Development of A Hydrologic Model for Flood Prediction And Suitability Analysis of Boro Crop Cultivation for Sunamganj Haor Region Using HEC-HMS

The Haor region in Bangladesh, covering one-fourth of the entire region, is highly vulnerable to flash flood. Sunamganj is one of the worst affected Haor areas. During the Rabi season, these Haor areas are occupied by Boro crop fields, where water remains stagnant or experiences flash flooding. These flash floods spill onto low-lying flood plain lands in the region, inundating standing Boro crops, causing loss of thousands of hectares of lands. Although these sudden water surges primarily occur due to heavy rainfall in the Meghalaya region, it is necessary to develop a hydrological model for the low-lying areas in Bangladesh. Additionally, a strong and dedicated early warning system of flash flood forecasting with a considerable leading time for Sunamganj Haor region is essential to mitigate this enormous loss. In this study, we set up a hydrologic model of the Sunamganj Haor basin using Hydrologic Engineering Center, Hydrologic Modeling System (HEC-HMS). The discharge data for the Sunamganj station (Station ID: SW269) was collected from the Bangladesh Water Development Board (BWDB), while the hydrologic data originated from ERA5 (ECMWF Reanalysis v5) data available on the Copernicus Climate Data Store. The methodology involved delineation of catchment areas and stream networks, and setting up the model for different methods of different hydrological processes, including the Simple Canopy Method, Simple Surface Method, Initial & Constant Loss Method, SCS Unit Hydrograph, Recession, and Muskingum. The proposed model provides a sufficient lead-time forecast and computes the stage threshold condition for Boro crop cultivation. The model was calibrated and validated using historical observed data of two significant flood years 2007 and 2010 respectively. The model's accuracy was assessed using statistical measures, including Coefficient of Determination (R 2 ), Nash Sutcliffe Efficiency (NSE), Root Mean Squared Error Ratio (RSR), and Percent Bias (PBIAS). The parameters for different hydrologic methods were such that the determination coefficients and coefficients of agreement for all the flood events surpassed the acceptable threshold, including the R 2 values 0.9976 and 0.9631 for calibration and validation respectively. The calibrated and validated model was further applied to generate a flood hazard map categorizing the area into Threshold Zone, High Risk Zone, and Low Risk Zone. The Hazard Map, particularly relevant for Boro crop cultivation, considered the critical water level for the Surma-Meghna River and identified areas susceptible to flooding.