 (i) below; (iv) sediment samples to be taken at or near each tank site for determination of expected sediment levels in the reser- voir for all tanks with more than 1,000 ha CCA. 5.08 Hydrology. Actual streamflow data observed at the tank site and correlated with observed rainfall in the catchment over a period of at least three years should be used wherever possible to estimate the annual and monthly design runoff from a catchment. In order to improve the hydrological data base for TIPs to be constructed in the last years of the proposed project as well as for schemes to be built thereafter, a long term streamflow and rainfall gauging program would be started immediately (paras 5.25-5.28 below). In cases where actual measurements of streamflows are not yet available (during the first years of the project) or where they will not be available in the future (for small TIPs), runoff estimates should either be based on data from nearby catchments with similar characteristics for which observations are available, or on the use of Strange's method (see Annex 6) after proper classification of the catchment. A more accurate system of runoff yield estimation, based on "continuous water balance methods," would be developed and adapted to Karnataka conditions in the first year(s) of the proposed project (para 5.29 below). It would, when successfully implemented, replace the methods mentioned above for ungauged or poorly gauged catchments. The hydrological basis for the planning and design of TIPs would thus be considerably improved after two to three years. 5.09 In addition, rain gauges should be installed in the catchment area of all TIPs serving more than 150 ha and daily rainfall records should be main- tained. (Self recording gauges for TIPs with more than 1,000 ha command.) For other TIPs, the nearest existing rain gauging station should be used. Where existing tanks are located in the catchment area of a proposed TIP, a 25% return flow can be assumed from an existing upstream irrigation scheme. The impact of the proposed TIP on existing downstream schemes should be care- fully assessed. For an estimate of the net impact, a 25% return flow from the proposed TIP can be assumed. If the estimated reduction of inflows to - 24 - the existing downstream scheme would exceed 5% of the design yield for that scheme, the proposed TIP would have to be submitted to IDA for review and approval. 5.10 Otb>r Planning Criteria. (i) To avoid erosion of the catchment and silting of tanks, the need for catchment protection works should be assessed, based on maps depicting slopes, soils, land capability, erosion and vegetative cover of the catchment. Appropriate measures should be outlined and included in the project proposal. They would be part of the TIP and eligible for reim- bursement from the proposed IDA credit. (ii) GOI's project preparation and monitoring cell has prepared, in cooperation with GOK, standard cropping pat- terns, which should be suitably modified for each TIP in order to take account of soil surveys and the existing cropping pattern in the proposed area; special care should be taken against an under-estimation of the probable paddy area in the future with project. (iii) The project formulation cells established under the project should, in cooperation with concerned departments, carry out surveys on the existing agricultural supporting services in the area and propose arrangements for any necessary strengthening of such services. 5.11 Water Requirements: (i) The modified Penman method 1/ should be used to estimate crop water requirements; (ii) monthly evapotranspiration rates (ETo) for each agroclimatic zone should be taken from zone-wise ETo estimates prepared by GOI's Project Preparation and Monitoring Cell; (iii) based on long-term rainfall records in the proposed command area or at a nearby station, effective rainfall would be calculated as 50% of the 50% dependable rainfall during the monsoon, and as 65% of such rainfall during the non-monsoon season; (iv) irrigation requirements for paddy should include allowances for land preparation, nursery and transplanting (200-250 mm), deep percolation (2-4 mm/day) and for maintaining a water depth of about 50 mm; (v) preplanting irrigation requirements should be scheduled for upland Kharif and late Rabi crops; (vi) the conveyance efficiency should not be assumed higher than 85%; (vii) assumed field irrigation efficiencies should be based on soil characteristics and should not exceed 90% for paddy and 70% for diversified (upland) crops, (viii) a month-by-month tank operation study for the design year should be carried out. 5.12 Dams, Embankments and Appurtenant Structures. Criteria and stan- dards for design and safety of earthen dams laid down by GOI and GOK and applicable to small dams should be observed in addition to sound and accepted engineering practice. Planning, design and construction of larger dams would be subject to recommendations and approval by a Dam Safety Panel (see paras 5.22-5.24 below). The design of overflows structures should also be based on results from the improved model for runoff estimation (continuous water balance methods), once this is developed, since it would be capable of more accurate estimates of peak runoff (para 5.29 below). Design Criteria 5.13 Distribution System: (i) Design and Layout of canaLs and iield channels should be completed down to the individual holding; for very small 1/ FAO Irrigation and Drainage Paper No. 24. - 25 - holdings, water would be supplied to I ha; (ii) the head capacity of supply channels should be designed according to a formula which takes into account the expected mix of paddy and upland crops (between 0.7 liter/second and 1.2 liter/second per ha); (iii) the design should permit rotational water supply and small systems should be designed for daylight irrigation only; (iv) the complete system should be lined down to turnouts serving about 8 ha subunits. 1/ Lining should be extended, if necessary, so that the distance between lined channels and farm gates does not exceed 300 m; (v) part of the field channels should be lined in case of permeable soils or of slopes exceeding 2%, so that the unlined length of a channel does not exceed 100 m; (vi) all outlets and turnouts should be gated; (vii) field channels should have adequate structures, especially protective drop structures; (viii) measuring devices should be installed at the outlets from the tank and main channels and at each 1 ha outlet; (ix) cross regulators in the main channel would permit full discharge at take-off points when the channel is flowing at less than 50% capacity. 5.14 Drainage System: (i) Adequate cross drainage works should be built along the main channels, as well as cut-off drains on the up-hill side of such channels; and (ii) in the command area, a system of drainage channels designed to drain a runoff at a rate of three liter/second/ha should be built and connected to natural drains. 5.15 Communication: (i) An all-weather access road should be provided to the tank; (ii) according to size of channels, all-weather service roads or motorcycle/bicycle paths would be built along main and distributary channels. Implementation Criteria 5.16 Construction Practices: (i) Canal embankments and earthfill around structures should be thoroughly compacted by mechanical equipment; and (ii) to ensure proper settlement of foundation in minor channels, sections would be thoroughly saturated by ponding or exposure to one season of rainfall before lining. 5.17 Implementation: (i) Construction periods should be planned not to exceed two years for smaller TIPs (< 400 ha), three years for medium size TIPs (< 1000 ha) and four years for larger ones, excluding time needed for prelim- inary works such as construction of approach roads, which should not last longer than one year; (ii) large TIPs should be commissioned in blocks not more than 400 ha in size; (iii) water should not be released until all field channels have been completed and the TIP (or a section of it) has been com- missioned; (iv) lining should be tested by ponding before commissioning; if seepage losses of more than 3 cusec/M sq ft occur, lining should be rectified; and (v) commissioning would include tests to determine whether water reaches all parts of the CCA. 1/ An economic justification of lining is given in Annex 8. - 26 - 5.18 Cost Estimates: (i) Estimates should be based on current depart- mental rates or recent local bid prices; (ii) they should include physical contingencies of 10% for work already designed in detail and 20% for all other works, and (iii) should include cost of land acquisition, settlement of displaced families (if any) and of approach and service road^. Operation Criteria 5.19 Water Release: (i) Discharge from outlets/turnouts to field channels should normally be about one cusec, and not less than one half cusec; (ii) water release to farmers should initially follow a fixed schedule 1/ prepared for each season; and (iii) releases