Applicability assessment of the CASCade Two Dimensional SEDiment (CASC2D‐SED) distributed hydrological model for flood forecasting across four typical medium and small watersheds in China.

Hydrological modelling is a critical tool for preventing and mitigating severe flood disasters. This study aims to assess the applicability of a physically based distributed hydrological model, CASCade Two Dimensional SEDiment (CASC2D‐SED), for flood forecasting in four typical medium and small watersheds across three hydroclimatic zones (semi‐arid, semihumid, and humid) in China. CASC2D‐SED model has the best performance in Xianbeigou (semi‐arid), followed by Luanchuan (semihumid), and Shujia (humid), while has the relatively lowest performance in Chengcun (humid). It tends to underestimate a large portion of the discharges in the watersheds, indicating that the Green‐Ampt infiltration‐excess model, the only runoff generation scheme in CASC2D‐SED model, imposes restrictions on hydrological simulation without further accounting for subsurface and groundwater flows. This limitation is particularly obvious in the semihumid and humid regions. Our analysis suggests that combining infiltration‐excess and saturation‐excess runoff generation mechanisms and accounting for subsurface and groundwater flows in CASC2D‐SED will further enhance its capacity for flood forecasting and hydrological simulation for semihumid and humid hydroclimatic zones. [ABSTRACT FROM AUTHOR]