The thermal coupling between the atmosphere and the subsurface on the Qinghai-Tibetan Plateau (QTP) governs permafrost stability, surface energy balance, and ecosystem processes, yet its spatiotemporal dynamics under accelerated warming are poorly understood. This study quantifies soil-atmosphere thermal coupling ((3) at the critical 0.1 m root-zone depth using in-situ data from 99 sites (1980-2020) and a machine learning framework. Results show significantly weaker coupling in permafrost (PF) zones (mean (3 = 0.42) than in seasonal frost (SF) zones (mean (3 = 0.50), confirming the powerful thermal buffering of permafrost. Critically, we find a widespread trend of weakening coupling (decreasing (3) at 66.7 % of sites, a phenomenon most pronounced in SF zones. Our driver analysis reveals that the spatial patterns of (3 are primarily controlled by surface insulation from summer rainfall and soil moisture. The temporal trends, however, are driven by a complex and counter-intuitive interplay. Paradoxically, rapid atmospheric warming is the strongest driver of a strengthening of coupling, likely due to the loss of insulative snow cover, while trends toward wetter conditions drive a weakening of coupling by enhancing surface insulation. Spatially explicit maps derived from our models pinpoint hotspots of accelerated decoupling in the eastern and southern QTP, while also identifying high-elevation PF regions where coupling is strengthening, signaling a loss of protective insulation and increased vulnerability to degradation. These findings highlight a dynamic and non-uniform response of land-atmosphere interactions to climate change, with profound implications for the QTP's cryosphere, hydrology, and ecosystems.

Classifying a given landscape on the basis of its susceptibility to surface processes is a standard procedure in low to mid-latitudes. Conversely, these procedures have hardly been explored in periglacial regions. However, global warming is radically changing this situation and will change it even more in the future. For this reason, un-derstanding the spatial and temporal dynamics of geomorphological processes in peri-arctic environments can be crucial to make informed decisions in such unstable environments and shed light on what changes may follow at lower latitudes. For this reason, here we explored the use of data-driven models capable of recognizing locations prone to develop retrogressive thaw slumps (RTSs) and/or active layer detachments (ALDs). These are cryo-spheric hazards induced by permafrost degradation, and their development can negatively affect human set-tlements or infrastructure, change the sediment budget and release greenhouse gases. Specifically, we test a binomial Generalized Additive Modeling structure to estimate the probability of RST and ALD occurrences in the North sector of the Alaskan territory. The results we obtain show that our binary classifiers can accurately recognize locations prone to RTS and ALD, in a number of goodness-of-fit (AUCRTS = 0.83; AUCALD = 0.86), random cross-validation (mean AUCRTS = 0.82; mean AUCALD = 0.86), and spatial cross-validation (mean AUCRTS = 0.74; mean AUCALD = 0.80) routines. Overall, our analytical protocol has been implemented to build an open-source tool scripted in Python where all the operational steps are automatized for anyone to replicate the same experiment. Our protocol allows one to access cloud-stored information, pre-process it, and download it locally to be integrated for spatial predictive purposes.