Identifying factors limiting legume biomass production in a heterogeneous on-farm environment
MetadataShow full item record
Douxchamps, S., Frossard, E., Uehlinger, N., Rao, I., Hoek, R. van der, Mena, M., Schmidt, A. and Oberson, A. 2012. Identifying factors limiting legume biomass production in a heterogeneous on-farm environment. Journal of Agricultural Science 150(6): 675-690.
Permanent link to cite or share this item: http://hdl.handle.net/10568/33527
Multipurpose legumes provide a wide range of benefits to smallholder production systems in the tropics. The degree of system improvement after legume introduction depends largely on legume biomass production, which in turn depends on the legumes’ adaptation to environmental conditions. For Canavalia brasiliensis (canavalia), an herbaceous legume that has been recently introduced in the Nicaraguan hillsides, different approaches were tested to define the biophysical factors limiting biomass production on-farm, by combining information from topsoil chemical and physical properties, topography and soil profiles. Canavalia was planted in rotation with maize during two successive years on 72 plots distributed over six farms and at contrasting landscape positions. Above-ground biomass production was similar for both years and varied from 448 to 5357 kg/ha, with an average of 2117 kg/ha. Topsoil properties, mainly mineral nitrogen (N; ranging 25–142 mg/kg), total N (Ntot; 415–2967 mg/kg), soil organic carbon (SOC; 3–38 g/kg) and pH (5•3–7•1), significantly affected canavalia biomass production but explained only 0•45 of the variation. Topography alone explained 0•32 of the variation in canavalia biomass production. According to soil profiles descriptions, the best production was obtained on profiles with a root aggregation index close to randomness, i.e. with no major obstacles for root growth. When information from topsoil properties, topography and soil profiles was combined through a stepwise multiple regression, the model explained 0•61 of the variation in canavalia biomass (P < 0•001) and included soil depth (0•5–1•70 m), slope position, amount of clay (19–696 kg/m2) and stones (7–727 kg/m2) in the whole profile, and SOC and N content in the topsoil. The linkages between topsoil properties, topography and soil profiles were further evaluated through a principal component analysis (PCA) to define the best landscape position for canavalia cultivation. The three data sets generated and used in the present study were found to be complementary. The profile description demonstrated that studies documenting heterogeneity in soil fertility should also consider deeper soil layers, especially for deep-rooted plants such as canavalia. The combination of chemical and physical soil properties with soil profile and topographic properties resulted in a holistic understanding of soil fertility heterogeneity and shows that a landscape perspective must be considered when assessing the expected benefits from multipurpose legumes in hillside environments.