Persistent extreme precipitation covering a large area usually causes severe flooding disasters in China, but how to depict it and what are the possible causes are still open questions. With Climate Prediction Center global unified gauge-based analysis of daily precipitation and NCEP/NCAR daily reanalysis dataset from 1979 to 2019, summer regional pentad extreme precipitation (RPEP) is defined according to the threshold of the 95th percentile of pentad precipitation with more than 5% land grids coverage in eastern China. While the definition of RPEP highlights the climate features of both the persistence and the regionality of extreme precipitation, it is distinctly different from the previous definitions that mainly reflect the synoptic aspects with daily data and have strictly temporal-spatial constraints. Four categories of RPEPs are objectively identified by K-means cluster analysis, i.e., South China (SC), South of Yangtze River (SYR), Jianghuai River (JHR), and North China (NC). Along the Yangtze River (SYR and JHR), intensity and area of RPEP are positively correlated with each other, and with the increase of RPEP intensity, its center of gravity tends to move eastward in all the four cluster regions and southward in Jianghuai River and North China, respectively, and vice versa. The RPEPs mostly persist for one pentad but can reach up to two to three pentads at most, and along with the duration of RPEP, its intensity and area are both enhanced accordingly. Furthermore, the frequency of RPEP increased significantly since the late 1990s in SYR, JHR, and SC. Associated with RPEP, strong pentad-mean convergence and ascending motion occur in the middle-lower troposphere, and except for SC that is dominated by the local low-pressure and cyclone anomalies, the other three cluster regions are all forced by the western Pacific subtropical high to the southeast and weak low-pressure trough to the north, and the low-level anticyclone anomaly to the southeast transports abundant water vapors to the RPEP regions accordingly. Besides, all the RPEPs are closely in accordance with obvious subseasonal oscillations, especially the 10–30-day and 30–60-day oscillations, which can be regarded as the potential sources of RPEP predictability in eastern China.
Read full abstract