Simulation of Flood Inundation Extent by Integration of HEC-HMS, GA-based Rating Curve and Cost Distance Analysis

Flood inundation modeling is crucial for analyzing floods and understanding the impact of rainfall on runoff. Traditional hydraulic models require substantial data and technical skill, which creates difficulties in situations with limited data and expertise. To reduce computational cost and data requirements, the study utilizes HEC-HMS, a Genetic Algorithm (GA)-based rating curve, and cost-distance analysis to develop a flood inundation model in Tlawng river basin. The HEC-HMS SMA model was calibrated for daily streamflow from June 2017 to May 2020 and then cross-validated from Nov 2018 to May 2020. The model streamflow predictions were evaluated using statistical parameters such as Root Mean Standard Deviation Ratio (RSR), Percent Bias (PBIAS), Nash–Sutcliffe Efficiency (NSE), Root Mean Squared Logarithmic Error (RMSLE), and Mean Absolute Error (MAE). The values obtained were 0.4 (calibration) and 0.5 (validation) for RSR, 5.78% (calibration) and 9.28% (validation) for PBIAS, 0.85 (calibration) and 0.775 (validation) for NSE, 0.18 (calibration) and 0.21 (validation) for RMSLE, and 9.92 (calibration) and 7.12 (validation) for MAE, respectively. Then, a GA-based rating curve was developed to convert the simulated discharge into depth at basin outlet. Furthermore, cost-distance analysis tool was used to determine flood inundation extent. Modeled flood inundation extents were accurately validated by observed GPS points of flood marks. The horizontal distance between the modeled inundation and the actual flood mark was 9.48 m, while the vertical distance coincided precisely at an elevation of 40 m MSL. The study demonstrates a practical and efficient way to assess flood risks and develop suitable mitigation strategies in the Tlawng river basin.