Watershed Modeling Using Soil and Water Assessment Tool (Case Study on Nashe Lake of ETHIOPIA)

Abstract

Fincha Amarti Nashe multi-purpose hydropower project, and above of 322km2 , dam volume more than 448 million cubic meters and (the project) is located about 364km North West of Addis Ababa from capital town of Ethiopia, 50km from Zonal capital Shambu, in the Blue Nile river basin and having the latitude 9°35' to 9°52'N and Longitudes 37°00' to 37°20'E. The spatially distributed data (GIS input) needed for the Arc SWAT interface include Digital Elevation Model (DEM, Resolution of 30mx30m), land use/cover data, soil and the weather (climate) data, and AVSWAT Hydro meteorological and hydrological data were the base data for Watershed modeling. The Sensitivity analysis of the lake Nashe is highly sensitive to Lake Precipitation, river inflow and Evaporation and the total twenty seven parameters were considered for the model of parameterization sensitivity analysis; only ten of them were effective for monthly flow simulation analysis. For this research 30 years historical data were used for Nashe watershed. However, the calibration was run for 17 years (1985-2001) where the first one-year 1985 is used to “Warm up” the model and the measured data of stream flow of 13 years period of (2002-2014) were used for the model validation process. During the stream flow calibration the coefficient of determination 2 R and the Nash-Sutcliffe equation has been applied for model testing between simulated and observed flows and calculated on the monthly basis was 0.81 and 0.80 respectively, and again the correlation coefficient ( 0.84 2 R  ) and the Nash-Sutcliffe ( NS  0.74 ) shows a good agreement validation for the between the observed and simulated values. Means the evaluates that the 2 R ranges from 0.0 to 1.0 with higher values indicating better agreement and the values of NS ranges minus infinity to 1.0, with higher values indicating better agreement.