for upland crops should be rotated every ten days. 5.20 Tank Irrigation Committee (TIC). A TIC should be formed imme- diately after construction of a tank has been sanctioned. The TIC should be consulted by the Irrigation Department (ID) before finalizing the design of a TIP. Although the ID would keep, for the time being, full responsibility for the physical maintenance of the system above the irrigation outlets, the TIC would share with the ID the responsibility for operation (water schedul- ing) and possibly maintenance of the system. Until the proposed Study on Water User's Organizations (para 5.26) would be completed, GOK should, on an experimental basis, initiate formation of such committees in at least three selected new TIPs. Economic Criteria 5.21 Individual TIPs should have an economic rate of return exceeding 12%, based on estimated annual costs and benefits discounted over 50 years. To facilitate economic evaluation, total financial cost per cubic meter of water available in the design year at the reservoir outlet should be compared to a critical limit derived and tabulated for the agroclimatic zones by IDA (Annex 2, Appendix). (These cut-off rates should be increased each year by the estimated rate of inflation for civil works.) A TIP could be approved if its unit cost of water is below the critical limit for the respective zone. Critical cost limits were derived in such a way as to assure a minimum rate of return of 12% for the individual TIP, based on benefits projected for each agroclimatic zone by IDA. GOK may submit a TIP to IDA for special approval in cases when GOK deems the project justifiable although its cost exceeds the cut-off rate. Dam Safety 5.22 Design and construction standards for the comparatively small dams of most TIPs would follow sound engineering practice, and adequate quality control would be providel under the proposed organization (para 6.03 below). 1/ A mote flexible water supply schedule should be started after an initial period of fixed rotations. The more flexible schedule would be geared to the needs of the actual crops grown below each outlet. The Tank Irrigation Committee would gradually involve itself in this more sophisticated scheduling. - 27 - However, there would be a number of TIPs which due to their larger size would warrant special attention to design and construction standards of dams and related structures, as well as to regular maintenance. A special Dam Review Panel has been constituted by GOK to undertake periodic and comprehensive reviews of plans, designs and construction procedures of these iarger dams or embankments with the objective of ascertaining and reporting on their general safety and performance. In the following cases this review would be necessary: (a) if the proposed height of the embankment above streambed exceeds 10 m, or; (b) if the gross storage behind the embankment exceeds 2.5 Mm at full reservoir level, or; (c) if the proposed location of the embankment is such that significant destruction would occur in case of its failure. 5.23 The Panel would review and evaluate all data and design and construc- tion procedures and make specific recommendations to GOK concerning embankment safety, foundation, abutment integrity and spillway adequacy. This review would be required specifically at the time when foundation preparation work has been completed and before embankment placement has progressed very far, and at such other times during construction as is necessary or desirable. The Panel's role should not be viewed as one of assisting or guiding the executing Irrigation Department, since responsibility for proper design and construction, including quality control, rests with GOK. Rather it should be considered as a review body of technical experts whose judgment and opinions on particular technical matters would carry considerable weight in the evalua- tion of proposed actions by the project administration. Agreed terms of refer- ence (TOR) for the Panel, which have been more or less standardized for other IDA-financed projects, are included as Annex 4. The TOR require that the Panel submit a written report containing their review, comments, and recom- mendations within 60 days following each convening of the Panel. The report should be made available to IDA and to such bodies within GOI which may be assigned safety-of-dam responsibilities for their review. 5.24 An assurance has been obtained from GOK, that it would maintain its Dam Review Panel with TOR, experience and qualifications agreed with IDA for the above mentioned purposes. Any costs incurred for activities of the panel in relation with the project would be eligible for reimbursement under the proposed credit. Rainfall and Stream Gauging Network 5.25 To improve the data base for planning of future TIPs, about 1000 new fiberglass raingauges would be installed to supplement approximately 220 raingauge sites that presently meet Indian Meterological Department (IMD) standards. At present there are only 31 self recording raingauges in the State; according to IMD standards about 10% of the 1,200 sites would be self recording. The raingauge sites would be installed in small catchments which already have or for which there are proposals to install stream gauging stations. - 28 - 5.26 There are presently 124 gauge-discharge measuring sites on streams within the State; discharges at 89 sites are measured with current meters and 35 by recording reservoir levels. However, most of the above sitef have been established in catchments with drainage areas greater than 5000 km . In order to improve the data base for smaller catchments it is proposed to add 100 new stream discharge measuring sites throughout the State. Appropriate equipment would be financed under the proposed credit (Annex 1, T-7). 5.27 As an additional basis for estimation of evapotranspiration rates for various agro-climatic zones within the State, to be used in crop water requirement calculations, it is recommended that six new hydrometerological stations be established within the State to supplement presently available data. There are presently 45 such stations within the State. In addition, evaporation data should be supplemented in six of the existing sites by installing Class "A" pan evaporimeters. Equipment for this purpose would be financed from the credit (Annex 1, T-7). 5.28 It is proposed that the raingauge and hydrometerological stations be operated and maintained by the IMD while the stream gauging program be carried out by the Irrigation Department. The data is critically needed to improve estimates of water supplies by allowing the introduction of new methodologies. Improved estimates of water supplies will assure a sound investment program in the future. Development of an Improved Hydrological Forecasting System 5.29 Improvements in the estimation of catchment runoff over traditional empirical formulae can be made through use of modern hydrological techniques and computing equipment. The development of "continuous water balance" simulation/estimation models has produced excellent results in many parts of the world. Such models take into account daily distribution and intensity of rainfall, rather than monthly or annual aggregates only, as well as such catchment characteristics as potential evapotranspiration, soil properties, vegetation, topography, drainage channels and man-made changes. Recent tests of a continuous water balance model in the Narmada Basin of Madhya Pradesh State have shown its usefulness for tropical monsoon environments. In order to make this technology available to Karnataka for the proposed project and other future projects, a computerized hydrological forecasting system would be developed, tested, calibrated and commissioned under the coordination of the Karnataka Engineering Research Station at Mysore and with technical assistance from an experienced hydrological institute, under terms of reference to be agreed with IDA. Simultaneously, training would be provided by the institute for about two local hydrologists in the use of the model and the computing equipment. The cost of technical assistance and training is estimated not to exceed about US$70,000, the cost of necessary computing equipment (desk- computer and plotter) to be about US$35,000. These costs would be reimbursed from the proposed credit. Water Users' Organization Study 5.30 One of the objectives of the project is to increase farmers' parti- cipation in the operation of TIPs. A study of possible approaches to water users' organizations would be undertaken and completed by March 31, 1983. It - 29 - would investigate existing farmers' organization and cooperation within tank command areas, in order to recommend means for establishing viable and effec- tive participation of irrigators in operation and maintenance of TIPs financed under the project. Under IDA supervision, the study would be carried out by the Indian InLtitute of Management (IIM) in Bangalore, with assistance from the Minor