[Spatial Correlation Network and Driving Factors of Tourism Ecological Resilience in the Yangtze River Delta Urban Agglomeration].

Based on the framework of resilience theory， the level of tourism ecosystem resilience is systematically evaluated， and the structural characteristics and driving factors of its spatial correlation network are revealed， which is of great significance for practicing the concept of green development and promoting high-quality development of the tourism industry. The level of tourism ecological resilience in the Yangtze River Delta urban agglomeration from 2008 to 2022 is measured using a comprehensive level evaluation model， and its spatial network characteristics and driving factors are revealed by using the modified gravity model， core-edge analysis method， spatiotemporal geographic weighted regression， spatial econometric， and threshold model. The results show that： ① The tourism ecological resilience of the Yangtze River Delta urban agglomeration presented the spatial correlation characteristics of "local tightness and overall looseness." The connection between the "Shanghai-Nanjing-Hangzhou-Suzhou" sub-network and other nodes was polarized， and the first connection nodes mostly appeared in core cities， showing the core-edge characteristics of the three-layered. ② Factors such as economic development， industrial structure， transportation infrastructure， human capital， opening up to the outside world， and the opening of high-speed railways significantly drove the spatial correlation of tourism ecological resilience and showed obvious belt-like or sheet-like distribution patterns. ③ The impacts of industrial structure， transportation infrastructure， opening up to the outside world， and high-speed rail opening on the spatial correlation of tourism ecological resilience had spatial spillover effects. In addition， the impacts of urbanization and marketization on the spatial correlation of tourism ecological resilience had threshold effects， showing a nonlinear characteristic of increasing marginal effects.