Climate change and variability impact on agriculture in the West African Semi-Arid Tropics (WASAT). At present WASAT farmers are most concerned by climate variability which shows the greatest consequences towards the northern drier end of the region. Relative variability, number of existential droughts, species loss and variety turnover are highest there. This paper presents experiences made and approaches developed in the framework of a Research for Development (R4D) project aiming at preparing WASAT farmers to deal with climate constraints. It is argued that agro-phytodiversity management is a reasonable approach to deal with climate variability but that it needs better social organisation to conserve a sufficient number of crops and varieties. Optimised participation and empowerment of farmers in the R4D continuum leads to faster progress with respect to innovation testing, adaptation and sustainable adoption.Keywords : agro-phytodiversity, farmer exchange visit, participatory research, scientific approach, varietal diversity. R esum eAdaptation a la variabilit e et au changement climatique dans la recherche pour le d eveloppement visant l'Afrique de l'Ouest : la n ecessit e d'un changement de paradigmesLa variabilit e et le changement climatique ont un impact sur l'agriculture dans la r egion tropicale semi-aride de l'Afrique de l'Ouest (WASAT).A l'heure actuelle les agriculteurs de cette r egion sont les plus concern es par la variabilit e du climat qui a des cons equences importantes a l'extr emit e nord, plus s eche, de cette zone. La variabilit e relative, le nombre de s echeresses, la disparition des esp eces et le changement des vari et es sont courants. Cet article pr esente les exp eriences et les approches men ees dans le cadre d'un projet de recherche pour le d eveloppement (R4D) visant a pr eparer les agriculteurs de la r egion a mieux faire face aux contraintes climatiques. Il est g en eralement admis que la gestion de l'agrophytodiversit e est une approche raisonnable pour faire face a la variabilit e du climat, mais celle-ci a besoin d'une bonne organisation sociale pour mieux conserver un nombre suffisant de cultures et de vari et es. La participation optimis ee

Pearl millet (Pennisetum glaucum (L.) R.Br.) is the major staple crop produced by subsistence farmers in the West African Sahel, but its panicle yield is low because of poor seedling establishment in low-nutrient soils. Seedball is a cheap seed-pelleting technique that combines sand, loam, seeds and optionally wood ash or mineral fertiliser as an additive to enhance early growth of pearl millet under infertile soil conditions. The aim of this study was to investigate the effects of seedball technology on pearl millet crop establishment and panicle yield on-farm under Sahelian subsistence conditions. Over 2000 on-farm (2015–18) trials were conducted in 65 villages of the Maradi region in Niger. Conventionally sown and seedball-derived pearl millet crops were grown by using ‘farmer-optimised’ simple split-plot designs with three treatments: (i) farmers’ practice as control; and seedballs (2.0 cm diameter, made from 80 g sand + 50 g loam + 25 mL water + 2.5 g seeds as standard recipe) that contained either (ii) 3 g wood ash or (iii) 1 g mineral fertiliser (nitrogen, phosphorus, potassium; NPK) as effective nutrient compounds. In 2016–18, participating farmers could opt for one of the seedball treatments. Panicle as well as stover yield data were collected and compared with respect to seedball type (wood ash vs NPK), sowing depth (shallow vs deep), sowing time (wet vs dry), weed management (complete vs partial), local soil type (texture range sand to loamy sand), cropping system (sole vs mixed), and farmer. Results showed that seedballs do not suppress seedling emergence. Seedball treatments produced fewer but longer and denser panicles. Wood ash-amended seedballs showed a higher panicle yield increase relative to their site controls than NPK-amended seedballs. However, the average panicle yield of NPK-amended seedballs was higher than that of the wood ash-amended seedballs. The treatment factors wet sowing, partial weeding, sole cropping and farmers showed higher panicle yield. The seedball technology increases pearl millet panicle yield by ∼30% in the Sahel; it is simple and based on local materials.

No abstract