1.1 Context at Appraisal
According to several archeological and historical evidences, Yemen is the first country in the world where spate irrigation was practiced. This unique system witnessed its zenith during the Shebean period in the first millennium BC. The intense development of trade after the Islamic period may have promoted the spread of spate irrigation from Yemen to other arid and semi-arid regions. By the late 1990s, use of Yemen’s water resources was deemed to be unsustainable, due not only to the overexploitation of the non-renewable groundwater resources but also to the neglect of traditional irrigation systems fed by the (relatively renewable) spate waters. To help address this issue, the Government of Yemen (GOY) through the Ministry of Agriculture and Irrigation (MAI) undertook to prepare an Irrigation Improvement Project (IIP) and to implement it with assistance from the International Development Association (IDA) under an Adaptable Program Lending (APL) credit. The project was planned to be prepared and implemented in two phases, namely (i) Phase I, covering schemes in two wadis (Zabid in Hodeidah governorate and Tuban in Lahej governorate) with a total command area of about 26,000 ha, and (ii) Phase II, covering schemes in five other main wadis (Bana, Hassan, Mawr, Rima’a and Siham) with a total command area of about 64,000 ha, and/or smaller schemes in other wadis (Hijr, Ahwar, Meifa’a, Surdud, Harad and Raysan) that would meet the selection criteria.
IIP Phase I preparation studies were themselves undertaken in two phases. Phase 1 comprised an inventory survey, characterization and preliminary study of spate irrigation schemes in five wadis (Mawr, Zabid, Tuban, Bana and Hassan). Through a ranking analysis, two of these wadi schemes (Tuban and Zabid) were selected for more detailed study in Phase II. The rationale was that the two selected schemes would provide the best opportunity to serve as models or pilots for developing and demonstrating the participatory irrigation management (PIM) approach to decentralization and transfer of irrigation scheme management responsibilities to appropriate local institutions. To execute the preparation studies for the selected Wadi Zabid and Wadi Tuban schemes, a Project Preparation Unit (PPU) was formed within the MAI. National and international consultants were engaged by the PPU to undertake the studies in Yemen. A Japanese grant managed through the World Bank provided the principal funding for project preparation. GOY, the World Bank, IFAD and FAO/IC contributed additional funding and/or resources.
Following on from project preparation, the IIP Phase I Project Appraisal Document (PAD) was finalized and issued on August 9, 2000. Financing of the total project cost of US$25.60 million was planned to be provided by: (i) an IDA credit of SDR 16.2 million (US$21.3 million equivalent, 83.2%); (ii) direct contributions from beneficiaries totaling US$1.2 million (4.7%), and (iii) GOY contributions from its own resources amounting to US$3.10 million (12.1%). The project implementation period was planned to be 5 years.
Government-level institutional arrangements for project implementation, as envisaged at appraisal, included: (i) the MAI as the agency to assume overall authority and responsibility on behalf of the GOY, (ii) a Steering Committee (SC) to oversee implementation, comprising the Minister of the MAI as chairman and five further key persons including the project director; (iii) a central Project Management Unit (PMU) in Sana’a for coordination and general supervision, led by the project director and staffed by technical, procurement, financial, administrative and monitoring and evaluation (M&E) personnel; and (iv) two Project Implementation Units (PIUs), one in Zabid and one in Tuban, for supervising implementation activities at the project sites, each staffed by a director, an engineer, an agriculturalist, an institutional specialist and an accountant. The PIUs were to work closely with the relevant regional agencies, the Tihama Development Authority (TDA) in Wadi Zabid and the Lahej Regional Agricultural Office (RAO) in Wadi Tuban.
The project commencement date was January 1, 2001 and the credit effectiveness date was January 18, 2001. Earlier, in July 2000, the PMU in Sana’a was established with a director and staff including specialists (civil servants or consultants) in the fields of irrigation, agronomy, institutional development, publicity, procurement, financial management and M&E. PIUs in Zabid and Tuban were formed at the same time. The PMU and PIUs remained fully operational during the execution of the project. Also, in July 2000, the SC was established under the chairmanship of the Minister of the MAI and included the Deputy Minister of the MAI, the Deputy Minister of the Ministry of Planning and International Cooperation (MOPIC), the Deputy Minister of the Ministry of Finance (MOF), the Director General of the General Directorate of Irrigation (GDI), the Director General of Planning & Monitoring MAI, and the PMU director member/secretary. Project implementation was eventually carried out over a total period of 8 years with essentially no change in total project cost. The project closure date was December 31, 2008.
