Mixed research-farmer assessments of legume integration scenarios in upland rice-based cropping systems in Madagascar

Summary In Madagascar, a three-year mixed-methods research approach was conducted through a partnership with a network of neighboring farmers. The objective was to explore the integration of legumes into biannual rice-based rotations. Thirty-six legume integration scenarios were jointly selected and assessed on field at the rotation scale over two cropping seasons, crossing (1) cropping intensity, represented by four legume integration modalities: sole crop (SC), intercropped with maize (IM), intercropped with cassava (IC), and intercropped with both maize and cassava (IMC); and (2) legume type, consisting of nine cropping options: four pulses, four cover crops, and four controls without legumes. Over the two cropping seasons, mean gross incomes per rotation were significantly higher in SC (US$273) and IC (US$301) systems compared to IM (US$210) and IMC (US$228). Carbon inputs were significantly greater in SC (1.16 t C ha−1) and IM (1.06 t C ha−1) systems than in IC (0.84 t C ha−1) and IMC (0.90 t C ha−1). During participatory field sessions, farmers gave the highest acceptance scores to SC systems (score: 809), followed by IC (610), IM (575), and IMC (428). Regardless of cropping intensity, the two traditional pulse species led to higher gross incomes and greater farmer acceptance, but with lower carbon inputs compared to the introduced pulse and cover crops. A composite assessment revealed that economic and social performances may match, while economic and environmental performances are never related. This original approach provided valuable insights for both researchers and farmers for feasible solutions at both regional and farm levels.