Sediment Yield Modeling And Evaluation Of Management Practices In Case Of Middle Awata Watershed, Southern Ethiopia

Abstract

Soil erosion is a worldwide problem that causes land degradation and long-lasting challenge in Ethiopia. Sediment yield of a watershed is influenced by the watershed characteristics (LULC, soil type and slope), climatic conditions and surface runoff. Middle Awata watershed is one of the Ethiopian watersheds which is susceptible for the problem of soil erosion in the river system. The sediment yield rate of the watershed and the spatial distribution of sediment yield at the subbasin level are necessary in order to create alternative watershed management methods. This study aimed at modelling of sediment yield and assessing management practices using Soil and Water Assessment Tool (SWAT) model in Middle Awata Watershed. Spatial data and weather data were major inputs for the model. The land use land cover map was created using ERDAS image classification of satellite images. The model simulation was done by dividing the watershed of total area 1912Km2 into 37 sub basins and 294 Hydrologic Response Units (HRUs) for the period of 31years (1988-2018). The model performance evaluation was carried out using Sequential Uncertainty Fitting (SUFI-2) algorithm in SWAT-Calibration Uncertainty Programs (SWAT-CUP) on monthly basis. The model was calibrated for the period of 18 years (1988- 2005) including the first two years as warm-up period and validated for 8 years (2006-2013) for both stream flow and sediment yield by using more sensitive parameters. The model performance was evaluated by coefficient of determination (R2 ), Nash Satcliff Efficiency (NSE), Root Mean Square Error Standard Deviation Ratio (RSR) and Percent bias (PBIAS) to be quit acceptable for both stream flow and sediment yield estimation in the watershed. It has resulted in (R2=0.76, NSE=0.75, RSR=0.51 and PBIAS= 5.6%) for calibration and (R2=0.75, NSE=0.74, RSR=0.5 and PBIAS=2.7%) for validation of stream flow. Whereas (R2=0.69, ENS=0.66, RSR=0.58 and PBIAS=3.7%) for calibration and (R2=0.67, ENS=0.65, RSR=0.61 and PBIAS=5.6%) for validation of sediment yield. The simulated annual average sediment yield was 34.543×103 ton/yr at outlet of the watershed. The determination of spatial distribution of sediment yield was identified the first 12 upstream sub watersheds to be soil erosion prone area. Evaluation of the selected BMPs (Filter strip, Contouring, Grassed waterway and Terracing) concluded that application of parallel terracing is the most recommended soil erosion reduction option with about 47.05% of reduction for almost all the critical sub basins of the watershed.