Key points are not available for this paper at this time.
Arid and semiarid areas account for more than half of China, have fragile ecological environments, are sensitive to global climate change and human activities. Due to the advantages of wide coverage and high resolution, multi-sources remote sensing precipitation products play an important role in monitoring precipitation in areas where rainfall gauges are scarce. Therefore, evaluating the performance of different precipitation products becomes very important. Here, the annual and daily average precipitation data from different precipitation products in China were analyzed from 2000 to 2020. Nine precipitation datasets are included: two reanalysis datasets and seven remote sensing datasets. The results show that CHIRPS (Climate Hazards group Infrared Precipitation with Stations) is the best product for precipitation in arid and semiarid China, and the mean annual precipitation correlation coefficient between CHIRPS and observed data is 0.82. CPC (CPC Global Unified Gauge-Based Analysis of Daily Precipitation) shows less dispersion and deviation in the daily precipitation, and the correlation coefficient between CPC and CN05 (observation data) daily precipitation is 0.92. In addition, the performance of precipitation products is tailored to local conditions, with MSWEP (Multi-source weighted-Ensemble Precipitation) evaluating precipitation poorly in Northwestern China but better in the areas with more precipitation. Extreme precipitation in China has shown an increasing trend in the last 20 years, with a very significant increasing trend in extreme precipitation in semi-arid areas and a constant trend in extreme precipitation in arid areas. The PERSIANN (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks) is the best product for extreme precipitation in arid and semiarid China.
Liu et al. (Fri,) studied this question.