Water beneath glaciers in Antarctica can influence how the ice slides or melts. Airborne radar can detect this water, which looks bright in radar images. However, common techniques cannot identify the water's size or shape. We used a simulator to show how the radar image changes based on the bed material, size, and shape of the waterbody. This technique was applied to a suspected waterbody beneath Thwaites Glacier. We found it may be consistent with a series of wide, flat canals or a lake.

We explore how the shape of the land and underwater features in Scandinavia affected the former Scandinavian ice sheet over time. Using a computer model, we simulate how the ice sheet evolved during different stages of landscape development. We discovered that early glaciations were limited in size by underwater landforms, but as these changed, the ice sheet expanded more rapidly. Our findings highlight the importance of considering landscape changes when studying ice-sheet history.

Associated with climate change between the Last Glacial Maximum and the current interglacial period, we reconstruct the behaviour of the southwestern Laurentide Ice Sheet, which covered the Canadian Prairies, using detailed landform mapping. Our reconstruction depicts three shifts in the ice sheet’s dynamics. We suggest these changes resulted from ice sheet thinning triggered by abrupt climatic change. However, we show that regional lithology and topography also play an important role.

We developed an algorithm to estimate snow mass using X- and dual Ku-band radar, and tested it in a ground-based experiment. The algorithm, the Bayesian-based Algorithm for SWE Estimation (BASE) using active microwaves, achieved an RMSE of 30 mm for snow water equivalent. These results demonstrate the potential of radar, a highly promising sensor, to map snow mass at high spatial resolution.

Ice crystals often show a rod-like, vertical orientation in snow and firn; they are said to be anisotropic. The stiffness in the vertical direction therefore differs from the horizontal, which, for example, impacts the propagation of seismic waves. To quantify this anisotropy, we conducted finite-element simulations of 391 snow, firn, and ice core microstructures obtained from X-ray tomography. We then derived a parameterization that may be employed for advanced seismic studies in polar regions.

Individual satellite passes instead of daily means of sea ice concentration are used to correct the sea ice model forecast in the Barents Sea. The use of passes provides a significantly larger improvement of the forecasts even after a 7 d period due to the more precise information on temporal and spatial variability contained in the passes. One major advantage of the use of satellite passes is that there is no need to wait for the daily mean availability in order to update the forecast.

This paper introduces an automated sea ice mapping pipeline utilizing a multi-task U-Net architecture. It attained the top score of 86.3 % in the AutoICE challenge. Ablation studies revealed that incorporating brightness temperature data and spatial–temporal information significantly enhanced model accuracy. Accurate sea ice mapping is vital for comprehending the Arctic environment and its global climate effects, underscoring the potential of deep learning.

In 2018 we drilled an 18 m ice core on the summit of Grigoriev ice cap, located in the Tien Shan mountains of Kyrgyzstan. The core analysis reveals strong melting since the early 2000s. Regardless of this, we find that the structure and temperature of the ice have changed little since the 1980s. The probable cause of this apparent stability is (i) an increase in snowfall and (ii) the fact that meltwater nowadays leaves the glacier and thereby removes so-called latent heat.

The age of the surface blue ice in the Grove Mountains area is dated to be about 140 000 years, and one meteorite found here is 260 000 years old. It is inferred that the Grove Mountains blue-ice area holds considerable potential for paleoclimate studies.

The specific surface of snow dictates key physical properties and continuously evolves in natural snowpacks. This is referred to as metamorphism. This work develops a rigorous physical model for this evolution, which is able to reproduce X-ray tomography measurements without using unphysical tuning parameters. Our results emphasize that snow crystal growth at the micrometer scale ultimately controls the pace of metamorphism.

Rock glaciers in ice-rich permafrost can be discriminated from debris-covered glaciers. The key physical phenomenon relates to the tight mechanical coupling between the moving frozen body at depth and the surface layer of debris in the case of rock glaciers, as opposed to the virtually inexistent coupling in the case of surface ice with a debris cover. Contact zones of surface ice with subsurface ice in permafrost constitute diffuse landforms beyond either–or-type landform classification.

We use a sea ice model to reproduce ice growth observations from two buoys deployed on coastal sea ice and analyze the improvements brought by new physics that represent the presence of saline liquid water in the ice interior. We find that the new physics with default parameters degrade the model performance, with overly rapid ice growth and overly early snow flooding on top of the ice. The performance is largely improved by simple modifications to the ice growth and snow-flooding algorithms.

On the Antarctic Peninsula, there is a small bay that had sea ice fastened to the shoreline (fast ice) for over a decade. The fast ice stabilized the glaciers that fed into the ocean. In January 2022, the fast ice broke away. Using satellite data we found that this was because of low sea ice concentrations and a high long-period ocean wave swell. We find that the glaciers have responded to this event by thinning, speeding up, and retreating by breaking off lots of icebergs at remarkable rates.

We use the ice surface expression of the Gamburtsev Subglacial Mountains in East Antarctica to map the horizontal pattern of valleys and ridges in finer detail than possible from previous methods. In upland areas, valleys are spaced much less than 5 km apart, with consequences for the distribution of melting at the bed and hence the likelihood of ancient ice being preserved. Automated mapping techniques were tested alongside manual approaches, with a hybrid approach recommended for future work.

