THE EFFECT OF LAND USE LAND COVER CHANGE ON HYDROLOGY OF BILATE WATERSHED, SOUTHERN RIFT VALLEY BASIN OF ETHIOPIA.

Abstract

Land use and land cover change (LULCC) has been a key research priority throughout the world with multi-directional impacts on both human and natural systems. The prime goal of this study is to assess the effect of land use land cover change on streamflow of Bilate watershed, which is located in the southern rift valley basin of Ethiopia with a total area of 5252 km 2 . Specifically, the study analyzed the spatial-temporal changes of land use land cover using 1985, 2000 and 2007 land use and land cover maps. This study plays a role to build the ability of planners, researchers and other stake holders to formulate and implement sound policies to minimize undesirable future land use land cover impacts on streamflow. The land cover changes within the Bilate watershed were examined through classification of satellite images with integrated use of ERDAS imagine 2014 and Arc GIS 9.3. SWAT model was used to investigate the impact of land cover change on streamflow of the study area. The model was set up using readily available spatial and temporal data, and calibrated using the measured discharge. The land cover maps were developed using the Maximum Likelihood Algorithm of Supervised Classification. The accuracy of the classified maps was assessed using Confusion Metrics. Land cover change analysis has shown that the cultivated and Settlement area has increased from 34.78% to 64.33% as well as urban areas from 5.92% to 11.45% between 1985 and 2007, while shrub land has decreased from 35.99% to 14.67% and Grass land from 23.39% to 9.55%. Cn2, Esco and Gwqmn were the most sensitive parameters identified for the stream flow of the study area and used for calibration of the model. The model calibration was carried out using observed stream flow data from 01 January 1990 to 31 December 2000 and a validation period from 01 January 2001 to 31 December 2007. Both the calibration and validation results showed a good agreement between measured and simulated stream flow with the coefficient of determination (R 2 ) of 0.80 and Nash-Sutcliffe efficiency (ENS) of 0.75 for the calibration, and R 2 of 0.76 and ENS of 0.72 for the validation periods. The evaluation of SWAT model showed that annual surface runoff increased by +41mm and ground water flow declined by -44mm between 1985 and 2007.Therfore, the model results showed that the stream flow characteristics changed due to the land use/cover changes during the study periods.