1.2 Original Project Development Objectives (PDO) and Key Indicators
The PDOs of the APL Phase I project were to ensure: (i) sustainable and efficient water conveyance, distribution and use in the two spate irrigation schemes of wadis Tuban and Zabid, through rehabilitation and PIM; and (ii) increase in agricultural productivity and rural incomes, through implementation of an intensive agriculture demonstration program (ADP). The key indicators are as follows:
Key Indicators PAD Targets
Indicators for both PDOs (i) and (ii):
Number of established and empowered water user organizations (attributed mainly to Component B (PIM), as explained below).
Incremental rural areas benefiting from the spate rehabilitation / modernization and flood protection works (attributed mainly to Component A but also to Components B, C and D, as explained below in Section 3.2).
443 WUGs (230 Tuban and 213 Zabid); 32 WUAs (16 Tuban and 16 Zabid); 2 ICs (1 Tuban and 1 Zabid).
Improved irrigation and flood control on 26,000 ha (11,000 ha Tuban and 15,000 ha Zabid).
Indicator for PDO (ii) only:
Increased productivity per hectare and farmer income (attributed mainly to Component C (ADP) but also to Components B and D, as explained below in Section 3.2).
Increased productivity and incomes from 5,000 ha of demonstration farms.
1.3 Revised PDO (as approved by original approving authority) and Key Indicators, and Reasons/Justification.
There has been no considerable revision as to the key PDOs and respective indicators. However, some of the indicators were rephrased or simplified per the approved Development Credit Agreement (DCA) amendments that have taken place during the project lifetime. Sections 1.7 and 2.3 below elaborate on these DCA amendments and on the associated revision of the indicators, respectively.
1.4 Main Beneficiaries
The direct project beneficiaries identified in the PAD are the farmers, farm workers and families whose livelihoods will have been improved by the project interventions. These beneficiaries were estimated to total 150,000 people in 27,000 households. Other identified beneficiaries include:
a) The MAI including regional agencies;
b) Water user organizations established and empowered by the project, including canal-level water user groups (WUGs) and water user associations (WUAs) and two wadi-level Irrigation Councils (ICs); and
c) Non-irrigation (domestic) water supply users, mostly in the surrounding rural areas, benefiting from the reduced over-extraction of groundwater resulting from the project.
1.5 Original Components
The project’s original components, all relating to interventions in the Wadi Tuban and Wadi Zabid areas, were:
a) Rehabilitation, Improvement and Protection of Spate Irrigation Infrastructure (Component A), including rehabilitation and improvement works for irrigation systems, flood protection works and roads, together with associated surveys, hydraulic modeling, detail design and construction supervision;
b) Irrigation and Environment Management and Participatory Irrigation Management (Component B), covering: (i) establishment of and support for PIM organizations, (ii) development of spate irrigation management systems including water management information systems (MIS), flood warning systems, spate management model and hydrological monitoring, (iii) provision of operation and maintenance (O&M) equipment, and (iv) environmental mitigation and enhancement measures, including groundwater and hydraulic monitoring, upper watersheds study, and Wadi Tuban soil salinity/sodicity study/mapping;
c) Intensive Agriculture Demonstration Program (Component C), covering large scale demonstrations for agriculture improvements and associated extension services development and support including technical assistance (TA); and
d) Institutional Strengthening and Capacity Building (Component D), covering irrigation sector institutional and legal framework development support, Phase II project preparation, and establishment of and support to the PMU and PIUs.
1.6 Revised Components
None, but there were changes within components as explained in Section 1.7 below.
1.7 Other Significant Changes
A number of amendments to the DCA) were made during the course of the project. Firstly, in August 2003, the DCA was amended to allow for Component A improvements covering:
a) Rehabilitation (including asphalting) of existing service roads for improved canals and access roads connecting the project-served villages and market centers within the project area;
b) Community participation works, through permitting civil works of less than US$10,000 per contract and up to an aggregate amount not exceeding US$1,200,000 to be executed directly through the concerned WUAs; and
c) Shopping for small works, through permitting works of less than US$30,000 per contract and up to an aggregate not exceeding US$500,000 to be procured under lump-sum fixed-price domestic contracts awarded on the basis of shopping for three quotations.
