Calibration and evaluation of SWAP model under PRD irrigation: a case study, Pakdasht, Iran

ABSTRACT Crop models are effective tools for irrigation management, yield prediction, and improving water productivity. However, variability in irrigation regimes and agronomic practices challenges accurate simulation across regions. This study aimed to calibrate the Soil-Water-Atmosphere-Plant (SWAP) model for maize in the Pakdasht region (Iran) and evaluate its performance under Partial Root-Zone Drying (PRD) irrigation. Calibration was conducted under full irrigation (100% water requirement), while validation used deficit irrigation treatments (75% and 55%). Model performance was assessed using plant height, leaf area index, and biomass yield. Under full irrigation, the model showed high accuracy for LAI (R2 = 0.97, RMSE = 0.20, NRMSE = 6%). Although performance declined under deficit irrigation, acceptable accuracy was maintained (R2 = 0.88). Plant height was well predicted across all treatments (R2 = 0.88-0.99). Biomass yield showed excellent agreement under full irrigation (R2 = 0.98) but lower accuracy under deficit irrigation, particularly at 75% PRD (R2 = 0.44). Overall, the SWAP model reliably simulated vegetative growth under PRD irrigation, while biomass prediction showed greater uncertainty under severe water stress.