Traditional flood season obtained by statistically analyzing long series of hydrologic data lacks of the consideration of factors such as climate change and extreme weather.To this end,based on the meteorological data of precipitation p,average temperature T and sunshine time data SD from 1962 to 2008 in Xiaoyi City,the Mann-Kendall method is firstly used to determining the years of mutation in each climatic element,and then taking the year of mutation as cut-off point,the optimal partition of Fisher method combined with PCA method is used to determine the index weight for staging the flood season.The flood season of Zhangjiazhuang Reservoir in Xiaoyi City is staged by above procedure,and the results are compared with that of traditional staging methods.The results show that,(a) the optimal partition of PCA-Fisher method has good applicability in flood season staging;and (b) climate change has a significant impact on the stage results of flood season,and the main flood season is shorter than the series before the climate change year.%传统汛期分期通过对长系列的降雨、径流资料的统计分析得到,缺乏对气候变化和极端天气等因素的考虑.为此,对1962年～2008年孝义市降水量p、平均气温T和日照时数SD等气象因子的年数据序列,采用Mann-Kendall法来检测各气象因子下的气候突变年,以此突变年为分界点,再采用Fisher最优分割法,结合PCA法确定所选取的能够反映汛期水库流域洪水特征的各项指标权重,对孝义市的张家庄水库进行汛期分期,并与传统的分期方法进行对比的结果表明,PCA-Fisher最优分割法在汛期分期中有较高的适用性;气候变化对汛期分期结果具有重大影响,降水量减少,地区的气候突变年后主汛期较突变年之前短.
展开▼