About two years later, in October 2005, the DCA was amended and the closing date was extended from June 30, 2006 to June 30, 2007. Then, in June 2007, the DCA was further amended and the closing date was extended from June 30, 2007 to December 31, 2008. The extensions were to allow for: (i) completion of civil works; and (ii) applying some project design changes aimed at improving project impact, as explained below. The changes were effected through DCA amendments only (rather than through first-order project “restructuring” involving changes to the PDOs or substantial project design changes). These DCA amendments were as follows:
a) October 2005 (as a result of the “Mid Term Review” undertaken in late 2004). To help expedite the then-lagging civil works (Component A), the amendment allowed for increased farmer ownership via administration/execution of community-level spate irrigation works at the community level, through clarification of the 2003 amendment relating to WUA contract works. The clarification defined the arrangements for cost sharing (in cash and in kind). Details are provided in Annex 2. The amendment also simplified the disbursement condition included in the original DCA which specified that civil works could start only after WUA establishment. Scheme headworks and main canals were deemed to be public infrastructure, and the disbursement condition was amended to be applicable only in the case of the (quasi-private) secondary and community-level tertiary canals. This allowed for headworks and main canal works to be started in advance of establishment of WUAs for the lower-order canals; and
b) June 2007 (to make use of project cost savings). It was decided that available unused project funding should be directed towards applying lessons learned and piloting fresh development ideas in a third wadi (Wadi Ahwar in Abyan Governorate). Wadi Ahwar was shown to enjoy more favorable enabling conditions than either Wadi Zabid or Wadi Tuban. Activities introduced into the project included: (i) completion of feasibility study and designs for rehabilitation of the wadi’s spate diversion system and flood protection works (including application of fresh ideas for developing the potential for “conjunctive use” involving also groundwater usage to help to improve the socio-economic viability of spate irrigation), all to be implemented under the recently-approved Water Sector Support Project (WSSP); and (ii) execution of small urgent works for protection of villages from flooding and for rural water supply. Details are provided in Annex 2.
Why/how were the funds allocated to Component B significantly decreased, and those allocated to Component A significantly increased, without compromising the PDOs?
Per the ICR review meeting, the IDA task team was requested to elaborate on why/how the IDA/GOY funds allocated to Component B were significantly decreased, while those allocated to Component A were significantly increased, without affecting the PDOs. The sections below provide elaborations. Throughout the project lifetime (FY2000 to FY09) there have been three main causes which inevitably changed the funds allocated to Components A and B, as follows.
(1) Cost reductions under Component B and cost increases under Component A:
Project cost savings in IDA commitments were estimated at US$3.5 million in FY07/FY08 (comparing cost estimates between FY2000 and FY2007). These savings helped to reallocate funds across project components, particularly from Component B to Component A. The origins of the cost savings were:
a) Savings in goods and equipment, due to GOY/MAI obtaining water gates from a parallel Japan-funded grant rather than from IIP. This reduced the IDA allocation intended for funding the Goods/Equipment under Component A and also helped reduce corresponding IDA funds allocated to Component B(iii) (“provision of O&M equipment”, which would have cost US$3.8 million per the PAD). To effect these savings, IDA disbursement Category 1 on Goods/Equipment was officially revised from SDR 2.4 million (per original DCA) to SDR 0.9 million; and
b) Other technical savings (in consultancy services), particularly due to resorting selectively to hiring national rather than international consultants across IIP components. At appraisal, in hindsight the appraisal teams envisaged a sizable number of international consultants (man/days) particularly under Component B. Gradually, as the project progressed (and particularly after the Social Mobilization Teams achieved progress under Component B), the MAI and IDA teams reworked the consultancy plans by FY04, in order to make use of the on-the-job knowledge/skills accruing to the national consultants and to MAI/PMU/PIUs. For instance, GDI executed the “upper watersheds” study (under Component B(iv)), which would have otherwise cost US$0.3 million (per the PAD) if executed by international consultants. The same applies to the groundwater and hydraulic monitoring activities and the “soil salinity/sodicity” study in Tuban (which per the PAD would have cost US$0.6 million and US$0.2 million respectively), both under Component B(iv).