Irrigation Wing Staff of the Irrigation Department. The terms of reference for the study have been developed and discussed between GOK and IDA, and are given in Annex 3. Project Costs and Financing 5.31 The proposed project would support a four year time-slice of the GOK construction program of about 120-160 new TIPs. The proposed IDA credit would finance expenditures on TIPs planned, designed and constructed in accordance with the agreed criteria (paras 5.07-5.21) during the period 1981/82 to 1984/85. Total project cost would be Rs 650 M (US$77.4 M) with a foreign exchange component of Rs 84.5 M or about 13% of total cost. Taxes and duties are insignificant and therefore not included. Price contingencies have been estimated as about 23% of base cost for the tank construction program. Price contingencies for equipment are estimated as about 10%. They are negligible for technical assistance (study and forecasting model). % of Base Local Foreign Total Local Foreign Total Cost --- Rs M--------- -------US$ M- Construction of TIPs 456.3 68.3 524.6 54.3 8.1 62.5/a 99.4 Rainfall and Stream Gauging and Computing Equipment 3.0 0.5 3.5 0.4 - 0.4 0.5 Technical Assistance and Training 0.8 - 0.8 0.1 - 0.1 0.1 460.1 68.8 528.9 54.8 8.1 63.0 100.0 Price Contingencies 105.4 15.7 121.1 12.5 1.9 14.4 565.5 84.5 650.0 67.3 10.0 77.4 /a Rows/columns may not add due to rounding. 5.32 The proposed IDA credit of US$54.0 M (SDR 43.5 M) would finance about 70% of total project cost including the foreign exchange component (US$10.0 M) and about 65% of local cost. All expenditures would be met by GOK from its development budget which includes a GOI contribution. An assurance has been obtained from GOK that it would promptly provide funds to execute the project in accordance with the schedule of expenditures (Annex 1, T-9). IDA would annually exchange views with GOK on the implementation program for the construction of TIPs. The proposed allocation of the proceeds of the credit is presented in Annex 1, T-10. An assurance has been obtained from GOI that proceeds of the credit would be channeled to GOK in accordance with standard arrangements for development assistance with the States of India. - 30 - Procurement 5.33 Civil works under the project would be individually small and scat- tered over the whole State. Construction would be seasonal and carried out through labor-intensive methods. Consequently, it would neither be feasible nor economic to group them into large contracts. Present practice of GOK is to award most of the construction works for dams, spillways and main canals after local competitive bidding (LCB), and the practice would be followed under the proposed project. The Central Water Commission of GOI has recently developed guidelines for procurement procedures for civil works which have been approved by IDA. GOK would adopt these guidelines for procurement under the proposed credit. For contracts costing more than the equivalent of US$200,000, review by IDA of tender documents and award recommendation would be required prior to award. It is expected that due to the remoteness and small size of some works stand&rd LCB procedures would not attract any contrac- tors' response. In these cases, unit price contracts would be awarded on the basis of rates established by competitive bidding each year. Some minor works would be carried out through GOK's Employment Affirmation Scheme. Departmental work would be used only when required by safety or quality considerations. The stream and rain gauging equipment and the computing equipment would be procured under LCB procedures acceptable to IDA. Disbursement 5.34 Disbursements would be made for: (i) 100% of foreign exchange cost for goods directly imported; (ii) 100% of local manufactured goods procured ex-factory; (iii) 70% for other locally procured goods; (iv) 100% of expendi- tures on technical assistance and of cost of training; and (v) 85% of expendi- tures on civil works. Disbursements against payments less than Rs 300,000 for civil works and less than Rs 150,000 for equipment would be made against statement of expenditures submitted by GOK. The same arrangement would be applied to disbursements on departmental work, without cost limit per payment. Supporting documents for these payments would be retained by GOK and made available for inspection by IDA during the course of review missions. The procedures for statements of expenditures have been well established under the on-going Karnataka Irrigation Project (Cr. 788-IN) and are satisfactory to IDA. The estimated semi-annual disbursement schedule is presented in Annex 1, T-ll. It is expected that disbursements would be completed by March 31, 1986. Accounts and Audits 5.35 The project would be subject to normal Government control and audit- ing procedures which are satisfactory. A resident Audit Officer, representing the State Accountant General would audit the project accounts. Statements of account certified by the Accountant General or his designee supported by an itemized account and a summary of expenditures would be submitted to IDA as soon as possible after the end of each fiscal year. Assurances have been obtained from GOK that: (i) it would maintain separate accounts on project expenditures; (ii) it would ensure that these accounts would be audited annually by the Accountant General; (iii) the resulting reports would be submitted to IDA as soon as possible but not later than nine months after the end of each fiscal year; and (iv) that it would make complete accounts and financial statements available for inspection during IDA review missions. - 31 - VI. ORGANIZATION AND MANAGEMENT The Irrigation Department 6.01 The Irrigation Department (ID) is headed by a Special Secretary, Irrigation who reports directly to the Chief Secretary, GOK. The department's organization is flexible and adjusted to the workload. At present, it has ten branches, each headed by a Chief Engineer with functional and/or regional responsibilities. Branches important for the proposed project are: (a) The Water Resources Development Organization (WRDO) with headquarters at Bangalore is entrusted with general estab- lishment matters of the Irrigation Department and other items such as water disputes pertaining to Krishna, Godavari and Cauvery, river gauging, preparation of the Water Year Book, maintenance of water accounts of river valleys, investigation, planning, design of major and medium surface irrigation projects, and inspection of machinery and equip- ment of the Department all over the State. There are 3 circles and 4 divisions, with sub-divisions in the field. (b) The Karnataka Engineering Research Station acts as an advisor and consultant in solving the problem of field engineers in charge of irrigation and hydroelectric projects and as test house for field problems. Its activities include design and research on dams and water conveyance systems, studies on return flow and silting of reservoirs, as well as studies on crop water requirements. The station has 5 divisions. (c) The Minor Irrigation Wing is in charge of planning, imple- mentation and operation and maintenance of irrigation schemes with less than 2,000 ha command area (minor schemes). It is divided into two regional branches, each headed by a Chief Engineer: (i) Minor Irrigation (South) oversees minor schemes in 11 districts. Its headquarters are at Bangalore. (ii) Minor Irrigation (North) is in charge of minor schemes in the remaining 8 districts of the State. Its head- quarters are at Dharwar. The Minor Irrigation Wing (MIW) 6.02 Existing Organization. The direct responsibility for planning implementation, operation and maintenance of tank irrigation projects (TIPs) would rest with the Minor Irrigation Wing of ID. As mentioned above, the MIW has two Chief Engineers who report directly to the Special Secretary, Irrigation. The two Chief Engineers are assisted by 10 Superintending Engineers (SEs). Six are in charge of investigation, construction, operation and maintenance (OEM), and monitoring of the minor irrigation works; two are responsible for preparing designs and cost estimates. Each Superintending - 32 - Engineer is in charge of 2-3 districts and has about 3-8 divisions headed by Executive Engineers (EEs). Each-EE has about 5 subdivisions headed by Assistant Engineers (AEs) and is supported by engineering and clerical staff. At present, the MIW has 2 design cells, 8 investigation divisions and 36 construction and O&M divisions dispersed throughout the State. 