Document of The World Bank Report No: 30929 IMPLEMENTATION COMPLETION REPORT (PPFI-P9060 PPFI-P9061 IDA-29000) ON A CREDIT IN THE AMOUNT OF SDR 18.2 MILLION (US$26.3 MILLION EQUIVALENT) TO THE UNITED REPUBLIC OF TANZANIA FOR A RIVER BASIN MANAGEMENT AND SMALLHOLDER IRRIGATION IMPROVEMENT PROJECT December 28, 2004 AFTS2 Africa Region CURRENCY EQUIVALENTS (Exchange Rate Effective June 2004) Currency Unit = Tanzanian Shilling (Tsh.) 1 Tsh. = US$ .001 US$ 1 = 1,107 Tsh. FISCAL YEAR July 1 - June 30 ABBREVIATIONS AND ACRONYMS ASDP Agriculture Sector Development Program BWB Basin Water Board BWO Basin Water Office CAS Country Assistance Strategy CTB Central Tender Board DCA Development Credit Agreement DCFT District Council Facilitation Team FAO Food and Agriculture Organization GOT Government of the United Republic of Tanzania IDA International Development Association IFAD International Fund for Agriculture Development MAFS Ministry of Agriculture and Food Security MATI Ministry of Agriculture Training Institute M&E Monitoring and Evaluation MTR Mid-Term Review MWLD Ministry of Water and Livestock Development NCB National Competitive Bidding NIDP National Irrigation Development Plan PADEP Participatory Agricultural Development and Empowerment Project PCU Project Coordination Unit PWO Principal Water Officer RBMSIIP River Basin Management and Smallholder Irrigation Improvement Project SAR Staff Appraisal Report SDR Special Drawing Right SMUWC Sustainable Management of the Usangu Wetland and its Catchment project TANESCO Tanzania Electric Supply Company Limited UDSM University of Dar es Salaam WRD Water Resources Department WRM Water Resource Management WUA Water User Association WUF Water User Fee Vice President: Gobind T. Nankani Country Director Judy M. O.Connor Sector Manager Karen Mcconnell Brooks Task Team Leader/Task Manager: IJsbrand H. de Jong TANZANIA RIVER BASIN MANAGEMENT AND SMALLHOLDER IRRIGATION IMPROVEMENT PROJECT CONTENTS Page No. 1. Project Data 1 2. Principal Performance Ratings 1 3. Assessment of Development Objective and Design, and of Quality at Entry 2 4. Achievement of Objective and Outputs 6 5. Major Factors Affecting Implementation and Outcome 14 6. Sustainability 14 7. Bank and Borrower Performance 17 8. Lessons Learned 19 9. Partner Comments 19 10. Additional Information 20 Annex 1. Key Performance Indicators/Log Frame Matrix 21 Annex 2. Project Costs and Financing 22 Annex 3. Economic Costs and Benefits 24 Annex 4. Bank Inputs 30 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 31 Annex 6. Ratings of Bank and Borrower Performance 32 Annex 7. List of Supporting Documents 33 Annex 8. Smallholder Irrigation Improvement 34 Annex 9. Government of Tanzania Implementation Completion Report 38 Maps IBRD Numbers 27610, 27611 and 27613 Project ID: P038570 Project Name: TZ:RIVER BASIN MGM.SMAL Team Leader: IJsbrand H. de Jong TL Unit: AFTS2 ICR Type: Core ICR Report Date: December 27, 2004 1. Project Data Name: TZ:RIVER BASIN MGM.SMAL L/C/TF Number: PPFI-P9060; PPFI-P9061; IDA-29000 Country/Department: TANZANIA Region: Africa Regional Office Sector/subsector: Irrigation and drainage (55%); Central government administration (45%) Theme: Water resource management (P); Environmental policies and institutions (P); Participation and civic engagement (S); Administrative and civil service reform (S) KEY DATES Original Revised/Actual PCD: 11/01/1994 Effective: 08/09/1996 12/05/1996 Appraisal: 06/26/1995 MTR: 12/31/1999 03/09/2001 Approval: 07/11/1996 Closing: 12/31/2002 06/30/2004 Borrower/Implementing Agency: GOVERNMENT/MAFS; GOVERNMENT/MWLD Other Partners: STAFF Current At Appraisal Vice President: Gobind T. Nankani V. E. Kimberley Jaycox Country Director: Judy M. O'Connor James W. Adams Sector Manager: Karen Mcconnell Brooks Sushma Ganguly Team Leader at ICR: IJsbrand H. de Jong James Coates ICR Primary Author: Malathi S. Jayawickrama 2. Principal Performance Ratings (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HL=Highly Likely, L=Likely, UN=Unlikely, HUN=Highly Unlikely, HU=Highly Unsatisfactory, H=High, SU=Substantial, M=Modest, N=Negligible) Outcome: S Sustainability: L Institutional Development Impact: SU Bank Performance: S Borrower Performance: S QAG (if available) ICR Quality at Entry: S Project at Risk at Any Time: No 3. Assessment of Development Objective and Design, and of Quality at Entry 3.1 Original Objective: Context. Tanzania's ability to manage scarce water resources became a national issue in the early to mid 1990s. Following the liberalization of economic activity in 1986, annual GDP and exports grew at over 4 percent during 1986-94, with increased production of food and traditional export crops, and a solid rise in nontraditional exports. New opportunities in agriculture, and the greater demand for water for irrigation and hydropower, together with the long dry season and several years of less-than-average rainfall, contributed to water scarcity and conflicts, while the lack of information on water quantity and quality, and an inadequate framework for tackling cross-sectoral water issues severely constrained sustainable water resource management. Tanzania has defined nine river basins for water resources administration. The Rufiji is the largest basin, covering 177,000 square kilometers, or 18 percent of mainland Tanzania. Irrigation and hydropower account for 99 percent of all abstractions in the Rufiji and Pangani basins, with hydropower located downstream of irrigated areas and urban water abstraction points. This makes water use for hydropower highly subtractable from other uses. It is vital that the national hydropower system, which is entirely located in the Rufiji and Pangani rivers, gets adequate water and is operated optimally. The costs of interruptions (load-shedding) in the electricity supplies are estimated at several thousand shillings per kilowatt-hour in lost production and wastage of raw materials (Ministry of Water, Water Resources Economics Report I, 1996). Agriculture is the foundation of the Tanzanian economy and accounts for about half of GDP and exports. Its performance has a significant effect on income and poverty levels. Food crop production, which dominates the agricultural economy, contributes about 50 percent of agricultural GDP. Agricultural yields are highly dependent on water resources. Crop yields are substantially higher in areas where there are irrigation systems and adequate water flow, especially in rice production (Ministry of Water, Water Resources Economics Report I, 1996). Smallholder farmers practice on-farm water management through traditional furrows and small basins that are prone to water logging, and via poorly-designed canals that lose water. The irrigation efficiency in such schemes is low, estimated at less than 10 percent in the wet season and 14 percent in the dry season, although efficiency on a basin level, due to repeated re-use of water, can actually be high (Assessment of Irrigation Efficiency in Traditional Irrigation Schemes in Pangani and Rufiji Basins, 2002). Allocations for farmers at the tail-end of the canals are often uncertain. Inefficient water use and wastage in upstream irrigation schemes have also left inadequate supply in both hydropower reservoirs, contributing to reductions in planned hydropower production. As only about 0.1 percent of water goes to domestic uses, water savings must come from efficiency improvements in the irrigation schemes (and enhanced operating routines of the hydropower system). The Government's 1991 National Water Policy does not adequately address water management and allocation, cross-sectoral interests, and sustainable river basin management. Nor does it provide a framework for preventing and resolving conflicts among competing users, and regulating the demand for water. RBMSIIP was prepared and implemented in response to these issues. Objectives. The project's objectives were: (i) to strengthen the Government's capacity to manage water resources and address water-related environmental concerns at the national level, and in the Rufiji and Pangani River Basins; and (ii) to improve irrigation efficiency of selected smallholder traditional irrigation schemes in these two basins (SAR, June 5, 1996). RMBSIIP sought to achieve this through a stronger institutional, regulatory and incentive framework for basin management, and enhanced stakeholder participation in basin management and irrigation scheme operation. - 2 - RBMSIIP's design aimed to improve WRM, and increase societal benefits from water use by establishing the capability to: (i) collect and analyze the information needed to establish how much water is available, when and of what quality; (ii) administer a system of water rights, pollution standards and other regulations on water use; (iii) create awareness among water users of the consequences of their actions on downstream users, and build consensus for representative management of water; and (iv) provide incentives, organizational support and financing for increasing water use efficiency, especially among small, traditional irrigators too poor to invest in the necessary infrastructure improvements. The approach planned to focus on existing schemes, and enhance the capabilities of the Water User Associations that manage them. These objectives were consistent with the Development Vision 2025, the Poverty Reduction Strategy Paper, and the Bank's Country Assistance Strategy of May 23, 1996. The project was expected to contribute to fiscal stability and sustain financing of the scarce water supply, by improving irrigation infrastructure and increasing tariffs that had encouraged inefficient water use. It also sought to alleviate poverty, by raising small farmer incomes, reducing food insecurity, and encouraging women's involvement in village irrigation committees and off-farm income generation activities. 3.2 Revised Objective: The project objectives were not revised during implementation. 3.3 Original Components: The project had two components as outlined in the 1996 SAR: (1) River Basin Management (US$10.6 million). This component sought to improve the national framework for comprehensive WRM, and strengthen WRM in the Pangani and Rufiji basins. (a) Strengthening National Water Resources Management (US$4.8 million) included five areas: (i) Policy Framework, to improve the national framework for water management by: (a) funding task forces and studies to update water-related legislation; (b) strengthening MWLD's Water Resources Department; (c) strengthening the offices of the PWO and BWOs; and (d) funding high-priority hydrological studies that addressed national issues. Financing also included incremental travel and operating costs, per diems and consultancies needed to conduct these activities. (ii) Legislative Framework, to support a multidisciplinary Task Force, led by MWLD, to conduct a review and lead public discussion of all legislation pertaining to basin management. The institutional framework and legislation for WRM emphasized four issues: (a) water rights; (b) water charges; (c) protection of water against pollution; and (d) broadening stakeholder participation in basin management. (iii) Strengthening WRD, and the Office of the PWO. WRD has two sections--hydrological and hydrogeological services, and monitors water availability and use across the country. With offices at Regional level, it administers the data collection networks, gathers information, and forwards it to the central office in Dar es Salaam where they are further processed, analyzed and disseminated. The project planned to review the functions of the Department, and provide training, specialized equipment to improve national monitoring, and vehicles. RBMSIIP also planned to strengthen the Government's regulatory functions by providing equipment, training and technical assistance to the offices of the PWO and BWOs in the Pangani and Rufiji Basins. (The office of the PWO was abolished in the new institutional framework. See Section 4.2.) - 3 - (iv) Special studies, to understand features in the two basins, which were important to basin hydrology and environment, natural resource management in the Great Ruaha sub-basins, and river basin modeling. (v) Strengthening water quality monitoring, by improving the national monitoring network, upgrading laboratories and equipment in Iringa and Tanga, which serve the two basins, and financing civil works, vehicles and incremental operating costs. (b) Strengthening Basin Management (US$ 5.8 million). In addition to improving water management capabilities at the national level, RBMSIIP focused on needs in the two basins, including: (i) strengthening regulatory functions of the BWOs, and providing BWO staff at the basin and regional level with training, vehicles and equipment to assist them in information-collecting and development activities; and (ii) rehabilitating the hydrometric networks. (2) Smallholder Irrigation Improvement (US$20.3 million). This component targeted low-income smallholders within selected traditional irrigation schemes, to improve scheme management and infrastructure. Better water management and infrastructure were expected to increase water use efficiency, enhance access to and distribution of water in these schemes, and free excess water to downstream users. The project planned to rehabilitate 25-30 schemes covering about 7,000 hectares over the six years. Activities included: (a) Participatory Scheme Selection and Management (US$2.4 million), to finance five multidisciplinary teams from the Irrigation Department deployed in the two basins to work with the targeted schemes, to enhance scheme management and bring them to a position where they could make use of the infrastructure improvements offered under the project. (b) Engineering Works on Schemes (US$15.2 million), to provide Government financing for scheme civil works necessary to enhance irrigation infrastructure and efficiency. The total cost of infrastructure improvements was estimated between US$1,500-2,000 per ha. GOT was to contribute about 80 percent of the total cost of civil works, while farmers were expected to contribute 20 percent, through labor and materials. (c) Mitigation of Negative Environmental Effects (US$0.7 million), to create capacity to conduct environmental evaluations of irrigation schemes proposed for support under RBMSIIP and other donor-assisted projects in the Irrigation Department. This activity planned to finance a four-person cell in the Department to monitor interventions in smallholder schemes, and ensure that negative environmental effects were avoided or minimized. (d) Irrigation Sector Capacity Building (US$1.9 million), to strengthen the capability of the private and public sectors to undertake smallholder irrigation development. Planned training for private contractors included design and management of labor-based construction methods. The project also aimed to train staff in the Irrigation Department to carry out its `core' functions, including selecting schemes (through participatory methods), assessing the feasibility of interventions, organizing WUAs and supporting them until they ran smoothly. 