长江流域降水极值时间序列的分布特征

Distribution feature of time series of extreme precipitation over the Yangtze River Basin

在1960-2005年长江流域147气象观测站汛期4-9月逐日降水资料基础上,通过计算逐站大于95th强降水及其间隔天数、小于1.27mm/d的持续天数,分析长江流域降水极值时间序列的时空分布特征,并建立概率分布模式.研究发现,长江上游四川盆地附近及中下游鄱阳湖流域东南部是汛期强降水中心,也是长江流域强降水最集中发生的地区.汛期降水强度小于1.27mm/d的天数,在上游干流、岷沱江流域、乌江上游地区为多.但此处干旱持续天数最短,干旱形式并不严重.而在金沙江上、下游,洞庭湖流域,鄱阳湖流域东南部支流及下游干流区干旱持续天数较长.长江流域大于95th强降水的间隔天数与小于1.27mm/d的干旱持续天数服从Weibull-Ⅱ型分布.分布参数变化的模式较准确的反映降水极值时间序列的时空变化特征.

Bascd on daily precipitation data of 147 stations in the Yangtze River Basin during the flood season （April to September） for 1960-2005, extreme high precipitation （percentile〉 95th） and its time gap, days with precipitation〈 1.27mm/d and its duration were calculated. Accordingly, spatial distributions and temporal sequences of extreme precipitation events had been analyzed, and the probability model was established. Research results showed that Sichuan Basin in the upper reaches and southeastern Lake Poyang Catchment in the mid-lower reaches were the high precipitation center of the Yangtze River Basin with centralized extreme event. Total number of days with precipitation〈 1.27mm/d is more in the mainstream section, Minjiang-Tuojiang Catchment and upper Wujiang Catchment than any other area of the Yangtze River Basin. However, due to shorter lasting period of events with〈 1.27mm/d, it did not create serious drought situation in theses areas. Whereas, areas in the upper and lower Jinshajiang（upper reaches of the Yangtze River） Catchment, Lake Dongting Catchment, southeastern Lake Poyang Catchment, and lower main stream section, were more likely to be stricken by prolonged events with〈 1.27mm/d and droughts caused. By applying Weibull model with the method of maximum likelihood estimator on the time gap of extreme high precipitation and lasting period of 〈 1.27mm/d, it was proved that Weibull-II can simulate well the probability distribution of timing of extreme precipitation events over the Yangtze River Basin. Changes ofWeibull-II parameter can reflect well the spatial and temporal variation of timing of extreme precipitation events.