6.03 Proposed Organization (Annex 1, C-1). To improve the implementation capacity of the Minor Irrigation Wing under the project, a Project Formulation Cell (PFC) has been established in each of the offices of the two Chief Engineers, Minor Irrigation. The two PFCs would prepare tank projects for appraisal according to criteria (paras 5.07-5.18) agreed between GOK and IDA. In addition, two Quality Control Divisions under the supervision of SEs, Design have been established to ensure appropriate standards in the construc- tion of tank projects. Staffing of these divisions and cells will be completed by March 31, 1981. The Minor Irrigation Committee 6.04 Under the chairmanship of the Special Secretary, Irrigation, a Minor Irrigation Committee (MIC) consisting of the Deputy Secretary, ID; the Chief Engineers, Minor Irrigation; the Chief Engineer, Water Resource Development Organization; the Director of the Karnataka Engineering Research Station; the Director of the Department of Agriculture and experts from the Indian Institute of Management, the Institute of Social Change and the University of Agricultural Sciences would be established to be responsible for appraising and supervising TIPs. The MIC would supervise the works of a Special Appraisal and Supervision Circle (SASC) headed by a Superintendent Engineer who would be also Secretary of the MIC. He would be supported by four divisions: two Appraisal Divisions (AD), a Technical Monitoring and Evaluation Division (TMED), and an Operational Monitoring and Evaluation Division (OMED). The proposed organization of the MIC is shown in Annex 1, C-2. 6.05 The two ADs would be responsible for appraising the proposed TIPs. Each of these two divisions will be staffed with one EE, three to four AEEs and an Agriculturalist (Annex 1, C-2). The TMED would supervise TIPs under implementation and would initiate technical review and evaluations of the agreed criteria (assumptions on irrigation efficiencies, design of the irrigation and drainage networks and construction methods). It would liaise with the Institute of Engineering in Karnataka to incorporate results of technical research into the implementation of TIPs. The OMED would monitor the agro-economic aspects as well as O&M of TIPs under operation. In addition, it would carry out surveys to assess the effectiveness of water allocation procedures and the appropriateness of the recommended cropping patterns. In cooperation with local agricultural authorities it would monitor the develop- ment of crop yields under the project for operating TIPs and induce appropriate measures to be taken in the case of adverse developments. It would also liaise with other research and evaluation agencies in charge of overall evaluation of the minor irrigation program. The staff of these two divisions will include an Executive Engineer, 4-5 Assistant Engineers; an Agricultural Economist, and 2 Statistical Officers (Annex 1, C-2). The MIC and SASC have already been established. Staffing for the latter will be completed by March 31, 1981. - 33 - Procedures for Processing of Tank Irrigation Projects 1/ 6.06 Project Identification. Potential sites for tank irrigation projects (TIPs) below 1,000 ha would be identified by the existing 8 Minor Irrigation Investigation Divisions. TIPs above 1,000 ha would be identified by the existing 4 Major Irrigation Investment Division under the Water Resources Development Organization. Preliminary reports will be submitted to the concerned Chief Engineers (CEs). After consultation with concerned District Offices, the CEs will approve full scale investigation and prepara- tion of TIPs. 6.07 Project Preparation. Hydrological, geological, topographical and soil investigations would be carried out by the Irrigation Investigation Divisions (Minor and Major). These divisions would also be responsible for detailed designs of the reservoirs and the irrigation systems. The two Project Formulation Cells, which would prepare projects for appraisal according to criteria agreed between GOI, GOK and IDA, would be responsible for producing Project Preparation Reports (Annex 5A). Each report would include designs, agricultural plans, costs estimates and a tentative imple- mentation plan. These reports would be cleared with GOK's concerned depart- ments at the District level as well as at the State level. After clearance, the projects would be submitted for appraisal. 6.08 Project Appraisal. The two Appraisal Divisions of the Special Appraisal and Supervision Circle (SASC) would be responsible for appraising the proposed TIPs at the sites and ensure that the designs are technically sound, the hydrology and geology as well as designs of large dams have been reviewed by the Dam Safety Panel and its recommendations have been followed; the proposed cropping patterns are suitable for the command area; design and operation of the system have been discussed and agreed with farmers; arrange- ments for strengthening of agricultural supporting services are adequate; implementation schedules are realistic and the projects are economically viable. The Appraisal Divisions would confirm and adjust cost estimates as necessary and prepare Project Summaries (Annex 5B) to be submitted to the MIC. 6.09 Project Approval. On the basis of information provided by the Project Formulation Cells in the Project Preparation Report and by the Appraisal Divisions in the Project Summary, the Minor Irrigation Committee (MIC) would approve or disapprove the proposed TIPs. An assurance has been obtained from GOK, that project summaries of approved projects and minutes of all MIC meetings would be promptly furnished to IDA. 6.10 Project Supervision. The following activities would be carried out: (a) Annual Implementation Review. The Superintending Engineer (SE) of the SASC would prepare an annual report to list TIPs under preparation and implementation (Annex 5C). Special attention would be given to budgetary allocations and staff availability for each TIP in order to ensure that adequate funds would be provided in accordance with implementation schedule. 1I/ For a graphic presentation of the project cycle, see Annex 1, C-3. - 34 - (b) Annual Tank Performance Review. The Superintending Engineer of the SASC would also prepare an annual summary of the performance of TIPs built under the project, including data on operation and maintenance, on areas actually irrigated and on cropping patterns (Annex 5D). Annual implementation programs and annual performance summaries would be submitted to IDA by Ma-ch 31 of each year. An assurance to this effect has been obtained from GOK. (c) Progress Review: Technical Implementation. The Technical Moni- toring and Evaluation Division (TMED) would be responsible for supervising TIPs under implementation. Because of the large number of TIPs scattered throughout the State, field visits to each one will not be possible. Consequently, TMED would visit every TIP above 1,000 ha and above 25% of TIPs below 1,000 ha twice a year. It would submit to the MIC a supervi- sion summary for each TIP visited to report the progress of construction (as compared to planned implementation schedules), change in costs, if any, compliance to agreed criteria and procurement problems encountered during construction, if any (Annex 5E). For the remaining 75% of TIPs below 1,000 ha, the concerned implementing divisions would prepare progress reports twice a year for submittal to the MIC. (d) Progress Review: Operational Implementation. The Operational Monitoring and Evaluation Division (OMED) would be responsible for supervising the agro-economic aspects as well as O&M of TIPs under operation. The OMED would visit the same projects as those for implementation review twice a year. It would submit to the MIC a supervision summary for each TIP visited to report irrigation benefits, adequacy of agricultural sup- porting services as well as the quality and effectiveness of operation and maintenance (Annex 5F). For the remaining 75% of the TIPs below 1,000 ha, the concerned implementing divi- sions would prepare progress reports twice a year for submit- tal to the MIC. IDA Involvement 6.11 An assurance has been obtained from GOK that the procedures set forth in paras. 