3.4 Revised Components: As mentioned in Section 3.2, the project objectives were not revised. However, the MTR in March 2001 concluded that partial restructuring and budget revisions were necessary to adjust project activities to meet - 4 - the development objectives, and to integrate environmental quality objectives. The rehabilitation program under SII was scaled down to 15 schemes, and the credit savings were transferred to complete work under the RBM component, including consolidating achievements until the MTR, further strengthening the two BWOs, and implementing specific activities to resolve a number of water use and allocation problems in the two basins. MWLD and MAFS agreed on the budget revisions between the two components, and these minor adjustments were included in the revised budget after the MTR. 3.5 Quality at Entry: RBMSIIP was one of the first efforts to translate the Bank's 1993 Water Resources Management Policy into practice in Africa. The policy signified a new paradigm in mandating Bank support for a more comprehensive approach to WRM, recognizing that water resources need to be managed to meet multiple uses rather than as inputs to specific sectoral activities. RBMSIIP was also the first IDA investment in smallholder irrigation in Tanzania. It attempted to link two Ministries to provide an integrated approach to WRM, and address common areas of concern. This presented a challenge, both in terms of design and implementation. The project's timing and design address the Government's priorities. RBMSIIP was prepared during a time of heightened national interest in available water resources. Issues such as the Pangani Falls, conflicts surrounding Usangu and the Great Ruaha River, and load shedding following the drying up of hydropower reservoirs in 1994, had raised several queries. Was Tanzania overextending its water resources, and how could resources be used more efficiently to meet existing demands, and provide opportunities for limited expansion? The 1991 Water Policy addressed mainly supply issues, and did not deal comprehensively with cross-sectoral linkages, allocation, and participatory aspects. RBMSIIP responded well to these priorities. The project design incorporates key recommendations of previous investments supported by other donors in Tanzania, and IDA investments outside Tanzania. In particular, RBMSIIP integrates important lessons from a NORAD project in the Pangani basin, on the need for farmer involvement in WRM. It also incorporates guidance from an Operations Evaluation Department (OED) review of ten irrigation projects worldwide, including the recommendation to exploit the potential synergy between irrigation and catchment improvement, and to seek O&M financing directly from irrigators (World Bank Report 13676, 1994). The design was also based on the Rapid Water Resources Assessment (DANIDA/World Bank, 1995) that identified the river basins requiring priority action. Pangani and Rufiji, the two basins retained under RBMSIIP, had the most acute conflicts related to intersectoral water allocation between agriculture, environment and hydropower. The project design indicates a high level of understanding and innovation, and captures a number of win-win opportunities. The principal economic users of water in the two basins are hydropower and irrigation. In 1996, when RBMSIIP was appraised, more than 85 percent of Tanzania's electricity came from hydropower. Practically all the hydropower-generating capacity in the country was in the Rufiji and Pangani basins, with hydropower located downstream of irrigated areas. The levels of water utilization for irrigation (and wastage) in the years prior to the project had left inadequate supply in both reservoirs, and caused reductions in planned hydropower production. RBMSIIP evaluated the returns to hydropower production, basin management, and greater irrigation efficiency, and captured the opportunity for substantial economic gains, to both hydropower and smallholders, from enhanced upstream water management and irrigation infrastructure. Potential increases in crop productivity also provided strong incentives for farmers to buy into the idea of water savings. Finally, project design seized the opportunity to establish a framework for rational WRM through a National Water Policy that is unique in SSA, institutional reforms and legislation. This framework - 5 - included forming a platform for stakeholder involvement in water management, while also reducing the transaction costs of conflict resolution, issuing secure water rights, enforcing water user fees, and establishing basin water organizations responsible for resource management at the basin level. This framework is expected to improve water rights administration, provide incentives for transparent mechanisms for allocation, and enable a reallocation in water use towards high-return uses, increasing overall economic welfare. While RBMSIIP achieved and possibly surpassed its objectives in terms of training, the basis for staff training is unclear. A training needs assessment during the early stages of the project might have helped to identify some of the human resource bottlenecks. For example, the project failed to keep up with the growing demand for data, database management and the maintenance of hydrometric stations. The Government had expressed the need for a stronger capacity-building component during design, in particular, to strengthen river basin management. RBMSIIP's quality at entry is considered satisfactory, based on its response to the Government's priorities, its integrated approach to WRM, and its innovative win-win design. 4. Achievement of Objective and Outputs 4.1 Outcome/achievement of objective: RBMSIIP spearheaded the review of the policy, institutional and legal framework, and laid the foundations for strengthening national and basin WRM. Cabinet adopted the new National Water Policy in July 2002. The Policy integrates principles of multi-sectoral use, treats water as a social, economic and environmental good, recognizes financing of WRM through the application of a Water User Fee, and promotes autonomy at the basin level. Policy implementation strategies are being developed as part of sector-wide reforms. The National Water Sector Development Strategy--a part of a sector-wide reform process supported by GTZ--was finalized in September 2004, and is awaiting Cabinet approval. The review of legislation was done in Tanzania for the first time. It was a learning process, and Tanzania had to search for experiences of other countries. WRM reforms, involving many conflicting interests, are inherently difficult undertakings. The policy formulation process facilitated by RBMSIIP was even more substantial than the project had envisioned, engaging all the key players--agriculture, water, and hydropower--through extensive dialogue and consultations. Technical and stakeholder workshops included all classes of water users, including livestock and fisheries. The need to ensure wide stakeholder participation and endorsement of reforms delayed the process. Also, the decision to combine the originally-foreseen three separate policies (Water Resources, Urban Water Supply and Rural Water Supply) into one National Water Policy, and the different stages of development of the three separate policy papers, led to additional delays. However, water and WRM are now firmly entrenched as key priorities in the National Development Vision for Tanzania. The Government has adopted the new Water Policy, which is being extensively disseminated. Broad ownership has also been achieved. This is a significant accomplishment. RBMSIIP has also created much enthusiasm by the stakeholders to participate in various water management interventions. WUAs and/or User Groups are being formed on a demand-driven basis, including in the Usangu catchment. These groups play an important role in managing water resources at local and catchment level. It is hoped that these organizations would become a source for democratic election of legitimate representatives of users to Basin Water Boards. Annex 1 gives the project's Key Performance Indicators. The main achievements under the RBM - 6 - component include: l Rehabilitation of 43 hydrometric and 11 weather stations in the Rufiji basin, and 28 hydrometric and 8 weather stations in the Pangani basin, and the development of databases; l Provision of staff training, computer facilities, and vehicles and equipment, which have resulted in an increase in the volume of data handled; l Forming of 51 local water user groups and two apex, pilot subcatchment organizations, in collaboration with BWOs and District Councils; l Survey of 98 percent of water abstractors in the Pangani and Rufiji basins. In the Rufiji, 1514 abstractions with or without Water Rights have been identified. Of these 69.4 percent or 1050 have Water Rights. In the Pangani, a 1992 basin-wide survey indicates a total of 3359 abstractions with or without Water Rights. Of these, only 38.4 percent or 1292 have Water Rights; l Collection of Water User Fees (excluding TANESCO Royalty fees) amounting to US$58,800 in the Pangani and US$52,600 in the Rufiji; l Specialized studies in the two basins, revealing how the hydrology of Great Ruaha and Pangani Rivers functions and how they react to human activity, to provide the basis for designing future interventions; l Creating awareness with respect to pollution, and attempting to change industry habits through dialogue. Pollution issues in both basins were handled through dialogue (as the legal costs of charging polluters were higher than the fines set by GOT in the Water Act). No pollution fines were collected. The SII component surveyed all schemes and identified 613 schemes in the Pangani and 122 in the Rufiji. Thirty four schemes were selected for possible improvement, compared to forty suggested in the SAR. Due to considerable delays in scheme rehabilitation, only 15 schemes (8 in Pangani and 7 in Rufiji) were recommended for improvement, compared to the SAR target of 15 schemes in each basin. The main achievements under SII are listed below. (Annex 8 gives details on the measurement of irrigation efficiency, crop production and yields, and farmer incomes.) l Irrigation efficiency has more than doubled on all schemes. Average efficiency has increased from 8% to 19% in the wet season, and from 11% to 27% in the dry season; l 5,317 farming families have benefited from improved irrigation and drainage facilities on 15 schemes in the two river basins, covering a total area of 5,059 ha; l Rice yields and total production have increased in both Pangani and Rufiji basins--average yield has more than doubled, from 1.98 t/ha to 5.27 t/ha in Pangani, and from 1.46 t/ha to 4.06 t/ha in Rufiji. Average maize yields have increased from 1.06 t/ha. to 4.86t/ha. in Pangani, and from 1.08 t/ha. to 3.34 t/ha. in Rufiji; l Annual household farm incomes have increased from an estimated US$425 to US$1,500 in Pangani, and from US$340 to US$1,100 in Rufiji (current prices- Scheme level Project Completion Survey, Environmental Resources Consultancy (ERC), Dar es Salaam, June 2004). It must be noted, however, that the ultimate impact on incomes depend on market situations. Increased yields could result in lower producer prices and incomes if farmers are not able to market their crops. This emphasizes the need for extension agents to advise on marketing as well as on production. l With higher earnings, farmers have been able to cater for most of their needs. Many farmers are now able to pay school fees for their children and afford better health care. Farmers have also assisted in improving school facilities. The ERC survey of each scheme found that farmers have built better homes, purchased bicycles (also observed by the ICR team), and agricultural inputs, and a number of them have opened other avenues of business; l 1,674 farmers (1,052 men and 622 women; over 100 farmers per scheme) have been trained in scheme water management, crop production techniques, agro-business and financial management, and leadership skills. Farmers indicated that training in crop technology and water management has contributed significantly to improved yields; - 7 - l Five MAFS/MWLD multidisciplinary teams have been formed to assist schemes, and ten District Catchment Facilitation Teams (DCFT), one for each district, have been formed to replace these; l Farmers have been empowered to manage and maintain their own systems, through the establishment and registration of 19 Irrigators Organizations (11 as Irrigators Cooperative Societies and 8 as Irrigators Associations); l Due to improved water use efficiency, downstream water users have benefited from the re-establishment of base flows in several areas. In addition to the gains for existing schemes, several measures taken under RBMSIIP are expected to benefit future irrigation interventions. For example, other smallholder irrigation projects are using the participatory planning, design and implementation procedures established under RBMSIIP. The environmental cell, established within MAFS, is assisting with Environmental Impact Assessments in other agricultural development programs. Although the SII component rehabilitated only 15 schemes, and some of the RBM accomplishments were delayed, the performance indicators listed in Annex 1 show a solid record of accomplishment under the project. RBMSIIP has undoubtedly strengthened GOT's capacity to manage water resources through critical reforms, and contributed to increased irrigation efficiency and family incomes in the improved schemes. The project is therefore rated satisfactory in achieving its development objectives. 4.2 Outputs by components: Component 1: River Basin Management (US$10.6) (a) Strengthening National Water Resources Management Policy, Institutional and Legislative Framework. The policy review identified several weaknesses in the previous institutional framework. The PWO is the Secretary to the Central Water Board, and the BWO is the Secretary to the Basin Water Board. The PWO authorizes water rights, collects various fines and water charges, and carries out the provisions of the law in basins designated as being of "national" importance by MWLD. In the case of the Pangani and Rufiji basins, these responsibilities were designated to the BWOs. This arrangement did not facilitate coordination between the Office of the PWO and WRD. The project had been supporting the PWO, WRD, and BWOs, when it became evident that these structures were parallel and led to inefficient resource allocation. The position of PWO was abolished in the streamlined institutional framework. The new framework also includes an efficient organizational structure for WRD, the BWBs and BWOs. The final cornerstone of water sector reforms addressed by RBMSIIP--review of legislation, has been achieved. Implementation of the new Policy, however, is an ongoing process. The challenge now is to maintain commitment and resources to continue good WRM, and accelerate dissemination of the policy to different water user communities and interest groups. Strengthening WRD. The revised functions and organization of WRD was approved in September 2004. RBMSIIP provided WRD with three vehicles, a photocopier, five computers, a seismic system, and other equipment, and supported the BWO as indicated in section (b) below. Following these improvements, water management, including allocation, monitoring and enforcement, has improved. Water allocation decisions are now based on data on water resources and use collected by the project. Field offices that have been established or improved are close to water use areas, enhancing monitoring and contact with users. Greater awareness, especially in the Pangani and Rufiji basins, has contributed to more applications for water rights. Although still low, WUF billing and collections have increased compared to before RBMSIIP. Special Studies. The main technical outputs under RBMSIIP were: (a) review of value of water, economics and financing; (b) River Basin Management Modelling (RBMM) and water balances; and (c) - 8 - study on revision of effluent standards, water quality standards, & Environmental Impact Assessment (EIA). RBMSIIP supported two other studies to assist in understanding specific features: (d) the Sustainable Management of the Usangu Wetland and its Catchment (SMUWC) project, co-financed by DFID, was implemented during September 1998 to March 2002, in response to the seasonal drying up of the Great Ruaha River in the Ruaha