Temperature and deformation response under the influence of continuous typhoons in seasonal permafrost rainfall-induced landslide evolution

Traditional landslide early-warning systems usually focus on displacement, rainfall, and stress, neglecting temperature effects. Rock deformation and failure processes are fundamentally characterized by energy dissipation, which can manifest through thermal changes. It is essential to explore whether temperature response inside landslides could help predict instability. This study investigates a representative landslide in southeastern Jilin Province, which has repeatedly experienced intense disturbances due to consecutive typhoon events. By employing a comprehensive multi-source monitoring strategy, including digital elevation model differencing (DoD), underground instrumentation, and detailed field surveys, we tracked temperature fluctuations, internal deformation, and surface displacement of the landslide. Our findings demonstrate that exceptionally intense rainfall events, occurring approximately once every sixty years, nearly reactivated landslides previously considered stable. Reactivated landslides exhibited accelerated movement, presenting severe hazards to infrastructure and nearby communities. Significantly, temperature within the landslide mass displayed distinctive fluctuation patterns at the onset of instability: Sharp Peak & Gentle Slope (SPGS) and Roller-Coaster (RC). These thermal signatures correlated strongly with rainfall intensity and deformation rates. To explain these observations, we propose two theoretical frameworks: the ambient heat-driven SPGS fluctuation patterns and the spontaneous heat-driven RC fluctuation patterns. The coupled temperature-accumulated rainfall-rainfall intensity analyses showed that: Temperature data can establish joint thresholds for cumulative rainfall and rainfall intensity. Further validation of the SPGS fluctuation patterns is essential for practical application in rainfall-induced landslide early-warning systems. Further research and validation efforts are essential to conclusively determine the predictive value and reliability of temperature changes induced by rainfall in forecasting landslide initiation.