基于水动力与模糊综合模型的洪水风险评价

Flood risk assessment based on hydrodynamic and fuzzy comprehensive model

针对传统洪水风险评价模型指标的单一性,不能多方面综合评价洪水风险的不足,创建多指标模糊综合评价模型(Multi-index Fuzzy Comprehensive Evaluation Model, MFCE)。该模型使用水动力耦合模型计算洪水淹没过程,利用GIS处理地形和社会影响因子,结合完整的多指标模糊综合评价体系评估研究区洪水风险等级。以江西省修水县为例,评价该区50a洪水风险。结果表明:(1)在传统的情景模拟法中,高风险区域集中在淹没深度大、淹没时间长的地区。由于洪水灾害给城区带来的损失远大于山体和林地,所以本文所建模型结果中的高风险区域主要集中在城区和公共建设密集的地区,提高了评价模型的准确性和可信度。(2)相比于传统模型,MFCE模型考虑了经济和复杂地形的影响,部分地区风险等级改变:传统模型评价为高风险的山体和林地(57.13%)和较高风险的农田(28.36%)风险等级降为低风险。城区周边较高风险地区(9.65%)提升至高风险。(3)改进模型模拟结果为高风险、较高风险、中等风险、较低风险和低风险面积分别为0.37 km^(2)、0.62 km^(2)、0.49 km^(2)、1.46 km^(2)和0.96 km^(2)。由此可见,所构模型可以提高洪水风险评估结果的全面性和可靠性,为洪水防洪减灾规划和灾害风险分析提供新的依据和方法。

In view of the single index of the traditional flood risk assessment model, it is unable to comprehensively evaluate the flood risk in many aspects. In this paper, a multi-index Fuzzy Comprehensive Evaluation Model(MFCE) is established. The model uses hydrodynamic coupling model to calculate the flood submergence process, uses GIS to process the terrain and social impact factors, and evaluates the flood risk level of the study area with a complete multi-index fuzzy comprehensive evaluation system.Taking Xiushui County of Jiangxi Province as an example, the 50 a flood risk is evaluated. The results show that:(1) In the traditional scenario simulation method, the high-risk areas are concentrated in the areas with large submergence depth and long submergence time. As flood damage to urban areas is much greater than mountain and forest land, the high-risk areas of the model built in this paper are mainly concentrated in urban areas and areas with dense public construction, which improves the accuracy and credibility of the evaluation model.(2) Compared with the traditional model, the MFCE model changes the risk level of some regions considering the impact of economy and complex terrain. The traditional model evaluates the risk level of mountain and forest land with high risk(57.13%) and farmland with relative high risk(28.36%), however, they are decreased to low risk in this model. Areas with relative high risk(9.65%) around urban areas increase to high risk.(3) The simulated results of the improved model show that the areas of high risk, relative high risk, medium risk, relative low risk and low risk are 0.37 km^(2), 0.62 km^(2), 0.49 km^(2), 1.46 km^(2) and 0.96 km^(2) respectively. It can be seen that the model can improve the comprehensiveness and reliability of flood risk assessment results, and provide a new basis and method for flood control planning and disaster risk analysis.