The relationship between regional precipitation change and warming is an important open issue in climate change physical science. Because precipitation in China has strong sensitivity to warming, quantitative assessment and projection on the responses of precipitation and its extremes in a warming world are crucial for better understanding of regional climate change and helpful for regional adaption to climate change. For this reason, based on simulations of 24 models from Coupled Model Intercomparison Project Phase 5 (CMIP5), this study assesses the ability of the models in simulating the responses of annual mean precipitation and its extremes to warming over China and its subregions, and then projects their change under the RCP4. 5 and RCP8. 5 scenarios that represent respectively a medium-low and high radiative forcing. The annual mean precipitation is defined as the total amount of precipitation from January to December. The precipitation extremes are measured by the R95p (very wet days) and R99p (extremely wet days) indices, which are defined by the Expert Team on Climate Change Detection and Indices (ETCCDI). According to the definition of ETCCDI, the R95p and R99p refer to annual total precipitation when the daily precipitation exceeds the 95th and the 99th percentile of the wet day precipitation, respectively. Eight subregions determined by administrative boundaries and societal and geographical conditions, i. e., NEC(Northeast China), NC(North China), EC (East China), CC (Central China), SC (South China), SWC1 (Tibetan Plateau), SWC2(Southwest China), and NWC(Northwest China), are used in this study. The model performance is validated through the comparison for the time period from 1961 to 2005 between the historical simulation and the gridding observation dataset with a horizontal resolution of 0. 25 degrees X 0. 25 degrees in latitude and longitude. Quantitative analysis shows that the CMIP5 multi-model ensemble (MME) can generally capture the spatial features of the temperature, mean precipitation and precipitation extremes as well as the relationship of precipitation and its extremes with temperature over China. However, it underestimates the response of mean precipitation while overestimates the response of precipitation extremes over China region in historical period. The CMIP5 MME also has some abilities in reproducing the responses of the mean precipitation and its extremes to the warming over the subregions of China, and better performance can be found for the precipitation extremes. Under the RCP4. 5 and RCP8. 5 scenarios, concurrent with the temperature rising, the mean precipitation and precipitation extremes are projected to increase consistently over China. As the regional mean temperature rises by 1 degrees C, the mean precipitation will increase by 3. 5% and 2. 4%, and the R95p will increase by 8. 0% and 11., respectively. The response of R99p is much more sensitive, respectively with an increase of 15. 3% and 21. 6%. For the subregions of China, they all show positive response and the regional difference will decrease in the future. Moreover, the sensitivity of the precipitation extremes to the warming is higher than that of the mean precipitation. The stronger the precipitation extreme is, the higher sensitivity it will have. Besides, the response of the mean precipitation to the warming is larger in Northern China than in Southern China. The largest increases in R95p and R99p are projected in the Tibetan Plateau and Southwest China, indicating an increasing risk of heavy rainfall and floods.
