Evaluation on the uninhabited island ecological sensitivity based on unique features, various disturbances, and ecological conditions

The ecological sensitivity in uninhabited islands denotes the susceptibility to degradation and low resilience in ecosystems arising from their unique features and exposure to various disturbances. Evaluating this sensitivity is particularly significant for implementing effective island integrated management strategies. We developed a robust evaluation model incorporating three critical dimensions: unique island features including form, isolation, terrain, and land cover; various disturbances consisting of natural factors such as meteorological, oceanic, geological, and biological influences along with diverse anthropogenic factors; and ecological conditions primarily represented by vegetation and soil characteristics. Through the integration of comprehensive fieldwork and advanced remote sensing data, this model was systematically applied to assess the uninhabited islands within northern China’s Miaodao Archipelago. The results demonstrated clear spatial heterogeneities in ecological sensitivity across two spatial scales. At the broader archipelago scale, quantitative assessments revealed that a majority of islands exhibited medium sensitivity levels, with both island area and distance to inhabited islands showing positive correlations with improved ecological conditions and enhanced disturbance resistance. When examining individual islands at finer scales, zones with medium, high, low, very low, and very high levels of uninhabited island ecosystem status followed a descending order of areas. Comparative analysis indicated that natural stresses exerted substantially greater influence on ecological sensitivity than human activities, with terrain characteristics and plant diversity emerging as the predominant factors driving spatial variance. While human impacts occurred at relatively low intensities overall, they nevertheless produced notable localized effects on sensitivity distribution patterns. These multi-scale spatial variations strongly emphasize the necessity for implementing precisely targeted conservation measures including establishing extensive and continuous ecological monitoring, carefully controlling and optimizing exploitation activities, and conducting scientifically informed ecological restoration projects to maintain uninhabited island ecosystems. The successful application of this evaluation model has conclusively verified its comprehensive nature, unique methodological approach, and ability to capture spatial heterogeneity, while simultaneously demonstrating its strong potential for assessing ecological sensitivity across diverse bedrock uninhabited island ecosystems worldwide.