摘要
Sorghum (Sorghum bicolor (L.) Moench) is an important staple food in semi-arid tropics which contribute to food security and poverty alleviation in Benin. However, its production is seriously facing enormous abiotic and biotic stresses including climate variability. To document its cultivar diversity, the impact of climate change on its production and diversity, the adaptation strategies developed by farmers and the performance of landraces, 22 villages were randomly selected and surveyed in the Department of Atakora in the northwest of Benin using participatory research appraisals (fields and granaries visit, individual and group survey using questionnaires, Four cells or distribution and extent analysis, etc.). Data were analysed through descriptive statistics (frequencies, percentages, means, etc.) to generate summaries and tables at different (villages, individuals) levels and also through multivariate analysis (cluster analysis) and the results figures. In total 8 constraints affecting sorghum production were recorded among which striga proliferation, soil poverty and climate change effects (drought; excess of rain; delay, insufficient and irregular rainfall, etc.) were the most important. Subject to synonyms, 89 sorghum landraces were inventoried and their number varied from 4 to 17 (7 on average) per village. The Shannon–Weaver diversity index (H) estimated at 3.02 indicated high sorghum diversity in the study area. Many landraces were threatened. The relatively high rates (16.7–88.2 %; 40.9 % on average) of cultivars loss recorded per village are proofs of menace of genetic erosion and necessity of developing conservation strategies. Inadaptability of cultivars to climate variability (39.6 % of abandoned landraces) was among others the most important reason justifying the loss of diversity. Crop rotations, growing of early maturing and/or drought tolerant landraces, establishment of fields in lowlands are the most important strategies developed by farmers to mitigate impacts of climate change (low productivity, increase of damages from storage insects, early drying of plant leaves etc.) on the crop. Participatory evaluation of the existing landraces led to the identification of some high-performing ones that are resistant/tolerant to diverse abiotic and biotic stresses. Farmers’ cultivars preference criteria were identified and prioritized. Results of this study are useful to policy makers, agricultural extension services of the different districts, genetic resources specialists and breeders in order to improve sorghum production in Benin.