Degradation of near-surface permafrost can pose a serious threat to the utilization of natural resources, and to the sustainable development of Arctic communities. Here we identify at unprecedentedly high spatial resolution infrastructure hazard areas in the Northern Hemisphere’s permafrost regions under projected climatic changes and quantify fundamental engineering structures at risk by 2050. We show that nearly four million people and 70% of current infrastructure in the permafrost domain are in areas with high potential for thaw of near-surface permafrost. Our results demonstrate that one-third of pan-Arctic infrastructure and 45% of the hydrocarbon extraction fields in the Russian Arctic are in regions where thaw-related ground instability can cause severe damage to the built environment. Alarmingly, these figures are not reduced substantially even if the climate change targets of the Paris Agreement are reached.

Mean annual ground temperature (MAGT) and active layer thickness (ALT) are key to understanding the evolution of the ground thermal state across the Arctic under climate change. Here a statistical modeling approach is presented to forecast current and future circum‐Arctic MAGT and ALT in relation to climatic and local environmental factors, at spatial scales unreachable with contemporary transient modeling. After deploying an ensemble of multiple statistical techniques, distance‐blocked cross validation between observations and predictions suggested excellent and reasonable transferability of the MAGT and ALT models, respectively. The MAGT forecasts indicated currently suitable conditions for permafrost to prevail over an area of 15.1 ± 2.8 × 106 km2. This extent is likely to dramatically contract in the future, as the results showed consistent, but region‐specific, changes in ground thermal regime due to climate change. The forecasts provide new opportunities to assess future Arctic changes in ground thermal state and biogeochemical feedback.

Abstract. Monitoring the thermal state of permafrost (TSP) is important in many environmental science and engineering applications. However, such data are generally unavailable, mainly due to the lack of ground observations and the uncertainty of traditional physical models. This study produces novel permafrost datasets for the Northern Hemisphere (NH), including predictions of the mean annual ground temperature (MAGT) at the depth of zero annual amplitude (DZAA) (approximately 3 to 25 m) and active layer thickness (ALT) with 1 km resolution for the period of 2000–2016, as well as estimates of the probability of permafrost occurrence and permafrost zonation based on hydrothermal conditions. These datasets integrate unprecedentedly large amounts of field data (1002 boreholes for MAGT and 452 sites for ALT) and multisource geospatial data, especially remote sensing data, using statistical learning modeling with an ensemble strategy. Thus, the resulting data are more accurate than those of previous circumpolar maps (bias = 0.02±0.16 ∘C and RMSE = 1.32±0.13 ∘C for MAGT; bias = 2.71±16.46 cm and RMSE = 86.93±19.61 cm for ALT). The datasets suggest that the areal extent of permafrost (MAGT ≤0 ∘C) in the NH, excluding glaciers and lakes, is approximately 14.77 (13.60–18.97) × 106 km2 and that the areal extent of permafrost regions (permafrost probability >0) is approximately 19.82×106 km2. The areal fractions of humid, semiarid/subhumid, and arid permafrost regions are 51.56 %, 45.07 %, and 3.37 %, respectively. The areal fractions of cold (≤-3.0 ∘C), cool (−3.0 ∘C to −1.5 ∘C), and warm (>-1.5 ∘C) permafrost regions are 37.80 %, 14.30 %, and 47.90 %, respectively. These new datasets based on the most comprehensive field data to date contribute to an updated understanding of the thermal state and zonation of permafrost in the NH. The datasets are potentially useful for various fields, such as climatology, hydrology, ecology, agriculture, public health, and engineering planning. All of the datasets are published through the National Tibetan Plateau Data Center (TPDC), and the link is https://doi.org/10.11888/Geocry.tpdc.271190 (Ran et al., 2021a).

The thermal state of permafrost affects Earth surface systems and human activity in the Arctic and has implications for global climate. Improved understanding of the local-scale variability in the global ground thermal regime is required to account for its sensitivity to changing climatic and geoecological conditions. Here, we statistically related observations of mean annual ground temperature (MAGT) and active-layer thickness (ALT) to high-resolution ( ∼ 1 km 2 ) geospatial data of climatic and local environmental conditions across the Northern Hemisphere. The aim was to characterize the relative importance of key environmental factors and the magnitude and shape of their effects on MAGT and ALT. The multivariate models fitted well to both response variables with average R 2 values being ∼ 0.94 and 0.78. Corresponding predictive performances in terms of root-mean-square error were ∼ 1.31 • C and 87 cm. Freezing (FDD) and thawing (TDD) degree days were key factors for MAGT inside and outside the permafrost domain with average effect sizes of 6.7 and 13.6 • C, respectively. Soil properties had marginal effects on MAGT (effect size = 0.4-0.7 • C). For ALT, rainfall (effect size = 181 cm) and solar radiation (161 cm) were most influential. Analysis of variable importance further underlined the dominance of climate for MAGT and highlighted the role of solar radiation for ALT. Most response shapes for MAGT ≤ 0 • C and ALT were nonlinear and indicated thresholds for covariation. Most importantly, permafrost temperatures had a more complex relationship with air temperatures than non-frozen ground. Moreover, the observed warming effect of rainfall on MAGT ≤0 • C reverted after reaching an optimum at ∼ 250 mm, and that of snowfall started to level off at ∼ 300-400 mm. It is suggested that the factors of large global variation (i.e. climate) suppressed the effects of local-scale factors (i.e. soil properties and vegetation) owing to the extensive study area and limited representation of soil organic matter. Our new insights into the factors affecting the ground thermal regime at a 1 km scale should improve future hemispheric-scale studies.

Ongoing climate change is causing fundamental changes in the Arctic, some of which can be hazardous to nature and human activity. In the context of Earth surface systems, warming climate may lead to rising ground temperatures and thaw of permafrost. This Data Descriptor presents circumpolar permafrost maps and geohazard indices depicting zones of varying potential for development of hazards related to near-surface permafrost degradation, such as ground subsidence. Statistical models were used to predict ground temperature and the thickness of the seasonally thawed (active) layer using geospatial data on environmental conditions at 30 arc-second resolution. These predictions, together with data on factors (ground ice content, soil grain size and slope gradient) affecting permafrost stability, were used to formulate geohazard indices. Using climate-forcing scenarios (Representative Concentration Pathways 2.6, 4.5 and 8.5), permafrost extent and hazard potential were projected for the 2041–2060 and 2061–2080 time periods. The resulting data (seven permafrost and 24 geohazard maps) are relevant to near-future infrastructure risk assessments and for targeting localized geohazard analyses.