While the above are deemed positive (efficiency-caused) cost savings, on the negative side the following issues have also resulted in reducing the cost of Component B while increasing the cost of Component A (as explained further in Section 2.2 below):
a) Subcomponent B(ii) on “improving spate water management” (US$0.9 million per the PDA) was not duly implemented (particularly in Wadi Tuban), due to the lack of capacity at NWRA and MAI/GDI needed for the two agencies to cooperate toward executing the activities, and
b) Funds allocated to Category 2 on civil works (for Component A) were increased due to escalated price/physical contingencies, often associated with the delay in contract execution (which attributes to lack of capacity of national contractors as noted below in Section 2.2).
(2) Community contribution to capital and O&M costs:
Also, starting FY05, the community contributions (which thanks to the 2003 and 2005 DCA amendments were made “targeted”, to sustain the feeder subsystem) helped to reduce the respective IDA allocation to Components A and B (subcomponent B(iii) on “provision of O&M equipment”). These contributions amounted to a total of US$0.3 million in FY09. Empowering the communities through these two DCA amendments also helped achieve (procurement/efficiency related) cost savings across project components.
(3) Lately added GOY funds and utilization of the unallocated IDA category:
Corresponding to the aforementioned savings from IDA, in FY07 the GOY additionally allocated US$0.30 million to support Component A (particularly for Wadi Ahwar). Also the project utilized the unallocated category, by reducing it from SDR 1.6 million to SDR 200,000 in FY08 in order to top up Category 2 on civil works (Component A).
The above cost savings/reductions considerably helped in funding the new activities in Wadi Ahwar, per the 2007 DCA amendment as explained above.
The changes to the M&E indicators associated with the above design changes are presented below in Section 2.3.
Justification for not undergoing a first-order project restructuring:
Had the above changes, reallocations and simplifications been made all at once at some point in time, GOY and IDA teams might have considered a first-order project restructuring . Since much of these changes occurred sporadically throughout the project lifetime, the GOY and IDA teams effected them by resorting to amending the DCA and the respective project documents (including ISRs during FY04 to FY09), per the following.
The DCA amendments and category reallocations (August 2003 and October 2005) were effected in consistence with the following guidelines:
“Minor changes are approved by Country Directors (CDs). Such changes may include modifications limited to changes in executing units or denominations, implementation plans and schedule changes, new action plans to bring the project back on track, new reporting requirements (or other adjustments to improve implementation), loan closing dates, implementation dates that under the terms of the legal agreement may be put into effect by notice, or a reallocation of loan proceeds that do not affect the project’s design, scope, or expected outcome. Minor changes are proposed, approved and documented through the routine ISR updating process and can be counted towards pro-activity for purposes of project “upgrading” in the Region’s judgment, but not for labeling as restructuring. These cases may also require legal amendments”.
However, the DCA amendment made in June FY07 was effected through the Regional Vice-President’s (RVP’s) approval, which is consistent with the following recent guidelines (2006):
“Second-order restructurings are those where the project design or implementation arrangements are significantly modified (e.g., including reallocations of inputs and/or changes in outputs) but there is no change in PDOs or associated outcome targets. Under the new procedures, second-order restructurings are approved by Regional vice presidents (RVPs)”.
2. Key Factors Affecting Implementation and Outcomes
2.1 Project Preparation, Design and Quality at Entry
a) Conformity with country and sector policies and strategies. The project formulation conformed well to the Bank’s Country Assistance Strategy (CAS), primarily through its strong provisions for: (i) improved water management and irrigation infrastructure sustainability; and (ii) enhanced agricultural productivity for poverty reduction. It also addressed well key government irrigation sector priorities relating to sustainability improvements for spate irrigation systems and corresponding institutional decentralization and user participation.
