Abstract: Based on the data of hail, gale, thunderstorm and lightning of 2481 weather stations in China from 1961 to 2016, the spatial and temporal distribution characteristics, periodicity and climate abruption characteristics of four kinds of severe convective weather in China were analyzed by various mathematical statistics methods. The results show that: In time, the days of four kinds of severe convective weather in China are decreasing, and hail and thunderstorm days are characterized by "first increasing and then decreasing". Hail, gale, thunderstorm and lightning days respectively present oscillation cycles of 35 years, 23 years, 12 years and 14 years in China, and the hail and thunderstorm days mutated in 2002 and 1992. In space, the QinghaiTibet Plateau and western Sichuan are the high value distribution areas of the number of hail, gale and thunderstorm days. The South of the Yangtze River are also the high value distribution areas of the number of thunderstorms. Southern China and the southwest region with the same latitude are the high value distribution areas of the number of lightning days. In terms of the trend, the number of hail days is decreasing mainly in the QinghaiTibet Plateau.The trend of gale days is in the spatial pattern of a decrease in the east region, invariability in the central region, and an increase alternating with  a decrease in the west region. The number of thunderstorms is increasing in Tibet, north China, Chongqing, Zhejiang and the northwest of Heilongjiang and that  of lightning days is decreasing in the south of the Yangtze River. In terms of the fluctuation, hail days fluctuates markedly in the southeast and gale days fluctuates wildly in the east of Hu Huanyong line. The fluctuation characteristics of thunderstorms days are large in the northwest and small in the southeast . In addition to the small fluctuation in the Northern Xinjiang and Southern China, lightning days fluctuates widely  in other regions in China. 

Key words: disaster weather, hailgalethunderstormlightning, spatiotemporal pattern, characteristic of periodic oscillation, mutation test