Global Sensitivity Analysis and Evaluation of the DSSAT Model for Summer Maize (Zea mays L.) Under Irrigation and Fertilizer Stress

Sensitivity analysis (SA) can identify the most critical parameters for crop growth model output, thus helping to improve model calibration efficiency. However, when combined with different production conditions, especially adverse conditions such as water stress and fertilizer stress, parameter sensitivity remains unclear. This study (i) assessed the sensitivity of the output response of the CERES-Maize model to the input parameters, particularly the effect of water and fertilizer stress on the SA results and (ii) evaluated the model performance based on the SA results. The results indicated that water stress had a considerable effect on SA, whereas nitrogen stress had little effect on SA. P5, G3, and P2 had significant effects on yield, maximum aboveground biomass (AGB), daily AGB, daily leaf area index (LAI), and daily actual evapotranspiration (ETc). Under water stress, soil drainage rate, soil runoff curve number, and photosynthesis factor greatly affected the output response of CERES-Maize. Compared with the calibration of maize cultivar coefficients, CERES-Maize with additional consideration of soil parameter calibration was more accurate. The model evaluation results revealed that the simulated LAI, yield, and soil water content were consistent with the actual measured values. These findings can provide a reference for the calibration of CERES-Maize model parameters under water and fertilizer stress.