b) Relevance and appropriateness of project objectives, components and design. The project’s two PDOs, presented at Section 1.2 above, reflect well and succinctly the priority development areas indicated above. The four project components likewise explicitly address the priorities of physical and operational water management improvements and sustainability coupled with increased agricultural production for enhanced rural community incomes. With regard to the two wadi spate irrigation schemes selected for project interventions, there was at the time of project preparation a full awareness of highly significant and heavily entrenched socio-economic inequities prevailing in the Wadi Zabid area. It was also clear from the various interactions with system users and prospective beneficiaries at that time that the potential for achieving a successful project outcome from the point of view of sustainable and equitable water management arrangements in that wadi area could be considered rather limited. It might have been better to select a less problematic pilot scheme for the Phase I project. On the other hand, there was a strong desire to proceed with interventions in this otherwise high priority scheme and to thereby ameliorate to the extent possible the unfavorable situation of the disadvantaged downstream scheme users.
c) Soundness of institutional and implementation arrangements. The formulated institutional arrangements for project implementation, covering a central PMU and two wadi-based PIUs overseen by a SC chaired by the Minister of MAI, as described at Section 1.1 above, would seem to have been both simple and effective. Also, seemingly most appropriate were the principles formulated for establishment of water user organizations, particularly WUAs at secondary canal level, and of joint system management entities, namely the ICs at wadi level, all aimed at helping to secure the long-term sustainability of the improved schemes. The requirement that WUAs be established prior to disbursements for corresponding civil works, which was set as a credit condition, became an important issue during implementation. The condition was aimed at ensuring that there would be participation, consensus and commitment on the part of the users relating to planned scheme improvements and subsequent systems O&M before the making of heavy physical infrastructure investments, and it arose largely as a consequence of the perceived problematic inequitable socio-economic situation in the Wadi Zabid area. An intensive user mobilization and WUA formation program was formulated and planned for execution during the first year of the project, in parallel with participatory planning and design of scheme improvements, and a substantial budget provision was made to cover this and subsequent PIM developments. In the event it appears that the overall program and process was not given the expected prioritization and importance, and that the corresponding budget provision was reallocated largely to physical works items. This comment extends also to most of the various other formulated non-works interventions (e.g. for water resource monitoring and management improvements, O&M equipment provisions, environment-oriented operations, and institutional strengthening). A notable exception seems to have been the ADP, which is reported to have been fully implemented and to have led to significant positive agricultural productivity impacts. It may be that the generally poorer-than-expected implementation performance on institutional development aspects, and the delays in disbursements for civil works, resulted largely from implementation period constraints, as outlined in Section 2.2 below. On the other hand, some may consider that, during project preparation, a greater degree of subtlety, flexibility and/or pragmatism in this area could perhaps have been applied, as it seemingly was later on, following the mid-term review, when the WUAs establishment condition for disbursements for civil works was modified to allow works for headworks and main canals to proceed in advance.
d) Application of previous lessons learned. The comprehensive nature and relatively large scope of the designed PIM and institutional development components of the project resulted from previous lessons learned about the need for beneficiary participation and joint management in the development and usage of irrigation systems in order to achieve sustainability over the long term. It remains to be seen whether the reduced importance given to this in terms of priority and scope of interventions during implementation will still lead to the desired long-term sustainability of the physical infrastructure investments and of the related enhanced agricultural productivity.
e) Identification and mitigation of risks. Risks relating to potential farmer and government agency shortfalls in terms of capacities and contributions were identified and addressed as part of project formulation. At the more fundamental level of underlying government commitment to irrigation sector institutional reform and support of corresponding improvement measures, it was established that these were indeed generally in place. What was perhaps insufficiently addressed was the risk that such commitment and support would not continue into implementation at the same high level and would not translate into timely and effective program implementation actions.
f) Choice of lending instrument: After considering the water management “Log-Frame”, the preparation/appraisal teams considered the APL as the most relevant instrument because IIP’s phase 1 (APL1) needed to test fresh ideas which were not tested before in Yemen’s coastal wades. These included concepts such as the PIM concept (Component B) and the “more-crop-per-water-drop” concept (Component C). Proven successful, APL2 would then replicate these ideas on other wades which use spate water.
