Evaluation of rainfall datasets from CHIRPS and WorldClim in Meghalaya, India

Limited ground-based rainfall information has posed significant obstacles to hydrological, agricultural and related studies in Meghalaya, India.In this study, rainfall products were validated against ground-based rain gauge data from the India Meteorological Department, over three seasons.Both qualitative categorical metrics -probability of detection, false alarm rate, critical success index and frequency bias index, as well as quantitative statistical measures including Pearson correlation coefficient (r), Nash-Sutcliffe efficiency, relative bias, RMSE-observations standard deviation ratio, were computed to evaluate rainfall from CHIRPS and WorldClim.The quantitative statistics showed that the total rainfall estimated by CHIRPS provided better estimates than WorldClim during pre-monsoon (February-May) and post-monsoon (October-January).Neither dataset could provide good estimates of rainfall during the monsoon season (June-September).From this study, we found that CHIRPS is a reliable alternate source of rainfall data for Meghalaya with a limited rain gauge network.