Ecosystem health responses to tourism activities in the Qinghai-Tibet Plateau: Nonlinear relationships and mechanism

Rapid tourism development in ecologically fragile regions poses a risk of exceeding ecosystem self-regulation capabilities. To promote the coordination of conservation and development, this study aims to explore the nonlinear relationship between ecosystem health and tourism activities. This study established a tourism activity intensity index and employed a machine learning model to simulate ecological responses on the Qinghai-Tibet Plateau. The results reveal that the response of ecosystem health to tourism activity is characterized by distinct nonlinear features. While ecosystem health maintains relative stability under low-to-moderate disturbance, it suffers accelerated degradation once the intensity exceeds critical thresholds. Specifically, quantitative thresholds were identified at the kilometer-grid scale (0.134) and the trajectory-point scale (0.337), suggesting a scale-dependent resilience mechanism. Natural geographical conditions and tourism organization patterns jointly shape diverse nonlinear response patterns. These insights bridge the gap between theoretical carrying capacity and quantitative estimation, offering practical references for defining the safe operating space in high-altitude areas. • Enables high-precision estimation of tourism activity intensity • EH shows nonlinear threshold responses to tourism intensity • Cross-scale analysis reveals pronounced spatial heterogeneity • Four threshold response types identified with tipping-point risks • Provides insights to reconcile conservation and tourism in fragile regions