Is AquaCrop a useful tool for rapeseed growth and yield prediction in semi-arid regions: model evaluation under different water-saving using long-term weather data

Crop models are widely used to simulate and predict crop production under different field conditions. Therefore, the AquaCrop model was used to simulate the rapeseed growth and yield production under different water regimes during 3 years of field experiments. The data of the first year were used for model calibration, and the data of the other 2 years were used for model validation. The sensitivity analysis was also conducted to find out the most effective input variables on the model’s output. Thereafter, the AquaCrop model was applied to simulate the agronomic and irrigation traits of rapeseed using long-term data (1978–2011). Statistical analysis between the measured and simulated canopy cover as R2, NRMSE, and d of 0.88, 24%, and 0.97 for calibration and 0.70, 31%, and 0.91 for validation, respectively, illustrated the acceptable ability of AquaCrop model for rapeseed growth. In addition, similar results were observed for crop evapotranspiration with R2, NRMSE, and d as 0.57, 29.3%, and 0.86 for calibration and 0.78, 29.5%, and 0.92 for validation. The values of d and NRMSE for grain yield were 0.97 and 12.4% for calibration and 0.87 and 12.8% for validation, respectively. AquaCrop model was highly sensitive to minimum growing degree day for full biomass production and minimum air temperature range on simulating rapeseed grain yield. Based on the application of the AquaCrop model for the long-term, the best time to irrigate rapeseed without significant reduction in rapeseed grain yield and grain water use efficiency is when 60% and 70% of soil water is depleted, respectively. For better performance of the AquaCrop model, it is suggested to include different root distribution functions. In general, the performance of AquaCrop in simulating rapeseed production under semi-arid conditions is acceptable and can be used as a powerful tool for decision-making.