2.2 Implementation
From its commencement in 2000, project implementation was constrained by design-related difficulties (particularly the pre-requisite condition of establishing the WUAs prior to design and execution of the civil works) and/or by other major externalities. These either slowed down implementation progress or made it difficult for the impacts of conventional irrigation system improvements to “trickle down” to a large number of end use beneficiaries. These difficulties and externalities included:
a) Abnormal unpredictable droughts (particularly in Wadi Tuban, which was deemed to have suffered a 1-in-50 year drought several times in the last decade);
b) Frequent social unrest in the south (Wadi Ahwar);
c) Sluggish national contractors (particularly those who had other larger and more lucrative contracts with the GOY and hence lacked incentives to expedite their IIP contracts);
d) Chronically inequitable upland-lowland water rights issues in the west (Wadi Zabid, associated largely with the regressive social and land distribution situation prevailing there);
e) The inherent difficulties of spate water management, stemming from the inability to readily store spate water, due to the abrupt and erratic nature of flash floods, and to therefore prevent large losses of water to the sea or to desert fringes, leading to the situation whereby incremental agricultural benefits cannot often offset the large investments needed to improve the physical systems; and
f) The insufficient capacity at NWRA (the water regulator) and MAI/GDI (the water user), particularly affected the implementation of Subcomponent B(ii) on “Improving Spate Water Management”. This issue has been insurmountable in the medium term, but has been addressed per the design of the recently-approved WSSP (2009-2014).
In response to the challenges presented by these constraints, and to try to raise the return on project investments, efforts to re-focus and diversify the IIP components began in 2003. A heavy focus on the irrigation subsector and on costly conventional rehabilitation of spate irrigation infrastructure was modified to develop and follow a more comprehensive cross-sectoral approach (mostly through the 3 DCA amendments explained at Section 1.7 above). Decisions and actions taken on this covered:
a) Rehabilitation and asphalting of key service roads for canals as well as farm-to-market roads, to raise the return on investments and to improve the well-being of the rural inhabitants;
b) Improvement of community engagement through the indicated DCA amendments;
c) Rehabilitation of flash flood and environment protection works;
d) Revisiting the ADP component with a view to increasing crop yield per cubic meter of water (rather than only increasing yield per hectare);
e) Empowering the ICs and WUAs to apply the GOY-enacted Water Law;
f) Preparation (through an international QCBS bid) of a master plan for surface-/groundwater conjunctive use in a third “more promising” wadi (Wadi Ahwar), aiming at low-cost harvesting of the spate water and increasing the inter-seasonal subsurface storage volumes (useful for non-irrigation uses and for counteracting sea water intrusion). The master plan is to be executed under the recently-approved WSSP; and
g) Addressing the capacity and inter-agency cooperation issues between NWRA and MAI as part of the design of the recently-approved WSSP.
Consequently, by the mid of FY2008, IDA’s Implementation Status and Results (ISR) Implementation Performance (IP) and PDO ratings had been raised from “MS” to “S” because there had been major progress in all of the aforementioned undertakings and components and in meeting most of the target outcomes or DCA milestones. The only exception to the latter used to correspond to the spate irrigation infrastructure component (Component A). The mixed performance on this component (due to the above-listed difficulties and externalities including tardiness of national contractors) had been the main reason for “MU” and “MS” ratings on previous ISRs (from late FY06 to late FY07). However, even with this component there was good progress gradually achieved during FY2007 and FY2008 because 
i) 70% of the lands targeted in the PAD were irrigated (due largely to high spate floods in the 2007 season, especially in Wadi Zabid); (ii) most of the main civil works were complete, and (iii) ICs and WUAs helped to address the water rights issue, particularly in Wadi Tuban. Further details are given in Annex 2.
2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization
In the first 5 years of IIP implementation, the PMU employed an M&E officer who assisted the MAI and the IDA ISR teams with M&E. In the last 3 years however, the PIUs in Zabid and Tuban became the main direct sources of M&E data flow to the PMU, while the PMU’s M&E officer gradually phased out due to health reasons.
The M&E activities were supposed to be supported by MIS and Geographic Information System (GIS) applications. These, together with flood warning systems, spate management models and hydrological monitoring systems, made up the spate irrigation management improvement package provided for under the project’s Component B, and were established and tested. However, the project was unable to make these high-technology systems function as intended. The needed hydrological data could not be obtained, and the generated results could not be applied, because of a lack of cooperation between the MAI/IIP and the Ministry of Water and Environment (MWE)’s National Water Resources Authority (NWRA), particularly in Wadi Tuban. The NWRA, as the water resources regulation agency, is mandated to monitor and assess water resources availability. Measures to resolve this information flow and interagency cooperation issue have been formulated as part of the design of the recently-approved WSSP.
