Content area
Abstract
Maize (Zea mays L.) is an important food crop in Ethiopia but yield is low due to numerous biotic, abiotic and management constraints. Conservation agriculture (CA) and better nitrogen management in targeted technology extrapolation domains (TED) could reduce some of these constraints. Generation of good agronomic practices for all of the diverse TED of Ethiopia through field research alone is not feasible due to resource scarcity in Ethiopia but use of crop simulation models coupled with geographic information systems (GIS) may greatly complement field research. A robust procedure was developed for the application of geospatial modeling of CA and N management practices in Ethiopia. Field study results indicated improvements in soil properties and crop yield may require some yr of CA before crop yield and soil benefits are achieved. Evaluation of CERES-Maize, CROPGRO-Dry bean, and CROPGRO-Soybean crop models under different cropping conditions suggested their suitability for simulating maize and legume responses to N rates and CA in Ethiopia. Split application of N at planting, at 40, and 60 days after planting greatly reduced N leaching and slightly improved maize yield at all TED. Either CAr (reduced tillage with 30% of crop residue retention and 75 kg N ha –1 under maize-soybean/dry bean rotation) or CAr+N (CAr but with 100 kg N ha–1) may be used for sustainable maize production across the target TED. Model generated maize N response coefficients varied for the conventional and conservation agriculture production conditions and the coefficients can be applied to optimize N fertilizer use at their respective TED in Ethiopia.