This study models present and future rock wall temperatures in the mountains near Sisimiut, creating knowledge on mountain permafrost in Greenland for the first time. Bedrock is mostly frozen but also has temperatures near 0 ^oC, making it very sensitive to climate changes. Future climatic scenarios indicate a reduction in frozen rock wall areas. Since mountain permafrost thaw is linked to an increase in landslides, these results call for more efforts addressing mountain permafrost in Greenland.

Climate warming is thawing permafrost that makes up palsa (frost mound) peatlands, risking ecosystem collapse and carbon release as methane. We measure this regional degradation using radar satellite technology to examine ground elevation changes and show how terrain roughness measurements can be used to estimate local permafrost damage. We find that over half of Sweden's largest palsa peatlands are degrading, with the worse impacts to the north linked to increased winter precipitation.

This paper focuses on predicting Arctic-wide sea-ice thickness using surrogate modeling with deep learning. The model has a predictive power of 12 h up to 6 months. For this forecast horizon, persistence and daily climatology are systematically outperformed, a result of learned thermodynamics and advection. Consequently, surrogate modeling with deep learning proves to be effective at capturing the complex behavior of sea ice.

Understanding snow phenology (SP) and its possible feedback are important. We reveal spatiotemporal heterogeneous SP on the Tibetan Plateau (TP) and the mediating effects from meteorological, topographic, and environmental factors on it. The direct effects of meteorology on SP are much greater than the indirect effects. Topography indirectly effects SP, while vegetation directly effects SP. This study contributes to understanding past global warming and predicting future trends on the TP.

Here we validate soil temperatures from eight reanalysis products across the pan-Arctic and compare their performance to a newly calculated ensemble mean soil temperature product. We find that most product soil temperatures have a relatively large RMSE of 2–9 K. It is found that the ensemble mean product outperforms individual reanalysis products. Therefore, we recommend the ensemble mean soil temperature product for the validation of climate models and for input to hydrological models.

The West Antarctic Ice Sheet is losing ice at an accelerating pace. This is largely due to the presence of warm ocean water around the periphery of the Antarctic continent, which melts the ice. It is generally assumed that the strength of this process is controlled by the temperature of the ocean. However, in this study we show that an equally important role is played by the changing geometry of the ice sheet, which affects the strength of the ocean currents and thereby the melt rates.

Our results indicate millennial-scale oscillations of the last Fennoscandian Ice Sheet (FIS) in northern Poland between ~19000 and ~17000 years ago. Combined luminescence (OSL) and ¹⁰Be dating show the last FIS left basal tills of three ice re-advances at a millennial-scale cycle: 19.2 ± 1.1 ka, 17.8 ± 0.5 ka and 16.9 ± 0.5 ka. This is the first terrestrial record of millennial-scale palaeo-ice margin oscillations at the southern fringe of the FIS during the last glacial cycle.

A large fraction of the mass loss from marine-terminating glaciers is attributed to frontal ablation. In this study, we used a 3D ice flow model of a real glacier that includes the effects of calving and submarine melting. Over a 30-month simulation, we found that the model reproduced the seasonal cycle for this glacier. Besides, the front positions were in good agreement with observations in the central part of the front, with longitudinal differences, on average, below 15 m.

Accurate knowledge of firn grain size is crucial for many ice sheet research applications. Unfortunately, collecting detailed measurements of firn grain size is difficult. We demonstrate that scanning firn cores with a near-infrared imager can quickly produce high-resolution maps of both grain size and ice layer distributions. We map grain size and ice layer stratigraphy in 14 firn cores from Greenland and document changes to grain size and ice layer content from the extreme melt summer of 2012.

Ice fracture can cause glacier crevassing and calving. These natural hazards can also modulate the flow and evolution of ice sheets. In a new study, we use a new high-resolution dataset to determine a new failure criterion for glacier ice. Surprisingly, the strength of ice depends on the mode of deformation, and this has potential implications for the currently used flow law of ice.

Information about snow depth is important within climate research but also many other sectors, such as tourism, mobility, civil engineering, and ecology. Climate models often feature a spatial resolution which is too coarse to investigate snow depth. Here, we analyse high-resolution simulations and identify added value compared to a coarser-resolution state-of-the-art product. Also, daily snow depth extremes are well reproduced by two models.

We use the IMAU firn densification model to simulate the 21st-century evolution of Antarctic firn air content. Ice shelves on the Antarctic Peninsula and the Roi Baudouin Ice Shelf in Dronning Maud Land are particularly vulnerable to total firn air content (FAC) depletion. Our results also underline the potentially large vulnerability of low-accumulation ice shelves to firn air depletion through ice slab formation. 

Antarctic sea ice maintains year-round snow cover, crucial for its energy and mass budgets. Despite its significance, snow depth remains poorly understood. Over the last decades, Snow Buoys have been deployed extensively on the sea ice to measure snow accumulation but not actual depth due to snow transformation into meteoric ice. Therefore, in this study we utilize sea ice and snow models to estimate meteoric ice fractions in order to calculate actual snow depth in the Weddell Sea.