6.06-6.10 would be adopted. IDA's involvement would be to spotcheck the decisions of the Minor Irrigation Committee on a selective basis. The MIC would be required to submit to IDA minutes of MIC meetings and project sum- maries of approved TIPs. IDA would select randomly at least two TIPs a year to appraise in depth. In any case, if IDA disagrees with MIC's approvals of certain TIPs, IDA would n.t disburse against expenditures claimed for those TIPs. 6.12 IDA supervision missions would: (i) assess the quality of the appraisal divisions' works; (ii) spot-check in the field whether agreed criteria have been followed; (iii) review procurement procedures; (iv) check the financial reports kept by GOI and GOK; (v) assess whether the established - 35 - criteria continue to be appropriate; and (vi) examine the content and timeli- ness of reports submitted by the SASC to the MIC and by MIC to IDA. Farmers' Organization 6.13 In order to improve on overall tank operations, in a system designed to higher standards of efficiency and operation, a Tank Irrigation Committee would be formed for each new tank command area, immediately after a TIP has been sanctioned for the respective location. The Committee would be heard and consulted before finalization of detailed designs, so as to ensure a more active participation of farmers in "their" tank development right from the beginning and to integrate farmers' views of a proposed project into final design. After construction, at least twice a year before each irrigation season, local staff of the Irrigation Department would consult and agree with the Committee on the details of scheduling water delivery for the coming season. The Department would remain in charge of physical maintenance of the tank and distribution/drainage system, but would share the responsibility for O&M with the Committee. Each Committee would consist of an appropriate number of farmer representatives, an agricultural officer and a junior engineer. The junior engineer would be responsible for preparing schedules for water delivery according to the decision of the Committee. The appropriate organizational structure of the Committee and farmers' organization for water sharing would be studied under the project (para. 5.26). Until such study is completed, GOK would initiate the formation of a Tank Irrigation Committee on an experi- mental basis for at least three TIPs. An assurance has been obtained to this effect. Maintenance of Tank Irrigation Projects 6.14 Until a decision on structure, functions and responsibilities of Tank Irrigation Committees (TICs) is implemented (which would include a deci- sion on sharing of the responsibility for as well as the physical execution of the maintenance of distribution systems between farmers on one side and the Irrigation Department (ID) on the other side), the ID would continue to maintain the canals, structures and drainage networks of TIPs. Physical main- tenance requirements would, however, be reduced considerably by lining of the distribution systems. Maintenance of field channels would be carried out by the farmers themselves and would be supervised by the ID's staff. Operation and maintenance of dams and appurtenant structures would remain with the ID irrespective of final arrangements for the sharing of responsibilities with the TICs. Cost Recovery 6.15 Capital and O&M Costs. The per hectare investment cost for individual tank projects varies with size and topographical conditions of a project. On average, capital cost would amount to about Rs 21,300 per ha, including the construction of tank, distribution and drainage system and on-farm works, land acquisition, physical contingencies and overhead (para 5.04). Operation and maintenance is assumed to cost about Rs 85 per ha annually. Thus, total annualized cost, at 10% interest over 50 years, would be about Rs 2,230/ha. - 36 - 6.15 Charges Related to Irrigation. Prevailing water rates, last estab- lished in January 1981, vary with crops from Rs 60/ha for most upland crops, to Rs 128/ha for paddy, Rs 120/ha for cotton and Rs 560/ha for sugarcane. On the average, a watercharge of about Rs 106/ha of CCA can be expected. Land revenue differs for rainfed and irrigated land; the project would result in an incremental revenue of Rs 24/ha. Other local taxes and cesses are calculated as a 60% addition to land revenue. Considering all water related charges, annual cost recovery would amount to about Rs 144/ha of CCA, or to about 6.5% of annual capital and O&M cost. 6.17 Other Taxes. Other taxes, which are at present easier to asses and collect are given greater importance by the Government. In Karnataka, a tax of 4% is levied on the sale of all agricultural products. Sales of fertili- zers and agrochemicals also carry a tax of 2%, and sales of most certified seeds are taxed by 4%. In addition, a 2% market fee is levied on all produce sold through regulated markets, of which 0.8% is used for the operation of the market while 1.2% is collected by the state. Sales taxes and market fees would increase GOK's revenue by about Rs 380 per hectare brought under irrigation. Together with water-related charges, tax revenue would increase to a total of about Rs 524 per ha, covering all O&M cost and about 20% of annual capital cost, under the prevailing system of watercharges and taxes. Income tax payable on incomes exceeding a certain threshold, especially by medium and large farmers, would also contribute to higher revenues, but this effect has not been estimated. 6.18 Farmers' Ability to Pay. Cost recovery has to be compared to what farmers can reasonably be expected to afford to pay, considering their incre- mental benefits from irrigation and the necessary incentives for them to participate in this scheme. For this purpose, "project rent" has been cal- culated as incremental net income per ha, less appropriate rewards to family labor, entrepreneurship and allowance for cultivation risk (Annex 1, T-l1). At full development, the estimated average project rent would amount to about Rs 2,800 per ha. Direct recovery would thus result in a recovery/rent ratio of 5%. This leaves ample scope for increases in watercharges without reducing incentives to use irrigation water. 6.19 Limits to Higher Charges. Cost recovery with present levels of water charges would be unsatisfactory under the proposed project, both with regard to farmers' estimated potential ability to pay and to the percentage of total cost recovered. Under the present system, a State-wide uniform rate is charged (for a given crop on a per acre/per season basis) in all public irrigation schemes because of administrative difficulties of assessing benefits individ- ually for each farmer. Large variations in benefits do not only occur between different farmers in the same command area (e.g., due to farm size, soil quality and location within the scheme) but also between projects and between different types of schemes (large and medium surface water schemes, tanks, etc.). It can be observed in existing projects that the water supplied to tail-end farmers is often unreliable--due to design and condition of the distribution system--and that benefits to irrigators are fairly low. Thus, water rates are set at a low level to give tail-end farmers sufficient incen- tive to use the water, although the "average" farmer would be able to pay substantially higher charges. The same reasoning applies also to qualitative - 37 - differences between irrigation schemes and types of irrigation. Until it is technically and administratively feasible to implement volumetric water charges, which better reflect the benefits accruing to each farmer, water charges will remain at an inherently low level. 6.20 Study of Modified System of Watercharges. Under the proposed project, design standards and operational criteria for water distribution aim at an improved water supply, with essentially equitable and measurable quantities to be delivered to all participating farmers on a rotational basis. This would allow methods of charging for irrigation services more closely related to the amount of water actually used by individual farmers (e.g., by recording the time periods of actual flow to a holding). It would also facilitate an increase in the overall level of water charges and their collection (for these new tank irrigation schemes), since higher reliability of service and the link between charges, volume of water received and benefits derived therefrom can be expected to change farmers' attitudes towards payment for such service, particularly in view of rather high expected benefits. As charging for irrigation services is an important element in the relationship among farmers as well as between the Irrigation Department and the farmers, the proposed study on Water Users' Organization would include recommendations as to what methods of assessing, administrating and collecting water charges within te envisaged organizational framework would be suitable to promote equitable charging, efficiency of water use and acceptable cost recovery, given the improved design and operation of distribution systems. The recommendations of the study would also take into account the political and legislative feasibility of special charging methods and/or levels for irrigation schemes with improved design and operation. 