城市景观格局与自然灾害生态风险研究——以深圳市为例

Taking Shenzhen City as a case study, with a special focus on such natural disasters as typhoon, flood, soil and water loss, landslide, and debris flow, this project aims to make a full understanding of the correlation between urban landscape patterns and natural disasters, and to quantify and simulate ecological risk associated with natural disasters. With the application of MATLAB software, RS, GIS and Spatial modeling technologies, the influencing mechanism of urban landscape patterns on natural disasters is fully discussed. In details, urban landscape patterns are quantified with landscape types, patch area, and landscape metrics. The methods of correlation analysis and mutation inflexion-point are applied to quantifying the spatial correlation between urban landscape patterns and natural disasters, and furthermore to make clear the correlation thresholds at various spatial scales. The GWR (Geographical Weighted Regression) model is also used to analyze the spatial non-stationarity of spatiotemporal relations between urban landscape patterns and natural disasters. Based on the integrated evaluation model for ecological risk associated with urban landscape patterns and natural disasters, the spatial differentiation of ecological risk in the study area is discussed, together with uncertainty analysis. Different scenarios on social and economic development are also applied to simulating urban landscape change and associated ecological risks, so as to put forward the spatial pattern optimization for natural disaster prevention and ecological risk mitigation. Therefore, the application of this project will be helpful to the development of landscape ecology in China, to expending the research areas of urban landscape ecology and ecological risk, to enrich the landscape approach for the research on couple human and nature system, and to fulfill the practical demanding for urban disaster prevention and ecological civilization construction. In sum, it is of great scientific and practical importance.

围绕城市景观格局、自然灾害相互影响及其生态风险评估与模拟的关键科学问题，本项目以深圳市为案例区，重点考虑台风、洪涝、水土流失、滑坡、泥石流等自然灾害，基于遥感、GIS与空间模拟技术及MATLAB等软件平台，综合考量景观斑块类型、空间格局及其组合特征变量，系统分析城市景观格局与自然灾害的空关联及其多尺度阈值、空间非平稳特征，深入探讨城市景观格局对自然灾害的影响机理，构建基于景观格局的城市自然灾害生态风险综合评估模型，明晰研究区生态风险空间分异特征并开展风险不确定性分析，模拟不同社会经济发展情景下城市景观格局与自然灾害生态风险的时空动态，探索基于景观格局优化的城市自然灾害生态风险防范途径。本项目研究有助于在实践中发展中国景观生态学，拓展城市景观生态学与生态风险的研究领域，丰富人类-自然耦合系统研究的景观途径，服务于城市防灾减灾与生态文明建设的实践需求，具有十分重要的科学及实践意义。

本项目以我国典型快速城市化地区深圳市为例，基于其人口高度集中、气象灾害多发、早期开发强度较大等特点，开展深圳市城市景观格局与自然灾害生态风险研究。利用长时间序列气象数据、统计数据、遥感数据以及空间化属性数据分别就洪涝、土壤侵蚀、城市热岛、极端气候、热带气旋等深圳市当前主要的灾害与生态环境质量退化的问题，结合研究区景观格局变化特征，深入探讨其与生态风险间的作用机理。. 项目围绕城市景观格局、城市自然灾害生态风险综合评估与情景模拟等关键科学问题开展，主要成果体现4个部分：(1)通过研究区景观格局特征变化归纳深圳市城市景观格局演变的驱动机制；同时分别对热带气旋、极端气候等灾害的时空动态特征进行分析，进而完善研究区景观格局与城市灾害时空数据库；(2)基于景观格局特征，分析其对城市内涝、土壤侵蚀及大气污染程度的影响，探讨不同尺度的自然灾害影响阈值，进而说明基于景观格局与强弱风险源间的复杂关系及综合因素所导致的风险空间变异性；(3)分别构建了暴雨洪涝、高温热浪及极端气候等灾害的风险评估模型，同时结合地质灾害、人类活动强度、植被空间分布、水土流失威胁、地表热环境影响、经济社会发展情况、景观类型空间变异性、生态用地连通程度、城市开发对周边的干扰、景观类型间的增益效果、绿色植被空间恢复程度、社会经济发展程度等城市发展现状、景观格局特征与灾害威胁等因素，构建城市生态风险综合评估模型，并对深圳市城市生态风险进行评估；(4)由土地政策与景观格局视角，设置不同的情景，以说明不同情景的其生态效应，同时基于城市安全与灾害防范需求，开展防灾避险空间分析与景观生态安全格局源地识别，进一步提出深圳市城市景观结构优化策略。. 本研究通过揭示城市景观格局与自然灾害的相互影响机理，系统的研究城市自然灾害生态风险，进而完善城市生态风险评估的相关理论与方法，并为深圳市提供城市灾害防范的景观生态安全保障。