Estimating for Subsurface Temperature in the Arctic: Study Case in the Miellajokka Catchment, Northern Sweden

Context & Objectives: The northern lands are experiencing a generalised increase in soil temperature, resulting in permafrost thaw and subsequent fast changes on water, heat and matter fluxes in these areas. This triggers many important consequences, including infrastructures destabilisation and release of greenhouse gases. Spaceborne thermal imaging can provide extensive and high-resolution information about the temperature of the arctic continental surfaces. Providing subsurface temperature maps at the scale of a catchment and understanding its interactions with the surface conditions is highly needed for studies of the climate warming induced arctic changes, including permafrost thawing.

Methods: In this study, we used downscaled meteorological data from Nordic Gridded Climate Dataset (NGCD), topographic maps, a land cover map of the region derived from Sentinel-1 and Sentinel-2 data and downscaled Sentinel-3 Land Surface Temperature (LST) images. These surface conditions were combined through a regression model with ten stations of in situ soil-temperature and water content observations positioned along an altitudinal gradient across the Miellajokka watershed, Abisko, Northern Sweden.

Results: We generated soil temperature surface maps for the Abisko region, covering an area of about 52 km² at 300 m spatial resolution. We studied the behaviour of top-layer soil temperature according to climatic conditions, water content, soil properties and surface vegetation.

Conclusion: The developed methodology aims at allowing using satellite images, as thermal observations, for deriving key information about soil thermal regime in the Arctics. By developing this kind of approach, the arctic science community may get tremendous benefit from the future launching of high-resolution TIR observation missions such as TRISHNA and LSTM, for instance for permafrost modelling and climate change impacts assessment.