6.21 On the basis of such recommendations, GOK would, by March 1984, review the method and the level of changing for irrigation water supply in new tank irrigation projects and would, after the consideration of IDA's com- ments, introduce an appropriate system and leve of such charges, with the objective of recovering the operation and maintenance cost of such tanks fully, and the investment cost to the extent possible, given the need for incentives to participating farmers and given their payment capacity. An assurance to this effect has been obtained from GOK. VII. BENEFITS AND JUSTIFICATION General 7.01 Under the proposed project, 120-160 individual tanks would be built, covering a command area of about 25,000 ha. About 60% of this area would be located in dry and drought prone areas with less than 800 mm of mean annual rainfall. The project would reduce some of the dependence of agriculture on - 38 - the vagaries of rainfall, reduce farmers' risks and their risk aversion, 1/ increase production of food and cash crops and provide farmers with higher and more regular incomes and employment. Not only would new irrigation facilities be provided, but simultaneously also higher standards of design and operation would be introduced in these projects. The independence of projects from major sources of irrigation water permits their location in places where otherwise no irrigation would be possible, thus helping to correct some of the existing imbalance in irrigation development. The main benefits from the project are summarized below: Summary of Expected Benefits Benefiting farm families 7,500-8,500 Increase in net irrigated area 25,000 ha Increase in cropping intensity from 102% to 161% Increase in foodgrain production 63,000 t per year Increase in cashcrop production Rs 105 M Increase in per capita income 300% Increase in farm employment 3 M mandays/year Increase in value added Rs 109 M Employment in construction during 5 years 35-40 M mandays 7.02 Production Benefits. Annual production increases generated by the project at full development (about 1990), are estimated at about 63,000 tons of cereals and pulses, valued at Rs 69 M (+340% over "without project situa- tion"), and Rs 105 M of cashcrops (+570%), mostly cotton, oilseeds, chillies, sugarcane and mulberry. The incremental annual value added to the economy at those production levels is estimated at about Rs 109 M per annum, not counting induced value added in production of material inputs or in the processing industries. An additional benefit, which is difficult to quantify, would be generated in the form of fish production in new tanks. This is current practice in most existing tanks and would be expected under new tanks also. 7.03 Employment Benefits. Overall direct employment effects will largely stem from the provision of irrigation in presently rainfed areas. Employment on benefiting farms would increase from about 1.8 M mandays to about 4.7 M mandays per year under the project, which is equivalent to the creation of about 12,000-13,000 full time jobs. Most of this employment would occur in the form of increased on-farm activities for family members, thereby reducing the need to rely on uncertain and often remote work opportunities. Medium and larger farmers would hire additional labor, mostly from underemployed rural workers, at a rate of some 1.2 M mandays annually. It is estimated that con- struction of tanks would provide, over a span of four to five years, between 35 M and 40 M mandays of employment for unskilled workers. 1/ Farmers would still be depending on annual variations in rainfall, since the source of irrigation (the volume of water in the reservoir) is also directly depending on rainfall. The risk reduction refers rather to the bridging of dry spells during the rainy season in average and good rainfall years, and to an assured water supply to those areas that receive irrigation in the dry season in any given year (based on actual reservoir levels before the start of the dry season). - 39 - Income and Poverty Impact 7.04 The project is directed towards farmers which are at present vir- tually without any irrigation, more than 60% of them living in areas which are considered moderately or severely droughtprone. Incomes under these circumstances are near the subsistence level, which is reflected in estimated farm budgets for some typical farm sizes without the project (Annex 1, T-ll). Net family incomes (in cash and kind) in the dry and transitional zones vary from Rs 1,800-2,200 for a marginal farm to Rs 7,000-11,000 per year for a large farm. Taking into account family size on such farms, per capita incomes vary roughly from Rs 400 to Rs 1,300 per year. In the high rainfall zones, per capita income ranges from about Rs 600 to Rs 1,200 per year. 1/ Given an estimated rural poverty income threshold of about Rs 780 per year and given with the farm size distribution as illustrated in para 1.16, some 70% of farm families likely to benefit from the project live presently under poverty con- ditions. Projecting farm budgets for the future with project (Annex 1, T-11), assuming average zonal cropping patterns and varying availability of family labor, farm family incomes (including off-farm earnings) would increase to about Rs 6,000 for a small farm, Rs 23,500 for a medium size farm and about Rs 45,000 for a large farm, translating into average per capita incomes of about Rs 1,300, Rs 3,700 and Rs 5,700, respectively. The incidence of project benefits is likely to follow the present pattern of landholding size. It is thus anticipated that absolute poverty would be almost eliminated among parti- cipating families. Not much is known about income levels of rural landless laborers in Karnataka, but a survey in the Tungabhadra command area revealed that the number of days worked annually by landless workers increased from about 130 to 230 as a result of irrigation. With the observed difference in average wages for operations on rainfed and irrigated land (Rs 5 vs Rs 6.3), incomes of landless laborers (who belong generally to the poorest of the poor), are expected to increase at least two-fold in places affected by new tank irrigation. Assumptions for Economic Analysis (see Annex 7 for details) 7.05 For each agroclimatic zone, cropping pattern, crop yields and input requirements were projected towards the expected year of full development (1990), both for the "with project" and "without project" situation. Cropping patterns for the "without project" situation are based on present rainfed cropping practices; those under the project have been derived from observed patterns under irrigation in each zone, taking into account agroclimatic characteristics, farmers' preferences when switching from rainfed to irrigated production, but also GOK's policy of promoting production of certain crops (oilseeds, to a lesser extent pulses) and of restricting use of valuable water for water-intensive crops, such as paddy and sugarcane in areas of low rainfall. Cropping patterns so deri-ed do not represent necessarily an optimum choice in the sense of maximizing 'eturns to water (the limiting factor), since such an outcome could only be expected from strictly enforced, optimal water distribu- tion practices. Thus, the share of paddy and sugarcane is always assumed to 1/ Only a small percentage of the farms in the high rainfall zones are larger than 3 ha. Population density on the majority of farms is very high, resulting in comparatively low per capita incomes. - 40 - be somewhat larger than advisable under the climatic conditions. Assumed irrigation intensitiSs (average 145%) result in average water3requirements of about 9,500-10,500 m /ha of CCA in dry3zones, 10,000-11,000 m /ha for transi- tional zones and about 13,000-14,000 m /ha for high rainfall zones, reflecting different conditions of likely supply. Crop yields for rainfed crops have been assumed slightly higher than at present to take into account likely increases over the next decade. Irrigated yields reflect the assessment of what would be feasible with presently available varieties and average present cultivation practices under irrigation. Cropping patterns, yields, inputs and crop budgets for different agroclimatic zones are presented in Annex 1, T-13 and T-14. 7.06 Economic input and output prices were projected for 1990, in constant 1980 Rupees, according to standard Bank practice, i.e. they were derived from projected world market prices after adjusting for transport, handling, pro- cessing and quality differences. A standard conversion factor of 0.8 was applied to domestic cost elements. Economic wages for unskilled farm labor were estimated as about two thirds of financial wage rates. Reflecting an observed wage differential, wages in irrigated agriculture are assumed to be about 25% higher than in rainfed agriculture. 7.07 Project Cost was derived as a weighted average construction cost per cubicmeter of water available in an average year at the reservoir outlet, the weights being the expected shares of different tank 3size categoriSs (Annex 1, T-3). Such cost varies typica ly with size from Rs 1.50/m to Rs 2.20/m ; the average is about Rs 1.97/m . In order to take into account other project related cost difficult to estimate on a unit basis (access roads, catchment protection works, land shaping, etc.), this value has been increased by 15% to Rs 2.26 per cubic meter. A construction conversion factor of 0.78 was applied to this cost, reflecting the assumed shares of unskilled labor, non-traded inputs and import content and their respective economic cost. Cost of opera- tion and maintenance of a TIP was &ssumed as Rs 85/ha, multiplied by a conver- sion factor of 0.7 to reflect a high content of unskilled labor. Creation of a reservoir causes loss of agricultural production in the submerged area. On the average, an area equivalent to about 15% of CCA is estimated to be sub- merged; a loss of rainfed production of Rs 1,050/ha of CCA was assumed to occur due to the project. 