Stress-Responsive Water Productivity Improves AquaCrop Simulation of Winter Wheat Under Water-Stress Conditions

Accurately simulating wheat production under water-limited conditions remains a major challenge for AquaCrop and other models used in irrigation management. In this study, the hypothesis of variable normalized water productivity (WP*) was evaluated for winter wheat under water-stress conditions. The original AquaCrop (AquaCrop-OS) model was applied across five irrigation treatments over two growing seasons. The AquaCrop-OS model successfully simulated canopy cover dynamics and soil water content, the latter with acceptable accuracy. However, consistently overestimated aboveground biomass and grain yield under moderate to severe water stress, with normalized root mean square error (NRMSE) values of 15.0% and 16.0%, respectively. The enhanced model version, AquaCrop-WM, which incorporates a water stress-responsive adjustment to WP^*, substantially reduces these discrepancies, lowering NRMSE to 7.9% for biomass and 8.7% for grain yield. In this improved model, a dimensionless coefficient (W_s) is used to control the sensitivity of water productivity to stomatal stress. The W_s value for winter wheat was estimated to equal 3.63, which is lower than the previously reported value for maize (5.32), indicating winter wheat’s moderate sensitivity to progressive water limitation.