National Park, starting in 1993. SMUWC increased the understanding of the hydrological and environmental process in Usangu, and provided input to the policy review process. SMUWC findings were widely disseminated (and can be found under www.usangu.org); and (e) Hydrological phenomena, floods & droughts studies. The floods studies assessed flood damage to irrigated lands and infrastructure, and prepared maps showing the flood-prone areas in the Pangani and Rufiji. The droughts studies determined the potential impacts on crops and yields of catchments, river flow characteristics, and obtained parameters for water use policy formulation. These studies were contracted to the Water Resources Section of the Civil Engineering Department of the University of Dar es Salaam (UDSM). Both SMUWC and the UDSM studies provided insight into the Great Ruaha River system, and contributed to the policy review process. Recommendations of these studies should assist in designing correct strategies for the management of the Great Ruaha River system. Other studies completed under RBMSIIP include: (f) a preliminary groundwater assessment & recharge mechanism, to provide input to policy, improve the database, and design a groundwater monitoring network; (g) catchment degradation and conservation; and (h) an assessment of Tanga, Arusha, Moshi, Iringa, Mbeya, Central Water Laboratory. Several studies planned at appraisal were not completed. These are: the initial study on two other basins, which was differed at MTR; the study on Economic Evaluation of Usangu Wetlands; and the impact of water crisis in Ruaha National Park and possible remedial action. Strengthening Water Quality Monitoring A water quality monitoring network was designed and implemented to obtain knowledge of water quality in the rivers, assess water pollution caused by various activities, and design pollution control measures. The Central Water Laboratory (CWL) and Regional Water Laboratories (RWL) at Mbeya and Iringa in the Rufiji Basin, and at Tanga, Moshi and Arusha in the Pangani Basin were assessed. The project rehabilitated the CWL and RWL at Mbeya, Iringa, Tanga and Arusha, and provided laboratory equipment and chemicals as a starting point. The Moshi Lab was not yet operational and needed a much larger investment. As these services could be obtained from a nearby lab in Arusha, work on the Moshi lab was deffered. Although the labs have been rehabilitated and modestly equipped, their operational capability is still hampered by inadequate finances, transport and personnel. Future strengthening of water quality monitoring programs need to take this into consideration. RBMSIIP reduced water quality monitoring in the Rufiji, as SMUWC undertook intensive monitoring in Usangu in the Rufiji Basin. SMUWC found that very few agrochemicals were used within the catchment, even on the large rice farms, and there was no serious evidence of agrochemical or other chemical pollution. In the Pangani, pollution was found near the urban centres of Arusha and Moshi. The main sources were domestic wastewater and industrial effluents. Polluting industries were issued warnings. Intensification of agriculture and greater agrochemical use to increase future crop yields could increase pollution, and may necessitate establishing an organic analytical lab, at least at the Central Water Laboratory. (b) Strengthening Basin Management Basin Office Strengthening. The project provided the Pangani and Rufiji BWOs with equipment, vehicles and renovation/construction of office facilities. Two staff from Zonal Water Labs of Mbeya (Rufiji) and Tanga (Pangani) were trained in Integrated Environment and Water Management (IEWM) at Delft Hydrology Institute in Denmark. The Director of the Water Lab Unit participated in a study tour on WRM in Zimbabwe and South Africa, with other ministry staff, District Executive Directors and Regional - 9 - Administrative Secretaries of Iringa and Tanga, who are members of the Pangani and Rufiji Basin Technical Committees. In addition, the Pangani BWO participated in a WRM study tour in the Netherlands. Hydrometric Networks Rehabilitation RBMSIIP rehabilitated 43 hydrometric and 11 weather stations in the Rufiji Basin, and 28 hydrometric and 8 weather stations in the Pangani. Preliminary surveys had proposed 65 and 44 hydrometric stations for Rufiji and Pangani respectively, however, as some stations were in good condition, the numbers were reduced to 43 and 28. The project also funded the development of databases, computing facilities, vehicles and equipment, and training. This has increased the volume of data handled. The hydrometric stations are now equipped with staff gauges (as control sensor), and either a pressure transducer sensor or shaft encoder sensor, each equipped with a data logger. Seventy technicians received training in basic computer skills and thirty hydrologists and technicians were trained in equipment installation and management. Overall, the hydrometric network in the two basins has been reasonably reinstated. The data are used for project design and operations, research, water allocations, monitoring of pollution, and water use and modeling. Installation of new technology equipment went well, but many technicians needed training in station maintenance. Updating of databases was also slow due to low manpower and equipment capacity. Component 2: Smallholder Irrigation Improvement (US$20.3 million) (a) Participatory Irrigation Scheme Selection and Management Scheme Identification. This began with a survey of all schemes, which identified 613 in the Pangani and 122 in the Rufiji. The survey also provided useful input into the national Irrigation Master Plan inventory of existing schemes. Five multi-disciplinary teams from MAFS/MWLD ranked the identified schemes using the NIDP priorities as the basis--technical aspects, economic/financial aspects, ease of rehabilitation, agricultural and management aspects, social aspects--but adapting them to ensure more weight to beneficiary participation in the selection process. There was much debate, however, on the appropriateness of the scheme selection criteria embedded in the NIDP for RBMSIIP. NIDP focuses on user participation, and does not reflect the integrated catchment management approach followed by RBMSIIP. Scheme selection exacerbated conflicts in some cases. Rehabilitation of the first four schemes was delayed considerably due to inexperienced contractors and weak supervision. The time taken by the Central Tender Board (CTB) to process proposals from both contractors and consultants caused substantial delays. Scheme Improvement Assessment Reports (SIAR) were prepared for the final 15 schemes, containing information on topographic surveys, socio-economic studies, water resources assessment, soil survey, environmental evaluation, preliminary designs, cost estimates, and economic and financial viability assessments. MAFS staff and farmers worked closely to finalize designs, which were evaluated by a consulting firm acting as an independent reviewer. The Government chose Districts as the focal point for all field-level interventions. As in the Project Implementation Plan, ten DCFTs, one for each district, were formed to replace the five MAFS/MOWLD multidisciplinary teams, and provided the focal points for District Councils. DCFTs worked closely with the authorities on all aspects of scheme improvement and development, including the resolution of water conflicts and queries relating to water rights. The teams received training in facilitation skills, participatory approaches, water management, and O&M of irrigation schemes. This should enable them to backstop the already-assisted schemes, and also assist with developing other schemes. Formation/Upgrading of Irrigators Organisations. Experience suggests that schemes with strong IOs and good leadership maintain schemes well, and address aspects related to high production. The project facilitated formation and/or upgrading and registration of nineteen IOs, and provided them with basic office - 10 - facilities and furniture. They were also trained in water management, crop production and marketing, budgeting for O&M, payment for water user fees, business farming, and management skills. DCFTs facilitated this process and assisted with all field activities related to RBMSIIP. Water Rights. All IOs of the fifteen improved schemes have water rights. Some schemes like Mombo had its own official water right before the project, while others formalized them during implementation. DCFTs and SIIP engineering staff worked closely with the BWOs to relate water diversions to crop demands. This enabled water to be effectively provided to all parts of the scheme within the agreed and often informally-established water right. The Igomelo scheme was only able to irrigate 50 percent of the command area before improvement. In some schemes, where dry season flows were extremely small, abstractions were adjusted so that IOs were fully aware of what could be irrigated in a one-in-five dry year. All schemes were provided with visible "transparent" measuring structures equipped with staff gauges close to the headworks. The IOs have been trained in this form of flow measurement, enabling farmers to understand the meaning of water rights. Gender Mainstreaming. The ratios of men to women participating in the IOs were 75:25 in the Pangani, and 65:35 in the Rufiji. This distribution indicates that there is an attempt at gender balancing and mainstreaming of women into the management and leadership of the organizations (Project Impact Assessment, June 2004). (b) Engineering Works on Schemes. For implementation, the schemes were divided into three groups, with the first two (A and B) being subjected to pre-qualification of contractors. The pre-qualification process contributed to significant delays (see Tendering Process in Annex 8). Group C adopted post-qualification of contractors. Contractors carried out all major works, with farmers completing selected minor works such as tertiary drains and field canals. All works were completed by the end of the Project, following an 18-month extension. Preparation of design and tender documents, and the tendering process were new to many Irrigation Department staff, and required training. This delayed procurement of consultancy services and contractors for the first four group A schemes, but improved as staff gained experience through on-the-job training. Supervision of contractors was also new to MAFS Engineers. To increase their capacity and provide on-the-job training, one Civil Engineer and a surveyor were assigned to each scheme to work under the international consultant's supervision Engineer. A total of 15 Engineers and 15 Surveyors received valuable supervision experience in this manner. Several engineers and surveyors are now successfully supervising construction contracts in others programs like PIDP (IFAD), ASPS (DANIDA), and PADEP (IDA). Scheme Investments. The initial inexperience of MAFS engineers was reflected in cost overruns experienced by all schemes. The final total cost of the civil works contracts, excluding consultancy services, but including farmers' contributions, was US$8.39 million, or US$1,659 per hectare. When design review and construction supervision costs are included, the total cost for the 15 schemes rises to US$10.0 million or US$1,977 per hectare. This is within the SAR estimate of US$1,500-2,000 per ha., although on some schemes it appears that works were not undertaken, as this would have increased costs per hectare above US$2000. Farmer Contributions varied between 5-20% of the total investment costs. Low values reflected the relatively high final contract price compared with initial estimates used to determine farmers' inputs. Each scheme has devised its own mechanism of attending to O&M activities and Water User Fees. (c) Environmental Mitigation Measures. An Environmental Cell Unit (ECU) was established in MAFS to conduct Initial Environmental Evaluations (IEE) of irrigation schemes supported by RBMSIIP and other donors. The unit conducted IEE on 24 schemes in both basins, including the 15 selected schemes, and 9 additional schemes from the Pangani and Rufiji basins, an Environmental Audit of the first four IEE of irrigation schemes group A, and developed four Environmental Management Manuals for farmers. At the - 11 - start of the project guidance for the ECU was inadequate, but improved during implementation. Initially, national environmental guidelines were not in place, and World Bank guidelines and safeguard polices were not readily available to ECU. This led to the updating of the IEEs to comply with both the GOT and World Bank guidelines. In compliance with the Bank guidelines and regulations, environmental audits (EA) were completed on all 15 schemes. (d) Irrigation Sector Capacity Building. Staff and Private Sector Training. All staff involved in RBMSIIP received training. In addition to on-site training of Engineers and Surveyors, staff attended a wide variety of training courses. 244 staff from the districts, irrigation zones, and Ministry headquarters were trained in tendering and cost estimates, scheme management, environmental planning and impact assessment, stakeholder analysis, financial management, drip irrigation, monitoring of performance indictors, and Integrated Pest Management. Local contractors. SIIP commissioned the National Construction Council (NCC) to study the capabilities of local contractors to undertake construction of irrigation schemes. Of 100 local contractors, only three had experience understanding repairs and rehabilitation of irrigation systems, including its major hydraulic structures. The SAR had intended to provide training to local private sector contractors, but this was limited to three two-day awareness creation seminars on irrigation opportunities conducted by NCC in 1997, and attended by 190 local private contractors and consultants. Farmer Training. This sub-component was very successful. The SAR had envisaged training about 450-600 committee members (15-20 for each Irrigators Association or Cooperative Society) from 30 irrigation schemes. Training was increased to benefit more than 100 farmers in each scheme, as well as groups from outside the two basins, such as irrigators from Lower Rufiji and Kibondo District. Formal training sessions were conducted mostly at the Kilimanjaro Agricultural Training Centre (KATC) in Moshi, MATI, Igurusi, and Nzihi Folk Training Centre in Iringa. Topics covered included: (i) scheme water management; (ii) crop production techniques (paddy, maize, beans, vegetables); (iii) agribusiness and financial management; and (iv) scheme management and leadership skills. There were also outreach programs to the 15 improved schemes. By project closure, a total of 1,674 farmers (1052 men and 622 women) had been trained on a wide range of issues. During the ICR mission, farmers on several schemes indicated the tremendous value of the training they had received under the project, in terms of adopting new technologies and achieving higher yields. Farmers requested that training be continued. 