1/ 7.08 Value of water and project benefits. Cost-benefit analysis is based on an evaluation of cost and returns to water (as the main benefit-producing factor); thus the incremental net return per unit of water used (value of water) has been calculated. 2/ It varies between agroclimatic zones from 1/ 80% of average estimated economic net return of Rs 1,300/ha, since part of the submerged area can be cultivated when tank is temporarily dry. 2/ Water requirements were calculated according to modified Penman method and agreed criteria (para 5.11) for each agroclimatic zone, based on estimated average rainfall (50% dependable) for the zone and projected cropping patterns. - 41 - about Rs 0.50 tg Rs 0.60 per m of water (at reservoir outlet), averaging about Rs 0.55/m (Annex 1, T-15). The average over all zones virtually does not change with varying assumptions on irrigation intensities; therefore, project benefits depend only on the actual quantities of water made available per ha of CCA. For the calculation of expected benefits, the release and use of water as well as the incremental production were assumed to follow an ini- tial build-up over five years (after completion of construction), reflecting the gradual adoption and adjustment process of participating farmers. Economic Rate of Return. 7.09 After distributing construction cost over an estimated average implementation period of three years, and discounting costs and benefits of water over 50 years, the economic rate of return is about 20%, as against an estimated opportunity cost of capital of 12% for India. This economic rate of return reflects the viability of an average TIP. Estimated economic cost and benefit streams for an average ha of command area are shown in Annex 1, T-16. Project Risk 7.10 Uncertainty of Average Reservoir Inflows. I/ Methods presently used in Karnataka for estimation of annual inflow into a reservoir are based on formulae which estimate runoff yield of a catchment on the basis of mean seasonal rainfall in the catchment, given certain catchment characteristics. Although the proposed project would improve the data base for runoff yield estimates by actual streamflow measurements and a more accurate prediction model, there will always be errors in prediction to varying degrees, which cause a potential risk to a project's viability by reducing the benefits (which depend essentially on the average volume of water available per ha of command area) below their appraisal level. However, since there are no indications that the presently available methods are significantly biased towards either under- or overestimation, the overall project outcome as an average over many dispersed tank irrigation projects faces a diversified, i.e., low risk, even with those less accurate runoff estimation methods. 7.11 The results from a study undertaken by GOK to test the reliability of runoff estimates according to Strange's curves (Annex 6) by comparing actual with estimated runoffs in 12 catchments in Karnataka indicates that the actual inflow into a reservoir may be assumed to be not less than 65% of the estimate with 90% probability. For the proposed project, a reduction of average water supply by 35% relative to appraisal assumptions would still result in an economic rate of return of about 14%. 1/ The uncertainty or risk referred to here is not related to year-to-year variations in rainfall (see footnote to para 7.01), but to the long term average water supply for an individual TIP and for all TIPs as an aggre- gate. The evaluation of benefits is based on such long term average supply; consequently, this risk affects the long term benefits from the project. - 42 - 7.12 An attempt was made to illustrate the consequences of under- and overdesigning of a typical project relative to various levels of actual reser- voir inflow. For this purpose, an approach based on decision analysis was taken; for details on the assumptions and results see Annex 7. The results indicate that better prediction methods achieve, of course, a reduction of the (opportunity) cost of uncertainty, i.e. of risk, but do not necessarily lead to different decisions on the optimum size of a project, unless the expected benefit/cost ratio is low. 1/ If the problem were whether to delay a project in order to collect more information and to reduce risk, or to go ahead based on less reliable information, the decision should certainly be in favor of the latter, since the expected cost of delay in terms of net benefits lost 2/ (about 30% of "appraisal NPV" for three years delay, at a 12% discount rate), would always outweigh the benefits gained in terms of reduced opportunity losses. In addition, the risks associated with imperfect methods of runoff estimation do not seem more serious than those resulting from possible errors in other parameters (rainfall, yields, prices, water requirements, project cost). There is little chance of average water supply shortfalls being so serious as to reduce the expected net present value to zero. However, the introduction of improved methods of prediction and actual runoff measurements at the tank site would reduce the expected cost of uncertainty considerably 3/ and would lead, on average, to better design decisions and economic returns for later schemes under the project and thereafter. 7.13 Sensitivity to Basic Assumptions. The project's economic viability was tested for its sensitivity to deviations from basic values for key para- meters, such as crop yields at full development, input costs, output prices, water requirements, and cost of and delays in construction, etc. Correspond- ing changes in the economic rate of return are tabulated below: 1/ "Optimum" size (as against a size based on deterministic design decisions) is the project size with the highest expected net present value. Gener- ally, the "optimum" size in this approach is slightly lower than the appraisal size. (This is due to the fact that expected benefits increase only slightly for project sizes larger than the appraisal size and reach soon a maximum, while project cost varies approximately in proportion to size.) For low B/C ratios, a more accurate prediction method would result in appraisal sizes closer to the optimum size than less accurate methods. 2/ A delay in construction would not affect the rate of return; it would, however, change the NPV as estimated at present, i.e., at the time of undelayed construction. 3/ The expected cost of uncertainty (also known as expected opportunity loss, expected value of perfect information or simply expected risk), is the difference between the expected net present value for the optimum size under uncertainty and such value under certainty. - 43 - Change Economic Rate of Return (%) (%) Base Case - 20 (a) All yields at full development - 20 14 (b) Input cost at full development + 20 18 (c) Net return per ha, without project + 40 18 (d) Economic output prices - 20 15 (e) Cost of construction + 40 15 (f) Average construction period +100 17 (g) Average water requirements per ha + 20 17 (h) Water supply per ha of CCA - 20 17 (i) Conveyance efficiency - 20 17 Other parameters, such as particular values of conversion factors, economic wages, O&M cost or loss of production due to submergence have no significant effect on the sensitivity of the economic rate of return. Even combinations of substantial changes of benefit and cost parameters would not be likely to reduce the economic rate of return to 12%. 7.14 Switching Values. The returns to and quantity of water and the cost of construction play an important role in determining economic viability. Therefore those values of these parameters have been examined (switching values) which would, ceteris paribus, reduce the economic rate of return to the estimated opportunity cost of capital. These values are estimated as follows: Value of water at the tank outlet Basic assumption: Rs 0.55/ 3 Switching value: Rs 0.30/m (-46%) Supply of water from the tank Basic assumption: 10,800 m3/ha/year Switching value: 5,880 m /ha/year (-46%) Cost of construction per m of releas-ble water: Basic assumption: Rs 2.26/mi Switching value: Rs 4.31/m3 (+91%) Construction period Basic assumption: Average: 3 years Switching value: Average: more than 10 years - 44 - A shortfall of 46% in the incremental net returns per cubic meter of water or in the supply of water per ha of CCA and a 90% overrun in construction cost are not considered likely. The average construction period is not likely to more than double relative to basic estimate, since most sub-projects would be relatively small and GOK's capability of implementing a relatively modest program of construction of about 120-160 tanks is not questioned. 