In Wadi Zabid, the IIP executed a civil-works contract for drilling four groundwater monitoring wells. The data from these wells, together with further needed water resources information obtained by the Wadi Zabid PIU with help from the TDA, were useful for monitoring of groundwater drawdowns over the life of the project. The groundwater drawdown is an important indicator for the project’s Component B Environmental Management Plan (EMP), and when reduced it would indirectly reflect IIP success in improving the management and use of surface water. The good cooperation with the TDA in Wadi Zabid was in contrast to that with the NWRA branch in Aden which is mandated to address Wadi Tuban, and lacked the capacity and the incentive to cooperate with the Wadi Tuban PIU in the provision and collection of water information. It is noted that the rainwater and flood flow gauges in Wadi Tuban, for which NWRA was responsible, were sabotaged several times during the project’s lifetime.
Also under the project’s Component B EMP, a water resources assessment was undertaken as part of an upper watershed study executed by the GDI with TA from IDA and FAO/CP. The results of the assessment were helpful in the M&E exercises undertaken by the PMU and IDA, particularly for Wadi Zabid.
To help secure sustainability of IIP interventions, the project provided assistance to the two ICs to closely monitor the performance of the established WUAs. Three broad performance indicators were established, namely (i) institutional, (ii) financial, and (iii) technical. Details of the adopted community-based M&E approach are given in Annex 2.
Commensurate with the design changes presented above in Section 1.7, MAI and IDA teams started by 2005 to revisit and simplify the M&E results framework as per the PAD. Per the PAD Annex 1 (dating back to 2000):
a) The APL and PDO indicators became no longer consistent with the APL milestones per the (amended) legal DCA; and
b) Some of the key indicators stated in Annex 1 (main-system spate schemes completed, cost sharing completed) became no longer directly attributed with the indicators listed in the PAD main text (on page 4: increased water availability, adequate financing of O&M of spate schemes, decentralization of government service).
The following Table 1 presents the revised and simplified M&E indicators.
Table 1: Revisited/Simplified Indicators
(comparing the PAD with the ISRs of FY05 to FY09)
APL and PDO indicators per the PAD of 2000 (Annex 1 Results Framework)
Revisions and simplifications associated with the design changes, per the DCA amendments. Reflected in MAI progress reports and in IDA ISRs starting FY05.
APL purpose 1: Sustainable spate water management:
APL purpose 1 indicators:
• GOY budget contribution reduced to zero
• Farmer organizations manage the schemes
PDO leading to APL purpose 1:
• Effective spate water control and conveyance structures, and
• Effective PIM
Indicators reflecting the PDO:
• Head-works and main canals of spate schemes rehabilitated, and
• WUAs and ICs established and agreed cost sharing paid
APL purpose 2: Improve agricultural productivity:
APL purpose 2 indicator:
Increased crop yield and cropped area
PDO leading to APL purpose 2:
Improved rural incomes
Indicator reflecting the PDO: increased agricultural productivity of spate schemes
PAD indicators no longer precise, as per the DCA amendment in 2005 GOY would still shoulder the O&M for the main system, while WUAs gradually takeover the subsystem. Hence, ISR PDO indicator was simplified to reflect the corresponding DCA milestones, thus: “Spate water management improvements by at least 80%” (inferred by the PMU reports on the % increase in on-farm water availability across the wadi).
Indicators no longer precise, as the 2003 DCA amendment added the rehabilitation of roads and community/subsystem works, and after the 2005 DCA amendment re cost sharing. Hence the ISR intermediate indicator was simplified accordingly. Targets were set at:
• A range 80% to 100% completion of all project-introduced infrastructure as consistent with the DCA, and
• 100% formation and empowerment of the agreed number of WUAs and ICs (32 WUAs and 2 ICs 2).
PAD indicators not precise due to unclear cause-and-effect attribution (increased productivity helps to increase rural incomes). Thus, the ISR PDO indicator was simplified to reflect the number of hectares improved due to ADP (for which yield has increased by at least 30%), as compared to the PAD target of 5,000 hectares.
