IMPACT OF LAND USE AND LAND COVER CHANGES ON STREAM FLOW AND SEDIMENT YIELD: CASE STUDY ON GIDABO RIVER CATCHMENT, RIFT VALLEY BASIN ETHIOPIA

Population growth, deforestation and higher demand for farm lands have effect on natural resource of the Gidabo River Catchment, Rift Valley Basin. Assessing the problem and finding solution at catchment level is use full to minimize the effect of the problem. Decision support tools are needed for successful assessment of the hydrology and soil erosion process for planning and application of soil and water conservation measures. In this study the effect of land use and land cover changes on stream flow and sediment yield is observed. Soil & Water Assessment Tool (SWAT) model was calibrated & used to evaluate changes on sediment load and stream flow. Two SWAT model set up were run to evaluate the impacts of land use changes on stream flow and sediment yield using two land use for year 1990 and 2013 of the study area. Sensitivity, calibration, validation and uncertainty analysis was conducted using Sequential Uncertainty Fitting–version 2 (SUFI-2) in SWAT-CUP (Calibration and Uncertainty Program) using sediment yield and stream flow historical data. Model calibration and validation was conducted for both land use using measured data near to the two land use by using multi gauge calibration approach in monthly base using SWAT-CUP. For 1990 land use calibration was from 1989-1995 and validation was from1997-2000, for 2013 land use calibration was from 1997-2002 and validation was from 2004-2006. Results from calibration for both land use show acceptable range (0.50 to 0.70 for NSE and 0.55 to 0.72 for R2) between observed and simulated stream flow and sediment yield in monthly base. The model was able to bracket 63 to 87% of the observed data in acceptable uncertainty band (r-factor, 0.68 to 1.28). The results of validation were acceptable (0.51 to 0.76 for NSE and 0.6 to 0.78 for R2). The model bracket 51% to 83% of the observed data in acceptable uncertainty band (r-factor, 0.58 to 2.13). The annual sediment load at Gidabo dam site was found to be 435.26 ton per square kilometer per year. Three scenarios are developed to observe the impact of land use changes. Based on this 53% of change in forest land to agriculture land resulted in 67% and 29% increase in sediment load and stream flow respectively, 19% of change in forest land to agricultural land resulted in 43%and 13% increase in sediment load and stream flow respectively. Change in 35% of grass land into agricultural land increase the sediment yield by 9% and stream flow by 3%