基于情景分析和突变理论的洞庭湖蓄滞洪区洪水风险评估研究

With the pushing of China's urbanization process, the flood storage and detention areas as an important part of flood control system face the embarrassment of maintaining flood control function and development. It would lead to a heavy losses to use flood storage and detention area in a wrong way. Because the flood storage and detention area has already been the weakest link of China's current flood control system construction, it becomes particularly important to evaluate the flood risk of flood storage and detention area. For now, the research of the flood storage and detention area's flood risk evaluation usually concentrates more on model simulation than index analysis. Because of existing strong subjectivity and many qualitative components, the accuracy and rationality of evaluation result need to be improved. In this project, three large flood storage and detention areas, Gongshuangcha, east Datong lake and Qianliang lake of Dongting lake, are chosen as experimental zones to carry out quantitative evaluation research of flood risk on refined scale and explore to build risk evaluation index system from using risk factors selected from three aspects: disaster-causing factor, hazard-formative environment and hazard-affected body based on space measure. High-precision flood routing model is built to evaluate the flood risk quantitatively and research of flood scenario simulation and analysis is developed by integrating high-resolution remote sensing technology. This project try to evaluate the fuzzy membership degree of flood risk index to solve the uncertain problem of risk factors' optimum sorting based on catastrophe theory. This research develops particular study on flood risk refined evaluation of flood storage and detention area from the view of integration of the knowledge of various subjects and expect to acquire the breakthrough of this difficult point.

随着我国城市化进程的推进，作为防洪体系中重要组成部分的蓄滞洪区面临着保持防洪功能和开发建设之间的尴尬境地。蓄滞洪区一旦运用损失严重，业已成为当前我国江河防洪体系建设中最为薄弱的环节,开展蓄滞洪区洪水风险的评估尤为重要。目前，蓄滞洪区洪水风险评估的研究多“重模型模拟，轻指标分析”，评估结果主观性强、定性成分多，准确性、合理性有待提高。本项目选择洞庭湖共双茶、大通湖东、钱粮湖三个大型蓄滞洪区作为试验区，开展精细尺度下的洪水风险定量评估研究，以空间可度量为原则从致灾因子、孕灾环境、承灾体三个方面遴选风险因子，探索构建风险评估指标体系；通过集成高分辨率遥感技术，建立高精度洪水演进模型，定量评估洪灾损失，开展洪水情景模拟与分析研究；基于突变理论，评判洪水风险指标模糊隶属度，解决风险因子优选排序的不确定性问题。本研究从学科知识集成的角度对蓄滞洪区洪水风险精细评估开展专门研究，以期对该难点有所突破。

蓄滞洪区作为一种特殊的洪水风险区域，评估其洪水风险可以明确蓄滞洪区内不同区域居民避洪安置方向和安全建设模式，亦可规划蓄滞洪区工程设施和安全设施建设方案，为洪水风险管理政策和措施的拟定奠定基础。在当前蓄滞洪区开发程度日益加剧，行蓄洪风险不断增强的大背景下，开展蓄滞洪区洪水风险评估的研究具有重要的理论和实际意义。. 本研究以洞庭湖蓄滞洪区作为实验区，通过采集研究区机载LiDAR点云和高分辨率航空遥感数据，从洪水灾害孕灾承灾信息提取、高精细洪水淹没模拟模型构建、单体建筑物洪水灾害损失评估、蓄滞洪区洪水灾害风险综合评估四个方面展开研究，通过建立一套蓄滞洪区洪水灾害风险评估的技术与方法，期望能够完善和丰富高精度、定量的洪水灾害风险评估理论，同时能为蓄滞洪区洪水灾害风险管理政策的制定和实施提供技术支撑。主要研究内容如下：. （1）基于机载LiDAR数据的蓄滞洪区洪水灾害孕灾承灾信息提取。提出了一种适用于蓄滞洪区等广大农村地区的基于三维分形维数分析的建筑物信息提取方法；提出一种基于高分辨率数字地形数据的蓄滞洪区农田、田埂和沟渠信息提取方法；提出一种基于高分辨率数字地形数据的堤防中心线和堤顶堤坡参数提取方法，并用于堤防破坏程度分析。.（2）集成高精细数字地形数据的蓄滞洪区洪水淹没模拟技术与方法研究。提出一种高精细数字地形数据下的集成分块压缩追踪法和二维洪水演进模型的洪水淹没模拟方法，通过对二维洪水演进模型执行空间插值和分块压缩追踪，生成高精细洪水淹没模拟计算结果。.（3）蓄滞洪区单体建筑物洪水灾害损失评估。本研究参照《住宅设计规范(GB50096-2011)》国家标准，基于建筑物高度信息将建筑物类型分为砖木结构建筑物和砖混结构建筑物两类。通过分别构建洪灾损失率曲线，再基于单个建筑物洪水淹没水深，计算了蓄滞洪区内全部建筑物的洪灾损失值。.（4）基于突变理论的蓄滞洪区洪水灾害风险综合评估研究。尝试引入突变理论，按照突变级数评价法开展蓄滞洪区洪水灾害风险评估的研究。