气候变化对江西省主汛期降雨极值时空差异的影响研究

Study on the impact of climate change on the extreme rainfall distribution in main flood season of Jiangxi Province

【目的】为了研究气候变化对江西省主汛期降雨极值分布的影响,【方法】以江西省1960—2017年的逐日降雨资料为基础,采用多种检验方法检测突变点,并对突变点的显著性进行检验,使用Fisher最优分割法对气候突变前后的汛期进行分期及确定主汛期时间域,并对气候突变前后主汛期降雨极值分布与设计降雨进行探讨。【结果】结果表明:(1)1960—2017年江西省降雨量突变年为1991年。(2)气候突变后最大一日降雨量、暴雨天数及大雨天数普遍较气候突变年前大,尤其在广昌地区最大一日降雨量显著增加,增幅可达到228 mm,广昌-樟树-南昌-玉山线以西区域内暴雨天数显著增加,波阳-南昌-修水线以北所在区域内大雨天数显著增加,增幅都可达到17 d。(3)气候突变前后韦布尔分布对降雨极值拟合最优,概率点距相关系数与和确定系数分别达到了0.994 8与0.988 9,且韦布尔分布得到的降雨极值大于P-Ⅲ分布,气候突变后的主汛期设计极端降雨值小于气候突变前主汛期设计极端降雨值。【结论】在对江西省主汛期降雨进行设计时,要注意对洪水的防范。

[Objective] The goal of the study is to get the influence of climate change on the distribution of rainfall extremes in the main flood season in Jiangxi Province. [Methods]Based on the daily rainfall data of Jiangxi Province from 1960 to 2017, a variety of test methods are used to detect the mutation point, and the significance of the mutation point is tested. The Fisher optimal segmentation method is used to stage the flood season of climate abrupt change and determine the time domain of main flood season, with discussing the distribution of extreme rainfall during the main flood season with abrupt climate change and the designed rainfall. [Results] The results show that:(1) 1991 is the abrupt change year of precipitation in Jiangxi Province during 1960—2017.(2) The maximum daily rainfall, heavy rain days and heavy rain days after the abrupt climate change are generally larger than those before the abrupt climate change, especially in the Guangchang area, with an significantl increase of 228 mm. Guangchang-Zhangshu-Nanchang-Yushan line with the number of heavy rain days in the west area increased significantly, and the number of heavy rain days in the area north of the Boyang-Nanchang-Xiushui line increases significantly, with an increase of 17 days.(3) Weibull distribution is the best fitting to rainfall extremes before and after climate abrupt change, and the correlation coefficient and determination coefficient of probability point distance reach 0.994 8 and 0.988 9 respectively, and the rainfall extremes obtained by Weibull distribution are greater than that of P-Ⅲ distribution, and the designed extreme rainfall values in main flood season after climate abrupt change are smaller than that before climate abrupt change. [Conclusion] The prevention of flood should get attention when it is designed in the main flood season in Jiangxi Province.