The rapidly warming climate on the Qinghai–Tibet Plateau (QTP) leads to permafrost degradation, and the thawing of ice-rich permafrost induces land subsidence to facilitate the development of thermokarst lakes. Thermokarst lakes exacerbate the instability of permafrost, which significantly alters regional geomorphology and hydrology, affecting biogeochemical cycles. However, the spatial distribution and future changes in thermokarst lakes have rarely been assessed at large scales. In this study, we combined various conditioning factors and an inventory of thermokarst lakes to assess the spatial distribution of susceptibility maps using machine-learning algorithms. The results showed that the extremely randomized trees (EXT) performed the best in the susceptibility modeling process, followed by random forest (RF) and logistic regression (LR). According to the assessment based on EXT, the high- and very high-susceptibility area of the present (2000–2016) susceptibility map was 196,222 km², covering 19.67% of the permafrost region of the QTP. In the future (the 2070s), the area of the susceptibility map was predicted to shrink significantly under various representative concentration pathway scenarios (RCPs). The susceptibility map area would be reduced to 37.06% of the present area in RCP 8.5. This paper also performed correlation and importance analysis on the conditioning factors and thermokarst lakes, which indicated that thermokarst lakes tended to form in areas with flat topography and high soil moisture. The uncertainty of the susceptibility map was further assessed by the coefficient of variation (CV). Our results demonstrate a way to study the spatial distribution of thermokarst lakes at the QTP scale and provide a scientific basis for understanding thermokarst processes in response to climate change.

Climate change is causing permafrost in the Qinghai–Tibet Plateau to degrade, triggering thermokarst hazards and impacting the environment. Despite their ecological importance, the distribution and risks of thermokarst lakes are not well understood due to complex influencing factors. In this study, we introduced a new interpretable ensemble learning method designed to improve the global and local interpretation of susceptibility assessments for thermokarst lakes. Our primary aim was to offer scientific support for precisely evaluating areas prone to thermokarst lake formation. In the thermokarst lake susceptibility assessment, we identified ten conditioning factors related to the formation and distribution of thermokarst lakes. In this highly accurate stacking model, the primary learning units were the random forest (RF), extremely randomized trees (EXTs), extreme gradient boosting (XGBoost), and categorical boosting (CatBoost) algorithms. Meanwhile, gradient boosted decision trees (GBDTs) were employed as the secondary learning unit. Based on the stacking model, we assessed thermokarst lake susceptibility and validated accuracy through six evaluation indices. We examined the interpretability of the stacking model using three interpretation methods: accumulated local effects (ALE), local interpretable model-agnostic explanations (LIME), and Shapley additive explanations (SHAP). The results showed that the ensemble learning stacking model demonstrated superior performance and the highest prediction accuracy. Approximately 91.20% of the total thermokarst hazard points fell within the high and very high susceptible areas, encompassing 20.08% of the permafrost expanse in the QTP. The conclusive findings revealed that slope, elevation, the topographic wetness index (TWI), and precipitation were the primary factors influencing the assessment of thermokarst lake susceptibility. This comprehensive analysis extends to the broader impacts of thermokarst hazards, with the identified high and very high susceptibility zones affecting significant stretches of railway and highway infrastructure, substantial soil organic carbon reserves, and vast alpine grasslands. This interpretable ensemble learning model, which exhibits high accuracy, offers substantial practical significance for project route selection, construction, and operation in the QTP.