基于复合极值模型的辽宁省极端降水重现期研究  

Study on the return period of extreme precipitation in Liaoning Province based on the Compound Extreme Value Model

作　　者：程攀 杨萌 孙虹雨 张燕[5] 肖光梁 朱宪龙 CHENG Pan;YANG Meng;SUN Hongyu;ZHANG Yan;XIAO Guangliang;ZHU Xianlong(Chengdu University of Information Technology,Chengdu 610225;Liaoning Meteorological Disaster Monitoring and Early Warning Centre,Shenyang 110166;Key Opening Laboratory for Northeast China Cold Vortex Research,Shenyang 110166;Liaoning Meteorological Observatory,Shenyang 110166;Liangshan Yi Autonomous Prefecture Meteorological Bureau of Sichuan Province,Liangshan 615000)

机构地区：[1]成都信息工程大学大气科学学院,成都610225 [2]辽宁省气象灾害监测预警中心,沈阳110166 [3]东北冷涡研究重点开放实验室,沈阳110166 [4]辽宁省气象台,沈阳110166 [5]四川省凉山州气象局,凉山615000

出　　处：《暴雨灾害》2022年第6期662-670,共9页Torrential Rain and Disasters

基　　金：辽宁省气象局科研项目(ZD202219,201902,BA201909)。

摘　　要：极端降水重现期的研究可提升对降水规律性的认识,对预防城市内涝、农田渍害等有重要意义。利用辽宁省14个国家气象观测站1960—2020年5—10月的逐日降水资料,选取第98个百分位值作为极端降水指标,采用泊松-广义帕累托复合极值模型对辽宁省极端降水日数和极端日降水量进行拟合并计算其重现期降水量。结果表明:辽宁省极端降水日数和极端日降水量分别服从泊松分布和广义帕累托分布(均通过α=0.05的显著性检验);辽宁省极端日降水量从西北向东南方向逐渐增多,其中丹东市的极端日降水量和极端降水日数均为最高;泊松-广义帕累托复合极值模型和经验频率法计算的重现期降水量拟合度较高,不同重现期下辽宁省东南部极端日降水量值均为最高,辽河流域中部极端日降水量较小,西北部极端日降水量最小,辽宁东部山区和辽宁西部丘陵地带极端日降水量高于辽宁中部平原区,且不同重现期分布状态一致。The study on the return period of extreme precipitation can improve the understanding of precipitation regularity, which is of great significance for preventing urban waterlogging, farmland waterlogging. To study the characteristics of the probability distribution of extreme precipitation in Liaoning Province, we use the daily precipitation data from 14 national meteorological observational stations in Liaoning Province during May to October of 1960-2020, and select the 98th percentile as the threshold of extreme precipitation. The Poisson-generalized Pareto compound extreme value model fits the extreme precipitation days and extreme precipitation amount and estimates the precipitation amount during the return period in Liaoning Province. The results show that the days with extreme precipitation and the extreme precipitation amount in Liaoning Province follow the Poisson distribution and the generalized Pareto distribution, respectively(both passed the statistically significant test of α =0.05). Extreme precipitation shows a gradual increase from the northwest to the southeast in Liaoning Province. The amount of extreme precipitation and the days of extreme precipitation are highest in the Dandong city. The Poisson-generalized Pareto compound extreme value model and the empirical frequency method can fit the precipitation amount during the return period very well. Under different return periods, the extreme precipitation amount in southeastern Liaoning Province is the highest.The extreme precipitation amount is relatively small in the middle of Liaohe River Basin, and is the smallest in the Northwest. The extreme precipitation amount in the mountainous areas of eastern Liaoning and the hilly areas of western Liaoning is higher than that in the plains of central Liaoning, and this spatial distribution is consistent for different return periods.

关 键 词：极端降水 泊松分布 广义帕累托分布 复合极值模型 经验频率法 

分 类 号：P468.024[天文地球—大气科学及气象学]