Matching improved maize production technologies to the resource base of farmers in a moist savanna

Methods that permit rapid assessment of crop production technologies may help in fine-tuning arid focusing research. The Decision Support System for Agrotechnology Transfer (DSSAT), simulation models were used to study the maize technologies including varieties, residue management, time of planting, and inorganic N fertilization for farmers' conditions in a tropical moist savanna. Linear regression analysis showed that simulated grain yield, and days to silking and physiological maturity of the varieties (OBA SUPER2 and TZUTSR-WC5) approximated observed measurements. Across various component technologies, OBA SUPER2, a longer duration maize variety, yielded more than TZUTSR-WC5 a medium duration variety. Farmer attitude to planting time was satisfactorily explained by simulation. Planting early (April-May) matches the production strategy of farmers with no fertilizer input. This strategy offers more possibility of obtaining higher yields than planting at the regular time (mid May-June). Appropriate crop residue management, and practices that ensure high soil organic carbon increase the productivity of no-fertilizer farmers. In systems where high levels of fertilizer N are used (120-150 kg/ha), obtainable yields are higher but responded less to alternate planting windows and soil fertility status. (C) 2003 Elsevier Science Ltd. All rights reserved.