Abstract
This study aims at developing a physically based semi-distributed rainfall-runoff model in the HEC-HMS platform to predict the historical and future stream flow of the Dhaka River basin. This model adopted and integrated several physio-hydrographic parameters as input data, such as LULC, HSG, DEM, observed stream flow, historical and projected future precipitation of the ACCESS-CM2 ensemble of the Coupled Model Intercomparison Project Phase 6 (CMIP6). The stream flow predictability of the HEC-HMS is observed subject to the model functions and simulation controls. After developing the model architecture, during calibration, a Nash-Sutcliffe efficiency (NSE) of 0.78 and a coefficient of determination (R2) of 0.81 were found, which indicates the simulation efficacy of the model setup. Furthermore, in the validation phase, the model demonstrated its robust performance, with R2 = 0.80 and NSE = 0.78. The study showed the predicted yearly peak discharge of about 341685.8 m3/s, 330017.4 m3/s, and 315588.9 m3/s under the SSP1-2.6, SSP2-4.5, and SSP5-8.5 scenarios, respectively. Here, the Mann-Kendall and Sen's slope tests were conducted to analyze the daily, monthly, and yearly stream flow trends and they substantiate a significant increase in predicted daily streamflow under both the SSP1-2.6 and SSP2-4.5 scenarios and a gradual increase in monthly discharge in May under the SSP5-8.5, as well as in July and August under the SSP1-2.6. Outcome of this study underscores the model’s robustness and contributes vital perceptions for flood control and mitigation strategies.