[Ecological network construction and resilience evaluation considering typhoon disaster sensitivity of coastal regions in Beibu Gulf, Guangxi, China].

Coastal ecosystems are typical ecologically fragile zones, the ecological security and spatial stability of which face severe threats from typhoon disasters. Taking the Beibu Gulf coastal region of Guangxi as research zone, we integrated typhoon disaster sensitivity assessment into the process of ecological source identification and resis-tance surface construction. Based on the minimum cumulative resistance model, we extracted ecological corridors and constructed a coastal ecological network oriented toward disaster adaptation. Following complex system theory, we established an "element-structure-function" resilience evaluation model to assess the resilience of the ecological network, and compared it with traditional ecological network construction methods. The results showed that a total of 15 ecological source areas were identified, covering 3297.31 km2. Twenty-three ecological corridors were constructed, with a total length of 734.62 km and an average length of 31.94 km, breaking the traditional pattern of ecological networks concentrated in inland mountainous regions and extending ecological space to coastal risk zones. At the structural level, the improved ecological network demonstrated higher connectivity, closure, and line-point ratio by 18%, 63.6%, and 19.5%, respectively, compared to the traditional ecological network. Network effectiveness increased by 41.2%, with significant improvements in connectivity and stability. At the functional level, the robustness index of the improved ecological network rose from 0.49 to 0.56 compared to the traditional network, indicating higher resilience. The coverage of source areas and corridors in high-sensitivity typhoon zones increased from 8.5% and 6.2% to 53.6% and 20.0%, respectively, indicating stronger disaster prevention buffering capacity and regional ecological security support. The findings expanded the theoretical framework of ecological networks and would provide scientific support for ecological network optimization and disaster risk management under disaster scenarios.