Studies On The Variation Characteristics And Experimentation With Simulation Of East-China's Extreme Precipitation

Based on the 210 stations data of daily precipitation from 1953～2002 over Eastern China. Bring in an effective parameter estimation method, probability distribution model is used to fit the extreme precipitation, then analysis it's temporal-spatial distribution and climatic character. Taking the simulation tests of extreme precipitation to now and future climate by using Monte Carlo Method ,to find some change rules.12 regions are divided by REOF. Confirm the best sampling time length of extreme event, and found that based on one day precipitation one to two day precipitation could represent the extent of main quantity in one process of extreme event, and then analysis the climatic characterics of extreme precipitation during the time length. The asymptotic distribution is fitted using the Gumbel Distributions, L-moment is an effective method to parameter estimation, with which the fitting precision are improved. Define the region extreme-value. The results show that, as a whole, the trend of extreme precipitation over Eastern China changes little, but large difference is exsit between different ages and different regions. The spatial distribution of current extreme precipitation and extreme value with a return period indicate that extreme precipitation decreases from southeast to northwest, there is also a high value in Two-lakes basin, HuangHai bay and Liaotung peninsula.On the backfround of globle warming, Monte-carlo Method is used to simulate the character of extreme precipitation, the result showed that it is feasible to predict the future character by change the mean value.