引发山洪的降水特征及动态阈值研究

Research on precipitation characteristics and dynamic critical thresholds of flash flood disasters

降水是引发山洪灾害的最重要因素。本文应用2013年1月到2016年12月的全国山洪灾害数据、土壤含水量数据及历史降水量数据,分析全国及典型区域(云、贵、川、渝)的山洪灾害及暴雨时空演变特征,初步探讨了基于灾害前期1 h、6 h、12 h及24 h的降水与山洪灾害的关系,研究山区流域的暴雨特征及引发灾害的静态阈值计算方法,同时,结合灾害前期不同时段土壤含水量特征,以10~50 cm深度的土壤含水量推求雨量动态阈值,建立了山洪灾害动态降水阈值推求方程。同时,选取了一定数量的样本对该方程进行检验,结果表明,当灾害发生前一段时间的土壤含水量信息确定时,检验样本中对应时段的实测降水量的值大于本文建立的方程计算出的动态降水临界阈值的样本占总样本的73%以上,是较优的阈值雨量的判别方法。

Serious harms of flash flood can be caused by heavy precipitation. Based on flash flood disasters, soil moisture information and the historical precipitation data, etc. from January 2013 to December 2016, the temporospatial distribution characteristics of precipitation and flash flood over China and in several typical regions (Yunnan Province, Guizhou Province, Sichuan Province and Chongqing Province) are pre liminarily analyzed. The method of static critical precipitation of flash flood is developed by establishing the relationship between flash flood events and precipitation at different time-scales (1 h, 6 h, 12 h and 24 h) before each flash flood event. Combined with the characteristics of the relative soil moisture in different layers (10~50 cm) at different time period before disasters, the dynamic critical precipitation threshold is calculated as well. The dynamic critical precipitation threshold model of flash flood in national and regional domains is proposed in the pa per. The results show that the models perform well, and the hit ratio of the dynamic model reaches 73%, which is better than the static model.