Evaluation of Satellite-Based Precipitation Estimates over an Agricultural Watershed of India

摘要

Due to lack of watershed-wide hydrologic data from in situ platforms, whether they are real time or historical, water management has been quite challenging. Benefiting from the high spatio-temporal resolution and near-global coverage, satellite-based precipitation estimates (SPEs) provide an opportunity to enhance present hydrologic prediction capability for a watershed or river basin. In the present study, three different SPEs i.e., PERSIANN, CMORPH, and TRMM precipitation datasets have been assessed over gauge-based India Meteorological Department (IMD) gridded dataset employing statistical and contingency table methods for an agriculture-based Marol watershed of India. The detailed evaluation has been carried out for the years 1998-2013 on daily, monthly, seasonal, and yearly basis with a spatial resolution of 0.25×0.25 latitude/longitude. SPEs performed reasonably well with the IMD gauge-based gridded dataset, although significant biases were observed. In general, the rainfall detection capabilities of SPEs were found better during monsoon season than non-monsoon season and annual time scale. A good correlation was found between watershed-averaged monthly precipitations of TRMM dataset against IMD gauge-based gridded dataset (R~2 = 0.77) as compared to CMORPH (R~2 = 0.63) and PERSIANN (R~2 = 0.56) datasets. Overall, performance of TRMM precipitation dataset was found superior over CMORPH and PERSIANN datasets, however, due to error and bias in SPEs considerable inter-annual and seasonal variations were observed. The analysis indicates that the use of SPEs can be a compensating approach after suitable bias correction and it has potential for hydrologic simulation in data-sparse regions.