| Precipitation extremes have become more frequent and shown a continuing increase trend under global warming. Precipitation extremes may cause floods, debris flow and other nature hazards, which have a bad effect on people’s lives, the economic development and the protection of ecological environment. Because of the inadequate drainage facilities, the great deal of human ingenuity, the fast process of urbanization and the incomplete of land use, precipitation extremes have a worse influence in developing countries like China. This phenomenon attracted much attention of scholars at home and abroad, and became a hot point in research field.The precipitation data in 756 stations between 1960 and 2009 in this research were downloaded from “the China Meteorological Data Sharing Service System”, the 1km grid of GDP data and population density data in 2010 were downloaded from “Data Sharing Infrastructure of Earth System Science”, and the three levels of data on land cover types in China in 2009 were provided by USGS. Define the threshold of precipitation extremes is the start point of the precipitation extremes’ research. First, we used the Detrended Fluctuation Analysis(DFA) method, the percentile method and the Pearson-III probability distribution to define the threshold of precipitation extremes in each station. Through the comparison of three methods, DFA was selected to define the thresholds of precipitation extremes in China. And according to the actual precipitation and nature conditions the proposed thresholds of precipitation extremes in four natural regions(Northwest China, Qinghai-Tibet Region, North China and South China) were given. Based on the thresholds defined by DFA, spatial and temporal variations of extreme precipitation amount, frequency and intensity were analyzed. And the Mann-Kendall test was applied to observe the tendency variation of precipitation extremes. Besides, the risk model and Scenario Analysis method were chosen to study the risk of precipitation extremes. We chose the extreme precipitation amount and extreme precipitation frequency as two index, and precipitation extreme scenarios with return periods of 5, 10, 20, 50, 100 and 200 years were designed to analyze the hazard of precipitation extremes. Economy and population were chosen to analyze the vulnerability of precipitation extremes. At last, use the hazard of precipitation extremes and the vulnerability of precipitation extremes combine with the risk model, to analyze the risk of precipitation extremes, and finish the risk regionalization of precipitation extremes under different return periods in China. The results show that:(1) The thresholds defined by the 90 th percentile method and the 95 th percentile method are very similar, and they are much lower than the thresholds defined by the 99 th percentile method. The thresholds defined by the Pearson-III probability distribution were almost twice that of the thresholds defined by the 99 th percentile method. Thresholds calculated by the DFA method are much higher than that calculated by the percentile method, and very similar to the thresholds with the return periods of two years.(2) The spatial distribution of the thresholds of precipitation extremes are similar to the spatial distribution of annual precipitation. And the spatial distribution of the thresholds defined by DFA are similar to the spatial distribution of the thresholds with the return period of two years, and much similar to the spatial distribution of annual precipitation. The high values are all in the edge of Sichuan Basin, southern China and lower-and-middle section of Yangtze River.(3) Through the comparison of the DFA method, percentile method, Pearson-III probability distribution method, combined with the characteristics of the spatial distribution of annual precipitation, and considering the long term reaction of the climate system, this study suggested that: the thresholds defined by DFA method is better to understand the precipitation extremes on a national scale.(4) Considering the thresholds define by DFA, the annual precipitation, the precipitation frequency, the climate, the terrain and the drainage in each region, we suggested that the thresholds in four regions should be as follows: in South China 80 mm is appropriate, in North China should be 50 mm, in the Qinghai-Tibet region 25 mm and in Northwest China 20 mm.(5) The temporal variations in four regions are: the extreme precipitation amount in Qinghai-Tibet region, South China and whole China have shown an increasing tendency since the 70 s, Northwest China and North China shown fluctuation changes; the extreme precipitation frequency have shown an up-going tendency in the Qinghai-Tibet region and South China since the 70 s, while shown volatile changes in North China, Northwest China and all over the country; the extreme precipitation intensity have shown an increasing fluctuation since the 80 s in the Qinghai-Tibet region, North China and South China, while have had a decreasing tendency in Northwest China since 2000.(6) Through the MK test, the extreme precipitation amount, frequency and intensity of most stations in South China, the Qinghai-Tibet region and Northwest China had an increasing tendency, while three indexes in North China had a decreasing trend. But most trend in each station is insignificant, among four regions, the vary tendency of stations in the Qinghai-Tibet region is the most significant.(7) The amount and intensity of precipitation extremes decreasing from the southeastern coastal areas to northwestern inlands, the values in southern China and the lower-and-middle section of Yangtze River are higher, while in Northwest China and the north of Qinghai-Tibet region are lower. The frequency of precipitation extremes is randomly distributed, high value mainly shows in the upper reaches of the Yellow River, the middle and lower reaches of Yangtze River, Sichuan Basin, Hainan province, and the junction of Heilongjiang and Inner Mongolia. Low value mainly located in the bulge of the Yellow River, south of Northwest China and southwest of the Qinghai-Tibet region.(8) Under each return period, the hazard level of extreme precipitation amount decreasing from the southeastern coastal areas to northwestern inlands, but the hazard level of extreme precipitation frequency is randomly distributed. The highest-risk areas are southern China and the middle and lower reaches of Yangtze River. The higher-risk areas are North China and South China. The hazard level of extreme precipitation amount in Northwest China and north of the Qinghai-Tibet region is the lowest. The hazard level of extreme precipitation frequency is higher in the north of 43 degrees north, Yangtze River and part of the southern China. The comprehensive hazard level is high in southern China and the middle and lower of Yangtze River under every return periods.(9) The comprehensive, economy, population and land use extreme precipitation vulnerability are all higher in the east and lower in the west. The vulnerability in North China Plain, Yangtze River Delta region, costal region and capital cities like Beijing, Shanghai, Guangzhou, Wuhan, and Chongqing are higher than other regions, but much lower in the west of Northwest China and Qinghai-Tibet Region.(10) Under each return period, the extreme precipitation risk level decreasing from southeastern coastal areas to northwestern inlands. The higher risk regions are mainly in the Yangtze River Delta region, North China Plain, Sichuan Basin, southern China and costal region, the lower risk regions are located in the west of the northwest China and northwestern Qinghai-Tibet region. |
PDF Full Text Request