7.15 The proposed project faces the risks normally associated with irrigation projects, the overall risk being widely diversified over a large number of individual projects. The risk is considered acceptable; the project would be economically viable. VIII. RECOMMENDATIONS 8.01 Agreements have been reached with GOI during negotiations that the proceeds of the credit would be channeled to GOK in accordance with GOI's standard arrangements for development assistance to the States of India (para. 5.26). 8.02 Agreements have been reached with GOK during negotiations on the following points: (a) Tank Irrigation Projects (TIPs) would be planned, designed, approved, implemented and operated according to the agreed criteria (para. 5.06); (b) (i) GOK would maintain its panel of experts with qualifications, experience and terms of reference (Annex 4) acceptable to GOK and IDA to review the plans and designs of all dams, including their related structures constructed under the Project, whose proposed height of the embankment above streambed exceeds 10 m, Phose proposed gross storage behind embankment exceeds 2.5 Mm at full reservoir level, or whose proposed location of the embankment is such that significant destruction could occur in case of failure; (ii) it would cause the panel to conduct periodic reviews during the design and construction to examine whether any new grounds for making changes in the design of the such dams have become apparent (para 5.24). (iii) GOK would cause the dams mentioned above to be period- ically inspected in accordance with sound engineering practice in order to determine whether there are any deficiencies in the condition of such structures, or in the quality and adequacy of maintenance or methods of operations of the same, which may endanger their safety. - 45 - (c) A study would be completed by March 31, 1983 according to terms of reference agreed between GOK and IDA (Annex 3) with the purpose: (i) to investigate whether the formation of water users' organizations would contribute to an increase of the farmers' participation in the operation of the TIPs; (ii) to recommend means for establishing viable and effective participation of beneficiaries in operation and maintenance of the TIPs; and (iii) to recommend a system of water charges in accordance with objectives agreed with the Association (para 5.30). (d) GOK would promptly provide funds from its development budget to execute the project in accordance with the schedule of expenditures set forth in Annex 1, T-9 (para. 5.32); (e) (i) GOK would maintain separate accounts on project expen- ditures; (ii) it would ensure that these accounts are audited annually by the State Accountant General or his designate and that the resulting reports would be submitted to IDA as soon as possible but not later than nine months after the end of each fiscal year; (iii) it would make complete accounts and financial statements available for inspection during IDA review mission (para 5.35); (f) GOK would promptly furnish to IDA Project Summaries (see Annex 5B) for each TIP approved by the Minor Irrigation Committee (MIC) as well as minutes of each MIC meeting (para 6.10); (g) GOK would submit to IDA by March 31 of each year an Annual Implementation Program (see Annex 5C) and an Annual Performance Summary (see Annex 5D), (para 6.10); (h) individual tank irrigation projects would be processed according to agreed procedures (para 6.11); (i) GOK would initiate, on an experimental basis, the formation of a Tank Irrigation Committee in each of at least three selected TIPs. Committees would be formed before start of constrution, in accordance with TOR and with a composition to be agreed with IDA (para 6.13); and (j) With the objective of ensuring recovery of annual operational and maintenance costs and, to the extent possible, cost of infrastructure investments, having consideration to incentives for, and payment capacity of, farmers, GOK would (i) consider- ing the recommendations of the study specified under (c) above, by March 31, 1984, review the method of charging for irrigation water supply as well as the level of water and water related charges in tank irrigation projects in Karnataka, and (ii) as soon as feasible thereafter, introduce and enforce an appro- priate system and level of such charges after paying due consideration to IDA's comments, if any, on the above mentioned review and the study (para 6.21). - 46 - 8.03 With the above conditions and assurances, the proposed project would be suitable for an IDA credit of US$54 M on standard IDA terms. The Borrower would be the Government of India. -47- ANNEX 1 Table 1 INDIA KARNATAXA TANK IRRIGATION PROJECT Performance of Selected Major Crops (Annual Growth Rates in Percent) 1955/56-1977/78 1955/56-1966/67 1967/68-1977/78 Area Yield Production Area Yield Production Area Yield Production Rice 0.8 2.0 2.8 2.6 0.4 3.0 -0.6 2.0 1.5 Jowar -1.8 4.1 2.4 0.9 2.9 3.8 -0.3 3.4 3.7 Ragi 0.2 0.9 1.1 2.2 -5.0 -3.1 -0.1 7.4 7.4 Wheat 1.0 6.1 7.1 -1.0 0.0 -l.0 1.5 6.4 8.5 Pulses 0.4 2.7 2.9 -0.3 1.8 1.4 0.3 3.0 2.9 Foodgrains -0.2 3.0 2.7 0.8 1.0 1.7 -0.1 3.7 3.4 Oilseeds 0.5 0.1 0.5 -0.7 -1.7 -2.3 0.8 -1.3 -0.7 Sugarcane 4.7 1.1 5.7 5.8 3.7 9.5 4.4 -2.3 2.2 Cotton -0.4 3.4 2.9 -1.4 0.3 -1.4 -0.7 7.9 8.0 -48- ANNEX 1 Table 2 INDIA KARNATAKA TANK IRRIGATION PROJECT Contribution of the Irrigation Sector to Agricultural Growth In Karnataka 1/ (A) Increase in the Irrigated Sector (+98%) RsM Share in Total (%) 1. On existing irrigated areas (gross 0.97M ha) (a) Due to yield improvements +1,170 34.1 (b) Due to changes in cropping pattern -280 -S.2 (c) Interactive effects of (a) and (b) -100 -2.9 +790 23.0 2. On new irrigated areas (gross 0.59M ha) (a) Due to area increase alone +2,140 62.4 (b) Due to yield improvements +720 21.0 (c) Due to changes in cropping pattern -170 -5.0 (d) Interactive effects of (a), (b) and (c) -50 -1.4 +2,640 77.0 Total irrigated sector: +3,430 100.0 As per cent of total increase: 78.3 (B) Increase of Production in Rainfed Sector (+11.8%) 1. On existing rainfed areas (gross 9.5 M ha) (a) Due to yield improvements +1,750 184.2 (b) Due to changes in cropping pattern -350 -36.8 (c) Interactive effects of (a) & (b) -70 -7.4 1,330 140.0 2. On decreased rainfed area (gross -p.38 M ha) (a) Due to change in area alone -320 -33.7 (b) Due to yield improvements -70 -7.4 (c) Due to changes in cropping pattern +10 +1.1 (d) Interactive effects of (a), (b), and (c) 0 0 -380 -40.0 Total Rainfed Sector +950 100.0 As per cent ot total increase 21.7 (C) Total Increase in Both Sectors (+38.0%)2/ (a) Due to shift of areas from rainfed to irrigated agriculture +1,600 36.5 (b) Due to yield improvements +3,010 68.7 (c) Due to changes in cropping pattern -670 -15.3 (d) Due to change in total area cropped +230 5.3 (e) Interactive shift/yield effect +420 9.6 (f) Other interactive effects -210 4.8 4,380 100.0 1/ In constant 1975/76 prices. The year 1975//6 represents actually an average of the years 1974/75 to 1976/77, the last year for which complete production data are available. 2/ Fertilizer use alone would explain about 35% ot the overall increase of Rs 4,380 M, when assuming a production increase of Rs 8 due to 1 kg of nutrient. However, a distribution of this effect among irrigated and rainfed agriculture, new and existing areas has not been attempted. INDIA KARNATAKA TANK IRRIGATION PROJECT Typical Unit Cost Estimates for Tank Irrigation Projects (Rupees per na or UA; Size Item 20-100 ha 100-200 ha 200-500 ha 500-1000 ha 1000-2000 ha Average (Weight: 34%) (Weight: 32%) (Weight: 27%) (Weight: 1%) (Weight: 6%) (Weighted) A) RESERVOIR Dam 9,000 7,000 5,500 4,000 3,000 7,005 SPillway 5,400 4,200 3,300 2,400 1,800 4,203 Buildings, Plant 300 700 1,000 1,500 1,500 701 Land Aquisition 750 750 750 750 750 750 Subtotal 15,450 12,650 10,550 8,650 7,050 12,659 B) DISTRIBUTION SYSTEM 4: Main Canals & Distributaries 900 1,800 2,500 3,200 3,600 1,805 (down to 40 ha, lined) Minor Channels 990 990 990 990 990 990 (down to 8 ha, lined) Field Channels 200 200 200 200 200 200 Outlets 70 70 70 70 70 70 Drainage 280 280 280 280 280 280 Land Aquisition 100 100 100 100 100 100 Subtotal 2,540 3,440 4,140 4,840 5,240 3,445 Base cost 17,990 16,090 14 90 13 49'0 12,290 16,104 C) GENERAL Physical Contingencies (15%) 2,700 2,410 2,200 2,020 1,840 2,416 Engineering & Supervision (15%) 3,100 2,180 2,540 2,330 2,120 2,778 Total 23,790 21,280 19,430 17,840 16,250 21,298 INDIA EARNATARA IANR IRRICATION rEtJECT Elemnt and Features of Six Selected Tank irrigation Projests Project: Dangen.halli Vadedoddi Attinori handead Kailuvoddh.lla GNbur Lc-atibn in scate East Central tatre. e SE NW H1il North Centr- l West C-e-ral North-ast Protect ToPo-a-phy fmdnlatint onAd "-stly level L-nl-nt-me loon is level to Le-el wIth a
Groupe de la Banque mondiale · Staff Appraisal Report
India - Karnataka Tanks Irrigation Project
Voir le document original
Le texte intégral est hébergé par l’organisation qui le publie. lawenc.com indexe les métadonnées et renvoie vers la source officielle.
Texte intégral
Informations clés
Organisation
Groupe de la Banque mondiale
Type de document
Staff Appraisal Report
Pays
Inde
Source
Banque mondiale