Spatial and Temporal Variations of Precipitation Extremes and Seasonality over China from 1961~2013

Using the 0.5° × 0.5° gridded Chinese ground precipitation dataset from 1961~2013, spatial and temporal variations in precipitation extremes, total precipitation, the seasonality of precipitation and their linkages in the context of climate change are investigated using the Mann-Kendall trend test, Pettitt change-point test and correlation analysis. The investigation focuses on four extreme indices, i.e., the annual maximum number of consecutive dry days (CDD), the annual maximum number of consecutive wet days (CWD), the annual total precipitation when daily precipitation is greater than 95th percentile (R95pTOT), and the maximum 1-day precipitation (RX1day). The results show that precipitation extremes increased in northwestern China, especially Xinjiang, Tibet and Qinghai (CWD, R95pTOT and RX1day), and scattered parts of southeastern China (R95pTOT and RX1day), but decreased over considerable parts of southwestern China (CWD) and some small parts of northern China (CWD, R95pTOT and RX1day); the spatial patterns of the trends in precipitation extremes and that of total precipitation exhibit considerable similarity over China, which indicates the close relationship between changes in precipitation extremes and total precipitation; change points are detected in different periods ranging from early 1970s to early 2000s for different regions and extreme precipitation indices, and the spatial patterns of the abrupt changes of extreme indices are similar to those of the trends in extreme indices; the concentration index (CI) is strongly positively correlated with R95pTOT and RX1day in most areas in northern China (from the northeast to the northwest) and southwestern China (including Sichuan, Chongqing Guizhou and Guangxi), which means for these regions, the temporal heterogeneity of daily precipitation over a year is dominated by heavy rainfall amounts. The seasonality index of precipitation (SI) is positively related to R95pTOT and RX1day over most areas above 30° N, indicating that heavy precipitation events have a better chance to occur in places with a strong seasonal variation in annual precipitation in these areas, but for most areas below 30° N, the positive relationship is not significant.