Evaluation of Global Gridded Precipitation Datasets Accuracy over Urmia Lake Basin, Iran

The present study aims to analyze the accuracy of four global gridded precipitation datasets namely CRU, GPCC, PCDR, DEL over Lake Urmia basin, Iran. To this end, observed data have been collected from six synoptic stations namely Urmia, Mahabad, Takab, Tabriz, Maragheh and Saqqez over a period of 30 years (1984-2013). Evaluation was done based on Nash-Sutcliffe Efficiency (NSE), correlation coefficient (CC), Root Mean Square Error (RMSE) and Bias. Results on all scales (monthly, seasonal and annual) showed the proper performance of the GPCC. For example, in annual scale, GPCC, outperformed other databases with NSE, CC and RMSE values of 0.87, 0.94, and 31.12 mm respectively. According to the results, performance of the CRU is acceptable. This dataset has been performed better than the GPCC, based on Bias. Also PCDR''s and DEL''s performance were unacceptable in all time scales.