ABSTRACT Flooding, one of the most frequent and devastating natural disasters, threatens life and property. Accurate flood simulation underpins flood prevention, early warning, and disaster mitigation. Traditional rainfall–runoff models typically utilize a fixed Curve Number (CN) during parameter calibration; however, this simplified approach cannot represent complex runoff processes driven by antecedent soil moisture, limiting forecasting performance. Thus, this study applied the semi-distributed HEC-HMS model to the Fenshui River Basin in Southeast China. Automatic parameter optimization with the Nelder-Mead algorithm was integrated with parameter sensitivity analysis and land-use evaluation. The study compared flood simulation accuracy and hydrograph characteristics under two calibration strategies: uniform parameter calibration and antecedent moisture condition (AMC)-based classification. Results showed AMC-based calibration improved model accuracy, with the mean relative error of peak discharge during validation reduced to 9.05% and NSE acceptance increased to 85.7%, outperforming the 31.76% peak error from the uniform parameter calibration. Furthermore, introducing AMC classification reduced the peak flow error by approximately 22.7 percentage points and decreased the runoff volume error from 41.96 to 15.29%. These findings highlight the effectiveness of AMC-based calibration for the HEC-HMS model, particularly in humid regions represented by the Fenshui Basin, providing technical support for regional flood warning and management.
Xiang et al. (Fri,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: