FLOOD RISK MAPPING USING HEC-RAS MODEL: CASE STUDY ON WAJA WATERSHED IN RIFT VALLEY BASIN CENTRAL ETHIOPIA REGION, ETHIOPIA

Abstract

Flood is among the most devastating natural disasters worldwide, significantly affecting human lives and property. The current study conducted on the Waja River floodplain aimed to model and maps the flood inundation, flood hazard, flood vulnerability, and flood risk associated with flooding in the area. To achieve this objective, various data sources were utilized, including meteorological, hydrologic, and topographic data collected from different organizations. The study employed several tools and materials, including the HEC HMS and HEC-RAS models, GIS software, GPS devices, and metering tape. The HEC HMS model was used to analyze flood hazard and risk by developing inflow design floods for different return periods. The model was calibrated and validated using actual stream flow data. During model calibration the NSE value was 0.75, Percent Bias (PBIAS) was 2.02, coefficient of determination (R2 ) was 0.78, and Relative Mean Square Error (RMSE) was 2.03. During the validation period, the model achieved an R2 of 0.77, NSE of 0.76, PBIAS of 1.64, and RMSE of 1.3. After calibration and validation, the annual maximum precipitation from rainfall data was extracted to develop frequency storms for different return periods. These storms were then used as input for the HEC HMS model to generate flood hydrographs. The HEC-RAS model, combined with the flood hydrographs, was used to produce flood inundation maps, which were visualized in ARC-GIS software for detailed analysis. The results of the study indicated that for return periods of 10, 25, 50, and 100 years, the areas inundated by floods were 3030 ha, 3364 ha, 3520 ha, and 3683 ha, respectively. Additionally, the maximum flood depths were found to be 6.3m, 9.2m, 12.6m, and 14.45m for the respective return periods. The maximum flood velocities were 3.8 m/s, 4.7 m/s, 5.5 m/s, and 6.8 m/s for the same return periods. Flood hazard maps were derived from the depth, velocity, and duration of floodwaters, revealing that 35% of the flooded area was categorized as having very high and high hazard, while approximately 65% was classified as medium and low hazard. The flood vulnerability map classified approximately 17% of the flooded area as having high and very high vulnerability. About 18% of the flooded area fell into the moderate vulnerability class. The majority of the flooded area, approximately 65%, had low and very low vulnerability. By combining the flood hazard and vulnerability information, the study developed a flood risk map. The results showed that 24% of the area fell into the high and very high-risk categories