The Use Of GIS And Remote Sensing Techniques To Analyse Water Balance Of Lake Bogoria Under Limited Data Conditions

Abstract

Analysis of water balance for Lake Bogoria in Kenya using GIS and remote sensing techniques is presented in this study. Due to limited meteorological and hydrological data, long term averages of mean annual rainfall, streamflow and potential evaporation were used. Land cover of the catchment was derived from Landsat TM of 1995 upon which the cover coefficients for determining actual evapotranspiration were estimated. The lake has a surface area of 35 km² with catchment area of 785 km². The water balance model used in the study is that of a single reservoir whose components consist of inflows to and outflows from the lake. The inflows consist of streamflows, direct rainfall on the lake and discharges from hot springs. The mean magnitudes of these components on annual basis were found to be 31.00 Mm³, 24.29 Mm³ and 28.38 Mm³ respectively giving a total of 83.67 Mm³. The outflows included actual evaporation from the lake, domestic, livestock and irrigation abstractions. The estimated mean annual values of these outflows were 68.75 Mm³, 0.7 Mm³ 1.2 Mm³ and 13.02 Mm³ respectively. Domestic and livestock abstractions were estimated based on human and livestock population while abstractions from irrigation were estimated from the difference between inflows and outflows. All the data were analysed in the form of thematic maps in raster format using computation algorithms within the GIS framework. The results show how this technology can be used to provide reasonable estimates of water balance of a reservoir under limited data conditions. The estimated results can be used as a starting point in developing water resources management and planning systems.