PEAK FLOW PREDICTION BY USING MACHINE LEARNING MODELS AND HYDROLOGIC HAZARD ASSESSMENT FOR KESSEM DAM (CASE STUDY OF KESSEM WATERSHED, AWASH BASIN, ETHIOPIA

Abstract

In dam systems, flood is a peak flow that the major problem and cause of dam failure in the form of overtopping phenomena, especially for embankment dams. A quantitative hydrologic risk assessment of the dam is used for dam safety evaluation to deciding whether existing structures provide adequate levels of safety or not, and what modifications are justified to improve the dam's safety. This thesis considered the Kessem River watershed of the Awash basin, located at 9o 8‘45" latitude and 39o 55'31" longitude at the southern end of the Afar rift in the Afar region of Ethiopia, and focused on the accurately techniques of peak inflow prediction to the reservoir, including the semi quantitative assessment of hydrologic hazard for Kessem Dam using machine learning predictive models and RMC-RFA software. The SDSM model was used to downscale from the CanESM2 output GCM to study area and project rainfall, maximum and minimum temperature climate data for future. The future LULC scenario in the watershed is estimated based on an assessment of LULC change over the last two decades with reclassified LULC using Landsat images. The most recent three RNN (LSTM, Bi-LSTM, and GRU) machine learning predictive models with hybrid to SCS-CN model were used for simulation of Kessem river flow. The hydrologic hazard curve for semi-quantitative risk assessment was developed by uploading the peak inflow events to the reservoir using RMC-RFA software. After training the models in Python 3.9, results show that the Bi LSTM model was a powerful predictive model with acceptable Nash Sutchcliffe Efficiency and Root Mean Square Error (NSE = 0.968 & RMSE = 3.873) for training and (NSE = 0.744 & RMSE = 17.547) for testing in the study area. Peak daily inflow events to the reservoir are a predicted to be 467.72 m3 /s, 435.88 m3 /s, and 513.55 m3 /s occur in 2035, 2061, and 2090 year, respectively. The hydrologic hazard analysis results show that 2,823.57 m3 /s and 935.21 m, 2,126.3 m3 /s and 934.18 m, and 11,491.1 m3 /s and 942.11 m peak discharge and maximum reservoir water level during the periods of 2022-2050, 2051-2075, and 2076-2100, respectively, for 0.0001 APE. As concluded, Kessem dam may potentially be overtopped by a flood with a return period of about 10,000 years during the period of 2076–2100. So, the dam required further risk analysis study and dam safety modification in order to control this probable failure mode during the indicated tim