Model-based comparison of crop yield responses to water, nitrogen, and weed management using APSIM and AquaCrop
摘要
Agricultural productivity in sub-Saharan Africa is limited by inefficient water, nitrogen, and weed management. This study investigates their interplay on crop yield under resource constraints, using two crop simulation models, AquaCrop and APSIM, to guide sustainable practices. Twenty-seven scenarios, generated from three nitrogen levels (high, moderate, low fertility), three weed management regimes (fortnightly, monthly, mid-season removal), and three maximum allowable soil water depletion (MAD) thresholds (40%, 70%, 100%), were simulated for maize and potato in a 3 × 3 × 3 factorial arrangement within a Randomised Complete Block Design. Simulation outputs were analysed using ANOVA at the 5% significance level, with means separated by Duncan’s Multiple Range Test. Model outputs were compared using correlation coefficients (r), Root Mean Square Error (RMSE), and Mean Absolute Error (MAE). Maximum maize yield at maturity was 10,755 kg ha⁻1 for scenarios with 100 kg N ha⁻1, fortnightly weed removal, and MAD of 70% or 100%. Maximum potato yield at maturity was 9020.95 kg ha⁻1 for 150 kg N ha⁻1, fortnightly weed removal, and MAD of 70% or 100%. These yields were not significantly different from other high-performing scenarios. At maturity, predicted grain yield from maize using APSIM and AquaCrop had low correlation (r = 0.41) but no significant difference (RMSE = 3847.41 kg ha⁻1; MAE = 2870.80 kg ha⁻1). Potato root yields at maturity showed medium correlation (r = 0.65) but significant difference. These results underscore the strategic importance of integrated water, nutrient, and weed management for maximising productivity while conserving resources. This comparative modelling approach provides a robust decision-support framework for enhancing agricultural sustainability under climate variability.