2.4 Safeguard and Fiduciary Compliance
Safeguard policies. The only safeguard policy triggered was Environmental Assessment (OP 4.01, BP 4.01, GP 4.01). The Environmental Category was: B, Partial Environmental Assessment and EMP Requirements. At appraisal there were no major environmental issues arising from the partial EA completed for the IIP. The project was to have beneficial impacts on the environment, since it would increase the reliability of spate irrigation water and reduce overdraft of groundwater aquifers. However, due to changes in the pattern of water distribution as a result of rehabilitation of spate irrigation infrastructure, it was thought that reduction of recharge to certain areas could occur. On the other hand, improved water distribution was expected to result in a reduction of groundwater use in those areas. These benefits were deemed to outweigh any minor adverse environmental impacts (soil quality, waterlogging, water quality) resulting from agricultural activities in the improved areas.
Results from implementation of the EMP: The project has been satisfactory in the area of EMP execution. With regard to “hardware”, all of the village/wadi protection (or so called “environmental protection”) civil works have been completed. On the “software” side, progress has been satisfactory on the main items addressed by the EMP: (i) groundwater monitoring; (ii) environmental awareness and participatory management; and (iii) building the knowledge base on soil salinity/sodicity in Wadi Tuban. However, the cooperation between NWRA-Aden and Lahej RAO, the Wadi Tuban scheme operating agency, needs to be improved to allow for easy and timely provision of and access to the NWRA-Aden spate data, especially from the wadi’s Dukaim station. The upper watershed management study activity progressed well and included a remarkably good contribution from GDI. The study and mapping of soil salinity and sodicity in Wadi Tuban, including formulation of mitigation alternatives, was finalized, and study results were presented at a participatory workshop. The major finding of this study was that irrational groundwater use should be phased out and should be replaced by spate water usage and conjunctive use, since otherwise the over-drafting of groundwater beyond the safe yield level will result in further intrusion of seawater and hence further salinization of the wadi’s fertile soil. The Tuban IC has attempted to respond to this finding through prioritizing spate irrigation diversions to the saline land areas, in order to ensure the needed flushing of salts. Details on environmental and social monitoring and related capacity building issues are given in Annex 2.
Procurement. Per the last two ISR missions and the ICR mission the procurement rating has been “Satisfactory”.
Procurement performance: The procurement performance has been rated satisfactory in the last four years. The PMU has complied with the DCA. The post-review reports reflected the capacity of the PMU/PIUs to undertake procurement per WB guidelines. Also, Yemen IPR (Independent Procurement Review) reflected a satisfactory performance. The lessons learned are documented in Section 6 below.
Financial management (FM) and disbursement. Per the last two ISR missions and the ICR mission the FM rating has been “Moderately Satisfactory”.
FM and disbursement performance: The FM performance was initially rated satisfactory until December 2006, when the rating was downgraded to moderately unsatisfactory. Afterwards, the PMU complied with the DCA, particularly the financial covenant related to submission of annual audit reports and audited quarterly interim financial reports. The audit reports were unqualified. The Audit Report Compliance System reported no accountability issues from all the audit reports. The management letters included some recommendations to enhance the FM and accounting of inventory, asset register, advances, etc., which the PMU satisfactorily applied to its practices. The lessons learned are documented in Section 6.
2.5 Post-Completion Operation/Next Phase
As mentioned above, interventions in a third wadi, Wadi Ahwar in Abyan Governorate (7,000 ha), were added to the IIP during the later period of the project to serve as an advance introduction to the recently-approved WSSP (2009-2014). The IIP closing date extension from June 2007 to December 2008 was perceived by GOY and IDA to be a “bridging phase” between IIP and WSSP, rather than a fully-fledged APLII. The IIP’s APL Phase II will defacto be incorporated into the WSSP. During the bridging phase, uncommitted IDA funds resulting from Phase I cost savings were used to pilot fresh ideas relating to cross-sectoral conjunctive use of surface- and groundwater in the Yemeni coastal plains. Based on an extensive comparative study of 5 candidate wadis , Wadi Ahwar was selected for use of these funds for urgent flood protection, canal cleaning and village water supply works, and for execution of a major feasibility study for wadi-wide interventions to be implemented under WSSP.
