Long-Term Evolution of Permafrost across the Qinghai–Tibet Plateau: Perspectives from Multimodel Ensembles and Machine Learning

Abstract The Qinghai–Tibet Plateau (QTP) is highly vulnerable to climate change, with permafrost changes exerting substantial impacts on regional ecosystems, hydrological stability, and cryospheric water resources. This study combines Coupled Model Intercomparison Project phase 6 (CMIP6) data with four machine learning algorithms, including support vector regression (SVR), categorical boosting (CatBoost), light gradient boosting machine (LightGBM), and deep neural network (DNN), to construct optimal models for predicting permafrost extent and maximum seasonal soil freeze depth (SFD) across the QTP. The relative importance and marginal effects of predictors were further evaluated using the random forest (RF) analysis and Shapley additive explanations (SHAP). Permafrost evolution from 2025 to 2100 was projected under four shared socioeconomic pathway (SSP) scenarios (SSP1-2.6, SSP2-4.5, SSP3-7.0, SSP5-8.5). Results revealed that the DNN ( R 2 = 0.961) outperforms other models in capturing frozen ground dynamics. Key drivers of SFD were identified as summer–autumn precipitation, temperature, and nonfrozen season soil moisture. Permafrost is projected to continuously degrade into seasonally frozen ground, with transition zones expanding from plateau margins toward the interior, leading to increased spatial fragmentation. SFD is expected to decline at rates strongly correlated with emission scenarios, with the most pronounced reductions along current permafrost margins and gradually extending inland. Notably, permafrost stability in high-altitude regions will face increasing challenges. The northern Tibetan Plateau, the margins of the Qaidam Basin, and the foothills of the Gangdise Mountains were identified as high-risk zones for frozen ground degradation under continued carbon dioxide (CO 2 ) emission trajectories. Significance Statement Frozen ground is rapidly degrading due to climate warming, yet few studies have examined its interannual variability and underlying response mechanisms. This study develops an integrated modeling framework combining Coupled Model Intercomparison Project phase 6 (CMIP6) multiscenario projections and focuses on the Qinghai–Tibet Plateau (QTP) as a representative alpine region. The framework enables high-resolution prediction and causal interpretation of frozen ground dynamics, capturing future spatiotemporal changes and elucidating the key driving factors and the underlying mechanisms. The findings are essential for understanding permafrost response in cold regions and provide a scientific basis for infrastructure planning, water resource management, and ecological risk mitigation under warming scenarios.