4.3 Net Present Value/Economic rate of return: Project benefits fall into two inter-dependent categories. In the first case, the returns from the SII component came from increased yields due to rehabilitated infrastructure and other factors, such as seasonal risk reduction, the improvement of communal scheme management, farmer training in water management, and in particular, better access to inputs and knowledge required for agricultural intensification. In the second case, due to improved irrigation efficiency and the implementation of the RBM component, positive externalities arose for a variety of downstream users. At appraisal, it was assumed that the main economic benefit is additional hydropower generation. The financial and economic rates of return have been re-calculated for each component based on incremental benefits (see Annex 3). In general, since itemized information was not recorded for specific activities on a yearly basis, costs were estimated on the basis of yearly disbursements and maintenance costs during the life of each component. The overall economic rate of return (ERR) for the project, estimated at 16 percent, is lower than the appraisal estimate (the NPV is Tsh. 2,934 million compared to Tsh. 19,112 million in the SAR). - 12 - The ERR of the SII component, 10.4 percent, is somewhat lower than the appraisal estimate. This reflects: (i) the delays of the rehabilitation works (the first schemes where fully rehabilitated in early 2002); and (ii) a significant reduction of the rehabilitated area of smallholder schemes (5,059 ha compared to 7,000 ha planned at appraisal). Also the ERR of the RBM component, estimated at 20.9 percent, is lower than at appraisal. In addition to the factors mentioned above, this is due to: (iii) the revision of the field requirements of water for crop production; and (iv) a conservative estimate of the price of water for hydropower generation. Table 1. Financial and Economic Rates of Return FRR ERR Component SAR ICR SAR ICR Overall (SII + RBM) ... ... 21.5% 34.0% 16.0% Smallholder Irrigation Improvement (SII) ... 21.8% 12.1% 10.4% River Basin Management (RBM) ... 21.2% 49.2% 20.9% It is important to note that the allocation of project benefits to each component is somewhat hypothetical. In practice, the SII and RBM components are inherently linked to each other. The baseline calculations presented here follow the SAR methodology in order to make appraisal and ICR estimates comparable. The effects of a different allocation of the components' benefits are discussed in the sensitivity analysis in Annex 3. Changing the assumptions of the allocation of project benefits demonstrates that a key driving force generating the benefits may be attributed to the improvement of irrigation efficiency. Using alternative scenarios on project life and a reduction in incremental benefits, the ICR estimate was found to be sufficiently robust. Other benefits for downstream users (other than hydropower generation) are discussed in the sensitivity analysis in Annex 3. The inclusion of wetlands is considered to be an important modification, as it became evident during the project that benefits arising from improved scheme irrigation efficiency include a variety of downstream users, in particular those in the Usangu catchment. Overall, the consideration of wetlands and other forms of downstream uses slightly increases the ERR of the RMB component, but does not lead to dramatic changes in the overall ERR. 4.4 Financial rate of return: Table 1 presents the FRR, estimated at 21.5 percent (NPV Tsh. 8,414 million), which is not reported in the SAR. 4.5 Institutional development impact: The project had a substantial impact on institutional development. Tanzania now has a new National Water Policy and a streamlined institutional framework, and different levels and categories of stakeholders have received training on the new legislation. These have built capacity to enforce water issues. The operationalization of the new institutional framework, capacity building at all levels of WRM, and the establishment of Catchment Organizations have strengthened mechanisms for coordination and integrated basin planning, and developed appropriate linkages with ongoing Local Government reforms. RBMSIIP has also enabled clear and professional communication between MWLD and MAFS on water allocations, and how to determine and enforce them. This working relationship has increased the understanding of how water is used for irrigation in the basins. Stronger BWOs and BWBs have markedly improved the abilities of BWOs to plan, allocate, operate and manage water resources in the two basins, and in some instances, resolve water allocation conflicts. BWOs - 13 - also work with DCFTs in solving disputes, examining new applications for water, and assessing environmental implications. In addition, the establishment of catchment water organizations that link with WUAs and the Irrigation Organizations enables all users to participate in water use discussions, and provides a mechanism for addressing areas of community concern. 5. Major Factors Affecting Implementation and Outcome 5.1 Factors outside the control of government or implementing agency: The intense competition over water in the target basins may have contributed to the perceptions of some farmers that basin management is a ploy to deprive them of exercising their historical (customary) rights to use water for irrigation. Farmers also view themselves as the losers in their trade of water for hydropower. Since electricity is primarily provided for urban domestic and industrial use, farmers claim they have not seen the benefits of hydropower production. In addition, the location of the Pangani Basin Water Office within TANESCO premises at Hale has aggravated these misconceptions, and eroded the strength of participation in basin management. The project has moved the Pangani office to Moshi, to dispel these fears that have become disruptive. Future water resources interventions need to take these issues into consideration. 5.2 Factors generally subject to government control: Government commitment contributed to the success of water sector reforms, and, in particular, to establishing effective basin management. GOT was also persistent in pursuing the adoption of the new Water Policy. The Policy explicitly recognizes several key principles that are fundamental to successful WRM, including the joint management of both water quantity and quality on a day-to-day basis, establishing integrated participatory management approaches, and adopting the river basin as the unit of planning. Despite other government institutional reforms and restructurings, efforts were made to ensure that project implementation arrangements were minimally affected. 5.3 Factors generally subject to implementing agency control: The SII component was rated 'unsatisfactory' from May 2001 to March 2003, due to lagging implementation on rehabilitation works. One of the major causes of delay throughout the project was the time taken by the Central Tender Board to process proposals from both contractors and consultants. MAFS engineers were new to the supervision of contractors, contributing to cost overruns in all schemes. 5.4 Costs and financing: Project Costs. The total project cost, including contingencies, was estimated at US$30.75 million over six years. The project closed on June 30, 2004, following two extensions of one year and six months. Actual total project cost over approximately 7.5 years was US$24.81million. Although 97.4 percent of the Credit was expended, the lower-than-predicted total cost is due to changes in the exchange rate between the SDR and the US dollar. (The exchange rate in May 1996 was SDR 1 = US$1.45. At the time of the ICR it was SDR 1 = US$1.347.) Project Financing. IDA financing was estimated at SDR18.2 million, equivalent to US$26.3 million (or 85.5% of total). GOT was expected to provide US$1.53 million (5% of total) and the farmers contribution was estimated at US$3.0 million (10% of total.). IDA credit disbursements totaled SDR17.73 million (US$23.88 million equivalent) on October 31, 2004, the last date for approval of credit disbursements, leaving SDR469,000 (US$720,000 equivalent) undisbursed, and US$74,053 in the Special Account (to be refunded). GOT and farmers contributed US$3.51 million and US$0.94 million respectively. 6. Sustainability - 14 - 6.1 Rationale for sustainability rating: River Basin Management--Institutional Sustainability. Policy implementation and institutional reforms are an ongoing process. Institutional sustainability will depend on GOT's commitment to continue support to integrated WRM and basin water organizations. The new Policy and legal framework provide the necessary mechanisms for stakeholder collaboration, and the institutional framework supports effective participation in all aspects of WRM at the different levels (National, Basin, Catchment, District and Community/Water User level). Institutional reforms have transformed MWLD and have strengthened its capacity to support basin water organizations. Also, GOT has recently established BWOs in all remaining basins, confirming its commitment. Staff who were contracted under RBMSIIP's project coordination team have now rejoined the Department, and should contribute to continuity and sustainability. River Basin Management--Financial Sustainability is unlikely at currrent levels of funding. Better management of basin water resources and financial sustainability requires improved funding. The main functions of the BWO include: (i) general regulatory functions (data collection and monitoring, inspections and control of abstractions and discharges, collection of fees and fines, and information and awareness creation); and (ii) individual-based functions (granting of water rights and issue of discharge permits). In the Pangani basin, a study of financing mechanisms shows that the main sources of funds over the past three years (2000/01 - 2002/03) have been, on average, the WUF (36%), central government (MWLD) (29%), TANESCO royalties (20%), and funds from RBM (12%). Support from stakeholders to address specific issues amounted to 1 percent. To obtain government funds, the BWO has to submit a case to the Ministry every year, but there is no guarantee that money will be available. This depends on the Ministry's other commitments and priorities. During the three consecutive years up to 1999/2000, the Pangani basin did not receive any government funding except for personal emoluments (Annex 7 - Turpie, Ngaga and Karanja, 2003, p. 86). The recurrent budget required to maintain sustainable WRM in the Pangani for the financial year 2003/04 is estimated at Tsh. 380 million, and the financing gap is about 27-30 percent. Due to inadequate funding, both basins face problems such as: water users abstracting more water than allocated in their water permits; use of water without formal water permits; inadequate monitoring of inefficient water use by abstractors; inadequate enforcement mechanisms; inability to formulate integrated planning; and inadequate human resources. Allocations from government are too small to allow the BWO to carry out their duties satisfactorily. In some years, actual releases from the government have been even smaller than the allocations (NORPLAN, 2000). Given the pressures on the government's budget, and priorities within MWLD, the feasible approach to funding basin management is to focus on the WUF and royalty payments. These are also funds that GOT allows to be collected and retained by the Ministry/Basin offices. Retaining WUF by the BWOs is an incentive to collect more revenue from fees, and explains the improvement in granting water rights and collecting WUF. This is a move towards greater decentralization and financial autonomy of WRM within a common national framework. On the other hand, the fact that BWOs are allowed to retain WUF partly explains why the government may not give priority to allocating sufficient funds to the management of basin water resources. In principle, WUF /royalty should be the most important source of revenue, as TANESCO and irrigation (small- and large-scale) use over 90 percent of the water abstractions. There is an enormous capacity to increase the revenues from user fees, due to the large degree of non-payment, and due to the fact that most users are currently not charged for water use at all. In the Rufiji, of a total of about 1,514 abstractions, 1050 are billed, and of these, only 70 percent pay. In the Pangani, of a total of 3,359 abstractions, about 1292 are billed, and 40 percent pay. The NORPLAN 2000 report finds that the royalty paid by TANESCO could also be increased without significantly affecting TANESCO's costing. In 1997, the royalty amounted to 0.1 percent of TANESCO's electric power sales and about 0.14 percent of the cost of sales. In 2002, royalty payments amounted to 11 percent of the - 15 - generation costs and about 2 percent of the total cost of sales. The TANESCO royalty payment is independent of their production or water use. If TANESCO is charged a WUF based on water abstracted, as for other stakeholders, they would be given an incentive to optimize production according to water aspects in the basin. The study finds that unless the WUF is doubled several times, it will have no impact on hydropower production. Also, given TANESCO's monopoly situation, costs may relatively easily be passed onto customers. The study also concludes that the current levels of WUF are too low to have a "significant" effect on water allocation, and recommends a substantial increase in the WUF for irrigation. Raising fees, however, will meet some resistance from water users. TANESCO and industries are already complaining about inadequate water, while large-scale farmers complain about excessive taxation. Therefore, most of the fees collected must be reinvested in the management of water, to justify the increase, and for users to appreciate the efforts made. This means the BWO has to spend more time on the general regulatory functions mentioned above. If these activities are improved, large users like TANESCO are likely to benefit from enhanced water management in the basins. MWLD is currently reviewing fees for different categories, taking into consideration the increased demand for water, higher pollution, and inflation (Turpie, Ngaga and Karanja, 2003, p. 91). If the WUF is to become an effective allocation instrument, the rates must be relatively high, especially for small- and large-scale irrigation, to provide incentives for water-saving measures. In addition to raising user fees, strategies to improve the financial and sustainable management of basin water resources include investments to enhance billing and fee collection in terms of manpower and resources, to enable the basin office to identify all live water abstractions, facilitate them to obtain water rights and bill them, strengthen fee collection mechanisms, and monitor to ensure compliance. There is also a need to create incentives for timely payment. Financial sustainability of the RBM component will depend on all these factors. Smallholder Irrigation Improvement component--Overall Sustainability. With farmer training, the increased understanding of water management and water use efficiency, and the introduction of water rights charges, IOs appear to have realized that water has value and costs, and that each user needs a water right in relation to available resources and other users. All improved schemes have formalized IOs that are able to meet water rights charges and fund O&M through income from increased yields. Crop Production Plans that provide guidance on crop selection for each season, combined with the training that farmers have received, should enable them to continue obtaining higher yields, and contribute to funding O&M. Proper O&M of irrigation infrastructure facilities by farmers is critical to sustainability. However, farmers need technical guidance from engineering staff, particularly prior to the start of the maintenance period. Structures and canal systems could fall into disrepair if the essential preventative maintenance of scheme structures and lined canals is not supported. This should be provided by the District Irrigation Officers (DIOs) under the revised structure. RBMSIIP--Environment Sustainability. The project aimed to improve irrigation efficiency from 15 percent to 30 percent through investments in infrastructure modernization, establishment of irrigation and water user associations, and capacity