【摘要】 研究海岛型城市在自然-社会复合胁迫下的生态脆弱性，为陆海统筹视角下的生态空间管控提供科学依据。以福建省平潭综合实验区为研究区，构建压力-活力-组织力-复原力（PVOR）模型，利用博弈论组合赋权法确定综合权重体系，集成遥感、统计与地理空间多源数据，开展2000—2020年生态脆弱性长时序演变监测及影响因子分析。结果显示：研究区生态脆弱性指数（EVI）从2000年的0.52持续攀升至2020年的0.64，重度脆弱区面积占比增长21.3%，生态环境退化显著，脆弱性上升区域主要集中在城镇建设开发区、填海造陆区及人工海岸线周边区域；生态脆弱性表现出较强的空间集聚特征，全局Moran’s I值从0.929升至0.958(P<0.05)；高-高集聚区主要分布在海坛街道的城镇中心，且呈扩张趋势，低-低集聚区主要分布在北部君山一带；单因子与交互探测显示，生态质量（q=0.59）、生境质量（q=0.63）、岛屿开发强度（q=0.53）和景观破碎度（q=0.31）为主要影响因素，且因子间的交互作用均大于单个影响因子的贡献度。更多还原

【Abstract】 Accurate assessment of ecological vulnerability for island cities facing a combination of natural and anthropogenic stresses is crucial for maintaining insular ecosystem integrity and guiding sustainable resource utilization. In this study, the Pingtan comprehensive experimental zone of Fujian Province was selected for case study and we explored the spatiotemporal distribution of ecological vulnerability in the zone based on the Pressure-Vigor-Organization-Resilience(PVOR) conceptual model, developed as part of this project. Meanwhile, we analyzed key influencing factors of ecological vulnerability using geodetector. The construction of the PVOR model was based on data from remote sensing images, statistical surveys, and geospatial information in the study area at five-year intervals from 2000 to 2020. The weighting of each indicator was established using game theory combination weighting. Applying this framework, we conducted long-term spatiotemporal evolution monitoring of ecological vulnerability in the study area from 2000 to 2020 and analyzed key influencing factors. Results show:(1) The comprehensive ecological vulnerability index was 0.52, 0.53, 0.58, 0.59 and 0.64, respectively, in 2000, 2005, 2010, 2015and 2020, a consistent increase. The ecological vulnerability in Pingtan was generally classified as mild to moderate, accounting for approximately 70% of the total study area. However, the high-vulnerability area expanded by 21.3%, and environmental degradation was evident. The areas of rising ecological vulnerability were primarily concentrated in urban construction and development zones, land reclamation areas, and regions surrounding artificial coastlines.(2) Ecological vulnerability exhibited strong spatial clustering and significant spatial autocorrelation characteristics, with the global Moran’s I value increasing from0.929 to 0.958(P<0.05). High-high clustering areas(regions with high vulnerability surrounded by highvulnerability areas) were primarily concentrated in the urban center of the Haitan subdistrict and were expanding, while low-low clustering areas were mainly distributed in the Junshan area in the north.(3) Geographical detector analysis revealed that ecological quality(q=0.59), habitat quality(q=0.63), island development intensity(q=0.53), and landscape fragmentation(q=0.31) were the key factors influencing ecological vulnerability. Further, the interactive effects between factors exceeded the individual contributions of any single factor and the interactive effects were non-linear. Our study will provide a scientific basis for ecological restoration and spatial planning in island cities.更多还原