Comprehensive report and evaluation : cereals production II (685-0235), project extension (March 1985-December 1987), USAID/Senegal

Results were mixed. AF targets for tree planting, extension, and training met with good success. Woodlots were started in many project villages; survival rates ranged from 60% in better protected woodlots to 34% in double line windbreaks. Natural regeneration of Kad (Acacia albida) had an 89% survival rate. AF training events for extensionists were held both in-country and through short-term study tours in the United States. Project personnel also benefited from stateside training in information management and farming systems research. Extension activities in project villages (e.g., woodlots, orchards, windbreaks, windrows, individual plantings, live fencing) were successful as well, although a complicating factor was the lack of water and materials for compost pit construction and well digging. On the other hand, soil and water conservation activities did not take place as planned, due to problems with local contractors, and there was only marginal success in developing AF and SC techniques. Funds programmed for adaptive research were used instead for technical studies carried out by Senegalese firms. Collaboration among the six different agencies involved in the project was complicated by budget restrictions, and research-extension cooperation did not occur as designed. In fact, project management proved to be the most challenging aspect of this pilot effort, due to the project's small size and diversity, the number of participating agencies, and USAID/S's direct procurement approach. Some 15 lessons are gleaned from this project; they stress the need for streamlined, field-based management procedures, the importance of considering village dynamics when designing AF projects of this type, and the need to develop more sophisticated improved maintenance procedures for tree plantings. Other specific lessons are that (1) simply planting trees does not adequately demonstrate their value; (2) tree protection is a good indicator of village interest in different plantings and species; (3) barbed wire fencing is probably easier and cheaper to use than local fencing materials; (4) SC activities require accurate data, so preliminary studies should be performed before these activities are designed; and (5) AF is not a sufficient response to the problems of soil degradation.