ESTIMATION OF SEDIMENT FLOW TO GRAND ETHIOPIAN RENAISSANCE DAM RESERVOIR AND SEDIMENT REDUCTION METHOD USING SWAT MODEL: ABAY BASIN, ETHIOPIA

Abstract

Reservoir Sedimentation is the gradual accumulation of sediment from a river into the reservoir. As part of a long-term optimal sediment management program, it is important to predict sediment flow into the Grand Ethiopian Renaissance Dam, assess impacts on the reservoir, and maintain dam capacity. The overall objective of this study was to estimate sediment yield, map the watershed in terms of sediment production, and determine the most suitable BMPs for the GERD Basin by using SWAT model. The watershed is delineated, calibrated and validated at the El Deim station where the watershed area is 174,580.12 Km2 and by using the calibrated model simulate the sediment and flow at the GERD axis (172,250 Km2 ). For the model 45 metrological stations were used 42 is from NMI and the other 3 is from NASAPOWER Website with appropriate bias correction and the LULC and soil map is taken from ENTRO after comparing with the data’s available at ( FAO , Map server Ethiopia, USGS. and Alaska Edu). Model calibration result for flow of (2002-2006) gives coefficient of determination (R2 ) and Nash-Sutcliffe simulation efficiency (NSE) 0.92 and 0.91 respectively and for validation period (2007-2010) the model gives 0.88 and 0.88 for R2 and ENS values respectively. Model Sediment calibration gives R2 and NSE 0.72 and 0.72 respectively and validation period gives R 2 and NSE 0.67 and 0.67 respectively. This result indicates that the recorded values show good agreement with model simulated value for both stream flow and sediment yield. The rainfall runoff and sediment yield relationship of the catchment has been determined for each sub-basin of the watershed. Also, the spatial distribution of sediment yield, the temporal distribution of rainfall runoff, and sediment yields of the catchment were determined and the hotspot area that is sensitive to soil erosion was identified. The model result indicates that the simulated annual average sediment yield for the GERD Basin was 130.83 t/ha/yr. and the long-term average annual sediment loads obtained from the SWAT model at EL Deim was 159.37 million ton/year. This study reveals that 59.55% of the GERD Watershed area is classified as extremely erosion prone, 1.32% and 6.68% as very high and high erosion prone, 3.44% as moderately prone to erosion, and 20.01% as very low prone to erosion. Three different manage mencenario’s wear proposed from them Strip-cropping is found to be the best management practice that needs to be employed on the selected hot spot areas in order to increase the service life of the GERD Reservoir