building. Water in excess of what is needed for optimum crop production was returned as much as possible to the river through improved drainage. Additional wastage was reduced through the project supported catchment work, i.e. the Water Managers forum in the Usangu Catchment. The higher irrigation efficiency, as a result of the above efforts, had a positive impact on water availability to downstream water users, including the Usangu wetlands and the Ruaha National Park. During RBMSIIP, a review of Bank Safeguard Policies had recommended specific action to remain in compliance with the Natural Habitats Safeguard. The negative environmental impact of expanded irrigation on the Usangu wetlands and Ruaha National Park during the dry season had been established. In consequence, and to avoid additional stress to the natural habitats of the areas, measures were agreed with - 16 - MAFS and MWLD to reduce the scope of improvement works, as well as the number of schemes. Works were to include rehabilitation of off-takes and of drainage infrastructure; concrete weirs were no longer to be constructed. The Usangu has attracted nation-wide attention by a public commitment from the Government to restore flows in the Great Ruaha river by 2010. Sustainability of RBMSIIP is rated as likely, based on the the overall sustainability of the SII component, especially the strong response of farmers in operating and maintaining their irrigation schemes and observed yield increases. The financial sustainability of the RBM component, however, depends on improved financing mechanisms at the basin level, including efficient water right administration, economic WUF and royalty payments, and government funding to cover some of the general regulatory functions of BWOs. 6.2 Transition arrangement to regular operations: The new Water Policy has to be effectively executed to continue successful WRM. A human resource assessment needs to be carried out to determine the staffing needs and capacities to implement the policy, legal and institutional framework. The assessment should address: (i) the shift from a regional to a basin perspective; (ii) the changing staff profiles (with no new hydrologists being hired and many close to or already at retiring age, and shifting roles of the regional hydrologists and water engineers to basin water officers and hydrologists); (iii) emerging needs for staff with training on water law, economics, international relations, international water law, and environmental planning; and (iv) the capacities of the national academic institutions (such as UDSM, Sokoine University, Dar Technical College and the WRI) to meet the emerging needs of the water sector. Some of these issues are examined under the Bank's ongoing Tanzania Country Water Resources Assistance Strategy. RBMSIIP has invested in establishing a network for collectiong hydro-meteorological data in the two basins. The extent to which O&M of the stations can continue to be financed is determined by the value of the data to various users and uses. Increasing the value of the data involves knowing the demand for such data. While the project has established hydrometric stations, and has collected, stored and processed data, there is a need to move beyond data collection, and respond in a more demand-driven way to data collection and supply. This requires further thinking on the `market' for hydro-meteorological data, including the potential users and uses of the data, the extent to which demand can be satisfied, and what is required to better serve potential clients. Such an assessment would provide a coherent context and clarification not only to the kind of data needs and their level of analysis, but also to the identification of staffing requirements and capacities, training needs and educational programs, the contents and location of databases, and issues related to data collection technology. The IDA-funded Tanzania Water Sector Support Project plans to address some of these issues. 7. Bank and Borrower Performance Bank 7.1 Lending: The Bank's performance during preparation was satisfactory for all the issues discussed in Section 3.5. 7.2 Supervision: Project supervision is rated satisfactory. The project was granted two extensions--a one year extension following the MTR to complete the delayed rehabilitation works, and a six-month extension in late 2003, for the procurement of a limited number of specific items. The Bank fielded 11 supervision missions (and one ICR mission) during the seven-and-a-half years of implementation. Missions were regular, although project records show no supervision missions between February 2000 and March 2001. The MTR was postponed from late 2000 to March 2001 due to Bank Budget problems. In general, missions included - 17 - experts in WRM, irrigation, agriculture and environment. Supervision reports and Aide Memoires provided a comprehensive review of implementation progress and clearly indicated required follow-up actions. At one point during the period May 2001 to March 2003, when the SII component was rated 'U', the project seemed at risk and was borderline unsatisfactory/satisfactory. The team discussed the rating, including with the manager, and decided to retain the 'S' rating in light of good progress in rehabilitating hydrometric stations and staff training under RBM, and farmer training under SII, and to communicate the unsatisfactory elements clearly in the Aide Memoire. A dam safety inspection of the Nyumba ya Mungu hydropower dam was recommended during implementation, but not carried out under RBMSIIP. (No other smallholder irrigation improvement projects downstream of the dam can be financed by Bank funds unless it has been proven that the dam complies with international dam safety regulations.) 7.3 Overall Bank performance: The Bank's overall performance is satisfactory, with good design and supervision. The only issue is that a change in the project management arrangements, as a result of the correct application of the Bank's rules on hiring civil servants under consulting contracts financed under Bank credits, led to reduced project staff morale and effectiveness of implementation of RBMSIIP. Borrower 7.4 Preparation: The Borrower's preparation performance is assessed as satisfactory. One issue is that IDA was not kept informed of a parallel FAO initiative in the water sector, before, during or after RBMSIIP's appraisal, nor when Ministry of Water staff were in Washington in 1996 to work on the water sector study. Following the RBMSIIP Supervision Mission in April - May 1997, IDA expressed concern (in the Aide Memoire) over the apparent lack of transparency in related activities that was resulting in inconsistent approaches. This also raised several questions about the credibility of IDA's dialogue with the Ministry, and its involvement with the Ministry in many activities within the sector. The FAO project was subsequently cancelled, and its outputs folded into RBMSIIP. 7.5 Government implementation performance: As indicated earlier, achievements in policy, institutional and legal framework were delayed due to the long but necessary stakeholder consultations, and the decision to combine three policies into one National Policy. However, GOT's commitment to these difficult reforms, involving many conflicting interests, has helped to establish WRM as a key priority in Tanzania's National Development Vision. Satisfactory Government implementation performance is also attributed to the close collaboration and coordination of all key players through the three management committees that were established: the Inter-Ministerial Steering Committee; Technical Steering Commitees; and the Basin Technical Committees. These committees met as scheduled to provide policy and technical guidance. Also, the Project Coordinators for both components remained on RBMSIIP for the entire seven-and-a half years, contributing to sound implementation. 7.6 Implementing Agency: Overall, the implementing agencies, MAFS and MWLD performed satisfactorily. The main bottleneck, as mentioned in Section 4.2, was the delay on the part of MAFS in scheme construction. The long drawn bureaucratic process in commissioning suitable international consultants, the tendering procedures, and the relative inexperience of MAFS engineers in supervising contractors, contributed to significant delays in the construction of Group A (4 schemes) and B (8) schemes. This necessitated reducing the total number of schemes to be rehabilitated. 7.7 Overall Borrower performance: Overall Borrower performance is rated satisfactory based on the sound design and GOT's strong - 18 - commitment to reforms. During the first two years of the project, insufficient Government counterpart funds and delays in releasing budgeted counterpart funds caused concern for IDA, however, the issue was resolved. 8. Lessons Learned RBMSIIP's approach of targeting synergies between water resources management and irrigated agriculture remains valid, given that irrigated agriculture is the largest and least efficient water abstractor, and is located upstream of all other users. Addressing irrigation development from a river basin, multi-stakeholder perspective enables to firmly place on the agenda the impact of irrigation abstractions on up- and downstream water use, establish mechanisms for resolving inter- and intra-sectoral water allocation conflicts, and provide incentives for increasing irrigation efficiency and productivity. RBMSIIP put into practice the Bank's 1993 Water Resources Management Policy, which called for a more comprehensive approach to WRM, incorporating cross-sectoral perspectives, environmental concerns, and stakeholder participation. Project design was sound, and overall, RBMSIIP helped to understand the complexities of such interventions and the achievements. Policy and institutional reforms, and the transformation of Basin Water Organizations into accountable, transparent, financially autonomous and independent organizations are slow and difficult, but necessary for instituting changes. Understanding the local context is critical to WRM, and this needs to be assessed independently in each area. This experience has necessitated a consideration for customary law assessments in the review of the legal framework. The project has demonstrated that limited but targeted investments in irrigation and drainage infrastructure improvement can provide an adequate incentive in terms of increased productivity as a "carrot" to encourage farmers to reduce water wastage and to consider the water demands other users. Investment of approximately US$2000 per hectare in upgrading small-scale irrigation schemes using participatory methods, when coupled with adequate agricultural advice, can bring a measurable improvement in the welfare of smallholders at acceptable rates of return. Farmers are also willing to contribute significant amounts of their own time towards the costs of rehabilitation works, and are capable of operating and maintaining their own irrigation schemes. However, they require periodic support from Scheme Extension Agents and District Irrigation Officers to establish O&M procedures and appropriate cropping calendars. The participatory approach of RBMSIIP, involving farmers at all stages, is a slow but essential part of the process of ensuring improved scheme performance and sustainability, and a demand-driven spirit among beneficiaries. Scheme Improvement Assessment Reports (SIAR) provided a good basis for smallholder scheme developments. Under the future arrangements, districts should be encouraged to prepare these reports for any schemes being considered for assistance under district plans. This will ensure that all aspects, particularly those relating to conflicting water demands and availability, are addressed. 9. Partner Comments (a) Borrower/implementing agency: See Annex 9 for Government of Tanzania ICR. (b) Cofinanciers: - 19 - (c) Other partners (NGOs/private sector): 10. Additional Information - 20 - Annex 1. Key Performance Indicators/Log Frame Matrix Outcome / Impact Indicators: 1 Indicator/Matrix Projected in last PSR Actual/Latest Estimate Basin Management Component: (a) River Guaging Stations Operational Pangani 28; Rufiji 43 Pangani 28; Rufiji 43 (b) Basin Office Staff Trained under Project Pangani 10; Rufiji 15 Over 70 technicians trained in basic computer skills, and 30 hydrologists and technicians trained in equipment installation and maintenance. Also, Water Lab Unit officials trained in both Pangani and Rufiji (c) Vehicles and Equipment provided Pangani US$1.8m; Rufiji US$2.6m Pangani US$1.38m; Rufiji US$1.92m Smallholder Irrigation Component: (a) Smallholder schemes assisted by Pangani 40; Rufiji 40 Pangani 20; Rufiji 14 multidisciplinary teams (b) Water User Groups registered (as Pangani 8; Rufiji 7 Pangani 9; Rufiji 10 irrigation association/coop society) (c) Smallholder Schemes rehabilitated (i) Schemes (number) Pangani 15; Rufiji 15 Pangani 8; Rufiji 7 (ii) Area irrigated (ha.) Pangani 3500; Rufiji 3500 Pangani 2904; Rufiji 2155 Output Indicators: 1 Indicator/Matrix Projected in last PSR Actual/Latest Estimate Basin Management Component: (a) Water Rights Issued (Number) Pangani 3000; Rufiji 2200 Pangani 1292; Rufiji 1050 (b) Economic User Fees collected (US$) Pangani 26700; Rufiji 21700 Pangani 58800; Rufiji 52600 (does not include Tanesco Royalty fees) (c) Pollution Fines collected No estimate in PSR None collected (see Section 4.1) Smallholder Irrigation Component: (a) Irrigation Efficiency (%) Pangani 30%; Rufiji 30% Pangani 27%; Rufiji 27% (b) Family Income from Irrigation in improved Pangani US$1100; Rufiji US$740 Pangani US$1500; Rufiji US$1100 schemes (US$/ha) (c) Farm families in improved schemes Pangani 3500; Rufiji 3500 Pangani 3472; Rufiji 1845 (d) Rice yields in improved schemes Pangani 5; Rufiji 4 Pangani 5.27; Rufiji 4.06 (weighted average tons/ha) 1End of project - 21 - Annex 2. Project Costs and Financing Project Cost by Component (in US$ million equivalent) Appraisal Actual/Latest Percentage Component Estimate Estimate of (US$ million) (US$ million) Appraisal A. River Basin Management 1. Strengthening National Water Resources Mgmt. 4.48 3.89 86.8 2. Strengthening Basin Management 5.37 4.64 86.4 B. Smallholder Irrigation Improvement 1. Participatory Scheme Selection and Organization 2.21 2.16 97.7 2. Engineering Works on Schemes 13.00 11.73 90.2 3. Mitigation of Negative Environmental Effects 0.66 0.60 90.9 4. Irrigation Sector Capacity Building 1.72 1.79 104.1 Total Baseline Cost 27.44 24.81 90.4 Physical Contingencies 2.03 Price Contingencies 1.28 Total Project Costs 30.75 24.81 Total Financing Required 30.75 24.81 Note: The total Project Cost (actual/latest estimate) of US$24.81 million is equivalent to SDR 18.45 million at the exchange rate of SDR 1 = US$1.347 (time of ICR). This rate was SDR 1 = US$1.45 in May 1996 (SAR). Project Costs by Procurement Arrangements (Appraisal Estimate) (US$ million equivalent) Expenditure Category Procurement Method 1/ Total Cost ICB NCB Other 2/ 1. Civil Works - Contracts 3.158 9.474 0.000 12.632 (3.158) (9.474) (0.000) (12.632) 2. Civil Works - Farmers 0.000 0.000 2.962 2.962 (0.000) (0.000) (0.000) (0.000) 3. Vehicles & Equipment 5.132 0.307 0.264 5.703 (5.019) (0.300) (0.258) (5.577) 4. Consultants 0.000 0.000 5.345 5.345 (0.000) (0.000) (5.345) (5.345) 5. Training 0.000 0.000 0.822 0.822 (0.000) (0.000) (0.822) (0.822) 6. Operating Costs and 0.000 0.859 2.433 3.292 Allowances (0.000) (0.500) (1.416) (1.916) Total 8.290 10.640 11.826 30.756 (8.177) (10.274) (7.841) (26.292) - 22 - Project Costs by Procurement Arrangements (Actual/Latest Estimate) (US$ million equivalent) Expenditure Category Procurement Method 1/ Total Cost ICB NCB Other 2/ 1. Civil Works - Contracts 2.63 8.846 0.431 11.907 (2.63) (8.84) (0.000) (11.47) 2. Civil Works - Farmers 0.000 0.000 2.03 2.030 (0.000) (0.000) (0.000) (0.000) 3. Vehicles & Equipment 1.851 0.648 0.975 3.474 (1.65) (0.60) (0.85) (3.10) 4. Consultants 0.188 0.416 3.298 3.902 (0.000) (0.000) (3.12) (3.12) 5. Training 0.000 0.000 1.377 1.377 (0.000) (0.000) (1.37) (1.37) 6. Operating Costs and 0.000 0.000 2.120 2.120 Allowances (0.000) (0.000) (1.416) (1.416) Total 4.699 9.910 10.925 24.810 (4.17) (9.81) (6.84) (20.47) 1/Figures in parenthesis are the amounts to be financed by the Bank Loan. All costs include contingencies. 