Tracking a Thawing World: Using Data Science to Understand Norway Permafrost Change

June 29, 2026

By Aubrie Engen, University of Alaska Student and SDSC Communications Intern

Across the hillsides of Norway, a slow but significant transformation is underway: permafrost is thawing. In northern regions, where soils repeatedly freeze and thaw, these changes can reshape entire landscapes. Norway’s permafrost-covered terrain offers researchers a unique opportunity to better understand how climate-driven changes may affect frozen ground systems around the world.

JohnPaul Sleiman, a recent Ph.D. graduate in Earth and Environmental Sciences from the University of Rochester, is conducting a six-month research project focused on permafrost geomorphology in Norway. His goal is to quantify how seasonal freeze-thaw cycles influence land deformation caused by water movement in permafrost-affected environments.

To do this, Sleiman is analyzing open-access geospatial datasets collected 22 previously mapped sites across Norway. 

“These locations have experienced solifluction, a process in which water-saturated soil slowly moves downhill, causing erosion and structural changes in landscapes underlain by frozen ground,” Sleiman explained. “By examining these sites, we hope to better understand how warming temperatures and changing moisture conditions are altering terrain stability.”

Sleiman said that the project relies on advanced computational tools to transform massive amounts of environmental data into meaningful insights. Using DeCODER (Democratized Cyberinfrastructure for Open Discovery to Enable Research) analysis frameworks, Sleiman processes and evaluates large geospatial datasets to identify patterns of landscape change.

A key component of the research is the use of GMSTAR, an Interferometric Synthetic Aperture Radar (InSAR) tool built on Generic Mapping Tools. GMSTAR allows researchers to generate highly detailed, time-series visualizations of land movement, revealing changes in terrain with millimeter-level precision. These “high-precision time series,” as Sleiman describes them, provide a replay-like view of how landscapes evolve over time.

Caption: Red dots are 10% of the InsAR points and the blue lines are the lobes 

Caption: InsAR points velocities colored by movement 

To gain a more complete picture, the deformation data is compared with satellite observations of soil moisture, temperature and other environmental variables. By integrating these datasets, researchers can identify links between unusual weather patterns and changes in ground stability, helping to explain how permafrost landscapes respond to a warming climate.

“Our research offers several benefits because datasets such as these will increase our capabilities for assuring preventative weather and land disaster safety measures, as well as predicting changes in landscape and ground stability,” Sleiman said. “The tools in this project allow scientists to continue to compile massive, informational datasets which can be made open-source in an effort to contribute to a global understanding of how the world’s terrain is morphing under warming global temperatures.”  

The project’s computational notebook is available on Zenodo. This project was partially funded by  DeCODER.