Sahel water data network and management II (AGRHYMET, project no. 625-0940) : mid-term evaluation

External evaluation covers 1982-9/85 and is based on document review, site visits, and interviews with project, counterpart, and donor personnel. Both the project and the overall AGRHYMET program have produced several positive achievements, but neither has progressed as quickly as expected. Logistic and technical problems - attributable to the contractor, the U.S. National Oceanic and Atmospheric Administration (NOAA), which has in particular failed to provide sufficient technical staff - have slowed procurement of computer and telecommunications (TC) equipment; 8 of 10 planned computer systems are operational at the AGRHYMET Regional Center (ARC) and at three National AGRHYMET Centers (NAC's), while TC equipment is in place at 6 NAC's. TC for both data collection and transmission is inadequate, and the computer system neither adapted to nor large enough for the region. Lack of a method to collect crop and pasture data has been another key constraint. Nonetheless, decadal bulletins of fair to good quality are being produced by NAC's, as are both decadal and monthly bulletins by the ARC. Two key achievements of this Phase II project have been the creation of multidisciplinary working groups and implementation of a pilot project ("Haute Vallee") in Mali in which use of AGRHYMET data has increased millet, sorghum, and arachide yields by up to 45%; similar pilots are being launched elsewhere in the Sahel by other donors. ARC training (300 graduates) has quickly developed a trained cadre for NAC's, but U.S. computer training has been insufficiently practical and also troubled by lengthy delays, high costs, and attrition (7 participants). Only 9 of the 31 participants have thus far returned to NAC posts (where, as a result, computer maintenance has been inadequate). Key NAC and ARC positions are still held by expatriates, which has not only put a strain on donor funding (virtually frozen for the last 4 years), but has added to administrative difficulties at the ARC, and most importantly, hindered attempts to institutionalize the program. Three prior, other-donor evaluation teams lacked technical skills and failed to note NOAA's inadequate performance. USAID technical oversight has also been remiss, monitoring personnel have been inexperienced and/or overworked, and recordkeeping and reporting poor.