2/ Includes civil works and goods to be procured through national shopping, consulting services, services of contracted staff of the project management office, training, technical assistance services, and incremental operating costs related to (i) managing the project, and (ii) re-lending project funds to local government units. Project Financing by Component (in US$ million equivalent) Percentage of Component Appraisal Estimate Actual/Latest Estimate Appraisal Bank Govt. Total Bank Govt. Total Bank Govt. A. River Basin Management 1. Strengthening National Water 4.48 0.49 4.97 3.89 0.79 4.68 86.8 161 Res. Mgmt. 2. Strengthening Basin Mgmt. 5.37 0.58 5.95 4.64 0.96 5.60 86.4 165 B. Smallholder Irrigation Improvement 1. Participatory Scheme Selection and Mgmt. 2.21 0.19 2.40 2.05 0.11 2.16 92.8 57.9 2. Engineering Works on Schemes 13.00 15.03 9.92 0.87 11.73 76.3 873 3. Mitigation of Negative Environmental Effects 0.66 0.66 0.59 0.01 0.60 89.4 25.0 4. Irrigation Sector Capacity Building 1.72 0.04 1.76 1.03 0.77 1.80 60.0 97.5 Price Contingencies 0.87 0.79 Total Project Costs 28.31 2.09 30.77 22.12 3.51 26.57 78.1 176 Total under Engineering Works includes farmer contributions of US$ million 2.03 (Appraisal) and 0.94 (Actual/Latest). - 23 - Annex 3. Economic Costs and Benefits Identification of Project Benefits 1. Project benefits fall into two inter-dependent categories. In the first case, the returns from the Smallholder Irrigation Improvement (SII) component came from increased yields due to rehabilitated infrastructure and other factors, such as seasonal risk reduction, the improvement of communal scheme management, farmer's training in water management, and in particular, better access to inputs and knowledge required for agricultural intensification. In the second case, due to the improvement of irrigation efficiency and the implementation of the River Basin Management (RBM) component, positive externalities arose for a variety of downstream users. At appraisal of the project, it was assumed that the main economic benefit is additional hydropower generation. Other potential benefits for downstream users, including the Usangu wetlands, are discussed in the sensitivity analysis. 2. Indirect benefits of RBM, which are not included in the following analysis are: (i) reform of the regulatory framework for WRM; (ii) improved institutional and analytical capacity of implementing agencies; (iii) strengthening of information and resource monitoring in the Rufiji and Pangani water offices; (iv) enhanced forecasting capacity; and (v) rehabilitation of the hydrometric network in the basins. Important indirect benefits of SII that are not included in the analysis are: (vi) better access to schooling for children; and (vii) better provision of healthcare through improved incomes. While many serious health issues in the project area remain, field evidence suggests that the prevalence of malaria, diarrhea and others diseases is declining. Production Benefits 3. The production impact of SII is presented in Table 1. Crop yields for rice rose from about 1.5 t/ha to 4.1 t/ha for the Rufiji basin and 2.0 t/ha to 5.3 t/ha for the Pangani basin. Crop yields for maize rose from about 1.1 t/ha to 3.3 t/ha for the Rufiji basin and 1.1 t/ha to 4.9 t/ha for the Pangani basin. The figures are weighted averages based on a minimum of 20 harvest sample cuttings on each of the 15 schemes in the two basins, with an average area of 10 square meters per sample. Estimates from the Project Coordinating Unit (PCU) of the SII component suggest that crop yields for tomatoes and onions have also increased significantly. Table 1. Production Impact of SII Component Average Yields (t/ha) SAR ICR Baseline With Project Baseline With Project Rufiji Basin Rice 1.8 4.0 1.5 4.1 Maize ... ... 1.1 3.3 Tomatoes ... ... 2.0 3.0 Onions ... ... 2.5 3.5 Pangani Basin Rice 2.0 5.0 2.0 5.3 Maize ... ... 1.1 4.9 Tomatoes ... ... 2.0 4.0 Onions ... ... 2.5 4.0 4. In the SAR, incremental benefits were measured mainly for rice (assumed to cover 95% of the project - 24 - area). The estimates were based on two schemes deemed to be representative for the Rufiji and Pangani basins. The ICR includes two modifications to measure production benefits more precisely. First, incremental benefits are calculated for all 15 improved irrigation schemes. Second, maize, onion and tomato production are included in the analysis. Based on SII estimates, rice is assumed to cover 50% of the project area, followed by maize (40%), onions (6%), and tomatoes (4%). SII component estimates suggest that there were no major changes in cropping patterns due to the project. Total cropping intensity, however, improved from 60% to about 95%. Hydropower Benefits 5. Improved irrigation efficiency, (which increased in the dry season from about 11% to about 27%) and the implementation of the RBM component monitoring and enforcing water rights, greatly helped to reduce the irrigation and distribution losses of water. Particularly in the dry season, it allowed a better redirection of water to downstream uses. In line with the SAR, the effect of water savings is illustrated by using its capacity to generate additional energy in hydroelectric power plants, located below the irrigated areas. Consultations with implementing agencies and field visits suggest that spillage of Mtera and Kidatu dams was negligible during the project cycle. Consequently, the full amount of the estimated water savings was monetized. 6. In consultation with the implementing agencies, the ICR assumes an average annual field requirement of 5,250 cubic meters/ha of water for crop production, and values water at 34 Tsh/cubic meter (based on secondary sources. For details see end of this Annex). The most important hydroelectric power plants, located in the Rufiji sub-basin, are Kidatu (204 MW) and Mtera (80 MW). They deliver about 85% of electricity to the national grid. In fact, data from TANESCO show that hydropower generation from these two plants increased by about 28% from 1996 to 2003. Financial and Economic Rates of Return 7. The financial and economic rates of return have been re-calculated for each project component based on incremental benefits. In general, since itemized information was not recorded for specific activities on a yearly basis, costs were estimated on the basis of yearly disbursements and maintenance costs (labor and other inputs, including projected future maintenance costs) during the life of each component. The overall ERR for the project is estimated at 16% (NPV Tsh. 2,934 million), and is lower than the appraisal estimate (see paragraph 8). Table 2 also presents the FRR, estimated at 21.5% (NPV Tsh. 8.414 million), which is not reported in the SAR. Table 2. Financial and Economic Rates of Return FRR ERR Component SAR ICR SAR ICR Overall (SII + RBM) ... ... 21.5% 34.0% 16.0% Smallholder Irrigation Improvement (SII) ... 21.8% 12.1% 10.4% River Basin Management (RBM) ... 21.2% 49.2% 1/ 20.9% 1/ The ERR for the RBM component is erroneously reported at 34% in the SAR. A re-calculation, based on the original SAR project data, suggests that the correct value is 49.2%. 8. The estimated rates of return for both components are summarized in Table 2. The SII ERR is 10.4% and is somewhat lower than the appraisal estimate. This reflects: (i) the delays of the rehabilitation works (the first schemes where fully rehabilitated in early 2002); and (ii) a significant reduction of the rehabilitated area of smallholder schemes (5,059 ha compared to 7,000 ha planned at appraisal). Also the - 25 - ERR for RBM, estimated at 20.9%, is lower than at appraisal. In addition to the factors mentioned above, this is due to: (iii) the revision of the field requirements of water for crop production; and (iv) a conservative estimate of the price of water for hydropower generation. It is important to emphasize that the allocation of project benefits to both components is somewhat hypothetical. In practice, the SII and RBM components are inherently linked to each other. However, the baseline calculations presented here follows the SAR methodology in order to make appraisal and ICR estimates comparable. The effects of a different allocation of the components benefits will be discussed at the end of the following sensitivity analysis. Sensitivity Analysis 9. Table 3 presents a sensitivity analysis of the ERR, using alternative assumptions about the project life (scenarios 1 and 2), and a reduction in incremental benefits (scenario 3). These scenarios were chosen in consultation with stakeholders, as they represent events that are more likely to occur. Other scenarios, including more optimistic ones, were purposely not considered. Overall, the baseline estimate was found to be sufficiently robust, as the ERR remains stable in a range of about 13% to 17%. Table 3. Sensitivity Analysis ERR Assumptions of Scenario Overall SII component RBM component 1. Project life of 15 years 12.7% 5.9% 18.5% 2. Project life of 25 years 17.1% 12.1% 21.6% 3. A 10% reduction in incremental benefits 13.9% 8.5% 18.6% 4. Value of water saved not only includes hydropower generation, but also the value of watersheds, Ruaha National Park and other uses 11.2% 10.4% 11.5% (lower bound scenario) 5. Value of water saved not only includes hydropower generation, but also the value of watersheds, Ruaha National Park and other uses 17.8% 10.4% 24.0% (upper bound scenario) 6. A different allocation of the value of water 17.8% 25.3% 6.9% saved to the SII and RBM components (based on data from scenario 5) 7. Baseline estimate (Table 2) 16.0% 10.4% 20.9% ______________________________________________________________________________________ 10. Table 3 also presents a different calculation of the value of water saved for downstream users. This is considered to be an important modification, as it became evident during the project cycle that benefits arising from improved irrigation efficiency not only benefit hydropower generation, but also a variety of downstream users in the Pangani and Rufiji basins, in particular those in the Usangu catchment. Providing a lower and upper bound estimate of the economic value of water covers potential methodological as well as data uncertainties. The upper bound estimate (scenario 5) suggests that consideration of wetlands and other forms of downstream uses increases the ERR of the RMB component, but does not lead to dramatic changes in the overall ERR. The lower bound estimate (scenario 4) slightly decreases the returns to RBM, possibly due to an underestimation of the genuine value of water for downstream users. 11. Finally, Table 3 presents a change in the methodology regarding the allocation of the project benefits - 26 - (scenario 6). In the previous calculations, following the SAR methodology, a more efficient use of water is credited entirely to the RBM component. The rationale is that an improved river basin management system enforces and monitors water rights for the highest alternate user. In practice, however, a more efficient use of water should be allocated as well to the SII component because the benefits or water saved arise from improved irrigation efficiency. Based on data from scenario 5, the modifications involve the following steps: (i) the value of water saved is calculated using the average costs of water for all sectors compounded. In addition to the benefits from crop production, these benefits are now attributed to the SII component. Assuming that there would be no RBM component, average costs are used here as a proxy for a 'wild' allocation of water; (ii) The benefits of the RBM component are calculated separately, according to the priorities stated in the National Water Policy (i.e., first to human use, second to environmental use, and third to all other uses). Acknowledging that the RBM component leads to a more efficient allocation of water, the difference between (i) and (ii) is now attributed to the RBM component, in accordance with actual water consumption. The exercise demonstrates that the ERR of the SII component increases significantly, and that a key driving force generating the project benefits may be reasonably attributed to the improvement of irrigation efficiency. However, it should be stressed that both components are inherently linked to each other, and that the genuine benefits of the RBM component are notoriously difficult to measure. As such, one should use the scenario as indicative of broad patterns, rather than for the exact numbers produced. Underlying Assumptions 12. In addition to the discussion in the previous paragraphs, the following are relevant clarifications of the underlying assumptions: (a) All costs and benefits are expressed in Tanzanian shillings (Tsh.) at constant 2004 prices; (b) In line with the SAR, a 12% discount rate is used throughout the analysis. Moderate variations of the discount rate have a negligible impact on the ERR; (c) Due to delays in scheme rehabilitation in the SII component, the first schemes where fully rehabilitated in early 2002, i.e. year 6 of project implementation. However, benefits of the SII component are assumed to start in FY 2000-2001, i.e. year 5 of project implementation. This reflects the fact that the project did not only increase water use efficiency through infrastructure rehabilitation, but in particular, provided substantive training for farmers to increase crop production. Field visits during project implementation strongly support this assumption. Even in partly rehabilitated irrigation schemes, farmers were already benefiting from increased training opportunities; (d) In consultation with the implementing Ministries, smallholder irrigation O&M costs are estimated on an annual basis of 5% of construction and operating costs, equivalent to about 3% of average annual SII component costs. Similarly, RBM O&M costs are estimated at 45% of operating costs, equivalent to about 6% of average annual RBM component costs. Moderate variations of O&M costs do not have a significant impact on the overall ERR; (e) Table 4 gives the financial prices and the allocation of water for downstream users (including the wetlands). The data are based on direct use values for agricultural production or household consumption. It includes: (i) hydropower; (ii) domestic use; (iii) lowland livestock; (iv) lowland traditional furrows; (v) agro-industrial use; (vi) aquatic ecosystems; and (vii) the Ruaha National Park. The lower bounds for aquatic ecosystems are proxied by water user fees for small-scale irrigation. The upper bounds are proxied by the use value of lowland livestock. In the absence of detailed data and baseline studies, the assumption - 27 - here is that the next alternative use for wetlands is either small-scale irrigation (lower bound) or livestock production (upper bound). Equally, assuming that the next alternative use for the Ruaha National Park would be livestock production, the incremental benefits are roughly proxied by their direct agricultural use value; (f) The focus on production effects, ignoring as such important ecological functions and non-use benefits of wetlands, and the use of proxy indicators, may underestimate the genuine economic value for downstream users. However, the assumptions here may be tolerable for two reasons. First, estimates from the implementing Ministries and the Rufiji and Pangani water offices--taking into account the effects of evaporation and detangled nature of the river system during the dry and wet seasons--suggest that water savings through improved irrigation efficiency mainly benefit hydropower generation, (which is in line with the original assumptions of the SAR), and only to a lesser extent for other downstream uses. Second, by using an upper and lower bound scenario, the evaluation exercise explicitly takes into account data uncertainties; Table 4. Value of Water Saved in Different Uses and Project Impact on Water Saved Through Improved Irrigation Efficiency 1/ Tsh/m3 Rufiji Basin 2/ Pangani Basin Type of Use Lower Upper % Total Savings, % Total Savings, Bound Bound of Use mio m3 of Use mio m3 Hydropower 34 46 72.00 11.14 54.32 8.10 Domestic use 1,200 1,500 0.20 0.03 1.50 0.22 Lowland livestock (cattle, goats) 479 926 0.04 0.01 1.28 0.19 Lowland traditional furrows, small-scale irrigation 35 109 27.20 4.21 42.43 6.33 Aquatic ecosystems 35 926 0.16 0.02 ... ... Ruaha National Park 479 926 0.39 0.06 ... ... Sugar estates 32 101 0.01 0.00 0.20 0.03 Commercial flower farms 3,500 5,300 ... ... 