Watershed Hydrological Responses to Changes in Land Use and Land Cover at Hangar Watershed, Ethiopia.

Abstract

Land use/land cover change has been responsible for altering the hydrologic response of watersheds leading to impact stream flows. The various water resources project planning and implementation need the knowledge of factors influencing watershed hydrology. Therefore, this study analyzed the land use/land cover change from 1987 to 2017, and the effect these changes have had on the hydrology of the Hangar watershed, which is the tributary of the Didessa River Basin, Ethiopia. The study was accomplished through integrating ERDAS imagine (2015) software and the SWAT model. The land use/land cover data (Landsat-5 TM, Landsat-7 ETM+ and Landsat-8 OLI_TIRS, for the year 1987, 2001 and 2017 respectively) acquired from the website of USGS. The ERDAS imagine (2015) software used to generate the land use/land cover maps through the Maximum Likelihood Algorithm of Supervised Classification. The results indicated that, cultivated land and built-up area expanded (by 23.7%, 6%, and 28.2%) and (by 22.2%, 19.3%, and 37.2%) for LULC of 1987-2001, 2001- 2017 and 1987-2017 respectively. Whereas, there was a decrease in forest (by 20.4%, 15.1% and 32%), rangeland (by 28.1%, 12.6% and 37%), grassland (by 64.9%, 6.1% and 67%) and water body (by 1.9%, 53.8% and 55%) for LULC of 1987-2001, 2001-2017 and 1987-2017 respectively. In addition to the land use/land cover data, the input data used for the SWAT model simulation were the DEM data, soil data, and climatic data. Parameters sensitivity analysis, calibration and validation of the SWAT model carried out by the SWAT-CUP through SUFI-2 program. Sensitivity analysis has indicated that the curve number is the most sensitive parameter that could affect the hydrology of the watershed. The model calibrated and validated using measured streamflow data of 13 years (1990-2002) and 9 years (2003-2011) respectively. The SWAT model performs well for both calibration (R2 = 0.87, NSE = 0.82 and PBIAS = +1.4) and validation (R2 = 0.89, NSE = 0.88 and PBIAS = +1.2). The result after simulation indicated that the annual total water yield of the watershed decreased which is 790.26mm for LULC of 1987, 777.38mm for 2001and 766.08mm for 2017. The annual simulated stream flow through the study period is increased for wet (by 3%, 4% and 7%) and short rainy season (by 2, 5% and 7%) whereas, decreased for dry season (by 2%, 1% and 2%) for LULCC of 1987-2001, 2001-2017 and 1987-2017 respectively. Unless the proper watershed resources management implemented, the increase in runoff has implication for increasing soil erosion and sedimentation. The increase of wet season flow may result flooding, and the reduction of dry season flow may affect water scheme practice. Therefore, curving the changes of LULC towards increasing vegetation cover is very necessary in order to reduce surface runoff that contribute to wet season flow and increase infiltration that supply groundwater from which dry season/baseflow is contributed.