Modeling Rainfall-Runoff Dynamics Using HEC-HMS: Case of Gololcha Watershed

ABSTRACT Accurate simulation of rainfall–runoff processes is essential for flood prediction and sustainable water resources management, particularly in watersheds that exhibit high seasonal variability in streamflow. The Gololcha watershed in the Wabe Shebele River Basin of Ethiopia frequently experiences high runoff during intense rainfall events, which leads to flooding in the surrounding areas. Understanding the hydrological response of the watershed is therefore important for effective flood risk mitigation and the planning of hydraulic structures.The main objective of this study was to model the rainfall–runoff process of the Gololcha watershed and evaluate the performance of the HEC-HMS model for streamflow simulation and peak flood estimation. Daily rainfall and streamflow data for the period 2000–2020 were used together with a 12 × 12 m Digital Elevation Model, land use/land cover, and soil maps to derive the physical characteristics of the watershed. Missing rainfall data were filled using XLSTAT by linear regression of a nearest neighbor approach was applied to fill the missing data, and areal rainfall was computed by the Thiessen polygon method. HECGeoHMS was applied to delineate the watershed and generate model parameters such as sub-basins, stream network, curve number, and lag time. The Soil Conservation Service Curve Number method, SCS Unit Hydrograph, monthly constant baseflow, and Muskingum routing method were used for loss estimation, runoff transformation, baseflow simulation, and channel routing, respectively. Model calibration and validation were carried out using observed streamflow data.The model performance was evaluated using Nash–Sutcliffe Efficiency (NSE), coefficient of determination (R²), and percent bias (PBIAS). The results showed good agreement between observed and simulated flows, with NSE values of 0.809 and 0.845, R² values of 0.813 and 0.92, and PBIAS values of 11.97% and 14.08% for calibration and validation, respectively. Flood estimation based on 24-hour rainfall for different return periods indicated that peak discharge increases from 106.8 m³/s for the 2-year return period to 630.3 m³/s for the 500-year return period, revealing the high flood potential of the watershed. Overall, the study confirms that the HEC-HMS model is a reliable tool for rainfall–runoff simulation and design flood estimation, and the results provide valuable input for flood risk management and watershed planning in the study area.