Farmer preferences and legume intensification for low nutrient environments

Improved varieties of legumes adapted to nutrient deficiency have the potential to improve food security for the poorest farmers. Tolerant varieties could be an inexpensive and biologically smart technology that improves soils while minimizing fertilizer costs. Yet other technologies that improve productivity and appear to be biologically sound have been rejected by farmers. To translate benefits to smallholder farmers, research on low-nutrient tolerant genes and crop improvement must keep farmer preferences and belief systems in the forefront. We review farmer participatory research on legume-intensification and soil fertility management options for smallholder farmers in Africa, including recent results from our work in Malawi and Kenya. We suggest that indeterminate, long-duration legumes are the best bet for producing high quality residues, compared to short-duration and determinate genotypes. This may be due to a long period of time to biologically fix nitrogen, acquire nutrients, photosynthesize and grain fill. Also, the indeterminate nature of long-duration varieties facilitates recovery from intermittent stresses such as drought or pest pressure. However, indeterminate growth habit is also associated with late maturity, moderate yield potential and high labour demand. These traits are not necessarily compatible with smallholder criteria for acceptable varieties. Malawi women farmers, for example, prioritized early maturity and low-labour requirement, as well as yield potential. To address complex farmer requirements, we suggest the purposeful combination of species with different growth habits; e.g. deep-rooted indeterminate long-duration pigeonpea interplanted with short-duration soyabean and groundnut varieties. On-farm trials in Malawi indicate that calorie production can be increased by 30% through pigeonpea-intensified systems. Farmers consistently indicate strong interest in these systems. In Kenya, a 55% yield increase was observed for a doubled-up pigeonpea system (a double row of pigeonpea intercropped with three maize rows) compared to traditional, low density intercrops. However, the need for improved pigeonpea varieties with high intercrop suitability, including reduced early branching, was highlighted by a farmer preference study in the same area. These examples illustrate the potential for participatory research methodologies to drive biophysical research in farmer-acceptable directions.