Extreme Weather Events Diagnostic Techniques

With global warming became more and more obvious, global climate change has received wide attention from the governments, scientists, and the public. Extreme climate events is a hot topic in the research work of climate change. Under the background of global warming, the extreme events of China shows characteristics such as high intensity and hard to predict. In this paper, using state of the art knowledge of statistics, we try to study the change of frequency and magnitude of extreme events of China from a brand new angle, and the main contents is summarized as follows:Using the daily precipitation data of 740 stations in China from 1960 to 2000, the analysis on the variations and distributions of the frequency and the percentage of extreme precipitation to the annual rainfall have been performed in this paper. Results indicate that the percentage of heavy rains (above 25mm/day) in the annual rainfall has increased, while on average the day number of heavy rains has slightly reduced during the past 40 years. In the end of 1970s and the beginning of 1980s, both the number of days with extreme precipitation and the percentage of extreme precipitation abruptly changed over China, especially in the northern China. By moving t test, the abrupt change year of extreme precipitation for each station and its spatial distribution over the whole country are also obtained. The abrupt change years concentrated in 1978-1982 for most regions of northern China while occurred at various stations in southern China in greatly different/diverse years. Besides the abrupt change years of extreme precipitation at part stations of Northwest China happened about 5 years later in comparison with that of the country's average.The very essence of extreme events detection is to separate abrupt climate events from its background. Traditional extreme events detection method sometimes would failed to detect all the extreme events when several extreme events clustered together because of the restriction of climatic background description method. Based on daily and monthly temperature records from 1957 to 2001 of China, using climate extremes detecting method based on average and median, we have analyzed the transition and distribution of frequency of temperature extremes. Results indicate that: while the average number of extreme high temperatures events per year has been increased in the past 50 years; the counterpart of extreme low temperature events decreased. Except for some stations which located in southwest China, the frequency of extreme low temperature events of most regions over China decreased. The spatial pattern of transition of frequency of extreme high temperature events showed southeast coastal regions decreased, while those inland regions increased.The frequency of record-breaking high temperatures became more and more under the background of intensifying global warming. What is the probability that current temperature record will be surpassed in the next few years? Answers are provided given a variety of simple statistical models for temperature. Annual global temperature is taken to be the realization of a random variable and a model describes the temperature random varible. The answers illustrate how record beaking is influenced by alternative model specifications. Estimates for the probabilty of a record are shown to range widely. If annual temperature is independent and identically distributed then a new record is unlikely. But probability increase rapidly if there is a rend or autocorrelation.