基于历史事件的暴雨洪涝灾害损失概率风险评估——以湖南省为例

A Probabilistic Approach of Assessing Rainstorm-induced Flood Loss Risk Based on Historical Event Records—Case Study in Hunan Province,China

在中国每年因暴雨洪涝灾害造成的经济损失非常惊人,洪涝灾害风险越来越被政府和保险公司所关注.本文基于湖南省1978—2007年的历史灾害事件记录,对暴雨洪涝灾害所造成的直接经济损失进行概率风险评估.通过蒙特卡洛模拟生成事件损失表,在此基础上模拟直接经济损失超越概率曲线,包括最大损失超越概率曲线(OEP曲线)和累积损失超越概率曲线(AEP曲线),最后应用OEP和AEP曲线计算不同超越概率下每年的最大可能损失与累积损失,以及年损失期望值.研究表明灾害造成的损失并不一定与致灾强度成正比,对于灾害频发的地区,AEP曲线能提供决策者更正确的信息.研究同时发现对于不同重现期,洪涝灾害造成损失风险的空间分布有差异.重现期为千年一遇时,益阳市的损失风险级别最高;重现期为百年一遇时,怀化市和郴州市的损失风险最大,总损失将超过60亿元;重现期为五十年一遇和十年一遇时,损失风险最大的也是怀化市.张家界市在各种重现期下的损失最轻.文中所得评估结果可作为风险图绘制的基础,政府和保险公司等决策者可据此获知面临暴雨洪涝灾害的直接经济损失风险,从而制定相应的风险管理和保险业务发展策略.

The economic losses caused by rainstorm floods are tremendous every year in China. Flood risk is increasingly being recognized as a concern in both government and insurance company. This article employs the probabilistic model to assess the risk of direct economic losses caused by rainstorm floods in Hunan province based on historical event records from 1978 to 2007. The Monte Carlo simulation is carried out to generate the flood event table and develop exceedance probability curves for flood losses, ：including Occurrence Loss Exceedance Probability（OEP） and Aggregate Loss Exceedance Probability （AEP）curves. The article finally utilizes these curves to calculate the Probable Maximum Loss（PML） and Total Losses at different return periods,and to estimate Average Annual Loss（AAL） of every city in Hunan province. The study reveals that the intensity of hazards is not necessarily proportional provide more accurate information for flood-prone to the magnitude of disasters. AEP curves can areas. Meanwhile, the results indicate that the spatial distribution of flood losses is nonuniform at different return periods. For the 1000-year return period,Yiyang will become the highest risk city of flood losses;for the 100-year return period, Huaihua and Chenzhou will have largest losses among all cities, the total losses will both exceed 6 billion RMB;for the 50-year and 10-year return periods, Huaihua will remain the highest risk city of flood losses. Conversely, Zhangjiajie is the city of lightest economic losses at various risk levels. The assessment results are the data foundation for risk mapping. It can be used as a reference for policy-makers to make effective flood risk management plan. It could also serve the insurance industry for business development strategies.