0.27 0.04 1/ Data refers to 2004. Estimated project's impact on water savings through improved irrigation efficiency are from the Ministry of Water and Livestock Development, based on data from the Rufiji and Pangani water offices. 2/ Including the Usangu catchment. (g) Production data, financial and economic prices, and the amount of saved water for different users are based on the measurements (or estimates) provided by the implementing agencies. For non-traded commodities, a standard conversion factor of 0.9 is used to convert financial prices of into economic values. This conversion factor is also used in the case of data scarcity; (h) Other key data sources used for this analysis include Government of Tanzania Implementation Completion Report, project documents, various consultant reports and studies, and separate estimates prepared by the implementing Ministries. Price estimates for water in Table 4 are taken from secondary sources (including NORPLAN, 2000, Developing Water Economics: Technical Report Economics and Financing, consultant report to the Ministry of Water and Livestock Development, by A. Kristiansen; IUCN, 2003, A Preliminary Economic Assessment of Water Resources of the Pangani River Basin, Tanzania: Economic Value, Incentives for Sustainable Use and Mechanisms for Financing Management, Eastern Africa Regional & Pangani Basin Water Office, by J. K. Turpie, Y. M. Ngaga and F. K. Karanja; - 28 - and the Ministry of Water and Livestock Development `Water Users Fees' database). In general, the mean value of water in different uses was proxied by dividing the net return of products in Tsh. per unit by estimates of water requirements in cubic meters per unit of production. The price estimates for major project inputs and outputs, displayed in Table 5, are from the SII component Project Coordination Unit. Table 5. Financial and Economic Prices Item Unit Financial Economic Outputs Rice Tsh/kg 380 204 Maize Tsh/kg 138 119 Tomatoes Tsh/kg 240 240 Onions Tsh/kg 230 230 Seeds Paddy Tsh/kg 120 114 Maize Tsh/kg 1500 1425 Onions Tsh/kg 16,000 15,200 Tomatoes Tsh/kg 67,000 63,650 Fertilizers (average) Tsh/kg 273 260 Labor Labor (simple) Tsh/man-day 1000 1000 Labor (technical) Tsh/man-day 2500 2500 Machinery costs (ox, tractor rental) Tsh/ha 17,667 14,133 - 29 - Annex 4. Bank Inputs (a) Missions: Stage of Project Cycle No. of Persons and Specialty Performance Rating (e.g. 2 Economists, 1 FMS, etc.) Implementation Development Month/Year Count Specialty Progress Objective Identification/Preparation 09/19/1996 3 TTL (1); AG. OPERATIONS (1); OPER. SPEC. (1) Supervision 05/05/1997 3 WATER RESOURCES S S SPEC. (WRS) (1); AG. SERV (1); OPER. OFF. (1) 03/31/1998 4 TTL (1); FIN. ANALYST (1); S S WRS (1); AG. OPER. OFF (1) 11/03/1998 3 TTL (1); SR. FIN. ANALYST S S (1); AG. SERVICES (1) 05/11/1999 5 TTL (1); FIN. ANA. (1); AG. S S SERVICES (1); PROC. (2) 02/03/2000 4 AGRICUL. (1); WRS (1); FIN. S S ANA (1); IRRI. SPEC (1) 03/09/2001 7 TTL (1); WRS (1); ENV (1); S S HYDROLOGIST (1); FIN. MNMT (1); AG. RES. (1); AG. SERV (1) 10/12/2001 5 TTL (1); WRS (3); AG. SERV S S (1) 06/07/2002 5 TTL (1); WRS (2); AG. SERV S S (1); ENV (1) 02/07/2003 4 TTL (1); WRS (1); AGRICUL. S S (1); ENV. (1) 10/11/2003 3 TTL (1); WRS (1); FIN. (1) S S 04/01/2004 5 TTL (1); WRS (2); RURAL S S DEV. (1); FIN. MNGT (1) ICR 07/01/2004 5 IRRI/TTL (1); IRRI (1); S S ENV (1); ECON (1); OPER OFF (1) (b) Staff: Stage of Project Cycle Actual/Latest Estimate No. Staff weeks US$ ('000) Identification/Preparation 108.3 426.8 Supervision 169.2 913.9 ICR 17.5 57.4 Total 295.0 1398.1 - 30 - Annex 5. Ratings for Achievement of Objectives/Outputs of Components (H=High, SU=Substantial, M=Modest, N=Negligible, NA=Not Applicable) Rating Macro policies H SU M N NA Sector Policies H SU M N NA Physical H SU M N NA Financial H SU M N NA Institutional Development H SU M N NA Environmental H SU M N NA Social Poverty Reduction H SU M N NA Gender H SU M N NA Other (Please specify) H SU M N NA Participation and Civic Engagement Private sector development H SU M N NA Public sector management H SU M N NA Other (Please specify) H SU M N NA - 31 - Annex 6. Ratings of Bank and Borrower Performance (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HU=Highly Unsatisfactory) 6.1 Bank performance Rating Lending HS S U HU Supervision HS S U HU Overall HS S U HU 6.2 Borrower performance Rating Preparation HS S U HU Government implementation performance HS S U HU Implementation agency performance HS S U HU Overall HS S U HU - 32 - Annex 7. List of Supporting Documents Staff Appraisal Report Tanzania River Basin Management and Smallholder Irrigation Improvement Project, Report No. 15122-TA, June 5, 1996, Agriculture and Environment Operations, Eastern Africa Department, Africa Region. Aide Memoire/Supervision Reports Reviewing Completion of Activities under PPFs and Project Startup Status (September 1996) Aide Memoire Supervision Mission (May 1997) Aide Memoire Supervision Mission (March 1998) Aide Memoire Supervision Mission (October/November 1998) Aide Memoire Supervision Mission (May 1999) Aide Memoire Supervision Mission (February/March 2000) Aide Memoire Supervision Mission (February/March 2001) - Mid-Term Review Supervision Mission (September/October 2001) Aide Memoire Supervision Mission (May/June 2002) Draft Aide Memoire Supervision Mission (January/February 2003) Aide Memoire Supervision Mission (October 2003) Aide Memoire Supervision Mission (April 2004) Aide Memoire Supervision/ICR Mission (June/July 2004) Aide Memoire Studies 1. RBMSIIP, SII Component, Implementation Completion Process, Environmental Resources Consultancy (ERC), June 2004. 2. Sustainable Management of the Usangu Wetland and its Catchment (SMUWC), Final Report, March 2001 and all studies on www.usangu.org website. 3. Pangani Basin, River Basin Management Benefits study. 4. A Preliminary Economic Assessment of Water Resources of the Pangani River Basin, Tanzania: Economic Value, Incentives for Sustainable Use and Mechanisms for Financing Management, Eastern Africa Regional & Pangani Basin Water Office, by J. K. Turpie, Y. M. Ngaga and F. K. Karanja, November 2003. 5. RBMSIIP Initial Environmental Examination Scheme reports, Environmental Cell Unit, 2000-04. 6. Rufiji Water Basin Office Status Report, June 2004. 7. RBMSIIP, SII Component, Assessment of Irrigation Efficiency in Traditional Smallholder Schemes in Pangani and Rufiji Basins, Tanzania, December 2002. 8. Government of Tanzania Implementation Completion Report, June 2004. 9. RBMSIIP, RBM Component, Project Implementation Report, December 1996 - March 15, 2004, (presented to the 14th Rufiji and Pangani Basin Technical Committee Meetings), Mr. Washington Mutayoba, May 2004. 9. NORPLAN, 2000, Developing Water Economics: Technical Report Economics and Financing, consultant report to the Ministry of Water and Livestock Development, by A. Kristiansen; IUCN, 2003. 10. Ministry of Water, RBM Component, Water Resources Economics Report I, Inter-Consult Ltd. of Tanzania and NORPLAN A.S of Norway, April 1996. 11. World Bank, "A Review of World Bank Experience in Irrigation", Report 13676, November 1994. United Republic of Tanzania/Ministry of Water, "Rapid Water Resources Asessment, Volumes I and II, DANIDA/World Bank, January 1995. - 33 - Additional Annex 8. Smallholder Irrigation Improvement The SII component focused on activities related to training of irrigators in the areas of crop production, scheme and water management, leadership and financial management, assisting farmers in formation/upgrading and registration of Irrigators Associations to be legal entities, acquisition of Water Rights, and infrastructure improvement. Table 1: Final 15 Schemes selected for Rehabilitation No. Scheme Name District Irrigated Area (ha) Irrigated Crops Pangani Basin 2904 1. Mombo Korogwe 220 Paddy 2. Lekitatu Arumeru 464 Paddy, Maize, Beans, Tomatoes, Water melon 3. Soko Moshi (R) 370 Paddy, Maize, vegetables 4. Mahenge Korogwe 300 Paddy, Maize 5. Kambi ya Tanga Arumeru 410 Maize, Cassava, Tomatoes, Beans 6. Lemkuna Simanjiro 430 Paddy, Maize, Water melon 7. Kivulini Mwanga 410 Paddy, Maize, Tomatoes, Beans, Water melon 8. Longoi A Hai 300 Maize, onions, Beans, groundnuts, Tomatoes Rufiji Basin 2155 1. Luganga Iringa (R) 300 Paddy, Maize, Tomatoes, groundnuts, Paprika 2. Ipatagwa Mbarali 542 Paddy, Maize, Beans 3. Igomelo Mbarali 300 Paddy, Maize, Onions, Tomatoes 4. Ruanda Majenje Mbarali 450 Paddy, Maize, Beans, Onions, Tomatoes 5. Mangalali Iringa (R) 104 Maize, Onions, Beans, Tomatoes, Paprika 6. Nyamahana Iringa (R) 109 Paddy, Maize, Tomatoes, Paprika, Sweet potatoes 7. Mapogoro Iringa (R) 350 Paddy, Maize, Chickpeas, Beans, Onions, Tomatoes Operation and Maintenance. IOs are responsible for preparing yearly O&M budgets for irrigation systems, and for collecting fees for both Water Right and O&M costs. Some schemes combine the two fees. All Water Right fees are per farmer and O&M fees are per acre, but on those schemes where the two are combined, the basis is unclear (Table 2). The IOs and all members agree on the amount charged. The two fees have caused confusion on some schemes. Farmers also seem puzzled about depositing funds in their bank accounts for future use and investments. This is reflected in the proportion of farmers who pay fees. Farmers contribute cash and free labor, and they do not understand the need for cash to repair damaged structures. The District Irrigation officers (DIO) need to clarify this, as they are responsible for establishing seasonal O&M programs with the IOs and for identifying technical needs that farmers sometimes overlook. This appears to be a weak link and requires further post-project support. Delays in preparing O&M manuals have also contributed to the problems experienced by IOs and DIOs. Relationship with outsiders and Conflict Resolution. IOs are responsible for solving emerging conflicts within and around the schemes. The impact assessment study noted that water conflicts remain a problem in nearly all schemes, except Longoi. Conflicts have declined in Lekitatu, Lemkuna, Ruanda Majenje Mapogoro, Nyamahana, Mangalali, and Luganga following formalization and training of IOs. Most conflicts are related to seasonal variations in flows and the unwillingness of some farmers to accept water reductions while paying fees for Water Rights. - 34 - Table 2. Details of Water Right and O&M Fees Charged by IOs Basin Scheme Year IO O & M No. of Percent Members Water Formed O&M Fee3 Members Paying Right Fee5 Period Kambi ya Tanga 2001 No1 NA 650 NA Yes 2,000 Per Year Lekitatu 1998 Yes 3000 4 612 24% Yes *5 Per Year Longoi 2001 Yes 5000 240 75% Yes 2,000 Per Season Pangani Soko 1998 Yes 8000 4 349 17% Yes *5 Per Season Kivulini 2001 Yes 3000 142 100% No NA Per Year Lemkuna 1998 No2 NA 89 NA Yes 3,000 Per Season Mombo 1979 Yes 16000 429 47% Yes 500 Per Season Mahenge 1998 Yes 10000 4 300 3% Yes *5 Per Season Igomelo 1998 Yes 6000 4 280 71% Yes *5 Per Season Ruanda Majenje 1997 Yes 1000 180 6% Yes 1,500 Per Year Ipatagwa I 1998 Yes 1000 264 18% Yes 1,000 Per Year Rufiji Ipatagwa II 1998 Yes 1000 264 18% Yes 1,000 Per Season Mapogoro 1997 Yes 3000 4 150 100% Yes *5 Per Year Nyamahana 1997 Yes 1000 215 93% Yes 2,050 Per Season Mangalali 1992 Yes 4000 4 90 100% Yes *5 Per Year Luganga 1997 Yes 5000 4 163 100% Yes *5 Per Season 1Members volunteer labour to cover O&M costs. 4This fee includes the Water Use Fee. 2Lemkuna will begin to collected fees this year. 5Covered by joint O&M and Water Right Fee 3Per acre. Source: Impact Assessment Field Data, June 2004. Document Preparation. When the project started, preparation of design and tender documents and the tendering process was new for many Irrigation Department staff. Training took more time than planned, delaying the procurement of consultancy services and contractors for the first four group A schemes. However, this was considerably reduced as staff gained experience through on-the-job training. Tendering Process. Pre-qualification of contractors was required for Group A schemes. This included annual minimum turnover, experience in similar works, personnel capabilities, equipment capabilities, and firm's financial position. Most local firms lacked experience in constructing similar irrigation projects, and none of the local contractors were able to meet the minimum requirements. The Bank was therefore asked to lower its criteria and it agreed with two local companies being awarded the first four contracts. In retrospect, this was not a good decision, as the contractors were unable to undertake the works to the required standard and did not mobilize sufficient resources, mainly due to their poor financial resources. Their inexperience also surfaced in delays in starting the works, and poor contract and personnel management. For the second Group B contracts, pre-qualification took place under National Competitive Bidding (NCB) and minimum criteria were strictly adhered to. Four internationally-based contractors were awarded the contracts. There were still some problems with mobilization and quality control, but in general, the performance was much better than the first group A contracts, and construction was completed in a relatively shorter period. For the final Group C schemes, the pre-qualification was replaced by post qualification of contractors to reduce the bidding process period. Three local contractors were awarded the works, and completed the works just before the Project ended. One of the major causes of delay throughout the whole project was the time taken by the Tanzanian Central Tender Board (CTB) to process the proposals from both the contractors and consultants. Irrigation Efficiency. One of the main objectives of the project was to raise irrigation efficiency from below an estimated 15 percent to an average of 30 percent after scheme improvement. Data were not - 35 - collected at the start of the project, but MAFS and MWLD staff measured irrigation efficiency on 8 out of 15 schemes before the improvements took effect, and following recommendations made in the MTR in March 2001. These schemes were: Mombo, Soko, Mahenge, Nyamahana, Mangalali, Igomelo, Ipatagwa and Ruanda Majenje. Data collection involved the measurement of canal discharges at different levels of irrigation systems (conveyance and field canals), using current meters. Results indicated that the average irrigation efficiency was 8 percent in the wet season and 11 percent in the dry season. Repeat measurements were made in 2004, following physical improvements to diversion and control structures and conveyance and distribution systems on all the selected schemes, and the training of farmers in irrigation