西南地区极端降水时空变化及概率分布特征研究

Spatiotemporal Characteristics Analysis and Statistical Distribution of Extreme Precipitation in the Southwest Region

摘要: 基于西南地区1961~2012年101个气象站的逐日降水资料，本文研究了该区极端降水的时空变化规律及其概率分布特征。结果表明：西南地区极端降水频数整体呈不明显的下降趋势，气候倾向率仅为−0.14次/10a，但极端降水量对总降水量的贡献率呈极显著增加趋势。西南地区极端降水尽管具有突变特征，但均没有通过显著性水平检验。此外，极端降水具有55a、35a、21a和13a的周期变化规律，且不同周期的振荡能量明显不同。极端降水高值区主要集中在云南西南部、贵州大部和四川盆地中部，其中四川省乐山站的降水量及其变差系数均为高值中心。广义极值GEV分布和Gamma函数分布均能较好地拟合西南地区的极端降水，但Gamma函数分布的拟合效果更优。

Abstract: Based on the daily precipitation data from 101 meteorological stations, this paper investigates the spatial and temporal characteristics of extreme precipitation and analyzes its probability distribution characteristics in Southwest China from 1961 to 2012. The results show that the overall frequency of extreme precipitation in the southwest region showed an insignificant downward trend, with a climate inclination rate of only −0.14 times per ten years, but the contribution rate of extreme precipitation to total precipitation showed a significant increasing trend. Extreme precipitation had mutation characteristics in the Southwest, but none of them had passed the significance level test. Extreme precipitation exhibited periodic variation patterns with cycles of 55a, 35a, 21a, and 13a, and the oscillation energies of different periods were significantly different. High values of extreme precipitation are mainly concentrated in the southwest of Yunnan, the majority of Guizhou, and the central Sichuan Basin, with Leshan station in Sichuan province being the center of high values for both precipitation amount and its coefficient of variation. The Generalized Extreme Value (GEV) distribution and the Gamma function distribution both exhibit excellent fitting performance for extreme precipitation in the southwestern region, with the Gamma function distribution demonstrating a superior fit.