Satellite-based estimation and validation of monthly rainfall distribution over Upper Ganga river basin

Abstract. Water is one of the most precious natural resources for all living flora and fauna. 97.5% of water on the Earth is sea water, the remaining 2.5 % is fresh water of which slightly over two thirds is frozen in glaciers and polar ice caps. The unfrozen fresh water is mainly found as groundwater, with only a small fraction present above ground or in the air. Since one of the main source of water is rainfall. Therefore, proper information on rainfall and its variability in space and time is required for better watershed planning and management and other applications. In Himalayan basin, the rain gauge network is relatively sparse with uneven distribution. Hence, there is lack of proper information on rainfall patterns of this region. The main advantage of satellite derived rainfall estimation over rain gauge derived rain data is that these provide homogenous spatio-temporal rainfall information over a large area e.g. Upper Ganga river basin region. Therefore, a better understanding of the rainfall patterns of this region is required for better disaster mitigation. The objectives of this study are to evaluate the reliability of Tropical Rainfall Measuring Mission (TRMM) 3B43 V7 derived high resolution satellite product to study the rainfall distribution over the Upper Ganga river basin. TRMM 3B43 V7 derived monthly rainfall data is analyzed and the monthly rainfall product is validated and correlated with IMD (Indian Meteorological Department) gauge station's rainfall data. The monthly rainfall data of 15 years i.e. from 1998 to 2012 is used in the study. Statistical indices can be used to evaluate, compare and validate satellite rainfall data with respect to gauge rainfall data. Statistical indices used in this study are Correlation Coefficient (r), Mean Square Error (MSE), Root Mean Square Error (RMSE), and Average Percentage Error (Avg. % Error). Most of the rainfall in the study area occurs in the months of June, July, August, September and October. The isohyets were prepared using gauge rainfall data and are matched with the spatially distributed rainfall surface prepared from TRMM satellite data for all the months of the rainy season of the study area. Kriging spatial interpolation method was used to generate the spatially distributed rainfall surface. From the results it was observe d that they matched fairly well with each other showing high spatial correlation. The monthly rainfall result showed that TRMM data is underestimated with low accuracy, though TRMM data and rain gauge data have positive correlation.