water management. Irrigation efficiency was estimated to have improved from an average of 8 percent to 19 percent in the wet season, and from an average of 11 percent to 27 percent in the dry season. The greatest improvements were achieved in conveyance efficiencies. The results are in line with the SAR targets, although there is still significant scope for improving efficiencies at field level (Table 3). This can be achieved through continued training and follow up by the DIOs and Zonal offices, although it is expected to be a slow process as farmers take time to understand what is involved. Table 3. Results of Irrigation Efficiency Measurements Season Description Baseline Improved Average Range Average Range Average Increase % Conveyance 0.29 - 0.53 0.40 0.70 - 0.89 0.84 110 Wet Canal 0.55 0.55 0.55 - 0.80 0.65 18 Field 0.32 - 0.40 0.34 0.32 - 0.40 0.35 3 Overall 0.05 - 0.09 0.08 0.15 - 0.22 0.19 138% Conveyance 0.45 - 0.73 0.60 0.72 - 0.92 0.85 25 Dry Canal 0.48 - 0.66 0.60 0.62 - 0.87 0.74 23 Field 0.32 - 0.40 0.38 0.32 - 0.50 0.44 16 Overall 0.10 - 0.16 0.11 0.24 - 0.26 0.27 145% Crop Production. Crop productivity in traditional schemes was very low. Baseline data indicated that the average yield of paddy for both basins was around 2.0t/ha. Together with infrastructure improvements, farmers received training in crop production techniques, agro-business, and the introduction of high-yielding varieties of paddy such as IR54 and IR56. This was supported with on-farm agronomic demonstrations that produced excellent results with yields of over 7 t/ha at Mombo and Ipatagwa schemes. Observed yields have been increasing over the recent years and the results from farmer fields have shown a steady improvement. Table 4 shows average paddy yields for all schemes, with the average yield in the Pangani and Rufiji reaching 5.3 t/ha and 4.2 t/ha respectively. Similar data for maize is given in Table 5, and shows that average yields have increased from 1.1 t/ha to 5.6 t/ha for the Pangani, and from 1.1 t/ha to 3.4 t/ha for the Rufiji. (These are based on a minimum of 20 harvest sample cuttings on each scheme with an average sample area of 10m x 10m.) Yield data have been obtained from test cutting examinations on each scheme. Random sampling is made of ten different spots from the same area, and samples are weighed at 14 percent moisture content (using moisture meter) to establish the average actual yield per unit area. Crop production plans are produced annually in Swahili on all 15 improved schemes, and these are used to guide farmers and extension staff on appropriate production programs for each season and crop zone. These have proved very successful, with most farmers following the planting and harvesting dates and other details. Most rehabilitated schemes are producing paddy and maize, but other crops including - 36 - watermelon, tomatoes and pepper (paprika) have been introduced recently. Table 4. Average Paddy Yields for All Schemes Paddy Yield (t/ha) River Basin Irrigation Scheme Baseline March Increment 2004 1. Mombo 2.2 6.0 3.8 2. Lekitatu 2.0 5.6 3.6 3. Soko 2.0 6.0 4.0 Pangani 4. Lemkuna 1.8 6.0 4.2 5. Mahenge 2.0 4.0 2.0 6. Kivulini 2.0 4.0 2.0 Average Yield 2.0 5.3 1. Luganga 1.8 5 3.2 2. Ipatagwa 1.5 3.5 2 Rufiji 3. Igomelo 1.6 5 3.4 4.Ruanda Majenje 1 3.5 2.5 5. Nyamahana 1.8 4 2.2 Average Yield 1/ 1.5 4.1 Table 5. Average Maize Yields for All Schemes Irrigation Scheme Maize Yield (t/ha) Basin Baseline March Increment 2004 1.Mombo 9 2. Soko 1 6 5 Pangani 3. Kambi ya Tanga 1 4 3 4. Longoi A 1 3 2 5. Lemkuna 1.2 6 4.8 Average Yield 1.1 4.9 4.5 1. Luganga 1 3.25 2.25 2. Igomelo 1.5 3 1.5 3. Ipatagwa 1 3 2 Rufiji 4. Nyamahana 0.8 3.5 2.7 5. Ruanda Majenje 1 4 3 Average Yield 1/ 1.1 3.3 2.25 1/ In Tables 4 and 5 above, average yield refers to average weighted yields. Weights are by plot size. Family incomes. Increased crop yields have enabled formers to have enough food for family consumption and sell excess for cash income. Average family incomes reflect the degree of farmer involvement, and have increased from US$425 to US$1,500 in the Pangani basin, and from US$340 to US$1,100 in the Rufiji Basin (at current prices) respectively (Project Completion Surveys, ERC Environmental Resources Consultancy, Dar es Salaam, July 2004). Communities indicated that spending for health and education has increased significantly in the villages, improving overall livelihoods. - 37 - Additional Annex 9. Government of Tanzania Implementation Completion Report 1. BACKGROUND 1.1 INTRODUCTION The Development Credit Agreement (DCA) for the River Basin Management and Smallholder Irrigation Improvement Project (RBMSIIP), which became effective on December 5, 1996, was part of a practical and phased approach, which was adopted by the Government of Tanzania (GOT) to develop and implement its strategy for managing water resources. The project had two sub-components mainly: (i) River Basin Management (RBM) which was implemented by Ministry of Water and Livestock Development (MWLD) through Water Resources Department (WRD), and (ii) Smallholder Irrigation Improvement Component (SII) which was implemented by the Ministry of Agriculture and Food Security (MAFS) through Irrigation Department. The total project cost was estimated at US$ 30.7 million in 1995/96 of which SDR 18.2 million (equivalent to US$ 26.3 million, at an exchange rate of 1SDR=US$ 1.45) was IDA credit, US$ 2.96 million was farmers contribution and US$ 1.44 million was GOT contribution. The RBM sub-component was valued at US$ 10.6 million with contingencies, and SII sub-component was valued at US$ 20.1 million with contingencies. 1.2 PROJECT OBJECTIVES The objectives of the project were: (i) to strengthen the Government's capacity to manage water resources and address water related environmental concerns at national level and in Rufiji and Pangani River Basins, and (ii) improve irrigation efficiency of selected smallholder traditional irrigation schemes in the above mentioned river basins. The project consisted of the following parts to achieve the objectives: A) RBM Component (1) Strengthening national water resources management through improving the legal and regulatory framework and broadening participation in river basin management; (2) Carrying out a review of the policy, institutional framework and legislation for water resources management focusing on water rights, water charges, and protection of water against pollution and broadening participation in the institutional framework; (3) (a) Carrying out a review of the functions and organization of the WRD, in order to structure its activities in light of the increased awareness of water-related issues and increase in responsibility and volume of activities; and (b) strengthening of WRD's capacity to monitor water availability and use, through (i) acquisition of specialized equipment and vehicles; (ii) construction and rehabilitation of office facilities in Dar Es Salaam and Rufiji River Basins; and (iii) provision of training and technical advisory services to improve monitoring capabilities; (4) Strengthening of regulatory functions in the water sector through: (a) the acquisition of equipment and vehicles; (b) provision of training and technical advisory services to the offices of the Principal Water Officer, and the Basin Water Officers in the Pangani and the Rufiji River Basins, and (c) rehabilitation of the offices of the Principal Water Officer in Dar Es Salaam. (5) Carrying out studies on: (a) features of the Pangani and Rufiji River Basins which are of importance to Basin hydrology and environment; (b) issues related to natural resources management in the Great Ruaha sub-basins; (c) regulation of water use, and assessing the environmental impact of large scale farming enterprises engaged in both river basins, and (d) two other basins. (6) Strengthening the national water quality network to enable it to carryout effluent monitoring of ambient conditions and groundwater monitoring through: (a) replacement and upgrading of equipment and facilities in MWLD regional laboratories serving the Pangani and Rufiji Basins, and (b) improving of facilities at MWLD Central Water Laboratory. - 38 - B) SIIP Component (1) Improvement of management capacity and updating of infrastructure for targeted traditional smallholder irrigation schemes in the Pangani and Rufiji River Basins. (2) Strengthening of community organizations, to enable them to manage improved irrigation schemes, through the provision of training and technical advisory services. (3) Establishment of a unit in MOA's Irrigation Department to conduct environmental evaluation of irrigation schemes proposed for support under this and other interventions in irrigation. (4) Strengthening MOA's capacity to carry out successful smallholder irrigation development, through the rationalisation of MOA's Irrigation Department, acquisition of specialised equipment and vehicles and the provision of training and technical advisory services. (5) Strengthening capacity of the private sector to carry out successful smallholder irrigation, through the strengthening of water user groups and training of private sector contractors on technical features of construction of irrigation works and the use of labour based construction methods. (6) Carry out research on issues identified as impediments to the development of the sustainable and profitable cropping systems and management of smallholder irrigation in the Pangani and Rufiji. (7) Increase water use efficiency to 30 percent. 2.0 PERFORMANCES OF THE PROJECT COMPONENTS 2.1 RBM PERFORMANCE ACHIEVEMENTS AND CONSTRAINTS A) Achievements The following were achieved during the implementation process of RBM component: i. Reviewed the policy, legal and institutional framework as envisioned in Staff Appraisal Report (SAR) and adopted National Water Policy, in July 2002. The policy provides the framework for achieving equitable and sustainable availability, utilization and management of water resources for socio-economic development. Policy implementation strategies are under development as part of sector-wide reform process. The effort to disseminate the policy to the broad-based stakeholder is still going on; ii. Improved information base by investment in rehabilitation of water gauging stations, undertaking of specialized studies and establishment of databases for water resources management; iii. Detailed and streamlined institutional framework and organizational structure was prepared. The new framework aims at facilitating the participation of different categories and levels of stakeholder in decision-making. It also included streamlining the organization structure for the DWR, the Office of the PWO, and BWOs; iv. Based on Amendment of Water Act No. 10 of 1981 the Minister responsible for Water declared nine river basins for-water administrative purposes intended to minimise conflicting interests among Government institutions related with water resources management. In addition the Water Act established two executive boards that are Central Water Board and Basin Water Board, the later being with autonomous both financially and administratively as specified by Section 7(2) of the Act 8 of 1997. To-date, five out of nine Basin Water Boards has been established, and the sixth has its Officer appointed. It is anticipated that the remaining Basin Water Boards and Offices will be established in the future; v. Several studies were undertaken by SMUWC project, UDSM, Ms NORPLAN etc on hydrological and environmental features of the Pangani and Rufiji River Basins, natural resources management in the Greater Ruaha, regulation of water use and assessment of environmental impact of larger scale farming. The findings of these studies recommends to promote integration of water resources management through popular participation of broad-based water users or water user's organisations; vi. The Central Water Laboratory and Regional Water Laboratories at Mbeya and Iringa (Rufiji Basin), Tanga and Arusha (Pangani Basin) were assessed, rehabilitated and equipped; vii. Training as one of pillar of the capacity building was undertaken by attending courses on environment; water management; project management, monitoring & evaluation and procurement, Stakeholder Analysis and Water Resources mapping, planning approaches, Training of Trainers, offered in Canada, Denmark, - 39 - USA, Zimbabwe, South Africa and the Netherlands; viii. Rehabilitation of 71 hydrometric stations (28 from Pangani basin and 43 from Rufiji basin) and 20 weather stations (8 from Pangani and 11 from Rufiji basin) and one at MWLD headquarter to be transferred later to a more suitable position was undertaken; ix. Groundwater data and information from the basins were gathered and analyzed. This involved groundwater monitoring network, geophysics and hydrological survey which were carried out in order to identify potential areas and suitable locations for drilling of monitoring boreholes; x. Assessments of water user and water rights indicated that there were 849 abstractions with water rights, 1861 without water rights and 196 applications for water rights in Pangani river basin. While in Rufiji basin there are 1514 abstractions out of which 1050 were with water rights and 464 without water rights; xi. The Pangani basin water user database software were redesigned to accommodate Sigi, Umba and Mzimbazi catchments. While in Rufiji basin water user and water rights database software were re-installed by International Communication Systems Ltd; xii. Awareness creation and sensitizations were carried out in order to increase participation of water users and other stakeholders in the protection and conservation of water resources, and, xiii. More than 64 participatory assessments for water resources at local level in Weruweru, Rau, Rundugai springs, Muwe (Pangani basin), Mkoji and Halali (Rufiji basin) sub-catchments and 51 local water user groups and two pilot sub catchments organizations in Mkoji and Halali were formed. B) Constraints There were a number of constraints mainly as follows: i. Existence of broad-based stakeholders with conflicting interests on water resources management; ii. Limited resources both human and physical at all levels. 2.2 SII Performance Achievements and Constraints A) Achievements The original sub-component of the SIIP as described in the SAR para 2.27-2.36 were (i) Participatory Irrigation Scheme Selection and Management (ii) Engineering Works on Schemes (iii) Environmental Mitigation Measures and (iv) Irrigation Sector Capacity Building. The following were achieved during the implementation: i. A total of 715 (613 in Pangani and 122 in Rufiji river basins) large and small scale schemes/abstractions for irrigation purposes were identified, while 34 high ranking schemes were selected for possible improvement instead of the anticipated 40 schemes in the SAR. Followed by scheme improvement assessment studies, which involved a more detailed study of a scheme than the initial reconnaissance process. Ten (10) District Catchment Facilitation Teams (DCFT) one for each district were formed to replace the five MAFS/MWLD multidisciplinary teams. A total of nineteen (19) Irrigator Organizations (11 as Cooperative Societies and 8 as Irrigators Associations) were registered and 15 are at different stages towards formalization. ii. The target was to improve between 25
Группа Всемирного банка · Implementation Completion and Results Report
Tanzania - River Basin Management and Smallholder Irrigation Improvement Project
Открыть оригинал документа
Полный текст размещён на сайте публикующей организации. lawenc.com индексирует метаданные и ведёт на официальный источник.
Полный текст
Основные сведения
Организация
Группа Всемирного банка
Тип документа
Implementation Completion and Results Report
Страна
Танзания
Источник
Всемирный банк