高州市不同历时降雨组合的风险概率研究

Analysis of Risk Probability of Different Duration Rainfall in Gaozhou City

基于目前水文频率分析中的不确定性问题,以高州市马贵站1965-2016年的降雨资料为例,利用P-Ⅲ型曲线对高州市年最大1 h降雨量与相应各历时下降雨量进行了频率分析;构建年最大1 h与相应6 h、12 h、24 h降雨量的联合分布模型,并计算三种组合下的两种风险率大小。结论指出:条件概率P2均明显大于条件概率P1;各降雨组合的P1和P2随年最大1 h降雨量的增加而增大,即当H1为117.45 mm取得最大值,其中P1(H>H6|H<H1)与P2(H>H6|H>H1)达到最大值48.5%和71.2%。在设计重现期增至10年及以上时,其所对应的风险率均显著降低至20%以下。因此在推求设计暴雨和设计洪水时,利用Copula函数估算不同降雨组合的风险概率,对于准确合理预测遭遇强降雨事件,准确计算设计暴雨值,保证居民财产安全有着重要的意义。

Based on the rainfall data from 1965 to 2016 in Maogui station of Gaozhou, the P-Ⅲ curve is used to analyze the frequency of the maximum 1 h rainfall and the corresponding daily rainfall in Gaozhou. The annual distribution model of 1 h and the corresponding 6 h, 12 h, 24 h rainfall distribution model was established. We also calculated the two kinds of risk ratio of three combinations. We have got the conclusion that the conditional probability P2 is significantly larger than the conditional probability P1; The value of P1 and P2 of each rainfall combination increased with the increase of the maximum rainfall of 1 h,P1（H〉 H6 ｜ H〈 H1） and P2（H 〉H6 ｜ H〉 H1）reached the maximum value of 48. 5% and 71.2%. When the return period is increased to 10 years and above, the corresponding risk rates significantly reduced to less than 20%. Therefore,it is important to estimate the risk probability of different rainfall combinations by using Copula function to estimate the storm value and ensure the safety of residents＇ property.