Abstract: Enhancing the resilience of forest ecosystems is critical for promoting sustainable forestry and regional collaborative governance. To optimize sustainable forestry strategies and strengthen ecological civilization construction, we constructed an evaluation index system for measuring the resilience of forest ecosystems in 31 provinces of China from 2004 to 2021 based on the pressure-state-response (PSR) framework, using the entropy method to determine indicator weights. We established a spatial association network through a modified gravity model and applied social network analysis along with quadratic assignment procedure (QAP) to analyze the structural characteristics and driving factors of the network. The results showed that the overall resilience of China’s forest ecosystems increased from 0.13 in 2004 to 0.18 in 2021. There were regional variations across the eastern, central, western, and northeastern regions. The overall network connectivity of China’s forest ecosystem resilience required improvement, with individual network structures exhibiting pronounced “core-periphery” characteristics. Henan, Hubei, and Hunan occupied core positions, while Jilin, Heilongjiang, and Tianjin were situated at peripheral locations. Forest ecosystem resilience spatial association network could be divided into four major blocks, characterized by relatively strong internal connections within each block and relatively weak interactions between blocks. Differences in regional total forestry output value, geographical adjacency relationship, forestry ecological construction investment, and forest coverage exerted a strong positive influence on the establishment of the spatial association network, while differences in the proportion of forestry tertiary industry in GDP exerted a weak negative influence. Our results suggested that we should rely on the location advantages of core provinces and regions, strengthen cross-regional cooperation and resource sharing, and give play to the spatial radiation effect of core regions. Meanwhile, we should optimize the spatial allocation of forestry resource elements and develop targeted implementation plans.

Key words: forest ecosystem resilience, spatial association network, social network analysis method, quadratic assignment procedure