Due to water scarcity,analyzing the current situation of the water resource and monitoring its dynamic changes through remote sensing have practical significance for Africa.It is difficult to establish a complete time series database and make full use of the non-remote sensing data by a single remote sensing data source.With the aid of MODIS NDVI products(MOD13),a 15-year(2000-2014)lake surface area dataset of the middle part of the valley was obtained by a combined method of band ratio and threshold segmentation.In the same time series,based on the TRMM/3B43 product and the MODIS ET product(MOD16),information of rainfall and the terrestrial evapotranspiration was obtained.Further,their spatial distribution and their annual and interannual fluctuation were analyzed.Finally,the characteristic of the lake area variation was depicted and its relationship to the climatic factors such as the rainfall and the terrestrial evapotranspiration was discussed,respectively.The results revealed that the lakes with different area magnitude in the study area,occupying 80.64%of the total surface lake area across the African continent,had a tendency of decrease between 2000and 2014.The rainfall in the study area also decreased drastically in the same time series and had variation regularity which reflected the climate model in the East African Great Rift Valley.The comprehensive analysis of the annual variation of rainfall,terrestrial evapotranspiration and lake area demonstrated a coincident changing pattern which indicated the climatic factors had significant effect on the lake surface area change in the study area.