Sensitivity Analysis and Calibration of SWMM Model Parameters for Irrigation District Using the Morris Screening Method

This study uses the runoff coefficient method to evaluate the sensitivity of Storm Water Management Model (SWMM) parameters to rainfall events with varying return periods in the Jiaokou irrigation district, while calibrating the model to improve accuracy. A SWMM model was constructed for the district, and eight key hydrological parameters were selected. The parameters were calibrated using a 5-year rainfall event. The parameter combination corresponding to a runoff coefficient of 0.281 was selected as the optimal calibration result for the SWMM model. The model was validated using rainfall events with return periods of 1, 3, and 8 years. The sensitivity analysis revealed that Manning’s roughness for impervious areas (N-Imperv), depression storage for impervious areas (Destore-Imperv), and depression storage for pervious areas (Destore-perv) had significant impact on the total runoff. N-Imperv and Manning’s roughness for pervious areas (N-Perv) were highly sensitive to peak runoff, and drying time (DT) showed no sensitivity under any conditions. The simulated runoff coefficients at the validation stage were 0.254, 0.273, and 0.297 for the 1, 3, and 8-year rainfall events, respectively, which were close to the calibrated values. The results demonstrate that the SWMM model accurately represents the hydrological characteristics of the Jiaokou irrigation district, providing robust technical support for flood simulation and flood control in the region. It also offers practical guidance for flood simulation and management in other irrigation districts. This study focuses on identifying key sensitive parameters and constructing a verifiable model framework, emphasizing its engineering application value.