Glacier-permafrost interactions in Alpine environments influence geomorphological processes, making it essential to understand the relationship between subsurface structures and surface morphodynamics for predicting landscape evolution under climate change. This study uses applied geophysics and remote sensing methods to investigate this relationship and to reveal the distribution of different ground ice types, as well as the related surface morphodynamics in three alpine glacier forefields.

Active layer thickness (ALT) is a sensitive indicator of the thawing Alaskan frozen soil, which may lead to increased greenhouse gas emissions, vegetation changes, and infrastructure damage. This study represents a multi-scale assessment of ALT spatial variations using observations including intensive field sampling, and drone, airborne and satellite remote sensing. Our study allows for improved interpretation of remote sensing and process-based ALT simulations for the changing Arctic.

Some glaciers flow slowly for many years before dramatically accelerating. This is usually a sign of a surge. Theory and observations suggest that surges occur in certain climates. The Antarctic Peninsula has such a climate yet only one surge has been observed there. We present detailed observations of three glaciers that surged recently. We explore how climate change over the past hundred years, and projected climate change up to 2150, has and will affect surging behaviour on the Peninsula.

We take advantage of ICESat-2 satellite, which measures surface heights with nominal resolution of meters to tens of meters, to characterize a near pan-Arctic winter sea ice leads, meter-scale elongated cracks in ice, from 2018 to 2024. We find that lead fractions are higher near the ice edge and sizes follow a power-law distribution. Four distinct features in the temporal evolution of lead fraction are also identified, with fraction increase associated with larger leads.

We developed a new statistical approach to track monthly changes in the surface height of the Greenland Ice Sheet using radar data from the CryoSat-2 satellite (2011–2025). The method lets seasonal and long-term variations emerge directly from the data, creating a consistent record of elevation change that helps reveal how Greenland responds to ongoing climate change.

Using satellite imagery and machine learning, we created the first Greenland-wide dataset of slush (waterlogged snow) from 2016 to 2024. Slush was widespread and varied strongly between years and across Greenland, becoming most extensive during high melt seasons. It was most common where melting was greater and snow had less capacity to store water. These findings show that slush should be included in models of future melting and sea-level rise.

We evaluate polarimetry for snow depth retrieval on Arctic winter transport routes. Over landfast ice and tundra, traditional dual-frequency (Ku/Ka) methods often fail. We demonstrate that cross-polarization (VH) provides a more reliable marker for snow/ice and snow/ground interfaces. For lakes, we successfully resolve air, snow, and ice interfaces to retrieve snow and ice thickness simultaneously. These results establish a baseline for future polarimetric satellite missions.

Through a series of direct shear tests, this study analyzes the variation of shear strength parameters (cohesion and internal friction angle) in compacted snow under different conditions of density, sintering time, and temperature. A Genetic Algorithm-Back Propagation neural network model was subsequently developed to establish systematic benchmark values for these parameters. This work provides essential data and a predictive framework for the reliable design of snow structures in cold regions.

Atmospheric rivers (ARs) are narrow corridors of intense heat and moisture that can deliver extreme weather to the Arctic. We investigate a record March 2022 event in Svalbard using measurements of air and precipitation to track its origin and evolution. While the AR brought unusual winter rain and snow melt, it also delivered substantial snowfall that partially offset glacier mass loss. These findings show that the impacts of ARs on Arctic glaciers are more complex than typically reported.

This study presents an observation-driven sea ice albedo parametrisation by discovering an equation using symbolic regression, an interpretable machine learning method. Leveraging satellite and reanalyses data, our discovered equation identifies high sensitivity to thin snow and the weighted temperature difference between sea ice surface and 2 m air as critical to determine sea ice albedo. Our findings contribute to improving Arctic climate projections and understanding.

Passive microwave radiometers are the primary tool for monitoring Antarctic sea ice, but their reliability decreases in the marginal ice zone between the pack ice and the open ocean. We simulate satellite observations to assess the impact of varying physical parameters on sea ice concentration retrievals. The largest errors are due to flooded/wet snow, thin ice, and roughened ocean surfaces. These findings can improve our interpretation of satellite observations and forecast sea ice changes.

Understanding past ice sheet behaviours is key to predicting future ice sheets in a warming climate, but meshing geological observations with ice sheet models is challenging. This manuscript models the sedimentary system of the Finnish Lake District Ice Lobe during its retreat about 12 000 years ago, when extensive water flow developed under the ice, and connects models with observations. The approach opens a path to improve knowledge of previously glaciated regions and systems under ice today.