Evaluation of Six Satellite Precipitation Products over the Chinese Mainland

Satellite precipitation products have been applied to many research fields due to their high spatial and temporal resolution. However, satellite inversion of precipitation is indirect, and different inversion algorithms limit the accuracy of the measurement results, which leads to great uncertainty. Therefore, it is of great significance to quantify and record the error characteristics of different satellite precipitation products for their better application in hydrology and other research fields. In this study, based on CN05.1, which is a set of site-based interpolation data, we evaluated the accuracies of the six satellite precipitation datasets (IMERG-E, IMERG-L, IMERG-F, GSMaP, CMORPH, and PERSIANN-CDR) at different temporal scales (daily, monthly, and yearly) in mainland China for the period from 2001 to 2015. The results were as follows: (1) In terms of mean precipitation, IMERG-F was superior to other data in all areas. IMERG products and PERANN-CDR performed better than other products at all scales and were more suitable for precipitation research in mainland China. Site correction can effectively improve the accuracy of product inversion, so IMERG-F was significantly better than IMERG-E and IMERG-L. (2) Except PERSIANN-CDR, all precipitation products underestimated precipitation in the range of 1-4 mm/day and had a high coincidence with CN05.1 in the range of 4-128 mm/day. (3) The performance of six types of satellite precipitation products in summer was better than that in winter. However, the error was larger in seasons with more precipitation. (4) In the Qinghai-Tibet Plateau, where there are few stations, the inversion of precipitation by satellite products is closer to the actual situation, which is noteworthy. These results help users understand the characteristics of these products and improve algorithms for future algorithm developers.